The financialisation of housing across five case study countries - synthesis report

266800

FESSUD FINANCIALISATION, ECONOMY, SOCIETY AND SUSTAINABLE DEVELOPMENT Large Collaborative Project, Social Sciences and Humanities D8.26 Two synthesis reports for the water sector and the housing sector bringing together findings from the case study countries

Due date of deliverable: 31/05/15 (M44)

Completion date: 31/05/15 (M44)

Start date of project: 01/12/2011

Duration: 60 months

Deliverable lead contractor: 6 UEP

Author(s): Kate Bayliss, SOAS, University of London Mary Roberston, University of Leeds

Dissemination Level: PU

Deliverable Status: Completed

2

This project is funded by the European Union under

the 7th Research Framework programme (theme SSH) Grant Agreement nr 266800

Preface This deliverable contains two reports submitted to meet the requirements for Deliverable 8.26 (D8.26) under Task 6, Work Package (WP) 8 of the EU-Funded FP7 FESSUD project. The papers provide syntheses of the findings from the series of case studies prepared for D8.25 exploring the systems of provision (sop) for water and for housing in the UK, South Africa, Poland, Portugal and Istanbul. The papers submitted for D8.25 of FESSUD, on which this Deliverable is based, are as follows: Water: K. Bayliss: Neoliberalisation of Water in South Africa, SOAS, University of London, with support from F. Banda and G. Isaacs, CSID, University of the Witwatersrand, Johannesburg P. Lis: Financialisation of the Water Sector in Poland, Poznań University of Economics N. Teles: Financialisation and Neoliberalism: The Case of Water Provision in Portugal, CES, University of Coimbra K. Bayliss: The Financialisation of Water in England and Wales, SOAS, University of London G. Yilmaz & Ö. Çelik: Case Study: Rethinking Istanbul Waters through Systems of Provision, Middle East Technical University, Ankara Housing: Çelik, Ö., A. Topal and G. Yalman (2015): Fınance and Housing Provision in Istanbul, Middle East Technical University, Ankara Isaacs, G. (2015): Housing System of Provision – South African Case Study, CSID, University of the Witwatersrand, Johannesburg Lis, P. (2015): Financialisation of the System of Provision Applied to Housing in Poland, FESSUD Working Paper No. 100, Poznań University of Economics Robertson, M. (2014): Case Study: Housing and Finance Provision in Britain’, FESSUD Working Paper No. 51, University of Leeds Santos, A. C., N. Serra and N. Teles (2015): Finance and Housing Provision in Portugal, FESSUD Working Paper No. 79, CES, University of Coimbra The above case study papers, already submitted, provide a detailed analysis of the sop for water and housing in these selected countries. This Deliverable does not aim to reproduce the material contained in the case studies. Rather, the objective is to provide a comparative overview of the main findings. These synthesis reports draw on the D8.25 papers as background material and, where reference is made to country examples, these papers are not cited on each occasion. Other sources have been added to supplement and update the case papers as appropriate. The findings in this report also contribute to three thematic papers prepared for FESSUD Deliverable D8.27 on the following topics: Neoliberalism, Financialisation and The Role of the State.

3



D8.26 Synthesis Report Part 1:

The System of Provision for Water in Selected Case Study Countries1

Kate Bayliss, SOAS, University of London

Abstract This paper provides a synthesis of the main findings from five comparative case studies prepared for Deliverable 8.25 for the EU research programme, Financialisation, economy, society and sustainable development (FESSUD). The studies – conducted in England and Wales, Poland, Portugal, South Africa and Istanbul - review the system of provision (sop) for water, exploring the way that sector outcomes are shaped by relations between agents, themselves embedded in historically evolved social and economic structures and processes. The research has focused on the role of finance in the sop. The paper shows that the policy approach adopted in each of the cases is remarkably similar. All of the locations introduced neoliberal sector reforms in the 1980s and 1990s. All have stated a commitment to cost recovery pricing although the way in which this is applied varies across locations. Similarly the extent of other neoliberal reforms, such as decentralisation and privatisation varies widely. England and Wales is an extreme outlier in the nature and depth of privatisation, with all the water and sewerage companies listed on the London Stock Exchange in 1989. Several companies have since been de-listed and are owned privately, some by financial companies. The financialisation of the sector is far more profound here than elsewhere. In the other case study locations, privatisation takes the form of a lease or concession contract, and the extent of implementation has been less widespread. Using the sop approach, the paper explores the way that the delivery of water is contested among agents. Neoliberal policies are presented as scientific and politically neutral by their proponents but agents in the sector have competing priorities. Contestation is particularly prevalent where private companies are involved in service delivery but there are also tensions between different state agencies involved in the provision of water. Pricing is a key area of conflict, with upward pressure from privately-owned water companies while some municipal providers strive to keep charges down for political and/or social reasons. Meanwhile external agencies, such as the European Union, also place policy requirements on governments. The findings suggest that cost recovery pricing is not affordable for many households and, in places, the sector is under strain from unpaid water bills. Authorities have adopted different approaches to social policy, with the cost recovery approach to pricing applied more strictly in some locations than others, but measures to support those who struggle to pay their bills are often inadequate. The case studies highlight diversity and conflict in the role of the state, as different agents compete for economic and social control. The neoliberal policies adopted are not neutral. Rather, there are

1 This paper builds on the extensive work of the authors of the country case studies. In addition thanks are due to the following for comments on earlier drafts: Piotr Lis, Nuno Teles, Ozlem Celik, Galip Yalman, Mary Robertson and Ben Fine.

4



winners and losers. Outcomes emerge from embedded power relations which are specific to individual locations and peculiar to water.

1 Introduction This paper brings together findings from case studies carried out under Task 6 of Work Package 8 of the EU-funding research programme Financialisation, economy, society and sustainable development (FESSUD). These case studies have examined the system of provision (sop) for water in selected locations. According to the sop approach, sector outcomes emerge from relations between agents which are themselves embedded in historically evolved social and economic structures and processes. This is in contrast to orthodox economic approaches which view the world in terms of deviations from an idealized, market-like condition, subject to correction through regulation or otherwise. Originally devised in connection with consumption studies, this segment of the FESSUD research programme aims to extend the sop approach to consider public sector systems of provision with particular reference to housing and water (for more on this see Bayliss, Fine and Robertson 2013). One of the key principles of the sop approach is that consumption is not the spontaneous outcome of decisions made by rational individuals but is inherently vertically linked to production processes. Participants in the sop have diverse and often competing interests with more or less permanent resolutions highly contested, and contestation continuing to evolve. These agents operate within structures, relations and processes which are far from neutral, and power relations shape the outcomes of the sop. For the sop approach, each commodity has its own material culture (MC) which is unique in time and location and derived from the commodity itself and the context in which it is provided and consumed. The MC of water is distinct from that of housing. This is in contrast with orthodox approaches that consider outcomes to emerge from the combined actions of optimising rational individuals with all commodities treated in the same way subject to conditions governing supply and demand. The factors that shape cultural systems have been grouped by Fine (2013) under ten headings (known as the 10Cs): Constructed, Construed, Commodified, Conforming, Contextual, Contradictory, Chaotic, Closed, Contested and Collective. The relevance and usefulness of the different Cs will vary, depending on the type of good, the sop and the reason for investigation.2 Water has certain properties that affect its sop. It is an input into virtually all aspects of social and economic life not just in its own right but also as an input into industry, agriculture and energy. There is no substitute. It flows downhill (unless pumped) and sometimes has to be shared across regional and international boundaries. It is heavy to transport relative to value and so tends to be used close to source. Delivery is capital- 2 See Bayliss, Fine and Robertson 2013 for more details on these and how they relate to the provision of water and housing.

5



intensive, relying on networks of pipes and pumps that are not easily moveable, so investments are long-term. There are considerable scale economies and delivery is usually monopolistic. Ensuring supply can be challenging due to variability in rainfall, and provision is affected by pollution and climate change. The selected case studies show the importance of context. Each case presents a different set of issues and constraints when it comes to water. Differences are particularly prominent in the social and economic history as well as demographics and geological context. E&W is a high-income country with a sophisticated financial sector and long-established privatisation programme. Portugal, an EU member since 1999, is regarded as being part of the EU southern periphery where countries have faced challenges with loss of competitiveness and rising external deficits. Poland has been in the process of transition from a planned to a market economy since 1990. South Africa has also been through a major transition after the end of the apartheid in 1994, with extensive state investment to address the inequalities of the previous regime. The case study from Istanbul stands apart from the others in that it is a city (rather than a national study) with a very high population density. The case studies also present significant diversity in the geological aspects of water provision. South Africa and Istanbul are highly water-stressed and have constructed major infrastructure to divert water for long distances to urban locations. Elsewhere, water is plentiful and used close to source. The MC of water has changed over the past three decades as part of a global shift towards a more “neoliberal” ethos in the provision of basic services. In most of 20th century Europe, water production was the preserve of the state which provided investment to ensure universal access, and there were cultural and symbolic features associated with the expansion of water infrastructure at this stage (Gandy 2004; Swyngedouw 2005; Bakker 2007). In the UK, a rapid expansion in connections to the water network was funded largely by governments partly through local taxes and partly through concessional loans from central government (Fisher et al 2005). Since the late 1980s, across the case studies (and elsewhere), the provision of water has been framed in a more commodified form. That is to say there is greater attention to water pricing as a tool to control the demand for water with increasing attention to water metering and prices that recover costs. Service provision is increasingly presented as a market with providers, if not necessarily in the private sector, encouraged to adopt business-like approaches to management. This policy shift is couched in an ideology of greater “water efficiency”, itself with increasing emphasis on the notion of scarcity – both of water and of finance. Policy discourse has shifted from one of abundance to one where resources are in short supply. This shift in narrative is presented as justification for the greater attention to financial and demand management that neoliberal practices provide. While the case studies present diverse socio-economic and geographical contexts, they have all adopted a “neoliberal” ethos to water policy,3 although there is considerable variation in way in which this ethos has been adopted in practice. All countries state a commitment to “cost recovery” pricing (critically discussed in detail

3 The nature of neoliberalism, with application to water and housing, is the subject of a separate thematic report.

6



below) but implementation of other aspects of the traditional neoliberal package (such as decentralisation and privatisation) have proved more challenging to apply to the water sop. E&W is an extreme outlier. Here privatisation took the form of divestiture with water and sewerage companies listed on the London Stock Exchange. Several companies have since been de-listed and a significant proportion of water provision is in the hands of global financial investors. In the other case study locations (and most of the rest of the world), water privatisation takes the form of a lease or concession contract and the extent of implementation has not been widespread, despite policy efforts. The sop approach shows that the neoliberal framing of water provision has benefitted some interest groups over others since the early 1990s. Far from providing a neutral policy package in which “markets” are shaped to improve environmental and societal outcomes, the case studies show that control and use of resources and finances are contested in the sector. There are winners and losers from the neoliberalisation of water. Outcomes stem from (newly) embedded power relations and these are specific to the individual case study locations. Consistently, however, the case studies suggest that neoliberalism has favoured powerful economic and political interest groups while low-income households (and labour although this is not covered extensively in the case studies) have lost out. The paper is structured as follows. The following section reviews the country findings on water production. This is followed by a review of sector finance. Financialisation is covered to some degree but this has been limited outside E&W. The paper then turns to water consumption in the case studies, exploring the ways in which end users access water with particular attention to pricing and affordability. The subsequent section considers the role of the state both in terms of the institutional framework and in balancing the competing interests of agents in the sop. Section 5 concludes.

2 Production Societies have always been organised around the consumption and distribution of water. In modern times, water production and distribution has become based on extensive capital investment in pipes and pumps. These are long-term investments, constructed and used over decades. Investments have been shaped by the political and economic systems in which they are located. Current sops have emerged from decades of evolving practices combined with geological, political and social imperatives. The long-lasting nature of water infrastructure means that there is a considerable lag between the prevailing political paradigm and that which produced the infrastructure (Mosse 2008). All the countries studied have seen a fairly rapid expansion of access to water funded and implemented by the state and/or donors. In E&W this took place at the start of the last century. In Poland, Portugal and South Africa, it has been more recent. Both Portugal and Poland were required to increase access under directives from the European Union (EU). In Portugal this has been since 1986 and in Poland since 1993, each with substantial EU funding. In South Africa, the country’s infrastructure has been skewed towards a white minority. With the end of apartheid, the ANC

7



government focused on an extensive investment programme to redress the inequality of the previous regime. The result has been a substantial increase in access since 1990, particularly in rural areas (Table 1), although “access” can involve varying quality of connection and of the water itself. Table 1: Water consumption Piped on to premises % of population

Urban Rural Total 1990 2012 1990 2012 1990 2012

Poland4 97 99 73 96 88 98 Portugal 96 100 83 100 88 100 South Africa 85 93 16 57 52 79 Turkey 91 99 51 97 75 99 United Kingdom 100 100 98 98 100 100 WHO/UNICEF Joint Monitoring Programme for Water Supply and Sanitation Country files (www.wssinfo.org)

Water is abstracted from surface sources (rivers and lakes) and (under)ground sources (aquifers) for consumption by households, agriculture and industry. This is more or less a universal aspect. There is not much diversity across the cases in this regard. Water technology is not one of rapid change. Some cities have been using infrastructure that is over a century old. What do vary across locations and over time are the institutional structures and processes of water provision. All the case studies have, to varying degrees, brought in substantial sector reforms over the last three decades. The water sops studied are situated on something of a sliding scale of commodification, globalisation, corporatisation and privatisation of water production which can be considered as comprising a neoliberal package of policy reforms. Neoliberalism is also typically associated with decentralisation based on the supposition that it will increase local accountability and strengthen cost recovery policies (see below on cost recovery and Herrera and Post 2014 on decentralisation of water services). However, in the case studies, the trend is towards more rather than less centralisation to reduce the impact of fragmentation of service provision. E&W stands apart from the other case studies. Here, water is provided by ten private water and sewerage companies (and some smaller water-only companies) with boundaries based on river basins. In contrast, in each of Portugal, Poland and South Africa, delivery of water is largely the responsibility of municipalities although some large volume (industrial) consumers access water directly from bulk sources. Each of these countries has undergone similar reforms which aim to structure production so that costs and revenues are ring-fenced, sometimes under a corporation owned by the state (known as corporatisation), and this can facilitate the introduction of private investment at a later stage. Other initiatives are the adoption of “cost recovery” pricing (required under the EU Water Framework Directive (WFD)), and in some cases privatisation with concession contracts (Poland, Portugal and South Africa) and divestiture (E&W). These processes are explored below. 4 These data are all from the Joint Monitoring Programme which compiles indicators using several sources but there are inconsistencies with some national level data. The Central Statistical Office in Poland puts the percentage of the population connected to the public water supply at 88% for 2011 (GUS 2013).

8



2.1 Processes: (un)bundling and corporatisation Water production has been substantially restructured in all countries. While there are similar processes observed in terms of commodification and privatisation, the sops are packaged in different ways as countries vary in the nature of horizontal and vertical integration. In E&W and Poland, the provision of water is vertically integrated, with the same organisation responsible for provision from the source through to end-user. In Portugal, however, the supply of bulk (ie water abstraction, treatment, elevation and adduction) water has been separated from retail (storage and final distribution to end consumers including tariff setting and collection) water in a process known as “deverticalisation”. In South Africa there is a similar separation of bulk and retail water as well as a third level of horizontal stratification in the sop with an additional category known as “raw water” which applies to untreated supplies, consumed directly from the water source. Users of raw water include large industries, mines and irrigators. Notwithstanding these structural variations, water institutions in these countries have all undergone a process of corporatisation if to different degrees. The water sector in Portugal was substantially restructured in 1993. Prior to this date, water provision had been the exclusive responsibility of local municipalities - with the exception of Lisbon where water was managed by the state-owned enterprise Empresa Publica de Aguas de Lisboa (EPAL). This localised control at the, now elected, municipal level was significant in the country after the 1974 revolution. However, the sector was highly fragmented with some 300 municipalities responsible for all aspects of water provision from abstraction through to end-users. The restructuring in the 1990s took the form of separation of bulk water from retail water. Bulk water provision was consolidated through the creation of a series of companies across the country. In each of the bulk water companies, the controlling stakeholder was a newly-created state holding company, Aguas de Portugal (AdP), with a 51% stake, while the municipality (or a group of municipalities where the bulk provider served more than one) had a 49% ownership share. This represented a process of centralisation and consolidation AdP was and continues to be owned by the state (with the state-owned development bank, Parcaixa, SGPS, holding a 19% shareholding (AdP Annual Report 2013)). Although ownership was to remain in public hands, it was expected that reforms would bring in principles of private sector style management. While the state retained ownership of the water sector, the new corporate concessionaires assumed control over businesses with a high degree of institutional and budgetary independence. The expectation was that new management, supposedly independent from political pressures, would enhance efficiency. And, for Portugal, this was also motivated by converging to European standards to meet the conditions set for accessing funding from the EU. The new forms of public management were intended to turn policy making into a technocratic process devoid of any political content and subject to financial constraints. Corporatisation of water introduced corporate accountability in management practices which served to embed financial practices further in the sector. At the retail level, water in Portugal continues to be controlled by municipalities. Since 1993 they have had autonomy to raise finance independently, and some have introduced private concession contracts known as “public-private partnerships” (PPPs). There continues to be tension between municipalities and AdP which goes

9



back to the separation of bulk from retail water and was not popular with many municipalities in Portugal. About a third of municipalities have refused to give up their control over bulk water and continue to operate an integrated system. The above-mentioned processes of reforming segregation and decentralisation of water provision have affected financing. Investment at the municipal level in Portugal is more expensive than investment in the bulk sector because borrowing is on a smaller scale. Interest rates have been considerably higher at the municipal level than for the bulk providers where AdP has a stake because the scale of borrowing gives them access to cheap finance such as loans from the EIB which are not available at the municipal level (see below). However, many private operators in the retail sector have still managed to make a profit (see below). In E&W a similar corporatisation process took place in the 1970s. Under the 1973 Water Act, water became the responsibility of Water Resources Authorities. These operated on the river basin, rather than municipal level. They were ring fenced so that funds could no longer be diverted into the local authority budget. From 1973 they were obliged to operate on a cost-recovery basis, and investment finance could be raised by borrowing from central government. After 1983 they were allowed to borrow from private capital markets. These changes shaped the sector that was privatised in 1989 (see below). In South Africa, the water sector was restructured in the mid-1990s after the end of apartheid. The institutional framework has not changed substantially but the ethos of cost recovery as well as a commitment to ensuring universal access come under the 1997 Water Services Act. As mentioned above, South Africa has three tiers of water production (raw, bulk and retail water). In part, this has evolved as a result of the geographical and socio-economic development of the country. South Africa is classified as highly water-stressed but the country’s areas of economic activity are not aligned with water availability. Water security has been achieved by large-scale engineering and a well-developed system of dams to divert water to where it is most used. Gauteng Province, where the largest urban area, Johannesburg, is located, imports 88% of its water and relies on water from Lesotho via the Lesotho Highlands Water Project. The users of raw water are those who take it directly from source or from large infrastructure, such as large industries and mines. In addition raw water is taken by bulk water companies who treat it before selling it on to municipalities or to other industrial consumers. Raw water infrastructure is financed “off-budget” largely under the control of the Trans Caledon Tunnel Authority (TCTA) a state-owned enterprise (SOE) established in 1986. There is a charge for raw water based on the costs of the infrastructure and it is distributed across those who use the infrastructure, rather than out of general taxation. TCTA raises private (and sometimes concessional) loan finance for infrastructure and allocates repayment costs across end users, securing financial commitments in advance. This means that the costs of specific infrastructures are allocated just to the users (mines, industries, water boards and water service authorities). This is based on the neoliberal pricing strategy known as “user pays” whereby the costs are allocated directly to users (rather than financed out of general taxation).

