Chap01.ppt, consumption, saving and investment

Chapter 1
Consumption, Saving
and Investment
© Pierre-Richard Agénor
The World Bank

 Consumption and Saving
 Investment

Basic facts (Figure 1.1):
 Gross domestic saving rates in developing
countries are higher than the rates in industrial
countries (highest in Asia).
 Investment rate follows similar pattern (highest in
Asia).
 Foreign saving is particularly large in Sub-Saharan
Africa.

Figure 1.1
Saving and Investment Rates, 1976-97
(Percentage of GDP)
United States
European
Union
Japan
Developing
countries
Africa
Asia
Middle East
and Europe
Latin America
and Caribbean
35 30 25 20 15 10 5 0
Gross domestic savings (S) Gross domestic investment (I)
0 5 10 15 20 25 30 35
1976-83 1984-91 1992-97
S-I
-0.9
-2.2
-1.5
0.2
0.3
0.5
0.6
2.8
2.4
-0.8
-1.9
-1.9
-4.6
-3.0
-4.4
0.5
-1.7
-1.3
3.4
-4.4
-1.6
-3.1
-0.4
-2.6
Source: International Monetary Fund.

Consumption and Saving
 Keynesian Approach
 Permanent Income Hypothesis
 The Life-Cycle Model
 The Basic Framework
 Age and Dependency Ratio
 Other Determinants
 Income Levels and Income Uncertainty
 Intergenerational Links
 Liquidity Constraints
 Inflation and Macroeconomic Stability
 Government Saving

 The Debt Burden and Taxation
 Social Security, Pensions and Insurance
 Changes in the Terms of Trade
 Financial Deepening
 Household and Corporate Saving
 Empirical Evidence

Keynesian Approach
 Consumption is a function of disposable income:
C = (1 - s)(Y - T),
C current consumption, Y - T disposable income;
0 < s < 1 marginal propensity to save.
Merits:
 First approximation in empirical macroeconomic
models.
 A reflection of the behavior of consumers subject to
liquidity constraints.

The Permanent Income Hypothesis
 Consumption is a function of permanent income.
 Example : Consumers are identical and live for only
two periods, 1 and 2. Assume perfect foresight.
 Budget constraint for period 1:
A1 - A0 = Y1 - T1 + rA0 - C1 (1)
A: stock of financial assets, Y - T disposable
income;
r: real interest rate (constant).

 Budget constraint for period 2, in the absence of
bequest:
C2 = Y2 - T2 + (1+r)A1 (2)
 Eliminate A1 from (1) by using (2).
 Yield the household’s intertemporal budget
constraint:
C2 Y2 - T2
C1 + ––––– = (1+r)A0 + (Y1 - T1) + ––––––––
1+r 1+r
(3)

Simple version of the model:
 Household’s objective is to maintain a perfectly
stable (or smooth) consumption path, C1 = C2.
 Divide its lifetime resources equally among each
period of life.
 Amount consumed by the household in each period
is equal to its permanent income, Yp .
 Yp : level of income that gives the household the
same present value of its lifetime resources as that
implied by its intertemporal budget constraint.

 Using equation (3) intertemporal budget constraint is
 Then Yp becomes:
Yp Y2 - T2
Yp + ––––– = (1+r)A0 + (Y1 - T1) + ––––––––
1+r 1+r
1+r Y2 - T2
Yp = (–––––){(1+r)A0 + (Y1 - T1) + ––––––– }
2+r 1+r

(Y1 - T1 )+ (Y2 - T2 )
Yp = –––––––––––––––––––––
2
 If A0 = 0 and r = 0, YP becomes an exact average of
present and future disposable income.
Implications:

Saving (in period 1) is the difference between disposable
income and permanent income.
S1 = Y1 - C1 = Y1 - Yp

 Transitory income :
YT = Y1 - Yp
 This forms the basis for a number of empirical
tests.

 Importance of variations in the structure of
income during life cycle.
 Figure 1.2: stylized pattern of income, consumption,
and savings predicted by the standard life-cycle
model.
The Life-Cycle Model

Figure 1.2
Income, Consumption, and Saving
in the Life-Cycle Model
Source: Adapted from Deaton (1999, p. 42).
Age
A
Income
Retirement Death
Consumption
Borrowing
Income,
Consumption,
and
Saving
Saving
Dissaving
Retirement
income
Consumption
B
B'
C
C'
Dissaving

Basic Framework:
 Two period framework.
 Life time budget constraint :
C2
C1 + ––––– = W1 (4)
1+r
W1: life-time wealth.

