Institutions and Development

The Ronald Coase Institute

Institutions and Development

Mary M. Shirley

President, The Ronald Coase Institute



Institutions

• Institutions are the “humanly devised constraints that structure human interaction” (North, 1990).

• Institutions include formal rules (laws, regulations) and informal norms (habits, unwritten codes of behavior).


• Institutional variables have large, statistically significant correlations with growth

• Trump geography or trade in horse races between variables (Rodrik, et al. 2002)


Questions I’ll Address

1. How do institutions determine development?

2. Can foreign aid improve institutions?

3. Can foreign aid avoid institutions?

4. What can be done about damaging institutions?


Why is this important?


Development is the exception

• Most people in the world live in underdeveloped countries


World’s population living in developed vs. underdeveloped countries, 2006

DC

LDC

World Bank, DC = >$11,115 GNP p.c. PPP in 2006

85%


Underdevelopment persists

• Most people in the world live in underdeveloped countries

• This has changed little over the past two decades


World’s population living in developed vs. underdeveloped countries

0

10

20

30

40

50

60

70

80

90

1979 2006

DC

LDC

DC

LDC


Underdevelopment persists

• Despite large amounts of aid

• And large aid community


Aid Community in 2005(gross aid* in billions of dollars)

0

20

40

60

80

100

120

Multilats. Bilaterals NGOs21 36 17,428

$51

$120

$15

*$188 gross vs. $121 net. OECD


Role of institutions in development


What Kinds of Institutions Are Needed for Development?

Encourage exchange by lowering transaction costs, providing framework for competition & transparency


Reduce transaction costs

• Contracts & contract enforcement mechanisms,

• Commercial rules & norms, information disclosure

• Beliefs and norms that favor trust, shared values, etc.


Institutions Needed for Development

Encourage exchange by lowering transaction costs, providing framework for competition & transparency

Direct the power of the state toward protecting property & individuals, not exploiting them


Control the state

• Constitutions, electoral rules, federalism

• Political checks & balances

• Norms and beliefs encouraging civic mindedness, probity in public service, rule of law, etc.


Institutional frameworks

1. Usually durable2. Endure because

• Powerful people benefit • Humans resist changes in habits &

beliefs

3. Changes are often idiosyncratic & experimental


E.G. idiosyncratic change: China’s TVEs

• State owned enterprises run by private investors

• Overseas Chinese ties to village substitute for secure property rights

• Regular payments of profit shares to government officials


“Capitalism with a red hat”


Can aid improve institutions?


No evidence aid can stimulate growth

Rajan & Subramanian, 2005 find no statistically robust association looking at:



Rajan et al., 2005 find no statistically robust association looking at:

• 10, 20, 30, 40 time periods




• 10, 20, 30, 40 time periods

• Only aid directed at growth




• 10, 20, 30, 40 time periods


• Only countries with good policies




• 10, 20, 30, 40 time periods


• Only countries with good policies

• Different regions, areas


No evidence aid can improve institutions

• No effect on institutions (see cites in Burnside & Dollar 2004)


Evidence aid can harm institutions

• Higher aid associated w/ lower bureaucratic quality, corruption, less rule of law (Knack 1997)




• Increased rent seeking in less open economies & those w/ large public sector (Economides 2004)




• Increased rent seeking in less open economies & those w/ large public sector (Economides 2004)

• Rent seeking, corruption, waste (Bauer 1991, Kanbur 2000, Easterly 2002)


Evidence aid harms institutions

• Higher aid associated with slower adoption of market oriented reforms (Heckelman & Knack 2005)


Evidence aid harms institutions

• Higher aid associated with slower adoption of market oriented reforms (Heckelman & Knack 2005)

• Negative association with democracy (Djankov et al, 2006) & transition to economic freedom (IMF 2005) – But, reverse causality?


Cross Country Regressions

• Obvious flaws



• Obvious flaws

• Problems in establishing causal relationships



• Obvious flaws

• Problems in establishing causal relationships

• But – can rule out hypotheses consistently rejected in most specifications


World Bank’s evaluation department judged its projects to improve public administration, rule of law, etc. to be “largely ineffective” (2005)


Why?

• Samaritan’s Dilemma


Why?

• Samaritan’s Dilemma (Buchanan 1977)– Payoff highest to Samaritan if Samaritan

provides aid and beneficiaries respond by exerting higher effort


Why?