10



This approach to financing is in contrast to the integrated pooling of finance observed elsewhere, for example in vertically integrated water utilities. Furthermore this financing strategy only applies to new infrastructure and the result is that water is provided relatively cheaply to large volume consumers while those that receive water after it has been processed by water boards and municipal providers (ie households) pay a higher price. The country’s 2013 water strategy envisages mobilizing more private sector finance for the “economically viable portion of water resource development; that is water supplies to users who can afford to repay loan finance, such as industries, mines and power generation and domestic users receiving high levels of water services” (NWRS2 DWA 2012b, p. 86). This approach which largely by-passes government spending (although there are some transfers from the state Department for Water Affairs (DWA)) also means that wealthy (‘economically viable’) users do not have to engage with government financing but are encouraged to contribute to separate private financial structures for their own consumption. These also have the option of paying a premium to be in the bracket of “high assurance user” to ensure a more reliable water supply. Thus the application of the neoliberal pricing policy of “user pays” leads, potentially, to the hiving off of provision for the most wealthy who finance and use their own infrastructure separately from the users for whom provision is not necessarily “economically viable” (ie profitable). In South Africa there are twelve bulk water providers (Water Boards (WBs)) that buy raw water, treat it and distribute it to industry, agriculture and municipal providers. There is considerable diversity in the economic health of water boards. Some have very high debts while others are profitable, in part due to the economic health of the region of the country which they serve. Water is then sold by the bulk WB to the Water Service Authority (WSA) (usually the municipality) which provides water to end-users. The country has 152 WSAs. Water is one of several services provided by municipalities. Paying for bulk water is a municipal cost item and water revenue goes into the general finance pool for municipal revenue. So water can cross-subsidise other municipal services. The vertical segregation has meant that water users are considered as separate consumer groups. For the South African government, bulk raw water infrastructure has been planned for the needs of a specific sector “to the exclusion of other water users” so that the planning of bulk water infrastructure has not taken account of the water needs of communities and rural households.5 The result has been the construction of infrastructure and distribution networks that bypass these communities. Wealthy mineral production sits next to shack housing where residents lack basic services. In Poland, in contrast, there is extensive horizontal segregation. Following reforms in 1990, water management was decentralised with responsibility devolved to local authorities (gminas). There are now 2,479 of these in the country. While the municipality is responsible for the provision of water there has been a step to distance the provision of water from municipal councils by establishing separate entities for water provision. The municipal authorities are required to delegate water management to separate organisational forms and there are now 1,807 of these in the country with some providing services to more than one gmina. Of these, 656 take the form of

5 Department of Water Affairs General Notice, Notice 888 of 2013, No.36790

11



“commercial law companies” and most of which (543) are majority-owned by the local authority (gmina). Other commercial law companies were owned by domestic capital and some had a share of foreign capital. In addition there are 582 organisations described as “budgetary establishments”; 286 “water companies” and 244 are “natural persons running business activities” (see Lis 2015, p.19 for more details). Even if water delivery is delegated, it is still the responsibility of the local authority. Water management in Poland then operates on a very small and fragmented scale in some cases. Efforts to create a separate entity for the management of the water has not always been successful and, in a number of cases, the water enterprise is owned by the gmina. In Warsaw, the municipal enterprise was converted into a public limited company – the Municipal Water and Sewage Company which is described as a joint stock company whose shareholder is the city of Warsaw. Similarly, in 2005, Aquanet SA was created as a joint stock company providing water and sewage services for the city of Poznan. Politics has been overt in the management of water in Poland with local government election candidates offering the promise of cheap water if they are elected (Lis 2015). In Turkey the provision of drinking water was originally the responsibility of municipalities with the “Law on the Waters’ in 1926. However, in the wake of Great Depression, control became centralised. The Bank of Municipalities was established in 1933 to support the financing of municipalities’ investments. Yet it soon became clear that giving public loans to the municipalities was not enough to eliminate their financial difficulties. The Development Board of Municipalities was established in 1935 to provide drinking water to municipalities with a population of more than ten thousand under the auspices of the Ministry of the Interior. In the post-war era, there was further institutional restructuring as a new financial institution came into existence with enhanced capability to provide drinking water for municipalities, irrespective of the number of their inhabitants. With the merging of the Development Board of Municipalities and the Bank of Municipalities, the Bank of Provinces was established in 1945 with responsibility not only for the provision of finance to municipalities for infrastructural investments including water and sewage systems, but also to provide technical support for such projects. The establishment of the Bank was followed two years later by the foundation of the Municipalities Fund, which augmented the financing capability of the Bank and remained as the main source of municipal finance for water systems. Investments in water and sewage systems constituted the major portion of the allocations made by the Bank for the following decades. Moreover, Water Administration units were established in 1947 in the three major cities of Istanbul, Ankara and İzmir. The General Directorate of State Hydraulic Works (DSI), established in 1953, was given the task in 1960 of developing and financing water systems for municipalities with a population of under 3000 inhabitants and all villages. DSI is a national level institution that used the sources of the Treasury but the municipality was expected to contribute in cash or in kind (equipment and labour force) to DSI investment programmes in water systems. After 1964, this task was taken over by the newly-established Ministry of Village Affairs. With the rise in urban population of major cities due to increased migration from the rural areas, DSI was to be brought back into the provision of water for major cities with a population of more than one hundred thousand people, including Ankara and Istanbul, given the inadequate financial

12



capabilities of the municipal administrations concerned. This meant effectively the centralisation of the provision of water as DSI was to assume responsibility for the planning and construction of water and purification systems for the use of households as well as of industry. The financing of these activities would entail the provision of loans to the municipalities with 30-year maturity without any interest as proscribed by Law no: 1053 enacted in 1968. However, the management of the water and purification systems would be undertaken by the municipal administrations once they were completed by DSI (Çınar, 2006a, 2006b). With the transition to neoliberalism from 1980 onwards, another round of institutional restructuring was initiated. This entailed a new division of labour among the institutions concerned, namely, Ministry of Village Affairs, the Bank of Provinces, DSI and municipal administrations. The restructuring brought in qualitative changes in the financing of water and sewage systems and increases in prices. With the establishment of Greater City Municipalities for cities such as Istanbul in 1984, the role of the Bank of Provinces in the provision of water diminished significantly. This signalled the transition from Water Administration units to a new model called ‘ISKI model water management’, initiated first in İstanbul, and subsequently reproduced in other Greater City Municipalities. İstanbul Water and Sewerage Administration (ISKI) was initially established as a separate institution to meet one of the conditions attached to receiving loans from the World Bank in 1981. Then in 1984, ISKI became an institution of the Greater Istanbul Municipality. The main distinction of the ISKI model was its policy on the pricing of water, bringing in ‘the user pays’ model. The case studies show that the structures of the water sectors continue to evolve. In the UK, vertically integrated private water companies are going to be required to separate their bulk water production activities from retail as the retail part is due to be subject to competition from 2017 for business customers. Meanwhile in Portugal, taking advantage of the current significant financial constraints in some municipal providers, there is a plan to integrate AdP into some municipal provision, thereby vertically integrating the sop. And bulk water is being consolidated with nineteen corporate entities being reduced to just four. There is also consolidation in South Africa with twelve bulk entities being reduced to nine. In Portugal the aim of consolidation is to reduce the gap between bulk water tariffs of coastal regions and the interior (where tariffs are higher) so there will be more pooling and greater regional pooling of costs and charges. In South Africa there is very weak capacity in some bulk providers, in part reflecting the socio-economic context in which they operate. The weaker are being de-established and in some cases their operations taken over by better performing WBs.

2.2 Processes: Privatisation There has been some privatisation in each of the case study locations although the E&W case stands out with the sector having been entirely in the private sector for the past 25 years. Here, water companies were privatised by listing on the London Stock Exchange. This process has not been replicated in any other country. In E&W, infrastructure investment is entirely the responsibility of the private company. Ownership stakes in the companies are bought and sold. Since privatisation in 1989 there has been considerable change in the ownership structure and its forms. Most of

13



the companies have been delisted from the stock exchange and are owned privately. Four out of the ten water and sewerage companies are now owned by “special purpose vehicles” (SPVs) which are shell companies put together by financial investors. These have set up complex off-shore corporate group structures which have allowed the water utilities to amass very high debt levels while staying just within the boundaries of the legal regulatory framework. These companies have added some of the debt that they used to acquire the company to the debt of the water utility with the interest payments covered by consumers’ bills. This is a more profound transition than observed in the other countries. Elsewhere in the case studies (and in the world) privatisation in the water sector usually takes the form of a concession or lease contract (the distinction being which party is responsible for ownership of infrastructure) for a number of years. Despite extensive efforts, privatisation has not been widespread. Reforms in Portugal in 1993 have allowed the entry of private capital in the sector and this has taken different forms. In some municipal companies, private capital holds only minority stakes in corporate concessionaires and the municipality remains the majority stakeholder. This approach is intended to leverage more private finance. The country now has 29 retail private concession contracts out of 380 managing entities and two bulk level private concessions at the municipal level out of sixteen managing entities. However, while private concessions are few in number they are mostly in densely populated areas and so they cover 13% of the population in the retail sector. Three major Portuguese and Spanish construction companies have come to dominate the water privatisations in Portugal: Aquapor, Indaqua and AGS. Aquapor has the strongest presence with numerous municipal and multi-municipal concessions across the country as well as minority stakes in some municipal companies. Aquapor was originally part of the ADP group but started to bid for municipal retail concessions and to reorganise the internationalisation of AdP, ultimately being privatised in 2008. The company is now owned by the construction firms DBB, AGS and Bragaparques. The other two firms are also owned by construction firms for which diversification arose out of the use of idle funds following the stagnation of the housing construction market over the previous fifteen years. Overall these companies have found the water sector to be profitable. The private concessionaires AdPlanalto, AGSPFerreira and AdPFigueira have achieved operating margins of 60.4%, 55.9% and 50.5% - well above the threshold recommended by the regulator, ERSAR. But privatisation has proved expensive in Portugal. The case study cites an audit report that shows that the municipalities bear most of the financial and operational risk, for example from changes in costs due to changes in the reference “Euribor” rate. This is in part attributed to failings in the capacity of the state agencies to negotiate contracts and the absence of monitoring units to supervise implementation. In Poland, the water and sewerage function has been established as separate, at arm’s length from the local authority. There have been few concession contracts apart from a thirty-year concession signed in 1992 to a joint venture company owned 51% by the French company, SAUR and 49% by the City of Gdansk called Saur Neptun Gdańsk (SNG).6 Elsewhere in the country, privatisation did not take off despite policy efforts

6 www.sng.com.pl

14



with this aim. In Poznan for example, there was a plan to privatise the water supply in 1996 but this was rejected. According to Hall, Lobina and Motte (2005) the city council rejected the privatisation proposal on the grounds that the city had already improved the efficiency of its water services and had obtained investment finance from the EIB. While water providers in Poland have been established separately from the local authority, large-scale privatisation has been limited. The high level of fragmentation of the sector is a challenge for privatisation, and a process of consolidation is difficult due to the separate price setting process in each gmina. The establishment of a central regulator, discussed further below, would be a step towards more widespread privatisation in the sector (Lis 2015). In South Africa municipalities are able to subcontract water services to private providers, but this has not occurred on any scale. Of the privatisation initiatives in the 1990s and 2000s only two long-term concessions remain (in Nelspruit and Dolphin Coast), and these have both now been brought under the control of a single owner, Singapore company, Sembcorp. Elsewhere in South Africa, three contracts signed in the Eastern Cape in 1999 were either terminated or not renewed. Johannesburg had a management contract with the French multinational, Suez, for five years from 2001 but this was not extended when it expired in 2006. While privatisation in the form of concessions has not been widespread, there has been smaller-scale private involvement, for example, with contracts to build and operate wastewater treatment plant. In South Africa, Veolia has a 20-year contract to provide water treatment through the Durban Water Recycling Project.7 In Warsaw, Veolia Water Solutions and Technologies built a wastewater treatment plant in 2013. The project cost 769 million euros with 40% funded by the European Cohesion Fund and 60% by the Municipality of Warsaw.8 Of the privatisations that have taken place, there has been some consolidation of ownership and private water companies are part of global conglomerates. Sembcorp which now owns the two South Africa concessions also owns a water-only company in Bournemouth in England. Veolia sold its Portuguese concessions in 2013 to Beijing Enterprises Water Group (BEWG) Ltd which was incorporated in Bermuda as an exempted company with limited liability and the shares are listed on the Hong Kong Stock Exchange.9 There are parallels with the English privatisation pattern as the Beijing company, that took over the utility CGEP from Veolia, is in part financing this with loans from the new shareholders. The shareholder loan will be paid interest annually by the utility, CGEP, to its parent holding company. So, the utility pays interest to the owners of the company on funds used to buy the company. Interest is another form of shareholder distribution along with dividends. And the interest is tax deductible.10

7 www.veoliawaterst.co.za/municipal-water-treatment 8 Veolia Press Release, 29 March 2013. 9 “BEWG successfully acquired Portugal assets of Veolia Water” Press Release, Beijing Enterprises Water Group Ltd, Hong Kong, 25 March 2013. 10 “BEWG successfully acquired Portugal assets of Veolia Water” Press Release, Beijing Enterprises Water Group Ltd, Hong Kong, 25 March 2013.

15



Each privatisation contract is associated with numerous risks. These include the risk that the infrastructure will not be built on time or to specification; that the currency will fall in value while funds have been raised in foreign exchange; that the cost of key inputs (eg power) will increase; that interest rates will increase; that demand for water will not be as predicted; that end users will not pay their bills; that the government will change the rules of engagement or adjust prices to reduce revenues. Privatisation brings together agents with competing objectives and the allocation of risk is subject to contested negotiations. Investors want low risk while the procuring authorities want them to bear high risk. The reality of risk allocation is complex, emerging in part from the bargaining positions of agents involved and this is different across locations and sectors and risks change over the course of a project. The process of privatisation itself creates and shapes risks. Where the private sector is remunerated on the basis of units sold, a fall in consumption will lead to a shortfall in revenue (see section 4 on Consumption). Privatisation contracts are based on assumed revenue streams derived from anticipated demand. While this is generally fairly predictable with water, changes in the way it is provided, for example, increases in pricing and metering, can lead to reductions in demand. In the privatisation literature, this is known as “demand risk”. The case studies showed that generally, contracting firms were insulated from this risk. In Poland, in 1993 the private water company in Gdansk installed over 800 meters. By 2005 there were 30,000 meters installed. This led to a rapid decrease in demand in Gdansk, thereby reducing revenue for the investor. Household consumption fell by 15% in a single year in 1995. The price was increased several percent to compensate SAUR’s associated losses (as an ‘exceptional circumstance’) (de la Motte 2005). Consumers then had to pay a higher price as a direct result of reducing their consumption. In Portugal, all the contracts had to be revised with most of the amendments referring to adjustments of expected demand as this was overstated in the initial contracts. Most cases extended the contract period. This means that the firm’s revenues were maintained and paid over a longer period to compensate for shortfalls from demand contraction. In E&W private water firms are supposed to encourage customers to use less water under the Abstraction Incentive Mechanisms (AIM). Furthermore the expansion in the use of metering is also intended to reduce consumption. However, the Revenue Correction Mechanism (RCM) is intended to compensate firms for loss of such revenue and allows firms to increase prices in the next price review to compensate for a fall in demand (CCWater 2013). These cases are discussed in more detail in the section on consumption but the point here is that privatisation changes the way in which consumption affects agents. While a reduction in the volume of water consumed may be generally desirable for environmental reasons, with a private investor this becomes a revenue loss which reduces shareholder returns. If water were in the public domain, costs and expenditures could be revised accordingly. However, where private investors are expecting a return, they need to be compensated for a fall in demand if this in not anticipated.

16



The way in which private capital is engaged in water is limited in the case studies considered. E&W is an extreme outlier. In the other case studies, water providers have been created as a stand-alone corporatized entity, run along private sector lines. Water privatisation, where it has occurred, has taken the form of a fixed term lease concession but has not gone as far as transferring ownership of infrastructure to private investors. Outside E&W, privatisation has not been widespread due to limited profitable opportunities, government reluctance and public opposition. In Portugal, the privatisations that have occurred have been profitable but limited to the higher income areas. It is more difficult to attract private investors to rural areas with low incomes and higher costs. The private sector has already creamed off the most lucrative privatisation contracts which may limit the scope for further private transactions. In South Africa there was a strong anti-privatisation protest movement and, as the discussion below indicates, many end users struggle to pay for water which is less attractive for investors. There were some short-term management contracts in the country which ran their course (for example in Johannesburg) and now water is in the public sector. This fits with global trends. Water is the sector that has attracted the least of all private investment according to the World Bank Private Participation in Infrastructure (PPI) Database. The sector has the most cancellations or projects under distress. There is also a reported trend in the sector towards remunicipalisation in a number of cities across the world including Paris and Berlin (Kishimoto et al 2015). It is only in E&W where the structure of privatisation is deeply embedded and investors have very secure returns that a return to public water provision seems extremely unlikely. The involvement of the private sector in water in Turkey started in the middle of the 1990s through involvement of foreign finance. For example, some of the biggest water companies in the world were involved in the management and provision of water from Antalya Municipality for 10 years; for a dam in Izmit for 16 years; and for Cesme and Bursa as well. The World Bank, Europe Investment Bank (EIB) and Kreditanstait fur Wiederaufbau (KfW) had an important role in the involvement of the private sector in water management in Turkey. As well as the banks and finance corporations, multinational firms, i.e. Suez, Thames and Serco consortiums, have provided foreign loans to the water sector in Turkey. The involvement of these companies took the form of the ‘build-operate-transfer’ model (Çınar, 2006b). However, it seems so far there has not been any such private sector involvement in İstanbul for water provision.

