Chapter 16

download report

Transcript Chapter 16

Chapter 16
Adjustment and External Debt in
Low-Income Countries
© Pierre-Richard Agénor
The World Bank
1





Growth, External Debt, and Adjustment
The Debt Overhang and the Debt Laffer Curve
Measuring the Debt Burden
Debt Rescheduling
The HIPC Debt Initiative
2




Debt crisis of early 1980s had severe impact in
developing countries.
Closure of world capital markets led to inflationary
financing of fiscal deficits, a slowdown in
investment, and low rates of economic growth.
Servicing external debt continues to be a key
concern for policymakers in low-income countries.
Has led to a variety of proposals aimed at ensuring
that debt burden does not hamper capital formation
and growth.
3
Growth, External Debt,
and Adjustment
4
High foreign debt levels’ adverse impact on
savings, investment, and growth.
 Illustrating the dynamics: An extension of Villanueva
(1994).
 National income, Q, given by,
Q = Y - r *D *,
r*: cost of borrowing,
D*: stock of external debt.
5

National income per unit of effective labor, q, is then,
q = k - r*d*,

y = k: intensive form Cobb-Douglas production
function;
d* = D*/AL.
Foreign saving, S*, equals changes in net external
debt,
.
D* = S*
(4)
6

Domestic saving, S, equals Sp + Sg with,
Sp = s(1 - )Q
Sg = Y - G.

Changes in the capital stock,
.
K = I - K (7)
7

Domestic investment, financed by foreign and
domestic savings,
I = S = (Sp + Sg) + S* (8)

Using (4), (7) and (8),
.
.*
K + K - D = S (9)

Labor, grows exogenously,
.
L/L = 
(10)
8

Following Villanueva’s learning-by-doing, labor
efficiency,
.
A = (K/L) + A,  > 0.

(10) implies growth rate in efficiency units is given by,
.
.
A/A + L/L = k +  +  (11).
9

Change in foreign debt; dependent on difference
between, r (k-1 - ) and r*;
.
D*/D* = (k-1 -  - r*),  > 0
r* = rf* + (d*,), (13)
rf* : risk free rate
 : risk premium; function of a country’s stock of
debt in effective units of labor, d* and , exogenous
market perceptions.
10

Using (9), the change in the capital-effective labor
ratio is given by,
.*
(Sp + Sg) - K + D
- [k + ( + )]k
k=
AL
.


Model composes a system of two differential
equations in d* and k. See Figure 16.1 for graphical
solution.
Figure 16.2.
11
Figure 16.1
Foreign Debt and the Capital-Effective Labor Ratio:
Long-Run Equilibrium
D
k
A
K
~
k
B
E
K
D
~
d*
d*
12
Figure 16.2
Growth, Foreign Debt, and Fiscal Adjustment
D
k
E'
K
~
k
E
K
D
~
d*
d*
13
Implications:
 Fiscal adjustment reduces foreign debt burden and
has a permanent and positive effect on growth.
 Despite high marginal returns to capital, risk premium
may still be high enough to lead to credit rationing.
 In such cases, foreign aid can help both directly by
increasing aggregate investment and indirectly
through conditional aid that may induce policy
changes to reduce distortions.
14
The Debt Overhang
and the Debt Laffer Curve
15



Medium-term uncertainty associated with heavy
external debt burden can have adverse effects on
domestic investment--a phenomenon that the
model did not capture.
Helpman (1989), Krugman (1988), and Sachs
(1989): beyond a certain point, external debt levels
act as a marginal tax on investment; fraction of
return on capital goes towards paying back
creditors.
Debt overhang: when expected repayment on
foreign debt falls short of its contractual value.
16
Illustrating debt overhang:
 Consider an economy with current debt, D*, that must
be paid in the next period.
 Repayment given by,
R = min(D*, Y - Cmin).

Production function given by,
Ys = UsI, 0 <  < 1,
s: state of nature,
I: investment output
Us: productivity shock; high (UH) and low (UL).
17

If UH and UL differ such that,

YL = uLImax < uHI

Debt overhang exists if,
D* > YL - Cmin
18
Debt Laffer curve
 Relates a country’s nominal debt obligations to the
market’s expectation of repayments on those
obligations.
 Figure 16.3: Y-axis (D*), X-axis (V-market
expectations).
 Laffer curve drawn as a straight line up until a point
A, and concave down thereafter.
 : probability of repayment falls below unity at debt
levels above point A.
 Results in a point E > A, that represents a maximum
for V.
19
Figure 16.3
The Debt Laffer Curve
V
Zero probability
of default
E
A
45º
0
Source: Adapted from Krugman (1989b).
D*
D
20
Thus, levels of D* higher than E result in lower V;
country is said to be on the wrong side of the debt
Laffer cure:
 reducing total debt would increase its market value;
 reduction in the debt tax rate would increase debt
servicing.
Froot and Krugman (1989):
 Tried to estimate debt Laffer curve for 35 countries.
 Used secondary market price of debt as proxy for V.
 Debt overhang exists when additional debt increases
the discount rate to such an extent that the marginal
debt reduces the total market value of the debt stock.