 Suppose that the household’s preferences are
intertemporally additive :
u(C2)
U = u(C1) + –––––– (5)
1+r
U: life-time utility;
r: rate of time preference which measures the
degree of impatience.

C2
L = U + [C1 + ––––– - W1]
1+r
u’(C2) 
u’(C1 ) = , ––––––– = –––––.
1+r 1+r
 Maximization of (5) with respect to C1 and C2 subject
to the life-time budget constraint (4).
 By forming the Lagrangien expression:
: Lagrange multiplier.

The first order optimality conditions are given by:

 Combining these two equations obtain Euler
equation :
1+r
u’(C1 ) = –––––u’(C2)
1+
- - -1/ 
U = {C1 + C2 /(1+)}
 When  = r, we obtain C1 = C2. The model becomes
the simple version of permanent income hypothesis.
 Assume constant elasticity of substitution :

-1/ 1+r -1/
C1 = (–––––)C2
1+
 Taking logarithms of both side
1+r
ln(C2/C1) = ln(––––---)  (r - )
1+
 The elasticity of substitution between period 1 and period 2
consumption, , is :
 = 1/(1+).
 Euler equation becomes :

  measures the responsiveness of the change in
consumption between the two periods to changes in
interest rate, r.
 The effect of the change in r on consumption and
saving (in period 1) is indeterminate.
 Conflict between income and substitution effects.
 The greater is , the greater will be the reduction in
C1 (relative to C2) induced by a rise in r.
 If  is sufficiently large: effect of substitution
dominates; an increase in r reduces consumption
and raises saving.

Predictions of life-cycle model
 The young will save relatively little as they anticipate
increases in their future income.
 Middle-aged individuals, who are nearing the peak of
their earnings, tend to save the most, in anticipation of
relatively low incomes after retirement.
 The elderly tend to have a low, or even negative, saving
rate, although the desire to leave a bequest or to cover
the contingency of living longer than expected could
provide motivation for saving even after retirement.
Age and Dependency Ratio

 Implication: Aggregate saving rate will tend to fall in
response to dependency ratio, measured as
 youth dependency ratio (ratio of under-20 age
group to the 20-64 age group);
 ratio of the elderly to the working age population.
 Distribution of assets among the population
affects the consumption and saving patterns at the
aggregate level.
 The larger the share of total wealth held by the
middle-aged households, the higher the saving rate,
and the higher the growth rate of income in a given
country.


Remark : demographic factors such as the share of the working
population relative to the that of retired persons are likely to
explain only the long-term trends in saving.

Income levels and income uncertainty
 Recent empirical research has highlighted the fact
that at low or subsistence levels of income, the
saving rate is also low.
 Two implications:
 in low-income countries the response of saving
to changes in real interest rate is likely to be
weak;
 changes in income distribution can have
important effects on measured saving rates at
the aggregate level.

 Increased uncertainty regarding future income will
enhance the precautionary motive for saving.
Sources of income uncertainty :
 Many households in developing countries derive
their incomes from agriculture.
 In that sector, incomes can be subject to relatively
large fluctuations resulting from variations in
climatic conditions or changes in domestic and
the world prices of agricultural commodities.
 Macroeconomic instability.

 Intergenerational links are likely to be strong in
developing countries (role of extended family).
 These links can affect consumption and saving in
two ways:
 lengthen the effective planning horizon over
which households make their consumption and
saving decisions;
 affect household preferences (by affecting, for
example, the degree to which the marginal
utility of consumption).
Intergenerational Links

Liquidity Constraints
 Consumption smoothing requires well-functioning
financial markets to allow agents to borrow and
lend across periods.
 In many developing countries, well-developed
financial markets either do not exist, or not function
very well.
 Households often have limited access to credit
markets, and credit rationing may be pervasive.

