Versión de HTML Básico
Fiscal policy and economic growth: a simulation analysis for Bolivia
Analíti a
k
4
Revista de Análisis Estadístico
Journal of Statistical Analysis
An increase in the value added tax (third column) leads
to a substantial 17.6-percent increase in transfers as a share
of GDP, but this policy does not perform well in terms of
output. Total real output increases by just 0.13 percent, the
hydrocarbons sector is the only one to grow by more than
1 percent and the importables sector even declines by 0.63
percent. Despite the endogenous increase in the lump sum
transfer, the net welfare effect is negligible (-0.02 percent).
This can largely be explained by the 0.51-percent decrease
in consumption of importables. Private investment is also
adversely affected, with a decline of 1.23 percent. As for
public investment, it increases by 3.89 percent following
the simulated increased in the value added tax.
The literature on optimal fiscal policy states that capital
taxes should be nil. Here, a reduction in the capital tax
promotes aggregate capital accumulation, which in turn
increases the amount of capital used in each sector, and
thus increases output in each sector. The sector that ben-
efits most from a reduction in capital tax is the hydrocar-
bons sector which sees a 2.47 percent increase in output.
This happens because the hydrocarbons sector is intensive
in capital (
α
larger than 0.5). Manufacturing (importables),
the other capital-intensive sector, grows by 2 percent. No-
tice that the effects of an increase or a decrease in the capi-
tal tax are almost linear. When the tax on capital increases
by 10 percent, aggregate output decreases by 1.6 percent
as opposed to an increase of 1.56 percent when the capital
tax decreases by 10 percent. Production in the hydrocar-
bons sector is highly sensitive to variations in the tax on
the production in this sector. Notice that a 10-percent de-
crease in this tax increases output in this sector by 11.61
percent, while in the opposing simulation it decreases by
10.9 percent. What is interesting to observe in this simu-
lation is that government revenues rise due to a significant
increase in output in the hydrocarbons sector. Even though
the tax rate is lowered (column 6), transfers as a share of
GDP rise by 11.96 percent, which is similar in scale to the
increase in output in the hydrocarbons sector. Consump-
tion of tradable and non-tradable goods is also positively
affected, allowing the government to collect more revenue
via the consumption tax.
18
Finally, the last three columns of Table 4 show our simu-
lation of three combined scenarios, with transfers as a share
of GDP maintained at its level in the baseline scenario. This
is to analyze the increase in consumption or value added
tax required to compensate for the negative effects of tax
policies which reduce transfers as a share of GDP.
Among the three combined scenarios, the best scenario
combines a completely open economy (
τ
m
=
0) with a 3.72-
percent increase in the value added tax. The scenario that
combines a decrease in the capital tax with an increase in
the value added tax is also good in the sense that the im-
pact on all variables is positive, albeit limited, whereas the
scenario that combines an increase in the hydrocarbons tax
with an increase in the value added tax is undesirable be-
cause all variables are negatively impacted.
From these simulations, we can conclude that, in terms
of a tax policy, in Bolivia the best it is to liberalize the econ-
omy by reducing tariffs. This policy allows the economy
to grow by 3.3 percent and to experience welfare gains of
3.86 percent. If the government also wishes to maintain its
social policy based on transfers to households, it should
finance this policy by an increase in the value added tax.
This combined policy allows the economy to grow by 3.41
percent and to experience welfare gains of 3.91 percent rel-
ative to the baseline.
4.1.2 Expenditure and investment policy
Next, we analyze a fiscal policy based solely on public
expenditures and public investment. Recall that in our set-
ting, public consumption includes everything other than
investment. This means that health and education expen-
ditures are included in this variable, as are expenditures
such as wages and benefits for public workers. First, we
simulate a 10-percent increase in public expenditures. Sec-
ond, we simulate an increase in public investment as a
share of total government revenues. We simulate an in-
crease of this share that is equal to the average increase
over 2007 and 2008. Finally, we combine a 10-percent in-
crease in public expenditures with a 10-percent increase in
public investments.
People typically expect that an increase in government
expenditures or public investments positively affects out-
put. Cross-country empirical studies have generally found
that public infrastructure positively affects a country’s out-
put (
e.g
, Easterly and Rebelo (1993), Canning and Fay
(1993), Canning (1999)). This holds true for the case of
Bolivia. A 3.26-percent increase in public investment as a
share of government revenues (the average over the last
two years) respectively increases output and public invest-
ment by 1.58 percent and 5.19 percent. However, the op-
posite happens when the government increases its current
expenditures. A 10-percent increase in government expen-
ditures leads to a 1.86-percent decline in output and a 3.56-
percent decline in consumption. An increase in govern-
ment expenditure certainly weighs on the fiscal balance,
given that transfers as a share of GDP have to be reduced
by 51.55 percent.
The combined scenario shows that transfers to house-
holds are very sensitive to an increase in government ex-
penditures: they largely decrease, whereas aggregate out-
put, consumption and investment show positive signs. In
particular, public investment increases substantially, by
11.69 percent. In the first and third columns, it can also
be seen that the real exchange rate depreciates, although
this depreciation only positively affects outputs in the ex-
portable sectors in the combined scenario. In the case
18
See Chamley (1986) and Chari, Christiano and Kehoe (1994) for references on optimal fiscal policy.
Analítika,
Revista de análisis estadístico
, 2 (2012), Vol. 4(2): 57-79
67