21
Found that in 5 of the 12 low-income countries, high
debt-to-export ratios placed them on the wrong side
of Laffer curves.
Two problems with results:
 Cannot be used when there is no secondary market
for debt.
 Classens (1990), observed debt Laffer curves, while
concave, rarely actually slope downwards.

22
Deshpande (1997):
 Examined investment experience of 13 highly
indebted countries from 1971-92.
 Found significantly negative relationship between
external debt levels and investment.
 Argued that debt overhang impacts investment
through two channels:
 direct disincentive effect, fear of appropriation of
funds invested for debt servicing;
 indirect effect; via adjustment measures
undertaken to face debt-servicing difficulties, e.g.
imports cuts and decreased public sector
investment.
23
Depshande implications:
 Complementarity between public and private capital
outlays implies that fiscal adjustment programs
would tend to reduce private investment.
 However, fiscal consolidation that reduces
government spending, by reducing the crowding
out effect of public sector demand for credit, may
lead to an expansion in private investment.
24
Eliminating debt overhang and capital formation renewal
 Debt relief beneficial to creditors, if increase in market
value of debt, V , is sufficient to make up for the
decline the face value of debt outstanding.
 Formal model (Krugman, 1989), see Figure 16.3
dV/dx = - + (1 - )(dI/dx)I-1 > 0,
: loss in good states of nature;
dI/dx: increase in marginal investment due to debt
relief;
I -1: marginal product of investment.
25