2.3 Labour In E&W, the process of privatisation under Thatcher’s Conservative government in the 1980s was associated with a direct political agenda to break the power of trade unions. The restructuring of the sector into regional water providers created a more fragmented and therefore weaker union base in the sector. Although sector workers are part of national trade unions, employment terms and conditions are negotiated at the company level. The case study shows that a growing gap has emerged between payments to directors and expenditure on salaries and wages. In 1993 the remuneration of the highest paid director was in the region of 7 times the average

17



wage but by 2013 this ratio had risen to almost 30 reflecting a widening gulf between payments to senior executives and the employees in the sector. Directors’ remuneration is designed to ensure that their interests are aligned with those of shareholders. Senior staff are given shares in the company so they benefit financially from the payment of dividends and bonuses are awarded in part for improving shareholder returns. In Istanbul, there are three trade unions in the water provider, ISKI and they have seen their membership decline in the past decade as services have been subcontracted. The conditions of workers in subcontracting firms is precarious In Portugal, reductions in labour costs – both through wage cuts and downsizing - have to some extent counteracted increased financing costs during the current financial crisis (see below). Furthermore, union power has been weakened by sector restructuring. The corporatisation processes have involved the proliferation of “individual labour contracts” that are not covered by collective agreements. In addition, the current restructuring of the bulk sector will involve lay-offs, according to the trade unions.

3 Finance (and financialisation) The water sectors in the case studies have been financed over time by different combinations of government funding, donor grants, borrowing from banks and bond issuance. The case studies show different levels of financial complexity and involvement of the financial sector in the provision of water. As before, the experience of E&W is an outlier with the financial sector deeply involved in the sector, and ownership in the hands of financial companies. The E&W case showed all the “classic” aspects of a financialised sop with substantial rentier transfers, a big increase in proportion of revenue going to directors at the expense of labour, financial engineering, and revenue derived from financial practices rather than the production of water. This was not so significant for the other countries.

3.1 Capital investment finance The state has played a significant role in investment in water infrastructure in much of the twentieth century in most OECD countries (Bakker 2005). In the UK, the sector was largely in state hands until the late 1980s. In South Africa, central government grants continue to be disbursed to finance infrastructure to connect low-income communities to infrastructure networks. In addition to government finance, countries have benefitted from external grant funding. Both Portugal and Poland have received substantial investment finance from the EU. Between 1993 and 2012 28% of investment by AdP was funded by direct fiscal transfers from the EU, and support continues. Concessional loan finance has been important, for example from the DBSA in South Africa. All countries (but not Istanbul) have had loans from the European Investment

18



Ban (EIB) which has a specific water lending programme.11 The EIB lends to developed and developing countries and is the largest source of loan finance for the water sector compared with other International Financial Institutions (EIB 2008). In Portugal, AdP has seen a substantial increase in debt (from 744m euro in 2003 to 3,000m euro in 2013) and about 60% of this debt consists of loans from the EIB. In the case studies, the EIB lends to several water providers including the Durban water utility, eThekwini, and Welsh Water, Severn Trent and Southern Water in E&W. In Poland, the EIB has provided loans to the municipalities of Krakow and Warsaw for investment in infrastructure including for water.12 According to Lis (2015, p.36), the index of financing assets with equity capital (equity capital compared to total assets) was between 62 and 75% for the median of a group of water companies analysed in a study of 144 companies conducted by the Polish Waterworks Chamber of Commerce (IGWP). The study found that the larger the enterprise, the greater the share of equity capital in the balance of the enterprise. For Lis, this raises a fundamental question regarding the relationship between the rate of return on investments and the cost of the foreign capital involved, and this will determine to a large degree the return on equity capital and decisions concerning the structure of financing the assets of a water and sewage enterprise. The European Union plays a key role in providing finance regardless of the return on capital. Debt servicing costs with the financial surplus (net profit plus amortisation to total debt servicing ie capital instalments plus interest) amounted to 3.5-4.6% for the median of enterprises under analysis (from 7.7% to 14.3% for the average). The index of above 1.2% was obtained by 85% of enterprises. Although there are limitations with this index, it is assumed that enterprises did not have a problem with debt servicing. EIB loans have long maturities and preferential interest rates and their relative importance as a funding source for water is rising. The EIB lends to support EU water policy (EIB 2008). Bank lending supports sector consolidation with loans for the investments of service providers who operate at regional or multi-municipal levels. In Portugal, AdP was able to access EIB funding while smaller municipal water companies were not. In new member states, EIB loans have been used to support the creation of regional water utilities with appropriate operating and financial frameworks and necessary tariff reforms. The EIB states that it will “support cost recovery to ensure that service providers are financially sustainable” (p.9, EIB 2008). The EIB also supports policies to promote “Demand Side Management” which includes metering and pricing which are intended to improve what is termed “water efficiency” discussed in more detail below. Given the extensive reach of the EIB in the water sector, this would seem to be a significant means by which neoliberal hegemony in the sector can reach across countries. Bond issues to finance infrastructure have a long history with municipal finance and urban development. Infrastructure finance has always been interlinked with capital flows, and infrastructure bonds have been a core element of the development of modern capital markets (Gandy 2004). The case studies all showed that bond finance was standard practice to raise finance for infrastructure investment. In South Africa,

11 http://www.eib.org/infocentre/publications/all/eib-s-water-sector-lending-policy.htm 12 http://www.eib.org/projects/pipeline/2006/20060253.htm

19



the cities of Cape Town, Johannesburg and Ekurhuleni and Tshwane have issued municipal bonds mostly to finance large-scale infrastructure projects. Similarly in Poland, municipal bonds have been increasingly important. Galiński (2013) shows how these have been used to finance infrastructure in Poland, including water and sanitation in part to cover the country’s own contribution to EU investments. Mostly these were purchased by commercial banks. Although the municipal bond market in Poland is small relative to most of the EU, it is the largest bond market in the countries of Central and Eastern Europe. Public water companies, for example TCTA and Rand Water in South Africa, have issued bonds. TCTA is active in international financial markets raising money through the issue of bonds, long-term project loans and commercial papers. TCTA also uses derivatives to hedge risk exposure which is mainly the risk of changes in foreign currency exchange on the repayment of foreign loans and interest rate changes. In Portugal, AdP issued bonds of around 600 m euro to a very small number of foreign investors during the 2000s. This was to match long-term investments with long-term debt. The success of these bond market operations is attested by the low interest rates charged (1.8% interest rates in 2013). As with TCTA, AdP has also moved towards more sophisticated financial interventions with a number of interest rate and exchange swap derivatives mostly to protect against the variability of interest rates. Most of these derivatives were contracted with international banks (Citigroup, BBVA) as the domestic banking sector lacked the necessary know-how. However, these came under the spotlight due to losses made by a number of public enterprises involved in complex SWAP derivatives. Although AdP was one of the SOEs that was least affected in the country, the company still suffered notional losses of 25m euro in 2011 and 14m euro in 2012. The losses mainly related to interest rate swaps set up to try to hedge a rise in interest rates but after an intervention by the ECB in 2012, rates come down. Other SOEs suffered greater losses, as they had been involved in more complex derivatives. A new law has imposed stricter rules over contracting of derivatives for SOEs and these are now subject to approval from the Portuguese Treasury. The private water companies in E&W also issue bonds, and these have been involved in the most complex financial processes of all the case studies. The English water companies that are owned by financial institutions have created securitisation structures with special purpose vehicle companies established off-shore to allow extensive debt consolidation. These financial flows are difficult to trace, and complex group structures and inter-group company transfers cloud the picture further. These companies are the most deeply entrenched in the financial sector and some have securitised payments of water bills for decades to come. Generally, governments can access finance more cheaply than private companies. Governments are typically associated with a very low risk exposure and so they can secure finance at lower rates of interest. In addition, private investors also need to make a profit which pushes up costs. This was made explicit in the case studies where the sector regulators in Portugal and E&W calculate an expected rate of return for private investors which adds a premium to the rate of interest on government borrowing. In Portugal, the regulator recommends a target rate of return on capital based on the 10-year government bond market rate to which is added a risk premium

20



of 3%, In E&W, the estimated cost of capital for private water companies is based on the “risk free” return on government gilts to which is added a risk premium to compensate investors for the exposure to risk associated with capital markets. Although government borrowing is cheaper, this increases public debt which governments are seeking to avoid so decision-making can be biased towards privatisation. Largely, the nature and extent of financialisation of the water sector in the CSCs mirror the country’s wider experiences with financialisation. In the UK the financial sector is a core component of the economy, and the City of London is a global financial hub. The scale of financialisation of water is to a large degree driven by the companies that operate in this sphere (including investment banks, asset management firms and global financial companies). For Portugal, being in the Eurozone has directed the nature of financialisation as the country has benefitted from access to cheap foreign capital, and foreign bank loans have financed significant investment in the sector in the past 20 years. Financial liberalisation in Portugal opened the market to new private banks, and the privatisation of the banking sector and liberalisation of financial markets after 1989 were important for the development of a strong private domestic financial sector. The development of the financial sector was important for the development of capital markets which subsequently organised the wave of privatisations in the economy. However, financialisation in the country has been very uneven. AdP, the bulk water provider, has, with its scale of operations, become adept at financial practices. It has accessed various sources of credit and is well integrated into the international financial sphere. In contrast, retail systems are fragmented, diverse and more vulnerable to external pressure. There is therefore renewed pressure to integrate the municipality-controlled retail service into AdP. In South Africa, the nature and extent of financialisation is highly skewed, like the financial sector itself. Some water companies issue bonds and use derivatives (such as TCTA and Rand Water) while others serving poor areas with weak capacity lack any kind of financial sophistication and depend on central government support. Countries have fared differently from the global financial crisis and the crisis in the Eurozone. In 2011, Portugal was forced to request official financial assistance from the Troika13 to refinance public debt. In the water sector, there was greater pressure on municipalities to ensure that prices were at cost recovery levels. This process has been aided by a fall in labour costs in the public sector (a result of requirements of the country’s financial “bail-out”) which has off-set an increase in financing costs. However, the impact has been varied. The bulk water companies operating in areas that are more remote and sparsely populated and those with more recent investments have suffered losses due to high operating and financial costs. In E&W, the financial crisis was expected to make it more difficult (and expensive) for private investors to raise finance. This was a factor that came into the 2009 price review process when prices were set for the following five-year period, 2010 to 2015. As a result the regulator, Ofwat, allowed prices that were generous to investors to ensure that they would be able to finance their operations. In practice, the E&W water sector has been particularly attractive for investors offering both secure returns and a haven outside the turmoil of the Eurozone so the cost of financing for water firms has been lower

13 ECB, IMF and EU.

21



than anticipated in the price review and firms have made additional profits as a result, although the latest price review looks set to require a slight reduction in water tariffs. In Istanbul responsibility for water lies with the state-owned water company ISKI which is responsible for water pricing. The pricing model is based on the idea of “user pays” and ISKI aims to make a profit of 10%. This is in contrast to the old pricing model which was run by the local council and their decision on the tariff was without any profit seeking. All the other relevant water services were given to ISKI and in this way the central state subsidies were cut and the whole water provision process started to become market-based (Çınar, 2006a). The role of the DSI and Bank of Provinces weakened and local governments were empowered. However, after the 2000s, with the changes in the central state’s policies, a new round of institutional reorganisation was put into effect, thus DSI and Bank of Provinces gained their powers back. However, the important difference between the powers of the institutions in the past and today is the change in the form of financing. In the previous period, both sets of institutions were using state subsidies to fund water provision, but in the 2000s, the institutions of the state as well as those of the municipal authorities started to resort to borrowing from international financial markets for the financing of their investment in water provision (Çınar, 2006b).

3.2 Debt Debt has become far more significant in different ways and this was addressed in three of the case studies (E&W, Portugal and South Africa). There two aspects of debt in the sector that were addressed in the country studies. First, there is the debt incurred from borrowing by the water provider (public or private) and, second, there is debt that accumulates where water consumers have failed to pay their water bills. The increase in debt has been one of the most defining aspects of the development of sector financing in E&W. Some companies have greatly increased their net debt and the level of gearing (ratio of debt to equity) has increased dramatically since privatisation. In the latest price setting exercise for the water sector (PR14), the regulator, Ofwat, calculated the industry cost of capital based on an assumed gearing ratio of 60 to 70%. In the previous price review exercise in 2009, the estimated level of gearing for the sector was only 57.5%. There is diversity across the companies. The not-for-profit company, Welsh Water, has the lowest gearing level, at 61.7% while Yorkshire Water, owned by a group of largely financial investors based in Jersey, has the highest gearing ratio at 82.6%. Looking in detail at the components of gearing over the past decade, the case study shows that net debt has increased by an average of 74% while equity has declined by 37%. While the private water companies have invested extensively in the sector, the increase in gearing was not matched by an increase in fixed assets. Debt financing offers advantages over equity finance because it is cheaper and it is regarded as “tax efficient”. It is also clear that the increase in debt in some companies in E&W has been used to pay for dividends to shareholders (PWC 2013). These debts are not expected to be paid off any time soon with new debts taken on to pay off old ones. End users continue to pay the financing costs of such debts with almost one third of

22



household water bills going towards “return on capital” which covers interest and dividend payments. The interest charged in the annual accounts of England’s nine water and sewerage companies increased from £288m in 1993 to more than £2,000m in 2012 in real terms. In Portugal, during the 1990s, infrastructure investment was financed by borrowing and AdP was of pivotal importance in channelling external funding to the multi-municipal concessionaires that it controls operating in bulk water supply and waste water. AdP has had finance direct from the EU, long-term debt from EIB and bonds and short-term loans from the banking sector. While about 60% of AdP debt consisted of loans from the EIB, private banking debt accounted for about 20% of total debt in 2013 and includes loans from major foreign banks such as Deutsche Bank and DEXIA and domestic banks too. AdP also has recourse to bond markets (mentioned earlier). The high level of debt means that AdP is vulnerable to changes in interest rates, particularly to refinancing interest rates from the ECB. The cost of AdP’s debt increased from 2006 to 2008 due to the rise in the reference ECB interest rates which also influences the benchmark European interbank rate with EURIBOR. With the drop in ECB interest rates in late 2008 and 2009 the average interest rate dropped to the record low of 2.7% in 2010. However, in 2011 in the midst of the Portuguese sovereign debt crisis, the decoupling of domestic interest rates from the rest of the Eurozone is discernible. Interest rates rose from 2.7% in 2010 to 3.7% in 2012. Since then, and helped by the expansionist ECB monetary policy, interest rates have again dropped. Nonetheless, given the weight of debt in AdP, the bulk provider’s balance sheet and the renewed pressure for cost recovery tariffs, recent financial instability shows the vulnerability of consumers to fluctuations in international financial markets. Higher interest rates result in higher flows channelled from consumers, through tariffs, to the financial sector. The retail water sector in Portugal has been more vulnerable to fluctuations in financial markets than the bulk sector as the smaller scale of operations at the municipal level means that service providers have to pay higher rates of interest than AdP. Retail water companies have seen interest rates rise from 5.4% in 2001 to 7.4% in 2011. Private concessionaires were particularly affected by the rise in interest rates with higher levels of indebtedness (as measured by ratio of debt to assets). However, this has not prevented them from earning significant returns with operating margins in the region of 21.7% and considerably higher for some private concessionaires In South Africa, in contrast, debts of water boards (from borrowing) have fallen. Aggregate long-term debt has decreased from R7bn to R3bn between 2004 and 2011 while equity levels have almost tripled over the same period. This is mainly attributable to Rand Water and Umgeni Water reducing their debt levels substantially. While financialisation is not deeply embedded in the sector, a review by SALGA in 2013 indicated that a number of WBs were seeking prices to provide a higher net profit margin than previously. The justification for a higher margin is the need to obtain or maintain a minimum interest rate cover to satisfy lenders (SALGA 2013, p.8). The municipalities in South Africa are also not heavily indebted (at least in terms of loans) although they borrow, with some issuing bonds and taking out bank loans. The water utility in Durban, eThekwini, has stated its policy to minimise dependence on borrowing in order to reduce future revenue committed to debt servicing and redemption charges, and it maintains a gearing ratio below 50%.

23



These three case studies show diverse experiences with, and attitudes to, debt finance across and within countries. High debts are not an essential requirement for water financing, and some public utilities have made an effort to keep borrowing costs down. In E&W debt finance has grown dramatically for some companies and this is associated with a large increase in financing costs. While such an increase in any other expense (for example, labour) would be regarded as inefficient, distributions to the financial sector are not categorised in this way. Debts in the sector have also risen due to unpaid bills. Both Portugal and South Africa have a growing backlog of unpaid bills owed by municipalities to the bulk water provider. In Portugal, municipalities have often refused to adopt cost recovery principles despite pressure from the regulator, ERSAR. They have not increased tariffs and some do not charge for waste water at all. This has led to mounting debts owed to bulk concessionaires (eg AdP). These have escalated with rising bulk tariffs and deterioration of municipalities’ budgets with the new financial rules imposed by the troika memorandum. From 2007 to 2011 their debt almost doubled reaching 400m euros in 2011 and 535 million in 2013. In South Africa, there is a similar vulnerability in the sop. Municipalities owed US$320m for bulk water at the end of June 2014. The issue here is not that municipalities refuse to increase tariffs to cost recovery levels (as in Portugal) but that revenue collection is low at the municipal level. Analysis of municipal finances indicates that households have amassed substantial debts in unpaid water bills, and this has led to the municipalities’ failure to pay for bulk water. Reference is often made to what is termed a “culture of non-payment” which derives from protests in the apartheid era. In addition there are illegal connections among poor and wealthy communities. There is also compelling evidence to indicate that water tariffs are unaffordable for most of the population. In E&W, the proportion of household bills unpaid has been increasing and affordability is falling. These three case studies all raise concerns regarding increasing indebtedness. In E&W and South Africa, more households are struggling to pay their bills while, in Portugal, municipalities have kept prices low but are amassing debts to bulk water providers. Despite different structures and contexts, the financing of water in each of these sops is generating significant increases in indebtedness at different stages of the production chain which may be creating instability. In Poland water enterprises finance their activities primarily with their own capital. Foreign capital is mainly obtained from European Union funds with a relatively low share of commercial loans. Ownership concentration (with one gmina serving as an owner) led to obtaining a minimum net return on sales by an enterprise. The economic profit constituted a minor objective for such enterprises. In the majority, their owners did not aim at obtaining higher profits from the activities of water and sewage service enterprises and the realisation of the right to a dividend. According to Lis (2015, p. 34), such actions were “probably not accepted by the local communities”. In Istanbul, between 1998 and 2000, the investments of local governments in drinking water were as much as the investments of DSI and Bank of Provinces. However, almost half of the rising capacity of local government spending was gained from foreign finance. It was also evident in DSI’s finance sources that almost a quarter of

24



its investment was financed from outside the country and, in 2003, it has risen to cover half of the investments. However, the investments of Bank of Provinces remained in equities. While local government investments declined after 2000, central state institutions’ investment rose. The Bank of Provinces, as well as the DSI, started to increase their investment by obtaining foreign loans. For example, the Bank of Provinces signed a project worth 213 million Euros with the World Bank in 2006.