 Liquidity constraints affect the ability of
households to transfer resources across time
periods as well as across uncertain states of nature
relative to income.
 Consequence : consumption tends to be highly
correlated with current income, rather than
permanent income or life-cycle wealth.
 In the presence of liquidity constraints, financial
liberalization can have an adverse effect on saving
rates.
 Increased access to these markets will allow
individuals to bring forward their consumption
(reduce saving).

 If households are net creditors, an increase in the
inflation rate may lower real value of wealth. To
offset this, they raise their saving rate.
 The variability of inflation (measure of
macroeconomic stability) may affect saving by
increasing uncertainty about future income.
 Precautionary motive : a high degree of price
variability may lead to an increase in the saving
rate.
Inflation and Macroeconomic Stability

 Key feature of the life-cycle model: saving behavior
is directly influenced by households’ assessments
of their future disposable income.
 Key variable that affects these assessments is
government policy, particularly government saving
or dissaving.
 Three major interpretations of the relationship
between government and private saving:
Government Saving

Conventional view:
 Assume a fall in government saving (resulting from
a tax cut or a bond-financed increase in government
spending).
 This will tend to raise consumption and reduce
saving by myopic households (that is, households
who care solely about the present).
 Reason: they shift the tax burden from present to
future generations .
 Decline in government saving will lead to a decline
in national saving.

The Keynesian view
 Higher temporary government dissaving.
 Consumption and income increase in the presence
of under-utilized production capacity: multiplier
effect.
 Higher income will raise private saving.
 Whether or not this increase in private saving is
large enough to offset the initial decline in
government saving is a priori ambiguous.

The Ricardian equivalence view
 Predicts that a rise in the budget deficit resulting
from a tax cut will have no effect on the national
saving rate because private saving will rise by an
equivalent amount in anticipation of future tax
liabilities.
 If individuals are rational and far-sighted, they will
realize that a permanent rise in government
spending today must be paid for either now or later.
 They will increase saving by an equivalent amount.

 Critics for the assumptions of Ricardian
equivalence from the analytical point of view:
 consumers are far-sighted;
 successive generations are linked by altruistically
motivated bequests;
 consumers do not face liquidity constraints;
 taxes are nondistortionary.

 Empirical results for developing countries is
against Ricardian equivalence.
 Reason: although individuals may form
expectations about their future tax liabilities in a
systematic way, liquidity constraint may prevent
them from acting on these expectations.

Social Security, Pensions, and Insurance
 The availability of formal public pension and social
security schemes may cause to lower the private
saving rate.
 Channels implied by the life-cycle model:
 by redistributing income to the elderly;
 by reducing the need to save for retirement (if
there is no reduction of the retirement age);
 by curbing the need for precautionary saving to
cover the contingency of living longer than
expected.

 The impact of increased social security benefits on
national saving depend on the effect that such
changes on public saving.
 Private pension plans have been developed in
many developing countries in recent years.
 In principle, individuals should view their
contributions to funded private pensions as a
perfect substitute for other forms of saving.
 But, in practice, individuals do not seem to fully take
into account their pension contributions in
determining their saving behavior.


Result : introduction of private pension plans is often
accompanied by an increase in national saving rates.

Conclusion of Holzmann (1997) in the case of Chile.
 Availability of various kinds of insurance:

health insurance ;

unemployment insurance ;
 personal loss and liability insurance.
 They influence saving behavior.

To the extent that insurance plans limit expected
outlays for contingencies and emergencies, they tend
to reduce income uncertainty and therefore the need
for precautionary saving.

 Movements in terms-of-trade has an important
effect on saving.
 Harberger-Laursen-Meltzer Effect: predicts a
positive relationship between changes in the terms-
of-trade and saving, through their positive effect on
wealth and income.
Predictions:
 Temporary decrease in terms of trade leads to
decrease in current income compared to future
income thus leads to a decrease in saving.
 If permanent deterioration in the terms of trade
leads to reduction in both permanent and transitory
income, no effect on saving.
Changes in the Terms of Trade

Financial Deepening

Financial development may affect saving both directly
and indirectly:

reduction in cost of intermediation leads to
increase in the return to saving;

increased efficiency in the process of financial
intermediation leads to an expansion of investment
and stimulates the rate of economic growth;

increase in income leads to an increase in saving.

Figure 1.3: positive relationship between gross
domestic saving rates and an indicator of financial
deepening.

