Versión de HTML Básico
Carlos Gustavo Machicado y Paúl Estrada
Analíti a
k
4
Revista de Análisis Estadístico
Journal of Statistical Analysis
justment and conclude that fiscal stimuli through tax cuts
are more likely to increase growth than those based upon
spending increases.
1
A recent revival of this literature in developed coun-
tries, particularly in the United States, has been stoked by
the 2007-2009 financial crisis and the fiscal policy responses
that have been the basis of many of the recovery policies.
Feldstein (2009) indicates that, despite a recent general con-
sensus among economists that fiscal policy was not an ef-
fective countercyclical instrument, governments in Wash-
ington and around the world are now developing mas-
sive fiscal stimulus packages, supported by a wide range
of economists in universities, governments and businesses.
There has been a revival of the use of so-called fiscal policy
multipliers.
2
In Latin America and other developing countries, re-
cent literature has mainly sought to verify the idea that fis-
cal policy is procyclical, a puzzle that has sparked a grow-
ing theoretical literature in an effort to explain this ten-
dency. Gavin and Perotti (1997) were the first to draw
attention to the fact, while Talvi and Vegh (2005) claimed
that procyclical fiscal policy seemed to be the general rule
across the developing world. Recently, Ilzetzki and Vegh
(2008) find overwhelming evidence, using a battery of
econometric tests, to support the idea that fiscal policy in
developing countries is in fact procyclical.
A common feature in this literature is the focus on
tax and expenditure policies to the exclusion of analysis
of public investment policies, particularly those which in-
volve public infrastructure investments. As shown by As-
chauer (1989a, 1989b), infrastructure is an important source
of growth. These works concentrated on estimating the
production elasticities of government expenditures using
aggregated data, mainly for the United States
3
. Cross-
country studies have also highlighted the role of infrastruc-
ture for a country’s growth.
4
Papers in this literature have typically used regression
analysis on either growth accounting or on steady-state
equations. While these papers have been useful in pointing
out the importance of infrastructure, their methodologies
do not allow for analysis of important general equilibrium
feedback effects among key macroeconomic variables and
welfare.
It is in this context that this study examines the impact
of fiscal policy on output, consumption, investment and
foreign trade using a dynamic stochastic general equilib-
rium (DSGE) model for a small open economy with five
sectors and with the novel feature that firms in each sec-
tor use public capital or infrastructure as a production fac-
tor. These five sectors are the non-tradable sector (services),
the importable sector (manufacturing), hydrocarbons, min-
ing and agriculture, each of which are representative at the
level of the Bolivian economy. Among these sectors, the
government views the capital-intensive hydrocarbon sec-
tor as a strategic sector that will generate the resources
needed to combat poverty and underdevelopment.
In this study, we first analyze the macroeconomic and
sectoral impacts of changes in fiscal policy such as tax
structure and public infrastructure investments on: output,
consumption, investment, the trade balance and welfare.
Second, we identify the combination of fiscal policy instru-
ments that allows the government to sustain public social
transfers to households. Third we show that the fiscal pol-
icy alone is not sufficient to generate the output growth and
welfare gains needed to reduce poverty levels, as per the
Millennium Development Goals, which indicate that GDP
per capita should grow by more than 2 percent per year.
This is equivalent to an overall GDP growth rate of more
than 6 percent per year. A combination of effective provi-
sions of public capital and increased total factor productiv-
ity (TFP) is also needed. We provide the long-run results
for each simulation along with dynamic transitions for a
handful of selected cases.
The DSGE model is based on Chumacero, Fuentes, &
Schmidt-Hebbel (2004) and is modified to include pub-
lic infrastructure investments much like Rioja (2003) and
specifically includes different exportable sectors as seen
in Estrada (2006). We calibrated the model for the Bo-
livian economy and solved it using the second-order-
approximation technique developed by Schmitt-Grohé and
Uribe (2004). The advantage of this perturbation method is
that it allows for second-order effects, which feature heav-
ily in an economy with high levels of uncertainty.
An important aspect of the model is that it allows us
to derive precise quantitative information about the effects
that each scenario has on real output and welfare as well
as macroeconomic variables such as consumption, invest-
ment and output in the five sectors. The section with
the model simulation first reports the steady-state (long
run) effects then presents the dynamic effects on the com-
position of these variables. This is important if we con-
sider that, in recent years, the Bolivian government’s anti-
poverty policy has been based on transfers to households,
while it aims to use public investment as the main ap-
proach to promoting growth and welfare. Quantitative
measures of the impact of these policies are thus needed
to guide policymakers.
The paper is organized as follows: Section 2 briefly de-
scribes fiscal policy in Bolivia in recent years. Section 3
describes the dynamic general equilibrium model and its
1
There is a rich literature on the determinants and economic outcomes of large fiscal adjustments. A non-exhaustivelist includes Ardagna (2004),
Giavazzi, Jappelli and Pagano (2000), Lambertini and Tavares (2000), McDermott and Wescott (1996), Von Hagen and Strauch (2001), Von Hagen,
Hughes, and Strauch (2002), among others.
2
See Mauntford and Uhlig (2008), Alesina and Ardagna (2009), Cogan et.al. (2009), Ramey (2009), and Romer and Romer (2010), among others.
3
Munnell (1990) and García-Milá, McGuire and Porter (1993) use panel data to estimate production elasticities.
4
Easterly and Rebelo (1993), Ford and Poret (1991), Hulten (1996) and Canning (1998) among others.
58
Analítika,
Revista de análisis estadístico
, 2 (2012), Vol. 4(2): 57-79