4 Consumption of water As mentioned above the sop analysis aims to consider the chains of provision that connect production with consumption. As shown in Table 1 above, access to water has increased considerably and is approaching universal coverage in all of the countries except South Africa where low rates of access persist in some rural areas. Consumption of water is not just about the provision of physical infrastructure. It also needs to be affordable and so is determined by pricing and income levels. Table 2 shows the price of water in Euro/m3 for the largest cities in each of the case study locations along with the figure for GDP per capita. Although incomes vary substantially there is not such a diversity in water prices. In Warsaw and London, water is charged at a flat rate regardless of volume consumed while Lisbon and Johannesburg have an “increasing block tariff” (IBT) discussed below. Table 2: Water prices in the largest cities of the case studies Quantity

(m3)/month Currency* Euro/m3 GDP/capita

(Euro) London (£) 1.32 1.02 47,237 Lisbon (Euro) Up to 5 0.25 26,805 6 to 15 0.65 16 to 25 1.53 >25 1.93 Johannesburg (Rand) Up to 6 Free 13,529 6 to 10 5.84 0.42 10 to15 9.27 0.66 15 to 20 12.91 0.92 20 to 30 16.86 1.21 30 to 40 17.88 1.28 >40 21.98 1.57 Warsaw (PLN) 4.20 .99 40,232 Istanbul Household 4.13 1.37 *Converted to Euro at rate prevailing on 31.12.2014 (www.xe.com) Sources: Thames Water Metered Charges, EPAL Water Charges, Johannesburg Water Charges; Brookings Global Metro Monitor Map 2013-14.14

14 http://www.brookings.edu/research/reports2/2015/01/22-‐global-‐metro-‐monitor (Accessed 9th March 2015).

25



In Poland tariffs are set every year on the basis of “necessary revenues” which are allocated across recipients in different ways and the water and sewage enterprise chooses the structure and type of tariff according to different criteria. Water tariffs are supposed to cover all the costs associated with running the business activities of enterprises in the water sector and to “bring a decent profit” although water companies also benefitted from some subsidies from the local authority (Lis 2015, p.26). Over the past three decades, in many countries, water has become classified as an economic good rather than a public service. This process has its origins in what are known as the ‘Dublin Principles’ the fourth of which reads: “Water has an economic value in all its competing uses and should be recognized as an economic good”. The principles were formally adopted at the United Nations World Summit in Rio in 1992 and have since been repeated in numerous policy and strategy documents (for example the Global Water Partnership founded in 1996 by the World Bank, UNDP and SIDA). Significantly for most of the case study countries, this approach to water is incorporated in the EU’s 2000 Water Framework Directive (WFD) (Directive 2000/60/EC). According to the WFD “Adequate water pricing acts as an incentive for the sustainable use of water resources and thus helps to achieve the environmental objectives of the Directive”.15 This policy approach is supported by a discourse of scarcity both of resources (in the context of climate change) and of finance (particularly since the financial crisis). EU member states had a deadline of 2010 to establish water-pricing policies according to the terms of the WFD. The principles of cost recovery and “polluter pays” were to be applied to all water services (EU 2010). Cost recovery pricing in E&W was introduced in the 1970s and predates the WFD. In South Africa cost recovery was a core principle of the country’s post-apartheid water policy as articulated in the Ministry’s 1994 White Paper. As shown above, the involvement of the EIB also suggests the South Africa has a commitment to the EU’s policies of “water efficiency” and “demand management”. All countries in the study incorporate some notion of “cost recovery” in their pricing policy which requires that water tariffs are set at a level that covers the cost of production. This also known as “economic pricing”. The two purported advantages in this way of setting prices are that, first, the water utility will be financially sustainable and, second, consumers will become aware of the economic cost of water production and will adjust their consumption appropriately (ie “demand management”). Cost recovery is promoted as an antidote to politically-motivated pricing where low water charges are used to garner political support. To be effective in reducing demand, cost recovery also requires end users to be metered. All of these water sector reforms are portrayed as measures to improve “water efficiency” (see for example EIB 2008; Ofwat 2014). There are, however, major limitations when it comes to applying the concept of cost recovery. While presented as scientific and politically neutral, there is considerable discretion in how cost recovery is adopted and calculated in practice. The allocation

15 http://ec.europa.eu/environment/water/water-framework/info/intro_en.htm

26



of costs is a question of negotiation and contestation. The evidence from the case studies indicates a number of ways in which the approach is moulded to suit some agents over others. Cost recovery in practice creates specific distributional outcomes in the sop and raises a number of concerns. First, capital investment costs have proven to be particularly malleable when it comes to applying cost recovery pricing techniques. For example, in E&W, prices are not based on actual costs but on assumptions about future costs. The terms of the Price Review in 2009, where prices were set for the subsequent five years, were expressly generous to water companies in part because the financial crisis was expected to impede firms’ ability to raise finance. In practice, financing costs were considerably lower than assumed and private companies have benefitted financially, effectively creating a transfer of finance from end users to shareholders in the name of cost recovery. Second, in all of the case studies – and most of the world – the construction of water infrastructure predates the application of cost recovery pricing. Historical costs tend to be neglected in cost recovery approaches. In E&W, the capital valuation of infrastructure at privatisation was a fraction of its actual value so company shares were oversubscribed at privatisation and share prices increased rapidly. The costs of pre-existing infrastructure do not appear in cost recovery calculations which tend to centre on new investment costs. This is a sharp issue in South Africa where pre-existing infrastructure served the white elite in wealthy areas, and these cost are not recovered. However, new investment costs to roll out the network in areas previously without infrastructure are ascribed to less wealthy areas. Cost recovery in practice seems to mean that new investment is charged and older investment is not, advantaging those who are already provided for. Third, where companies are allowed to make a return on capital investment, there is a bias towards capital investment expenditure over other costs. In E&W the pricing structure has created a “capex bias” and in Portugal there have been tensions between AdP and municipal providers. AdP has been investing more than is required for existing water demand. The municipalities been required to pay the costs of this investment expenditure in the name of cost recovery even though the costs are higher than warranted by demand. Fourth, both E&W and Portugal have seen a change in their cost structures with a declining share of income going to wages and an increase to financing costs which reflects falling trade union power, troika conditionality in Portugal and an increase in rentier payments. Higher labour costs are seen as an indicator of less efficient production while higher interest payments are not judged in the same way. The ethos of cost recovery, while superficially neutral and scientific, is in practice supportive of this changing social structure and its adverse distributional implications. Fifth, an obvious contradiction in the “demand management” concept is that, a fall in demand, can lead to an unwelcome reduction in incomes for private operators that are remunerated on the basis of units sold (see above). This is overcome in E&W with a “revenue correction mechanism”, whereby loss of income due to a reduction in consumption is recovered in a price increase in the next price review period. Thus the application of cost recovery principles to promote demand management means in

27



practice that customers have to pay a higher price for reducing their consumption in response to price signals. If you try to save money by using less water your price will go up. The less you use the more you pay per unit. Consumers collectively may as well do nothing. Finally, outcomes can be regressive. It is more expensive to provide water to remote rural areas and to high-density slum areas located on the margins of cities than affluent urban areas or high volume industrial consumers. Those that face higher water charges are often the less wealthy communities. The effects depend on the extent of decentralisation. Greater pooling and centralisation can lead to a more equitable distribution of costs across users. In South Africa, the adoption of cost recovery has meant that large volume water users that can access bulk water directly pay a lower unit price than households. In practice water prices are heavily contested. In E&W the price review process entails extensive negotiations between water companies and the regulator and considerable amounts are spent on consultants by all parties. In Portugal, the sector has been a battlefield between municipalities and the central state mediated by AdP and ERSAR regarding the definition of water tariffs. In Portugal since 2014 the regulator has been given the power to set prices and impose these on municipalities. This has been in part the result of the financial crisis creating a more favourable climate for the imposition of cost recovery principles with an emphasis on the financial stress of the sector. According to Lis (2015), in Poland, the water and sewage service sector is not supervised by a central regulatory authority. It is the role of every gmina to supervise the functioning of water and sewage enterprises under the Act of 7 June 2001 on collective water supply and collective sewage discharge Countries vary in the measures they take to support consumption by low-income households. Table 2 above shows that Portugal and South Africa do so through a lower tariff for low consumption levels. In South Africa, low-income households (and in some locations all households) are eligible for a free basic amount of water equal to six cubic metres a month, known as “free basic water” (FBW) policy. There are, however, problems with this. The IBT structure requires metering and so does not work for the poorest households who may lack a connection (more of a problem in South Africa than Portugal). Furthermore, high household size can quickly push a household to higher levels of consumption and this is more common in low-income families. In Portugal, this is addressed to some degree with the Family Water Tariff which charges a decreasing tariff for higher tiers for households containing five or more members. There is also a social tariff based on proof of low household income in most water providers. However, their scope is very limited, reaching only around 4% of consumers while the poverty rate in the country is 18%. In addition, the average subsidy is rather small at around 2.2 euros a month per household.16 ERSAR expects to impose social tariffs to all providers in 2016. The IBT is not implemented in E&W. There is some limited financial support for low income households but water companies are restricted in the social support that they are allowed to offer. There has to be a business case for support meaning that the

16 http://www.epal.pt/EPAL/en/menu/customers/tariff/special-tariffs

28



costs of support for disadvantaged customers must be “cost-neutral” and must be outweighed by the reduction in debt recovery costs. When it comes to the human right to water, countries also vary in the treatment of households that fail to pay their water bills. In E&W it is illegal to disconnect a water supply for non-payment. This is not the case in the other countries. In Poland, a water and sewage enterprise can cut the supply of water or close a sewage terminal if the recipient of the services has failed to settle the payment due for two complete accounting periods (usually two months), or if there has been illegal consumption of water or illegal discharge of sewage. If the supply is disconnected, the water enterprise is obliged to provide a supplementary point for drinking water. This is usually outside the house but within the town. But there is no data on how many have been disconnected from the water supply systems for non-payment. In Istanbul, water supplies are disconnected for non-payment and the water meter is removed if the debt is not paid within six months. In South Africa, disconnection for non-payment is routine and this can wipe out the social benefit of access to a free basic water supply. Bond and Dugard (2008) cite World Bank advice to the first post-apartheid water minister, Kader Asmal, that there needed to be a “credible threat” of cutting the service if the country was going to attract private investment in municipal water provision. More recently, given the potential public health risks associated with disconnecting household water supplies, innovative approaches have been explored such as installing flow-limiters and pre-payment meters for water. Some households have managed to keep consumption low to stay under the FBW threshold but for many this has meant that life is dominated by keeping water consumption to a minimum. An increased policy focus on revenue management has led in some cases to the contradictory situation where, on the one hand, government-funded infrastructure is rolled out to connect low-income areas while, on the other, these are then disconnected for non-payment. For the sop approach this is an example of the conflictual and contested ways in which the sop is constructed. In E&W, bad debts in the sector due to non-payment of bills have been increasing at a significantly faster rate than for other utilities even though prices for other services such as energy have increased faster and use up a higher proportion of incomes. Here, end users cannot be disconnected and debt support agencies such as the Citizens Advice Bureau advise households prioritise more significant bills (such as housing costs). There is considerable evidence cited in the case study to show that those that do not pay their bills are poorer households. The regulator, Ofwat, is exploring the possibility of introducing some of the punitive measures that have been implemented in South Africa that fall just short of disconnection, such as flow limiters. The case study reports showed diversity in terms of affordability. In E&W, average household bills have increased by 40% in real terms since privatisation in 1989. Prices have remained more or less steady for the past five years. However, real wages have fallen substantially since 2009 and a rising proportion of households is struggling to pay for water. In 2011-12, 12% of households spent more than 5% of their income on water and sewerage bills compared with 8% in 2001/02. There has been an increase in bad debts in the sector, and this has been rising faster than debts for other utilities.

29



In contrast, in Portugal water charges are low and just 1% of household expenses is devoted to water supply and waste water, which is below the OECD threshold of 3%, although there is an upward trend in prices. For the bottom income quintile, water charges represent 1.5% of household spending and there is considerable diversity across the country. The ratio of lowest to highest charges is 1:31. of average monthly bill. Meanwhile, in South Africa there is evidence cited in the case study to show that water is not affordable for 63% of households in the country. The Report by the Department for Water Affairs states: “Given the high percentage of households in South Africa that are not expected to be able to pay for water, it is thus unsurprising that the revenue collected by almost all municipalities is insufficient to cover operating and maintenance costs and that almost all municipalities are heavily dependent on operating subsidies. This in turn affects the ability of municipalities to pay the water board for bulk water charges” (DWA 2014 p.12). These findings support the commentary above, indicating that, while tariffs are affordable in Portugal, they are too low to cover the costs of bulk water. In South Africa, tariffs would cover costs but they are not paid because they are unaffordable. The result is the same with the retail water segment accumulating debts to bulk water providers in both countries. The case study countries provide some support for low income households, and in 2010 the United Nations General Assembly explicitly recognized the human right to water and sanitation (UN Resolution 64/292) and acknowledged that clean drinking water and sanitation are essential to the realisation of all human rights. Explicitly this means that everyone has the right to a minimum level of water for basic consumption. This has been set at 50-100 l/c/d (enough to meet basic needs according to WHO) and it needs to be “affordable”. On this, the UNDP suggests that water costs should not exceed 3% of household income. Countries vary in the extent to which the right to water is reflected in national legal frameworks. South Africa is one of the few countries that make an explicit reference to the right to water in national legislation.17 The human right to water is reflected to some degree in sector policy (although not explicitly linked to human rights). Social policy for water varies across the countries. The IBT provides cheap water to low volume consumers and free water to some households in South Africa. In E&W, a (slightly) discounted price is provided to a number disadvantaged households that meet particular criteria. As with the rights-based approach, these polices are based on meeting the minimum required for basic needs. Where “equity” comes into policy debates, this relates to providing individuals with a small amount of consumption. There is little support for more substantive measures to improve equality of outcomes in the sector. Given that everyone has to consume water, alternative financing methods could lead to more equitable outcomes, for example financing out of progressive income tax or (as is still the case in much of the UK) water bills based on rateable property values. However, the neoliberal ethos, is enshrined in the EU WFD which rules out alternative approaches to sector finance.

17 See righttowater.info for details of the status of the right to water in different countries.

30



A legal challenge was raised in the courts in South Africa on the grounds that the FBW amount was not sufficient and the prepayment meters were unlawful as the South African Constitution guaranteed a right to water. After appeal it was ruled that 42 l/c/d is sufficient for the FBW amount and prepayment meters are lawful. 18 The rights-based approach then comes down to tweaking the limits of how much water is necessary for consumption. The wider inequalities escape scrutiny. The international water rights framework does not exclude cost recovery for water services and nor does it stipulate the provision of free water or the public ownership of water supply. It simply specifies that everyone is entitled to affordable water for personal and domestic uses. This basic needs approach, rather than calling for a fair distribution of water access and pooling of resources, only provides for the bare minimum for basic survival for the most disadvantaged and does not affect the high volume consumers. This could be considered to be not so much about redistribution or social equity so much as smoothing away the harshest social costs of neoliberalism.

5 Role of the state For the sop approach, the role of the state is understood in terms of relations between agents. This is reflected in but not limited to the institutional structures. The state is not a monolithic entity but has diverse and sometimes conflicting roles in the provision of water and these vary across countries. The state is the provider of water, the regulator, the financier and also a consumer. Even within some facets of state activity there can be conflict (for example with environmental and economic regulation). States themselves have to respond to supra-national agents. Most significant in the case studies is the European Union which sets policies in the WFD which need to be adopted by the governments of member states. A number of publications in the last decade have concluded that access to water is a political issue. Most cities produce enough drinking water to satisfy human health but the available water is often distributed in a highly unequal manner (Swyngedouw 2006, UNDP 2006). Access and affordability are political questions related to economic and political power rather than absolute scarcity. This is borne out in different ways in each of the case studies. States set the parameters which allow interest groups to access and use water resources in different ways with consequences for the distribution of wealth and power in society. The institutional structure varies in the extent to which the provision of water is separate from other operations of the state. In E&W, water companies operate independently of the state apart from regulation. Meanwhile, in South Africa, water is one of several basic services provided by municipalities with water revenue incorporated into the overall revenue base. In Portugal also for most municipalities, water revenue goes into the general finance pool. However, in both Portugal and South Africa municipal debts to bulk water providers have been rising. In SA, there have been suggestions that water-related central government subsidies to municipalities should be paid directly to bulk providers. In Portugal, the regulator, ERSAR, is fighting to require municipalities to ring fence water revenues and force those that are indebted to ADP to transfer at least half the tariff revenue to bulk

18 See court ruling, Mazibuko and Others v City of Johannesburg and Others, www.saflii.org

31



providers. In Poland water providers are regulated by the local authority (gmina) which is responsible for authorising tariffs but there has been an on-going discussion in the country about introducing a central regulator. Lis (2015) raises some concerns about the role of the gmina as the regulator of water. The gmina is not independent, often having a share in the ownership of the water provider and also subject to political pressure to lower the price of water and sewage. According to Lis (and citing the President of the Chamber of Commerce, p.31) water is still regarded as a political vote-winner. Local government officials keep prices down and use the promise of cheap water as an election banner. In addition, gminas often lack the technical capacity for effective regulation and the council of gminas failed to adopt a resolution on the authorisation of tariffs. In addition, the establishment of a central regulator raises administrative challenges, for example with such a large number of gminas (and there are 2,479) and each with their own tariff. There are different approaches to regulation in the countries studies. In E&W the economic regulator, Ofwat, has considerable influence on the sector, and there have been recent innovations to try to increase the role of consumers in corporate control with the creation of Customer Challenge Groups (CCGs), established at the behest of the state. Portugal also has an independent regulator, ERSAR, managing over 500 operators in the water and waste sector. Unlike Ofwat, ERSAR regulates water quality while in E&W this is the responsibility of a separate government agency, the Drinking Water Inspectorate. In Poland, water enterprises are regulated by local authorities which operate on such a small scale that many lack capacity for effective regulation, and there are calls for a national regulator. In South Africa, regulation is the work of different elements of government. The Department for Water Affairs is responsible for water but the Treasury sets limits on municipal borrowing. In Turkey, the role of the state in water provision and management has changed over time. Local governments were the main state bodies in the provision of water until the 1930s. Since then, water provision was centralised and different institutions (i.e. DSI, ISKI and Bank of Provinces) have shared the power of water provision and management. However, as Turkey has become more involved with global economic dynamics and developed closer relations with the World Bank, there has been a transfer of power to the local municipalities in order to source foreign debt and create public private partnerships. In South Africa large consumers are able to pay in order to ensure that they receive a more secure water supply than others (known as “high assurance” consumers). The country’s “economically viable” consumers of raw water are encouraged to finance their own water infrastructure “off-budget” so that they can by-pass government funding procedures. Since 1994 policy has been based on separating those that can afford services, and for whom providing water was ‘financially viable’, from those who cannot and who need to receive government subsidy. The state then is left with the provision of services that are not economically viable and does not have revenue to do so from the viable elements. The scope for cross subsidy is diminished. The state also faces competing priorities with pressure to promote economic development. These tensions are clearest in South Africa, where mining development threatens to pollute the water supply for Gauteng Province and poor water quality can have far-reaching impacts, for example on agricultural production. Yet the National Development Plan proposed expansion in agricultural activity even though water