Figure 1.3
Financial Deepening and Saving Rates
(Averages over 1980-95)
Source: World Bank.
Ratio
of
quasi
money
to
broad
money
stock
Gross Domestic savings (% of GDP)
0 5 10 15 20 25 30 35 40
10
20
30
40
50
60
70
80
90
Ghana
Malaysia
Algeria
Korea
Indonesia
Thailand
Bangladesh
Venezuela Costa Rica
Chile
Brazil
Tunisia
Panama
Peru
India
Nigeria
Zimbabwe
Côte d'Ivoire
Pakistan
Nepal
Jamaica
Bolivia
Philippines
Morocco
Colombia
Zambia
Tanzania

 Forgoing discussion focused only on saving by
households.
 This focus justified in the many developing countries
where private saving rates are essentially
determined by household behavior.
 However corporate saving (retained earnings) may
also be significant.
 They may respond to different variables than those
affecting the decisions of households.
Household and Corporate Saving

 Importance of this distinction in the aggregate
private saving depends on households’ responses
to higher corporate saving.
 If firms retain more earnings, households may have
less by a corresponding amount:
 In such conditions, households pierce the
corporate veil.
 Aggregate private saving behavior will largely
reflect household behavior.
 Example: Colombia (Figure 1.4).

Source:López-Mejía and Ortega (1998).
Household Saving
Corporate Saving
Total Private Saving
Figure 1.4
Colombia: Components of Private Saving, 1950-93
(Percent of GNP)
1950
1952
1954
1956
1958
1960
1962
1964
1966
1968
1970
1972
1974
1976
1978
1980
1982
1984
1986
1988
1990
1992
0
0.05
0.1
0.15
0.2

 Masson, Bayoumi, and Samiei (1995)
 Edwards (1996c)
 Dayal-Gulati and Thimann (1997)
 Loayza, Schmidt-Hebbel and Servén (1999).
Empirical Evidence

Masson, Bayoumi, and Samiei (1995)
 Used cross-country database (developing countries)
to study determinants of private saving.
Results:
 increase in public saving associated with higher
national saving, suggesting Ricardian equivalence
does not hold;
 decrease in age dependency ratio raises private
saving;
 increase in per capita income raises private saving;

 changes in real interest rate had no significant
effect on saving;
 increases in foreign saving increases both
investment and consumption;
 terms-of-trade windfalls have a positive but
transitory effect on saving.

Edwards (1996c)
 Used both developing and industrialized countries.
 Significant determinants of private saving rate:
 rate of growth of per capita income;
 monetization ratio (indicator of financial
deepening);
 foreign saving (negative effect);
 government saving (negative effect);
 social securities (negative effect).

Dayal-Gulati and Thimann (1997)
 Used both Southeast Asia and Latin America
countries.
 Determinants of private saving rate:
 terms of trade shock: positive effect;
 government saving: partially crowd out private
saving;
 social security expenditures: negative effect;
 fully funded pension schemes: positive effect;
 macroeconomic stability: positive effect;
 financial deepening: positive effect;
 per capita income: positive effect.

Loayza et al. (1999)
 Used an extensive cross-country database to study
determinants of private saving.
 A novelty of the analysis : distinction between
short- and long-term determinants of saving rate.
 This distinction is highly significant in their empirical
results.
Main findings:
 Macroeconomic uncertainty (the variance of
inflation) had a positive effect on private saving
rates. Consistent with the precautionary motive.

 Public sector saving had a negative but less than
proportional effect on private saving. Ricardian
equivalence does not hold in strict terms.
 Real interest rates had no significant effect on
saving.
 Terms-of-trade improvements were positively
associated with private and national saving rates.
Limitations:
 Effect of interest rates on saving may be nonlinear.
 Asymmetric effects of terms-of-trade.

Investment
 The Flexible Accelerator

The Cost of Capital

Uncertainty and Irreversibility
 Other Determinants of Investment
 Credit Rationing
 Foreign Exchange Constraint

The Real Exchange Rate

Public Investment

Macroeconomic Stability

The Debt Burden Effect
 Empirical Evidence

Flexible Accelerator
 Assumption: production technology is characterized
by a fixed relationship between the desired capital
stock and the level of output.
~
K = (K - K-1), 0 <  < 1.
 Suppose actual capital adjustment is:
K: desired capital stock; ya
: expected output.
~
~
K = ya
,  > 0,

 Gross private investment:
Ip
= K + K-1, 0 <  < 1,
: depreciation rate
 When  = 0,  = 1, and expected future output is
approximated by current output:
Ip
= y.
 It relates investment linearly to changes in current
output.
 Limitation: profitability, uncertainty, and the cost of
capital play no role.

