FORMULATION OF A REGULATORY FRAMEWORK FOR …siteresources.worldbank.org/.../4p.pdf · FORMULATION OF A REGULATORY FRAMEWORK FOR MUNICIPAL BORROWING IN SOUTH AFRICA FINAL REPORT Draft

4 December, 1998

1

FORMULATION OF A REGULATORYFRAMEWORK FOR MUNICIPAL BORROWING

IN SOUTH AFRICA

FINAL REPORTDraft

4 December, 1998

Matthew D. GlasserThomas H. CochranMichael DeAngelisMarlene HeskethRonald W. JohnsonChristiaan Johannes KappJohn E. PetersenK.P.S “Paddy” Roome

4 December, 1998

2

ACKNOWLEDGEMENTS .............................................................................................................................................................6

INTRODUCTION............................................................................................................................................................................7

THE ISSUE...................................................................................................................................................................................... 7THE PROCESS ................................................................................................................................................................................ 8

EXECUTIVE SUMMARY................................................................................................................................................................9

FUNDAMENTALS:......................................................................................................................................................................... 9SUMMARY OF RECOMMENDATIONS:...................................................................................................................................... 10

Sec 1: Issuing Debt...............................................................................................................................................................10Sec 2: Disclosure and Market Regulation.......................................................................................................................11Sec 3: Debt Structure...........................................................................................................................................................11Sec 4: Restrictions on Municipal Debt.............................................................................................................................11Sec 5: Authorising Debt ......................................................................................................................................................12Sec 6: Concessional Finance and Other Assistance......................................................................................................12Sec 7: Liquidity and Secondary Markets.........................................................................................................................13Sec 8: Monitoring, Intervention and Remedies ..............................................................................................................13Sec 9: Macroeconomic issues.............................................................................................................................................14Sec 10: Implementation.......................................................................................................................................................14

SECTION 1: ISSUING DEBT.....................................................................................................................................................15

1.1 BACKGROUND AND CURRENT SITUATION ...................................................................................................................... 151.2 KEY CONCERNS AND POLICY ISSUES................................................................................................................................ 161.3 WHO SHOULD ISSUE DEBT ? .............................................................................................................................................. 17

Should there be legal or regulatory distinctions between municipalities?..............................................................17Should there be issuers of local public debt other than municipalities? ..................................................................18What about public-private partnerships? .......................................................................................................................19

1.4 WHAT KIND OF DEBT ?....................................................................................................................................................... 201.4.1 General debt ................................................................................................................................................................201.4.2 Special pledges, limited obligations, and “ring fence” financing ...................................................................21

1.4.2.1 Pledges of assets and revenue streams .................................................................................................................... 221.4.2.2 Pledges of the power to set tax rates, utility tariffs and other levies...................................................................... 231.4.2.3 Revenue intercepts.................................................................................................................................................. 231.4.2.4 Enterprise financing................................................................................................................................................ 241.4.2.5 Special District Financing........................................................................................................................................ 25

1.5 RECOMMENDATIONS: ......................................................................................................................................................... 26

SECTION 2: DISCLOSURE AND MARKET REGULATION...............................................................................................28

2.1 BACKGROUND AND CURRENT SITUATION ..................................................................................................................... 282.2 DISCUSSION AND ANALYSIS............................................................................................................................................... 30

2.2.1 Disclosure By Whom...................................................................................................................................................302.2.2 Disclosure To Whom...................................................................................................................................................312.2.3 Disclosure When (Frequency of disclosure)..........................................................................................................322.2.4 The Content of Disclosure: .......................................................................................................................................322.2.5 A Central Repository for Disclosure.......................................................................................................................342.2.6 Registration and Enforcement of Disclosure:.......................................................................................................35

SECTION 3 DEBT STRUCTURE..............................................................................................................................................37

3.1 BACKGROUND AND CURRENT SITUATION ..................................................................................................................... 373.2 KEY CONCERNS AND POLICY ISSUES............................................................................................................................... 37

4 December, 1998

3

3.3 TERM..................................................................................................................................................................................... 373.4 CASH FLOW PROFILE.......................................................................................................................................................... 383.5 DERIVATIVES....................................................................................................................................................................... 393.6 RECOMMENDATIONS ......................................................................................................................................................... 41

SECTION 4 RESTRICTIONS ON ISSUANCE AND USE OF MUNICIPAL DEBT.........................................................42

4.1 BACKGROUND AND CURRENT SITUATION ..................................................................................................................... 424.2 DECISION POINTS – SHORT TERM DEBT ......................................................................................................................... 434.3 DECISION POINTS -- LONG TERM DEBT ......................................................................................................................... 44

4.3.1 Particulars of concerns about long term debt .....................................................................................................444.3.2 Restrictions on purposes of long term debt ..........................................................................................................454.3.3 Recommendations as to long term debt.................................................................................................................46

4.4 HOW MAY PROCEEDS OF LONG TERM BORROWING BE USED?................................................................................... 474.4.1 Key concerns and policy issues about use of proceeds.......................................................................................474.4.2 Pros and cons of restricting use of proceeds.........................................................................................................474.4.3 Recommendations as to use of proceeds................................................................................................................48

4.5 SUMMARY OF RECOMMENDATIONS ............................................................................................................................... 48

SECTION 5 HOW SHOULD BORROWING BE AUTHORISED?.....................................................................................50

5.1 BACKGROUND AND CURRENT SITUATION ..................................................................................................................... 505.2 DECISION POINTS................................................................................................................................................................. 50

5.2.1 Within the municipality ............................................................................................................................................51Debt policy .......................................................................................................................................................................... 52Capital improvement strategy ............................................................................................................................................. 53

5.2.2 Beyond the municipality ...........................................................................................................................................545.3 RECOMMENDATIONS ......................................................................................................................................................... 55

SECTION 6: CONCESSIONAL FINANCE AND OTHER ASSISTANCE.......................................................................56

6.1 BACKGROUND AND CURRENT SITUATION .................................................................................................................... 566.2 KEY CONCERNS AND POLICY ISSUES......................................................................................................................... 576.3 WHO SHOULD RECEIVE ASSISTANCE?....................................................................................................................... 576.4 WHAT KIND OF ASSISTANCE?.................................................................................................................................... 58

6.4.1 Technical Assistance and Training.................................................................................................................596.4.2 Financial Assistance..........................................................................................................................................60

6.4.2.1 Direct Lending Positions.................................................................................................................................... 616.4.2.2 Direct Interest Rate Subsidies ............................................................................................................................ 616.4.2.3 Guarantees.......................................................................................................................................................... 616.4.2.4 Transfer Payment Intercepts ............................................................................................................................. 62

6.4.3 Special intermediaries for small borrowers .........................................................................................................636.5 RECOMMENDATIONS .......................................................................................................................................................... 64

SECTION 7 LIQUIDITY AND SECONDARY MARKETS....................................................................................................65

7.1 BACKGROUND AND CURRENT SITUATION.............................................................................................................. 657.2 KEY CONCERNS AND POLICY ISSUES......................................................................................................................... 667.3 MARKET PSYCHOLOGY ............................................................................................................................................... 667.4 CONTINUING DISCLOSURE OF INFORMATION REGARDING BORROWERS........................................................... 67

7.4.1 Continuing disclosure by issuers.....................................................................................................................687.4.2 Rating agencies...................................................................................................................................................687.4.3 DBSA .....................................................................................................................................................................687.4.4 “In-house” Investor Credit Assessment Capability......................................................................................697.4.5 Project Viability ..................................................................................................................................................69

7.5 ACCOUNTING FOR INVESTMENT HOLDINGS ........................................................................................................... 697.6 TRADING SYSTEMS....................................................................................................................................................... 70

4 December, 1998

4

7.7 IMPROVING LIQUIDITY OF MARKET INTERMEDIARIES......................................................................................... 717.8 FORM OF DEBT .............................................................................................................................................................. 727.9 RECOMMENDATIONS................................................................................................................................................... 72

SECTION 8 MONITORING, INTERVENTION AND REMEDIES........................................................................................73

8.1 BACKGROUND AND CURRENT SITUATION ............................................................................................................. 738.1.1 Monitoring Provisions and Practices ...................................................................................................................738.1.2 Intervention Provisions and Practices..................................................................................................................748.1.3 Lenders’ legal remedies............................................................................................................................................768.1.4 Registration................................................................................................................................................................768.1.5 Provincial capacity...................................................................................................................................................778.1.6 Lack of enforcement ..................................................................................................................................................778.1.7 Summary......................................................................................................................................................................77

8.2 DECISION POINTS................................................................................................................................................................. 78monitoring .............................................................................................................................................................................78Remedies.................................................................................................................................................................................798.2.1 Market Forces.............................................................................................................................................................798.2.2 The need for a national approach..........................................................................................................................798.2.3 Monitoring and intervention: constitutional considerations ..........................................................................808.2.4 Intervention: its manner and extent .......................................................................................................................81

Contractual relationships and remedies ............................................................................................................................... 81Provincial oversight and instruction..................................................................................................................................... 83Financial Control Board or Receiver.................................................................................................................................... 83Bankruptcy .......................................................................................................................................................................... 86

8.2.5 Creditors’ remedies...................................................................................................................................................868.3 RECOMMENDATIONS ......................................................................................................................................................... 87

SECTION 9 MACROECONOMIC ISSUES..............................................................................................................................88

9.1 MUNICIPAL REVENUES...................................................................................................................................................... 889.1.1 Background................................................................................................................................................................889.1.2 Types of municipal revenues....................................................................................................................................889.1.3 Analysis ........................................................................................................................................................................90

9.2 OUTSTANDING MUNICIPAL DEBT .................................................................................................................................... 919.2.1 Sources of municipal funding..................................................................................................................................919.2.2 Categories of funding ...............................................................................................................................................92

9.3 DEMOGRAPHICS .................................................................................................................................................................. 939.4 SHOULD THERE BE TAX INCENTIVES TO MAKE BORROWING CHEAPER?.............................................................. 94

9.4.1 Background................................................................................................................................................................959.4.2 Undesirable Shifts in Tax Burden vs. Desirable Leveraging..............................................................................959.4.3 Decision points regarding tax incentives .............................................................................................................96

9.5 RECOMMENDATIONS ......................................................................................................................................................... 97

SECTION 10 IMPLEMENTATION...........................................................................................................................................98

ADDENDUM 1................................................................................................................................................................................99

INTERVIEWEES........................................................................................................................................................................... 99

ADDENDUM 2............................................................................................................................................................................. 108

PROPOSED MUNICIPAL BORROWING BILL ......................................................................................................................... 108

ADDENDUM 3............................................................................................................................................................................. 112

REGULATORY IMPLEMENTATION STEPS............................................................................................................................ 112SUPPORTIVE INTERVENTIONS .............................................................................................................................................. 114

4 December, 1998

5

ADDENDUM 4............................................................................................................................................................................. 116

SUMMARY OF MUNICIPAL DEBT MARCH 1997 – MARCH 1998....................................................................................... 116

ADDENDUM 5............................................................................................................................................................................. 117

INTEREST & REDEMPTION AS A PERCENTAGE OF INCOME............................................................................................. 117

ADDENDUM 6............................................................................................................................................................................. 119

AN OFF THE SHELF LOAN SYSTEM FOR SMALL MUNICIPALITIES................................................................................... 119

ADDENDUM 7............................................................................................................................................................................. 120

“WHAT SHOULD BE THE ROLE OF DBSA WITH REGARD TO MUNICIPAL BORROWING?”......................................... 120

4 December, 1998

6

ACKNOWLEDGEMENTS

Although errors of judgement or fact are the authors’ alone, our process of researching the state ofmunicipal borrowing in South Africa and coming to recommendations was greatly aided by thededicated work of a Reference Group convened by the Department of Finance on five separateoccasions. The group included both active participants in, and informed observers of, South Africanmunicipal markets. The members of this group included:

Junaid Ahmad, World BankRoy Bahl, Georgia State UniversityEnos Banda, Zader Financial ServicesJohn Douw, South Africa Local Government AssociationBarry Jackson, Development Bank of Southern AfricaBethuel Jwili, Institute of Municipal Finance OfficersMandla Maleka, Department of FinanceJackie Manche, Department of Constitutional DevelopmentGert Steenkamp, Department of FinanceDaan Wandrag, Imali CapitalRoland White, Department of FinanceLeon Winterboer, Banking Council of South AfricaAlan Yorke, Deloitte ToucheJohn Zohrab, Highway Technologies

Marie Du Plooy and Plaatjie Mahlobogoane were seconded to the team by the Department of Finance,and provided valuable help and insight throughout the process. Messrs. Bahl and Zohrab also workedas part of the team during our September visit, offering perceptive input and sound advice.

Roland White of the Department of Finance and Barry Jackson of the Development Bank of SouthernAfrica (DBSA) represented our principal clients, and put in endless hours briefing the team andreflecting with us on possible approaches. Our thanks to them, and to Maria Ramos, Director Generalof the Department of Finance; Dr Chippy Olver, Deputy Director General of the Department ofConstitutional Development; Dr Ian Goldin, Chief Executive of the DBSA; and Div Botha and MandlaGantsho, Executive Managers of the DBSA for their perspectives and advice. Thank you also to theDBSA for office space, facilities, and general support during the work.

Members of the Reference group, and staff of the Department of Finance, and the DBSA circulated tocolleagues some or all of our work over the course of several drafts. Though the number of reviewerswho passed on thoughtful comments is too large to name, the authors would like to express ourappreciation to many others who provided written and oral comments on various drafts.

Finally, we would like to express our appreciation to the officials and elected representatives,professional associations, bankers, insurers, asset managers, regulators, and other role-players who metwith us, informed us, and offered suggestions.

4 December, 1998

7

INTRODUCTION

The Issue

Historically, South Africa’s “white” municipalities could borrow to build infrastructure, though many ofthem relied on internal cash-flows to finance capital assets. Almost all consumers in these municipalitiespaid for the services they received, and these municipalities had no particular problems servicing thedebt they did incur. Local government was essentially an extension of provincial government, and it waswidely believed that provincial and/or national government would intervene to support any municipalitythat ran into financial trouble. Non-white municipal structures had little or no access to credit, builtmuch less infrastructure, and provided far fewer services.

Today, the fundamental rules have changed. Amalgamated municipalities have been created that includewell-served areas, unserved areas, and everything in between. They have been tasked with providinginfrastructure and services to all citizens. This will demand a lot of investment.1 National policy is thatthe bulk of these services should be provided by local government, and that private investment, bothequity and debt, should be attracted to help meet these needs. The private investment community hasbeen drawing back from the municipal sector, partly because of doubts about the financial viability ofthe new municipal structures, partly because some previous municipal debt has been repudiated by theborrowers, and partly because of uncertainty.

This report is intended as a step toward a future in which capital markets will channel private funds tomunicipal infrastructure projects. For this future to come to pass, municipalities must be creditworthy,which implies (1) that they will have adequate revenue to support borrowing, (2) that they will have themanagement and financial capacity and experience to make wise borrowing decisions, and (3) that therewill be financial and technical assistance for communities in need. This vision implies market allocationprinciples, i.e. that municipalities attract investment based on a project’s or a community’s financialviability, not a project’s social desirability. It further implies that the projects or the borrowers must besufficiently attractive to compete with other alternatives available to private investors.

There is no doubt that expectations for service delivery are immediate and extensive. The policiesarticulated in this report are not a recipe for immediate solutions to all of those pressing needs, but ratherfor sustainable and systemic solutions. The extent to which these systemic approaches must becombined with other mechanisms to achieve rapid and visible progress is a difficult policy issue,especially where interim steps may not be wholly consistent with the long term vision.

Attracting private capital back to the municipal sector, and increasing historic levels of private sectorlending for municipal infrastructure, is desirable:

1 Despite evident backlogs in infrastructure development, the municipal debt market in South Africa today is weak on both thedemand and supply sides. Many local authorities appear to have inferior budget and planning processes, leading to weak demandfor capital finance. Many investors have substantial reluctance to lend to municipalities, leading to weak supply. One estimateof the backlog is 80 billion rand, but much depends what is included as backlog.

4 December, 1998

8

• Because national and local governments have limited resources. Borrowing private money allowsmunicipalities (and the nation as a whole) to leverage the revenues they do have, building moreinfrastructure, and faster, than could be done on a pay-as-you-go basis.

• Because municipal borrowing from the private sector is consistent with the decentralised decisionmaking envisioned in the White Paper on Local Government and the Constitution. Localcommunities will set their own priorities, and apply to private capital markets for funding.

• Because market economics will cause funding to be allocated by private investors to thosemunicipalities and projects that are most creditworthy. This promotes healthy management andfinance practices at the local level and minimises the necessity for supervision by national orprovincial government.

The environment in which South African municipalities operate is changing. The focus of this report ison a set of policies and rules for municipal borrowing. While clearly pertinent to the effective functioningof a municipal credit system, issues involved in securing the financial viability of municipalities are largelybeyond the scope of this report. The recommended policies are intended to be robust enough toaccommodate various outcomes as regards the current process of municipal restructuring. Underlyingassumptions include:• A decentralised system of local government, generally as described in the White Paper on Local

Government and the Constitution• A municipal systems bill which is generally consistent with the recommendations of this report

As noted, the policy framework described in this report envisions a South Africa in which capitalmarkets channel private funds to municipal infrastructure projects, and in which eligible municipalborrowers are creditworthy. Being creditworthy implies that municipalities will have:• Stable, predictable, and sufficient revenues to support borrowing for capital investment, and• The management and financial capacity and experience to use debt responsibly.

While none of these critically important conditions wholly obtains today, the policies recommendedherein will help move South Africa in that direction and will be consistent with that future when it arrives.For the present, in addition to having sound policies as to municipal borrowing, it is important to providefinancial and technical assistance for communities in need to help them move toward that future.

The process

This report is the product of a five month long process undertaken in the second half of 1998 by theDepartment of Finance. The consulting team that prepared this report included both South African andinternational experts. The consulting team formulated the policy issues, made recommendations, andreviewed them at length with a Reference Group, during five separate working sessions over the fivemonth period. The Reference Group included both stakeholders in the South African municipal debtmarkets and technical experts (South African and international), and was chaired by Roland White ofthe Department of Finance. The members of the Reference Group are listed in the acknowledgementsabove. Individual and group interviews were conducted with dozens of additional parties, who are listedin Addendum 1.

4 December, 1998

9

EXECUTIVE SUMMARY

This executive summary is divided into two parts. The first part, entitled “Fundamentals” highlights themost important findings and recommendations of this report, focusing attention on those few key issuesthat are the most important and urgent. The second part summarises the recommendations of the reportin the order in which they are discussed, by topic.

Fundamentals:

The consulting team was engaged to suggest an appropriate policy, regulatory and institutionalframework for municipal borrowing in South Africa. The aim of this framework is to enable theexpansion of private sector investment in municipal debt. Separate efforts are underway to (1) facilitatepartnership arrangements with the private sector wherein private investors participate directly infinancing and/or operating infrastructure and (2) evaluate the role of concessional finance bydevelopment finance institutions.

As noted in the Introduction, the policies recommended in this report are intended to help achieve afuture in which capital markets channel private funds to municipal infrastructure projects – projects thatthe municipalities themselves have selected as representing the priorities of their citizens. This investmentby the private sector has the potential to build more infrastructure, and faster, than could be done withthe limited resources of national or local government alone. To promote private sector investment inSouth African municipal debt, the five most critical framework issues are:

1) The need to resolve outstanding amalgamation/boundaries questions.

Markets abhor uncertainty. There is a feeling among South African investors that municipalamalgamations to date have diluted security on formerly sound investments. Their concern about thepotential for further dilution is a key factor in investors’ current reluctance to lend to municipalities.

2) The need for clear creditor remedies.

Because municipalities seldom, if ever, defaulted under the old regime, the law on remedies in the eventof municipal default is underdeveloped. Creditors and borrowers alike will benefit from clarity about theconsequences of non-payment. There is a transition issue about existing creditors, and it isrecommended they be provided the same statutory remedies as new general creditors. In addition tostatutory specificity, the parties should have broad room to negotiate protections and remedies at thetime of borrowing.

3) The need to improve information systems.

Disclosure of municipal debts, defaults and security interests, along with improved accounting will enablemarket participants to distinguish between good and bad municipal credits. Currently, the municipalsector as a whole suffers from the failings of a relatively few. Until the permanent systems are in place

4 December, 1998

10

and operating well, important data contained in Project Viability can provide a good proxy, and shouldbe made public.

4) Predictable, adequate revenues for municipalities.

Without stable, predictable, and adequate municipal revenues, a framework for municipal borrowing isunnecessary. In fact, municipal borrowing may well be unwise. Municipalities which rely substantiallyon own-source revenues or which have the capacity to do so control their own destinies. Anymunicipalities that cannot generate sufficient own-source revenues will need intergovernmental transferswhich are predictable in amount and capable of being pledged as security, if they are to borrow.

5) Borrower capacity.

Municipalities must have the ability to budget, the ability to understand the impact of debt service and ofoperating and maintaining capital investments on their budgets, and the ability to plan strategically forcapital investment. They must be able to undertake analysis. Where these skills are lacking,municipalities are ill-equipped to engage the investor community. Technical assistance and training forelected representatives and appointed officials alike is needed.

Summary of recommendations:

This following paragraphs summarise the findings and recommendations of the report in the order inwhich they appear in the body of the report. For additional detail, please see the relevant section of thefull text.

Sec 1: Issuing DebtThis section deals with what municipal or quasi-municipal entities should be issuing debt, and what kindof debt should be issued (e.g. special vs. general obligation debt issues). Key recommendations are:

1) To avoid legal distinctions among municipalities as to their powers and duties with respect toborrowing

2) To permit both the superstructure and the substructure to borrow, if multi-level municipalgovernment continues

3) To clarify the legal status of, and remedies for, general obligation debt. Section 9 regardingremedies contains specific suggestions as to general creditor rights:a) The right to intercept funds due to a municipality from other spheres of governmentb) The right to trigger the imposition of an additional tax within the defaulting municipalityc) The right to trigger the appointment of a receiver or board to control expenditures and/or other

operations of the municipality.4) To allow the parties broad leeway to craft security provisions that meet their needs, providing that

minimum essential services are to be continued at all times. Minimum essential services must beclearly defined.

4 December, 1998

11

5) To make explicit the ability of municipalities to offer special security arrangements (e.g. ring-fencedor revenue bonds)

6) To expressly authorise tariff and rate covenants7) To expressly authorise contingent assignments of revenue (i.e. contractual revenue intercepts),

except such revenues as are expressly determined in advance to be necessary to provide minimumessential services.

Sec 2: Disclosure and Market RegulationThis section deals primarily with disclosure by issuers of debt, on the theory that better information willmake municipal capital markets operate more efficiently, and enable them to distinguish good creditsfrom bad. Key recommendations are:

1) To require borrowers:a) By law to disclose “all information material to an investment decision.”b) By regulation to disclose specified minimum informationc) By market standards to disclose sufficient information to comply with the legal “investment

decision” standard set forth in a) above.2) To require continuing disclosure at least annually, or on the occurrence of a “material event” that

could affect an investment decision.3) To make violation of the disclosure and registration requirements criminal violations, with liability

imposed on all participants. Elected representatives and appointed officials should have a duty ofreasonable inquiry in respect of compliance.

4) To apply disclosure standards and penalties equally to Bond Exchange and over the countertransactions.

5) To make Project Viability information available to the public.

Sec 3: Debt StructureSouth African municipalities have traditionally borrowed on very simple term structures. These do notnecessarily match revenue flows, and it is likely there will be a rapid introduction of new structures. Ingeneral, such new forms are useful, though there are risks for the unsophisticated. Keyrecommendations are:

1) To allow municipalities broad scope to structure their debt obligations to meet their needs.2) To avoid regulation that would prevent sophisticated municipalities from using products that allow

them to manage their assets and liabilities prudently.2) To adopt Department of Finance regulations banning the use of derivative instruments which create

leveraged, unhedged positions.

Sec 4: Restrictions on Municipal DebtThis section asks whether, and to what extent, the generally laissez-faire principles recommended in thisreport should be limited. Because of some instances of emerging and dangerous patterns of rolling overshort term debt, limitations are recommended. Key recommendations are:

4 December, 1998

12

1) To limit short term municipal borrowing to that amount required:a) To bridge operating cash shortfalls in anticipation of specific and realistic future income streams

to be realised within the fiscal year, andb) to bridge capital requirements in anticipation of specific and realistic grants to be received or

debt to be issued within the fiscal year.2) To require that short term debt be paid off annually, and remain paid off for some reasonable period

of time, with both borrower and lender responsibility for enforcement.3) To limit long term debt to capital investment in property, plant, and equipment.

Sec 5: Authorising DebtThis section deals with the question of what authorisations (internal to the municipality) and approvals(by other spheres of government) should be required for a municipality to incur debt. Keyrecommendations are:

1) To require that borrowing be authorised by municipal councils at the recommendation of the chiefexecutive.

2) To provide that short term borrowing may be authorised either by council resolution in eachindividual instance or by a general council resolution that the chief executive may borrow up tostated limits.

3) To require no provincial or national review or approval.

Sec 6: Concessional Finance and Other AssistanceThis section deals with the question of whether assistance should be provided beyond what the basicmarket approach can provide, and consider various alternatives. In South Africa, unlike manydeveloping countries, there is substantial aggregated domestic investment capital available tomunicipalities if the fundamental issues can be resolved. Key recommendations are:

1) To help municipalities access private capital primarily through technical assistance and training,which is geared to building skills and capacity. This technical assistance can be provided by anycombination of private and public entities, but could be funded by national government. Thisassistance is a form of subsidy, but one which helps borrowers and private lenders reach commonground.

2) To structure intergovernmental transfers with an eye to enabling leveraging of such transferswith private capital, and to provide that they will be pledgeable and interceptable, subject toappropriate penalties and administrative fees.

3) To approach guarantees and insurance, e.g. of specific maturities or against specific risks, withcaution. If government or the DBSA provides such guarantees, it should do so at cost, and perhapsthrough a self-financed mechanism.

4) To consider special intermediary institutions for municipalities with small borrowing needs andlimited market sophistication. In view of the private sector’s expressed interest in such pooling, it isrecommended that consideration of public institutions of this type be deferred until year after themore fundamental recommendations of this report have been implemented.

4 December, 1998

13

5) To avoid direct financial assistance by government (or DBSA), in the form of direct lendingpositions or interest rate subsidies.

Sec 7: Liquidity and Secondary MarketsLiquidity is important, especially for long term debt. Investors are more likely to lend for the long term ifthey know there is a secondary market for their holdings should they want to sell. Most or all of thefactors that will strengthen primary markets will also be important for increasing liquidity. Uncertainty isthe great enemy of markets. Key recommendations are:

1) To complete any further municipal amalgamation or demarcation as quickly as possible2) To improve disclosure, including

a) disclosure of Project Viability reportsb) disclosure of investor holdingsc) disclosure by borrowers as provided in Section 2

3) To consider trading structures in addition to BESA.

Sec 8: Monitoring, Intervention and RemediesRecent accounting reforms, once fully implemented, and debt-driven disclosure asrecommended herein, will limit the need for extensive information-gathering systems. Othermeasures are needed until then. The intervention process must be clear and simple. Keyrecommendations are:

1) To continue Project Viability monitoring until the disclosure systems recommended in Section 2and good accounting processes are established.

2) To develop the current provincial instruction process as the first level of intervention for troubledmunicipalities.

3) To provide a second level of intervention, in which the municipality or creditors have the right torequest the courts to appoint a receiver or financial control board where the provincial instructionprocess is unable to deal with the situation.a) A court-appointed receiver or financial control board should have broad powers.b) Minimum essential services should be protected

4) To provide general creditors with clear, simple remedies against a defaulting municipality, e.g.:a) The right to intercept funds due to the municipality from other spheres of governmentb) The right to trigger imposition of an additional tax within the defaulting municipality.c) The right to trigger the appointment of a receiver or financial control board, as described below.

5) To protect “minimum essential” municipal services, narrowly defined, e.g.:a) water needed for health and safetyb) sewage operationsc) refuse collection

6) To provide clear legislation governing municipal insolvency, includinga) creditors’ rights and prioritiesb) when and under what conditions debts and other municipal contractual obligations should be

dischargeable

4 December, 1998

14

Sec 9: Macroeconomic issuesGovernmental policy regarding municipal revenues will be a key determinant of municipalities’ ability toborrow. There is little interest at present in tax incentives. Key recommendations are:

1) To focus attention on municipalities’ need for stable, predictable, sufficient revenues. Without suchrevenues, municipal borrowing will not be viable.

2) To approach tax privileges with caution.a) If new privileges are to be pursued, a study should be undertaken to evaluate the leveraging

effects of any tax privilege against the revenue losses due to a tax exemption in the specific caseof South Africa.

b) Any tax privilege should be available equally to holders of municipal debt and to holders ofprivate debt incurred as part of a project to provide public infrastructure.

Sec 10: ImplementationTo implement the policy recommendations contained in this report, key recommendations are:

1) To draft and adopt a comprehensive Municipal Borrowing Bill. An outline for such legislation isincluded as Addendum 2.

2) To adopt regulations and take other implementation steps as discussed in Addendum 3.

In addition, supportive interventions could help stimulate an active private municipal capital market.These are also listed in Addendum 3, but have not been developed in further detail as a part of thisreport.

4 December, 1998

15

SECTION 1: ISSUING DEBT

1.1 Background and current situation

This report starts from the basic premise that enabling market-capable municipalities to access theprivate capital markets is the preferred solution to meeting their capital finance needs. This isthe approach taken by the White Paper on Local Government (see especially Section G of the WhitePaper). This approach:• Enables municipalities to leverage the funds available for municipal infrastructure by using private

capital to be repaid from municipal revenue streams.• Is consistent with, and enables, the decentralised decision-making envisioned in the White Paper

and the Constitution. Investment priorities are not distorted by financing incentives.• Helps ensure that funds are allocated by the market to those municipalities and projects that prove

themselves creditworthy. This promotes sound management practices at the local level.2

A private capital market can imply either a bank lending model or a municipal securities marketplace.3

As described in Section 7 of this report, active secondary markets can reduce the cost of municipalborrowing and make a greater supply of capital available.

Capital finance models that do not rely on private sector funding, , e.g. communal banks, involve morecentralism than the path South Africa has taken. They also make the market discipline South Africaseeks more remote, and offer less opportunity for leveraging than the approach embodied in this paper.The vision embodied in this report is that municipalities will use capital markets to convert streams ofcurrent revenue into capital for projects that are too large to finance on a pay-as-you-go basis. Variouskinds of market intermediaries can help improve access to markets for those that are potentially market-capable.