• Samaritan’s Dilemma (Buchanan 1977)– Payoff highest to Samaritan if Samaritan

provides aid and beneficiaries respond by exerting higher effort

– Payoff highest to beneficiaries if they get aid without increasing effort


Why?

• Moral hazard:

“…aid helps to ensure incompetent governments from the results of their actions, thus weakening their incentive to find alternative revenue sources or better policies.” Ostrom, 2002


Why?

Mismatch between characteristics of aid agencies & characteristics of institutional change


Characteristics of Institutions vs. Aid

Institutional framework

• Deep rooted & usually durable



Institutional framework

• Deep rooted & usually durable

Aid

• Focus on policy, organization, sector rules

• Three year projects

• Staff rotates every 3 -5 years

• Rewards for approval, not sustainability

• Accountability?



Institutional framework• Supported by power

elites & beliefs



Institutional framework• Supported by power

elites & beliefs

Aid• Requires support or

permission incumbent government

• Staff incentives to cooperate to win more project approvals

• Revolutionaries would be asked to leave



Institutional framework• Changes are often

idiosyncratic & experimental



Institutional framework• Changes are often

idiosyncratic & experimental

Aid• Focus on Western best

practice – no support to TVEs

• Focus on changes that are:– rapid, – defensible, & – provide benchmarks


Without improvements in damaging institutions aid

projects result in: pro forma reforms


E.G. Performance contracts

Government

State owned enterpriseForeign aid


Manager promises

productivityprofitabilityinvestmentother targets


Government promises

autonomy to achieve targets

$ bonus or recognition for good results punishment for bad results


Foreign aid agency promises


Performance Changes After the Introduction of Contracts

Return on assets

No change

Deteriorated0

4

8

12

Number of state-owned enterprises



Labor productivity

Improved

No change0

4

8

12




Total factor productivity

0

4

8

12


Improved

Deteriorated

No change

Shirley & Xu, 2000


G r a p h 7 : E f f e c t s o f P e r f o r m a n c e C o n t r a c t s o n T o ta l F a c to r P r o d u c t iv i ty i n C h in e s e S ta t e O w n e d E n te r p r i s e s

- 2 5 0

- 2 0 0

- 1 5 0

- 1 0 0

- 5 0

0

5 0

1 0 0

A v e r a g e P o s i t i v e N e g a t i v e

S h i r l e y & X u , 2 0 0 1

No Significant Positive Correlation with TFP in over 500 Contracts


Why such poor results?

• Weak targets,

• Poor information,

• No bonuses for good performance,

• No punishments for bad performance, or

• No autonomy to lay off workers or close plants


Why performance contracts?

Aid staff liked:



Aid staff liked:• Tangible action

(signing)• Tangible “success”

(achieve targets)• Projects possible w/o

privatization






privatization

Government liked:






privatization

Government liked:• Easy action (force

manager to sign)• Easy “success” (set

targets low)• Aid w/o political

costly actions (layoffs, privatization)


Bottom line: Fundamental changes in deeply rooted institutions do not happen because of outsiders’ money or pressures


Can aid avoid damaging institutions?


Problems

1. Measuring reform in institutions

2. Combating incentive to “move the money”


E.G. US Millennium Challenge Account

Created 2002

Assists only poorest countries that rank:



Created 2002


• top 50% of eligible countries on controlling corruption



Created 2002


• top 50% of eligible countries on controlling corruption

• top 50% on half of other criteria in each of 3 performance dimension


Dimension Performance Criteria

Governing Justly

1. Civil liberties2. Political rights3. Voice and accountability4. Government effectiveness5. Rule of law6. Control of corruption

Investing in People

1. Immunization rates (diphtheria & measles) 2. Public expenditures on health3. Primary education completion rate4. Public expenditures on primary education

Promoting Economic Reform

1. Inflation rate2. Cost to start a business3. Days to start a business4. Trade policy5. Regulatory quality6. Fiscal policy

Source: MCA Data on the Web at www.MCA.gov.


1. Current measures do not permit targeting aid to countries with improving institutions


Problems with measurement

Measures are not “actionable”



Measures are not “actionable”, instead they are

• Abstract – “rule of law”





• Static over time – governance





• Static over time – governance

• Uncorrelated with growth, e.g. political rights


Not institutions -- outcomes

• Outcome = secure property rights

• Institutions = laws, commercial code, norms of business, electoral constraints on government power, etc.