Cost of Capital
 View investment as depending inversely on the
user cost of capital.
Three components of user cost of capital:
 opportunity cost; measured by the interest rate the
firm would receive if it sold the capital and invested
the proceeds;
 cost resulting from the depreciation of the capital
good;
 capital loss (or gain) resulting from the fact that the
price of capital may be falling (rising).

 Cost of capital:
cK = PK [i +  - (PK / PK)],
i: interest rate; PK the price of one unit of capital;
: depreciation rate;
i - PK /PK : real interest rate measured in PK.
 When combined with the flexible accelerator:
K = ya
/ cK
~
 Investment is inversely related to the cost of capital
services.

 Limitation: it does not account for the impact of
uncertainty on the decision to invest.

Uncertainty and Irreversibility
 Under uncertainty, private investment decisions
may be significantly affected by irreversibility
effect (essentially due to sunk cost).
 Because of irreversibility of investment, waiting
has value as it gives firms the opportunity to
process new information before the decision to
invest is taken.
Servén (1997) model:
 Examine the effects of uncertainty and
irreversibility on investment.

Assumptions:
 Risk-neutral firm must decide whether to invest in
a project in which the initial cost is completely sunk
at the purchase cost PK at the beginning of period
t0 = 0.
 It yields a return of R0 at the end of that period.
 Future demand for the good generated by the
project is uncertain; as a result, the rate of return
on the project in period t = 1 and beyond, denoted
R, is also uncertain.

 Net present value of the anticipated return
stream of cash flows associated with the project:
V0  - PK + (1+)-h
E0R,
R0
1+i 1+i

h = 0

+ 1
2
i: Discount rate, taken to be equal to the rate of
return on an alternative investment, such as
riskless government bonds.
E0R: Given the information available at period 0,
the expected value of the future return.

 This can be rewritten as:
V0  - PK +
R0+ E0R /i
1+i

The conventional net present value criterion suggests
that the investment is profitable and thus should be
made as long as V0 > 0. After rearranging terms yields:
R0 - iPK + > 0 , (33)
E0 R - iPK
i
iPK: user cost of capital in the case where the
depreciation rate is zero.

 The presence of irreversibility requires taking into
account both the difference between the expected
return and the user cost of capital (Equation (33)).

With full reversibility of investment, the future would not matter;
the optimal decision rule would thus be to invest today as long as:
R0 - iPK > 0,
i.e. as long as the current return exceeds the user cost of capital.
(34)

 But although Equation (33) must hold in an ex ante
sense, it may not ex post; the reason is that there
is a nonzero probability that at some period t in
the future, the inequality Equation (34) may be
reversed, that is, R - PK < 0.
 The firm may thus be locked in an unprofitable
investment.
 There is, therefore, an incentive to delay
investment in order to learn more about the factors
affecting future return (in the present case, about
the state of market demand for the good produced
by the firm).

 To determine how uncertainty affects the decision
rule (33), consider first the case where the firm
knows for sure that uncertainty will completely
vanish in period t = 1 and that the project's returns
for t = 2,... will remain constant at the level
realized in the first period.
 Suppose then that the firm decides not to invest at
all today and to invest next period if and only if the
realized return exceeds the user cost of capital.

V1  Pr(R > i PK)
(1+i)-h
E0(R | R > i PK)
 In that case, the net present value of the
anticipated stream of cash flows:
- PK
1+i 1+i

h = 0

+ 1
2
{ }
Pr(R > iPK) : probability that the project's return
exceeds the cost of capital;
E0(R | R > iPK): expected value of R, conditional
on the project's return exceeding the cost of
capital.