32



supplies are likely to be insufficient. The case study shows that the state is pursuing several agendas simultaneously and these are in conflict with one another with resolution liable, in the case of water, to the better off and powerful. Water prices are set by the state. As shown above, this is far from a scientific or politically neutral process. Rather the price reflects a distribution of finance across end users and producers. As a result water pricing is a contentious issue. In E&W, pricing is the source of intensive negotiations between the regulator and the private water companies with the regulator wanting to keep prices down and companies pushing for as high a rise as possible. Prices are contested in different ways elsewhere. In Portugal municipalities disregard instructions from the regulator, ERSAR, to increase prices to cost recovery levels and keep them deliberately low. In Poland also, it appears that local authorities have manipulated water prices to keep them low in order to garner political support. Clearly cost recovery policy and rhetoric are distinct from practice. The case studies found considerable diversity both across and within the case studies regarding the capacity of state organisations. In Portugal, the state bulk water provider, AdP, operating on a national scale, has been able to acquire financial skills and accesses domestic and foreign financial markets. In contrast, some small municipalities have shown capacity constraints, for example in negotiations with private concessionaires and in monitoring contracts. In South Africa, some water boards have been heavily reliant on government subsidy while others (Rand and Umgeni) issue bonds on the capital markets. An assessment of water service authorities in South Africa found that most are not performing well. Revenue collection was the indicator with the greatest failings. Capacity constraints lead to a downward spiral of weak revenue and poorly maintained infrastructure and deteriorating quality. Water treatment plants, bulk water supply, reticulation and water storage systems are not well maintained. Many are in disrepair across the country and sanitation systems are often in a very poor condition. Weak state capacity has adverse impact on social policy. In South Africa central government grants have been underspent due in part because local municipalities have not been able to disburse the funds. This can mean that grant funding is reduced in the following year. Weak municipalities are less able to monitor private contractors, and this increases the risk of corruption and maladministration. This has also been the case with private concessions in Portugal. In South Africa, the municipalities that most need the funds are the least able to spend them. According to a WRC study in South Africa, “municipalities are continuously in a crisis management mode with limited management information and poor decision-making processes, financial and technical management” (McKenzie et al 2012 p x). In E&W the state has a regulatory function and does not provide water directly. Capacity levels are not regarded as a limiting factor in this sop. However the case study paper highlights some constraints and challenges in the regulatory process. The price of water is a locus of intense negotiation between the regulator and private water companies. The state’s position is affected by the practical challenges of dealing with the uncertainties of future costs, the need to ensure that the sector (and the country) is seen as an attractive investment opportunity and the impenetrable financial

33



engineering conducted by some private operators. While state capacity is not a constraint on the provision of water, the financial innovation in the sector is increasingly difficult to control and greater reliance is placed on external financial monitoring bodies such as credit ratings agencies to provide information of company performance. Each of the case studies shows that sector finances are manipulated for political ends in different directions. In E&W, private companies are pushing for high prices in order to increase shareholder returns. The state is faced with a balancing act in terms of meeting investors’ demands and consumer welfare. In Poland and Portugal, however, the municipalities want to keep prices low to suit local political ends. While cost recovery pricing has been adopted throughout the case studies, in practice this is open to manipulation to suit political ends. In Portugal the EU has a substantial impact on policy with three particular aspects that shape the water sop. First, the Maastricht Treaty fiscal criteria required a budget deficit of at most 3% of GDP which led to large constraints on public investment. As a result, direct public investment in water after 1993 was almost non-existent. At the same time, to meet the terms of the WFD, substantial investment was required and much of this was financed by direct fiscal transfers from the EU (between 1993 and 2012, 28% of investment by AdP was funded this way). Finally, the WFD requires that prices cover costs. This structure created a need to source new funding arrangements which included direct subsidies from the EU and loans from the EIB with long maturities and low interest rates. The need for more investment while setting limits on government borrowing leads governments to turn to the private sector. The neoliberal framework in Portugal has been driven more or less by a set of external constraints and opportunities. But the transformation has also been promoted by domestic agents and has been beneficial for particular segments in the country in particular financial and construction sectors. In E&W where the national infrastructure plan relies heavily on private investors, the state needs to ensure that their needs are prioritised (over end users). Companies go to considerable lengths and expense to lobby national governments and the EU to promote their interests and this must have some impact (or they would presumably stop). The case studies show diversity and conflict in the role of the state as different agents compete for economic and social control. In Portugal there is tension between municipalities and the regulator, in E&W between private water companies and the regulator, while in South Africa social tensions are rising as basic services are provided to large industrial consumers but denied to low-income households. States have responded to such tensions in different ways, deriving from historically-evolved structures and processes.

6 Conclusion This review of the financing of water shows that similar processes and structures can be observed to varying degrees across the countries, and these are underpinned by a

34



neoliberal ethos in each case. The sop approach highlights the contestation and conflict between agents that leads to diverse results from the implementation of similar policies. Contestation emerges over pricing with upward pressure in some locations (from private providers in E&W) and downward pressure in others (from local politicians in Poland). Contestation is also evident in the battle for control of water and this is particularly the case in Portugal where municipalities have been strongly resistant to the growing power of the regulator, ERSAR. The National Association of Portuguese Municipalities (ANMP) has been opposed to the role of ERSAR in setting tariffs and the privatisation of the waste services companies. ANMP also supports the on-going legal battle between some municipalities and private concessionaires in retail systems. There is also contestation over usage of water and this is mostly predominant in South Africa with economic expansion in the National Development Plan requiring more water than is available in the country. Water is directed to industries with economic and political weight while poor households often struggle to pay and many are disconnected. Despite the contradictions and questionable welfare impact of neoliberal water policies, opposition is not widespread outside South Africa. In Portugal, there is very little resistance to tariff increases. The resistance movements that exist are organised around the trade unions in opposition to privatisation. In South Africa, there have been extensive protests against the policy of disconnection and forced installation of prepayment meters with residents of Soweto mounting a legal challenge against the City of Johannesburg.19 In E&W water provision gets little media attention compared with other basic services that consume a larger proportion of the household budget such as housing and energy bills. This comparative study of water provision shows that while the countries have adopted similar neoliberal policies, these have played out differently depending on the context. The role of private financial capital has been highly significant in England and Wales while in South Africa the system of provision continues to be framed by the extensive inequality of the apartheid era and the country’s industrial structure. In Portugal the position of the country in the EU periphery has been significant and Poland’s relatively recent transition has shaped sector development. This study has also brought out some commonalities in the different contexts. The state has had a significant role in providing services, despite efforts to increase private management and finance. Even where provision is fully privatised in E&W, the state is closely involved through regulation. The policy paradigm that is prevalent in all of the case studies is one of linking costs and charges as closely as possible to the individual consumer, with metering and initiatives such as “cost recovery” and “user pays”. However, the discussion above demonstrates that water costs are not always clearly identifiable nor easily assigned to individuals or groups. Furthermore costs are likely to be highest for the most difficult to reach and those who need new investment. Taken to its logical conclusion, this policy approach risks charging the highest prices to those who can least afford it. Social policy in the sector is strongly underpinned by a business perspective and designed to smooth the rough edges off the

19 See court ruling, Mazibuko and Others v City of Johannesburg and Others, www.saflii.org

35



above structure. Equity is addressed in terms of meeting basic needs rather than addressing redistribution. This is, however, the only policy game in town, particularly with the weight of the IFIs ruling out alternatives. Neoliberal policies are widely depicted as politically neutral and scientifically objective but, in the water sector, the sop approach shows that there are winners and losers and outcomes stem from embedded power relations which are specific to the individual locations.

36



References

Bakker, K (2005) “Neoliberalizing Nature? Market Environmentalism in Water Supply in England and Wales” Annals of the Association of American Geographers vol 95, no 3, pp. 542–565.

Bakker, K. (2007) An Uncooperative Commodity: Privatizing Water in England and Wales Oxford: Oxford University Press.

Bayliss, K, B. Fine, and M. Robertson (2013) “From Financialisation to Consumption: the System of Provision Approach Applied to Housing and Water”, FESSUD Working Paper Series, no 2. Bond, P. and J. Dugard (2008) “Water, Human Rights and Social Conflict: South African Experiences” Law, Social Justice and Global Development http://www2.warwick.ac.uk/fac/soc/law/elj/lgd/2008_1/bond_dugard/bond_dugard.pf.

CCWater (2013) “Valuing Every Drop – How can we encourage efficiency and innovation in water supply?” Consumer Council for Water in England and Wales.

Çınar, T. (2006a) “Türkiye’de İçmesuyu ve Kanalizasyon Hizmetleri: Yönetim ve Finansman”, Su Yönetimi: Küresel Politika ve Uygulamalara Eleştiri (içinde), Tayfun Çınar ve Hülya K. Özdinç (Eds.), pp. 227-252, Memleket, Ankara,

Çınar, T. (2006b) “Neoliberal Su Politikaları Doğrultusunda İller Bankası, DSİ ve Belediyelerin Değişen Rolü”, TMMOB Jeoloji Mühendisleri Odası Bülteni, vol 3, pp. 70-78.

De la Motte, R. (2005) “WaterTime Case study – Gdansk. Poland” FP5 Research Project “WaterTime” Deliverable D20, University of Greenwich.

DWA (2014) “Water Boards as Regional Water Utilities (Institutional Arrangements to Facilitate Turnaround)” Department Water Affairs, South Africa.

EIB (2008) “The EIB’s Water Sector Lending Policy: Strengthening the EIB’s support for EU policy objectives in the sector” European Investment Bank, Luxembourg. EU (2006) “Pricing and long-term management of water” Communication from the Commission to the Council, European Parliament and Economic and Social Committee: Pricing and sustainable management of water resources [COM(2000) 477 - Not published in the Official Journal]. Updated 22.09.2006 http://eur-lex.europa.eu/legal-content/EN/TXT/?uri=uriserv:l28112. EU (2010) “Water is for life: How the Water Framework Directive helps safeguard Europe’s resources” Publications Office of the European Union, Luxembourg.

Fine, B. (2013) “Towards a Material Culture of Financialisation” FESSUD Working Paper Series, no 15.

Fisher, J., A. Cotton and B. Reed (2005) “Learning from the past: Delivery of water and sanitation services to the poor in 19th Century Britain” Background Report for WELL, Briefing Note 9, Water Engineering and Development Centre, Loughborough University.

Galiński, P. (2013) “Development of the Municipal Bond Market in Poland After 1989” Ekonomika vol 92, no. 2, pp. 122-136.

37



Gandy, M. (2004) “Water, Space and the Modern City” City vol 8, no 3, pp. 363-379. GUS (2013) “Ochrona środowiska 2013” Główny Urząd Statystyczny Warszawa.

Hall, D., E. Lobina and R. de la Motte (2005) “Public resistance to privatization in water and energy” Development in Practice vol 15, nos 3&4, pp. 286-301.

Herrera, V. and A. Post (2014) “Can Developing Countries both Decentralise and Depoliticise Urban Water Services? Evaluating the Legacy of the 1990s Reform Wave” World Development vol 64, pp. 621-641. Kishimoto, S., E. Lobina, and O. Petitjean (eds.) (2015) Our public water future: The global experience with remunicipalisation TNI, PSIRU, MDP and EPSU, Amsterdam, London, Paris, Cape Town, Brussels.

Lis, P. (2015) “Financialisation of the water sector in Poland” FESSUD Working Paper no 101.

McKenzie, R., Z. Siqalaba and W. Wegelin (2012) “The State of Non-Revenue Water in South Africa” Report for Water Research Commission, South Africa.

Mosse, D. (2008) “Epilogue: The Cultural Politics of Water – A Comparative Perspective” Journal of Southern African Studies vol 34, no 4, pp .939- 948.

NAO (2013) “Review of the VFM assessment process for PFI: Briefing for the House of Commons Treasury Select Committee” National Audit Office, London.

Ofwat (2014) “Consultation on retail controls for the 2014 price review” Ofwat Water Services Regulation Authority, Birmingham.

PWC (2006) “Hybrid PPPs: Levering EU funds and private capital” Report prepared by PricewaterhouseCoopers for PPIAF and the World Bank.

PWC (2013) “Cost of capital for PR14: Methodological considerations” Report prepared for Ofwat.

Swyngedouw, E. (2005) “Dispossessing H2O: The Contested Terrain of Water Privatization” Capitalism Nature Socialism vol 16, no 1, pp. 81-98.

38



Synthesis Report Part 2:

The system of provision for housing in selected case study countries20

Mary Robertson, University of Leeds Abstract This paper reviews five case studies prepared for Deliverable 8.25 for the EU research programme, Financialisation, economy, society and sustainable development (FESSUD). The case studies – carried out for the UK, Poland, Portugal, South Africa and Turkey – look at the systems of housing provision in each country, focussing on the ways in which they have been shaped by financialisation. The country findings are compared and contrasted across key components of the housing system – development and house purchase finance, land, production, the state, and consumption. For each component, the housing systems were found to be highly variegated, reflecting the particular social and economic conditions in each country. Supply structures, in particular, differed greatly, leading to divergences in the price responsiveness of housebuilding. Nonetheless, some common patterns are identified. A general trend towards commodification has taken the form of expanding owner-occupation, which has served to expand financial markets and incorporate households further within them. The goal of transforming housing into a private asset has been pervasive, though with varying degrees of success. The same is true of land, with land use in all the case study countries increasingly governed by the criterion of highest monetary value, and land and real estate used to appropriate ground rent.

Introduction

This paper provides a comparative overview of case studies of the housing systems in five different countries (the UK (Robertson, 2014), Portugal (Santos et al, 2015), Poland (Lis, 2015), Turkey (Çelik et al, 2015), and South Africa (Isaacs, 2015)).21 The case studies employ the systems of provision (sop) approach, which views the economy as divided into unique, integral and overlapping chains of activities – ‘sops’ - that underpin the provision of different goods. Consumption is said to be determined by the consumption cultures that emerge from the material conditions of provision within these sops.

Being the product of historical developments in different countries, sops are time- and place-specific and the approach is methodologically committed to investigating sops as such. This commitment is reflected in the wide diversity of housing practices, issues and outcomes that emerge across the five case studies, a diversity that is compounded by the way in which the case studies encompass not only countries at different stages of development but also, to some extent, different aspects of housing provision within these countries. For example, the three European countries considered – the UK, Portugal, and Poland – represent a core, a peripheral, and a transitional European country, respectively, while the Turkish case brings the perspective of a middle-income country and South Africa that of a developing country to

20 This work builds on the extensive work carried out for the five case studies. Grateful thanks for additional input and comments on earlier drafts to: Piotr Lis, Ozlem Celik, Gilad Isaacs, Ana Santos, Kate Bayliss and Ben Fine. 21 These case studies are referred to throughout without further citation.

39



bear, though all with their own specific characteristics. In addition to this, the UK, Portugal and Poland studies focus on owner-occupation, with other tenures, which tend to cater to lower income groups, dealt with only cursorily; Turkey focuses on Istanbul; South Africa on low-income housing.

In combination, the diversity of housing systems considered, and a commitment to studying them concretely, poses a problem of incommensurability across the case studies. The problem is avoided in virtue of two areas of common ground across the different countries. The first arises because any housing sop has certain core and irreducible components, namely, finance, land, production, labour, consumption and the state. While the form and nature of these components differ across the case studies, all the components are present in all of them in some way. The analysis below is thus organised around these essential activities. The second area of common ground concerns the forces and processes that have shaped the operation of the different housing sops over the periods considered. For while the sop approach shuns generalisation in favour of theoretically-informed specificity, it is also attuned to the global and systemic contexts within which different sops are shaped. This is reflected, for example, in the way in which the overlapping logics of capital accumulation, land value maximisation, and the expansion of financial markets are strongly felt in all the different case studies, albeit in highly differentiated ways.

For these two reasons, common themes and issues are present across the different case studies, providing interesting contrasts and comparisons. It will help subsequent analysis of these themes and issues to provide a brief summary of the findings of each case study. This is done in the next section. Subsequent sections compare the five different housing sops across the core areas of: finance, land, production, labour, the state, and consumption. The final section concludes by considering the extent, nature and impact of financialisation on the different housing sops in light of the preceding discussion.

Overview – demarcating the housing sops

UK

The UK case study focuses on the period since the 1980s, when two key changes occurred. First, the vast majority of the social housing stock was privatised through the discounted sale of local government-owned property to sitting tenants. With the aid of state-led discourses that promoted owner-occupation as ‘natural’ and innately desirable, this led to owner-occupation becoming the dominant tenure form and fed demand for mortgage lending. Second, mortgage market reform removed building societies’ privileged access to mortgage markets and opened them up to competition from banks. With the aid of increased international capital flows and, from the mid-1990s, low interest rates, the supply of mortgage credit increased dramatically.

Together, these two changes led to a vast increase in the amount of credit chasing residential land. Owing, on the one hand, to Britain’s restrictive planning system and, on the other, to the speculative nature of its private housebuilders, this demand tended to be channelled more into house prices than into housing supply. Housebuilders have made an increasing portion of their profit from land uplift rather than housing production, leading them to stagger output, and to a deterioration of employment conditions, skills, and technical progress. This pursuit of capitalised ground rent by housebuilders has been replicated by consumers, as homeowners have increasingly treated their home as an asset as well as a form of shelter,

40



leaving private and social renters doubly disadvantaged by both deteriorating housing conditions and exclusion from this form of wealth accumulation.

South Africa

The South Africa case study begins with the transition from apartheid in the early 1990s. The apartheid regime left a legacy of racial segregation whereby the black population was, by and large, confined to rural areas, shantytowns, and publically- or privately-owned hostels in townships on the periphery of urban areas or adjacent to sites of work, such as mines. It also left a one and a half million shortfall in housing units. While the post-apartheid African National Congress (ANC) government was committed in principle to a right to decent housing for all, the shape of post-apartheid policy was heavily shaped by the pro-business think tank, the Urban Foundation (UF). The UF sought the spread of homeownership among the black working and emerging middle class as an avenue to political acquiescence and consumerism. The resulting compromise between the different forces shaping post-apartheid housing policy was the mass provision of standardised serviced sites and/or basic (“top”) structures through a one-off capital subsidy, with an expectation that households would take responsibility for adding to and improving these structures over time by borrowing in mortgage markets.