The definition of municipalities that are market-capable and those that are not is tricky. This week’sdefinition can be outmoded by next week’s events. A year ago, the consensus in South Africa was thatno more than 40 municipalities were creditworthy. Today, the Infrastructure Finance Corporation

2 Communities making their own investment decisions, and lenders making their own lending decisions, are likely to build moreappropriate facilities. For example, when sewer plants in the US were built with grants, they were often over-engineered in anattempt to get larger grants. Capital was free. When communities consider the cost of capital in terms of the future need torepay, they are more likely to make sound expenditure decisions.3 Much of the municipal borrowing in South Africa’s past has taken the form of municipal securities, but in fact most investorsheld their municipal bonds to maturity, resulting in a market which for many purposes resembles a bank lending model.

4 December, 1998

16

(INCA) is considering lending to up to 120 municipalities.4 To help understand the issues, it makessense to categorise South African municipalities into three groups:5

1) Municipalities which do not require external assistance to obtain commercial loans or access tothe bond market – perhaps 100 to 150 municipalities currently are viable credits.6

2) Municipalities which require external assistance to access private capital markets; and

3) Municipalities which are poor candidates for private capital investment due to fundamentalweaknesses likely to persist for the mid- to long-term.7

These groupings are flexible and their relative size is very much affected by government policy. Withmunicipal amalgamation likely to bring the total number of municipalities down from over 800 to 500 orfewer, the make-up of all three groups can be expected to change again in the coming months.8

1.2 Key concerns and policy issues

The key concerns to be addressed in this section are simply stated:

1) Who should issue debt?

a) Should there be any distinction among municipalities?b) Should there be issuers other than municipalities?

2) What kinds of debt should be issued?

4 INCA (the Infrastructure Finance Corporation) is a private corporation that raises funds through equity and debt and makesloans to creditworthy municipalities. It provides indirect access to capital markets. It pools the risk, is itself rated and rateable,and is currently the primary vehicle for municipal investments by much of South Africa’s investment community. As anintermediary, INCA is in a better position to monitor borrowers than is a passive investor. The repackaging of individual loansinto a larger bundle adds to the viability of smaller credits and provides economies of scale.5 The categories used here relate to general obligation debt of municipalities. A similar analysis could be applied to specialobligation debt or project financing. Some projects are market capable, some are potentially market capable, and some should notbe financed.6 See section 9 of this report for a discussion of municipal revenues and Addendum 4 for data on municipal debt. Currently, 14 to17 South African municipalities are rated by Fitch IBCA. Both Duff & Phelps (44 local authorities) and CA Ratings arecollecting statistics bearing on the financial condition of 190 local units for private sector subscribers. All of these may have bondmarket or bank loan potential.7 According to one observer, about 1/3 of the present municipalities (250 to 300, primarily small and rural) are not financiallyviable and approximately 100 are “close to bankrupt,” a number that roughly corresponds with the 107 instructions that havebeen sent out in conjunction with Project Viability.8 If the number of municipalities is dramatically reduced, e.g. to 250 – 500, as some predict, it may be that substantially all ofthese new municipalities are potentially capable of borrowing. If there is more focus on creating pledgeable intergovernmentalrevenue streams, rather than one-time intergovernmental grants, the number of municipalities that can borrow could besubstantially increased

4 December, 1998

17

1.3 Who should issue debt?

As noted above, there are three kinds of municipalities:

• The first group has market access

• The second group does not currently have market access, but municipalities in this group have (orcan generate) revenues sufficient to their responsibilities, and thus are financially capable ofborrowing private capital. This group can be further divided into:

§ those that are big enough, and managerially sophisticated enough, to eventually attract privateinterest without government intervention

§ those that are too small or managerially unsophisticated to ever attract private lending withoutassistance

• The third group does not have and cannot generate sufficient revenues to build the infrastructure andprovide the services they require. Municipalities in this group do not have market access, andbecause of basically unsound finances, they should not.

In a nutshell, municipalities in the first two groups should have the possibility of using credit to developinfrastructure, while those in the third group should not. The market will define, as well or better thanany imaginable government regulation, which municipality is in which category. Governmental policiesregarding intergovernmental finance, as well as policies governing assistance to small andunsophisticated municipalities, will affect how the market assesses creditworthiness.

Should there be legal or regulatory distinctions between municipalities?

While it is useful to think of three groups of municipalities for analytic purposes, should thoseclassifications, or any other, be codified into law or regulation? Developed municipal credit marketseffectively classify municipal borrowers, and reflect credit assessments in the prices charged forborrowing. Even in developed markets, regulatory classification is sometimes practised by central orstate governments. In the United States, for example, some state governments differentiate amongmunicipalities through various classification mechanisms, and those differentiations can include differentialborrowing authorities. Following that, the market further differentiates based on assessments ofcreditworthiness.

South Africa has a tradition of classifying municipalities,9 and there may be some sentiment for limitingsome municipalities in their ability to issue debt. It is true that if a nationally significant amount ofquestionable municipal debt is issued, it could create distrust about municipal debt generally, and couldcause macroeconomic problems. Some municipal credit systems use classifications not to prohibit debt 9 Under the former system, there were two types of classifications. The first was among cities, towns, and villages, and thisdistinction resulted in different powers and authorities. The second was a grading of municipalities according to a complex seriesof factors such as urban services, income, etc. These classifications have now been abandoned.

4 December, 1998

18

for some classes but to differentiate among municipalities in terms of maximum debt, or more typically,maximum debt in relation to some revenue source, such as the property tax. However, in systems beingput into place for the first time, or radically redesigned, the tendency is to overuse classifications. Thestrongest argument against regulatory classification is that upward mobility between categories should beencouraged. Classifying a municipality in a way that encourages it to seek assistance and avoidresponsible borrowing on its own is exactly the opposite of the effect the government intends to have.Artificially limiting market access runs counter to the basic policy goal of maximising private investment.In South Africa, there seems to be no need to limit borrowing by governmental action – the mainproblem is lack of access to credit, not over-borrowing. Now that it appears to be understood thatthere are no sovereign guarantees, the market place has proven that it will be selective. Municipalitiesthat can attract private capital should be encouraged to do so. Recommendations as to restrictions onthe amount of debt municipalities can issue and for what purposes are covered in Section 4. Thequestion of recommended assistance to those that cannot access private markets on their own is dealtwith in Section 6.

Should there be issuers of local public debt other than municipalities?

Municipal services need not always be provided by municipalities per se. And borrowing need notalways be by municipalities per se. Internationally, when one speaks of “municipal debt,” one oftenincludes debt of quasi-municipal entities.10 These entities can be created by agreement of existingmunicipalities, or by legislation .11 For example, South Africa may provide that if several municipalitiesjoin together and create an entity to provide transport services, water services, or other municipalservices, that entity would have the power to borrow against its own revenues. Or it may find itexpedient to allow provinces to establish special service entities to provide municipal services inparticular areas, and to authorise these entities to borrow in their own name. These combined entitieswith their own borrowing power are one kind of “municipal service partnership” –wholly public“MSPs.”12 These entities may be given taxing power, or they may rely exclusively on fees and charges.The policy argument against allowing such issuers is that they can lead to a proliferation andfragmentation of local government and can cause diffusion of local revenue sources. The argument infavour of such issuers is that they allow services to be delivered by an appropriately crafted entity, withtargeted taxes, fees, and charges. Such quasi-municipal entities can serve an important function in 10 Legislation authorising such entities may not now exist, and it is beyond the scope of this framework to considertheirauthorisation. Because they can be useful, and because their use may well be expanded in future, this possibility should beconsidered as the municipal debt framework is being established.11 Broadly, there are three kinds of potential special-purpose arrangements:a. Funds, accounting arrangements, or entities within a municipality, whose revenues and expenditures are separated from thegeneral fund. These derive their power from the municipalities.b. Entities created by agreement of more than one municipality (e.g. to provide fire protection efficiently across a broad area).Their revenues and expenditures could be independent of the organising municipalities, but their borrowing powers would derivefrom the municipalities. Typically, any capital needs would be met be separate borrowings of the organising municipalities.c. Quasi-municipal entities created by legislation . These entities might provide municipal services (e.g. water development,disease control, or transport services) where needs do not necessarily relate to municipal boundaries. They would have suchpowers as are described in authorising legislation.

12 The Systems Bill is envisioned to provide a framework for municipal service partnership arrangements.

4 December, 1998

19

developing South Africa’s infrastructure, and it is recommended that a policy for municipal borrowingbe arrived at in a manner that facilitates the issuance of debt by such entities.

Other public-public partnerships could be created which do not have independent borrowing or taxingpower, in which case each participating public entity would raise its own share of any capital requiredthrough its own borrowing power.

What about public-private partnerships?

Another kind of municipal service partnership is the “public-private partnership.” This term covers awide variety of organisational and funding models for involving private participants in providing municipalinfrastructure. Various models have different implications for the issuance of debt. Some commonapproaches include:

• The use of private investment, in the form of municipal bonds or bank loans, to build, operate,and maintain publicly owned and operated facilities is the most common method for private sectorinvolvement around the world, is the main topic of this report, and involves issuance of debt bymunicipalities or other public sector entities.

• A service contract with a private entity to perform some specific service with regard to of publicfacilities is another form of public-private partnership. In this case, the capital investment is stillprovided by, and ownership stays with, the public entity. Examples include maintenance of roads,water, or power systems, collection of solid waste, and meter reading.

• A management contract with a private entity to operate and maintain public facilities is a thirdform. Again, capital investment is usually provided by, and ownership stays with, the public entity.

• Leases and concessions also give operation and maintenance responsibility to a private firm.Leases typically require the private firm to take the commercial risk and maintain the system, whilethe government provides the capital investment and retains ownership. Concessions typically goone step further and require the private firm to invest the original construction cost or significantcapital rehabilitation or expansion costs. The fixed assets are the property of the government, andare returned at the end of the lease or concession period.

• Franchises give one or more private firms the right to use public facilities or rights for privateoperations, e.g. cable television in public rights of way, or private trams on public streets.Investment and ownership of the infrastructure is private, though the public authorities may have theright to acquire the facilities at the end of the franchise period.

• Wholly privately financed and operated facilities, where the capital investment is provided bythe private entity can remain in private ownership (e.g. BOO or Build-Own-Operate arrangements)or ultimately public (e.g. BOT or Build-Own-Transfer)

4 December, 1998

20

As can be seen in each case, any debt required for capital investment is typically either a municipality’sdebt or a private entity’s debt, and that is as it should be. Joint public-private debt can create asignificant temptation for an irresponsible party to leave its joint obligor holding the proverbial bag.

1.4 What kind of debt?

Another way of asking “what kind of debt?” is to ask what kinds of security may be given by municipalborrowers. Security (in the sense of collateral) and remedies are aspects of the same question. Debtmarkets require clarity about both – before they invest, creditors want to know what their remedies orsecurity will be in the event of default. Efficiency in the debt markets requires not only clarity, but alsoeffective remedies and adequate security. Unless the available remedies and security are deemedadequate, the markets may set risk premiums that make credit unaffordable to many municipalities. Thequestion of what remedies should be available by operation of law to municipal creditors is important.So are voluntary security interests created by the contracting parties. As part of the new regulatoryframework for municipal borrowing, it will be necessary to spell out what powers municipalities have topledge assets, to pledge revenue streams, and to pledge the exercise of their powers to set taxes, tariffsand other levies. It will also be necessary to spell out how such security can be effected in the event ofdefault.

Until recently, most South African municipalities have given only unsecured general promises to pay, andmost lenders have accepted whatever remedies they have at law. Increasingly, however, the parties areexploring ways of enhancing a creditor’s security by contract. Pledges of particular physical assets, ofreceivables, and of particular revenue streams have all been used. The conditions and criteria adoptedby the Department of Finance under the Local Government Transition Act permit the local council to“furnish any security” for a loan.13 This is a broad, open-ended authorisation, whose limits have not yetbeen explored.

1.4.1 General debt

General debt probably cannot remain unsecured, as was historically the case. The former system relied,consciously or not, on the assumption that local government had the backing of central government, andwould not be allowed to fail. The investor community no longer accepts that view of the world, andinvestors are likely to require improved security for new loans. Since the government has nowunambiguously indicated that it will not guarantee municipal debt, holders of outstanding debt findthemselves wondering what security and remedies, if any, stand behind existing general debt. Because ofthe absence of default until recently, there have been few tests of the ability of creditors to enforce theseloans. Investors considering new loans ask themselves the same question.

One of the key questions therefore is what security comes with a general promise to pay, and whatremedies are available to enforce that promise. Questions of special security are dealt with below. This

13 Government Gazette 18764, 27 March 1998, R. 412, Sec. 2.(4).

4 December, 1998

21

section focuses on general obligations, incurred on behalf of the municipality as a whole, to benefit thepopulation at large.

Aside from the good faith of the municipality and the prospect of provincial or national assistance ifthings got difficult, general obligations of cities were historically backed primarily by the apparent abilityof creditors to seize physical assets of the municipality. This is not a sound system. Municipalities withassets unrelated to their municipal service responsibilities, e.g. commercial enterprises, are often betteradvised to divest themselves of the assets so as not to divert scarce management capacity to managingpotentially private activities rather than use those assets as collateral on loans. Other municipal assetsthat are used directly or indirectly in the provision of essential municipal services or discharge of othermunicipal responsibilities should perhaps not be risked as collateral.14 Though often desired by lenders,physical collateral also tends to divert the municipality’s attention from making sure its general revenues,or specific revenues related to the asset being developed with credit, are sound enough to supportborrowing.

The term “full faith and credit” originated in the US and is generally understood to mean more than ageneral, unsecured promise. Debt not backed by specific revenue flows, or general obligation debt,should probably be backed by a “pledge”15 of all municipal general revenues as a source of debt servicepayment. The municipality could be specifically obligated to use any and all of its general resources tomeet debt service obligations. Stronger and more specific remedies for general creditors are likely tostrengthen investor confidence.

1.4.2 Special pledges, limited obligations, and “ring fence” financing

In addition to the question of what security and remedies are implicit in a general promise to pay, thereis a decision to be faced as to whether the market should dictate the kinds of special security that canbe given to secure municipal debt, or whether there are policy reasons to limit the kinds of security thatcan be given. Some kinds of security arrangements that could be allowed include the following:

• Physical or monetary assets could be pledged

• Revenues could be pledged: Revenues from tariffs, fares, or rentals Revenues from particular taxes or special levies Revenues from intergovernmental grants or transfers

• The power to set tax rates, utility tariffs, and other levies could be pledged

• The parties could contract for revenue intercepts

14 At least in the absence of statutory or contractual provisions requiring that such assets continue to be used for delivering publicservices in the event of foreclosure by a creditor.15 Essentially, this pledge could be created by operation of law for all general debt.

4 December, 1998

22

Investors considering new loans to South African municipalities today want greater security than theyhave accepted in the past. Pledging particular revenue streams can afford today’s investors thatsecurity. Nothing in current legislation seems to prohibit such pledges, and the above-cited Departmentof Finance regulation would seem to authorise them. However, holders of existing debt worry about theadverse effect that fencing out of lucrative revenue streams will have on their positions as generalcreditors.

If revenue pledges are authorised, it may be advisable to distinguish between the rules that apply to oldgeneral debt and those that apply to new general debt.16 It has been argued with some force thatlenders did not until recently understand or anticipate the possibility of pledging specific revenues tospecific debts. If revenue pledges are to be authorised, there is also the question of whether they shouldbe in addition to, or instead of, the general obligation to pay.

1.4.2.1 Pledges of assets and revenue streams Municipal treasurers have expressed concern about broad language authorising the furnishing of securityfor debts. They are concerned that valuable civic assets may be unduly encumbered, potentially leadingto the loss of such an asset in the event of default.

To state the concerns somewhat more broadly, there may well be municipal assets or revenue streamsthat are important enough, on policy grounds , that they should be protected from a debt service pledge,in the analogy of “spendthrift” provisions in a trust. Examples could include:

• the equitable share transfers now being provided to cover municipal services for the poorestsegments of the population

• physical facilities deemed essential to the public health and safety

As a practical matter, lenders are likely to be reluctant to take indispensable physical assets as collateralon a loan. They may be willing to take marketable assets such as vacant land, office buildings, parkinglots, or sports facilities, but they are not likely to foreclose on essential service facilities, such as water orwastewater treatment plants, city streets, or fire stations.17 Lenders are more likely to secure municipaldebts by pledges of actual or potential revenue streams that are sufficient to cover the debt service.

To address the concern expressed by the municipal treasurers, a prohibition on pledging facilitiesdeemed essential to the public health and safety could be included with little impact on the market’sdevelopment of other useful security devices. If this restriction were included, minimum essentialservices should be clearly defined by law.18 They might include such water, sewer, and refuse collection 16 Although if the general creditor remedies are made strong, and available to “old” creditors, this probably gives them sufficientprotection.17 Foreclosure on such essential assets could raise constitutional issues, especially in the absence of a statutory framework toprotect essential services.18 Another approach would be to require an affirmative and binding determination of non-essentiality by the municipal council orsome other entity as a condition of a valid pledge. But such a finding buried deep in a Council Resolution would not address theproblem if the service were, in fact, essential.

4 December, 1998

23

as is necessary for human health and safety. Although assets relating to these services could bepledged, the law should require minimum essential services to be continued at all times. This would notbe a barrier to borrowing -- lenders are not interested in repossessing pipes in the ground – they areinterested in the revenue stream. Section 8 below discusses remedies and inter alia envisions areceiver or financial control board operating essential services and collecting revenues for the benefit ofcreditors.

A pledge of revenues from public utilities19 is appropriate to secure financing related to those utilities, butraises concerns if used to secure unrelated financing. At root, the concern is simple economics: when amunicipality subsidises general expenditures at the expense of utility charges, it results in a misallocationof resources. In economic theory, and usually in practice, the subsidising service is under-allocated, andthe subsidised service is over-allocated.

1.4.2.2 Pledges of the power to set tax rates, utility tariffs and other levies

There may be some doubt whether South African law currently authorises a local council to pledge thatit will set and maintain utility tariffs, tax rates, or other charges at a level sufficient to service a debt.20

However, it is clear that a municipal council may, and most municipal councils do, bind successors byway of a valid contract with third parties. This is established law and practice. Without such a legislativepledge to set and maintain rates or tariffs at sufficient levels, a revenue pledge is much less meaningful.Where the tariffs, rates, or charges to be used to retire a revenue bond can be increased or decreasedat the discretion of the local authorities, a covenant to set and maintain those tariffs, rates, or charges atadequate levels is a usual and useful corollary. Experience in many emerging market countries that havedeveloped municipal credit systems is that the biggest single cause of debt service default and arrears isthe failure to implement tariff increases that were planned to be the source of revenues. Municipalities inSouth Africa would benefit from clear legal authority to covenant future tariff or tax increases to securedebt. Municipalities may choose or not to use this mechanism, and negotiate on other bases with themarket, but they should have the legal authority to covenant if at all possible.

1.4.2.3 Revenue intercepts Municipalities could assign to creditors their interest in specified revenue streams from the national orprovincial government or other sources. Such revenue intercepts can provide a creditor with access to astable, predictable, and sufficient revenue stream in the event of a municipal default. However, unlessthere is a disincentive, it may be tempting for some municipalities to not pay their debts and forcecreditors to rely on intercepts. This would not encourage conscious priority-setting at the local budgetlevel. This problem can be addressed by assessing a significant penalty or administrative fee to theborrower, so that the municipality is always better off collecting its own revenues and paying its ownbills.

19 in the aggregate, municipalities generate approximately 60% of their income from service charges on electricity, water,wastewater, and refuse removal)20 Such provisions are known as rate covenants.

4 December, 1998

24

1.4.2.4 Enterprise financing

A common use for limited and special pledges of revenues and assets is for a self-supporting“enterprise” activity that will generate its own means of repayment and will not rely on recourse togeneral revenues.21 This “ring-fencing” involves the pledge of system or project revenues to repayment.In accounting terms, this implies the creation of a special fund to receive the specified revenues, whichare expended to meet costs associated with the enterprise, including the payment of debt. This conceptfocuses credit concerns on the viability of a particular project or system, rather than on the viability ofthe municipality per se. Poor or unsound municipalities can have viable enterprises, and vice versa.

There are several advantages to this type of financing:

• it establishes a private-sector like relationship between the costs of the service provided and theprice that they carry. This helps promote more efficient operations. (The relationship need not be anabsolute and can be modified in practice, but it has the benefit of assembling the costs and makingthe subsidies transparent.)

• If the utility has been run at a surplus to subsidise other governmental functions, then the added “tax“burden becomes evident.

• To the extent that general revenues have been subsidising the utility operation, the use of dedicatedrevenue structures will free up general revenues to help pay for other costs.22

• Management and operation of the revenue-producing facilities tend to be more efficient and thefacilities better kept. This can be encouraged by contractual provisions to protect income and value,paired with creditors’ active interest in assets and their operation.

• Where there are legitimate reasons to use general revenues as well as specific revenues, it may bebetter to use municipal general revenues to reduce the amount of debt incurred in the first place, forexample by making a municipal “equity” investment in the asset up front, and borrowing to build oracquire the rest of the asset, pledging only revenues produced by the asset, or even a part of theasset’s operations, to meet debt service requirements. This practice is common in WesternEuropean countries for many municipal utility operations.

But, critics cite several disadvantages to this type of financing:

21 This is the traditional notion of the enterprise revenue bond. There is a legion of variations on the idea; many of which havebeen designed in the U.S. to circumvent restrictions on tax-supported debt. The basic enterprise approach allows and encouragesallocation of full costs of services to the beneficiaries. Economically, this is desirable because it leads to efficient allocation ofscarce resources.22 This is a common situation elsewhere in the world, but in South Africa the subsidy more commonly runs from the utility to thegeneral fund, rather than the other way.

4 December, 1998

25

• “Asset stripping” is a term that conveys the concern faced, e.g. by creditors of municipalities thathave relied on a utility to generate subsidies for the general fund, when those revenues are insteadpledged to a utility-specific purpose.23

• Limited obligations may be thought to hinder redistribution of infrastructure and service priorities bykeeping potentially redistributable revenues for the benefit of an already privileged area.

• Future financing options are restricted. Such borrowing is a matter of contract between themunicipal sponsor, acting on behalf of the enterprise, and the investors.24 Investors typically willrequire restrictions that reduce the options of borrowers in the future.

Transparency would argue that utility surpluses used for cross subsidisation should not be buried, butshould be reflected in a specific tax or surcharge that reflects the added costs of cross-subsidies.Without this transparency it is not possible to see if the utility is operating at an economic optimum ingetting the most delivered service per unit of input. Having reached that optimum, the redistributionbecomes a clear added cost for some and benefit for others.

Specific and limited revenue pledges or revenues have not yet been widely undertaken in South Africa,although municipalities appear to be authorised to do so, according to regulations laid out by theDepartment of Finance.25 Such pledges are being proposed in conjunction with public-privatepartnership (PPP) or municipal service partnership (MSP) projects now being prepared. In both theU.S. and Western Europe, specific revenue pledges or limited obligation financing is the fastest growingtype of municipal credit. It is recommended that South Africa not only permit but encourage dedicatedrevenue obligation support for enterprise-like and utility borrowing when supported by user charges.

1.4.2.5 Special District Financing

A variant of special fund and ring-fence financing is special district financing. A special district is createdto provide infrastructure and services to a subset of the general population and to a geographic area thatdemands special types or levels of service. These districts have been used to provide urban services(water, sewer, and roads) to areas that are developing rapidly or that have special needs (such asdowntown areas). They may used to provide infrastructure or services or both.26 They are commonpractice in Western Europe and the United States.

23 Most economists would applaud the elimination of the cross-subsidy on allocation grounds.24 Common constraints are requirements that the borrower not pledge the same assets to another lender except under statedconditions (additional bonds test), that the revenues provide certain coverage of the debt service (rate covenant) and that revenuesbe retained for use on the facility and to the benefit of bond holder (closed loop). Negotiation of these restrictions and theassociated tests is an integral part of the borrowing transaction.25 Government Gazette 18764, 27 March 1998, R. 412, Sec. 2.(4)..26 A district may be “dependent” and overseen by the governing body of the municipality, “independent’’ with an autonomouselected or appointed board. In many areas storeowners or homeowners form associations to manage the district and levy charges.The key is the ability to levy taxes and charges and attach properties that don’t pay.

4 December, 1998

26

There is a myriad of examples, but the common theme is that if there are special benefits that can beascribed to a particular area, then the special district device provides a mechanism for recovering thecosts involved. Often special districts provide a way of transcending political boundaries or unitingseveral underlying jurisdictions into a single financing unit to provide a regional service.

Of particular note is the use of the special district where new development or extensive redevelopmentis occurring and where the property (as opposed to individuals) is the long-term beneficiary of theimprovements. For example, new public utilities and transportation, storm drainage, parks and similarimprovements increase the attractiveness of a geographic area and enhance property values. As thearea is improved by these expenditures, the amenities and access provided to it increase the propertyvalues as well as other indices of economic activity and worth. If the costs of the capital improvementsare borne by the public sector, then the public sector should have a way of “capturing” its investment.Such a value capture mechanism is possible with a special district that gears its taxes or charges to payfor capital improvements, the benefits of which were received by the improved properties.27

It is important to bear in mind that a special tax relies on good measurement and an efficient tax system,which itself is a furtherance of efficient government. In the United States, some special taxing distrustsare administered by private organisations that undertake the calculations and the billing acting as agentsfor and under contract to the host governments. The entire revenue raising mechanism is specified in theloan or bond agreement. This is a means of improving the administration of taxes, since the integrity andefficiency of the system becomes, in effect, a matter of contract with bondholders. In Western Europethe special district or special authority covering all or parts of more than one political jurisdiction hasfacilitated the subsequent privatisation of services, such as water utilities in the United Kingdom andFrance.

1.5 Recommendations: Key recommendations are:

1) To avoid legal distinctions among municipalities as to their powers and duties with respect toborrowing

2) To permit both the superstructure and substructure to borrow, if multi-level municipal governmentcontinues.

3) To clarify the legal status of, and remedies for, general obligation debt. The process forenforcement of the obligation should be clear. Section 9 regarding remedies contains specificsuggestions as to general creditor rights:a) The right to intercept funds due to municipality from other spheres of governmentb) The right to trigger imposition of an additional tax within the defaulting municipality

27 In its most common form in the U.S. and a few other places in Europe, the tax district uses property taxes (expressed as apercentage of taxable property value) or assessments (fixed dollar levies). But the district can use other bases to charge for thebenefit or service, including square or front footage or meters, number of vehicular trips (for roads), impervious surface (fordrainage), lumens of light (for lighting), residential bedrooms (for educational facilities), square footage of space (for parking), andso forth. The key is that there be a logical “nexus” between the improvement and the form of charge that is used.

4 December, 1998

27

c) The right to trigger the appointment of a receiver or board to control expenditures and/oroperations of the municipality.

4) To allow the parties broad leeway to craft security provisions that meet their needs, providing thatminimum essential services are to be continued at all times. Minimum essential services must beclearly defined.

5) To make explicit the ability of municipalities to offer special security arrangements (e.g. ring-fencedor revenue bonds)

6) To expressly authorise tariff and rate covenants7) To expressly authorise contingent assignments of revenue (i.e. contractual revenue intercepts),

except such revenues as are expressly determined in advance to be necessary to provide minimumessential services.28

28 An alternative would be a “prudent person” restriction on municipal pledges. If this approach is used, care must taken thatremedies not include an ab initio invalidation of the pledge. This would leave lenders in a position of contingent risk.

4 December, 1998

28

SECTION 2: DISCLOSURE AND MARKET REGULATION

2.1 Background and Current Situation

The operation of the marketplace in terms of exchange operations, off-market trading, clearing,settlement and supporting operations can affect the attractiveness of debt security sales versus otherforms of borrowing. Slow and imprecise functioning of marketplace mechanisms can drive uptransaction costs and add to the risks for investors. South Africa has a relatively sophisticated financialsector with a modern exchange and security clearing operation. Although the current securitiesregulation model was structured initially in contemplation of trading in exchange-based listed securities,there is a growing over the counter market.29

The issue is whether there are fundamental impediments to local government debt transactions. There donot appear to be, though there is substantial room for increased transparency and disclosure. Atpresent, the mechanical operation of the formal market for local government securities is largely a mootpoint. It simply does not operate. According to the Bond Exchange of South Africa (BESA), onlyR209 million in local authority bonds were traded on the secondary market in 1997, except for INCA,which had R13,228 million. This represents less than 0.5% of all bond secondary market trades.Furthermore, only R33 million were trades involving two local government issuers that are listed on theexchange. Present regulations require that any bond traded must be cleared through the BESA clearingmechanism but need not be listed on the Exchange. Listed bonds may be traded off the exchange butneed to be reported and cleared through the exchange.

The bond market is “dematerialised” which means that the system is full-book structure and thatevidence of ownership of individual shares (unless special arrangements are made) and of trades aresimply noted on the registrar’s books. The underlying bond certificate itself (if any) is immobilised andphysically kept in a central repository. Settlement is done electronically through agents with delivery offunds against bonds and is on an international T+3 basis. There is a fidelity fund (guarantee fund) ofR110 million to protect against defaults by members. Activity on the bond exchange is dominated byRSA and the large parastatal bond issues, although there are about 20 corporations and 4 municipalitieslisted. The only active trading municipal security is INCA.

The Financial Services Board (FSB) regulates the securities trading system. Although no review hasbeen done of the mechanical operations of the exchange, nothing has been reported that would blockthe functioning of the bond market. The marketplace is in transition, with a streamlining and potentialconsolidation of operations.30

29 The first South African stock exchange and securities acts were passed in 1947 following the approach of the U.S.’s 1933 and1934 Acts. The Act was modernized in 1989 (Financial Control Act). to include the derivatives and options markets. At thattime a Bond Market Association and a Bond Exchange were created. Recently, regulators noted that 60% of the bonds are tradedOTC away from the market and only 40% on the floor of the Exchange and that the latter share is dropping. About 70 to 80 %of all trading is in government bonds. 30 There are currently under consideration proposals to merge the South African bond, stock and futures exchanges into a singleoperation.

4 December, 1998

29

BESA has had little recent experience with local government debt, the market for which has beenvirtually moribund. It is amending its listing rules and disclosure document rules to make them moresuitable for municipal securities. Proposed disclosure requirements appear somewhat lax and anadaptation of the rules for corporate securities with exceptions thought applicable to municipalities.There appears to be no requirement for periodic financial disclosures to maintain a listing. The listing feefor both the initial listing and annually thereafter is 0.001% of nominal value, with a minimum of R1000and a maximum of R10, 000. Fees are subject to the VAT tax. Trades cost R9.5 per R 10000 in facevalue.

