Debt Relief, Transfers and Economic Development

1
Debt Relief, Transfers and Economic Development

• Presentation Development Economics
• 02.01.2006
• Johannes Boecker, Niklas Hoyer, Sabine Schröder,
  Sebastian Seiffert, Verena Hinze

2
Structure of the presentation
Introduction Debt relief Aid,
Allocation Effiency Conclusion

• 1. Debt relief
• Krugman,  Financing vs. Forgiving a Debt
  overhang 
• Sachs,  Resolving the Debt Crisis of Low-Income
  Countries 
• 2. Development aid and efficient allocation
• Foreign Aid - Facts and Figures
• Aid efficiency and allocation
• Conclusion and policy implications

3
First part Debt relief
Introduction Debt relief Aid,
Allocation Effiency Conclusion
4
Introduction
Introduction Debt relief Aid,
Allocation Effiency Conclusion

•  ... the massive debt forgiveness of the G7 will
  produce the same circumstances that have followed
  smaller debt write-offs. Venal or incompetent
  governments have merely run up a pile of new
  debt, squandering the money on unrealistic
  projects while creaming off spare cash into Swiss
  bank accounts.  (Arab News, Saudi Arabia
  12/06/05)
•  Doubling aid and cancelling Africa's debt are
  theoretically very attractive. They fail because
  they are based on a misguided faith that you can
  rely on human altruism to end human misery. 
  (Sunday Monitor, Uganda 12/06/05 )

5
Financing vs. Forgiving a debt overhang
Introduction Debt relief Aid,
Allocation Effiency Conclusion

• Paul Krugman, 1988

6
Illustrated example
Introduction Debt relief Aid,
Allocation Effiency Conclusion
Richie, a rich man.
Mike, a farmer.
A banker.
7
General structure
Introduction Debt relief Aid,
Allocation Effiency Conclusion

• Period 1 Mike starts with an inherited debt
  (given by Richie).
• He has to repay the debt to Richie.
• Hes not able to repay the debt, therefore Mike
  has to borrow new money.
• Period 2 Mike repays his borrowing with the
  income of the harvest of this period. If he can
  not, he pays all he can.

8
Easy example
Introduction Debt relief Aid,
Allocation Effiency Conclusion

• Harvests of the 2 periods are known.
• x1 and x2
• i opportunity cost of funds, D required debt
  repayment in period 1
• New borrowing if Dgt x1 , amount LD- x1
• Richies interest rater.
• The banker will supply the new money if

9
Easy example
Introduction Debt relief Aid,
Allocation Effiency Conclusion

• If
• Mike will not be able to meet his debt service.
• Richie should reduce Mikes obligations in order
  to avoid Mikes bankruptcy and to get the maximum
  of repayment.
• Without uncertainty the problem is
  straightforward.

10
Formal model
Introduction Debt relief Aid,
Allocation Effiency Conclusion

• Mike faces an uncertain future and is able to
  influence his harvest.
• Period 1 harvest x1
• Period 2 harvest x2 rdn effort

rdn rdnb or rdng
effort effl (0) or effh (0.5)
11
Formal Model
Introduction Debt relief Aid,
Allocation Effiency Conclusion

• Mikes profit
• P (actual payment in period 2) to Richie
• Profit function

12
Formal model
Introduction Debt relief Aid,
Allocation Effiency Conclusion

• His utility
• In the 2 situations
• A high r makes situation 1 more probable (Mike
  will choose low effort, low repayment).
• A low r will not assure a high repayment to
  Richie in favorable situations.

13
Debt overhang
Introduction Debt relief Aid,
Allocation Effiency Conclusion

• Mike ? indebted poor country
• Richie ? the creditor country
• Banker ? financial markets
• Creditors would like to fix a high r in order to
  assure a maximum repayment in all states, but by
  choosing this rate, the country has no incentive
  for high effort.