This model delivered one million housing units over six years and has continued to dominate housing provision for South Africa’s black population. However, it has run into a number of problems. Most notably, the poor’s limited access to credit has impeded their ability to upgrade their homes, and the state’s commitment to market-based distribution of land has confined provision to urban peripheries with concomitant problems regarding access to employment and infrastructure. Furthermore, while units were initially built by large construction firms, low profit margins due to rising quality requirements and building costs, led to their withdrawal. This, along with government’s more general desire to shift responsibility for delivery to local government, led to local authorities taking a more active role in delivery. The limited capacity of smaller construction firms, fiscal constraints, a stubborn commitment to providing free-standing units, corruption, and local government weaknesses held back delivery in this period and the housing backlog had grown to 2.1 million by 2014. In recognition of some of these problems, policy emphasis later shifted to ‘sustainable human settlements’ and tenure diversification, though implementation of these new policy dimensions has been limited. More generally, broader financialisation trends within the economy – in particular, increased investment in the middle and high end property markets spurred by speculative capital inflows – have meant that houses in the formal market are unaffordable to the working poor and lower-middle class. The promotion of housing as a ‘financial and economic asset’ by Government and pro-business groups – itself a manifestation of market-oriented policy – has failed in part because of the dysfunctional nature of township property markets.

Turkey

The Turkish case study focuses on Istanbul as the epitomy of processes of state-led gentrification that have dominated housing provision since the 2000s. The origin of these processes lies in the development of squatter communities, or ‘gecekondu’, arising out of mass urban in-migration in the 1950s-70s. In the absence of state-supported social housing, the gecekondu were socially, if implicitly, perceived as a solution to the problem of housing poor urban migrants, and property rights over settled land were granted through subsequent amnesties. This in turn facilitated the upgrading of gecekondu properties and capture of rising

41



urban ground rents by residents or local strongmen. The gecekondu co-existed with the ‘build-and-sell’ model utilised by small contractors to provide housing to the middle classes. Under this model, the contractor secured access to land at minimal capital outlay by promising the landowner a share of the proceeds of development. They then sold basement and first floor flats to finance production of higher level flats. The ground rents arising from rapid urban development, and an underdeveloped financial sector, meant housing provision was heavily shaped by land speculation in this period. However, rising land values, higher interest rates, and middle class relocation to new flats from the 1980s saw the gradual erosion of this model and its replacement by state-backed mass housing developments.

It was against this background that a newly-empowered Mass Housing Development Administration (TOKI) intervened to create a state-led construction boom in the early 2000s. TOKI acquired comprehensive land-use and planning rights, which it has utilised to deliver the mass production of housing units. This delivery has exhibited two controversial characteristics. First, TOKI has made high value, inner-city urban land available to private developers serving top-end markets, in exchange for a share of proceeds under its ‘revenue-sharing’ model. This revenue is then used to provide housing for lower- and middle-income groups on peripheral land, often with the aid of government-issued mortgage loans. However, despite the provision of low-cost housing being TOKI’s raison d’être, it is alleged that many of the dwellings built by TOKI are unaffordable to low income groups. Second, this high value, inner-city land has often been acquired through the displacement of gecekondu and dilapidated housing areas, including historical areas and affordable housing units, under the guise of urban regeneration. These features have made TOKI’s role in contemporary housing provision highly contested.

Portugal

The Portuguese case in part resembles that of the UK in documenting how financial liberalisation facilitated the growth of mortgage markets and owner-occupation. In Portugal, these processes were aided by state subsidies and tax breaks for mortgages, and by the imperatives created by European integration, which included cheaper credit resulting from lower levels of compulsory reserves. However, unlike the UK, bank lending to construction firms, lax land use regulation, and state investment in infrastructure meant that, in Portugal, the boom in mortgage lending was accompanied by a construction boom. This boom tailed off rapidly in 2002 as the result of the withdrawal of state support for mortgages and infrastructure investment, but a hospitable financial climate sustained mortgage lending and, consequently, the secondary real estate market and house prices. This has meant that those able to access the mortgage market – generally, those with higher incomes - have benefitted from both low housing costs and appreciating housing assets, if at lower rates than seen in the UK. The losers in the Portuguese housing sop have been those excluded from mortgage markets, who, notwithstanding a small number of state-provided socially rented housing estates on urban peripheries, have been confined to a rental sector that is either costly or rent-controlled and suffering from under-investment.

Poland

The study of Poland covers the period of transition from a socialist to a market system from 1990. Whereas the socialist era was characterised by heavy state subsidies, administered rents, and a prevalence of housing cooperatives, the period since 1990 has been characterised by privatisation and decentralisation. The rapid sale of dwellings to sitting tenants at heavily discounted rates accomplished a sudden shift to a model of housing provision in which

42



owner-occupation became dominant. In conjunction with financial liberalisation and state subsidies, the shift to owner-occupation led to a growth in mortgage markets, with a heavy presence of foreign currency (predominantly Swiss Franc) loans in the years 2000-12. Though the volume of mortgage lending was less dramatic than seen in the UK or Portugal, Poland exhibited one of the highest rates of mortgage market growth in the EU28. In this model, most new housing was provided by the private sector, which combined private development for sale and self-promotion, whereby households manage the delivery of their own houses.

While Poland’s ‘shock therapy’ accomplished a sudden shift to a market-based system, it also gave birth to a number of dysfunctionalities. Most new homeowners lacked the means to invest in the improvement of their property, meaning that the deterioration of the housing stock has continued. Foreign currency-denominated mortgages created distress for many mortgages when the Franc rose against the Polish Zloty in the midst of the financial crisis. As a result, the Polish Financial Supervision Authority introduced new regulations pertaining to these loans in 2011, leading to a marked decline in foreign currency mortgage lending since 2012. An under-developed planning system has favoured small-site development (often self-promotion) on urban peripheries, while impeding the preparation of larger urban sites, thus leading to chaotic urban sprawl with limited access to infrastructure. Meanwhile, responsibility for housing needy households was assigned to local authorities, but their ability to meet this responsibility has been impeded by their limited resources, in part reflecting the discounts to sitting tenants that accompanied privatisation.

Finance

Having provided an overview of the five case studies, I now turn to their analysis, beginning with finance. While finance may intervene in housing sops in multiple ways, it is useful to distinguish between the two broad categories of consumption finance (which concerns the role of finance in households’ demand and access to housing, principally in the form of mortgages) and development finance (which concerns the role of finance in housebuilding).

Consumption Finance

Mortgage market growth, as the counterpart to the promotion of owner-occupation, has been a feature of all the sops, though the extent and character of this growth have varied. The UK experienced the highest levels of mortgage lending. Total gross residential mortgage lending increased by 521% between 1995 and 2005, while the ratio of mortgage debt to GDP increased by 50%, from 53.3% to 80%, over the same period. In Portugal, household debt increased from 35% of disposable income in 1995 to a peak of 130.5% in 2009, with the share of housing debt in the total rising from 70% in 1995 to 81% in 2011. The average in Portugal was 71%, a bit below the EU average of 79%. In Poland, households’ housing debt grew from 0.2% of GDP at end of 1996 to 20.5% at the end of 2013. This is low compared to other EU countries but, nonetheless, represented a relatively rapid growth in the years to 2012, when Poland’s financial regulatory framework was tightened. The maximum LTV ratio in Poland ranged from 80-110% for loans made in Polish zloty and 70-120% for those made in foreign currency. Mortgage lending in Turkey has also displayed rapid growth from a very low base. There, the share of mortgages in total consumer credits increased from 4.2% to 37.5% over the period 2001-6.

For middle- and upper-income groups, the pattern is similar in South Africa, but poor black households – on which the South African study focuses – have been persistently excluded

43



from mortgage markets. For example, only 14% of subsidised housing was credit-linked over the period 1994-7. This reflected the poor quality and location of subsidised start homes - which meant they were not seen as providing sufficient collateral – as well as an unwillingness to take on housing debt among the poor. The pattern of mortgage markets catering to the better off is replicated elsewhere. Indeed, only in the UK was there anything resembling the US subprime market, and this disappeared after the financial crisis.

In all of the case studies, increased international capital flows creating a surplus of finance for lending was viewed as a factor in explaining increased mortgage lending. The state has also played a central role in facilitating mortgage lending. First and foremost this involved creating a regulatory environment that allowed mortgage lending to take place, but states have also supported mortgage lending more directly through subsidies. Mortgage subsidies were in place in the UK until the 1990s, and in Portugal and Poland until the 2000s. The state in Turkey continues to provide mortgages to low-income households.

In South Africa, intervention has been less direct. With subsidies for the poor focused on capital investment, the state has sought to encourage lending using the stick of veiled threats of nationalisation accompanied by the carrot of government pledges to stabilise the housing market. This gave rise to a Record of Understanding between government and banks in October 1994, but failed to significantly increase lending to poor households. As a result of the Record’s failure, the government briefly pursued a more adversarial approach, culminating with the 2002 Community Reinvestment Bill targeting discrimination in credit lending. However, this approach was quickly side-lined by the adoption of a new voluntary agreement between the government and the banking sector – the Financial Services Charter – which provided the basis for a new Memorandum of Understanding. Nonetheless, lending the poor households remains limited, and the South Africa case study argues that the country faces a chicken and egg dilemma with banks arguing that the government should first stabilise the housing market in order to create appropriate conditions for credit to enter, and the government arguing the credit entry was needed to stabilise the low-income housing market.

Mortgage markets in the case study countries have also developed specific features, reflecting the particularities of the countries concerned. As mentioned, the UK mortgage market has been the most securitised, which has meant that UK housing assets have been packaged up and traded on international financial markets. The UK also developed a pre-crisis subprime market, albeit one far more limited than that of the USA. In Portugal mortgage lending was encouraged by processes of European integration, which removed exchange rate risk and lowered interest rates. In Poland, foreign currency loans – predominantly in Swiss Francs - have been a major feature of the mortgage market, reflecting high interest rates in Poland and the appreciation of the Polish Zloty against the Franc. The ‘flight to safety’ on international markets in the midst of the financial crisis caused the Swiss Franc to appreciate, which in turn created repayment distress for many Polish mortgagees. However, the fall-out has been more contained than in other East European countries with high levels of foreign currency mortgages – the result of forex borrowers in Poland tending to be wealthier than those borrowing in Zloty, as well as regulation introduced in 2009 and 2011.22 Foreign currency loans have also been a feature of the Turkish mortgage market, though so far this market has proved more stable. Microfinance is a distinctive feature of the financial landscape of developing countries, though in South Africa its success in relation to mortgages has been limited. A combination of larger banks capturing the most secure loans and micro-lenders’ 22 http://blogs.ft.com/beyond-brics/2014/01/10/polands-lessons-for-hungary-on-forex-loans/

44



limited access to capital have led them to prefer small, short-term cash loans, and hence focus on consumption rather than housing loans.

Although the nature and extent of mortgage market growth has been highly variable across the five case studies, the analysis in this section points to three generalizable aspects. First, mortgage markets have been driven by lending to better-off households. Second, the goal of extending mortgage lending to worse off households has been pervasive, but has tended to be best achieved in wealthier countries with more mature mortgage markets. Third, variegation in the nature and size of mortgage markets to a significant extent reflects the character of mortgage market exclusion and attempts to address it. As will be discussed in more detail below, however, the implications of these general patterns for housing supply have exhibited considerable divergence.

Development Finance

Development finance concerns how housebuilding is financed. The form and arrangements for development finance vary across each country in ways that are key in determining whether the abundance of finance capital described in the previous section has tended to feed through more into housebuilding or house prices.

The UK stands out at one extreme among the case studies in having almost no direct state intervention in development finance.23 The majority of housebuilding is instigated by private developers funded through a combination of bank borrowing and equity issue. The UK case study notes an asymmetry between house purchase finance and development finance, with far more being channelled to the former than the latter, leaving housebuilders facing capital constraints. In combination with a restrictive planning system and the speculative character of housebuilding – both of which are discussed in more detail below – this has meant credit has fed through into prices more than supply.

South Africa is at the other extreme in that development finance for low income housing has been provided almost entirely by the state in the form of capital subsidies. Under this model, the state provided a one-off subsidy to developers overseeing the construction of housing for low income households. The subsidy was intended to cover the costs of a foundational housing unit, comprised of land and basic top structure, for each eligible household. This model facilitated a mass building programme, with one million units constructed in the first six years of the programme from 1994, but subsequently ran into capital constraints of its own and saw demand outstripping supply. The initial need for the subsidy to cover the cost of land confined subsidised building to urban peripheries where land is cheap, which has meant that inhabitants have difficulty accessing employment. More recently, land costs were excluded from the capital subsidy, but the state’s ongoing refusal to intervene in land markets has perpetuated the pattern of peripheral development. Resource constraints have also meant that the top structures are of questionable quality. Furthermore, the inability of most households eligible for the subsidy to access mortgages has meant that few have been able to upgrade and improve the top structure provided under the subsidy.

23 The exceptions are a very small amount of council house building – currently limited to a few thousand a year across the country – and the capital subsidies given to housing associations, though these have diminished from approximately 30% to 5% of housing association finance since 2010.

45



Portugal and Turkey fall somewhere between the UK and South Africa. Portugal is similar to the UK in that, with the exception of a small state-funded social sector, development is financed privately. However, it differs from the UK in that the surplus of finance capital following financial liberalisation and European integration has been channelled into construction as well as mortgage lending. The case study argues that Portugal’s entry into the single currency was a major reason for this. The claim that a strong euro penalised Portugal’s tradeable goods sector, and thus encouraged domestic capital to move into construction and real estate, needs to be considered in light of the pre-existing weakness of Portugal’s tradeable goods sector. However, it is plausible that the removal of the possibility of currency devaluation exacerbated this weakness. In any case, the portion of all business loans granted to the construction and real estate sectors increased from 10% in 1992 to 40% in 2008. The state also supported developers by making land available for development and building supporting infrastructure. All this meant that Portugal experienced both a construction boom and a house price boom, at least until 2002 when state support for mortgages was curtailed. The construction boom ended after this point, but house prices have proved resilient because the number of housing transactions was consistently higher than the number of completed dwellings over the period 2003-8. This reflected low interest rates, which maintained a speculative demand for housing. Although lending to new mortgagees tailed off, banks continued to expand mortgage facilities to existing borrowers by increasing the value of loans and offering loans for the purchase of second homes.

Turkey has also experienced both a house price and a construction boom, which again reflected capital being channelled into the construction sector as well as into housing demand in the form of mortgage lending. However, the state has played a more proactive role in financing development in Turkey than in Portugal, owing to the role played by TOKI. TOKI lacks direct state-funding but has used its extensive powers over land acquisition and use to provide land to large housebuilders catering to middle- and upper-income groups. With central urban land increasingly rare and expensive in Istanbul, particularly in the form of large sites, this represented a substantial investment contribution. Under TOKI’s revenue-sharing model, TOKI received a share of the profits of such developments in exchange for its land contribution, which it then deployed for building of lower income housing. In this way, TOKI has helped to finance housebuilding for both middle/upper- and lower-income groups.

In Poland housing demand has tended to outstrip supply, leading to rising house prices. Excepting the years 2001-2, when property prices fell, house price increases averaged 10-20% per year after 2002 and even 55-75% in 2006 (Kierzenkowski, 2008). This in part reflects fundamentals, and therefore reasons extraneous to finance. For example, over the period considered, demand was buoyed by rising wages and GDP per capita, while housing construction has been constrained by the emerging character of development firms and the planning system, which favours small-scale development (see below). However, financial arrangements have also played a role.

The two most important types of housing development in Poland are self-build and private development firms. State-subsidised loans for development are available to Associations of Social Housing, which build socially-rented housing, though this constitutes just 0.5% of the total housing stock (Kierzenkowski, 2008). For self-build, development and consumption finance dovetail to some degree, as self-build housing tends to be financed through mortgages, which are usually secured on the value of the land already owned by the borrower. Between 2007-12, a subsidy covering up to 50% of the interest paid on mortgage could be obtained by borrowers who did not own any other housing. However, the subsidy

46



was subject to price and size limits, which meant that, in the context of excess demand and rising house prices, these limits were rapidly exceeded by market prices, making the subsidy relatively ineffective. The subsidy scheme may even have contributed to inflating prices, as it also applied to those buying houses from private developers. Reflecting their relative infancy, Poland’s private developers have been characterised by a relatively large share of owner’s equity (40%). This was compounded by conservative bank lending policies to construction firms following a series of losses in the late 1990s. Development firms tend to be reliant on foreign capital and therefore vulnerable to changes in its costs or availability. The under-capitalisation of the private development sector, compounded by barriers to large-scale development created by the planning system have fed house price growth.

There is no unifying trend in housing development finance to correspond to the rise of mortgage lending in housing consumption finance. Nonetheless, and despite the tendency for literature on housing finance to focus on mortgage markets, it is clear from the case studies that the character and availability of development finance play a decisive role in shaping the housing system.

Land

Land features in all housing sops as the site on which housing is built; its role as such is defined by the way in which property rights in land can serve as a means to capture ground rent. Indeed, the processes of creating and appropriating ground rents are crucial factors governing the operation of housing sops and the behaviour of agents within them.

While some ground rent on residential land is appropriated from producers out of the value created through productive investment on that land, a substantial portion tends to be appropriated from the purchasers or consumers of that land, out of the circulation of revenues. The level of this portion of ground rent on residential land will reflect the relative balance between supply and demand for housing. An excess of demand over supply will, other things being equal, inflate ground rents, while an excess of supply over demand will reduce them. On the supply side, rent is determined by land availability and house building rates. On the demand side, in addition to wage levels, credit plays a key role in the determination of ground rent by inflating effective demand for land and driving up rents. Ground rents on particular pieces of land arise from their differential qualities, reflecting, for example, proximity to economic activity and desirable amenities, which are themselves subject to change in the course of urban development. Land use planning also plays a central role in determining and distributing ground rent. This may variously involve: inflating ground rent through restrictive planning laws; allocating and distributing rent by determining what can be built where; or lowering ground rent by making land available for lower value activities.

In all of the case study countries, the extent to which land use is governed by the pursuit of ground rents is high. This applies to land owned by the state and to land owned by private agents, and to how planning authorities make decisions about how land is used. In both cases, best value has become an increasingly important criterion governing land use. While the precise nature of land value determination, capture and distribution differ in each country, in all cases the growing commitment to the valorisation of land has had profound consequences for housing outcomes and wealth distribution.