Greatly improved disclosure of local government finances and operations will be critical to the re-ignitionof the South African (SA) municipal securities market. From the standpoint of creditworthiness andsecurity pledged, the market for local government securities that is to be re-established in South Africawill be much more information-driven and dependent than previously was the case. In prior days, therewas at least an implicit guarantee by the national government, and local government finances wereperceived as (and probably were) much stronger. Moving to this new regime will require establishingnot only market-oriented incentives for both borrowers and lenders to use the market but also clearguidelines as to how information channels will be established, managed and used.

The new governmental structure that is evolving in South Africa places severe burdens on borrower andlender alike. It will require large amounts of capital improvements, and will be dependent on localrevenue sources. It will also entail re-distributive transfers at the local government level. Localgovernance in South Africa has been highly reliant for revenues on commercial public utility operations(water and electric). The ability of these operations to generate resources to support themselves, andgenerate any surplus for general revenue purposes, depends heavily on efficient technical and managerialoperations. Unfortunately, non-payment of user fees and local taxes markedly decrease current cashreceipts and make questionable accounts receivable. These problems are well known in the aggregate,but the implications for particular local governments are just coming into focus. Recent defaults ongovernmental loans and financial scrapes in that sphere are a major disincentive to private investorsconsidering long-term loans to the sector.

General sectoral problems have led to uncertainty in fixed-income securities markets. Knowledge of theparticulars of each municipality or project will be needed to calculate risk versus reward and makeindividual investment decisions in a market setting. With the objective of helping local governments inaccessing private capital markets, setting the foundations for that process is a key purpose of this report.

Aside from the immediate needs of the bond market, effective democratic government will depend onregular reporting of the operating results and financial condition of governments in a consistent format.This is the theory behind the few existing (and a number of potential) reporting and auditing requirementsnow found in South Africa regarding local government finances. This section focuses on disclosure ofinformation that is intended for the securities market. However, as discussed below, because of theeconomic interests involved and the importance of thoroughness and timeliness, disclosure can havesubstantial beneficial impacts throughout local government operations.

4 December, 1998

30

2.2 Discussion and Analysis

Disclosure of information and other securities market regulatory practices can be broken down intoseveral component parts entailing who is responsible for disclosure, what information is provided, howfrequently it is provided, and the means of information distribution, including a central point for collectingand distributing information for the marketplace. Closely related topics have to do with generalquestion of market regulation and various operational issues such as the registration of securities, themechanics of listing securities on exchanges, and the oversight of underwriting and trading. Each ofthese subjects is a world unto itself; but the objective here is limited to recommending what areas ofcurrent market practice can reinforce or impede the revival and reformation of a market of localgovernment securities.

2.2.1 Disclosure By Whom

Normally, disclosures to the market are originated by the borrowers themselves, in this casemunicipalities. However, the borrowers can be assisted (or perhaps even largely superseded) incollecting, packaging and transmitting information by underwriters and brokers, and by the centralgovernment or provincial government authorities, depending on the nature of the information to befurnished. A fundamental issue is who is responsible for the payment of debt service. The party that isobligated to make timely and full payment is the appropriate party to make disclosures.31 A closelyrelated concept is that the party in control of decisions as to honouring obligations and/or that hascontrol over the information is the one responsible for information.32 However, a central regulatoryinstitution, whether governmental or established by the securities market, is important in making thisdisclosure information widely available.

Disclosure can be required by (1) central governmental fiat, (2) securities market regulation, (3) as a by-product of the operation of the market through contract between issuers and lenders (usually as part ofa borrowing contract that contains covenants as to issuer behaviour) or (4) market practice andconvention. This latter system is frequently called a voluntary or market-driven approach. Theenforcement of market-dictated disclosure is through investor expectations and habitual procedure andthe desire of issuers, in order to gain access and the most favourable terms, to provide lenders whatthey want. Most important, however, may be effective securities regulation so that borrowers,appointed officials, and elected representatives comply under fear of anti-fraud provisions of securitieslaw or similar legal obligations related to fair-dealing and fraud.33 As a rule, voluntary and market

31 In the U.S. system of municipal bond disclosure, the distinction is between the borrower (the issuer of the securities) and theobligated party (that party(is) that has a material legal responsibility to pay debt service on the securities in question). It is notuncommon to have more than one “obligated party” in a transaction.32 For example, a guaranty by a third party (such as the national government) has been seen as a reason to require less disclosureon the part of the actual borrower. That concept, however, has been rejected in U.S. practice where a guarantee or insurance) doesmot obviate the need the need for full disclosure by the borrower. in South Africa (as elsewhere) the custom evidently has beento lessen or relax requirements where the national government is the guarantor. As noted above, the change is the fiscal climatewill supersede that practice.33 South Africa’s existing securities laws and anti-fraud provisions are largely geared to transactions on a financial exchange, andwill need to be bolstered to cover over the counter markets.

4 December, 1998

31

dictated disclosure systems work best if there is an underpinning of securities laws and regulation thatprovides a baseline of disclosure and common procedures and penalties for those that are reckless oruntruthful.

2.2.2 Disclosure To Whom

Ultimately, the primary target of disclosure is the investor. The purpose of disclosure is to enableinvestors to make informed investment decisions. There is a second important purpose of disclosure,which is to enable the market to effectively allocate investment resources through pricing.34

Generally, most formal disclosure requirements are thought of as taking place where the issuer disclosesto the market through some readily accessible mechanism. Operationally, there may be importantrecipients of the information that “translate” or re-communicate the disclosure to actual investors. Themost important of these are the financial information services and rating agencies. For example in SouthAfrica, as elsewhere, rating agencies rate securities on their credit quality (usually defined as the risk ofdefault on the obligations.)

Three other distinctions regarding to whom disclosures need to be made are in order. First, disclosurein the securities market is hinged on investor protection, but not all investors need or want the samelevels of protection. Protection can be differentiated and conditioned by “suitability rules” which requirethat in dealing with the public, security salespersons must observe professional rules or guidelines thatrequire them to recommend only securities that are suitable given the circumstances and sophistication ofthe customer.

The second distinction is a “blanket” one where either the regulator or the issuer itself (and theunderwriter offering the securities) will make an a priori declaration that a security is limited to a certainclass of investors, typically those that are “sophisticated.” Definitions of sophistication differ, but thegeneral rule is that financial institutions and high net-worth entities and individuals qualify assophisticated.

A third distinction sometimes made has to do not with investor protection but with the countervailinginterest in diminishing burdens on certain classes of borrowers to permit easier access to the securitiesmarket. These lessened standards typically run to smaller issuers or those classes that are believed torepresent less risk. Traditionally, government securities have belonged to this lower disclosurerequirement, although that tradition has been eroded in the United States and elsewhere. Given therapid change in local government structure in South Africa and the financial stresses of manymunicipalities, it is difficult to support an argument for reduced disclosure standards for localgovernment. On the contrary, the widespread demands for more information by the investmentcommunity, the shaky performance of many municipalities, and recent actions by the central bank in 34 If the primary purpose of disclosure is avoid fraud and investor loss, then the emphasis can be on screening out high-risksecurities that regulators feel might cause loss to the investors. That substitutes a bureaucratic decision of that of the marketplace. The market approach to resource allocation requires that the market have sufficient information to decide the appropriaterate of return.

4 December, 1998

32

increasing the capital adequacy requirements for local government debt all argue for high standards ofdisclosure for instruments that would be sold in the public marketplace.

2.2.3 Disclosure When (Frequency of disclosure)

Disclosure relating to a securities offering is usually thought of as taking place in three phases:35

The first phase has to do with the “pre-sale” period where the there is notice that bonds are to beoffered for sale and documents are filed with the relevant regulatory body(is) or SROs (such as a stockexchange) and, upon approval (if required) solicitations are sent to underwriters or final investorsannouncing the characteristics and availability of bonds and the timing and terms of the transaction.

The second phase is where the bonds are actually sold and the terms of the transaction are struck. Atthis point, either the securities are issued and sent to investors (final buyers) or books set up (in a book-entry system such as is used in South Africa) and confirmations sent to final investors along withdocuments showing final terms.36

The third phase comes after the original offering and represents continuing disclosure to the bondholdersand/or “after-market” so long as the debt is outstanding. Such continuing disclosure may be on aperiodic and recurring basis (sending out or otherwise making available current financial reports andother operating data) and/or episodic disclosure when events occur that may have a material impact onthe borrowers’ creditworthiness. Disclosure reforms in the United States in the last two decades havefocused especially on improving the continuing disclosure of information as holders of existing municipaldebt found the value of their investments potentially threatened by subsequent debt issues by the sameentity.

Frequency of continuing disclosure is usually annually after financial reports are available. In the cases ofprojects, however, where bonds support construction of new income-earning facilities or properties,disclosure may be more frequent.

2.2.4 The Content of Disclosure:

Reflecting differing philosophies as to market regulation and the emphasis on disclosure, the scope anddetail of disclosure can vary markedly among securities markets and within markets, among types andsizes of issuers and debt issues. It bears repeating that in the prior South African market, there was littlethought given to the content of disclosure because of the simplicity of the security pledged and the 35 This description relates to a formal offering of securities and the details are dependent on the regulatory scheme in a particularcountry and for a particular class of issuer. Direct placement, such as a bank loan, typically is a more informal and flexibleprocess and frequently is not open to public view, particularly where bank secrecy obtains. As a general principle, bank secrecyand direct placements work against achieving disclosure and the operation of securities markets that promote competition andtransparency.36 In the US market, the process is standardised by “process” requirements in the municipal bond market and “process andcontent’ requirements in the private corporate securities market. Interestingly, the sovereign, the U.S. government has no suchdisclosure requirements.

4 December, 1998

33

implicit sovereign guaranty. New instrument forms and security pledges would require a mapping out ofnew disclosure content to reflect the new risk categories.37

The content can either be determined by a regulator’s detailed listing of contents of documents and asequence of presentation of various times or by a “flexible” standard that depends in its application towhat the issuer and its advisors and agents perceive as information that investors should find useful inreaching investment decisions (or a combination of the two, which is usually the case.)

Fundamental to continuing disclosure is the timely production of good financial statements on a timelybasis. Certain requirements are already on the books. According to existing law, South African localgovernment units have two central financial reporting requirements.

1) As required by Local Government Transition Act (LGTA) at 10(g)4(c) local governments are tosubmit their annual budgets to the Department of Finance Ministry for monitoring of the overallgrowth in local expenditure .

2) As required by LGTA at 10(g)2(e)(i), local governments are to submit end of the year financialstatements to the Auditor-General in a form prescribed by him within 3 months of the fiscal year.The earliest audited financial statements are available is, on average, currently nine months afteryear-end.

The accounting standards for local governments in South Africa are currently under review andrecommendations have been made to improve consistency and transparency. Were compliance withthese standards to be required in order for any borrower to undertake a loan (with the exception incertain de minimis borrowers), the adoption of the standards could be accelerated.

In addition to financial statements, and depending on the nature and characteristics of the project beingfinanced and the nature of the security pledged, additional information may be required for appropriatedisclosure.38 For example, a limited obligation or enterprise security (“ring-fenced”) that looked only tocash available after operating expenditures to repay debt would need disclosures on the operatingcharacteristics in order to judge how efficiently it was being operated and if there were any concernsabout future supplies, labour relations, lawsuits and the like. The list of potential items worthy ofdisclosure can be long and will be dictated by the nature of the operation and the security pledged.39

37 the expanded scope of the Project Viability data collection as well as the new department of Finance regulations regardingnotification of new loans are harbingers of new information needs.38 the word “appropriate” is used because once beyond a simple promise to pay (and likely even in that case), the informationneeded to assess risk will be situational to a particular municipality. For example, a municipality that relies heavily on utilityoperations will find its ability to pay debt heavily influenced by the financial health of those utilities. If costs of raw materials orlabour are rising rapidly and/or users are not paying their bills, then cash will be short. This may lead to inability to make timelydebt service payments.39 The listing of items to consider are found in various trade and professional publications. A good starting point for generic itemsis continued in the GFOA Disclosure Guidelines for State and Local Securities (Chicago, 1991).

4 December, 1998

34

Most investors and observers feel that information regarding the financial performance and health oflocal governments is needed in order to calculate and manage risk. Investors also need to know theextent of cross-subsidisation of activities by local governments (for example when the utility produces asurplus that is routinely used for general expenditures unrelated to the utility activity.)

Among investors, elected representatives, and government officials interviewed, there were deepconcerns about the managerial and technical efficiency of local government and what would be done toimprove operations. Management and labour relations will be key disclosure issues in any disclosuresystem that is implemented to meet the needs of the securities market.40

The involvement of third parties can change the nature of disclosure requirements and how they aresatisfied. For example, disclosure requirements for underlying borrowers may be lessened or modifiedwhen there is a guarantee by third party, although this is a debatable subject.41

2.2.5 A Central Repository for Disclosure

In order to ease burdens and improve distribution, disclosure by borrowers can be centrally collectedand available from a central depository. A continuing disclosure system can be based on informationroutinely received by that source on a recurring and/or an event-driven basis. In the U.S. that role isplayed in the municipal market by a limited number of officially sanctioned (but privately owned)repositories as well as a central repository operated by an SRO (the Municipal Securities RulemakingBoard, MSRB) Generally speaking, issuers after the initial offering can rely on filings to the officialrepositories in the fulfilment of their after-market disclosure responsibilities. By most reports, thatsystem, which relies on private vendors for distribution of information, appears to be working fine.42

A central repository for disclosure information is a natural extension of an expanded NationalIntervention Programme (Project Viability).43 In view of the relatively small size of the South African

40 There is ideally a dynamic quality to disclosure. What investors consider important will vary through time with changingcircumstances. At present in South Africa a pressing concern is the quality and depth of management and the ability to deal withpublic employee unions. Several interviewees mentioned that the biggest problem is not lack of resources but the efficiency andeffectiveness with which they were deployed.41 For many years in the U.S. market, the position was taken that where obligations were guaranteed or insured by another party,issuers were not subject to the same level of disclosure requirements as those not so enhanced. Practice and regulation has swungaround to the recognition that such enhancements are separate and that both the guarantor and the guaranteed entities havedisclosure requirements when bonds are offered to the public.42 The system of repositories used in the U.S. is a product of the unique circumstances in the United States that led to a numberof private repositories and one public repository run by the SRO (the MSRB). For most countries with more modest securitiesmarkets, the multi-unit system used in the U.S. would not be a good model. The concept of a repository that is run for thebenefit of the market by an SRO, is a good prototype, as are the various procedures that have been laid formulated.43 There is some concern among government officials that disclosures of Project Viability data for specific jurisdictions wouldinhibit or disrupt existing banking relationships, and would cause a negative impact on the reputation of municipalities. While theresults of any one period could be disruptive, and care should be taken in how the results are disclosed, the lack of disclosure alsohas negative impacts, visiting the perception of trouble on all municipalities, and limiting the possibility for potential investors todiscern differences. Banks can, of course, request data from their potential borrowers, but absent the Project Viability data, theinvestment community lacks points of reference.

4 December, 1998

35

investment community and the limited number of likely bond issues, the creation of a central repository,along with guidance on disclosure content would appear easily achievable.

2.2.6 Registration and Enforcement of Disclosure:

As noted, there are a variety of schemes that can be adopted regarding the degree of regulation of themarket and the enforcement of disclosure processes or content. Registration of debt issues, as acondition for listing bonds on an exchange (or for trading a security off the market) can be anenforcement mechanism.44 Registration can be simply a listing of the security, much the same as thelocal government loan reporting requirement recently mandated by the Department of Finance. Or, itcan carry detailed information requirements and an assessment of risk and quality before sale andtrading are permitted. As of now, South Africa has no such requirements, although the Bond Exchangeis developing standards for listing by governmental entities.

The ability of regulators to effectuate high standards of disclosure depends on the vigour with whichrequirements are enforced. In the case of document content requirements, enforcement depends on thequality and quantity or resources that are committed to that activity. If heavy requirements for disclosure(and/or listing requirements) rely on review (and/or administrative approvals) by individuals that are notprofessionally qualified, the results will likely be of little use.45

Disclosure requirements can be stringent but rely upon private advisory and information services toexamine disclosures and make informed judgements. In this case, not every investor will read everydisclosure document, but a limited number of qualified professionals will examine them and formopinions for which they are paid by investors. The leading example of this role is the rating agencies thatpost ratings (and in South Africa, compile statistical books) for issuers and issues and continue to keepthem under surveillance while the debt is outstanding.46

For effective market use of disclosures of material information, the widespread use of confidentiality inreporting financial matters will need to be abandoned (or restricted to emergency situations, in whichbondholders’ rights and interests are protected).47 It can be a requirement of issuance and orlisting/trading that the issuer enter into an agreement with bondholders to provide disclosure.

The national or provincial government could reinforce and/or effectuate disclosure requirements (andother reporting requirements) by withholding transfer payments in the absence of submissions of

44 Registration could be required of the borrower, the lender, or both.45 the lack of resources and qualified personnel at the provincial level, for example, is seen as major impediment to that spherehaving an active oversight role in local government finances.46 But, nothing is free. Regulators may not being doing the substantive reviews and forming opinions on the adequacy ofdisclosure themselves. But, they need to institute meaningful safeguards that those who do (such as financial advisors, ratingagencies and other information providers) are professionally qualified and are operating in an ethical fashion and not manipulatingthe market and are free of conflicts.47 For example, most bond default provisions allow a cure period to the issuer to correct the situation before the transgression isannounced and remedies become effective. On an equal level of concern is that the bondholders security will not be altered ordiminished without notice, agreement to modifications, and legal rights to recourse to obtain relief. .

4 December, 1998

36

information. Units that are too small and too poor to abide by disclosure regulations are unlikely to beaffected since they are unlikely to be securities issuers. A de minimis issue or outstanding debt sizerequirement could be used. Some caution is in order in withholding transfer payments for this purpose,however. There is a tendency to use withholding transfers for all kinds of purposes, not the least ofwhich is ensuring debt repayment. Each instance of using that leverage may make sense in isolation, butif the mechanism is used for many different purposes, the larger municipal finance and intergovernmentaltransfer system can become bogged down. This mechanism should be considered only in light of all theother possible uses of the mechanism.

The remodelled municipal securities market will not operate efficiently (and may not operate at all)without dependable and timely information concerning risk factors. Disclosure places a necessaryburden on would-be borrowers. Complete and timely disclosure is a necessary, if not a sufficient,condition for the operation of a municipal securities market.2.3 Recommendations Regarding Disclosure:

Key recommendations are:

1) To require borrowers::a) By law to disclose “all information material to an investment decision.”b) By regulation to disclose specified minimum informationc) By market standards 48 to disclose sufficient information to comply with the legal “investment

decision” standard set forth in a) above.2) To require continuing disclosure at least annually, or on the occurrence of a “material event” that

could affect an investment decision.3) To make violation of the disclosure and registration requirements criminal violations, with liability

imposed on all participants. Elected representatives and appointed officials should have a duty ofreasonable inquiry in respect of compliance.49

4) To apply disclosure standards and penalties equally to Bond Exchange and over the countertransactions.50

5) To make Project Viability information available to the public.

1) In addition, it would be good practice for an authority such as the Financial Services Board tomaintain a simple registry of municipal debt, security interests, and defaults. Although there is limitedsupport for such a registry among the participants and potential participants in the municipal debtmarket, this approach should be given continuing consideration as a municipal debt market re-emerges. The outline for a draft Municipal Borrowing Bill (Addendum 2) includes authority for theFSB to establish a registry if the need arises in the future.

48 Which will presumably evolve over time to reflect circumstances and management capacity49 E.g. disclosure which participants “knew or should have known” to be false or misleading could be the basis for liability.50 The disclosure, but not the registration, requirements could be waived by a sophisticated investor who holds securities to termor sells them only to other sophisticated investors who waive.

4 December, 1998

37

SECTION 3 DEBT STRUCTURE

3.1 Background and current situation

Debt structure is a matter that is largely determined by the market. In the past, municipalities were ableto borrow at rates and on terms that suited the structure of their assets, in a market where there wassome demand for municipal securities. As the potential supply of municipal securities began to outstripdemand, mainly as a result of higher perceived credit risk, lenders restricted municipal borrowers to theshort end of the market, and were able to set terms that are more favourable to the lenders.51

3.2 Key concerns and policy issues

The major concern is not to create regulation which interferes with the flexibility of both lenders andborrowers to structure debt in a manner which best suits both parties, or which favours one party overanother. What follows is a largely descriptive note on debt structure, which may serve to illustrate therange of instruments available to suit the profiles of both issuers and investors.

3.3 Term

The term to maturity of a debt instrument is ideally matched to the economic life of the asset it isfinancing. This is known as a hedged position. The amortising liability on one side of the balance sheetis matched by the depreciating asset on the other. Thus, infrastructure assets, such as water systems,roads, or municipal buildings, which typically have lives of 15 to 30 years, should theoretically befinanced with long term bonds of similar duration.52

From time to time, the shape of the yield curve or term structure may tempt issuers to prefer the useof shorter-term debt. Typically, the yield curve is upward sloping, i.e. the longer the term of the debt,the higher the interest payable. This is because investors want extra compensation for the lack ofliquidity of long term lending. However, if expectations are that the general level of interest rates isunsustainably high, and that rates are expected to fall, the yield curve may very well be inverted, withshort-term rates higher than long term rates. In such a market, borrowers would want to borrow short,so as not to lock themselves in to high rates for the long term.

When one considers credit spreads, term structure becomes more complicated. The price of municipaldebt includes a risk premium which reflects the credit and liquidity risk of the issue. For example, ifseven-year South African government bonds are trading at 15.80%, Durban bonds might trade between17.20% and 18.30% (known as trading 140 to 250 basis points over the curve, or as having a spreadof 140 to 250 basis points - a basis point being one percent of one percent). Typically, credit spreadsalso increase with the maturity of the bond. The spread for one-year paper could be as little as 75 basispoints (bps). 51 As previously noted, both supply and demand for municipal debt are weak.52 Matching asset life to debt term is also sound public policy because facilities can be paid for by those who use them.

4 December, 1998

38

The economic environment in South Africa in recent years has been one of relatively high real rates ofinterest, with frequent occurrences of an inverted yield curve. Thus, the activity of some of the largermunicipalities in the shorter end of the market may be influenced by pricing issues and general marketconditions, as well as by a reluctance on the part of lenders to buy long-dated debt.

Note that there is a trade-off between the lower rates typical of short-term debt, and refinancing risk.If the debt is shorter than the life of the asset, the issuer is exposed to refinancing risk, i.e. new debt mayhave to be raised during the life of the asset at a higher rate than the original loan. If the borrower’scredit risk has worsened, it may not be possible to refinance. Refinancing can of course work in favourof the borrower, where the general level of interest rates has fallen, or where the issuer’s credit spreadhas narrowed. In the case of general obligation bonds, the latter could happen as a result of improvedgeneral creditworthiness of the municipality. In the case of project finance, the construction and initialphases of operation are far riskier than later phases in a mature project, and it may be possible torefinance at lower rates. However, financiers are very aware of this, and rely on the later phases toprovide some compensation for the additional risk taken at the outset, so would probably reserve theright to refinance for themselves. All in all, maintaining an unhedged position is risky, and therefore notadvisable with public funds.

3.4 Cash flow profile

The cash flow profile of debt refers to the way in which the borrower pays and the lender receivesinterest and principal over the life of the liability. (As is evident from the discussion on zero couponbonds below, these may not necessarily coincide.) Common profiles are:

• The typical municipal bond is fixed rate and fixed term and pays a semi-annual coupon with theprincipal at the end. Such bonds can be more easily priced off the yield curve, as their profilematches that of government bonds, and are thus usually favoured by portfolio managers.

• Bank structured debt, because it is often linked to leasing finance, usually has an amortising profile,where the debt is serviced in equal instalments over the term of the debt, with earlier paymentscontaining proportionately more interest than capital. Use of this approach in municipal bonds isincreasing in the U.S.

• Zero coupon bonds (“zeros”) do not pay coupons. The investor realises interest by buying the bondsubstantially below its principal value. They can be issued by borrowers - in South Africa usually bybanks and the government, with some activity by parastatals. They can also be created syntheticallyby investment banks by stripping the coupon off a plain vanilla bond. Zeros can be sold to partieswho wish to secure a fixed amount of capital in the future without being exposed to reinvestment risk(or credit risk, if it is a government zero). The coupon stream from a stripped bond can be sold to aninvestor who is more interested in an annuity flow, although each individual coupon can beconsidered as a separate zero.

4 December, 1998

39

• Cash flow profiles of liabilities can be engineered to match the cash flows generated by the assets.They could be index-linked, where the revenue flows are expected to vary with an index, such asinflation or an input cost. In this case, interest payments can go up or down, depending on themovements of the index. Amortising payments, such as those described above, can be structuredwith an escalating profile, with lower debt service in the early years. This is quite common incommercial property finance, where rentals are expected to escalate, and could be appropriate forcertain municipal assets.

• As noted above, municipalities typically issue fixed interest securities. Bank structured loans can bemade at either fixed or floating rates to suit the borrower. The relative advantages of fixed versusfloating rate debt are similar to those regarding short versus long term debt. Floating rate debt impliescontinuous uncertainty about the cost of debt, but is appropriate where matching revenues vary withchanges in interest rates. This is not usually the case for municipalities.

• Again, as noted above, municipal bonds typically pay a semi-annual coupon, in line with conventionin South African bond markets. Structured loans can have varying coupon profiles - semi-annual,quarterly, even monthly - to suit the cash flows of the borrower.

The point to note about cash flow profile, is again to avoid being prescriptive, so as to allow thestructuring of cash flows which match the needs of both lenders and issuers.

3.5 Derivatives

“Derivatives” are financial instruments whose value is based on an underlying financial instrument orindex. Derivatives are practically limitless in number and are unregulated in many markets. In aderivative transaction, the municipality may be issuing a derivative instrument, acquiring one, or as ismost typically the case, doing both.

Investing in speculative derivatives can be dangerous, but using derivatives such as currency or interestrate swaps and default swaps can be prudent in certain circumstances. A blanket prohibition onderivatives may prevent municipalities from effectively managing their risks.

An interest rate swap is an agreement between two parties to exchange future flows of interestpayments. For example, one party agrees to pay the other party a fixed rate; the latter pays the formeran adjustable rate tied to a short-term index. A municipality may choose (perhaps because of its creditquality) to finance long-term assets with short-term loans, but will be concerned about rising interestrates. If it cannot hedge naturally, by matching the duration of its assets and liabilities, it can either acceptthis exposure (which is akin to speculation), or implement the desired change artificially by using swapsas a restructuring tool.

Default swaps are credit derivatives that allow the credit risk of a municipality to be traded separatelyfrom the risk-free component. Such instruments have been used by investment banks for municipalitiesand infrastructure projects. Commodity swaps allow the hedging of price movements in commodities.

4 December, 1998

40

These instruments can be useful for fixing the price of municipal inputs such as electricity (an electricityswap creates predictable future prices for the output). These have recently been structured for a largeindustrial concerns, and could be adapted for use by municipalities.

The above are examples of derivative instruments that could reduce risk for municipalities. Butderivative products can also carry risks that even sophisticated borrowers and lenders can fail toproperly assess.53

It is recommended that derivatives as a class not be prohibited as municipal investments. Instead, thosepermitted should be described by Department of Finance regulation, which might permit investment ininstruments with the following characteristics:

• Derivatives are to be used to reduce risk in future non-financial transactions54 normally carried on bythe municipality as part of its operations.

• Derivatives may not be leveraged by the use of borrowed funds or pledging unrelated assets.Notional positions in derivatives (the nominal principal amount) are not to be leveraged, i.e. financedby funds borrowed or security pledged only or primarily for the purpose of entering into thederivative contract, and are to be completely matched as to maturity (fully hedged).

• Derivatives are to be structured to provide certainty as to the future value of the notional principal orof the cash flows contracted for in the derivative contract. For example, a currency swap or interestrate swap would require that the municipality purchase a known future value or known paymentstream.

• Derivatives transactions may be entered into only with counterparties that are rated “investmentgrade” (e.g. BBB/Baa or higher) by an internationally recognised credit rating agency.

• Derivatives transactions may only be entered into by municipalities with a written and council-approved investment policy, independent oversight, and periodic valuation or “marking-to-market”of investments.

The above rules would permit forward currency swaps and interest rate swaps that afford present daycertainty as to future notional values (e.g. foreign exchange values), future cash flows (e.g. swappingvariable to fixed rate), or future notional principal value (e.g. swapping fixed to variable). Where there

53 The most sophisticated financial markets in the United States have not fully come to terms with derivative instruments andtheir risks. After a rash of state legislation in the wake of Orange County’s losses, regulators have grappled with detailedinvestment guidelines for public entities. Caution is recommended against too low a threshold of regulation, especially given themunicipal management capacity issue. Even sophisticated investors can fail to understand the risks of complicated arrangementsand risky counterparties. Witness the September 1998 near-collapse of Long Term Capital Management L.P. (LTCM).54 The phrase “non-financial transaction” is intended to capture the idea that there must be an underlying “real” business purposefor e.g. a swap. A swap should not be a vehicle for the primary purpose of earning income, but could be used for minimizingrisks involved in the business of providing municipal services.

4 December, 1998

41

is a fixed to variable swap, the duration of the swap should be the same as the maturity of asset pledgedto repay the notional value of the swap and that the pledged asset must not itself must not have beenacquired with borrowed funds or be otherwise pledged or encumbered (fully hedged and noleveraging).

Given the complexity of the possible derivative structures, these concepts should be developed intoregulations by the Department of Finance. These regulations would be authorised by law, and wouldgive substance to a legally-established prudent person rule. The Institute of Municipal Finance Officers(IMFO) and/or the Banking Council of South Africa can and should take an active role in monitoringthe effectiveness of these regulations and suggesting changes as appropriate.

3.6 Recommendations

Key recommendations are:

1) To allow municipalities broad scope to structure their debt obligations to meet their needs.2) To avoid regulation that would prevent sophisticated municipalities from using products that allow

them to manage their assets and liabilities prudently.55

3) To adopt Department of Finance regulations banning the use of derivative instruments which createleveraged, unhedged positions.

55 The question of proper use of municipal funds is broad and largely beyond the scope of this paper. See the discussion insection 4.4.3, infra.

