POLÍTICA FISCAL
Procyclicality of Fiscal Policy
in emerging countries: The cycle
is the trend
Gustavo Guelbert
FCE-UNC
While developed economies tend to run countercyclical fiscal
policies, Latin American countries have been characterized by
procyclical policies.
Esto ha sido probado empíricamente por diferentes autores
Explicaciones posibles:
1) High external debt causes severe constraints on the ability to
secure new loans, which forces countries to cut their budget deficits.
2) Another explanation is related to optimal behavior under political
constraints
This paper, tests a different channel, related to the characteristics of
business cycles.
In developing countries, the cycle is the trend: business cycles tend to
become persistent and thus to determine the fundamentals of economic
performance in these countries.
One possible channel is fiscal policy: in times of recession, the erratic
character of the crisis forces developing economies to cut expenditures,2
while the opposite occurs during booms.
1. PROCYCLICALITY OF GOVERNMENT EXPENDITURE UNDER
PERMANENT SHOCKS
Consider output and the real interest rate to be exogenous.
We take the tax rate as given and assume that government
expenditure
is endogenous. The government chooses its real expenditure, Gt, in
all periods (t = 1, 2,…) so as to maximize a utility function, with decreasing
marginal utility in government consumption:
where r is an exogenous interest rate, Y is the exogenous output level, g*
is the maximum level of government expenditure over output (G/Y), and
g is its actual level.
The intertemporal budget constraint is given by
where τ is the exogenous statutory tax rate and b0 is the ratio of
initial general government debt to output.
The first-order conditions are:
The optimal solution deriving from equation is to choose a smooth g in all
periods.
Before writing the solution, let us define the permanent value of
a variable X (with supra-index ~) as follows:
Plugging the optimal smooth value of G into the intertemporal
budget
constraint and taking the permanent value of output as defined we get
This equation states that the tax rate is set to finance the permanent level
of expenditure and the initial debt using the permanent level of output as
the tax base.
4
If there is an exogenous permanent shock on output, and given that debt
and the real interest rate are exogenous, the single way of restoring the
equality would be to adjust government expenditure.
In a recession (expansion), the equality requires cutting (rising)
expenditure; that is, it requires a procyclical fiscal policy.
This policy will be similar for both developed and emerging economies,
but the
outcome is different based on the degree of the permanent shock
and the economy’s response to it.
With regard to the degree of the shock, cycles may become persistent
in emerging markets (that is, the cycle is the trend), while they may be
purely transitory in developed economies. In this case, we would expect
fiscal policy to be acyclical (or countercyclical) in developed economies
and procyclical
in emerging markets.
With regard to the economy’s response to the shock, developed
and emerging countries may differ as a consequence of the risk
perception by economic agents.
5
To calculate the variable representing the phenomenon of the
cycle being the trend, the methodology is based on looking at the
variability of output over long horizons:
where yt = log (GDP per capita) at time t
and K is the amount of lagged differences.
Then correct the sample variance for small-sample bias by including a
degree-of-freedom correction term, T / (T – K + 1)
For each K, we calculate
where σ2Δ y is the value of σ2Δ K when K = 1. Thus, for all countries,
the
value of equation at K = 1 is one.
6
This value gives the ratio between the long-term variability of output
and the short-term variability, thereby providing a measure of the
extent to which the cycle is the trend.
The higher this coefficient, the more strongly countries are expected to be
affected
by changes in output.
7
2. DATA AND EMPIRICAL RESULTS
2.1 Empirical Specifications for Total Expenditure
We estimate the following types of regression:
where G is real government expenditure, CITT is “the cycle is the trend” variable,
dlog(POP) is the population growth rate; POP15 and POP65 are
the populations under 15 and over 65 years old, K is the average of the random walk
component, as explained above; RATIO refers to the ratio between the country’s GDP
per capita and the GDP per capita of the United States, both in purchasing power parity
(PPP) values; EMERGING is a dummy variable that equals one for emerging countries
and zero otherwise; and HyperInfl is a dummy variable that equals one when yearly
inflation is over 100% for two or more consecutive
years and zero otherwise 8
2.1.1 Budget Cuts
Cutting the budget in hard times is particularly painful, since it has a
negative impact on economic activity. Thus, it is interesting to study the
procyclicality of fiscal policy in recession periods. Table 1 summarizes
the number of budget cuts and whether they followed a recession
period (which would indicate procyclical behavior), the amount of
persistent budget cuts, and the depth of the budget cuts.
2.1.2 A Change in policy after the 1990s
The globalization of the 1990s exposed emerging countries to
international markets to an unprecedented degree.