In the UK, ground rents have been exceptionally high and rising over the period considered, owing to a combination of a restrictive planning system, low housing supply and abundant

47



mortgage credit. While planning authorities in the UK have been unwilling to make land available for uses below its highest potential value, a combination of a discretionary planning system, concentrated and opaque land ownership, and widespread ‘NIMBYism’ have combined to restrict the amount of land available for development. Planning has thus become a site of struggle between landowners and housebuilders, with each seeking to influence planning decisions over the allocation of land for development in ways that increases the ground rents that they are able to capture. Planning authorities intervene in this struggle, though who they benefit in doing so depends on the circumstances of the particular case at hand. For example, a restrictive approach to planning will tend to drive up land values, though the benefit of this to landowners is limited unless development is permitted on their particular piece of land. Similarly, housebuilders want planning permission to be granted for their own developments to allow them to realise ground rents, but not on land owned by others lest this drive down prices in the area. Existing homeowners will benefit from the inflation of house prices arising from a restrictive planning policy, but a more permissive policy will, other things equal, enable households to get on the housing ladder. The beneficiaries of particular planning decisions in the UK, therefore, vary case by case, but the recent pattern of high mortgage lending and low house building has tended to be to the advantage of landowners and existing homeowners, to whom large capital gains most commonly accrue.

A similar unwillingness to assign land to uses below its market value is evident in South Africa, where the state has refused to use powers available to it to expropriate land from private owners for the development of low cost housing, particularly in inner city areas. Even where prime land has remained vacant, it has been viewed as inappropriate for low-income housing.24 As discussed elsewhere, this has meant that state-subsidised, low-income housing has been confined to urban peripheries, with negative consequences for the viability of such developments. It has also meant that land-use policy in South Africa has reproduced the racial geographies of apartheid, with the regeneration of certain areas biased towards a (predominantly, though not exclusively) middle- and upper-classes, while the majority of new homes for poor blacks are built on urban peripheries. Of course, some inner-city areas continue to be “no go” areas for the wealthy because of their association with crime and poverty. But where inner-city regeneration has occurred, it has focused on business development and attracting better-off households from the suburbs, displacing the existing poor residents. Following Smith (2002), this process of gentrification should be viewed as a conscious strategy to develop globally competitive cities and attract international investment by mobilising inner-city real estate as a vehicle of capital accumulation: ‘the construction of new gentrification complexes in central cities across the world has become an increasingly unassailable capital accumulation strategy for competing urban economies’ (Smith, 2002 p. 443).

In Turkey, and, to a lesser extent, Portugal, the state has been proactive in making land available for higher value uses. In Turkey, this has meant utilising extensive powers of land acquisition granted to TOKI, most dramatically through the displacement of the gecekondu and historical area housing units, to free up large parcels of land for development by private developers. The central role played by TOKI in this process was a product of the shortage of large tracts of urban land suitable for mass housing development in the 1990s and 2000s, itself a product of rapid but small-scale urbanisation in earlier decades. It is noted in the case study that, in part by virtue of the imperatives created by the revenue-sharing model, TOKI’s 24 It is noted that this has allowed the owners of such vacant land to reap large speculative profits as investments in the surrounding area raises its value.

48



exercise of its land use powers is governed by considerations of profitability and rent extraction. This is why gecekondu residents have regularly been relocated to less valuable peripheral land. As in South Africa, urban regeneration is driven by gentrification, itself a strategy to attract investment in an accumulation process dominated by real estate. In addition to the removal of lower income communities from prime land, this strategy has involved shifting manufacturing to peripheries in order to make room for more high value service sector activities, and an increased emphasis on brand and place-making geared towards attracting international capital. More than in South Africa, however, in Turkey such an approach has coincided with a large expansion of housing supply, though increasingly in ways that have reinforced inequalities in land access.

The role of the state in Portugal has been more permissive, with lax land use controls allowing rapid urban development. But reassignment of land use has nonetheless proved an important rent redistribution mechanism, with planning authorities regularly capitulating to pressure from developers to fast-track planning applications. Local authorities have done this because they see such development as aiding local economic growth and employment, and improving their tax intake. In Portugal, as elsewhere, state-funded, low-income housing has been confined to peripheries, illustrating the state’s reluctance to use its own powers over land use to violate market-like allocation in which land is assigned to its highest value use. In Portugal, this reluctance was enshrined in a 1991 law that increased the amount of compensation required for public appropriation by shifting the basis on which land was valued from present (mostly agricultural) use to potential future use.

In Poland, the demolition of the strict hierarchy of planning controls in place under communism was followed by the devolution of planning powers to local government units. This devolution was shaped by the firm commitments to private property rights and the putative free market that accompanied transition. However, limited resources relative to an arduous procedure for adopting local plans, along with fear of conflict with existing real estate owners and the absence of penalties for failing to produce a plan, have meant that the majority of local authorities do not have plans in place. Indeed, the case study finds that less than a third of land in Poland is covered by a spatial plan. Planning permissions for land not governed by a plan are granted through piecemeal administrative decisions. This has favoured small-scale, and, especially, self-build, developments regarding which planning administrations tend to be permissive, while obstructing larger developments requiring more complex planning decisions. The result has been low density urban sprawl, particularly onto former farmland. The minimal rules governing the development of greenfield land lying outside of a spatial plan – which require that a neighbouring plot is developed with housing and that the site has access to a public road – has given this sprawl a ribbon-like character (Halleux et al 2012).

Across the case studies, then, land use is increasingly governed by the imperative of realising higher land values, leading to the growing subordination of use values to exchange values in land use decisions. It is notable that this finding holds true even where planning systems are restrictive. Such planning systems intervene extensively in determining land use, but they do so in a way that seeks to fit into the grooves of the market rather than subverting it.

Production

The question of production is concerned with who delivers new housing and how, and with what implications for housing outcomes. It is helpful to divide the case study countries into those that have experienced mass building over the period being considered and those that

49



have not. The former group comprises Turkey, South Africa, and Portugal, while the latter group consists of the UK and Poland. In Turkey and South Africa the state has – as discussed in the section on development finance - played a central role in facilitating production, though the construction itself has been carried out by the private sector. Portugal is somewhat distinctive in having experienced a building boom despite provision being largely private-sector led. This notwithstanding, the state played an important role in supporting Portugal’s housing boom, as was revealed by its abrupt end following the withdrawal of state support. In the latter group, consisting of the UK and Poland, production has been more comprehensively left to private agents. Even in these cases, however, the state has historically been a major contributor to the total housing stock, having previously provided large quantities of housing that were privatised over the period considered.

As discussed above, Turkey experienced a construction boom in the 2000s, mostly in the form of large-scale redevelopment projects. This state of affairs arose out of the breakdown in models of provision that prevailed prior to the 1980s. In the 1960s and 70s, housing provision in Istanbul was dominated, on the one hand, by gecekondus and, on the other, by small contractors (yapsatçı) who produced housing using the ‘build and sell’ model. Both models were characterised by small-site development with a strong speculative element. Gecekondus arose from urban migrants building informal settlements on squatted land. The lack of alternative forms of housing meant that these settlements were tolerated and, over time, amnesties were granted, legalising their status. As a result, gecekondus ceased to be precarious and marginal. Instead, they underwent physical upgrading and acquired commercial potential, becoming sites for speculation and accumulation. Squatter lords began to buy up peripheral land and sell tiny plots to migrants, meaning that gecekondus were no longer technically squatted.

Meanwhile, middle-income housing production was dominated by yapsatçı, who were responsible for half of all new housing units over this period. Being heavily capital constrained, yapsatçı acquired small plots of land by making profit-sharing deals with landowners, and then sold basement and first floor flats immediately, using the proceeds to fund the production of higher floors. The profitability of the model depended on the firm delaying the sale of higher level flats while their value rose as the result of rapid urban development, high levels of migration, and Turkey’s then under-developed financial sector which, in the inflationary environment of the 1970s, encouraged purchase of real estate as an ‘inflation-proof’ form of investment.

The gecekondu and yapsatçı models of development were undermined in the late 1970s by the depletion of city centre land and consequent rising land prices. These squeezed the profit margins of yapsatçı, while increasing their capital requirements. Urban land shortages also paved the way for the stigmatisation of gecekondu, which came to be viewed more as a means of enrichment of squatter lords than a way of housing the needy. Coincidentally, with the tailing off of construction business in the Middle East, larger construction firms began to turn their attention to domestic markets. However, the entry of big firms into the housing sector was contingent on the spread of larger-scale developments in which big firms were more competitive, in part because they made industrialised construction techniques feasible.

With small-scale production models increasingly dysfunctional and the state under pressure from big construction capital, in the 1980s policy shifted to the promotion of mass housing cooperatives. At the same time, higher-income groups began moving to gated communities. These both required larger and better-arranged urban environments as a precondition to emerge. The extensive land use powers granted to TOKI in 2003 should be viewed against

50



this backdrop, as a means of making available larger plots of land suitable to mass building, in the context of a squeezed land market. As described above, TOKI’s deployment of these powers in turn created a state-led boom through two channels. First, TOKI provided land to private developers constructing middle-income housing for sale on the market, in exchange for a share of revenues. Second, TOKI contracted private builders to develop housing for sale to lower income households using state-issued mortgages. As a result, TOKI built 419,000 units between 2003-10, and the average rate of growth of the construction sector was nearly twice that of GDP over the period 2001-7.

In South Africa, the government claims to have provided 3.7 million housing units through the capital subsidy model, a scale of building that represents a significant accomplishment. However, supply has continued to lag demand and the housing backlog of 1.5 million households that existed in 1994 has grown to 2.1 million households. Following an early production push, and in response to the limitations of a policy creating dormitory settlements, emphasis was shifted from quantity of housing units to quality. The more holistic goal of creating ‘sustainable human settlements’ equipped with infrastructure, amenities and employment opportunities was adopted and pursued concurrently through the in situ upgrading of informal settlements and the creation of new flagship communities. However, despite increases in the level of the subsidy, and the removal of land costs from its spending requirements, the unit-focused approach continued to dominate, while unit production fell. Policy has paid lip service to alternative tenures but they have remained marginal in practice.

Throughout the lifetime of the capital subsidy model, the actual building has been done by private construction firms. In the early stage of the programme – coinciding with large-scale delivery - construction was dominated by large firms, who took responsibility for instigating and delivering production, with private conveyancers responsible for dispersing subsidies. However, policy-driven increases in minimum standards and rising building costs – reflecting increases in the costs of wages and raw materials – eroded profit margins of big construction firms. This, combined with allegations of profiteering and shoddy production, and the availability of alternative avenues of business in the form of mega-projects, caused large construction firms to withdraw from low-income housebuilding. This coincided with a shift in the orientation of housing policy implementation towards local government and led, from 2002, to the responsibility for managing development being transferred from national to local government, though with only partial success (see section on the state, below). In response to the withdrawal of large construction firms and complaints from smaller construction firms that they had difficulty accessing work funded by the capital subsidy, local government delivery turned to smaller contractors. However, the limited capacity of such firms, and regulatory challenges facing local government delivery meant that production rates were lower in this period.

Portugal is the anomaly among the construction boom countries in that its building boom was private sector-led. In Portugal, the construction of housing also grew dramatically over the period considered, tripling from 40,000 to 120,000 per year between 1995 and 2002. Almost all of this housing was built by the private sector, whose share in total housing production averaged 89% between 1950 and 2002 and grew steadily throughout this period. The private construction sector incorporated private companies and direct development by households, with production relatively evenly divided between the two. However, individual promotion in practice encompasses a wide variety of production models, ranging from the informal to more market-like strategies such as subcontracting.

51



The Portuguese building boom was a product of the way in which finance was channelled to construction firms as well as mortgage lenders, as described above in the section on development finance. However, the state played an important facilitating role by making land available for development, investing in infrastructure, and subsidising mortgages. Following the ending of such subsidies in 2002, housing production fell from a peak of 122,000 in 2002 to 73,000 in 2004 and 28,000 in 2012.

The UK contrasts with these three countries in having persistently low housing construction rates. The absence of any significant supply-side intervention by the state since the 1980s is a factor in explaining this, but also crucial are the speculative structure of housebuilding firms in the UK and the UK’s restrictive planning system. UK housebuilders are speculative in the sense that they acquire land for development and build out on it without having a pre-arranged buyer. This means that any change in land values between its purchase and sale are a significant determinant of housebuilder profits. Housebuilder profits depend on buying land as cheaply as possible and selling it – with houses built out – for as much as possible, endowing housebuilder activities with a strong mercantilist element. The result is that housebuilders have a powerful incentive to stagger output. Flooding markets with new housing would lower prices and erode margins on all land owned by the housebuilder, so instead, where possible, housebuilders tend to build out on land that they own only gradually. Housebuilders’ ability to stagger output in this way is sustained by the planning system, which makes minimal land available for development and thus creates localised monopolies over planning permissioned land that housebuilders are able to exploit. Increased mortgage lending since the 1980s, and consequent upward pressure on land prices has raised the stakes in the mercantile aspect of UK housebuilders’ activities, and thus accentuated these restrictive tendencies relative to rising demand for housing.

Poland had a housing shortage of at least a million units over most of the period considered. The Polish case study finds that this deficit has been turned into a surplus of half a million units, but this view conflicts with findings reported elsewhere. For example, Kierzenkowski (2008) notes that in 2006 Poland had one of the lowest numbers of dwelling per 1000 inhabitants – at 337 compared to an average of 470 in the EU15 – and that Poland’s ratio of residential investment to GDP is also comparatively low (Kierzenkowski, 2008). The different time frame used by the two studies may account for some of the difference, as Kierenkowski was published in 2008 whereas the case study uses housing supply data that goes up to 2011. However, a later OECD study (OECD, 2011), which coincides with the data use in the case study, estimates a housing shortage of 1.5-1.8 million. Another possible reason for the disparity of estimates is household formation. The case study assumes a constant average household size of 2.8, whereas OECD (2011) hold that household size has been shrinking over the period considered. The latter view is supported by Happach whose competing estimates of household numbers in Poland are persistently larger than those in the case study.25 The case study’s inclusion of uninhabited dwellings in its calculation of the total housing stock also begs the question of why these dwellings are uninhabited. The poor quality of the housing stock in Poland – inherited from the socialist era, but exacerbated by the failure of privatisation to secure investments in improvements by new owners – is widely recognised to be a problem. If dwellings are uninhabited because poor quality has made them uninhabitable, then their inclusion in estimates of the total housing stock is questionable.

As a result of rapid privatisation during transition, the bulk of Polish housing is produced privately. Self-build accounted for almost half of all units produced between 1991 and 2013, 25 http://www.slideshare.net/dziarski/housing-policy-in-poland

52



reflecting the lack of formalised planning regulation. This prevalence of small-scale production may in part account for low supply-responsiveness, but the latter also reflects characteristics of private development of the industry. Capital constraints and the barriers that the planning system places on large-scale development have already been mentioned. These notwithstanding, a building boom emerged in the early 2000s, but ran into bottlenecks, with shortages raising prices in input and labour markets. Developers also cited lack of demand as a key factor holding back activity, though Kierzenkowski (2008) find that mark-ups and profit rates are high, and that developers sell units gradually, suggesting that firms are to some extent employing strategies similar to those in the UK.

Beyond the central question of housing supply, the case studies also point to interesting points of comparison in relation to the structure and organisation of the housebuilding industry. Both the UK and Turkey employ subcontracting extensively, primarily as a cost-cutting device. In the UK, subcontracting is also a response to firms’ speculative structure and capital constraints, which together create an imperative to free up capital to respond readily to land buying opportunities by minimising the amount of capital tied up in production.

The housebuilding industry is increasingly dominated by large firms in the UK, Portugal, and Turkey. In both the UK and Portugal, this is the result of mergers and acquisitions, which are encouraged by the cyclical character of the housebuilding industry. In Turkey, as explained above, it is due to the historical displacement of small by large construction firms, reflecting changing patterns of land availability and building models. The pattern is reversed in relation to low income housing in South Africa, where small firms have taken over from large ones. In response to low profit rates and high levels of bureaucracy associated with the capital subsidy model in low income housing construction, as well as the inherited biases of the sector, South Africa’s small number of very large – and white-controlled – construction firms have found mega-investment projects a more lucrative way of extracting value from the state.

Concerns about quality were raised in the UK, Poland, and South Africa, though its causes are different in each case. In the UK the prevalence of land speculation in determining firms’ competitiveness, and an accumulated housing shortage, have together allowed firms to neglect production quality. In Poland, low quality is more a reflection of the condition of the old, socialist-era housing stock and its quick-fire privatisation, which transferred responsibility for the stock to households who were by and large unable to finance improvements. In South Africa, the main sources of poor quality have been the limited value of the capital subsidy, corruption in the form of construction companies cutting costs to cream off value as profit, and, as in Poland, the inability of new homeowners to fund property upgrades.

Housing supply structures therefore vary significantly across the case studies, and exist in a relationship of mutual causation with other aspects of the housing sop. Even so, it is worth commenting on the key role played by the state in housing production. In all of the countries with large housing outputs – Turkey, South Africa, and Portugal – the state played a significant role, from leading development in the first of these countries through to supporting it in the last. More strikingly, the historical role of the state in providing housing that has subsequently been privatised means that, even in the UK and Poland, the state has been a decisive contributor to the overall housing stock.

53



Labour

The sop approach is unusual in treating labour as a central and integral element of provisioning. Its reasoning in doing so is not limited to labour’s obviously essential role in housebuilding. From this central role it follows, in addition, that the forms taken by labour processes and working conditions are shaped by and shape other components of sop. Most notably, in this context, labour conditions, in line with other components of the sop, have been transformed by processes associated with financialisation. These labour conditions, of course, are in turn heavily implicated in shaping consumption across innumerable other sops, through their effect on wages and lifestyle.

Despite all this, labour is rarely investigated in conjunction with consumption and other aspects of provisioning. Neoclassical economics’ preoccupation with methodological individualism and the realm of exchange has tended to preclude the kind of social-systemic analysis that this would require. Furthermore, the case studies struggled to find much detailed discussion of labour at a sectoral level. For the most part, labour is treated at an aggregate level, in the form of employment and wage levels, reflecting, perhaps, the confinement of labour processes to the production “black box” in microeconomics. Consequently, the discussion in this section is brief.

Across all the case study countries, labour in the housebuilding industry is casualised. In the UK and Turkey, this reflected the widespread use of subcontracting by housebuilding firms, which leads to short-term, casualised contracts and piecework. In Portugal, it reflected the cyclical character of the industry. There the boom and bust nature of the construction industry had a strong impact on the labour market, with construction’s weight in national employment falling from 12.2% in 2002 to 7.7% in 2012. The cyclicality of the industry was also a factor in Poland, where high rates of bankruptcy made construction employment relatively unstable. Casualisation has tended to diminish workers’ rights and employment conditions. Both Turkey and Poland reported high rates of delayed renumeration for construction workers, and Turkey reported, in addition, regular incidents of injury and death.

The State

The sop approach replaces the state-market dichotomy typical of orthodox economic approaches, with a recognition that the state is not only constitutive of markets but also intervenes in multiple and diverse ways along the chain of provision. It also views the state as encompassing a range of levels and bodies, which may sometimes operate in tension or conflict with each other. Such a view of the state is vindicated by the findings of the case studies.