4 December, 1998

42

SECTION 4 RESTRICTIONS ON ISSUANCE AND USE OF MUNICIPALDEBT

4.1 Background and current situation Most countries place limitations on the use of debt by local governments. A distinction is usually madebetween short-term debt and long term debt, and these are discussed separately here. Any dividing linebetween the two is arbitrary, but for the purpose of the current exercise short term debt is defined asthat which is due and payable with a year from the date incurred, and long term debt as that which isdue and payable more than a year after it is incurred.

Short term debt

Short term debt financing can be useful as a part of a municipality’s normal operations (e.g. to payoperating expenses before annual tax revenues have been received) and can also be used in anticipationof a particular non-recurring revenue (e.g. from the sale of assets, receipt of a grant, or issuance of long-term debt). Operating cash needs can be financed from a municipality’s working capital,56 or with theuse of borrowed funds. The former is usually cheaper and more reliable, but many municipalities inSouth Africa do not maintain adequate working capital funds.57

In South Africa today, municipalities may legally borrow to finance both routine and unusual short termneeds. They are required to settle their accounts by the end of the current fiscal year by legislation58 andwithin twelve months by the Constitution.59 The usual form of borrowing is a bank overdraft, thoughother kinds of short-term borrowing are also done. These short term borrowing techniques result inunsecured debt. Historically, the banking community allowed overdrafts as a matter of course. There isreported to be an increase in the use of overdrafts by municipalities, and in some cases, these overdraftsare not being settled as required at the end of the fiscal year. The growth of short-term debt is aconcern to the lender community and is an indicator of stress, both in specific cases and systemically inthe municipal finance system.

Long term debt

The use of long-term debt allows municipalities to acquire or build capital improvements more quicklythan can be done on a pay-as-you-go basis. It allows more equitable payment schemes, since userscan be made to pay for the capital cost of facilities over time. And borrowing from the private sectorleverages public funds to maximise the impact of public investment. But the use of long-term debt also

56 Or, under current accounting practice, from retained surpluses, general reserves, and special provisions57 New accounting rules will require that municipalities budget for working capital, though they will not be required to show it assuch on their balance sheet.58 Local Government Transition Act, Section 8 (a)(ii). The fiscal year is 1 July to 30 June59 Constitution, Section 230

4 December, 1998

43

has costs and risks. It inevitably limits a municipality’s budget flexibility in future years. Unwisely used,it can burden citizens with crippling tax rates or service tariffs.

In South Africa today, municipal borrowing is governed by the new Constitution, the Local GovernmentTransition Act of 1993, and Department of Finance Regulation No. R 412.60 Under this regulation,municipalities may borrow for “capital expenditure which has been budgeted for and approved by thecouncil.” Capital expenditures are defined relatively narrowly. In fact, long term borrowing bymunicipalities has essentially dried up due to lender concerns about the legal and financial environment.

4.2 Decision points – short term debt

For short-term debt, the major concern is that recurring expenditures be met from recurring revenues.Deficits that accumulate over time can reach levels that compromise a municipality’s ability to deliverbasic services. The policy issues are relatively simple: the challenge is to create rules that allowreasonable operational flexibility for good municipal managers while protecting the public against theconsequences of allowing debt to rise to unreasonable levels or to become permanent. The two biggestpotential problems with short term are:

1) it can reach unreasonable levels, requiring an unacceptably high proportion of revenues to bedevoted to debt service at the expense of public services, and

2) it can accumulate over time and become permanent. Eventually, when creditors perceive that thedebt has reached excessive levels, they can decide to withhold further extensions and the pyramid ofaccumulated short-term debt can collapse catastrophically, as happened to New York City in the1970s.

To protect the public against unreasonable debt levels, short-term borrowing could be limited to apercentage of the municipal operating budget, to the anticipated cash needs for a certain number ofmonths, or to revenues specifically foreseen to be receivable within a period of time. To protect againstaccumulations of debt over time, short-term borrowing could be required to be paid off in full at leastonce a year, with appropriate safeguards against immediate re-borrowing.

Up to the 1998/99 budget cycle municipalities used short-term bank overdrafts and credit facilities tofinance their working capital requirements (net current assets). In its 1998/99 budget directive, theDepartment of Finance instructed municipalities to provide their own working capital. The total amountprovided as contributions to funds for 1998/99 amounted to R2,2 billion, while debts increased by R2billion on the previous financial year. Collectively, it can now be concluded that municipalities are on acash funded basis and have eliminated deficit funding.

60 Government Gazette 18764, 27 March 1998, R.412

4 December, 1998

44

The income flows of municipalities, however, are not evenly spread over the financial year. A largeportion of property tax rates61 is paid annually. Many municipalities therefore still require additionalworking capital periodically. It can be argued that the current provision in the Local GovernmentTransition Act, which restricts municipalities to use short-term credit only in anticipation of a positiveincome stream, is a sufficient deterrent and requires no further legislation. This, however, does notprevent municipalities from going ahead with capital projects while they have not secured the funding --external loans or grant funding.

The consulting team’s original recommendation for a generous limit for short-term debt, i.e. that it belimited to not more than the coming year’s planned revenues, was met with concern. The referencegroup overwhelmingly felt such a limit was too generous, and encouraged an approach which would tieshort term borrowing to specific and realistic revenues to be received within the fiscal year. It isrecommended that short term debt be paid down to zero at least once each year. The responsibility forcompliance should rest with both appointed officials and elected representatives of the municipality.

It may be worth providing that lenders are also responsible for seeing that short-term municipal debtcomplies with these rules. At a minimum, lenders should be precluded from extending new short-termcredit if the borrower is not in compliance. Lenders may object to the additional burden this imposes tomake appropriate inquiry, but it is appropriate that both parties to the transaction exercise responsibility.Sanctioning and mandating the cutting off of additional short-term credit when that is appropriate mayeven help protect the lender from political pressure in hard cases.

4.3 Decision points -- long term debt

For long term debt, the policy issues are more complicated and more problematic. Any restrictions onlocal powers involve some substitution of national policy judgement for local flexibility and the naturalregulating effect of the market’s response to municipal borrowing. Among the restrictions that may beconsidered on national policy grounds are:

• restrictions on the purpose of borrowing• restrictions on the use of proceeds• restrictions on the amount of borrowing• restrictions on the type of security given or recourse available to the lender

4.3.1 Particulars of concerns about long term debt

It is generally accepted that long-term debt is appropriate for capital investment, and that as a matter ofpolicy, the term of the borrowing should be related to the useful life of the capital asset being built oracquired. More problematic is the question of whether multiyear debt should ever be allowed for otherpurposes. These purposes could include work-outs as part of a recovery package or extraordinaryexpenses related to the transition and restructuring now underway in the country.

61 Rates constitutes 20,1% of total income (June ’98 Project Viability figures).

4 December, 1998

45

4.3.2 Restrictions on purposes of long term debt

At least two approaches could be taken to long term debt: a permissive general authorisation for longterm municipal borrowing would allow municipalities to borrow for any public purpose authorised tothem by law. It would be up to the local council to decide what borrowing is wise and appropriate.

Or, municipalities could be authorised to borrow only for a limited list of public purposes. Settling onthe list of permitted purposes would be an exercise in itself. Possible items for such a list would be:

• building or acquiring a capital asset whose anticipated useful life will equal or exceed the term of theborrowing

• funding accumulated operating deficits as part of a legal or administrative restructuring, withprovincial approval

• funding other extraordinary needs, including but not limited to retrenchment packages62 or programsto abate natural or manmade disasters.

Competing interests are at play when one considers restricting local government’s use of long term debt.Favouring a broader authorisation are two factors:

• municipal finance is an evolving art – room to adopt new forms and techniques is desirable• given the national policy favouring decentralisation, responsible local managers should be allowed

maximum flexibility

Favouring a more specific list are two factors:

• clarity about what is permitted helps the market• there is a perception that the public needs to be protected from politicians or managers that might

use long term debt inappropriately.

The policies recommended in this paper are primarily aimed at facilitating municipal borrowing forcapital infrastructure. Given the strong concern expressed in the reference group about any otherpurpose for long-term borrowing, it appears that borrowing for other investments, however sound intheory, is a distraction to the primary objective. Accordingly, it is recommended that borrowing beauthorised only for the acquisition of identifiable capital assets (property, plant and equipment). If thereis good experience with municipal borrowing, the list of permitted investments could well be expandedto include such items as funding accumulated operating deficits in a work-out situation or fundingretrenchment programs which have demonstrable long-term benefits.

62 Many of those interviewed cited over-staffing as one of the biggest structural problems municipalities face. Meaningfulprograms for retraining and retrenchment are likely to require multi-year funding, but could well pay for themselves within a 3 to5 year period.

4 December, 1998

46

4.3.3 Recommendations as to long term debt

A recent International Monetary Fund publication has argued for a rules-based control of sub-nationalborrowing behaviour, with limits on the debt of individual jurisdictions that “mimic market discipline.”63

Such a test could be framed in terms of the maximum burden of annual debt service in relationshipto realistic and conservative projections of the revenues that are and will be available to pay debtservice. This approach depends on:

• the definition of “debt service,” which is used in the numerator64; and

• the definition of “available revenue,” which will be used as the denominator. The usual formulationis to use recurring revenues, netting out extraordinary items.

The argument against such restrictions is that they are an unnecessary interference with the market, andthat in particular cases it may well be necessary or desirable for municipalities to temporarily exceed anygiven ratio as it invests for the future or spends down accumulated reserves.

The argument in favour of such restrictions is that they protect the public from reckless borrowing byofficials or elected representatives who may not be sensitive to the long-term risks.

Although a limiting ratio can be set65 which provides a degree of protection, and is high enough not tointerfere with sound management in most cases, several observers warned that establishing such a limitwould be seen by municipalities as permission or endorsement to borrow to that limit.

Although ratio tests can be relatively easily applied to general fund debt, they will generally not makesense when applied to self-financing or “enterprise” projects, and they could restrict the ability of rapidlygrowing areas with bright prospects to expand services to meet demand. In keeping with the generaltheme of minimum necessary regulation, no ratio tests are recommended.

For now, term debt should be limited to capital investment in property, plant, and equipment. Ifexperience with municipal borrowing is positive, the power to borrow can be made more general in thefuture. In no event should long term debt should be used to finance current budget deficits.

63 Teresa Ter-Minassian, editor, Fiscal Federalism in Theory and Practice, Washington DC, International Monetary Fund, 1997,p. 171-172.64 For example, Addendum 5 is a chart, developed from Project Viability data, which shows the distribution of debt service(interest plus repayment of principal) as a percentage of income for 405 municipalities tracked by Project Viability. This dataincludes both internal and external loans, whereas new accounting standards will track external debt separately.65 A 15% limit was contemplated, but this could be too low for South African municipalities trying to meet substantial servicebacklogs A 25% limit was proposed, but there was concern that this would sanction over-borrowing. As shown in Addendum 5,current practice ranges up to nearly 30% in a few municipalities, but that figure includes internal as well as external borrowing.The mean debt service ratio for Project Viability core cities, including internal and external loans, is just under 10%.

4 December, 1998

47

4.4 How may proceeds of long term borrowing be used?

4.4.1 Key concerns and policy issues about use of proceeds

Given that long term borrowing is to be used only for capital investment, should there be any furtherrestrictions on the use of proceeds? Possible areas of concern include arbitrage and the use of publicfunds for private benefit.

Arbitrage

Arbitrage is the practice of investing borrowed funds at a return greater than the interest being paid tothe creditor. Arbitrage schemes can be tempting, but are almost always a bad idea. If the risk/returnrelationship is operating normally, and in the absence of tax-related effects, it should not be possible fora municipality to invest at a higher rate than it is paying without risk. The lessons of Orange County,California are one recent reminder of the dangers of seeking higher interest at the price of greater risk.

Private benefit

Some municipal debt systems try to restrict the use of public funds for private gain. In days of theAmerican Wild West, it was not uncommon for private railroad speculators to interest town fathers inhelping build a rail link – an early public-private partnership. Many of these joint ventures went bust,and led to state constitutional prohibitions on the use of public funds for joint ventures or to acquireequity interests in private ventures.

Where there is a tax exemption for interest on municipal debt, i.e. a subsidy for municipal borrowing, itis reasonable to restrict the use of the privileged borrowing and its implicit subsidy to acceptable publicpurposes. In the U.S. the tax exemption for municipal debt is taken away where use of proceeds failsto meet certain public use tests.

4.4.2 Pros and cons of restricting use of proceeds

Arbitrage

A tax exemption or privilege for municipal debt lowers municipalities’ cost of borrowing and wouldtherefore increase the possibility of arbitrage earnings and the temptation to borrow for investment.Since any borrowing does involve risks and limits future flexibility, borrowing for speculation should bediscouraged. Therefore, some municipal debts systems require the proceeds of borrowing to be spenton the stated purposes within a fixed period of time. Absent a tax privilege for municipal debt, there isno reason to expect wide scale borrowing for arbitrage, and therefore no need to restrict arbitrageexplicitly.66 There are accounting issues involved in any attempt to restrict arbitrage earnings. Tracing the

66 A an arbitrage restriction per se could do little harm, but could impose an analytic and regulatory burden. The limited capacityat all levels of government is probably better used elsewhere. More generally, it would make sense to restrict the investment of

4 December, 1998

48

proceeds of specific borrowing can be difficult, especially as financially challenged municipalities movemoney from one fund to another.

When interest rates change dramatically, opportunities to invest at positive earnings during theconstruction period may in fact exist, and there is no reason to prohibit such investments, as long as theypass normal “prudent investor” scrutiny.

Private benefits

Drawing the line between private and public benefit can be difficult, especially where a private companyis a key player in a community’s economic development strategy. Absent a tax privilege for municipaldebt, there is no compelling reason to limit the use of borrowed funds to any greater extent than otherfunds of the municipality.

4.4.3 Recommendations as to use of proceeds No special restrictions are recommended on the use of proceeds of borrowing. As with other municipalfunds, they should be held and invested with prudence and used only for authorised public purposes.The general provisions of the law on use and investment of public moneys should be reviewed with theexpiration of the Local Government Transition Act.,67 That Act properly relates to all municipal fundsand includes the proceeds of borrowing. Any regulations on the use of municipal funds should beincluded in upcoming municipal treasury legislation, rather than in the proposed Municipal BorrowingBill.68 Special provisions prohibiting arbitrage could be included in municipal borrowing legislation, butas noted, there are enforcement issues. And in the absence of a tax privilege for municipal debt, there islittle incentive for arbitrage, and no compelling reason to prohibit a conservatively invested municipalityfrom benefiting from a general market upturn in interest earnings, just as it would suffer from a generalupswing in interest charged.

4.5 Summary of Recommendations Key recommendations are:

1) To limit short term municipal borrowing to that amount required:a) To bridge operating cash shortfalls in anticipation of specific and realistic future income streams

to be realised within the fiscal year, and

the proceeds of borrowing, and of other public funds, to those investments which would be chosen by a “prudent person, withdue regard to the preservation of capital.”67 The LGTA, in Section 9, provides that subject to policies to be adopted by regulation, municipalities may invest in a relativelyshort list of instruments, including bank deposits and government securities. There may be a constitutional issue as to whetherreplacement legislation is actually needed on the expiry of the LGTA, in light of the broad powers granted to the municipal sphereby the new constitution, but that issue should properly be part of any municipal treasury debate.68 Such legislation might take an approach similar to the LGTA, with a basic list of permitted investments, and with authority forthe Minister to prohibit or restrict in general or special circumstances should the need arise.

4 December, 1998

49

b) to bridge capital requirements in anticipation of specific and realistic grants to be received ordebt to be issued within the fiscal year.

2) To require that short term debt be paid off annually, and remain paid off for some reasonable periodof time, with both borrower and lender responsibility for enforcement

3) To limit long term debt to capital investment in property, plant, and equipment.

4 December, 1998

50

SECTION 5 HOW SHOULD BORROWING BE AUTHORISED?

5.1 Background and Current Situation

The Constitution does not itself deal with the question of how a municipality authorises borrowing. Itforesees national legislation to deal with financial issues generally. This can and does include provisionsrelating to such authorisation.

The Local Government Transition Act, No. 209 of 1993 (LGTA) requires a municipality to compile andapprove an annual budget including capital expenditure. This budget must reflect the source of futurefinance.69 The municipality can only spend in accordance with its approved budget. The approvedbudget must be submitted to the Minister of Finance, who can exercise considerable influence andcontrol “in so far as it may be essential for the pursuit of the national economic policy,” by determiningmaximum expenditure limits for budgets or components of budgets. A budget that does not comply withmaximum expenditure limits is referred back to the council for consideration and amendment at the nextcouncil meeting. The LGTA requires the municipal budget to be approved by a two-thirds majority ofthe members of a council. A decision to incur expenditure requires a majority of the votes cast.70 Adecision to raise a loan may fall into this expenditure category.

When the LGTA expires, these provisions will be superseded by the Constitution, which provides thatthe municipal council must approve its budget and raising a loan by a vote of a majority of its members.All other questions are decided by a majority of votes cast.71 When the LGTA expires it willpresumably be replaced by other national legislation.

The policy reflected in the Local Government White Paper is to enhance local accountability,transparency and participation in all aspects of the municipal decision making process. In principle, thispolicy should include the budget process and the raising of loans, but no voter approval for anyborrowing is currently required. The notice requirement that currently applies is a post facto publicationthat a debt has been incurred. No advance public notification is required before a borrowing takesplace.

5.2 Decision points

Within the municipality, one can envision a spectrum of possible ways in which borrowing might beauthorised. From the least broad-based to the most, the possibilities for approving a loan can beorganised as follows:• by action of the executive• by action of the council• by action of the community at large, e.g. in a referendum.

69 LGTA, sections 10G(3) and (4)70 LGTA, sections 10G(3)(a) and (4)(a)71 Constitution, Section 160(3)

4 December, 1998

51

Beyond the municipality, one can envision approval of borrowing:• by provincial authorities• by national authorities

Each of these approvals could relate to a variety of issues, including the financial capacity of theborrower to pay the debt, the purpose of the borrowing, the form of borrowing, and consistency withnational economic policy.

Given all of these possibilities, what is the right path for South Africa?

5.2.1 Within the municipality

Borrowing authorised by the executive alone is the quickest and easiest way to borrow. This may wellbe appropriate for relatively small amounts and for relatively short-term borrowing, where the generalfinancial health of the municipality is not at risk. But it is not be appropriate for large amounts or longterms. As noted previously, borrowing can be a powerful tool, but it involves trading off future flexibilityin exchange for investment capital today. For large amounts of debt, the decision about how muchflexibility to trade away is a policy decision properly made by the council. Experience in other countriesclearly demonstrates that without council approval (where elected local councils exist), the probability offuture debt repudiation or refusal to enact necessary tax or tariff changes to meet debt serviceobligations increases dramatically.

Additionally, approval for borrowing may be submitted to the community at large for approval.Borrowing authorised by action of the community at large has several advantages:

• it encourages citizen participation in the decision making process,• it provides a philosophical basis for binding the community for the very long term, notwithstanding

changes in the elected council,• it confirms citizens’ consent to the subsequent attachment of municipal assets or income streams,

and the imposition of special rates or charges, as necessary to repay the debt

However, requiring voter approval for borrowing has its disadvantages:

• it adds to the time and expense of borrowing transactions• it can turn financial decisions into political battles

In the specific situation in South Africa, moreover, there are some additional problems:

• the majority of the participating citizenry does not own real property that can be taxed

4 December, 1998

52

• currently, the majority probably does not pay the service charges for municipal utilities in a timelyfashion72

• there may be constitutional limits on the extent to which the power of future councils can be limited,even by a general vote of the community.73

On the whole, approval for most borrowings is not recommended. It important that successor legislationto the LGTA provide for broad public participation, e.g. in the form of well-advertised public hearings,in the development of municipalities’ capital improvement plans, capital borrowing strategies, andspecific borrowings. This kind of participation can achieve many of the advantages of voter approvalwithout the disadvantages described above.

Therefore, it is recommended that borrowing generally be authorised by council at the recommendationof the chief executive. To the extent permitted by the constitution, it is further recommended thatcouncils be authorised to delegate to the chief executive the power to borrow for a short term up to alimit specified by council.74 For extraordinary situations, such as exceeding any general limits that maybe imposed on municipal borrowing, the creation of special taxing districts or the imposition of specialfees for debt service, it may be appropriate to consider voter approval, if the constitutional issues can beresolved.

Before long term borrowing is undertaken, it is recommended that each municipal council have in place:

• a written debt policy and• a capital improvement strategy.

Debt policy

The council of any municipality planning to issue municipal debt should adopt a written debt policy. Thispolicy helps establish limits and provide general direction to municipal executive officials in the planningand issuance of debt. Each actual long term debt issue should be specifically approved by the Council,but developing an overall debt policy assures that relevant policy questions are considered from abroader perspective. In addition, a debt policy lets municipal policy makers integrate debt planning withother long-term planning and financial objectives. A carefully crafted and consistently applied debtpolicy signals lenders and rating agencies that the municipality is committed to controlling its borrowing.Some of the items that should be spelled out as a matter of policy are:• What are acceptable levels of short and long term debt? The municipality must decide to what

extent it is willing to give up future flexibility. Debt issuance involves a trade-off. In exchange forfunds for current capital improvements, future spending is limited. The degree to which a

72 This has to be viewed as a transitional situation. A government which cannot convince its citizens to pay taxes or charges forservices cannot provide services for long.73 Section 151(2) of the Constitution provides that “executive and legislative authority of a municipality is vested in its MunicipalCouncil.”74 Section 160(2) provides that the raising of loans is a matter that Council cannot delegate. Whether this applies to existing shortterm borrowing practices is an open question.

4 December, 1998

53

municipality is willing to make these trade-offs depend on the urgency of its capital needs, itsexpected rate of growth, economic tends, and the stability of its overall finances.

• What are acceptable purposes for which debt can be issued? To what extent must there be apublic benefit? Must the improvement have a useful life of at least the term of the debt? Is waterand sewer debt acceptable, but not debt for infrastructure that does not generate revenue?

• To what extent, and for what purposes, will the municipality use general obligation debt vs. revenuedebt? Where the municipality can use revenue debt, it is less limiting of future flexibility. However,not all important projects generate reliable revenue streams to repay debt.

• What covenants, pledges, or security interests is the council willing to consider in order to makeborrowing possible and/or lower its cost of borrowing?

• For what purposes will the municipality use "pay-as-you-go" financing, and for what purposes will itconsider debt financing? A municipality should not issue debt for ongoing operations andmaintenance, nor for short-lived improvements or repairs. The municipality may want to shiftinfrastructure development costs to beneficiaries, e.g. through user fees, service charges, ordeveloper financing to the maximum extent possible.

• Will the municipality use variable rate debt, or will it only issue fixed rate obligations? In a highlyinflationary economy, fixed rate long term issues are not viable.

• For what term will debt be issued? For the near term, the market may be more of a limiting factorthan any policy the municipality might develop. However, the municipality will want to avoid issuingdebt for longer than the useful life of the improvement to be financed. At the same time, it will wantto spread the debt over a long enough period that the payments due in any given year aremanageable in light of the anticipated revenues.

• Will debt be placed competitively, or by negotiation? If the latter, what safeguards will assure themunicipality that it is borrowing at the lowest possible cost?

An organisation such as the Institute of Municipal Finance Officers (IMFO) could consider and adoptmodel guidelines to help municipalities develop meaningful debt policies at the local level.

Capital improvement strategy

Consideration of particular debt issues should be preceded by a long term capital investment planningprocess. The goal of such a process is to consider and prioritise a municipality’s capital needs, ideally inan open participatory process that will maximise the chances that the resulting capital improvements planwill: (a) represent a community consensus that is sustainable over the long term and over any change inleadership, (b) target the available resources on the most urgent needs, and (c) avoid the need toevaluate individual projects in a vacuum.

At a minimum, such a process will include analyses of capital needs, environmental considerations,financing considerations, project design and selection criteria, and the impact of planned improvementson annual costs for debt service, operations, and maintenance.

Integrated Development Plans (IDPs) required by the Local Government Transition Act (a requirementlikely to be survived in the Systems Bill or other legislation) will address most of the issues that should

4 December, 1998

54

be considered. The IDP process may be difficult for some municipalities, but wise use of debt wouldrequire a similar effort even if it were not mandated by law. As outlined in the White Paper on LocalGovernment, the IDP process includes:• An assessment of the current social, economic and environmental reality in the municipal area - the

current reality.• A determination of community needs through close consultation.• Developing a vision for development in the area.• An audit of available resources, skills and capacities.• A prioritisation of these needs in order of urgency and long-term importance.• The development of integrated frameworks and goals to meet these needs.• The formulation of strategies to achieve the goals within specific time frames.• The implementation of projects and programmes to achieve key goals.• The use of monitoring tools to measure impact and performance.

Wise planning would include specific consideration of how various capital improvements are to befinanced, how the borrowing capacity of the municipality should be used to that end, and a forecast ofthe effect of borrowing (and facilities built with borrowed funds) on future operating budgets, taxes,rates, and utility tariffs.

5.2.2 Beyond the municipality

National or provincial approval could be predicated on a variety of conditions, including:• the financial capacity of the borrower to pay the debt, as measured either by analysis or the

application of a formula or ratio specified by law or regulation75

• consistency with national economic policy, e.g. as to timing of the borrowing• the purpose of the borrowing• the form or formalities of the borrowing

In general, such reviews can control irresponsible borrowing at the local level, but any such review willtake time and will require the creation of capacity at the national or provincial level.

The marketplace, aided by the new accounting rules, the rating agencies and their improving analyticalcapacity, and individual investor reviews, will do the best job of testing the financial capacity of theborrower to pay the debt. There is no need to spend scarce resources to develop analytic capacity thatis already growing in the private sector. A certification that a given formula or ratio has not beenexceeded can be provided by a certified public accountant as well as by a governmental agency.

From time to time the national economy could require some form of temporary restraint on thecumulative volume of municipal loans.76 The Minister of Finance should implement this at national level. 75 See Section 4.3.3 of this report for a discussion of ratio tests and their possible uses76 In practice, this kind of restraint can be difficult to apply. Would a national restraint on cumulative involvement lead to a raceto the bank? Would there be some allocation scheme? How would the market know when the cumulative limit has been, or isabout to be reached? What happens to transactions in process the day the door is shut?

4 December, 1998

55

This concept is narrower than the existing provisions of the LGTA77 which envisage the Minister alsoexercising line-item control over individual municipal projects “for the pursuit of national economicpolicy.” Under this provision, the opinions of the Minister (or a designated functionary) would replacethe deliberations of the elected municipal council as to the purpose of the borrowing. Futurelegislation may well reflect a preference that these narrower matters should be left to the discretion ofindividual municipal councils to adjust within the complexities of the integrated development plan. Forthe purposes of a policy framework for municipal debt, no review of the purpose is needed.

That leaves the question of whether approval by provincial or national approvals as to the formalitiesof the borrowing is desirable. In the U.S. State of Texas, for example, local government issuers mustobtain approval from the Office of the Attorney General. That office reviews the proposed bonds andtheir supporting documentation, certifies their validity, and issues an opinion. This opinion is needed forthe bonds to be legally binding, and relieves individual purchasers of the necessity to inquire into theprocess by which the bonds were issued. This kind of certification can help build investor confidence,and relieve individual investors of some of the “due diligence” that would otherwise be required.However, it is not clear that under current conditions in South Africa, there is meaningful capacity at theprovincial or national level to devote to this task. Imposing a review requirement that cannot be quicklyfulfilled would harm the development of the market more than leaving the “due diligence” inquiry on theinvestor and/or private counsel. On the whole, the virtues of this approach do not currently outweighthe potential for delay and confusion. This could be revisited if issues of form and authority become asignificant problem.

5.3 Recommendations

Key recommendations are:

1) To require that borrowing be authorised by municipal councils at the recommendation of the chiefexecutive.

2) To provide that short term borrowing may be authorised either by council resolution in eachindividual instance or by a general council resolution that the chief executive may borrow up tostated limits.

3) To require no provincial or national review or approval.

77 LGTA sections 10G(4)(a) and (b)

4 December, 1998

56

SECTION 6: CONCESSIONAL FINANCE AND OTHER ASSISTANCE

6.1 Background and Current Situation

In considering assistance, this section starts from the perception that there are three groups ofmunicipalities.78 The first group includes municipalities that have the possibility of direct or indirectaccess to the capital markets, given a supportive policy and regulatory environment. The second andthird groups include municipalities that cannot access the market, even through intermediaries.

It is critical to be clear about why particular municipalities cannot access the capital markets.Distinctions in this regard are what differentiate the second and third groups. The second group ismade up of those which have or could have predictable, stable, and sufficient revenues to meet theirservice delivery responsibilities, but whose capital requirements are relatively small, or who lack theskills and capacity to develop and pursue borrowing options. The third group includes those that donot have, and cannot develop, adequate financial resources to meet their responsibilities. Borrowing willnot solve that problem. In fact, it will exacerbate the problem. No one should strive to create capitalborrowing programs for municipalities with inadequate revenue sources.

A substantial proportion of the nation’s population resides in the first group of municipalities, and asubstantial proportion of the demand for infrastructure financing will be generated within this group.79 Ifthe framework for municipal debt is done right, the markets should be able to finance capitalinfrastructure for this group of municipalities.

The size of the second and third groups depends on government policy choices. The morestable and predictable revenues are provided, the fewer municipalities will be in the third group. TheWhite Paper on Local Government recognises the need for certainty and adequacy in local revenues.The key question is how to support municipalities that do not have and cannot develop the resources tobe self-financing, e.g. because they do not have an adequate tax base. To the extent that South Africachooses to assist these municipalities through a predictable and stable system of intergovernmentaltransfers, they will have adequate local revenues. Municipalities’ revenue streams, from both localsources and intergovernmental transfers, can be used for capital investment (with or without borrowing).

If a community does have reliable streams of revenue, then it is definitionally in the second group, andsome form of access to borrowed capital should be available. Wholly private markets may still not 78 See section 1 of this report. This analytic scheme relates to the general financial condition of municipalities. Viable projectscould be found in any jurisdiction that are more or less independent of the jurisdiction’s general creditworthiness. A similar three-tiered analysis can be applied to project-specific debt.79 The concept of “demand” for debt financing is a slippery one in South Africa. “Demand” for infrastructure financing may differfrom the “need” for infrastructure for several reasons, including: (a) demand for market financing is likely to be generated by thosethat perceive themselves as being creditworthy; and (b) need for infrastructure (however measured) is generated by unmetrequirements or breakdown of existing infrastructure, irrespective of municipal creditworthiness. In light of the recent municipalhistory of South Africa, it is possible that market demand for financing may actually be negatively correlated with the need forsuch financing.