• Large margins of error


Kaufmann & Kraay 2002

Impossible to say w/90% certainty where 51 of 61 countries rank


2. Incentives of aid agencies are to “move the money”


Selectivity conflicts with incentives to more the money

• Fewer countries – By June 2007: only 8

of 61

signedexcluded



Little disbursed:– By June 2007: $71

million out of $3 billion allocated ($6 billion total to MCA)

DisbursedAllocatedUnallocated



Incentives to disburse– 13 allocated $310

million as “threshold countries”

DisbursedAllocatedTotal alloc.Threshold


What can be done about damaging institutions?


What can be done?

Raise capacity of countries to reform their own institutions


Societies cling to blatantly damaging institutions even when faced with crisis, but…

Institutional frameworks do change


Ideas and learning are powerful forces in overcoming beliefs and norms that inhibit change in institutional frameworks (North)


Intellectual capital

• New ideas in receptive environments, – E.g. Enlightenment in Europe

• Learning – E.g. Economists in China, Chicago

boys in Chile


Critical mass of well trained scholars in receptive circumstances (Shirley and Soto, 2007)


Case Studies


Case studies1. Chile:1975 Chicago boys

2. China:1993 “Socialist Market Economy”

3. Korea:1980 EPB economists

4. Taiwan:1958-61/1986 S.C. Tsiang, T.C. Liu, others

5. Argentina:1991 Cavallo boys

6. Indonesia:1966/1983 Berkeley mafia


Serious threat, economic problems Failure of past reforms

Change seen as

necessary by elites

Coherent alternative different from past reforms

Role model

Consensus view of group

Seen as experts?

Seen as not self interested?

Incentives to be truthful?

Local knowledge? yes

yes

yes

yes

Alternative paradigm

Channels to elites

Alternative is

known to elites

Alternative is

feasible

SCHOLARS RULING ELITES

Policy, paradigm changeParadigm is disseminate

d

REST OF POPULATION

Beliefs change

Institutional change

Outcomes seen as positive


Serious threat, economic problems Failure of past reforms

Change seen as

necessary by elites


Change seen as necessary by elites

• Threats & problems– E.g. Taiwan: Acceptance of China in UN

(1980’s)– E.g. Hyperinflation in Argentina (1989-1991),

Chile (1973-75),

• Failure past policies– Partial reforms in Taiwan, Menem’s &

Pinochet’s first efforts to stabilize in Argentina & Chile


Coherent alternative different from past reforms

Consensus view of group

Scholarly group with alternative paradigm



• Coherent & different from past reforms– Liberalization in Taiwan– “Chicago Model” in Chile

• Consensus view of group– Five economists’ paper in Taiwan– Cavallo Boys and Chicago Boys


Seen as experts?

Seen as not self interested?

Incentives to be truthful?

Local knowledge?


Lupia & McCubbins


Alternative paradigm Channels to elites

Alternative is known to elites

Alternative is feasible


Alternative is known to elites

Channels to elites

• Think tanks

• Publications (Vover a Crecer by Cavallo)

• Conferences (Milton Friedman in Chile)

• Media coverage (El Mercurio in Chile)

• Personal ties (Cavallo & Menem)


Paradigm is disseminated

Policy, paradigm change

Outcomes seen as positive

Beliefs, norms

change, new stakeholders

Public Elites


Paradigm is disseminated

• Vigorous public debate (even in Chile)

• Early success (lowering inflation in Argentina, Chile, spurring faster growth in Taiwan)

• Treated as new ideology (e.g. China’s socialist market economy)


Scholars in receptive circumstances:

• Identified barriers to reform

• Persuaded policy makers to adopt new paradigms and policies

• Informed public debate, contributing to changes in beliefs & institutions


Conclusion

• To develop countries need institutions that lower TCs and protect property and individuals

• Aid cannot strengthen – and may undermine – such institutions

• Aid cannot avoid institutions


Build Intellectual Capital

Local scholars and research can provide ideas and analysis that raise the potential for institutional change.

Institutions and Development

Technology

gross aid

foreign aid

institutions institutions

canharminstitutions

large aid community

large amounts of aid

role of institutions

kinds of institutions