 Comparing V0 with V1, the firm is better off investing today if :
V1 - V0 < 0,
a condition can be written as
R0 - i PK: cost of waiting, given by the net return
foregone in period 0 by not investing.
(R  i PK): value of waiting, given by the irreversible
mistake that would be revealed tomorrow if future
returns fall short of the user cost of capital.
PK > Pr(R  i PK)
E0(i PK - R | R  i PK)
i
(35)

 The expected present value of such mistake is
measured by the right-hand side of Equation (35):
 mistake is made with probability Pr(R  i PK);
 its expected per-period size, given today's
information, is
E0(i PK - R | R  i PK);
 because it accrues every period into the
indefinite future, it has to be multiplied by 1/i to
transform it to present value terms.

 Thus, condition (33) indicates that it is profitable to
invest immediately only if the first-period return
exceeds the conventionally measured user cost of
capital by a margin that is large enough to
compensate for the possibility of an irreversible
mistake.
 In other words: if the cost of waiting outweighs
the value of waiting.
 Implication of Equation (35): possibility that in the
future R may exceed iPK has no effect on the
investment threshold and thus no effect on the
decision to invest today.

 Reason for this asymmetry: option to wait has no
value in those good states of nature in which
investing would have been the right decision
anyway.
 Option value of waiting:
 = max(V1-V0 , 0).
 If V1 - V0 < 0, the option has no value, and the
optimal decision is to invest today (at period 0).
 In general, however, the option value of waiting can
be large, especially in a highly uncertain
environment.

 As a consequence, uncertainty can become a
powerful deterrent to investment even under risk
neutrality.
 Uncertainty may result from various domestic and
external sources, including a high degree of
volatility in aggregate demand, large movements in
the terms of trade and relative prices, and
incomplete credibility of adjustment policies.
 Increased macroeconomic volatility raises the
likelihood of bad outcomes (i.e. R  iPK) .
 Result: increase in the spread of the distribution of
future returns.

 This will raise the critical threshold that the marginal
productivity of capital must reach, and thus tend to
depress investment.
 But this does not always hold.
 If R0 is uncertain and the investment is partly
reversible, then higher uncertainty could hasten
investment, by making extreme favorable
realizations of R0 more likely.
 Reason: firm can avoid the impact of negative
outcomes on profitability by shutting down the
project (Bar-Ilan and Strange, 1996).
 Although both the value and the cost of waiting rise
with higher uncertainty, the latter rises by a greater
amount.

 The higher the degree of irreversibility (that is, the
higher the degree of asymmetry in investment
adjustment costs), the more likely it is that
uncertainty will have an adverse effect on capital
formation.
 Recent research on the relation between
uncertainty and investment: role of various other
factors;
 market structure;
 degree of risk aversion;
 capital market imperfections.

Caballero (1991):

Under asymmetric investment adjustment costs and
with risk neutrality, uncertainty and investment tend
to be positively related under perfect competition and
constant returns to scale.
 But they tend to be negatively related under
imperfect competition and decreasing returns to
scale.
Zeira (1990):

With risk-averse agents, uncertainty has an
ambiguous impact on investment.

The higher the degree of risk aversion, the more
likely it is that uncertainty will reduce investment.

Aizenman and Marion (1999):
 Negative link between uncertainty (or volatility) and
investment can result from the existence of a credit
ceiling.
 Such a ceiling may lead to a nonlinearity in the
investor's intertemporal budget constraint.
 This hampers the expansion of investment in good
times without mitigating the fall in bad times.
 This asymmetry may lead to a situation in which
higher volatility reduces the average rate of
investment.
 On purely theoretical grounds, the effect of
uncertainty on private investment is ambiguous.

 Because uncertainty affects investment through a
variety of channels and, depending on the degree
of risk aversion, market structure, and the nature of
adjustment costs, the relation between these
variables can be either positive or negative.
 The higher the degree of irreversibility, the more
likely that uncertainty will have an adverse effect
on capital formation.
 Increased macroeconomic volatility increases
likelihood of bad outcome of investment. This leads
to the value of waiting to rise thus tend to depress
investment.

 Beside to uncertainty, market structure, the degree
of risk aversion, and capital market imperfections
may also effect investment decision.