4 December, 1998

57

serve these municipalities because of the small size of their financing needs, because of their inability todo the analysis and planning necessary, or because of their inability to speak the language of the capitalmarkets. For this group, market intermediaries and technical assistance could be made available to helpbridge these gaps. This group is the major focus of this section.

6.2 Key Concerns and Policy Issues

Municipal borrowing in South Africa carries a stigma that is partly attributable to the turmoil anduncertainty caused by national policies. Policies of demarcation, local government restructuring, andamalgamation have helped close private capital markets to municipalities. It is a legitimate role and evena responsibility of national government to help provide an environment of marketability for local debt.Getting the macro policies right and in place is a big part of what needs to be done to help municipalitieshave access to borrowing. The focus of this section is on more direct assistance.

Several concerns and policy issues arise when considering assistance for municipalities in the two groupsthat are not presently market-accessible. These include:

1) Whether assistance of any kind should be rendered to help municipalities to access credit;

2) If so, what forms such assistance should take in order to encourage maximum possible privatecapital market participation and minimise or eliminate “crowding out” of private capitalproviders;

3) How should finite assistance resources be fairly allocated; and

4) What administrative and technical assistance channels should be used to deliver assistance?

6.3 Who should receive assistance?

Should assistance be provided to Group 1 municipalities? The argument could be made thatcapital infrastructure to serve historically under-served populations is so badly needed that even Group1 municipalities should receive some assistance. The counter-arguments are (1) that assistance to thisgroup allocates scarce resources to municipalities that already have market access, and (2) that attemptsto help Group 1 municipalities, e.g. by making credit cheaper, could reduce the private sector’s interestin providing capital, and thus reduce the total amount of investment capital available for South Africanmunicipalities.

Should assistance be provided to Group 2 municipalities? The main argument against assistance isthat it risks creating dependency and reducing market incentives. But since this group does not havemarket access, the need is clearer than with Group 1 municipalities. Therefore, it is important that anyassistance be used to leverage rather than reduce private sector involvement. By focusing on helping

4 December, 1998

58

Group 2 communities to access private credit sources, scarce resources can be used to provide a netincrease in the amount of investment capital available to South African municipalities.

Should assistance be provided to Group 3 municipalities? Here, the need for capital investmentmay be clear, but the need for debt financing is not. It would be both dangerous and futile to attempt toprovide capital market access to those municipalities that fall into Group 3, except in the context ofspecific and creditworthy projects. By definition, Group 3 municipalities are likely to be poorcandidates for private investment due to fundamental weaknesses likely to persist for the mid- to long-term. If they were able to secure loans, the additional burden of interest, increased operating costs, andfinancial management might not be sustainable. It is likely to be more effective to use such tools asstable intergovernmental transfers, capital construction or grants,80 targeted economic developmentassistance and jurisdictional amalgamation to fundamentally and permanently change their fundamentaleconomic status enough to move them into Groups 1 or 2.

6.4 What kind of assistance?

Assistance in accessing capital markets can take any of several basic forms. These range from technicalassistance of various kinds, through several possible forms of credit enhancement, to direct lendingpositions and interest rate subsidies that induce private market participants to join a transaction. At leastthree basic questions can help decision-makers judge whether to use a given tool in a given instance:

First, how efficiently and effectively does the specific tool leverage private sector investment in a giveninstance?

Second, (and closely related to the first) how likely is use of the proposed tool to crowd out privatesector capital?

Third, to what extent is use of the tool likely to increase the risk of moral hazard; that is how likely is itsuse likely to be misinterpreted as a form of State guarantee?

It seems reasonably clear that some crowding out by government-owned development lenders such asDBSA has occurred during the past two years. Efforts have been made by commercial marketparticipants and DBSA to agree on principles designed to prevent this from happening in the future, butthis matter, by its nature,81 will require continuing communications and vigilance at the policy andoperating levels of both assistance providers and the investment community.

80 Centrally allocated capital grants can also create issues of sustainability and appropriateness. 81 The ability of market participants to detect and prevent potential instances of crowding-out is inherently limited. This isparticularly true when it may well be in an astute borrower’s best interests to play assistance providers off against commercialinvestors and when formal or informal confidentiality agreements prevent the free exchange of information between commercialinvestors and assistance providers.

4 December, 1998

59

6.4.1 Technical Assistance and Training

In general, technical assistance and training which assists municipalities to become more credit-worthy isthe category most likely to efficiently leverage private sector capital investment, least likely to result incrowding out, and least likely to incur moral hazard risk. Technical assistance and training could beprovided in some or all of the following areas:

• Accounting and budgeting• Identifying capital investment projects• Analysing capital investment projects• Operating and managing facilities• Collecting and disbursing funds

Technical assistance and training should be provided in a way that maximises community participationand the chances that a community will eventually be able to do some or all of these steps for itself.There is always a tension between wanting to help and wanting to create local capacity. Whereverpossible, the bias should be toward creating local capacity. Technical assistance with capital planning,cash flow projections, and project management which allows the municipality to experience budgetconstraints, to practice matching revenues to expenditures, to consider how much to borrow, for whatpurposes, and how quickly to pay it back, is desirable.

One possible locus of technical assistance, especially for self-supporting projects, could be theMunicipal Infrastructure Investment Unit (MIIU), a Section 21 Company established by the SouthAfrican Government to help municipalities with financing and management of municipal services. MIIUcould provide training and assist with feasibility studies that would prepare municipalities to examinefinancing alternatives and consider options.

A targeted training assistance program has been developed by the Department of ConstitutionalDevelopment, in conjunction with the Local Government Education and Training Board. These courses,aimed at municipal officials and councillors, are intended to support the National InterventionProgramme (Project Viability), and are currently available only to municipalities that have been identifiedas troubled by Project Viability. These courses include:

• A refresher course for chief executive officers and treasurers• A formal training and monitoring course for finance officers without a qualification• A basic skills course for finance officers at the entry level.

Parliament has recently adopted a new Skills Development Act No. 97 of 1998, which will come intoeffect on a date proclaimed by the State President. This Act will require every sector to form its owntraining authority. The Local Government Education and Training Board and the Training Board forLocal Government will disappear and a new Sector Education Training Authority (SETA) will beestablished. This body will be responsible for education and training in the sector and will also controlthe quality of training by others. All providers in the sector will be required to register with this authority

4 December, 1998

60

as a quality assurance body. Material developed for Project Viability can follow this route once SETAhas been established. These materials can and should be expanded so that they are available to allexisting and emerging municipalities, not just those that have been identified as financially troubled.

In addition to government-sponsored training activities, the private sector offers some training andcapacity building to client municipal officials and elected representatives. In large part, this is acompetitive incentive offered by banks in order to attract deposit accounts and other municipal business.This practice is commendable, and should be encouraged. Educated consumers of municipal credit arelikely to be the most potent drivers in developing competitive private sector approaches.

6.4.2 Financial Assistance

Various forms of financial assistance can be envisioned to help Group 2 municipalities gain access toprivate credit. Indeed, many have already been used to some degree in South Africa. However, manyforms of direct financial assistance carry significant risks:

• they can foster dependency. The lure of cheap money provides an incentive to be needy, ratherthan an incentive to be as self-sufficient as possible.

• they can result in hidden subsidies. It is difficult to quantify the cost of contingent guarantees andenhancements.

• they can crowd out the private sector. Several municipalities told us that they went to DBSA forloans because DBSA offered the best interest rates82

• they can be less efficient at leveraging private sector resources than, e.g. technical assistance

Careful design and use of the financial forms of capital market access assistance can reduce these risks,but perhaps not eliminate them.

Another problem that can easily arise with financial assistance is that the sustainability of facilityoperating costs can be overlooked. In some fashion or other, financial assistance to assist with capitalaccess will reduce the cost of capital to the borrower (often from an “infinite” level when it wascompletely unavailable without the assistance). By helping the borrower access debt, the assistanceprovider is also likely to be helping the borrower take on increased facility operating costs. Borrowerswho had no access to capital from hard credit sources may not have been required to fully investigatethese costs and explicitly build them into their budget planning.

Perhaps the most fundamental principle governing the design of direct financial assistance is that it shouldbe targeted “down-market” of the (always-shifting) commercial credit markets.83 This implies that the

82 These municipalities may not have as much access to private capital as they believe, but the fact remains that they are tellingthemselves and others that DBSA is one bank among many, and that lower interest costs are their primary reason for borrowingthere.83 For example, INCA is reported to have rejected at least R1 billion in loan applications on credit grounds. This group of rejectedloans is in some way “down-market” from commercial debt markets. Some combination of technical assistance and/or financial

4 December, 1998

61

assisting entity would absorb one or more risks that are presently unacceptable to the private creditmarkets. This point has vital implications for the operating philosophies, business strategies, incentivesystems and other key characteristics of institutions such as DBSA that deliver this form of assistance.

6.4.2.1 Direct Lending Positions

One of the least efficient forms of financial credit market access assistance – and the one most likely tocrowd out private lenders, as well as invite moral hazard – is direct lending. However, as theInternational Finance Corporation’s A/B Loan syndication and certification structures havedemonstrated, substantial leverage efficiencies can still be achieved with such instruments, if they arecarefully designed and implemented.84

In order to maximise the efficiency of leverage and minimise the risks of crowding out and moral hazard,direct lending should be designed as much as possible to induce co-financing by the commercial lendingmarket participants. The smallest possible direct lending role required to achieve this objective will bothminimise the risk of crowding out and maximise the efficiency of the assistance rendered. Thus, forexample, if only a 5% junior lien position will induce the private market to join in a co-financing, theprovider of this form of assistance must be prepared to forego the institutional and personal satisfactionsof providing a larger facility. Personnel and other incentives of the assistance provider must be designedto support the goals of maximising efficiency (leverage) of the assistance provided rather than its sheersize.

6.4.2.2 Direct Interest Rate Subsidies

Interest rate subsidies are similar to direct lending positions in that they are likely to be relativelyinefficient uses of scarce resources and are more likely to lead to both crowding out and moral hazardthan indirect financial assistance (such as limited guarantees) or technical assistance. Nonetheless, theycan be useful tools, if they are carefully designed to insure that the smallest possible subsidy necessary toinduce private capital market participation is used, and if they are used only where the assistanceprovider can be reasonably assured that this is the only tool that will make the borrower credit worthy.

6.4.2.3 Guarantees

Guarantees are an indirect form of financial assistance. Due to their contingent nature, their cost is notalways easy to measure at the time the guarantee is given. Broad guarantees are a problem that is wellunderstood in South Africa. They lead to lax lending practices and hinder the creation of effective assistance might make these rejected loans worthy of reconsideration by market participants. Finding the minimum packagenecessary to make each rejected applicant commercially credit-worthy would be a relatively efficient means of using scarceresources and relatively unlikely to lead to crowding out. This might mean finding a way to eliminate a narrow risk (e.g.environmental risk) through provision of risk insurance. Or, it might mean taking a junior lien or even a pari pasu debt position inorder to comfort potential private lenders. Or, it might mean a guarantee on the “long end” of a debt structure if commerciallenders are able to provide short and medium principal maturities.84 DBSA believes that its loans can provide the private sector with sufficient comfort to co-finance transactions in which privatelenders would not otherwise be interested.

4 December, 1998

62

markets. However, guaranteeing specific risks and/or specific maturities may be worth consideration.The ability of an assistance provider to reduce or eliminate specific risks in a transaction (e.g. specificgovernmental performance, specified environmental hazards, etc.) or to back specific maturities that theprivate sector is unwilling to provide can leverage private capital investment. Properly designed andimplemented, such guarantees can reduce the risks of crowding out and moral hazard. Some of theWorld Bank’s guarantee operations have begun to demonstrate the utility of guaranteeing specific risksor maturities.

DBSA’s plans to provide partial risk and partial credit guarantees may be an appropriate means ofdelivering assistance in the form of third party credit enhancement. One means of reducing the risk thatthese and similar credit enhancement products will crowd out commercial lenders is to price themaccording to the risk so that borrowers (and commercial lenders) will see these forms of assistance asrelatively expensive and unappealing “last resorts.” It will also be vitally important that DBSA and otherproviders of such assistance stay in the closest possible touch with commercial capital providers so thatthey stay “down-market” of these players in a rapidly changing municipal debt market place. A creditenhancement product or business strategy that offered low risk of crowding out six months ago mayhave grave risks of doing so at present or six months from now.

Political risk insurance has been raised as a possible form of credit enhancement. In South Africa, thepolitical risk is not so much that government will pass some legislation or take some action that makes itlegally impossible for a borrower to repay debt. One obvious political risk is that through furtherconsolidation of municipalities, revenues will become less able than formerly to support debt service, insome jurisdictions. Ensuring against these risks poses two problems: (1) it may inhibit re-drawing ofboundary lines which is necessary to create viable municipalities, and (2) causation is very difficult. In agiven case, it might be very difficult to prove that the cause of default was revenue dilution as opposedto either poor municipal management or failure on the part of creditors to have insisted upon adequatesecurity at the time the loan was originated.

6.4.2.4 Transfer Payment Intercepts

This is another indirect form of financial assistance, but one that need not have any significant cost to thenational government.85 The presence in a transaction of transfer payment intercepts provides a form ofcredit enhancement. Such mechanisms may be more effective the more dependent the community is ontransfer payments from other spheres of government. For the many South African municipalities that arefinancially self-sufficient, a transfer payment intercept would be of modest credit enhancing value.However, as the national operating and capital assistance formulas are revamped and aid isconcentrated on the poorest (and often least credit-worthy) communities, this device may become amore powerful tool.86 A stream of stable, predictable intergovernmental transfers could be made

85 If the aid is to be provided anyway, making it pledgeable and interceptable does not add to the cost. Any administrative costscould be borne by the borrower. 86 Traditional rating agency analysis has given state aid intercepts only modest credit enhancing power. However, the power of astate aid intercept can be substantially increased if the state aid flow goes through a trustee-administered “lock box” arrangementin which debt-holders have first access to the revenue as a matter of course.

4 December, 1998

63

specifically pledgeable and interceptable. This would enhance creditworthiness. Significant penaltiesand/or administrative fees could discourage municipalities from relying on the intercept mechanism, andencourage them to manage their own debt payments. Such a mechanism would leverage private sectorfunding, rather than crowding it out.

6.4.3 Special intermediaries for small borrowers

International capital flows can be likened to high-tension power transmission lines. High voltages arenecessary to move power efficiently over large distances. But these transmission lines are eitherunavailable or dangerous to a small user of power (or capital). Special step down transformers areneeded to reduce the voltages to usable levels. INCA is one example of an institution that steps thevoltage down to a level that is appropriate for a broad range of users. For still smaller users, that needonly small and occasional capital flows, further step-downs may be needed. The power (or capital) canstill come from the grid. Separate facilities are not needed to generate the power (or capital), but ratherto intermediate and distribute it. The policy issue is whether some sort of special intermediary should becreated for municipalities that cannot access the market through existing intermediaries. As noted in theWhite Paper on Local Government, such special intermediaries should not replace existing commercialinstitutions, but should complement them.

Private sector representatives have indicated that they may be able to provide such intermediation,without the need to create a new agency or function for government. In principle, this is desirable, butsmall issues may not attract the market’s attention. It is recommended that the fundamental enablingenvironment (accounting, disclosure, reliable revenues, final boundaries and structure) be put in place,and that the usefulness of a government-sponsored (but not credit-subsidised) pool be evaluated a yearafter that is done.

Many kinds of intermediary models could be considered, including bond banks, bond pools, revolvingloan funds, and municipal lending institutions. Such an institution might borrow in its own name and usethe proceeds to purchase debt instruments of municipal borrowers, or it might package and securitisemunicipal debt instruments and make them directly available to the market.

For any model, there will be cost involved. These costs may include administrative costs, creditenhancement costs, or both. Properly designed, the cost will be less than outright capital grants, andwill have the virtue of helping elected local representatives and appointed local officials understand andpractice the trade-offs involved in debt finance.

The simplest kind of special intermediary could provide:

• access to capital market for those who could not otherwise access• savings on fixed costs of debt issuance• streamlined procedures• assistance with cash-flow projections• prestructured loan packages

4 December, 1998

64

In addition, as a matter of policy, the government will need to decide whether to offer direct financialassistance, such as:

• credit enhancement, e.g. in the form of debt positions or guarantees• interest rates at or below market rates

As discussed above, there are risks associated with such direct financial assistance. It is clearlyadvisable to avoid credit enhancements that reward dysfunctionality, are non-transparent, or crowd outprivate investment. However, if there were in place a stream of stable, predictable intergovernmentaltransfers for municipalities that lack the resources to be self-sufficient, these transfers could be madepledgeable and interceptable. This would enhance creditworthiness and leverage private sector fundingat little or no cost to the national government.

Technical assistance need not be provided by the same entity that provides the pooling function.

6.5 Recommendations

Key recommendations are:

1) To help municipalities access private capital primarily through technical assistance and training,which is geared to building skills and capacity. This technical assistance can be provided by anycombination of private and public entities, but could be funded by national government. Thisassistance is a form of subsidy, but one which helps borrowers and private lenders reach commonground.

2) To structure intergovernmental transfers with an eye to enabling leveraging of such transferswith private capital, and to provide that they will be pledgeable and interceptable, subject toappropriate penalties and administrative fees.

3) To approach guarantees and insurance, e.g. of specific maturities or against specific risks, withcaution. If government or the DBSA provides such guarantees, it should do so at cost, and perhapsthrough a self-financed mechanism.

4) To consider special intermediary institutions for municipalities with small borrowing needs andlimited market sophistication. In view of the private sector’s expressed interest in such pooling, it isrecommended that consideration of public institutions of this type be deferred until year after themore fundamental recommendations of this report have been implemented.

2) To avoid direct financial assistance by government (or DBSA), in the form of direct lendingpositions or interest rate subsidies.

An outline of a possible loan system for smaller municipalities is included as Addendum 6 to this report.A brief note on possible roles for the DBSA is included as Addendum 7 to this report. This note wasprepared at the request of the Department of Finance and DBSA to stimulate thinking on possibilities,but does not reflect a comprehensive evaluation of the DBSA or its roles.

4 December, 1998

65

SECTION 7 LIQUIDITY AND SECONDARY MARKETS

7.1 Background and Current Situation

Investors in any securities prefer liquidity and are normally willing to pay a substantial price (in reducedyield) for liquidity. This can be demonstrated in South African as in other markets around the world.87 Inthe past, bank loans have been assumed to be relatively illiquid debt instruments, but lenders in manycountries are increasingly interested in developing generic loan formats so that loans can be packagedand sold into the secondary market through such devices as Collateralized Loan Obligations (CLOs)and Collateralized Bond Obligations (CBOs). The team has heard interest among South Africa’sexisting and prospective municipal investors in these approaches.

The fundamental conditions necessary for a liquid secondary market include:

§ a shared interest by a significant number of participants in buying as well as selling securities (withprevailing market psychology determining the breadth and depth of that shared interest); and

§ some organised means of effecting trades (e.g. an over the counter system, an electronic exchange,an open outcall exchange, etc.).

Presently, there are few new municipal bond issues and few of the previously issued municipal bondissues are being traded. A more robust primary market for municipal short-term debt (such asoverdrafts or more formal loans of short fixed maturity) remains, although there is no evidence of tradingin short term instruments. In contrast, private sector short-term debt instruments such as bankers’acceptances do trade actively.

INCA presently remains the one important exception to the prevailing pattern of stagnation in SouthAfrica’s long term secondary municipal debt market. INCA itself is a form of liquidity for the pool ofloans it has originated. INCA has successfully floated its own senior debt and originated loans tolocalities (transactions executed as municipal bond issues purchased by INCA). INCA’s own debtdoes trade, albeit less frequently than its management and other market participants interested in owningand/or trading INCA debt would prefer.

There is a need to stimulate secondary municipal credit market activity in order to induce existing andprospective municipal credit market participants to invest given the magnitude of local government debtoutstanding, and given estimated future infrastructure financing requirements. To add additional liquidityin the short term, private financial institutions or the DBSA could package currently outstandingmunicipal debt for secondary purchasers, purchasing such outstanding stock as would be needed to 87 According to figures provided by the South African Reserve Bank (SARB), South African investors hold approximately R9billion of municipal bonds. Much of this debt was issued prior to 1989 when prescribed asset allocations were still in force, andinstitutional investors were mandated to purchase this debt. Furthermore, the central government of South Africa providedimplicit guarantees of municipal debt. Amalgamation and emphatic Government assertions that this debt is not sovereign –guaranteed have completely changed the nature of these securities in the view of virtually all existing and prospective investors.

4 December, 1998

66

generate a critical mass. If the private sector fails to act in this direction once the fundamentalsrecommended in this report are in place, government could consider undertaking such an operationthrough the DBSA.

In South Africa as in most places around the world, primary and secondary markets are so mutuallyinterdependent that any action that helps or hurts one will almost certainly help or hurt the other. Thus,policy discussions about liquidity and stimulation of secondary markets must always take cognisance ofthe broader framework for debt as a whole. However, this section will confine itself to a relatively smallnumber of issues that can be confined primarily to liquidity and the secondary market.

7.2 Key Concerns and Policy Issues

Key concerns and policy issues central to the stimulation of the secondary market include:

• Market psychology, including the factors that seem most central to the concerns of prospectivetraders in existing and/or new municipal debt;

• Continuing disclosure and the implications for holders of existing and new debt;

• Accounting for holdings by existing investors and how the “rules of the road” in this area can effectthe opportunities for trading; and

• Trading systems.

7.3 Market Psychology

“Market psychology” is a term that neatly captures the dominant perceptions of potential secondarymarket participants on the buy or sell sides. The performance of secondary markets is heavilydependent on market psychology. Market psychology can be substantially influenced by governmentaction but never completely determined by such action.

In the team’s view, much of the investor community remains anxious and sceptical about the willingnessand/or or ability of most existing or prospective municipal borrowers to pay debt service on long termobligations.88 Hence, there can be only modest improvements in secondary markets for municipal debtuntil the fundamental building blocks of a sound and stable sub-sovereign public finance system have notonly been conceived but also put effectively into place. The conceptualisation and design process is welladvanced, but implementation is still at an early stage.89 Fundamental scepticism in the investor

88 Again INCA is the notable exception, as the organisation’s leadership now believes that its market includes up to 120 potentialborrowers.89 There is sufficient objective evidence to support this element of the present market psychology. At least certain sub-sovereigncredit characteristics (e.g., local tax and fee collection rates, need for short-term borrowing to smooth cash flows, etc.) are either

4 December, 1998

67

community is responsible for the near gridlock in secondary markets. It is not likely to subside untilclear rules for municipal debt are established.

Of particular concern to secondary market psychology affecting the liquidity of outstanding bonds andloans is the uncertainty over municipal jurisdictional changes now widely anticipated in the near futureand how these changes will further affect the legal and financial status of debt issued prior to the initialamalgamation and/or debt issued between what now appear to be two stages of amalgamation. Findingways to clarify the status of both kinds of debt must be a high priority for significant secondary markettrading to occur. The most fundamental way to do this is to put hard credit culture building blocks intoplace through improvements in the legal, regulatory and institutional frameworks governing local debtissuance. It is vital that the jurisdictional changes now contemplated be made as swiftly as possible andthat these changes be permanent. (A lingering perception that further jurisdictional tinkering is possibleshould be avoided.)

Key Government decision-makers seem keenly aware of the relationship between the heath and stabilityof the overall system of local public finance and the ability of primary and secondary capital markets tofinance much of the infrastructure investment required by a citizenry with high expectations. Dramaticindications of the government’s understanding of this and related issues include the national andprovincial government’s roles in the Johannesburg debt crisis. Equally significant, but not as wellunderstood among investors and others are the many other actions of the National Intervention Programwhich have resulted in corrective instructions being issued by provincial Members of the ExecutiveCouncil (MECs) responsible for local government to local authorities. However, market participants arebecoming aware that many Instructions go unheeded and, until this is corrected, secondary marketpsychology will remain negative.

7.4 Continuing Disclosure of Information Regarding Borrowers

Secondary debt market participants need the most reliable possible methods of identifying existing andfuture risks so that they can price and allocate their capital in a risk-adjusted manner. Most emergingmarkets share the problem of little or no risk-differentiation among a variety of issuers or potentialissuers. In South Africa’s case, this lingering problem appears to be a legacy of the nation’s past use ofboth a prescribed investment regime and implicit or explicit sovereign guarantees of sub-sovereignpublic purpose debt. The accurate identification and estimation of risk is also constrained by the qualityand timeliness of information supplied by the entities being evaluated. This area needs substantialimprovement in South Africa as it does in virtually every developing sub-sovereign debt market.

What are the potential data sources for possible secondary market participants? In most organised debtmarkets, these include: continuing disclosure by the issuers themselves, independent rating agencies andgovernment agencies charged with some form of oversight for the given sector.

not improving or are continuing to deteriorate in much of South Africa, despite sustained central government attention andincreasingly rigorous collection efforts encouraged by the RSA.

4 December, 1998

68

7.4.1 Continuing disclosure by issuers

South Africa does not yet have any continuing disclosure requirements tailored to the needs of themunicipal securities market. While such requirements should be adopted in the mid- to long run inSouth Africa, it is unlikely that many municipalities could meet such rigorous continuing disclosurestandards in the near term. Instead, the team would recommend that short- to mid-term improvementsbe made using existing practices as the base. For example, it would be sensible at a minimum to requirethat any municipality with outstanding debt of more than one year’s maturity be required to publish theresults of its Project Viability questionnaire. Disclosure of Project Viability data for all municipalitieswould improve the market’s ability to distinguish good credits from bad, and would provide appropriateincentives to municipal officials and elected representatives..90 There should be a central repository forall initial issuance information, for continuing disclosure information, and for information concerningtechnical and financial defaults.

7.4.2 Rating agencies

Independent, objective and unregulated ratings of high quality are an essential component to thedevelopment of a vibrant hard credit marketplace. CA Ratings (affiliated with Standard and Poors),Fitch-IBCA and Duff & Phelps are now showing substantial interest in the municipal debt markets andat least two have developed rating contracts which provide for close on-going rating of the borrower.Each of the agencies has developed data bases of information on a substantially broader number ofmunicipalities than they have been called upon to rate and each makes an effort to recast data reportedto them into standard analytic formats. In the team’s view, the South African rating agencies have madesubstantial progress in the municipal field over the past two years. However, it also appears that not allsegments of the investor community are fully familiar with the rating resources and, hence, may not useratings agency information to facilitate secondary market trading (e.g. with use of private or "shadow"ratings, special studies and similar products). In addition, some investors expressed fundamentalreservations about the value of credit ratings in general. The former problem is likely to diminish as aresult of rating agency marketing efforts. In some cases, the latter problem will persist for the long runbecause it is borne of a fundamental distrust of the rating process world-wide, rooted in well-knowninstances in which the rating agency response to changes in credit quality have lagged behind rapidlymoving events.

7.4.3 DBSA

DBSA has also developed an internal rating system for its existing book of outstanding loans; analogousto that employed by the surveillance departments of banks and financial guarantors in fully developedmarkets. This information is not made available to other market participants or potential marketparticipants. Because this data is used directly in making credit decisions by a banking institution, itshould not be published. However, consideration should be given to finding ways that DBSA

90 There is some concern about the quality of the Project Viability data and indicators, but these concerns are more easily andquickly addressed if the information is made public.

4 December, 1998

69

information could be used in studies and analytic work without compromising the confidentiality of theDBSA’s credit decision-making process.

7.4.4 “In-house” Investor Credit Assessment Capability

In fully developed debt markets, neither institutional investors and nor guarantors rely solely on suchthird parties as rating agencies or government-sponsored monitoring programs to assess existing orfuture risk. However, in South Africa, only DBSA and INCA (and possibly a few other privateinstitutions) appear to be taking an active investor’s approach to this crucial issue and developing theirown systems for estimating and tracking changes in risk in the sub-sovereign sector. Most investorscontinue to take what appears to the team to be a more passive approach. This may reflect the marketpsychology issues discussed above.

In addition to the lingering legacy of the prescribed investment regime and implicit sovereign guarantee,there probably is a more practical reason why even some of the largest institutional investors have notdeveloped their own in-house systems for sub-sovereign risk assessment. Sub-sovereign securities, forexample, typically constitute a very small portion of each investor’s holdings and investors can beexpected to allocate scarce research resources to sectors in which their exposure is greater.

7.4.5 Project Viability

The National Intervention Programme (Project Viability) is another important means by which existingand future municipal and other sub-sovereign risks could be identified, if its findings were made availableto the investor community. The team recommends that Project Viability data be published on a regularand timely fashion.

7.5 Accounting for Investment Holdings

Full disclosure of institutional investor holdings facilitates trading and other market activity in more fullydeveloped capital markets.91 South Africa’s small and historically insular capital markets will facepressure for increased transparency as the world comes to South Africa and as South Africaninstitutions seek global credibility. Within South Africa, confidentiality and the lack of financialtransparency in normal financial reporting requirements constitute a major impediment to the successfulstimulation of the municipal credit secondary market. Only unit trusts (mutual funds) currently disclosewhat is held in their portfolios.92 The absence of requirements for full disclosure creates a market that isopaque and relatively immobile.

91 For example, in the U.S. insurance industry, full disclosure takes the form of a Schedule D filing, part of the normaldocumentation that insurance companies must annually file with their respective state Insurance Commissioners. The Schedule Ddiscloses on a line item basis all holdings of the insurance company at calendar year-end.92 This requirement may put unit trusts at a disadvantage compared to investors that are not required to disclose. Universaldisclosure by large institutional investors would remove this disadvantage.

4 December, 1998

70

Full disclosure of institutional investor holdings will bolster the credibility of rating agency judgements asthese holdings are subject to increased scrutiny by all market participants. As the credibility of ratingagency judgements increases, so will the market interest rate spread effects of these ratings.

Investor disclosure could be on a periodic basis, e.g. quarterly, with information required to be currentas of 30 days prior to filing, the focus is on a level playing field.

Although some large investors may be reluctant to disclose the extent of their municipal holdings, thegreater good is served by disclosure. If new disclosure rules were put in place, there could be someconsideration to limiting disclosure of the specifics of “old” municipal debt. For example, the aggregateamount of pre-amalgamation municipal holdings could be disclosed, rather than specific investments.

Investor disclosure allows second-tier investors to know what underlying instruments they are investedin – whether the institutional investor holding their portfolio fits their desired risk/reward points. With thecurrent capital structure so heavily dominated by banks and insurance companies, this may not seemimportant. But if the South African capital markets follow global trends and diversify, it will becomeincreasingly important to enable investors to make meaningful choices.

