Fiscal decentralisation, democratic institutions and inflation.
Iqbal, Nasir
This study examines the role of democratic institutions in an
attempt to explain the relationship between fiscal decentralisation and
inflation. The empirical analysis is based on time series data over
1972-2010 for Pakistan using the GMM estimation procedure. Three
different measures of fiscal decentralisation are used in order to
capture multidimensionality. The major findings of the study suggest
that expenditure decentralisation has a negative impact on inflation if
accompanied by democratic institutions. Revenue decentralisation,
however, has a negative impact on inflation even in the absence of
institutions, though institutions accentuate this effect. The role of
institutions, therefore, is important in realising the benefits of
fiscal decentralisation. Composite decentralisation has a negative and
significant impact on inflation. This implies that expenditure
decentralisation becomes effective when it is complemented with revenue
decentralisation. Intuitively, provincial governments become more
responsive when their expenditure needs are met with their own revenues.
JEL Classification: E31, HI1, H72
Keywords: Fiscal Decentralisation, Inflation, Institutions,
Pakistan
1. INTRODUCTION
Fiscal decentralisation (FD) is considered an effective strategy to
promote economic growth through controlling inflation [Martinez-Vazquez
and McNab (2003)]. The empirical literature, however, on the FD and
inflation nexus is scant and inconclusive. For example, King and Ma
(2001), using cross section data for 49 countries during the period
1973-1994, find a negative relationship between revenue decentralisation
(RD) and inflation especially for developed countries. Neyapti (2004)
extends this analysis by arguing that FD leads to lower inflation
provided monetary discipline exists and not necessarily otherwise. The
study, using a panel dataset for developed and developing countries,
concludes that RD is effective in easing inflation only if it is
accompanied by central bank's independence and local
accountability. Martinez-Vazquez and McNab (2006) show that expenditure
decentralisation (ED) decreases inflation only in developed economies
but not in developing countries. Thornton (2007) finds that if the RD
measure is restricted only to revenues over which local governments have
full autonomy, the impact of RD on inflation is negligible. Jalal, el
al. (2012) find that fiscal decentralisation appears to lead to a
decrease in inflation rate.
It is quite evident that further research is needed to explore
links between FD and inflation. The inconclusiveness of existing studies
can be attributed to the missing role of institutions. We can deduce
from existing literature that FD reduces inflation in developed
economies even in the absence of institutions, while in developing
economies, FD reduces inflation when it is supported by local government
accountability. The theorem of decentralisation (1) implicitly assumes
that the positive contribution of FD is linked with the inclusive
institutions that ensure accountability and transparency of governments
and public officials [Oates (1993)]. Recent advancement in the field of
FD makes this assumption explicit and incorporates the role of
institutions in the theorem of FD. We argue that FD and democratic
institutions reinforce and complement each other in determining
inflation. Absence of local accountability, lack of institutional and
administrative capacity and coordination problems are major factors that
make FD less effective in controlling inflation especially in developing
countries [Martinez-Vazquez and McNab (2006)]. Local autonomy in
collecting local revenues may be limited due to political considerations
[Neyapti (2004)]. Jalal, et al. (2012) find that the impact of
decentralisation on inflation is determined by the level of perceived
corruption and political institutions. We, therefore, hypothesise that
FD leads to lower inflation provided an appropriate institutional
framework exists, not necessarily otherwise. The objective of this study
is two-fold; first, to examine the impact of fiscal decentralisation on
inflation in Pakistan using time series data over the period 1972-2010;
second, to test the hypothesis that fiscal decentralisation leads to
lower inflation provided that a supportive institutional framework
exists, not necessarily otherwise.
The GMM approach is used for estimation due to possibility of
reverse causality and endogeneity among fiscal measures. This study
contributes to literature in multiple ways: First, to the best of the
author's knowledge, no study to date has investigated the
relationship between fiscal decentralisation and inflation for Pakistan.