Other Determinants of Investment

 Lack of development of equity markets makes
firms highly dependent on bank credit for working
capital needs and longer-term financing of capital
accumulation.
 When interest rates are highly regulated, excess
demand for credit will exist, forcing banks to ration
their loans.
 Since banks are imperfectly informed about the
quality of the investment project, this may also
lead to credit rationing.
 So beside to interest rate, quantity of credit should
be considered as a determinant of investment.
Credit Rationing

 Capital goods such as machines and equipment
must often be imported in developing economies.
 Investment may be subject to a foreign exchange
constraint if the foreign exchange needed to pay
for such imports may not be available due to
higher-priority needs.
Foreign Exchange Constraint

Real exchange rate affects private
investment through two channels:
 Demand side: real exchange rate depreciation
lowers private sector real wealth and expenditure
through its effect on domestic prices.
 This may lead firms to revise their expectations of
future demand and to lower investment through the
accelerator effect.
 Supply side: real depreciation raises the price of
traded goods relative to home goods.
The Real Exchange Rate

 It may stimulate investment in the tradable sector
and depress capital formation in the nontradable
sector.
 If the price of domestic factors of production
increases less than proportionately to the
domestic-currency price of final output, a real
depreciation will stimulate aggregate supply and
raise private investment.
 If a real depreciation raises the real cost of
imported capital goods, it will have an adverse
effect on private investment by raising the user
cost of capital or by dampening expectations of
future output through the accelerator effect.

Public investment has an ambiguous effect
on private investment as a result of two opposing
factors:
 Public investment may, by increasing the fiscal
deficit, crowd out private capital formation by
reducing credit available to the private sector or by
raising interest rates.
 Public investment in infrastructure projects may be
complementary to private investment.
 Figure 1.5.
Public Investment

Figure 1.5
Private and Public Investment Shares
(in percent of GDP)
Source: Aizenman and Marion (1999, pp. 173-74).
Note: Countries not identified in the figure are Peru, Paraguay, Bangladesh, Costa Rica, Nepal, Colombia,
Iran, Zimbabwe, Venezuela, Ecuador, Kenya, Sri Lanka, and Mali.
0 5 10 15 20 25
0
5
10
15
20
Public
investment
share
Panama
Uruguay
Thailand
Korea
Morocco
Tunisia
Brazil
Côte d'Ivoire
Indonesia
Pakistan
India
Philippines
Chile
Turkey
Tanzania
Ghana
Argentina
Nigeria
Private investment share
Guatemala
Dom. Republic
Mauritius
Malaysia
Papua New Guinea
Fiji
Bolivia
Madagascar
Malawi
Egypt
Guyana
Paraguay
Mexico
Haiti
El Salvador

 Irreversibility and asymmetric adjustment costs
cause macroeconomic instability to have large
negative effects on private capital formation.
 High level of inflation (characterizes
macroeconomic instability) may lower investment
by distorting price signals and the information
content of relative price changes.
 High inflation variability (translated into by
macroeconomic instability) may have an adverse
effect on expected profitability and if firms are risk
averse, their level of investment will fall.
Macroeconomic Instability

 Increase in policy uncertainty: risk-averse firms
reallocate resources away from risky activities
thereby lowering the desired capital level. By the
accelerator effect, this fall may translate into a
reduction in private investment.

High ratio of foreign debt to output may have
an adverse effect on private investment through
various channels.
 Resources used to service the public debt may
crowd out government investments in areas
where large complementarities exist between
public and private capital outlay.
 Domestic agents may want to transfer funds
abroad instead of investing them because of the
fear of future tax liabilities to service this debt.
Debt Burden Effect

 Discourage foreign direct investment by increasing
the likelihood that the government may resort to
the imposition of restrictions on external
payment.
 If foreign direct investment is complementary to
domestic private investment, the latter will fall also.
 When firms hold a large stock of foreign-currency
liabilities, they become vulnerable to exchange rate
movements.
 When a nominal depreciation raises, the burden of
debt and the risk of default increase. This may lead
domestic banks to tighten credit restrictions and
depress investment.

 Figure 1.6: negative relationship between debt
burden and private investment.