7.6 Trading Systems

Most or all of the institutional components necessary for a vibrant municipal bond secondary marketalready exist.93 Such existing institutional components include: The Bond Exchange of South Africa(BESA); broker/dealers; clearing systems; commercial banks; DBSA; financial planners; institutionalinvestors; issuers; jobbers; market makers; merchant banks; paying agents, and trustees.

The Bond Exchange of South Africa (BESA) is a licensed financial exchange and is charged with theresponsibility of regulating the bond market. The Exchange -- formerly, The Bond Market Association(BMA) -- was licensed in 1996, and operates within the framework of the Financial Markets ControlAct (FMCA) and a set of Rules and Directives approved by the Registrar of Financial Markets.

There are over 75 members of the Exchange including broker/dealers, issuers, insurance companies andbanks. In time, members with investor profiles may have to form separate dealer subsidiaries if theywish to continue to hold membership. The BESA market is designed for institutional investors or verywealthy individuals, because the minimum round lot is R1.0 million. Disclosure and other rulesgoverning formal listing designation are still in draft form at present. Only a handful of issuers ofmunicipal bonds are now formally listed. Most of the reported trades are in unlisted names.94

93 See Thomas H. Cochran, Michael Schaeffer & Kenneth R. Von Der Heiden, Achieving A Vibrant South African MunicipalSecurities Market, Report to the Republic Of South Africa, USAID & The Infrastructure Finance Corporation (INCA), TheUrban Institute, Washington, D.C., 199794 BESA members do report these trades to the Exchange even though the transactions were in effect executed “over the counter”(direct broker to broker trading not executed through an exchange). Listed or not, there is still a transaction fee that must be paidto Universal Exchange Corporation (UNEXcor) which is the approved clearinghouse.

4 December, 1998

71

Municipalities are not currently required to list their bond issues on the Exchange. The most recent draftof BESA’s “Listing Disclosure and Requirements” is dated October 28, 1997. Requirementscontemplated in the application process include the name of the issuer and financial instrument; the issuedate, coupon rate, payment periods, payment dates and maturity date; a copy of the resolution by thegoverning authority of the issuer authorising the issue of financial instruments; a copy of the provisions ofthe Act under which such financial instruments are to be issued and listed; a copy of any applicablegovernment guarantee in respect of the financial instruments; a copy of the placing document orprospectus; and copies of all marketing material used in connection with the original issue of the financialinstruments. The draft contains several other requirements and procedural suggestions.

Secondary market municipal bond transactions are legally required to be executed (“crossed”) throughthe Bond Exchange.95 However, the team was told that many trades are completed without the use ofBESA. In the long run, a formal bond exchange may not be the most useful medium for an efficient andactively traded primary or secondary municipal bond market. With respect to relatively small amountsof municipal debt issued by a wide variety of names, there may well be a lot of daily activity but rarely inthe same name. Such trading may ultimately prove to be more suited to the over the counter (OTC)markets as in the U. S. experience.

No technical shortcomings with trading systems explain the lack of secondary market activity. Becausesystems are in place and seem to be generally well run, there is no need for major changes in the tradingsystem. The demands of the market and the possibilities of technology will drive the development oftrading systems – government regulation of the forms of trading is not likely to be productive.

7.7 Improving liquidity of market intermediaries

Because INCA and similar institutions (if and when they emerge) provide a form of liquidity forborrowing clients, improving the liquidity of such entities’ debt instruments helps improve the liquidity ofthe municipal debt market. A number of steps can be taken to increase the liquidity of intermediaries’senior debt obligations. Such institutions can build their portfolio of investments through improvedprimary transaction origination, secondary market purchase and/or swaps, and potentially other means.Improved marketing and application procedures, a more predictive credit evaluation modelling capacity,use of new forms of structural and legal risk mitigation, increased use of revenue bonds, and use of serialbond structures for both intermediary and ultimate borrowers’ debt structures are among the methodssuch an institution can employ to build and maintain the credit strength of its portfolio.

Other steps can be taken which may improve the activity and liquidity of the broader sub-sovereignmarket in the near term. These include such actions as requirements that the valuation of securities heldin portfolios be done on a current market basis (usually referred to as “mark-to-market” requirements),improved disclosure of holdings by all classes of institutional investors, and improved information flows

95 A possible exception may be a swap between INCA and institutions using bonds in their portfolios in exchange for INCAstock that has been registered but not yet distributed.

4 December, 1998

72

to all sub-sovereign market participants). These improvements are likely to assist intermediary liquidityas they increase the activity and liquidity of the broader market.

7.8 Form of debtMunicipal securities are more tradable than bank loans. Since most of the current municipal lending is inthe form of bank loans, and much of this in the form of “structured finance,” it does not result in easilytradable stock. High interest rates in all sectors are one reason for the current emphasis on structuredfinance – as interest rates come down, the relative advantage of structure finance will be lessened. Butmore generally, as long as there are significant tax incentives to do bank lending rather than more easilytradable securities, the secondary market will suffer.

7.9 Recommendations

Key recommendations are:

1) To complete any further amalgamation or demarcation as quickly as possible.2) To improve disclosure , including

a) disclosure of Project Viability reportsb) disclosure of investor holdingsc) disclosure by borrowers as provided in Section 2

3) To consider institutional trading structures in addition to BESA.

4 December, 1998

73

SECTION 8 MONITORING, INTERVENTION AND REMEDIES

8.1 BACKGROUND AND CURRENT SITUATION

In South Africa, municipalities have the right to govern, on their own initiative, the localgovernment affairs of their community, subject to national and provincial legislation. The LocalGovernment White Paper recognises that to ensure the necessary levels of compatibility,uniformity and consistency, the national government must develop an overall framework for asystem of monitoring and oversight.96

The Constitution97 gives national and provincial government the legislative and executiveauthority to oversee the effective performance by municipalities of certain functions,98 byregulating the exercise by municipalities of their executive authority with respect to certain localgovernment matters.99 The provincial government has been given a key role in monitoringlocal government.

8.1.1 Monitoring Provisions and Practices

As previously noted, the Local Government Transition Act, No. 209 of 1993,100 (LGTA)provides for municipal budget submissions to, and monitoring by, the Ministry of Finance,paying particular attention to the maximum expenditure limits as determined by the Minister ofFinance in consultation with the Minister of Constitutional Development. Additionally, theMinister or the Member of the Provincial Executive Committee (MEC) may request anymunicipality to supply any information to him or her, a national department of State, aprovincial department, or body approved by him or her, and to determine the period withinwhich it must be supplied.101 The LGTA will expire at the end of April 1999 according toexisting constitutional provisions. However, its life will in all probability be extended until thenext local government general elections. Provision must be made to replace it with alternativelegislation in the near future.

The Ministers of Finance, Constitutional Development and the MECs tasked with theseportfolios approved the National Intervention Programme (Project Viability) on 19September 1996. The key elements of Project Viability are:

• the ongoing monitoring of municipalities;

96 Page 40 paragraph 1.3.197 Sections 139(1)(a) and (b); and 155(7). Sections 229(1)(b), (2)(b), and 230(1) provide for national regulation over certain fiscalpowers of a municipality. Section 215 requires national legislation, and section 216 gives powers to the national treasury toregulate certain financial affairs of municipalities98 Listed in Schedules 4 and 599 Listed in Part B of schedules 4 and 5, and any other matters assigned by legislation100 Section 10G(2)(e).101 Section 10.I

4 December, 1998

74

• management audits of municipalities identified by high level financial indicators as being in financialdifficulty;

• the introduction of a management support program if a municipality is unable to respond toinstructions issued by the provincial MEC;

• capacity building and training of councillors and officials; and• the introduction of a reporting framework on key financial indicators.

On November 14 1997, the Department of Constitutional Development issued regulations approved byMinMEC that require municipal Chief Executive Officers to submit regular reports containing specificfinancial indicators to their councils and Project Viability. The reports to Project Viability enable MECsto form an opinion regarding the finances of municipalities and, if warranted, justify intervention in termsof the Act. Failure to comply with the regulations exposes a municipal CEO to possible court anddisciplinary action by the municipal council.

The proposed Treasury Control Bill102 aims to set financial controls for all spheres of government,and to impose responsibilities and penalties on accounting officers and chief executive officers, as wellas their political heads.

Effective monitoring requires accessible, reliable information. The Constitution, the LGTA, otherexisting and proposed legislation and Project Viability all recognise the monitoring roles andresponsibilities of the national and provincial government with regard to municipal governance andfinance, and establish information systems to carry out these responsibilities. The existing informationsystem has been criticised, largely owing to lack of municipal management capacity and the consequentinability of many municipalities to respond meaningfully to data requests, but Project Viability managersbelieve the quality of information received from municipalities has improved significantly over the time.Uncertainty among capital market participants and government decision-makers about the extent towhich existing legislation is effective can only be addressed if national and provincial government actionis based on consistent, transparent and reliable information.

Borrowing from a competitive capital market should in itself help promote accountable municipalgovernment. In order to get access to this market, municipalities will need accurate and appropriatefinancial accounting and reporting systems. As previously noted, there should be full public disclosure offinancial information. This will increase transparency and promote public accountability and marketdiscipline.

8.1.2 Intervention Provisions and Practices

At present, intervention is governed by the Constitution and the LGTA103 which gives the MEC broadauthority to restore the finances of municipalities. Although the provincial government is primarilyresponsible for intervention, the national government may intervene together with the provincial

102 Chapters 4 and 6103 LGTA, section 10G

4 December, 1998

75

government. It may also intervene if a provincial government fails to do so, or to maintain economicunity essential national standards and national security, or to prevent unreasonable actions prejudicial toa province or the country as a whole.104

Under the Constitution,105 the provincial executive has the power to intervene when a municipalitycannot or does not fulfil an executive obligation in terms of legislation. It may take any appropriate stepsto ensure fulfilment of the obligation, including:

(a) issuing a directive to the council describing the extent of the failure, and stating anysteps required to meet its obligations; or

(b) assuming responsibility for the obligation to the extent required• to maintain essential national standards or meet established minimum standards for

a service;• to prevent that council from taking unreasonable action prejudicial to the interests

of another municipality, or to the province as a whole; and• to maintain economic unity.

If a provincial executive intervenes in this manner:(a) the intervention must end unless approved within 14 days by the MEC responsible for

local government affairs;(b) notice of the intervention must be tabled in the provincial legislature and in the National

Council of Provinces (NCOP) within 14 days of their respective first sittings after theintervention began;

(c) the intervention must end unless it is approved by the NCOP within 30 days of its firstsitting after the intervention began.

The NCOP must review the intervention regularly, and make appropriate recommendations to therelevant provincial executive.

The LGTA gives the MEC authority to take quite wide-ranging steps to restore to a state of health thefinances of municipalities that are in financial difficulty. However, time-consuming processes mustprecede intervention. The LGTA also provides, regarding unauthorised municipal expenditure, that theMEC responsible for Finance and the Auditor-General may give instructions to a council, and if theseare not implemented, the MEC may do so to restore the finances of the municipality.106

Project Viability has resulted in MECs issuing corrective instructions to municipalities that were found tobe in financial distress. Monitoring, problem identification and initial steps to restore sound managementare positive signals.

104 Constitution, section 100105 Constitution, section 139106 LGTA sections 10G(1)(k) and (m). Section 100. Section 216(2).

4 December, 1998

76

As indicated above, if a province does not fulfil its executive obligation in terms of legislation or theConstitution, the national executive may intervene. This is also limited in scope and duration, and mustend within 30 days if not approved by the NCOP.107

In March 1997, the national government announced special steps in response to municipal defaults onloans or bonds. Municipalities defaulting on loans will not be permitted to take on new loans oroverdraft facilities. Information on loan defaults will be shared throughout government agencies thathave lending facilities. These measures apparently do not apply to grants.

The Constitution provides that the transfer of funds to municipalities may be stopped if treasury normsand accounting practice are seriously or persistently breached.108

The existing procedure for intervention under section 139 of the Constitution is compatible with theconstitutional concept that local government is a sphere of government and not merely a subordinateagency.109 However, the section 139 procedure, which is confined to executive obligations, is laboriouswhere rapid and effective action is required.

The Interim Constitution (Act 200 of 1993) did not contain a corresponding section 139, and the useful– but long winded - intervention provisions contained in the LGTA were legislated under that formerConstitution. It follows that the extent to which they could be repeated in new legislation will depend ontheir constitutionality in relation to the new Constitution, particularly section 139.

8.1.3 Lenders’ legal remedies

In legal terms, lenders to municipalities can obtain court intervention in the event of default, and seek theattachment of assets. There is currently some legal uncertainty as to the scope and application of theState Liabilities Act, No 20 of 1957, in relation to municipalities as organs of state under the newConstitution.110

While existing legal mechanisms are of assistance, and have been successfully applied in some cases,lenders have expressed reluctance to use them for several reasons, including political sensitivities.

8.1.4 Registration

Registration is conceptually possible for:

107 Constitution, section 100108 Constitution, section 216(2).109 It may be noted that section 100 of the Constitution contains a similar provision in relation to national interventionin the provincial sphere, and that section 100 may authorise the national government to intervene in the provincialoversight of local government, at least where a province has repeatedly failed to effectively supervise localgovernment.110 This issue could seriously affect lenders’ remedies, and uncertainty about this issue could further limit new lending.

4 December, 1998

77

• debt issues,• security interests, such as liens or pledges,• negative covenants, and• defaults

At present, there is no structured method of easily identifying the existence of debts, security interests,or municipal defaults. Registering this information with a provincial or national depository could createaccess to objective and ready information. It could help prospective lenders identify existing liens andavoid conflicts. Lenders may justifiably require municipalities to limit future borrowing or future securitypledges, e.g. by requiring a municipality not to “ring-fence” or pledge certain identified revenues andassets. However, lenders currently have no ready method of enforcing negative covenants of this kind.A provincial or national based depository could deal with this problem by registering basic informationabout a debt issue and creating a duty to inquire. Lenders could be required to register this informationeach time they enter into a transaction, and failure to register could limit the enforceability of their rights,at least as to subsequent good faith lenders.

8.1.5 Provincial capacity

Effective monitoring and intervention by the provincial sphere of government is limited in certainprovinces by the lack of management capacity.

8.1.6 Lack of enforcement

There is a widespread perception of substantial disregard for existing constitutional and legislativeparameters for municipal finance, and almost non-existent enforcement by the private or public sectors.For example, it is common practice for municipalities to use "bridge financing" (or overdraft facilities) tofinance capital projects, despite legislation limiting this type of financing to cash flow requirements.Finance of this type is continuously rolled over; thus effectively creating long term debt from temporaryfinancing that must legislatively be paid within twelve months. This practice seems to have arisen fromthe municipalities’ need and inability to pay, and lenders’ reluctance to seek payment. Provincialintervention has not effectively addressed these matters.

8.1.7 Summary

The current system is deficient in the following areas:• lack of capacity in many municipalities to provide required reporting information;• lack of capacity of certain provincial governments to monitor information effectively;• inaccessibility of information to the general public;• lack of information about pledges and other security interests taken to secure debt, ;• lack of disclosure of financial information about municipalities, including information

relating to municipal defaults;• lack of clear remedies, and reliance on the provincial governments to effect

intervention;

4 December, 1998

78

• the cumbersome nature of section 139, and lengthy procedures under the LGTA;• reluctance of large institutional lenders to become embroiled in litigation that could be

resented by municipal residents as heavy-handed and punitive.

8.2 Decision Points

In deciding on a framework for borrowing and intervention, it is important to be clear whymonitoring is desired, and why intervention. What are the objectives to be achieved? Arethey related to municipal borrowing, which is the subject of this policy paper, or to municipalfinance more broadly?

monitoring

Should there be supervision and monitoring in the normal course of events, and if so, what kind and bywhom?

The national government has a justifiable interest in both municipal debt and municipal finances generally.The information needed to understand municipal debt and to identify potential triggers for intervention incases of borrowing-related troubles, is related to, but less than the information needed to understandmunicipal finances more broadly. Actual or anticipated failure to pay current expenses, including debtservice, is the area of commonality. Municipal debt monitoring (as is done, e.g. in Poland, NorthCarolina, and Pennsylvania) relates only to borrowing municipalities, while municipal finance monitoring(e.g. Philippines, Russia, Australia) relates to all municipalities, and may include budget and expenditurereview.

In the long term, most of what South Africa needs for municipal debt monitoring will be achieved byinitial and continuing disclosure requirements as recommended above, and by availability of audited,comparable financial statements (which is being pursued separately from this exercise). Once thistransparency is achieved, South Africa may not need any other municipal monitoring if existing lawsrequiring balanced budgets and annual retirement of short term debt are enforced.

Until such disclosure and accounting systems are in place, Project Viability is a good proxy, and resultsshould be made public.

Intervention:

Should there be detailed stages of intervention in the event of a problem? If so, what should be thetriggers for each stage, and what should happen in each stage?

Different stakeholders want different things from an intervention process. Lenders want their debt paid,ideally without bad publicity. Governments, both local and national, want services continued. These aresomewhat different goals, though there is overlap.

4 December, 1998

79

Intervention processes can be defined either by law or by contract. For a viable municipal borrowingframework, there is not, strictly speaking, a need for a detailed, statutory intervention process. Absentsuch a detailed process, the parties can define intervention and receivership processes contractually, andthose processes can be customised to the deal. However, there may be constitutional restrictions on theability of a council to contract for intervention, in which case statutory intervention rooted in theprovinces and judicial intervention may be the only constitutionally permitted avenues. And for nationaland local policy reasons beyond the borrowing framework, policy makers may want to spell out anintervention process by law or regulation.

If policy makers chose a detailed intervention process, it could be administrative and/or judicial.Triggers could be provincial or national action, municipal action, or creditor action.

Remedies

In cases of judicial intervention, what should be creditors’ remedies? This is perhaps the mostimportant decision point in the entire policy framework. Clear remedies help discipline borrowingdecisions – when actions have clear consequences, better decisions generally result.

8.2.1 Market Forces

The best option is probably to maximise the use of market forces to provide information,control access to the lending community, and coerce discipline. Information resources in theform of rating agencies and new accounting standards are evolving quickly. Some capitalaccess is being provided to the most creditworthy municipalities. Complete reliance on themarket may be premature, but care should be taken not to discourage market initiatives ordisrupt market incentives. Legal requirements that market participants disclose information willhelp the markets work more efficiently. Information systems are especially useful, as they willhelp the market to discriminate good credits from bad.

The White Paper on Local Government stresses the policy of reliance on market forces anddecentralised market relationships between borrower and lender, rather than centralcontrols.111 As noted there, this approach is in line with the decentralised orientation of theConstitution.

8.2.2 The need for a national approach

There is an issue whether each municipal or provincial government individually can regulate theproblems under consideration, or whether a wider common response is needed. A situation ofnational concern has been created by the extreme financial situation of many municipalities,

111 White paper on Local Government, Section G.3.1, page 122.

4 December, 1998

80

uncertainty as to the ability of many to meet their obligations, the concerns of the lendingcommunity as to existing municipal debt, and the inability of most municipalities to accesscapital.

National consistency when it comes to the exercise of intervention powers is thereforeimportant for stabilising the relationship between municipalities and financial markets. Thenational impact that municipal defaults may have on the local government capital market alsosuggests a broad-based solution. In addition, the scale of the problem may be such thatnational resources must be dedicated to efforts to resolve them. It is obviously in the nationaleconomic interest to have financially viable local government. Most of these concerns arereflected in section 44(2) of the Constitution, and weigh strongly towards stronger nationalinvolvement.

For these reasons, it is recommended that national legislation and guidelines on the process ofintervention should be developed. For constitutional reasons, the role of the provinces willneed to be respected.

8.2.3 Monitoring and intervention: constitutional considerations

The present understanding and practice it that provincial governments are designated tomonitor and intervene where appropriate. This is based on section 139 of the Constitution.The national government’s current role is essentially limited to acting after failure by aprovincial government.

The philosophic or political rationale for the present structure of intergovernmentalrelationships is understood. However, there is an issue whether section 139 by itself isadequate to the present circumstances of municipal indebtedness. Some additional processmay be needed. The Section 139 process is restrictive, lengthy and cumbersome. It is notwell suited in terms of process or content to deal with the urgent and unique issues of individualmunicipal indebtedness. Other options should be explored.

In this respect, it should be noted that the provision112 for provincial intervention relates tofailure by a municipality to perform an "executive obligation". It could be argued that failure tomeet financial liabilities is not simply failure of an “executive obligation”. A financial obligationthat takes the form of a debt is also a financial, moral, and contractual obligation, and mayrequire some additional process.

The Constitution appears to empower the national government to legislate in this area.113 Ifthis is correct, the door is open for effective and appropriate intervention in addition to theprovisions of section 139. In other words, national legislation may establish and empower one

112 Section 164113 Constitution, section 139(3)

4 December, 1998

81

or more authorities to deal with municipal indebtedness and the facilitation of access bymunicipalities to the capital market. This is especially so where the authority in question wouldoperate at the instance and under the supervision of the courts, whose jurisdiction extendsacross all spheres of government. .

8.2.4 Intervention: its manner and extent

The diagram below is intended to illustrate four stages of addressing troubled borrower-investorrelationships.114 Briefly stated, the stages are:• Where the borrower and investor have a contractual relationship that resolves any issues that may

develop, outside intervention is not needed.• Where the parties’ contractually specified structures fail, “Phase I” remedies are invoked, i.e.

provincial instructions and support.115

• Where “Phase I” succeeds, no further intervention is needed, but where it fails, “Phase II” remediesare invoked, i.e. a court appoints a receiver or financial control board. This includes an exhaustivelist of possibilities, from mild to drastic.

• Where even this intervention does not work, there is no alternative but liquidation of non-core assetsand discharge of debt.

These possibilities are discussed in more detail below.

Borrower toInvestor/LenderRelationshipsdesigned to enhancesecurity of the debt

Phase I Remedies:ProvincialInstructions andsupport

Phase IIRemedies:Court appointedReceiver/FinancialControl Board

MunicipalBankruptcy:Liquidation ofNon-coreassets andDischarge ofDebt

Doesnot

work

Doesnot

work

Doesnot

work

See DiagramA

Succeeds in >90+%of cases

See DiagramB

Succeeds

See DiagramC

Succeeds

Contractual relationships and remedies

Outside intervention is seldom needed where the parties to a borrowing transaction have structuredadequate provisions for security, and provided for various contingent remedies in the event of difficulty. 114 A note on vocabulary: there may be a single investor, e.g. a bank lender, or there may be many investors that have boughtmunicipal bonds. In this section, the term “investor” is intended to refer to both possibilities115 For the purposes of this scheme of creditor remedies, the provincial intervention process is “Phase I” intervention, and may beinvoked by creditor complaint. For the purpose of Project Viability, provincial intervention may also be triggered by high levelindicators that monitor municipalities’ financial health.

4 December, 1998

82

The policy framework recommended herein places primary reliance on borrower-investor relationshipsdesigned to enhance the security of the debt and to provide for resolution of difficulties. Diagram Abelow shows some of the possibilities:

• The parties may provide that some of the revenues that would otherwise go directly to themunicipality would be set aside, escrowed, or placed into a “lock-box” arrangement. Only afterdebt service and any reserve fund requirements had been fully satisfied would remaining revenues bepaid over to the municipality’s general find. Reserve funds, sinking funds, and rate covenants areother voluntary structures that can be created to ensure things go smoothly. The parties could alsoprovide that where the investors are bondholders, a trustee will act on their behalf to enforce bondcovenants, and that paying agents and registration agents to pay bondholders and keep track ofthem, increasing bondholders’ confidence.

• The parties may provide that in the event of a dispute, arbitration, mediation, or other forms ofalternative dispute resolution will be employed in an attempt to settle the outstanding issues. Theymay also create notice and negotiation mechanisms designed to forestall anticipated defaults beforethey occur.

• The parties may provide for “private” intervention, e.g. that an accounting firm or other operatingentity would take control of some portion of municipal operations in the event the municipalitydefaulted on a payment or on specified bond covenants.

Diagram ATypical Negotiated Arrangement

BORROWER

INVESTORS/LENDERS

REVENUE SOURCESRates

TariffsIGTsetc.

TrustarrangementsSinking fundsEscrows

Arbitration, mediation, ADR

Privateintervention

4 December, 1998

83

Provincial oversight and instruction

For the relatively rare instance when outside intervention will be required, this reportrecommends two phases of different character. Phase I is basically the provincial oversight andinstruction model now used in connection with Project Viability. When key performanceindicators raise concerns, or upon complaint by a creditor, the MEC would initiate an audit ofthe financial situation, operations, and credit control procedures of the municipality. Arising outof this, he or she could issue appropriate instructions to the council, including time frames.116

Diagram B below illustrates this process, showing the initial trigger, the audit, and theinstruction process. The three boxes to the right (management support program, capacitybuilding and training, and reporting framework) are not part of the creditor remedy processper se, but are ongoing elements of the current Project Viability process. Where the provincialinstruction process solves the problem, no further action is necessary. The province has nopower to stay the execution of creditor remedies provided by law or contract, though it cansuggest voluntary rescheduling or forbearance by a creditor.

Financial Control Board or Receiver

Because the province has no power to stay creditor remedies including execution on securityinterests, resort to the courts can be had at any point by:

116 It may be appropriate to provide a procedure for municipalities to request advice and guidance from the province prior to anyformal triggers.

DIAGRAM BPHASE I REMEDIES: PROVINCIAL INSTRUCTIONS AND

PROJECT VIABILITY

Ongoing Monitoringof High-levelIndicators(Project Viability)

Creditor Reports toProvincial Authorityeither Early WarningSigns or Late Payments

CompleteAudit/MECDecision

ProvincialInstructions

ManagementSupportProgramme

CapacityBuilding andTraining

ReportingFramework(OfficialstoCouncils)

Problemsolved

Problemsolved

4 December, 1998

84

• a creditor who believes the provincial oversight and instruction process is not adequatelyprotecting its rights, so long as the municipality remains in default of a payment obligationor other covenants, or

• a municipality that seeks a stay of execution or other equitable relief from a creditor’sclaims or actions to enforce a covenant or foreclose on security.

Thus, Phase II of the remedy and intervention process would be judicial action117, which couldinclude a court-appointed receiver or control board able to invoke a wide range of powersunder court supervision, as the situation warrants. These powers would include (generallyfrom the least intrusive to the most):118

a) Authority to approve any expenditures made by the municipalityb) Authority to establish a budget and a strategic planc) Authority to manage municipal staff, including hiring and firing where appropriated) Authority to take custody of and to operate municipal property, plant, and equipmente) Authority to establish tax ratesf) Authority to establish utility tariffsg) Authority to collect and control all revenues due to the municipality, from individuals, businesses,

other municipalities, or other spheres of governmenth) Authority to liquidate non-core assetsi) Authority to investigate and establish responsibility for mismanagement or fraud, and to make

recommendations for prosecuting offenders.j) Authority to petition the court for removal from office of any elected representative or appointed

official deemed to have been negligent in the performance of their public dutiesk) Authority to do any other thing that could have been done by the appointed officials or elected

representatives of the municipalityl) Authority to petition the court for suspension of labour agreements, where such suspension is

unavoidably necessary for the financial health of the municipalitym) Authority to petition the court for bankruptcy protection, including the rescheduling of loans and

the discharge of debt, where it is not reasonably possible for debt to be paid.119

117 The judiciary, as an institution, cuts across all three spheres of government and avoids any question of constitutionality.118 It should be noted that most, but not all, of these powers are also available to a province as part of the Phase I provincialintervention process, under which the province can do most of what the municipality itself could have done.119 National municipal insolvency legislation should be developed that is sufficiently severe to the parties, to persuadethem to seek the insolvency remedy a last resort, and not as a management tool. Increasing tariffs and taxes, andimproving collection of both, should be considered long before discharge of debt.

4 December, 1998

85

The upward arrows in the diagram are intended to illustrate feedback loops. Thus, thereceiver or board would start with relatively mild actions, e.g. approving expenditures,assesses the impact, and decides whether more strict measures are needed. More drasticmeasures, e.g. removing elected representatives or appointed officials from office, can only bedone by court order. The receiver would report back to the court periodically. The court’srole would be supervisory, not administrative.

The court-appointed receiver could be an individual or a specially-appointed financial controlboard with appropriate management expertise.120 The proposed Municipal Borrowing Billwould provide that absent bad faith or fraud, the receiver or board would be immunised fromlegal liability for the difficult decisions he, she, or it would likely have to make. And it wouldspecify that the extent of intervention should be appropriate to the context, and should bebased on the least intrusive means of dealing with the situation. A primary objective must beto restore democratic municipal governance as quickly as possible.

As with any court process, the court would have power to craft equitable remedies asappropriate. Thus, where the court finds that a well-managed provincial oversight andinstruction process is underway, and that a creditor’s rights are adequately protected, it maydecline to intervene further. In addition, where appropriate, it may grant a stay of execution ofa creditor’s remedies at the request of the municipality or province if it determines that theharm done by such a stay is warranted by the facts of the case.

120 E.g. a local business person.

DIAGRAM C

PHASE II REMEDIES: ACTIVE MANAGEMENT UNDER COURTSUPERVISION

Court Appoints Financial Control Board/Receiver

Financial Control Bd./Rec’r Assesses Situation, Takes Actions

Multiple RemediesEmployed Singlyor in Combination

4 December, 1998

86

Experience throughout the world shows that work-out situations require effectivecommunication and hands-on, active management, with the co-operation and involvement ofas many stakeholders as possible. The intervention process should provide an incentive forthe municipality and its creditors to work out a settlement, as well as the opportunity for themunicipality to take corrective measures to restore its viability.

Bankruptcy

In the very rare circumstance where the situation simply cannot be worked out, there is littlealternative but for the court to liquidate non-core assets, settle outstanding creditor claims foras many cents on the rand as possible, and discharge the remaining debt, and the proposedMunicipal Borrowing Bill should so provide. Such fundamental failure will likely reflect anunderlying problem that the municipality is simply not a viable unit of local government. Thequestion of disestablishment or disincorpororation then becomes relevant, but such governancequestions are properly addressed in the upcoming Systems Bill, not in the proposed MunicipalBorrowing Bill. The legal status of local government as a sphere of government and an organof state may influence the manner in which this issue is finally dealt with.

8.2.5 Creditors’ remedies

The legal ability of municipalities to contractually agree on creditor remedies is unclear. There aredifferent views on whether a contract to raise tariffs, e.g. would be enforceable. There is not likely to beclarity until a body of case law under the new constitution has emerged.