If dI/dx = 0, dV/dx = - < 0; barring investment effects,
debt relief never optimal from creditor viewpoint.
26
Measuring the Debt Burden
27
Debt Ratios
 Conventional measures:
 Stock of debt to output; debt stock can be a
misleading proxy, as it neglects both the duration
and interest rates on debt.
 Actual debt service to exports; a measure of a
countries foreign exchange constraints, low
value may reflect, in part, an unwillingness to
pay.
 Scheduled interest payments to exports.
 Second and third measure differ by concessions
made in rescheduling agreements and new lending.
28
Debt ratios that capture fiscal impact
 Two conventional measures that capture fiscal
impact of external debt burden:
 ratio of scheduled interest payments to
government revenue; capacity measure, ratio
considered low if less than .2, moderate at .2-.5,
high if above .5;
 ratio of scheduled interest payments to
government expenditure.
29
Alternative measures:
 Comparing present value of debt obligations against
the present value of estimated future export revenues.
 World Bank looks at present value of debt obligations
against current revenues.
 Practitioners consider ratios below 200% as
sustainable.
30
Three weaknesses of indicators:
 Data requirements complex; considerable uncertainty
over future exports; volume and prices.
 Results sensitive to assumed discount rate.
 Ratios provide no information about a country’s credit
risk profile.
31
Observed ratios
In many countries, conditions have worsened:
 In HIPC countries, external debt reaches $207 billion
in 1996, or 1.5 times GNP.
 Average ratio of foreign debt to exports was 450% in
1995.
 Ratio of actual debt service payments to exports was
17% in 1995.
 Ratio of actual debt service to government revenues
reached 70% in Uganda from 1992-94.
 Figures 16.4 to 16.8.
32
Figure 16.4a
Total External Debt Service
(In percent of GDP)
1980
1995
Africa
Asia
Benin
Bangladesh
Burundi
India
Cameroon
Indonesia
Côte d'Ivoire
Korea
Ethiopia
Gabon
Malaysia
Ghana
Nepal
Kenya
Pakistan
Malawi
Philippines
Nigeria
Sri Lanka
Tanzania
Thailand
Zimbabwe
0
5
10
15
0
1
2
3
4
5
6
7
8
9
Source: World Bank.
33
Figure 16.4b
Total External Debt Service
(In percent of GDP)
1980
1995
Latin America
Middle East and North Africa
Argentina
Algeria
Bolivia
Egypt
Brazil
Jordan
Chile
Mauritania
Colombia
Costa Rica
Morocco
Ecuador
Oman
Jamaica
Syria
Mexico
Tunisia
Peru
Turkey
Uruguay
Yemen
Venezuela
0
5
10
Source: World Bank.
15
20
0
2
4
6
8
10
12
34
Figure 16.5a
Total External Debt Service
(% of exports of goods and services)
1980
1995
Africa
Asia
Benin
Bangladesh
Burundi
India
Cameroon
Indonesia
Côte d'Ivoire
Korea
Ethiopia
Gabon
Malaysia
Ghana
Nepal
Kenya
Pakistan
Malawi
Philippines
Nigeria
Sri Lanka
Tanzania
Thailand
Zimbabwe
0
10
20
Source: World Bank.
30
40
50
0
10
20
30
40
50
35
Figure 16.5b
Total External Debt Service
(% of exports of goods and services)
1980
1995
Latin America
Middle East and North Africa
Argentina
Algeria
Bolivia
Egypt
Brazil
Jordan
Chile
Mauritania
Colombia
Costa Rica
Morocco
Ecuador
Oman
Jamaica
Syria
Mexico
Tunisia
Peru
Turkey
Uruguay
Yemen
Venezuela
0
20
40
60
80
100
120
0
10
20
30
40
50
Source: World Bank.
36
Figure 16.6
Heavily Indebted Poor Countries: Debt Indicators, 1992-94
(In percent)
Paid debt service/
Debt to exports ratio
( present value of external debt)
Angola
Angola
Benin
Benin
government revenue
Paid debt service/
(excluding grants)
scheduled debt service
Angola
Benin
Bolivia
Bolivia
Bolivia
Burkina Faso
Burkina Faso
Burkina Faso
Burundi
Burundi
Burundi
Cameroon
Cameroon
Cameroon
Central African R
Central African R
Central African R
Chad
Chad
Chad
Congo
Congo
Congo
Côte d'lvoire
Côte d'lvoire
Côte d'lvoire
Equatorial Guinea
Equatorial Guinea
Equatorial Guinea
Ethiopia
Ethiopia
Ethiopia
Ghana
Ghana
Ghana
Guinea
Guinea
Guinea
Guinea-Bissau
Guinea-Bissau
Guinea-Bissau
Guyana
Guyana
Guyana
Honduras
Honduras
Honduras
Kenya
Kenya
Kenya
Lao
Lao
Lao
Liberia
Liberia
Liberia
Madagascar
Madagascar
Madagascar
Mali
Mali
Mali
Mauritania
Mauritania
Mauritania
Mozambique
Mozambique
Mozambique
Myanmar
Myanmar
Myanmar
Nicaragua
Nicaragua
Nicaragua
Niger
Niger
Niger
Nigeria
Nigeria
Nigeria
Rwanda
Rwanda
Rwanda
SaoTomé & Prínc
SaoTomé & Prínc
SaoTomé & Prínc
Senegal
Senegal
Senegal
Sierra Leone
Sierra Leone
Sierra Leone
Somalia
Somalia
Somalia
Sudan
Sudan
Sudan
Togo
Togo
Togo
Uganda
Uganda
Uganda
Tanzania
Tanzania
Tanzania
Viet Nam
Viet Nam
Viet Nam
Yemen
Yemen
Yemen
Zambia
Zambia
Zambia
0
2,000
Source: United Nations.
4,000
0
20
40
60
80
0
20
40
60
80
37
Figure 16.7a
Heavily Indebted Poor Countries:
Debt to Exports Ratio and Per Capita GDP Growth Rate 1/
(In percent)
2,500
External debt measured at face value (1992-94)
2,000
1,500
1,000
500
0
-5
-4
-3
-2
-1
0
1
2
3
4
5
Per capita real GDP growth rate (1985-94)
Source: United Nations.
1/ Excluding Ethiopia, Cameroon, Congo, Nicaragua, Somalia, and Sudan.
38
Figure 16.7b
Heavily Indebted Poor Countries:
Debt to Exports Ratio and Per Capita GDP Growth Rate 1/
(In percent)
1,400
External debt measured in present value terms (1992-94)
1,200
1,000
800
600
400
200
0
-5
-4
-3
-2
-1
0
1
2
3
4
5
Per capita real GDP growth rate (1985-94)
Source: United Nations.
1/ Excluding Ethiopia, Cameroon, Congo, Nicaragua, Somalia, and Sudan.
39
Figure 16.8
Heavily Indebted Poor Countries:
Paid Debt Service/Exports and Per Capita GDP Growth Rate, 1985-94 1/
(In percent)
Paid debt service/exports of goods and services
40
30
20
10
0
-5
-4
-3
-2
-1
0
1
2
3
4
5
Per capita real GDP growth rate
Source: United Nations.
1/ Excluding Ethiopia, Cameroon, Congo, and Uganda.
40
External Solvency Measures:
 Current account deficit in an open economy defined
by,
CA = D* - D-1* = -(Y - A) - NT + i*D-1*