14
Changing the nature of claims
Introduction Debt relief Aid,
Allocation Effiency Conclusion

• 2 possibilities. Creditors can
• link repayment to some general measure of ability
  to repay (e.g. export revenues)
• link repayment to some measures of the shocks
  experienced by a country (state contingent
  claims)
• A state contingent claim is the best solution
  regarding incentives! Best for creditors and
  debtors.

15
Resolving the debt crisis of Low-Income
Countries by J. Sachs
Introduction Debt relief Aid,
Allocation Effiency Conclusion

• Bankruptcy laws
• overcome collective action problems
• Give fresh start
• Negotiations characterized by ad hockery and low
  degree of systematisation of international rules
• Lock in for decades in instability and
  impoverishment
• Possibly caused by excessive debt burden
• Debt relief guided by doing minimum to prevent
  outright desaster, but never enough to solve the
  debt crisis

16
(No Transcript)
17
(No Transcript)
18
(No Transcript)
19
(No Transcript)
20
History of debt reliefs from Paris Club
Introduction Debt relief Aid,
Allocation Effiency Conclusion

• 1975-88 rescheduling of debts
• 1988 (91) debt relief of 33,3 ( 50 ),
  multilateral debts can not be reduced, 20 (23)
  countries benefit
• 1994 debt relief of 67 for low income
  countries, min 50 debt relief for poorest,
  multilateral debts are not reduced, 30 countries
  benefit
• 1996 (99), HIPC initiative 70 (90) debt
  relief, reduction of multilateral debts allowed,
  debt export ratio 250 (150), 5 countries benefit

21
Reformation of system
Introduction Debt relief Aid,
Allocation Effiency Conclusion

• If MDGs taken as baseline, it is possible to
  calculate the level of net transfers needed to
  achieve a given level of output per capita in
  2015
• Medium-term plans by developing countries to
  scale up investments in health, education etc.
• Assessment of financial gaps that need to be
  bridged by ODA and debt cancellation

22
Reformation of system
Introduction Debt relief Aid,
Allocation Effiency Conclusion

• maximisation of debt repayment should not be
  major goal
• Fresh start, preservation of public functions
  achievement of broad development objectives
• Independent review panel objective evidence for
  most HIPC will be
• Complete cancellation of debts PLUS increased
  foreign assistance on a conditional basis to
  ensure that net resource flows in fact support
  the desired development objectives
• Monitoring of creditors and debtors by UN and
  Bretton Woods institutions

23
Conclusion - Debt Relief
Introduction Debt relief Aid,
Allocation Effiency Conclusion

• Krugman its rational to finance developing
  countries or to forgive debt, but incentives are
  important ? state contingent claims better
• Sachs Systematisation of debt relief necessary
  to achieve MDG
• Critique Budget constraint of donors implies
  that as debt relief increases, foreign aid must
  go down. (1994 high debt relief ?low aid)

24
Foreign Aid Data and Definitions
Introduction Debt relief Aid,
Allocation Effiency Conclusion
25
Foreign Aid Worldwide
Introduction Debt relief Aid,
Allocation Effiency Conclusion
DAC Members' net ODA 19902004 in Relation to GNI
and simulations of net ODA until 2010
26
Definition of Aid
Introduction Debt relief Aid,
Allocation Effiency Conclusion
Net Official Development Assistance (ODA)
Grants and concessional loans net of repayments
of previous loans to developing countries with
the main goal to foster economic and social
development
27
ODA of all DAC Countries in 2003, Net Distribution
28
Introduction Debt relief Aid,
Allocation Effiency Conclusion
German Foreign Aid
29
Foreign Aid Frameworks
Introduction Debt relief Aid,
Allocation Effiency Conclusion

• the model of poverty traps

• the two-gap-model (Chenery and Strout 1966)