The following legislatively prescribed remedies and protections are therefore recommended, in additionto whatever the parties may agree:

a) General creditors (including holders of existing debt) could be given:i) right to intercept funds due to municipalities from other spheres of governmentii) right to trigger imposition of an additional tax within the defaulting jurisdictioniii) right to trigger the appointment of a receiver to control expenditures and or operations of

the municipality, as described below.

b) Citizens would need protection for “minimum essential” municipal services, narrowly defined,e.g.:i) water needed for health and safetyii) sewage operationsi) refuse collection

At any time, including during Level One intervention, creditors would have the right to apply to court forexecution on their security interests and for judicial intervention. As part of a Level Two intervention orseparately, the court would be empowered to deal with insolvency and priority as between creditors.

4 December, 1998

87

The court would be empowered to discharge debt as a last resort where the municipality could nototherwise be made solvent.

8.3 Recommendations

Key recommendations are:

1) To continue Project Viability monitoring until the disclosure systems recommended in Section 2and good accounting processes are established.

2) To develop the current provincial instruction process as the first level of intervention for troubledmunicipalities.

3) To provide a second level of intervention, in which the municipality or creditors have the right torequest the courts to appoint a receiver or financial control board where the provincial instructionprocess is unable to deal with the situation.a) A court-appointed receiver should have broad powers.b) Minimum essential services should be protected

4) To provide that general creditors with clear, simple remedies against defaulting municipalities, e.g.:a) The right to intercept funds due to municipalities from other spheres of governmentb) The right to trigger imposition of an additional tax within the defaulting municipalityc) The right to trigger the appointment of a receiver or financial control board , as described

below.5) To protect “minimum essential” municipal services, narrowly defined, e.g.:

a) water needed for health and safetyb) sewage operationsc) refuse collection

6) To provide clear legislation governing municipal insolvency, includinga) creditors’ rights and prioritiesb) when and under what conditions debts and other municipal contractual obligations

should be dischargeable

4 December, 1998

88

SECTION 9 MACROECONOMIC ISSUES

9.1 Municipal Revenues

9.1.1 Background

The Constitution establishes national, provincial and local government as autonomous spheres that are"distinctive, interdependent and interrelated." The Constitution requires that nationally raised revenuesbe divided equitably between the three spheres of government.

This equitable division is to provide each sphere of government with the necessary funds to provide theservices and perform the functions assigned to it under the Constitution. This division is done after takinginto consideration each sphere's ability to raise its own revenues, the expenditure limits set by thegovernment to meet its stated macro-economic objectives and a careful weighing of the functions of thedifferent spheres of government. These considerations are then matched with the affordability constraintsset by the size and growth of the South African economy. After the vertical division has been done thelocal government share is horizontally divided among local authorities according to a formula. Theformula is designed to enable municipalities to deliver a package of basic services to low-incomehouseholds.

9.1.2 Types of municipal revenues

An understanding of municipal revenues is important in evaluating municipal capacity to service debt.South Africa’s local government financing system is under review and likely will undergo somesignificant changes in the next few years. Should the system become more centralised, then theframework recommended in this report will require significant adjustment. For example if localgovernments were less able to make discretionary decisions about committing resources to therepayment of debt, their ability to do autonomous borrowing and capital planning would becompromised. The assumption in this report is that the outcome of the current reforms will be a localgovernment sector with substantial fiscal discretion, i.e., with substantial ability to determine revenueflow. This assumption is consistent with the overall tenor of the recommendations in this report.

The framework for municipal borrowing needs to be flexible enough to accommodate changes in thefiscal system. The move to decentralisation may be gradual, and comprehensive local taxing powersmight not be in place for a few years. Boundary changes may affect which sphere of government orlevel of local government has access to which revenue sources, and both local governments and lenderswill have to get accustomed to the new realities of local government finance. Expenditure assignmentmay change, and this needs to be factored into the creditworthiness evaluation of local government.One hopes for rapid improvements in local government fiscal information systems, and a steep learningcurve in the management area. Local government discretion as regards the borrowing decision mayneed to be limited for certain types of local governments, although the current problem is moreavailability of long term credit than abuse of long-term borrowing.

4 December, 1998

89

The current revenues of municipalities can be classified as follows:

a) Tax based revenues include:Regional Services LeviesTaxes on propertyTaxes on goods and services (user charges)

b) Non-tax revenues include:Property incomeSurplus/deficit on trading accountsParking and garage feesInterestAdministrative fees and chargesRent: general and housingFines and forfeitsSewerage and cleansing feesLibrary feesMetro transport fees

c) Other revenues include:Endowment feesReceipts from other spheres of government for agency services rendered.

Capital revenue can be classified as follows:

a) Sales of assetsb) External and/or internal loansc) Capital transfers121

Current transfers may consist of the following:a) Transfers from other national departments (Water Affairs and Housing)b) Intergovernmental transfers122

c) Private sector: housing

121 Capital transfers were rationalised and restructured into the consolidated municipal infrastructure program (CMIP)in 1996, and phased in during 1997. CMIP provides grant funding for the installation, upgrading and rehabilitation ofmunicipal internal bulk and connector infrastructure, while the rural water infrastructure grant funds community waterand sanitation projects.

122 The intergovernmental transfers are defined as current operating transfers from national government to local government. Upuntil the end of the 1997 financial year, operating transfers to municipalities were made by means of intergovernmental grantsfrom provincial budgets, agency payments and implicit subsidies from national and provincial departments. The grant systemused was inequitable, inconsistent and not based on objective policy criteria. A transparent, formula-based system was thusintroduced and is operational as from the 1998/99 municipal financial year. The new system will be phased in over a period offive years for urban municipalities and seven years for rural municipalities.

4 December, 1998

90

9.1.3 Analysis

The following table sets out recent developments in local revenue and expenditure, as well asprojections to FY2000/01. These figures do not include trading service expenditures like electricity andwater provision.

Outcome Preliminary ProjectionsR billion 1994/95 1995/96 1996/97 1997/98 1998/99 1999/00 2000/01REVENUE: 17.6 20.5 21.4 24.3 24.4 27.0 28.9Tax revenue 6.7 7.4 7.5 8.2 8.8 9.5 10.0Non-tax revenue 7.4 8.7 8.9 9.8 10.4 11.5 12.5Capital revenue 0.3 0.3 0.4 0.4 0.5 0.5 0.6Grants 3.3 4.0 4.6 5.9 4.7 5.5 5.9

EXPENDITURE 18.5 21.3 22.2 24.3 25.9 28.4 30.2Current exp. 12.8 16.1 16.1 17.5 18.7 20.6 21.8Goods & services 11.7 14.1 15.0 16.3 17.4 19.3 20.4Interest 1.1 1.7 1.0 1.1 1.2 1.3 1.3Capital exp. 5.1 4.6 5.5 6.1 6.5 6.9 7.4Net lending 0.5 0.6 0.6 0.7 0.8 0.9 1.0Deficit/surplus -0.9 -0.8 -0.8 0.0 -1.5 -1.3 -1.3

Revenue % GDP 4.0 4.1 3.9 4.0 3.6 3.7 3.6Exp. % GDP 4.2 4.3 4.0 4.0 3.9 3.9 3.7Deficit % GDP -0.2 -.02 -0.1 0.0 -0.2 -0.2 -0.2GDP 444.9 497.3 556.0 613.0 669.3 734.3 809.6Source: Budget Review, 1998

In the foregoing classification, intergovernmental grants that are part of CMIP are counted as capitaltransfers, and intergovernmental grants that are part of the equitable share are counted as currenttransfers. The difference in classification is a little odd, since the amount of the transfer comes out ofcurrent national revenues in both cases. CMIP transfers are not capital transfers in the OECD sensethat national or provincial government are transferring an illiquid capital asset to municipal ownership, orselling a capital asset to receive revenue. This is more than an accounting issue, and raises the questionof why capital grants should be so strictly separated from operating grants. Arguably, this interfereswith local autonomy. If there is a restriction (legal or perceived) against using equitable share funds forcapital, this would discourage investment. It may be that more people can be better served by investingin a capital asset with equitable share funding rather than wasting money operating inefficient andoutdated facilities. Modernisation should not be discouraged. In practice, the distinction can be difficultto maintain and the current classification does not necessarily prevent current transfers from going toinvestment.

There appears to have been a substantial drop in capital spending in 95/96 as compared to 94/95, from28% of total spending to 22%. This may be due to new administrations’ weak management of capitalprogrammes and outgoing administrations holding off from capital projects.

4 December, 1998

91

9.2 Outstanding municipal debt

Commencing in January 1997 the Department of Finance initiated a project to determine the totaloutstanding debt of municipalities. Addendum 4 contains quarterly data for the period March 1997 toMarch 1998. The data for March 1998 is, however, at present still subject to possible revision.

Two surveys are conducted by the Department on respectively a quarterly and monthly basis wherebyinformation is requested from insurers and pension funds regarding outstanding loans (quarterly) anddefaults123 by municipalities (monthly). The Registrar of Banks requires banks to report on outstandingloans and defaults by municipalities as part of their quarterly reporting requirements to the Registrar. Thesurvey was supplemented by data obtained from the Reserve Bank on outstanding stock debt,Department of Housing, Public Investment Commission, Local Authorities Loan Fund, DevelopmentBank of Southern Africa and INCA. In order to preserve comparability estimates were made wheredeemed necessary.

After an initial surge during June 1997, mainly as a result of increased funding by banks, total debt ofmunicipalities remained fairly stable for the remainder of the period under review. Although theaggregate debt did not change significantly, despite the deteriorating financial position of municipalities ingeneral, changes did occur in the sources and categories of funding, as noted in section 8 above.

9.2.1 Sources of municipal funding

The local government debt market in South Africa is in a state of transition. The data for the periodMarch 1997-March 1998 is summarised in a table attached to this report as Addendum 4. As ofMarch 1998, an estimated R25 billion in outstanding municipal debt consisted of R14.4 billion in debt topublic lenders and only R7.5 billion in debt to the private sector. In the public sector, aside from R 9.1billion in housing loans, the biggest portfolio is held by the Development Bank of South Africa (DBSA).

The DBSA’s involvement in local government sector lending was increased with transfer of LALF loansto it as of 1 January 1998. Total loans to municipalities increased from R2 670 million at the end ofMarch 1997 to R3 528 million (32 per cent) on 31 March 1998. It would seem that as funding fromthe other public sector institutions becomes less readily available, the DBSA is expanding its role as asource of finance to local government.

Banks and insurers are the main private lenders to municipalities. The private sector's involvement inmunicipal debt declined by 8 percent from R10 658 million in June 1997 to R9 778 million in March1998. Bank loans dominate the limited supply of private sector credit. At the end of the period inquestion, banks held some R1.4 billion (including tax structured financing obligations) and had suchshort-term holdings grow by roughly 50% over the year. Funding by banks during the quarter endedJune 1997 rose by R1 502 million but banks then reduced their exposure over the following nine

123 Currently no consensus view exists of what would constitute default. The Department of Finance is working on a definitiontogether with other stakeholders in Municipal Finance issues.

4 December, 1998

92

months. Insurers and pension funds recorded a gradual but persistent decline in their role throughout theperiod under review. There has also been a marked shift toward INCA, which has seen its loanedamount increase from R257 million to over R700 million in the space of the year in question. Mutualfunds (unit trusts) do not appear to be a significant source of municipal lending.

Netting out the sectors among the long-term and short-term changes, it appears that the following isoccurring in the bond market:

• Banks are substituting direct loans to municipalities for securities and have increased their netholdings, largely as a product of their greater short-term lending. This has largely offset the declinein lending by insurance companies.

• The only other net new provider of private credit is INCA.

The general disinclination of private investors to hold municipal securities is confirmed by the data andby interviews. Aside from INCA there are no new listings of municipal bonds on the Bond Exchange.Virtually all private sector lending appears to be on the basis of direct negotiation with financialinstitutions and through bank loans, overdrafts, call bonds, and bankers’ acceptances. Holders ofexisting debt are reluctant holders and not interested in extending more long-term credit except in thestrongest cases where significant municipalities represent valued customer relationships for reasons otherthan their debt quality.

Once the fundamentals identified in this report are in place, there is still reason for concern about thesupply side of municipal capital markets. Finances in South Africa are remarkably concentrated andnon-diverse. At the end of March 1998, banks held R 578,9 billion of assets and long-term insurersheld R 497,5 billion. No other capital aggregations come close. There is cause for concern about“groupthink,” whereby a relatively small group of decision-makers convince themselves that themunicipal sector is not worth their attention. With a larger and more diverse assortment of potentiallenders, the market would certainly work better. It is difficult to know how much diversity is needed foreffective competition, i.e. for one entity to take risks the others are not taking, make profits, and thusbreak any herd mentality. If banking deregulation allows off-shore competitors effective access toSouth African markets, diversification and more effective markets are likely to follow.

9.2.2 Categories of funding

Financing through the issue of securities (of which the largest component - 70 percent - is marketablestock)124 continued to be a main source of outstanding financing for municipalities. However, there is adecline during the last three-quarters of the period under review, notwithstanding the increased

124 “Marketable stock” refers to stock that is quoted on the Bond Exchange or for which transfer registers are normally maintainedby the issuing municipality (according to the SARB a Central Depository institution has been developed in this regard).“Securities” as used in Addendum 4 refers to marketable, non-marketable and other securities (that are neither short- nor long-term).

4 December, 1998

93

importance of INCA as a source of funding. The impression is gained that most investors are notconverting to new issues when existing issues mature.

Funding by the National Department of Housing125, representing over 32% of short-term credit,constituted the biggest portion of credit extended to municipalities. Long and short-term loans providedby banks rose sharply during the June quarter, but subsequently recorded a generally downward trend.This was in contrast to the increase recorded in long-term loans by the DBSA, to which reference wasmade above.

Tax-structured finance, on the basis of data provided, continued to play a relatively insignificant role infunding throughout 1997 and the early part of 1998.

9.3 Demographics

Urban municipalities, including metropolitan municipalities, have easier access to funding for municipalprojects than rural municipalities. Loans granted by financial institutions, as reflected in the foregoingdata, cover the whole spectrum of local government.

The same underlying framework for debt is recommended in both urban and rural municipalities. Thebasic problem of converting current revenue into capital infrastructure is the same for all municipalities. Itis a given that local revenues are in transition, and that there will be a long-term difference in revenuesources between urban and rural municipalities. While urban municipalities will largely generate theirown revenues, many rural municipalities will likely depend on intergovernmental transfers. In eithercase, the framework should enable local officials to use some of their current revenue stream to supportborrowing. Where management sophistication is limited, private or public intermediaries could offer an“off-the-shelf” system, with clear, simple options for terms and amounts. See the model suggested inAddendum 6.

This approach avoids the need for central government to provide all of the capital needs of ruralmunicipalities – as with urban municipalities, the private sector can provide capital finds. This approachalso helps rural municipalities gain the management skills and perspectives they will need if they are todevelop their own revenue sources. Using the same basic approach for rural and urban municipalitiesavoids systemic disincentives.

125 The National Housing Act, 1998, extinguishes the debt owed by Municipalities to the National Department of Housing. Thiswill have a distorting effect on total local government debt outstanding, reducing the overall debt to about R16bn.

4 December, 1998

94

Demography: Preliminary estimates of urban and non-urban population by province126 (percentage)1996

RSA WesternCape

EasternCape

NorthernCape

FreeState

KwaZulu-Natal

NorthWest

Gauteng MpumaLanga

NorthernProvince

UrbanpopulationPercentage

Non-urbanpopulationPercentage

55,4

44,6

89,9

10,1

37,3

62,7

71,7

28,3

69,6

30,4

43,5

56,5

34,8

65,2

96,4

3,6

37,3

62,7

11,9

88,1

Based on the 1991 Census Statistics, the total urban population of South Africa was 17 551 745 andthe total non-urban population was 13 435 175. At that time it was still divided into the four provinces,self-governing territories and development regions.

9.4 Should There Be Tax Incentives to Make Borrowing Cheaper?

This section considers the question of tax incentives to encourage investment in municipal infrastructure.These incentives could be applied to municipal debt, to debt incurred to provide municipal services(whether of a public or private entity), or to savings and investment generally. South Africa’s IncomeTax Act No. 58 of 1962 contains a host of similar exemptions from taxable income for activitiesdeemed socially desirable,127 a number of deductions from income for capital investments deemeddesirable,128 and a range of other tax incentives.129 Although these privileges are embedded in existinglaw, adding new exemptions, deductions, or other incentives would likely be controversial.130

126 ‘Urban’ includes areas with some form of local authority, as well as areas of an urban nature without any form of localmanagement. All other areas are classified as non-urban. Residents of informal settlements immediately adjacent to the boundariesof a town are counted as ‘non-urban’. Non urban areas are all other areas not included under ‘urban areas’ or semi-urban areas’.127 These include: scientific, technical or industrial research under sections 10(1)(cA)(i)(aa) and 10(1)(cB)(i)(bb); providingcommodities, amenities or services to the State or public under sections 10(1)(cA)(i)(bb); promoting commerce, industry, oragriculture under section 10(1)(cA)(i)(cc); providing certain medical services under section 10(1)(cB)(i)(bb); nature conservationor animal protection under section 10(1)(cB)(i)(cc); activities of a cultural nature under section 10(1)(cB)(i)(dd); erection ofhousing under section 10(1)(cC)(i); providing residential accommodation for the aged under section 10(1)(cF); protecting andrehabilitating the environment under section 10(1)(cH); land development under section 10(1)(cI); and promoting and facilitatingthe distribution of agricultural products.128 See section 11 of the Income Tax Act, which provides such deductions for acquisition of machinery, plant and building,development and acquisition of trademarks, scientific research, the erection of employee housing, and the rehabilitation andprotection of the environment.129 E.g. section 12C regarding manufacturing operations, section 18A regarding educational funds, and section 37H regarding taxholidays for certain industries.130The Katz Commission has recommended eliminating existing incentives. This seems unlikely to happen quickly.

4 December, 1998

95

9.4.1 Background

An issue that seeks to be addressed is an issue of granting tax incentives or tax-exempt status on the‘coupon’ or interest earned by holders of municipal bonds. For example, in accordance with theAmerican Internal Revenue Service tax system, specifically section 103D of the Internal Revenue Code,interest on these instruments is usually exempted from federal income tax liability. This has enabledmunicipalities’ access to local capital markets and permitted them to borrow funds at a lower rate ofinterest than those prevailing in the commercial market place.

In South Africa, tax-structured arrangements (e.g. certain leasing arrangements) betweenknowledgeable lenders and sophisticated municipalities have provided a de facto tax benefit to somelenders. Unlike the American model, this arrangement benefits only a relatively large and sophisticatedlenders and borrowers. Such transactions account for a significant portion of the relatively limited newprivate lending to municipalities. Although the Department of Finance is generally aware of this situation,it has not been able to stop the practice.

Two questions arise:

1. At what cost to the fiscus does tax-incentivised municipal lending occur?• To the central sphere of government (due to an increase in compliance costs), and• to taxpayers who will face tax increases in one form or the other as the tax exempt status

translates into revenue foregone which must be collected elsewhere2. To what extent is this cost offset by the ability of the tax exemption to attract private investment in

municipal infrastructure?

9.4.2 Undesirable Shifts in Tax Burden vs. Desirable Leveraging

A risk is that tax relief for holders of municipal debt (who may be the more affluent taxpayers) willtranslate into higher tax burdens that will finally settle on less affluent taxpayers who do not have surplusfunds at their disposal to make these investments in the first place. Hence, the affluent will benefit whilstthe poorer sections of lower to middle income groups will face a higher tax burden as Government’srevenue needs must still be met at the end of the day.

However, this may be offset to the extent that new investment capital is attracted to public infrastructureinvestments that would not otherwise have been made. In other words, Government’s revenue needsmay be significantly reduced if the foregone tax is effectively leveraged into private investment capital forlocal infrastructure. If each rand of new private investment costs government only the marginal incometax foregone on the interest, that can be effective leveraging.131 The key question is whether and towhat extent the investment would have occurred but for the tax advantage. 131 Assume a 30% marginal tax rate for the taxpaying bondholder. Assume a 15% annual interest rate on the municipal debt.Each 100 rand of private money invested costs the central fiscus 4.5 rand per year that the debt is outstanding. This isconcessional finance by another name, because the government is effectively writing down the cost of borrowing, at a cost of 4.5rand per 100 rand invested. Some would see this as effective leveraging.

4 December, 1998

96

The Income Tax Act 58 of 1962 grants a general tax exemption on the first R2 000 of interest earnedby natural persons. This is arguably a very small amount, but from an affordability perspective it isunwise to raise this maximum limit further without a careful analysis of any offsetting benefits.

9.4.3 Decision points regarding tax incentives

Arguments for tax incentives include the following:

• Tax exemptions can stimulate the flow of investments by the private sector. Especially recently,South Africa has had very limited private investment in municipal infrastructure. Whatever theirdownsides, tax exemptions for municipal infrastructure have proven remarkably effective instimulating private investment in municipal infrastructure in the U.S.132

• Tax exemptions are consistent with allowing decentralised and market-compatible decisions aboutcapital allocation. In other words, they subsidise all capital investment by municipalities, or allmunicipal investment infrastructure by any party (depending on the option chosen). Local councilsdecide for themselves which capital projects are a priority, and investors impose market disciplineby investing only in creditworthy projects. Moral hazard issues inherent in other concessionaryfinance approaches are largely avoided.

Globally, fiscal experts are sceptical about tax incentives.133 In fact, both the OECD and the EuropeanCommission currently attempt to convince their members and also non-member countries that most taxincentives erode jurisdictions’ respective tax bases. Other arguments against implementing taxincentives or allowances can be summarised as follows:

• Tax concessions require higher tax rates and affect competitiveness: As noted above, taxpreferences narrow the income tax base, and these revenue losses may need to be offset with highertax rates (to the extent revenue requirements are not reduced by the effects of the tax incentive).High tax rates can render a country less competitive as an international investment destination.

• Tax incentives can be inequitable in granting preferential treatment to a few sectors, therebyburdening the other sectors with higher tax rates. For example, if municipal debt for infrastructurewere tax advantaged, but not private debt for municipal infrastructure, the public sector would havea competitive advantage in providing services to citizens.

An advantage to this approach to concessional finance is that the administrative costs are relatively small, and the market does theinvestment allocation, rather than a bureaucracy.132 Approximately one third of all municipal debt is held by individual taxpayers, and another third is held by taxpayers throughmutual funds, money market funds, and closed end funds.133 There is one interesting exception. It was, until recently, the position of the U.S. Treasury to recommend that sovereignsecurities not be given any tax advantage. However, since most such securities are held in book-entry form, tax compliance isvery effective when compared to alternative investments. Therefore, to put sovereign securities on a “level playing field,” theTreasury is now considering recommending preferential tax treatment for sovereign securities.

4 December, 1998

97

• Tax incentives can distort relative prices, leading to misallocation of scarce resources:municipalities will gain access to cheaper funds on the capital markets. This will make possiblemore municipal investments in infrastructure projects, at the expense of operational spending.134

• Tax incentives tend to distribute the tax burden unequally across different categories oftaxpayers. The principle of horizontal equity is compromised in the process. Some see taxincentives as subsidy payments to ‘preferred taxpayers’135 and not as an outlay reflectingexpenditure program intent. The concern is that these preferences could create a situation whereinfluential groups of taxpayers (probably the more affluent ones) pay less than their fair tax bill.Since different taxpayers have different levels of income, they will not benefit to the same extent.

• tax incentive schemes facilitate arbitrage activities. Controls on arbitrage would needed if taxadvantages for municipal infrastructure investment were implemented. Monitoring costs wouldincrease.

• It is hard to know whether a particular assistance should be delivered through the tax systemor by outright subsidies. If tax incentives are chosen, efficient budget program planning wouldsuggest that these programmes should be made as transparent as possible. Thus, a programbudgeting process could quantify such tax expenditures as “outlay equivalents.”136

9.5 Recommendations

Key recommendations are:

1) To focus attention on municipalities’ need for stable, predictable, sufficient revenues. Without suchrevenues, municipal borrowing will not be viable.

2) To approach tax privileges with caution.a) If new privileges are to be pursued, a study should be undertaken to evaluate the leveraging

effects of any tax privilege against the revenue losses due to a tax exemption in the specific caseof South Africa.

b) Any tax privilege should be available equally to holders of municipal debt and to holders ofprivate debt incurred as part of a project to provide public infrastructure.

134 The present “equitable share” program may unintentionally have the opposite effect: promoting current spending at theexpense of investment.135 The question is who really gets the benefit of tax advantages such as an interest exemption on debt for municipalinfrastructure. Certainly the rate offered to the bondholder by the borrower is dramatically and demonstrably lowered by the taxexemption. The taxpayer’s effective rate of interest is usually comparable to the effective rate he or she would receive on taxableinvestments. The majority of the benefit is received by the borrower, in the form of lower borrowing costs. 136 International experience accordingly suggests that expenditure planning achieves a higher efficacy if subsidy programmes aredesigned by appropriation committees rather than by tax committees that have no experience and expertise in structuring andplanning outlays. Some therefore argue that tax committees should limit their activities to the design of an equitable tax systemand should minimise the erosion of the comprehensive income tax base concept.

4 December, 1998

98

SECTION 10 IMPLEMENTATION

Implementation of the policy recommendations contained in this report will require a broad-based effortwith the support of those stakeholders in every sector that are interested in seeing active private capitalmarkets in South Africa.

The most essential implementation step is that a comprehensive Municipal Borrowing Bill be drafted andadopted. An outline for such legislation is included as Addendum 2. This outline describes the legislativepackage that forms the backbone of a new system for municipal borrowing.

In addition, a number of other implementation steps are necessary or advisable, including regulations,voluntary industry guidelines, and disclosure of available information. These other steps are discussed inAddendum 3.

Finally, various supportive interventions could be productive in stimulating an active private municipalcapital market. These are also listed in Addendum 3, but have not been developed in further detail as apart of this report. It is recognised that such efforts could require significant resources. Financialresources might need to be sought from the international donor community if not available domestically.

A systemic approach to the issue (including a co-ordinated program of all of the above elements) wouldbe ideal if resources can be found.

4 December, 1998

99

ADDENDUM 1

Interviewees

ANDERSON, Gerry – Head Financial MarketsFinancial Services BoardPO Box 35655,Menlo Park, 0101

BALFOUR, IanJP Morgan2 Maude StreetSandton Square

BOTHA, Div – Executive ManagerDevelopment Bank of Southern AfricaPO Box 1234Halfway HouseMidrand, 1685

BORAINE, AndrewCape Town City CouncilCivic Centre12 Hertzog BoulevardCape Town, 8000

BOTHA, Reiner – Manager: Retirement Funds and Friendly SocietiesFinancial Services BoardPO Box 35655Menlo Park, 0102

CANTER, AndrewRMB Asset Management (pty) Ltd.1 Merchant Place1 Fredman DriveSandton, 2146

COMBRINCK, J.Helderberg MunicipalityPO Box 3Strand

4 December, 1998

100

Cape Town, 8000

CUNNIFF, John -- CEORustenburg District CouncilPO Box 1993Rustenburg, 0300

DE CLERK, SageSouth African Reserve BankPO Box 427Pretoria, 0001

DITSHEGO, Tladi -– Evaluation Specialist : Operations Evaluation UnitDevelopment Bank of Southern AfricaPO Box 1234Halfway HouseMidrand, 1685

DLAMINI, Cleopatra – Project & Infrastructure FinanceInvestecPO Box 785700Sandton, 2146

DOUW, JohnSALGAPO Box 31Alabama, 21547

FLEMING, LindaEskom Pension FUND

GANTSHO, Mandla – Executive Manager (Finance)Development Bank of South AfricaPO Box 1234Midrand, 1685

GAULT, Ronald T. -- Managing DirectorJP MorganPrivate Bag X9936Sandton, 2146

GEBELA, S.Pietermaritzburg Transitional Local Council

4 December, 1998

101

GOLDIN, Dr. IanCEO and Managing DirectorDevelopment Bank of Southern AfricaPO Box 1234Halfway HouseMidrand, 1685

GOUGH, David -- Project ManagerWetton-Lansdowne-Philippi Corridor ProgrammeCape Town City CouncilPO Box 298Cape Town, 8000

GROBLER, Nick– Senior Manager Public Sector FinanceABSA Bank

GROTE, Martin – Tax Policy DirectorateThe Budget OfficeDepartment of FinancePrivate Bag X1115Pretoria, 0001

HEYMANS, ChrisManager: PolicyDevelopment Bank of Southern AfricaPO Box 1234Halfway HouseMidrand, 1685

HEYWOOD, DaveSouth Peninsula MunicipalityPrivate Bag X19TokaiCape Town, 7945

JACKSON, BarryDevelopment Bank of Southern AfricaPO Box 1234Halfway HouseMidrand, 1685mailto:[email protected]

4 December, 1998

102

JELLIMAN, NoelFirst National Bank

JONES, Allen -- Market OperationsBond Exchange of South AfricaPO Box 351Auckland Park, 2006

KELLERMAN, PankieGENSECTyger Park 4BelvilleCape Town, 8000

KING, DaveDuff & Phelps Credit Rating Co of Southern Africa (Pty) LtdPO Box 76667Wendywood 2144

KOLKER, Joel -- Housing & Urban Development DivisionUnited States Agency for International DevelopmentPO Box 55380Pretoria, 0007

KOTZE, DeonOostenberg MunicipalityPO Box 35BrackenfeldCape Town, 8000

KOTZO, M.Richards Bay

KRUGER, L.Inner West Local CouncilKUMAR, KrishNorth/South Control Local Council Durban

LABUSCHAGNE, LappiesMines Pension Fund

LAWRENZ, ArnoGENSEC

4 December, 1998

103

Tyger Park 4BelvilleCape Town, 8000

LE ROUX Jéan-Pierre -- Analyst – Sub-NationalsFitch IBCAPO Box 1985Randburg, 2125

LE ROUX T.A.Pietermaritzburg

LEVELL, MicRAND MERCHANT BANK

LEVER, Peter DirectorTreasury and Corporate ProjectsCorporate FinancePO Box 655Cape Town, 8000

LOMBARDC, Basie – Executive Officer: FinanceGreater Johannesburg Transitional ServicesMetropolitan Infrastructure and Technical ServicesPO Box 32243Braamfontein, 2017

MABILETSA, Edwin M.Mankwe Village Local Council Tribal Administration

MacDOUGALL, AndrewJP Morgan2 Maude StreetSandton Square

MALEKA, Mandla -- Director: Municipal Finance PolicyDepartment of FinancePrivate Bag X115Pretoria, 0001

MALI, LincolnPublic and Development Finance ForumBanking Council

4 December, 1998

104

PO Box 61674Marshalltown, 2107

MANN, JillTreasury – Structured ProductsRand Merchant Bank Limited1 Merchant PlaceSandton 2196

MANCHE, JackieDepartment of Constitutional Development87 Hamilton StreetPretoria, 0001

MARAIS, StephISCOR Pension Fund

MOLATUDI, S.A.Mankwe ZONE 4 Village Local CouncilLerome

MOLLENTZE, James -- Credit ManagerINCAPO Box 1153Johannesburg, 2000

MORRISON NEIL – Head of Global MarketsDeutsche BankPrivate Bag X9933,Sandton, 2196

MOSETE, J.M.Mankwe Zone 4 Liaison Officer

MVELASE, S.A.Uthukela Regional Council

NAICKER, VanessaIlembe Regional Council

NAIDOO, D.Inner West Local Council

4 December, 1998

105

OPPERMAN, LucasFormer Strategic Manager, FinanceGreater Johannesburg Metropolitan Council

PILANE, John M.Saulspoort Village Local Council Zone 4

PITYANA, NkulieEaston Grant:

PLAISTOWE, Tom– Assistant Manager (External Investments)Old MutualPO Box 878Cape Town, 8000

RENNIE GORDON – Manager : Financial MarketsFinancial Services BoardPO Box 35655,Menlo Park, 0102

ROOPNARAIN, R.NLC & IMFO

SCHENCK, PhilipCape Metropolitan Council44 Wale StreetCape Town, 8000

SCHOEMAN, Herman – Head: Short Term InsuranceFinancial Services BoardPO Box 35655Menlo Park, 0102

SCOTT, EddieBlaauwberg MunicipalityPO Box 35MilnertonCape Town, 8000

SEGODI, Samuel M,Mankwe Zone 4 Village Local Council

SMITH, Kobus (JG)

4 December, 1998

106

ABSA Group

SOMERS, Oscar – Specialist UnitDevelopment Bank of Southern AfricaPO Box 1234Halfway HouseMidrand, 1685

STEENKAMP, GertDepartment of FinancePrivate Bag X115Pretoria, 0001

STEPHENSON, ChrisABSA Group

STRYDOM, A.MSouth Local Council

TSHIDI, Dube – Head: Pension DepartmentFinancial Services BoardPO Box 35655Menlo Park, 0102

TSHITE, L.M.J.Mabodisa Village Local Council

TURREL, MikeDurban Metropolitan CouncilPO Box 1514Durban, 4000

VAN DEN BERG, J.C.Uthukela Regional Council

VAN DER MERWE, Piet --Deputy Director Municipal Finance PolicyDepartment of FinancePrivate Bag X115Pretoria, 0001

VENTER, Hennie – Executive Manager: FinanceKhayalami Metropolitan CouncilPrivate Bag X017

4 December, 1998

107

Kempton Park, 1620

VENTER, Louis – Head of public Finance Division (Economics Department)South African Reserve BankPO Box 427Pretoria, 0001

WANDRAG, DaanImali Capital9 Arnold RoadRosebank

WHITE, RolandDepartment of FinancePrivate Bag X115Pretoria, 0001

WHITTLE, ChristopherABSA Group

WILCOCKS, C.T.Ugu Regional Council

WINTERBOER, LeonPublic and Development Finance ForumBanking CouncilPO Box 61029Marshalltown, 2107

WORT, RichardCorporate Finance DepartmentRand Merchant BankSandton

YORKE, AlanDeloitte & TouchePO Box 243Durban, 4000

4 December, 1998

108

ADDENDUM 2

Proposed Municipal Borrowing Bill

1) Scope/Applicabilitya) The Bill would apply to borrowing by municipalities and by any quasi-municipal entities

subsequently authorised to borrow under the terms of the bill. This anticipates and allows laterlegislation authorising, e.g. special districts, to incorporate the Bill’s provisions by reference.

b) The Bill would apply to all borrowing by municipalities, whether by bank loan, overdraft, orissuance of municipal securities.