In countries that are in trouble but in which the governments succeed in
convincing foreign investors that the changes being made in the economy
will bring a relatively quick end to the badtimes, foreign investors will
perceive low stock exchange levels as an investment opportunity.
This may provide a new mechanism for a recovery: expectations may
change quickly, output may reverse, and governments will be less
dependent on performing budget cuts during recessions; that is, the
procyclicality of fiscal
policy would decline.
We define a dummy variable, D(90) that takes the value of one after
1990 and zero otherwise.
10
2.2 Empirical Results for Total Government
Expenditure
11
The coefficient of permanent shocks is insignificant for the developed
economies, whereas emerging markets have a coefficients of around 0.3
and statistically significant at the 1 percent level (column 2).
The coefficient rises during high times (column 4): in periods of parallel
reductions in G and Y, the coefficient increases by 0.2. We take these
results as the first evidence of our main hypothesis—namely, that GDP
shocks in emerging countries are associated with a procyclical reaction
in government expenditure. These results are confirmed using the other
methods.
Columns 3 and 5 present our results on whether there was a change in
expenditure behavior by emerging government after the 1990s, in
response to increasing globalization. In column 3, the coefficient is not
significant, but in column 5, it is significant at the 5 percent level.
This indicates that government expenditure was significantly less
procyclical in emerging countries after the 1990s, especially during hard
times.
12
2.3 Goverment Expenditure Composition (mirando los
componentes del gasto público)
TABLA 3: Mientras que el gasto en consumo público es
procíclico tanto en países desarrollados como en países en
desarrollo, es más procíclico en los países emergentes. Ello no
cambia significativamente en tiempos difíciles.
TABLA 4: Las transferencia del gobierno son procíclicas en
economías emergentes, patrón que es acentuado luego de los
90´s
TABLA 5: El gasto en capital tiende a ser procíclico incluso en
economía desarrolladas.
Posible explicación: Como los resultados de las inversiones
en capital se ven luego de varios años de producidas, los
políticos se ven tentados de iniciar obras públicas solamente en
13
períodos de abundancia, puesto que los resultados no se ven en
el corto plazo.
14
15
16
2.4 Other Issues to Consider
2.4.1 Foreing direct investment and international reserves
Una posible explicación a la mejor performance luego de los 90´s es que
los países ya no dependen tanto de sí mismos. La mayor exposición
producto de la globalización a inversores mundiales, suaviza las
opciones financieras de los países, por lo que ya no deben recortar sus
gastos tan profundamente en tiempos difíciles.
Una forma de medirlo es con la IED, pues se esperaría una prociclicidad
má suavizada
In table 7, we explore whether the FDI level has some explanatory
power for procyclical fiscal policies in emerging countries. For this
purpose, we performed two regressions, one using an interaction
variable between CITT and FDI (column 1) and another using an
interaction between CITT and a dummy variable, D−(FDI), that takes
the value of one when FDI is higher than the median for each group of
countries in each decade or 0 otherwise (column 2). Results are
significant and in the expected direction; for emerging countries with
high levels of FDI the coefficient of procyclicality decreases from 0.42
to 0.14 (column 2).
Kandil and Morsy (2010) find that international reserves help for
performing countercyclical policy in emerging countries. We use their
methodology for testing the role of international reserves and build a
dummy variable, D(Reserves), that takes the value 1 if the
international reserves at the end of the year are higher than the sum
of 3 months of imports (using average monthly imports of the
corresponding year).
Columns 3 and 4 show that the coefficients have the expected sign
and are significant at 10 percent.
3. SUMMARY AND CONCLUSIONES
This paper explores whether developed and emerging economies react
differently to persistent shocks to output. From a theoretical perspective,
we expected to find procyclical behavior in countries that are subject to
persistent shocks to per capita GDP—that is, they will increase
expenditure during booms and cut it during recessions.
We found that while government expenditure in developed economies is
not affected by these shocks (with the exception of government
investment), emerging countries do tend to pursue procyclical fiscal
policy in reaction to persistent shocks to per capita GDP. This is in line
with previous findings for investment, which show that both developed
and emerging countries act procyclically in this area.
However, procyclical policy in emerging countries is particularly evident
for total expenditure and is implemented in government investment,
consumption and transfers.
There are signs of a reduction in the extent of procyclical expenditure
policy in emerging countries after the 1990s, in response to increasing
globalization. Moreover, we found that countries with a high level of FDI
implemented milder procyclical policies, as did those included in
emerging markets indexes.