30
A Theoretical Dinosaur - The Two-Gap Model
Introduction Debt relief Aid,
Allocation Effiency Conclusion

• Savings Gap
• Investment necessary
• to achieve desired
• growth

Available domestic savings
Foreign Exchange Gap Import requirements for a
given level of production
Foreign exchange earnings
31
A Theoretical Dinosaur The Two-Gap Model
Introduction Debt relief Aid,
Allocation Effiency Conclusion

• The model is based on 2 key assumptions
• 1. stable linear relationship between investment
  and growth
• 2. foreign aid will primarily foster investment
  and only to a lesser degree increase consumption

AID INVESTMENT GROWTH
32
Empirical Assessment of the Models Implications
Introduction Debt relief Aid,
Allocation Effiency Conclusion

• The effect of aid on investment
• No clear positive correlation between aid and
  investment for all case studies
• But Harms and Lutz conclude that at the
  aggregate level a quarter of aid is translated
  into investments

33
Empirical Assessment of the Models Implications
Introduction Debt relief Aid,
Allocation Effiency Conclusion

• The effect of aid on savings
• Part of aid is used for consumption
• Aid may therefore lead to a reduction in domestic
  savings in relation to income
• Most studies conclude that aid has a negative
  effect on domestic savings
• The majority of studies finds a positive effect
  of aid on investment and growth

34
Introduction Debt relief Aid,
Allocation Effiency Conclusion
Zambia GDP per capita (PPP-adjusted) What might
have been and what actually happened
GDP/capita in US
35
Introduction Debt relief Aid,
Allocation Effiency Conclusion
Foreign Aid Efficiency
36
Introduction Debt relief Aid,
Allocation Effiency Conclusion
Views on Foreign Aid Efficiency
Because aid accrues to the government it
increases its resources, patronage, and power in
relation to the rest of society. This enhances
the hold of governments over their subjects and
diverts resources from productive economic
activities . (Peter Bauer, 1991)
Aid does not promote economic development for
two reasons poverty is not caused by capital
shortage, and it is not optimal for politicians
to adjust distortionary policies when they
receive aid. (Boone, 1996)
37
Introduction Debt relief Aid,
Allocation Effiency Conclusion
Views on Foreign Aid Efficiency

• Is aid per se ineffective?

• Or can we identify some fundamental forces that
  are responsible for that failure of aid in some
  countries and its success in others?

Money matters in a good policy environment
(World Bank, 2002)
38
Aid Efficiency and Allocation
Introduction Debt relief Aid,
Allocation Effiency Conclusion

• We examine 2 papers
• Aid, Policies and Growth (BurnsideDollar,
  2000) at a quick glance
• Aid allocation and poverty reduction (Collier,
  Dollar, 2002) a little more thoroughly

39
Aid, Policies, and GrowthBurnsideDollar, 2000
Introduction Debt relief Aid,
Allocation Effiency Conclusion

• Main questions
• Does aid have a positive effect on growth in the
  presence of good economic policies?
• Have donors systematically allocated assistance
  in favor of good policies?
• Has aid affected policies?

40
Aid, Policies, and GrowthBurnsideDollar, 2000
Introduction Debt relief Aid,
Allocation Effiency Conclusion

• Results
• Aid has little impact on growth
• Aid has a positive impact on growth in good
  policy environments
• Donors do not tend to favour good policy for
  bilateral aid
• Multilateral aid is allocated in favour of good
  policies

41
Aid Allocation and Poverty ReductionPaul
Collier, David Dollar, 2002
Introduction Debt relief Aid,
Allocation Effiency Conclusion

• Collier and Dollar, 2002 use a similar model as
  before to compare the derived poverty efficient
  allocation with the actual allocation of aid.
• Aid allocation with maximum effect on poverty
  depends on level of poverty and quality of
  policies
• With poverty-efficient allocation, productivity
  of aid would nearly double

42
Assumptions
Introduction Debt relief Aid,
Allocation Effiency Conclusion

• Aid is mainly used to reduce poverty.
• Donors have to take into account that
• Impact on growth depends on quality of economic
  policies and is subject to diminishing returns.
• Quantity of aid does not systematically affect
  the quality of policies.
• Aid is fungible (difficult for donors to direct
  aid).