2) Limitationa) The bill would provide that any borrowing by a municipality shall be the obligation solely of that

municipality. It would provide that neither provincial nor national governments, nor their officialsor representatives, may guarantee or pay any borrowing obligation of a municipality.

3) Authorisationsa) The Bill would authorise municipalities to borrow short term (i.e. for less than one year) no more

than the amount required:i) To bridge operating cash shortfalls in anticipation of specific and realistic future income

streams to be realised within the fiscal year, andii) to bridge capital requirements in anticipation of specific and realistic grants to be received or

debt to be issued within the fiscal year.b) Short term debt must be paid off annually, and remain paid off for some reasonable period of

time. Both borrower and lender are responsible for enforcement, and criminal penalties apply incase of violation.

c) The Bill would authorise municipalities to borrow long term (i.e. for more than a year) for capitalinvestment in property, plant, and equipment to be used for public purposes.

d) Council must authorise any long or short term borrowing by resolution stating the purpose of theborrowing.i) Long term borrowing must be specifically authorised in each instance by councilii) Short term borrowing may be authorised by instance or by a general authorisation that the

chief executive borrow up to stated limitsiii) Prior notice of a proposed borrowing resolution must be given, e.g. by publication at least 2

weeks in advance of consideration in a newspaper of general circulation in the municipalityiv) Borrowing resolutions must be adopted in open session of council

e) Municipalities must invest borrowed funds until they are needed for their authorised purpose aswould a prudent person with due regard for preservation of capital. Regulations to bedeveloped by the Department of Finance would define this “prudent person” rule further,specifically limiting the use of borrowed funds to invest in leveraged, unhedged derivativepositions.

4 December, 1998

109

f) The Bill would authorise municipalities to pledge their assets and revenue streams to securedebt, and to agree in debt covenants on methods by which creditors may realise on suchpledges, provided that:i) Assets required for “minimum essential public services” (to be defined in the bill, or perhaps

in regulations to be adopted by the Department of Finance) could be pledged, but thepledge would be subject to the requirement that any lien-holder or successor in possessionof such assets must continue to use such assets in providing essential public services untiland unless the assets are determined by the council or a court of law to be no longernecessary or useful for such purposes.

ii) Any pledge of public assets would require an affirmative determination by resolution ofcouncil that such assets are either not required for minimum essential public services, or areadequately protected.

g) The Bill would authorise municipalities to pledge tax rates, utility charges, and other fees andcharges to secure debt, and to covenant to maintain such rates, fees, and charges at specifiedlevels to secure debt.

h) The Bill would authorise municipalities to contract with creditors for:i) payment mechanisms such as escrows, “lock boxes,” and set-asides or voluntary revenue

interceptsii) special or limited obligation security arrangements, such as revenue debt or “ring-fencing”iii) alternative dispute resolution, such as arbitration and mediationiv) special interventions, such as private management, in the event of default, andv) such other arrangements as the council deems necessary and prudent to secure borrowing.

4) Disclosure:a) The provisions of this section would apply to any transactions in municipal debt instruments,

whether through an exchange, over the counter, or otherwise.b) The Bill would make the following conduct a criminal violation and actionable in a civil suit:

i) Any fraud in the offering, sale, or purchase of a municipal debt instrument,ii) Any untrue statement of a fact material to an investment decision, or omission of a fact

material to an investment decision, in connection with the offering, sale, or purchase of amunicipal debt instrument

iii) Any act, practice, or course of business which operates or would operate as a fraud ordeceit upon any person, in connection with the purchase or sale of a municipal debtinstrument.

c) The Bill would make the disclosure of material facts an affirmative duty of electedrepresentatives, appointed officials, lenders, brokers, underwriters, and other marketparticipants. All market participants should have a duty of reasonable inquiry in respect of factsand events material to an investment decision.

d) The Bill would require the filing of:i) a disclosure statement or prospectus before any long term municipal debt is incurred. An

exception should be provided if debt instruments will not be sold to retail investors.

4 December, 1998

110

“Sophisticated investors,” i.e. wealthy individuals and institutional investors should beallowed to trade among themselves without triggering filing requirements. 137

ii) periodic updates to the disclosure documentation. This requirement should be drafted sothat it can be satisfied by the filing of audited annual financial statements.

iii) additional disclosure upon the occurrence of an event material to an investment decision.e) The Bill would authorise the FSB to adopt implementing regulations, which could be written by

the FSB itself or by a self-regulatory organisation under FSB supervision:i) further specifying acts of fraud and deceit, and defining specific facts and events material to

an investment decision.ii) Establishing a repository for disclosure documentation, and establishing fees for filing and

accessing disclosure documentation. Such a repository could be privately operated underthe supervision of the Financial Services Board (FSB).

iii) Defining “sophisticated investors” who can hold and trade municipal debt without triggeringfiling requirements.

iv) Establishing a registry of municipal debts, security interests, and defaults, if the need arises inthe future.

5) Executive obligationa) In the Bill, municipalities would be declared to have executive obligation138 to:

i) Disclose as aboveii) Register debts, defaults, and liens as aboveiii) Pay debt service in accordance with their debt covenantsiv) Honour other covenants incurred in connection with debt issuancev) Pay off short term debt annually

6) Intervention and Remediesa) If a creditor has reasonable cause to believe that a municipality will default on its debt

obligations or is insolvent, the creditor may notify the provincial MEC, and provincial audit andintervention shall be in accordance with the constitution, Section 139

b) If a municipality defaults on its debt obligations and a creditor believes the provincial interventionprocess is not adequately protecting its rights, then at the instance of a creditor, a court isauthorised to:i) Intercept for the benefit of creditors any un-pledged funds due to the municipality from other

spheres of government, and to impose appropriate penalties and fees in connection withsuch intercept

ii) Impose a tax within the defaulting municipality for the benefit of creditors,

137 The exemptions appropriate to South Africa must be carefully crafted. Generally, short term loans and direct placements suchas bank loans or other direct loans from a small number of sophisticated investors would be exempt. Exempt bonds could not bere-offered to the general public, and any subsequent trading would be only between sophisticated investors. Any exemption fromthe material event disclosure rule should be approached with great caution, since material events can affect the quality of both longterm debt held by sophisticated investors and short-term debt held by any investor.138 This is to help clarify the right of provinces to intervene under section 139 in the event of a failure by the municipality tocomply with these obligations.

4 December, 1998

111

iii) Appoint a receiver or financial control board, provided that:(1) The court should select an individual or group with appropriate management expertise,(2) Absent bad faith or fraud, the receiver or financial control board shall be immune from

legal liability taken pursuant to the authority vested in it by the court,(3) The court should determine what level of intervention is appropriate, selecting the least

intrusive methods it believes to be necessary in the circumstances.(4) A primary objective of the intervention process should be to restore democratic

governance as soon as possible,(5) The court may authorise the receiver or financial control board to:

(a) approve any expenditures made by the municipality(b) establish a budget and a strategic plan(c) manage municipal staff, including hiring and firing where appropriate(d) take custody of and to operate municipal property, plant, and equipment(e) establish tax rates(f) establish utility tariffs(g) collect and control all revenues due to the municipality, from individuals, businesses,

other municipalities, or other spheres of government(h) liquidate non-core assets(i) investigate and establish responsibility for mismanagement or fraud, and to make

recommendations for prosecuting offenders.(j) petition the court for removal from office of any elected representative or appointed

official deemed to have been negligent in the performance of their public duties(k) do any other thing that could have been done by the appointed officials or elected

representatives of the municipality(l) petition the court for suspension of labour agreements, where such suspension is

unavoidably necessary for the financial health of the municipality(m) petition the court for bankruptcy protection, including the rescheduling of loans and

the discharge of debt, where it is not reasonably possible for debt to be paid.139

c) A municipality may itself apply to court where it seeks legal or equitable relief from a creditor’sclaims or actions to enforce a security interest, and may request the appointment of a receiver orfinancial control board.

d) In any judicial or court-supervised process, the court and the receiver or financial control boardshall use every reasonable means to assure that minimum essential services to the community arecontinued.140

139 Separate municipal bankruptcy legislation is recommended, which may also deal with the governance issues ofdisincorporation or disestablishment of non-viable municipalities.140 It is recommended that the Municipal Borrowing Bill or Department of Finance regulations narrowly define minimum essentialservices, e.g. “such water, sewer, and refuse collection as is necessary for human health and safety.” This definition would applyboth to the provision of the Bill that assets relating to minimum essential services could be pledged, provided that any creditor orsuccessor in interest in possession of such assets continues to use them in providing essential services, and to the obligation thatthe court and receiver or financial control board shall use every reasonable means to assure that services are continued.

4 December, 1998

112

ADDENDUM 3

Regulatory implementation steps

1) Disclosure Regulations

It is recommended that FSB adopt issuer disclosure regulations as authorised in the proposed MunicipalBorrowing Bill (MBB). One useful starting point is South Africa’s existing corporate disclosureregulations, but publicly traded municipal securities are fundamentally different than corporate securities.Recent experience in Poland has reconfirmed the need to treat these separately.141

The disclosure regulations would give life to the MBB’s “material information” disclosure requirement.The regulations would be general in nature, to allow room for industry practice to develop. The specificinformation included in any given transaction would be determined by the parties.

The regulation could be drafted by the FSB itself, with appropriate input from affected parties, or itcould be drafted by an industry-sponsored self-regulatory organisation (SRO) supervised by the FSB.

It is recommended that the regulation include:• A description of what is considered material information.• A description of who is obligated to make disclosure142

• A requirement that market professionals involved in the transaction certify that they believe thedisclosure to be true and complete.

• A requirement that the municipal officials and elected representatives certify that they believe thedisclosure to be true and complete.

• Requirements for periodic and “material event” disclosure• A definition of a “material event” for purposes of continuing disclosure• A description of what constitutes a “sophisticated investor” for whom full disclosure would not be

required.• A description of what disclosure documentation is to be filed, where, and how (i.e. creating the

repository).143

• A description of how disclosure documentation will be accessed and disseminated, and what feeswill apply.

141 Expert advice from the U.S. or Europe may be useful here.142 For example, in the U.S. market, a materially obligated party is any party obligated to pay at least 15% of the aggregate debtservice. It is therefore possible to have several obligated parties. Special provisions are made for enhancers and guarantorsallowing them to disclose by reference to information already on file.143 Ideally, the same stakeholders involved in the drafting of the regulation and the promulgation of guidelines would serve in anadvisory capacity in the establishment and operation of the repository. A newly established repository should be able to leap-frog the development process experienced in other countries and start with technically advanced receipt, storage, and retrievalsystems.

4 December, 1998

113

2) Disclosure and Documentation Guidelines

It is recommended that IMFO or some other appropriate body be encouraged to adopt disclosure anddocumentation guidelines, and to revise them periodically.144 It is further recommended that theseguidelines be written with the input of all affected parties. It is important that the committee draftingthese guidelines not be dominated by any one interest, and that drafts be widely circulated andthoroughly vetted.145 It is necessary to strike a reasonable balance between investor protection andcosts of compliance. These guidelines would provide guidance on at least three primary topics:• The design of the primary offering document (i.e. prospectus or official statement), including what

information should be included in various sections• The design of the current reporting document, the content of which would be specified at the time of

the initial offering, and the subsequent periodic provision of which would be a bond covenant orcontractual obligation of the municipality

• Recommended procedures, interpretations, and vocabulary which would represent acceptedstandard practice and which would supplement the rules enacted by the FSB.

These guidelines would serve as a checklist for what should be included in various documents and theirorder of presentation. Although standard formatting is not absolutely imperative, it does assist users inlocating information within the documents. There guidelines could also include guidance on the financialinformation to be included, geared to the type of security offered. For example, a revenue bondsecured on enterprise activity will disclose information concerning service area, supply factors, pricingpolicies, regulatory oversight, and so forth. The relevant market and operating factors will tend to beunique to the various types of revenue-producing activities, such as water/sewer, electric powergeneration and distribution, toll facilities, special taxing districts, public markets, and so forth.146 In thecase of self-supporting activity bonds, excerpts of engineering and marketing studies or entire reportsare included. There is also discussion of the financial statements to be included and related matters suchas ratings, enhancements, and the role of various professionals of the transaction. These disclosuredocuments are likely to be the most comprehensive and objective description of a municipality’sfinancial structure and situation.147

The guidelines can also be used to give procedural guidance. They can document “best practice”conventions in the municipal securities market. Such matters as the delivery of official statements todealers, incorporation of underwriting data in official statements or prospectuses, notification of trustees,

144 Expert technical assistance could also be useful in creating these guidelines.145 For example, in the US, the 1991 guidelines were reviewed directly by a committee that included some 30 individuals from allbranches of the market. Approximately 800 copies of the draft guidelines were circulated to interested parties, and 23 providedwritten comments. The Government Finance Officers’ Association (roughly equivalent to IMFO) acted as the secretariat forthese proceedings and published the resulting guidelines.146 Appended to the Disclosure Guidelines in the U.S. is a checklist of operating and financial performance data that are specificto the type of enterprise that was compiled by the professional organization of municipal credit analysts, one of the constituentparties to the development of the guidelines in the U.S.147 Although US municipalities are not subject to line-item disclosure requirements , they are liable under the anti-fraudprovisions, as are the professionals involved in the transaction. This legal exposure leads to a high level of accuracy and candor inmost documents, especially where risk elements are present

4 December, 1998

114

call notices, transfer agent and settlement dates, and the like were placed in the guidelines. Providedthat the sponsorship and authorship of the guidelines is broad-based, informed by experts from allsectors, and authoritative, defining such conventions and standards in one place is of great value tobuilding an efficient, rules-based market.

Supportive interventions

1) Project Viability Data

It is recommended that the Department of Finance release Project Viability data for all municipalities.The project questionnaire and data base contains at least one question that calls for a judgement orpersonal opinion of the respondent . Such responses might well be kept confidential to encouragefrankness, but the overwhelming majority of the information is relevant and useful. Experience aroundthe world is compelling that the quality of the data and the incentives for good management will bothimprove if this information is subjected to public scrutiny.

Until South African municipalities are generally reporting their financial positions in a timely fashion andin compliance with the new accounting standards, this is the single best source of information forcreditors and potential creditors.

2) Training and support

It is recommended that public and private training and support programs for elected representatives andmunicipal officials should be inventoried. Where they are inadequate, they should be systematicallyaugmented, revised and expanded. Training and support is especially critical in the following areas:

a) Municipal accounting and compliance with new accounting standardsb) Operational and capital budgetingc) Capital planning and debt strategyd) Project analysise) Disclosure and marketingf) Managing troubled finances

Without effective training and sustained development of human resources, the policy frameworkdescribed in this report is likely to take a very long time to bear fruit.

It is recommended that such specialised training materials as are developed with public or donor fundingbe made available to all training providers. An IMFO certification process could help raise the generallevel of municipal finance expertise. Practical, hands-on mentoring and support activities are asimportant as formal training opportunities.

4 December, 1998

115

3) Model debt policies

It is recommended that IMFO or another appropriate body be encouraged to adopt model debtpolicies for municipalities.

4) Pilot bond issues

Government could consider approaching international donors who might bear the cost of developingseveral pilot issues of tradable securities which, in the aggregate, constitute a large enough volume tostimulate a secondary market. The learning curve will be significant in preparing the first several issuesdesigned to comply with the new disclosure regulations. If this initial hurdle were overcome with outsideassistance, replication of that process by the private sector would be less costly.

5) Public education

It appears that there is substantial progress being made in establishing healthy municipalities in SouthAfrica, but that neither the public nor the financial community is fully aware of the situation. Forexample, Project Viability, the DBSA and private creditors are all successfully helping municipalitieswork out troubled debts. The good news of these successes has by and large not caught up with thebad news of defaults.

6) Off the shelf borrowing for small and emerging municipalities

Public or private efforts to develop small-scale lending for small, rural, and emerging municipalities couldhelp make the leveraging effects of borrowing available to all.

7) Intergovernmental transfers

Intergovernmental transfers could be specifically structured with an eye to enabling the leveraging ofsuch transfers through borrowing, and allowing the pledge and interception of such transfers for thebenefit of creditors, subject to appropriate penalties and fees in the event of involuntary intercepts.

4 December, 1998

116

ADDENDUM 4

Summary of municipal debt March 1997 – March 1998

R Millions

LENDER

Mar-97 Jun-97 Sep-97 Dec-97 Mar-98 Mar-97 Jun-97 Sep-97 Dec-97 Mar-98 Mar-97 Jun-97 Sep-97 Dec-97 Mar-98 Mar-97 Jun-97 Sep-97 Dec-97 Mar-98 Mar-97 Jun-97 Sep-97 Dec-97 Mar-98 Mar-97 Jun-97 Sep-97 Dec-97 Mar-98

1. NATIONAL

GOVERNMENT

1.1 Housing 9,095 9,073 8,912 8,907 8,900 9,095 9,073 8,912 8,907 8,900

1.2 Other depts 200 200 200 200 200 200 200 200 200 200

Subtotal 9,295 9,273 9,112 9,107 9,100 9,295 9,273 9,112 9,107 9,100

2. FINANCIAL

INSTITUTIONS

2.1 Public sector

2.1.1 DBSA 1,735 1,824 1,924 2,495 3,528 1,735 1,824 1,924 2,495 3,528

2.1.2 LALF ** 935 937 930 909 0 935 937 930 909 0

2.1.3 PIC 1,983 1,940 1,842 1,830 1,818 1,983 1,940 1,842 1,830 1,818

Subtotal 1,983 1,940 1,842 1,830 1,818 935 937 930 909 0 1,735 1,824 1,924 2,495 3,528 4,653 4,701 4,696 5,234 5,346

2.2 Private Sector

2.2.1 INSURERS 3,195 2,780 2,759 2,650 2,600 203 180 191 178 183 59 60 24 20 50 75 25 3,531 3,115 2,975 2,828 2,783

2.2.2 PENSIONS 114 82 156 153 149 36 36 34 32 35 48 50 5 5 5 203 173 195 185 184

2.2.3 BANKS 1,353 1,658 1,159 996 990 7 42 142 98 112 748 1,400 1,424 1,222 1,222 897 1,382 1,402 1,376 1,400 137 163 143 157 149 3,142 4,645 4,270 3,849 3,873

2.2.4 INCA 257 492 590 637 709 20 20 20 19 19 277 512 610 656 728

Subtotal 4,919 5,012 4,664 4,436 4,448 246 258 367 308 330 875 1,530 1,444 1,241 1,241 926 1,407 1,407 1,376 1,400 187 238 168 157 149 7,153 8,445 8,050 7,518 7,568

3. LOCAL GOVT 855 813 813 811 811 855 813 813 811 811

INTERNAL FUNDS

Subtotal 855 813 813 811 811 855 813 813 811 811

4. OTHER*** 1,930 2,213 2,224 2,298 2,210 1,930 2,213 2,224 2,298 2,210

Subtotal 1,930 2,213 2,224 2,298 2,210 1,930 2,213 2,224 2,298 2,210

5. TOTAL 9,687 9,978 9,543 9,375 9,287 1,181 1,195 1,297 1,217 330 11,905 12,627 12,480 12,843 13,869 926 1,407 1,407 1,376 1,400 187 238 168 157 149 23,886 25,445 24,895 24,968 25,035

TOTAL OUTSTANDING DEBT OTHER SHORT TERM LOANS TAX STRUCTURED FINANCE SECURITIES ANNUITY LOANS OTHER LONG TERM LOANS

4 December, 1998

117

ADDENDUM 5

Interest & redemption as a percentage of income

-

20

40

60

80

100

120

0 4 8 12 16 20 24 28 32

PERCENTAGE

NOOFLOCALAUTHORITIES

4 December, 1998

119

ADDENDUM 6

An off the shelf loan system for small municipalities

As envisioned in the White Paper on Local Government, assume a predictable flow of IGT revenuefrom the national or provincial government for those unable to carry out their assigned functions withown source revenues.

A special intermediary could help small municipalities with:

• capital planning -identification of capital needs -prioritisation of capital needs -analysis of revenues that capital projects could generate -analysis of O&M that capital projects will require -evaluating alternative packages (from a few simple alternatives)

• public participation

• bond issuance and supporting documentation

Micro-bonds issued through the intermediary could be in a standard format, in standard amounts, andwould contain a pledge of IGT and own-source revenue streams sufficient to cover debt service andpenalties.

The intermediary could consolidate groups of similar maturities and sell them to institutional investors.They might be sold at a premium or a discount, depending on current market conditions and whether asubsidised interest rate is desired.

4 December, 1998

120

ADDENDUM 7

“What should be the role of DBSA with regard to municipal borrowing?”

This is a question that has come up repeatedly during the preparation of this report. It is a differentquestion than “what concessional finance or other assistance should there be?” The latter questionassumes a tabula rasa. As noted in the body of the report, it is not recommended that anygovernmental or quasi-governmental entity, including DBSA, be designed or encouraged to act as ageneral provider of municipal credit. In a sustainable municipal borrowing system, such a role is best leftto the private markets. Governmental and quasi-governmental lenders run the risk of lending toborrowers that either should be receiving credit (and the accompanying financial discipline) from privatesources or should not be borrowing at all. If a governmental or quasi-governmental entity does assistmunicipalities in accessing private capital markets, such assistance should be surgical and designed tograduate borrowers fully to private markets as soon as practical.

DBSA appears to have two fundamentally conflicting missions: one the one hand it is to be self-sufficient, which means that it must not spend down its capital, and on the other it is to fund socialdesirable investments for which private capital is not available. As a recent DBSA publication notes,“[p]erforming a developmental role while remaining a viable financial institution poses a complexchallenge.” If DBSA lends to market-capable borrowers, it can point to its need to preserve capital. Ifloans go bad, it can point out that they were desirable social investments, and of course that comes atsome risk.

South African policy makers have two potentially conflicting desires: they would like to create asustainable framework for municipal borrowing from private sources, but also want to invest visibly andquickly to alleviate the accumulated backlog in demand for services. It is hoped that this report’sarticulation of that which is necessary for the former objective can help assure that efforts to achieve thelatter objective can be fashioned in a way that minimises the potential conflict.

This paper outlines a framework for municipal borrowing that envisions market-based interactionsbetween a pool of lenders and a pool of borrowers. As a practical matter, neither pool is very big rightnow. The lending community in South Africa is relatively small and tightly knit. It has largely come tothe view that the municipal sector is risky. The pool of qualified potential borrowers is also quite small.Few municipalities have done the capital planning needed to identify their needs for loan financing, andmany municipalities have been too busy coping with the stresses of amalgamation to consider long termstrategic needs. In a way, this paper proposes a set of rules for a game which will attract very fewplayers at the moment. Such players as there are now tend to be large, well-organised cities, such asDurban.

Given the basic premise of market-based borrowing, if DBSA is true to its first mission, it will lend tothe large, well-organised cities at market rates and preserve its capital. It is one more lender, providingmarket-based financing on the same sort of risk-reward calculus as other lenders, and is a welcomeaddition to the market . The more potential players, the more efficient the market place.

4 December, 1998

121

But if DBSA is concessional lender, funding projects that cannot attract private capital, it will sooner orlater spend down its capital as a portion of those projects go bad. And by lending at rates that do notreflect the risk involved, it could inject an addictive flow of low cost money into the system, thwartingmarket discipline. And in the meantime, there is the risk that it will crowd out private sector interest.

There are a variety of roles and activities that a quasi-governmental lender like DBSA could undertake.Not all of them involve direct lending. In general, to the extent that these activities are self-financing,they can do no harm and may be valuable. To the extent they are not self-financing, they are notsustainable and may cause more harm than good.

Some of DBSA’s potential roles are:

1) Technical assistance and training: DBSA has skills in project analysis. Making those skillsavailable through assistance to market participants on either the borrow or lender side is compatiblewith a market based capital model. Though there is some risk of crowding out other t.a. andtraining providers if the services are not priced at cost, the needs are so big that this is not animmediate problem.

2) Providing guarantees or insurance: in order not to thwart market discipline, such guarantees andinsurance should be provided at rates that genuinely reflect the risks involved. This is alsocompatible with market systems, and with the entry of private insurers into the market, which can beexpected in the near to medium term. The best way to insure risk-reflective pricing is to treat aguarantee and insurance operation as a separate, self-funded activity.

3) Intermediary for small borrowers : DBSA or another entity could pool small, good quality loansto achieve marketable size. Because the private sector could, in theory, do this as well, there is apossibility of crowding out private sector initiatives. A size limit, in terms of population, annualbudget, or issue size would limit the potential for competition with the private sector.

4) Long or junior positions : If it is true that private markets will currently lend only short- to medium-term, due to macro uncertainties, DBSA or another entity could provide:

a) financing for the longer maturities in a given issue,b) subordinated positions over the life of a loan,c) guaranteed take-out of some percentage of loans at some point in time (e.g. 80% take-out at 7

years).

The problem with these approaches is principally pricing. At some price, the private sector wouldprobably lend for long terms, or would lend the full amount required. As uncertainties diminish and asprojects or municipalities merit, such prices are likely to drop. Again, one would not want the DBSA tocrowd out the private sector. The danger of crowding out is lower if this activity is undertaken as aseparate, self-funded activity.

4 December, 1998

122

There would also be the need to describe what kinds of projects are eligible, to what amounts, and withwhat requirements. Clarity and transparency about the rules would be necessary to prevent arbitrary orrent-seeking decisions.

5) Rejected transaction finance: One way to keep the DBSA from competing with the privatesector is to allow first crack to the private sector. Before undertaking any municipal transaction, theDBSA would be required to advertise its intent. Or a borrower would have to show that it hadapproached, e.g. 3 lenders before coming to DBSA. These approaches at their best add time andtransaction cost. At their worst, they remove market discipline completely.

6) Direct lending: There is nothing wrong in principle with DBSA competing for investment positionswith private lenders. But if it does so at below-market rates with “free” or cheap governmentmoney, it can spoil the operation of the market. The potential for cheap funds will keep bothpotential borrowers and potential lenders from engaging fully in the market.

7) Special purpose lending, e.g. for environmental, transition or some other public-policy purposewhich the government was willing to make a priority. The has the virtue of restricting non-marketactivities to certain sectors or categories. Such an approach may limit private market activity in thegiven sector or category, but avoids crowding out in other sectors. Of course, below-marketfunding for any particular sector would tend to distort municipal priorities in favour of that sector.

In summary, there is no obvious way for DBSA to be active in the markets without the potential fordamage to the very discipline market mechanisms are intended to achieve, unless DBSA is bound bymarket rules, i.e. that it must, at least in the long run, get its capital from the capital markets and price itsloans accordingly.