A: domestic absorption;
NT: net transfer respeipts from abroad;
i*: interest rate on foreign liabilities.
This yields,
D* = (1 + i*)D-1* - Z
Z: net external surplus, the trade balance plus net
41
transfers.

Setting d* = D*/ Y, z = Z/Y and g: constant growth
rate of output, and applying the no ponzi game
condition,

d0*


h=1

1+g h
(
) zh
*
1+i
(23)
By (23) external solvency requires that current ratio
of net foreign liabilities to output be less than or equal
to the maximum level of debt that can be sustained by
prospective net external surpluses.
42
Debt Rescheduling
43
Two approaches to debt repayment (see Kletzer, 1988)
 Ability-to-pay approach:
 required debt service exceeds debt servicing
capacity;
 in general a solvency rather than liquidity
problem.
 Willingness-to-pay approach.
Two types of debt relief approaches:
 flow reschedulings;

stock-of-debt reschedulings.
44
Toronto terms:
 From October 1988-June 1991, terms of concessions
as outlined by the Paris Club:
 Option A, or partial cancellation. 1/3 of debt
canceled, remaining two thirds rescheduled over
14 years with 8 year grace period.
 Option B, or extended maturities. Debt
rescheduled over 25 years with 8 year grace
period.
 Option C, or concessional interest rates. Debt
rescheduled over 14 years, 8 year grace, with
interest rate set at market rate - 350 basis points.
45
London terms:
 December 1991-December 1994; 50% reduction in
NPV terms.
Naples terms:
 Since January 1995, terms of concessions as outlined
by the Paris Club:
 Eligibility; decided by creditors on individual basis.
 Concessionality; most receive 67% NPV
reduction on non-ODA (official development
assistance) debt. Countries with per capita income
greater than $500 and debt/exports < 350%
receive 50% NPV reduction.
 Coverage; non ODA pre-cut-off date debt decided
46
on individual basis.
Choice of options; DR (Debt reduction; 23 year
maturities with 6 year grace) or DSR (Debt
service reduction; 33 year maturities with
concessional interest rates).
 Capitalization of moratorium interest option.
 ODA credits; pre-cutoff date credits rescheduled
on the original concessional interest rates over
40 years with 165 years’ grace.
Uganda first low-income rescheduling country to
receive stock of debt rescheduling under Naples
term.


47
The HIPC Debt Initiative
48




For HIPCs, multilateral debt a large fraction of total
external debt, about 31% at the end of 1996, 42%
for bilateral official creditors and 12% for private
creditors.
HIPC Debt Initiative: joint World Bank-IMF
proposal, September, 1996:
addresses debt owed to multilateral official
creditors;
requires countries to show a track record of good
policy performance as monitored by the IMF and
the Bank;
6-year performance period, two stages of
implementation.
49
First Stage:
 Initial 3-year period; debtor country must establish
track record with Paris Club providing flow
rescheduling on Naples terms.
 Subsequent decision point reached after a threeyear track record of good performance.
 Debt sustainability analysis undertaken to
determine eligibility for further debt relief.
Three possible scenarios:
 If debt sustainability possible in three years, country
is deemed capable of exiting debt rescheduling
process.
 If sustainability deemed not possible, then country
50
becomes eligible for further debt relief.
In cases of doubt, flow rescheduling under Naples
terms.
Second Stage:
 Normally three years.
 Debt relief through flow rescheduling, up to 80% of
present value of Paris Club debt.
 At completion point of second stage, Paris Club
provides 80% reduction in NPV of debt stock.

51
Eligibility:
Debt sustainability analysis determines
eligibility and amount of debt relief based on
following criteria:
 ratio of NPV of debt-to-export ratio between 200250% by completion point;
 debt-service-to-exports ratio between 20-25% at
completion point;
 various measures of vulnerability;
 ratio of NPV of debt to fiscal revenue below 280%
with 2 other criteria: export-to-GDP ratio of at least
40% and fiscal revenue to GDP of 20%.
52