Source: World Bank.
Figure 1.6
Heavily Indebted Poor Countries:
Total Debt Service and Private Investment
(In percent, 1982-1995)
Private Investment (in percent of GDP), 1982-95.
0 5 10 15 20 25
0
5
10
15
20
Central African Rep.
Sierra Leone
Madagascar
Niger
Burundi
Ethiopia
Chad
Ghana
Bolivia
Zambia
Uganda
Nigeria
Guinea
Côte d'Ivoire
Burkina Faso
Cameroon
Kenya
Mauritania
Angola
Mali
Rwanda
Nicaragua
Senegal
Mozambique
Honduras
Vietnam
Yemen
Sao Tome and Principe
Togo
Sudan
Guinea-Bissau
Benin
Total
debt
service
(
in
percent
of
GDP),
1982-95.

 General empirical formation of a private investment
function :
(IP
/y) = H (y, cK, LP
/P, R*, IGI
, IGO
, z, z ,
,  , D*/y)
IP
/y: ratio of private investment to output;
y: income accelerator effect (captured by
changes in output);
cK: user cost of capital;
LP
/P: credit rationing (captured by the real stock
of bank credit to the private sector);
Empirical Evidence

R*: foreign exchange constraint (measured by
country's level of foreign reserves);
public investment, consists of investment in
infrastructure, IGI
, and other investments, IGO
, with
the former variable expected to have a positive
effect and the second an ambiguous effect;
z: real exchange rate, (has in general an
ambiguous effect);
macroeconomic instability, captured by the
variability of the real exchange rate, z , the level of
inflation, , and the variability of inflation, ;
D*/y: ratio of foreign debt to output.

 Analysis of the determinants of private investment
during the 1970s and 1980s in eight African
countries: four middle-income countries and four
low-income countries.
Results:
 Changes in real output (accelerator effect):
significant and positive impact on private
investment only in low-income countries.
 Public investment: positively related to private
investment in both groups; stronger
complementarity effect in middle-income countries.
Oshikoya (1994)

 Real exchange rate: positive and significant effect
in middle-income countries; negative effect in low-
income countries.
 Inflation rate: strong and unambiguously negative
impact in low-income countries; positive and
significant effect in middle-income countries.
 Debt service ratio: strong, negative effect on
private investment in both country groups.
 Macroeconomic uncertainty and instability:
(coefficients of variation of real output growth and
real exchange rate): negative effect on investment
during the 1980s.

 For Sub-Saharan Africa.
Results:
 Public and private investment are complementary.
 Lower inflation and real exchange rate variability
promote private investment.
 A high debt burden has an adverse effect on
investment.
Hadjimichael and Ghura (1995)

 Servén (1997, 1998): robust negative effect of
macroeconomic uncertainty on investment,
particularly when uncertainty is measured by the
real exchange rate.
 Aizenman and Marion (1999): negative relationship
between investment and macroeconomic volatility
measures (government consumption as a share of
output, nominal money growth and real exchange
rate).
 Figure 1.7.
Other Empirical Studies

Figure 1.7
Private Investment Share and Money Growth Volatility
Source: Aizenman and Marion (1999).
Note : The countries unidentified in the graph are Bangladesh, Guatemala, Nepal, and Sri Lanka.
1/ Money growth volatility is defined as the standard deviation of residuals from a first-order
autoregressive process for the narrow money growth rate.
0 5 10 15 20 25
0.005
0.01
0.015
0.02
0.025
0.03
0.035
Money
growth
volatility
1/
Panama
Malaysia
Thailand
Korea
Fiji
Mexico
Côte d'Ivoire
Indonesia
Pakistan
India
Nepal
Dominican Rep.
Chile
Costa Rica
Colombia
El Salvador
Tanzania
Ghana
Venezuela
Nigeria
Private investment share (in percent of GDP)
Haiti
Peru
Mali
Morocco
Iran
Egypt
Madagascar
Bolivia
Turkey
Ecuador
Uruguay
Malawi
Papua New Guinea
Philippines
Brazil
Argentina
Mauritius
Paraguay
Kenya
Zimbabwe

 Cost of waiting is available when as current return
on investment exceeds the user cost of capital and
equal to this difference.
 Value of waiting is defined by the irreversible
mistake that would be revealed tomorrow if future
returns fall short of the user cost of capital.

Chap01.ppt, consumption, saving and investment

More Related Content

Similar to Chap01.ppt, consumption, saving and investment (20)

More from SHIBABAWZEWDU (11)

Recently uploaded (20)

Chap01.ppt, consumption, saving and investment