In all five countries, state policy has exhibited a clear preference for private capital involvement in the building, financing and allocating of housing. However, this preference has not been pursued through a policy of laissez faire. Far from it, the preference has been attached to a particular form of market provision – owner-occupation – and in all cases the state has intervened heavily to realise and sustain this model. The most prevalent form of support has reflected the strong connection between owner-occupation and mortgage finance, with mortgage subsidies in place in UK, Portugal, Poland and Turkey, and a Record of Understanding, followed by the Financial Services Charter, between the state and banks in place in South Africa. In the UK and Portugal, mortgage subsidies were gradually withdrawn as markets became established, though the return of new forms of mortgage market support

54



post-crisis showed that the state continues to be willing to step in where necessary to sustain mortgage lending and borrowing.

The growth of owner-occupation has also been aided by other forms of support, most notably stock transfer, support for housebuilders, and land-use policies. In the UK and Poland, the dominance of the tenure was established through a rapid, once-off and heavily subsidised stock transfer of state-owned social housing to sitting tenants. In South Africa, the building of owner-occupied homes for the poor black population has been financed by the state through the capital subsidy. In Turkey the state has used authoritarian means to remove existing communities from valuable urban land and then used this land to collaborate with firms building mass developments for middle- and upper-income households. It has also employed contractors directly to build housing for sale to lower-income families. The Portuguese state also played a crucial role in supporting the construction of owner-occupied housing, through both land allocation and infrastructure investments. In Portugal, as in Poland and South Africa, support for private capital through land-use policies has also taken the form of non-intervention, with the state refusing to use powers available to it to expropriate land for the purposes of affordable housing. The most complex case with respect to land use policy is in the UK, where it is broadly governed by assigning land to its highest value uses, but where extensive regulatory powers mean planning authorities are subject to conflicting pressures from homeowners, housebuilders, and landowners, with varying beneficiaries.

Notwithstanding the pervasiveness of this commitment to opening housing to private capital, the promotion of owner-occupation and mortgage markets has been riddled with contradictions. The section on consumption, below, discusses the pressures that are, in some places, leading the private rented sector to expand, despite state efforts to the contrary. Here, I look at the different ways in which the state has adopted the responsibility for dealing with market fall-out, that is, for those whom the market does not adequately house. Again, this responsibility has been variously interpreted and realised, but it is present in some form in all case study countries. Residual forms of subsidised rental accommodation are a feature in Poland, Portugal and, to a lesser extent, South Africa and the UK, though independent bodies (although supported by the state) such as housing associations are more prevalent in the latter. In South Africa and Turkey state responsibility for housing marginalised populations has largely been incorporated into the pursuit of owner-occupation, with the state collaborating with private developers to build low-income owner-occupied housing and, in Turkey’s case, also providing mortgages for the purchase of that housing.

However, the state’s self-imposed responsibility to house the marginal has frequently come into conflict with – and been compromised by - the broader goal of establishing a market-based housing system. Tensions between different branches of government have been a recurrent facet of this conflict, as it is most often local authorities that are assigned the responsibility for housing the marginal, only to find their capacity to fulfil this responsibility undermined by other policy imperatives.

In the UK, the privatisation of the bulk of existing local authority housing stock was accompanied by regulatory changes that drastically curtailed local authorities’ ability to replace that stock. As the pressures of an under-supply of housing in the context of market-based distribution began to be felt, and the need for state action to house the market-excluded became more evident, these powers were not restored. In the late 1990s and early 2000s, support for housing associations was extended, but this support has since been continually cut. Local authorities’ responsibility to house the marginal has thus been undermined by constraints imposed on their capacity to act by central government. A further tension has

55



emerged between the policy goals of promoting owner-occupation and fiscal conservativism. Following the decimation of the social housing stock, those who are unable to purchase a house or afford market rents are reliant on rent subsidies in the form of housing benefit, while local authorities increasingly fulfil their obligation to house people that are vulnerable and homeless in temporary accommodation such as bed and breakfasts. The fall-out from owner-occupation is consequently leading to an escalating housing benefit bill and increased spending on short-life housing, which tends to be significantly more expensive than state-run dwellings.

A similar situation is evident in Poland, where the rapid sale of state-owned housing stock at heavy discounts to sitting tenants has left local authorities lacking the resources to meet its responsibilities regarding residual housing need. Similarly in Turkey, the shift to a focus on urban regeneration led to the concentration of land use control in the hands of TOKI after 2004, but the responsibilities and duties of the municipalities were not reduced to match.

South Africa also bears some resemblances to these cases. There, the right to decent housing was enshrined in the post-apartheid constitution. While the policy framework to enforce this right was designed at the national level, the responsibility for delivering policies was gradually devolved to the municipal level, where it ran into capacity constraints. While fiscal restrictions in the face of an ongoing de facto preoccupation with building housing units was a key limitation, another important factor was conflicts between local and other branches of government. On the one hand, provincial governments, who were reluctant to cede either political control over the housing programme or access to housing budgets, have blocked local authority accreditation – the process that would empower local authorities to take responsibility for the capital subsidy programme – and in many cases have retained the responsibilities for tendering and appointing contractors for themselves. On the other hand, the attempt to devolve housing responsibilities to local government coincided with a shift in emphasis in housing policy away from mass unit production and towards the creation of sustainable human settlements. In conjunction with the limited resources granted by central government – itself a reflection of the national government’s commitment to controlling fiscal spending – this has left local government under-resourced to meet their growing responsibilities. It is suggested that the design of the capital subsidy programme may have compounded the under-resourcing of municipalities, as the low density, peripheral developments typical of the programme are associated with higher infrastructure costs, inefficient use of serviced land, and lower rates of economic and employment growth.

To sum up this section, the sop approach rejects crude state-market oppositions, but the case studies show that state intervention is increasingly governed by a market-like logic with complex content and consequences. However, this logic has, to some degree, been self-undermining, as mounting pressures for new forms of state intervention have emanated from those unserved by market provision. Notwithstanding significant variety in state structure and behaviour across the case studies, these pressures have often been manifest through tensions between different branches of government.

Consumers

The dominant trend in housing consumption across the five countries has been the rise of owner-occupation and concomitant marginalisation of those excluded from the tenure. In all of the cases studies, owner-occupation has been heavily promoted by the state, often in association with the idea that homeownership should be a form of asset accumulation, which is politically pacifying and encourages consumerism. However, this promotion has been

56



contested and, in many cases, the marginalisation of those excluded from homeownership is leading to a resurgence of other tenures.

In the UK, owner-occupation increased from 50% of all housing in 1970 to 70% in 2006, though it has since shown a small decline. Among the case study countries, it is in the UK that the idea of homeownership serving as a form of asset accumulation has taken the firmest hold. Rising house prices in the context of abundant mortgage lending and limited supply have attached significant capital gains to homeownership. This has ensured that housing has proved a very lucrative investment for many people, and the goal of homeownership has become widespread. It should be noted, however, that this idea of housing as asset has had an uncomfortable coexistence with an association between homeownership and ontological security - an idea at tension with the investor-speculator mentality implied by treating one’s house as an asset. Furthermore, as mortgage lending became more restrictive in the aftermath of the financial crisis, the negative consequences of this model for those excluded from mortgage markets and therefore unable to get on the housing ladder have become more pressing. The 2011 census found that the private rented sector was expanding in Britain for the first time in almost a century, and a growing number of people face rising rents and deteriorating housing conditions, as private landlords seek to capture rising ground rents in a sellers’ market, in the context of accumulated shortages of social housing. A contradiction between housing’s dual role as asset and shelter has thus emerged, with the use of housing as an asset by some undermining the ability of others to access housing as a form of shelter.

The situation in Portugal bears some resemblance to that of the UK. Owner-occupation increased from 39% in 1960 to 73% in 2011. Furthermore, homeowners have, with the aid of mortgages, benefitted from both access to housing as shelter and wealth-enhancing capital gains on that housing. That house prices increases have been more contained in Portugal than they were in the UK has made the wealth gains associated with homeownership less extensive, but also meant that owner-occupied housing has been more affordable. Consequently, in 2012, only 7% of households with mortgages were spending over 40% of their income on housing, compared to 36% of renters. The higher portion of income spent on housing costs by private renters reflects a combination of the relative neglect of the private rented sector, which has led to rising rents, and the tendency for renters to have lower incomes. High rents in Portugal have been compounded by the relaxation of rent controls, which has led to the development of a dual rental market. In older, rent-controlled tenancies, rents are low but poor maintenance rife, whereas for newer tenancies the situation is reversed.

The negative distributional effects of homeownership are evident in the way in which homeowners in both Portugal and the UK have expanded their housing assets. In Portugal this has taken the form of a growth in second home-owning, which grew from 3% to 20% of dwellings between 1970 and 2011. In the UK, the accumulation of housing assets has predominantly occurred through the buy-to-let market, whereby homeowners use their own properties as collateral to purchase additional homes, which they then rent out. The rental incomes are used to pay off the mortgages, while the owner benefits from the capital gains that accrue to the property. The hoarding of housing in the context of rising house prices is placing homeownership out of reach of a growing number of people. Despite the clear advantages of homeownership in these countries, the rental sector has been expanding in both countries in recent years.

Poland experienced the most dramatic increase in owner-occupation – from 39% in 1988 to 82% in 2011 – as a result of the shock therapy privatisation that accompanied the transition

57



from communism. However, the distributional dynamics of this shift played out differently in Poland because of the heavy subsidies in place pre-transition. Under socialism, Poles paid between 2% to 12% of their income for housing expenses whereas, despite the discounts that accompanied privatisation, housing expenditures currently average 21.7% of household income for the population as a whole, rising to 36% for those below the poverty threshold. While owner-occupation is more arduous for the less well-off, and there remains a segment of the population reliant on the state for housing, the major source of growth for the rented sector was identified as the young having a preference for forms of housing that permit more mobility, rather than barriers to owner-occupation.

The growth of owner-occupation has taken a different form in South Africa and Turkey where owner-occupation has been the dominant form taken by social housing, rather than a replacement for it. Even so, this form of promotion has again had stratifying outcomes. In South Africa, attempts to make the poor black population homeowners have foundered in the face of high unemployment and poverty, which, in combination with the problems that beset the capital subsidy model, have meant that the supposed benefits of owner-occupation have not borne fruit for most recipients of the subsidy. The limitations of the capital subsidy approach were compounded by increased investment in the middle- and upper-class property markets, which, facilitated by increased flows of speculative capital, drove up prices in the formal housing market, putting housing in that sector out of reach of the poor and lower-middle class. As a result, most government-subsidised housing has failed to enter the secondary housing market and, when it has, it has often been at a lower value than the cost of production plus land value. Government-subsidised housing has therefore not served as a means of enabling poor households to move up the property ladder, and the goal of making housing a financial and economic asset for the poor black population has failed. Instead, housing has continued to be viewed by poor households as a physical shelter and an emotional asset, rather than a financial one. There has even been a significant level of deformalisation, whereby subsidy-recipients vacate their subsidised property and return to the informal rented sector.

In Turkey, the stratifying effects of the promotion of owner-occupation arose from the contingency of that promotion on the dispossession of residents in gecekondu. The displacement of squatters under the guise of regeneration was accompanied by a shift in public representations of squatter communities. Gecekondu had earlier been regarded as a necessary solution to housing poor migrants in the absence of any alternatives, and regarded with tolerance if not sympathy. However, the tightening of the urban land market, the formalisation of squats, and a policy-reorientation to urban regeneration, were accompanied by changes in attitudes towards squatters, who began to be portrayed by politicians and in the media as comfortable and middle class, and hence undeserving of either prime urban land or state support. Squatters themselves had little formal say in the process of regeneration, finding themselves either relocated to new TOKI estates on urban peripheries (if the gecekondu land was deemed valuable) or relocated in new developments on the same site. In both cases, relocated households were obliged to pay for their new accommodation.

The centrality of gentrification as accumulation strategy to Turkish state policy around housing has meant that Turkey provides the starkest manifestation of the:

‘fundamental geographical contradiction between the dramatically increased land values that accompany the centralisation of capital in the … metropolises, and the marginal, exurban locations where workers are forced to live due to pitiful wages on which that capital centralisation is built’ (Smith, 2002 p. 436).

58



As Smith predicts, this has led to heightened struggles around social reproduction and intensified forms of social control. Thus the displacement of gecekondu has met with heavy resistance, which has been at times violently repressed by the state. Interestingly, the case study finds that the character of resistance has evolved, with an earlier emphasis on community and collective consciousness being supplanted by discourses that emphasised notions of property ownership – a shift that mirrors a growing emphasis on property rights and ownership in state housing policy. Though Turkey provides the most extreme example of conflict over social reproduction and their violent repression, resistance – for example, to slum redevelopment – has also occurred in South Africa, particularly in response to slum clearance. Even the UK has seen a recent upsurge in housing struggles over the period since the case study was written (Robertson, 2015).

In sum, all of the case studies are characterised by a contradictory and uneven spread of owner-occupation, which is in some cases giving way to a resurgence in other tenures and, in others, is giving rise to struggles over social reproduction.

Conclusion – the financialisation of housing

While some common themes and issues have emerged in the above analysis, housing provision in the case study countries has taken very different forms in often wildly different contexts. What conclusions, then, if any, do the case studies allow as to draw about the nature and extent of the financialisation of housing? The answer to this question is complicated by the lack of a precise and agreed upon definition of what financialisation is. Opinions on this question range from those who advocate a narrow definition of financialisation as the proliferation and application of new financial engineering techniques such as securitisation, to those who argue that the term carries little analytical content, serving to obscure rather than illuminate, and should therefore be abandoned altogether. I here adopt an intermediary position, seeing financialisation as a somewhat chaotic concept encompassing a range of processes, including the growth of financial markets, the expansion of financial capital into ever more areas of social and economic reproduction, and the increasing subordination of policy-making to financial interests. With this somewhat permissive definition in mind, there are three areas in which housing provision across the case studies can be characterised in terms of financialisation.

First, the reorientation towards owner-occupation, which was seen in all countries, has served not only to commodify housing, but also to incorporate finance capital into its provision, by making most people’s access to housing contingent on taking out a mortgage. The international drive to increase owner-occupation has thus sought to open up new terrains for the expansion of financial capital. Of course, success on this front has been variable. At one extreme, in the UK, the expanding numbers incorporated into mortgage markets and increasing volumes of mortgage lending, have served as a basis for a growth of secondary mortgage markets. At the other, South Africa’s attempt to spread homeownership among the poor – which was explicitly associated with the goal of expanding the domain of financial capital into housing - has largely failed to facilitate the participation of the poor black population in financial markets.

The promotion of owner-occupation has also sought, conversely, to incorporate households into financial markets, by providing them with an economic and financial asset. However, the contradiction embodied in this goal has constrained the spread of owner-occupation. For the effect of opening up housing to the circulation of speculative financial capital has been to drive up house prices, putting homeownership out of reach of a section of the population. In

59



the UK and Portugal this contradiction is playing out through a recent resurgence in private renting following a long period of expansion of homeownership. In Turkey and South Africa it has manifested in a failure to incorporate poorer sections of the population into the formal and secondary housing markets.

Second, financial interests are served by conservative fiscal and monetary policies, geared towards, among other things, controlling general inflation and thus preventing the devaluation of debt. Such a macroeconomic policy environment is reflected in housing provision in all case study countries, for example, in the form of restrictions on state investment or of the use of public land being governed by best value. However, it should be noted that such a policy framework has been inconsistently enforced. For example, where it serves the interests of finance, states have proved willing to spend, as is evident in the presence of mortgage subsidies in most case study countries. Similarly, house price – as opposed to general – inflation has been tolerated and even encouraged, reflecting the asset role of housing, the importance of house prices in underpinning mortgage lending and, arguably, the role of house prices in feeding demand-led growth in the face of the neglect of real investment typical of financialised economies.

Finally, the subordination of housing needs to land valorisation, which was seen in all countries in different ways, should be viewed as both a manifestation of, and a response to, an increasingly financialised capitalism. Increased international capital flows and capital mobility have increased the size of rents up for grab, by allowing rents to be drawn from the global pool of surplus value rather than determined by local factors. This has strengthened pressure on states to compete to attract that capital through urban place-making and the development of ‘post-industrial’ central business districts, which segregate urban space to the exclusion of the poor. The growing integration of land and real estate markets, with land treated more and more like a financial asset, further feeds the extent to which land use is governed by the pursuit of ground rent. Rent appropriation and the imperative to assign land to its most valuable uses should, therefore, be viewed as a response to the globalisation of capital and the rebalancing of the economy towards the financial sector.

That common manifestations of financialisation are identifiable across the five highly diverse case studies is an indication of the veracity and usefulness of the concept. This notwithstanding, the content of the case studies testifies to its complex and variegated nature, and emphasises the multifaceted way in which financialisation shapes the different processes and agents involved in provision. It is helpful to conclude by echoing Haila whose discussion of ‘the different social agents involved in building provision (which, according to Ball are landowners, developers, building firms, building workers, building owners and final users)’ notes that

‘the increased influence of economic and monetary factors on land use, the integration of investment markets and the growing internationalization of the financial system, or what may be called the substitution of a logic of exchange value for a logic of use value in land-use decisions, constrain the power of all these agents.’ (Haila, 1988 p. 97).

On the basis of this comparative study, Haila’s statement must be modified by noting that the financialisation of housing has also increased the power of those agents best placed to access and control money and finance.

60



Bibliography

Çelik, Ö., A. Topal and G. Yalman (2015) “Fınance and Housing Provision in Istanbul” FESSUD Working Paper.

Haila, A. (1988) “Land as a financial asset: the theory of rent as a mirror of economic transformation” Antipode vol 20, pp. 79-101.

Halleux, J-M., Marcinczak, S. and van der Krabben, E. (2012) “The adaptive efficiency of land use planning measured by the control of urban sprawl. The cases of the Netherlands, Belgium and Poland Land Use Policy vol 29, no.4, pp. 887-898.

Isaacs, G. (2015) “Housing System of Provision – South African Case Study” FESSUD Working Paper.

Kierzenkowski, R. (2008) “Bridging the gap in Poland” OECD Economics Working Papers, No. 639, OECD Publishing.

Lis, P. (2015) “Financialisation of the System of Provision applied to housing in Poland” FESSUD Working Paper 100.

OECD (2011) “OECD Urban Policy Review, Poland 2011”, OECD Publishing

Robertson, M. (2014) “Case Study: housing and finance provision in Britain” FESSUD Working Paper Series no. 51.

Robertson, M. (2015) “Re-asking the housing question” Salvage, vol 1.

Santos, A. C., N. Serra and N. Teles (2015) “Finance and Housing Provision in Portugal” FESSUD Working Paper Series no 79.

Smith, N. (2002) “New globalism, new urbanism: gentrification as global urban strategy” Antipode vol. 34, no. 3, pp. 427-450.

The financialisation of housing across five case study countries - synthesis report

Documents