43
The mapping from aid to growth
Introduction Debt relief Aid,
Allocation Effiency Conclusion

• A simple linear regression model to explain
  growth
• Thus Growth(G) is a function of exogenous
  conditions(X), level of policy(P), level of net
  receipts relative to GDP(A), level of aid squared
  and interaction of policy and aid

44
The mapping from aid to growth
Introduction Debt relief Aid,
Allocation Effiency Conclusion

• By OLS Regression of different variants of the
  model the coefficients are estimated and tested
  for significance
• Policy coefficent significantly positive
• Interaction of aid and policy the most
  significant and positive
• Aid by itself not significantly different from
  zero
• Aid squared enters with a negative coefficient

45
The mapping from aid to growth
Introduction Debt relief Aid,
Allocation Effiency Conclusion

• Marginal impact of aid on growth
• The policy environment significantly and
  substantially determines how rapidly diminishing
  returns eliminate the marginal contribution of
  aid to growth

46
Poverty efficient allocation of aid
Introduction Debt relief Aid,
Allocation Effiency Conclusion

• Consider a world in which aid is given to
  maximize poverty reduction.
• Aid affects growth in good policy environment
• Policy and distribution of income within
  recipient countries are exogenous

47
Poverty efficient allocation of aid
Introduction Debt relief Aid,
Allocation Effiency Conclusion

• That gives us following objective
• Max poverty reduction
• Subject to
• y per capita income
• total amount of aid
• h measure of poverty
• elasticity of poverty reduction w.r.t.
  income
• N population
• i indexed countries

48
Poverty efficient allocation of aid
Introduction Debt relief Aid,
Allocation Effiency Conclusion

• Considering only interior solutions, the first
  order conditions for a maximum are
• Plugging in the we can solve for each
  countrys aid receipts as a function of its
  policy, poverty level, per capita income and
  elasticity of poverty

49
Poverty efficient allocation of aid
Introduction Debt relief Aid,
Allocation Effiency Conclusion

• Using the estimated coefficients we get
• The basic properties of the equilibrium can be
  easily graphed

50
(No Transcript)
51
(No Transcript)
52
Results from the model
Introduction Debt relief Aid,
Allocation Effiency Conclusion

• Actual allocation is inefficient since we have
  large variations in marginal productivity (for
  some countries even negative)
• Require reallocation between countries to
  equalize marginal productivities

53
Reallocating aid for poverty reduction
Introduction Debt relief Aid,
Allocation Effiency Conclusion

• Example
• In 1996 (Average) 1 Million US lifted 235
  people out of poverty
• Ethopia the estimated marginal productivity was
  1655 people per million US
• 1 million US allocated to Ethopia would have
  had a 7 times larger effect on poverty reduction
  compared to existing volumes

54
Reallocating aid for poverty reduction
Introduction Debt relief Aid,
Allocation Effiency Conclusion

• Overall impact
• Actual average cost of poverty reduction is 2650
  per person
• Efficient allocation would cut average cost to
  1387
• Thus, the productivity of aid could be doubled.

55
Conclusion
Introduction Debt relief Aid,
Allocation Effiency Conclusion

• At present, aid is allocated partly as an
  inducement to policy reform and partly for a
  variety of strategic and historical reasons.
• Since there is no evidence that aid affects
  policies, the allocation is inefficient.
• Instead of lifting 10 million people out of
  poverty, with efficient aid it could be 19 million

56
Easterly (2003) Can foreign aid buy growth?
Introduction Debt relief Aid,
Allocation Effiency Conclusion

• Background debate raised by the BD 2000 paper
• BD findings
• significant positive interaction between foreign
  aid and good policies
• furthermore, only in environments with sound
  fiscal, monetary and trade policies aid resulted
  in GDP growth

57
Prominence of BD 2000
Introduction Debt relief Aid,
Allocation Effiency Conclusion

• this hypothesis gained great prominence in
  following years, e.g.
• at the 2002 UN conference Financing for
  development in Monterrey,
• within white papers of national development
  agencies and even
• in the speeches of top-policy makers as Bush, who
  announced a 50 increase in US foreign aid

58
questioning the BD hypothesis
Introduction Debt relief Aid,
Allocation Effiency Conclusion

• include more recent data (1970-97) no support
  for the conclusion that aid works in a good
  policy environment
• when aid is defined as net ODA, then
  aid/policy-interaction is no longer statistically
  significant
• for longer periods (8, 12 or 24 years)
  regressions lead to more ambiguous results

59
dynamics inside the aid agencies
Introduction Debt relief Aid,
Allocation Effiency Conclusion

• aid buys growth as a sort of founding myth of
  aid bureaucracies
• moving of money considered more important than
  effective outcomes of aid projects
• Donors reluctant to see failures. Quite typical
  in this respect are various reports by the World
  Bank on economic perspectives in Subsaharan Africa

60
World Bank Economic Outlook
Introduction Debt relief Aid,
Allocation Effiency Conclusion

• 1981 Real gains in development and income in the
  near future
• 1984 Optimism justified by policy and
  institutional reform in some Subsaharan African
  countries
• 1986 Progress is clearly under way
• 1989 Important changes in policies and economic
  performance
• 1994 Great strides in improving policies
• 2000 Better economic management starts to pay off

61
in the context of such perceptions in aid
agencies, the BD (2000) paper was highly welcome
62
Selectivity
Introduction Debt relief Aid,
Allocation Effiency Conclusion

• Aid should be directed to where it can do good
• Past experience
• Countrywide success stories, e.g. Uganda, South
  Corea in 1960s
• Sectoral successes, e.g. elimination of diseases
  or improved family planning
• Numerous examples of inefficient or even adverse
  aid projects
• 2 possible strategies
• Ex-ante-conditionality
• Ex-post-evaluation

63
Ex-ante conditionality
Introduction Debt relief Aid,
Allocation Effiency Conclusion

• Conditionality normally focusses on macroeconomic
  indicators such as
• non-interference with market pricing,
• privatisation of state-run enterprises and
• liberalisation towards international trade
• Hence conditionality is rarely pursued
  consequently, e.g. IMF loans to Kenya over 20
  years despite poor economic policies
• Special bad performer funds at World Bank

64
Ex-post evaluation
Introduction Debt relief Aid,
Allocation Effiency Conclusion

• Evaluation has been a scarcely applied tool in
  aid agencies up to now, particularly while the
  project is still underway
• If at all existing, evaluations of World Bank aid
  programmes have not been published
• Other donors (UNDP, NGOs) also fail to evaluate
  systematically

65
Conclusions Aid Allocation/Efficiency
Introduction Debt relief Aid,
Allocation Effiency Conclusion

• Foreign aid does not significantly increase
  investment, and therefore growth
• Efficiency of aid depends on economic and
  political environment
• BD 2000 Aid combined with good policy has an
  effect on growth
• Easterly Internal dynamics in donor
  organisations more important than aid-growth
  interaction

66
Policy recommendations
Introduction Debt relief Aid,
Allocation Effiency Conclusion

• Systematisation of debt relief and foreign aid
• More research on disaggregate level with respect
  to various aspects of policies/institutions
• More case-to-case rather than general decisions
• Less bilateral foreign aid
• Increase foreign aid to good-policy-countries,
  otherwise project-specific aid
• Besides debt relief and increase in foreign aid,
  important to give fair chance for development by
  cutting of subsidies in donor countries

67
Thanks for your attention!