Second, this study employs GMM approach to estimation to control reverse
causality and endogeneity. Third, this study quantifies the role of
democratic institutions in explaining the fiscal
decentralisation-inflation nexus. Fourth, this study provides policy
recommendations which would help the policymakers in formulating better
economic policies for long run macroeconomic stability. It also informs
the policy-makers and practitioners about the strengths and weaknesses
of the process of fiscal decentralisation in Pakistan. In addition to
its relevance to policy-makers and practitioners, it also adds to the
academic discussion on the impact of fiscal decentralisation.
The rest of this paper is structured as follows: the conceptual
framework is explained in Section 2; the data, methodology and
econometric issues are explained in Section 3; Section 4 presents the
results of this study and Section 5 concludes the discussion and
provides some key policy implications.
2. CONCEPTUAL FRAMEWORK
According to Martinez-Vazquez and McNab (2003), fiscal
decentralisation promotes economic growth indirectly through maintaining
macroeconomic stability. (2) The literature on fiscal decentralisation
mostly uses price stability as a proxy for macroeconomic stability. (3)
There are controversies in literature as to whether fiscal
decentralisation promotes or impedes macroeconomic stability.
A number of authors have suggested that devolution of some
macroeconomic management policy measures to sub-national governments can
promote macroeconomic stability, not hinder it [Shah (1999); Rodden and
Wibbels (2002)]. Shah (2006) argues that fiscal decentralisation is
linked with enhanced fiscal and economic performance because a
decentralised fiscal setup provides a greater potential for the
development of macroeconomic governance than a centralised fiscal setup.
Public spending under a decentralised setup increases the economic
efficiency because sub-national governments have more precise
information about the preferences of the local communities that permits
non-uniform provisions of public goods and services in accordance with
the preferences of local citizens [Oates (1993)]. The process of
decentralisation is also associated with more accountability and
transparency in public service delivery [De Mello (2000)]. Existence of
local accountability leads to more responsible behaviour of tax-payers
that ultimately improves the effectiveness of local government
[Wasylenko (1987)]. This implies that decentralisation may lead to
macroeconomic stability via increased public sector efficiency [Neyapti
(2010)].
The fiscal decentralisation can exert positive impact on price
stability through the independence of the central bank. The existing
studies show that the credibility of the commitment to price stability
can be established if the monetary authority adheres to a set of formal
rules or if there is a guarantee that it is independent from pressures
from all levels of government [Shah (1994); Barro (1996)]. Shah (2005)
argues that the central bank under a decentralised system performs
better. Neyapti (2004) also argues that decentralisation and central
bank independence reinforce each other in controlling inflation. Revenue
decentralisation leads to lower inflation if it is accompanied by both
central bank independence and local accountability.
Another theory of decentralisation suggests that the process of
fiscal decentralisation does not affect the inflation directly, but it
keeps inflation rates constant, whether low or high, through making it
difficult to change fiscal or monetary policies [Tsebelis (1995)]. The
number of agents whose agreement is required for changing a policy is
increased in a federal structure. The sub-national governments are
sometimes provided with the right to veto the decisions made by the
central government. This in turn reduces the probability of changing
policy hence ensures continuity in the existing monetary and fiscal
policies which ultimately makes inflation rates constant. The final
macroeconomic outcomes, therefore, depend on policies which are
initially in place.
In countries where inflation rates are high, a decentralisation
process tends to perpetuate the underlying factors that cause high
inflation and hence make it difficult to achieve durable stabilisation.
On the other hand, in countries where inflation rates are low due to low
fiscal pressure and depoliticised monetary policy, a decentralisation
process further promotes stability via maintaining the inflation rate at
a low level. (4)
Various studies argue that fiscal decentralisation per se increases
macroeconomic instability or works as an obstacle to solving the
persistent fiscal imbalance due to potential disregard of budget
constraints by local governments under a decentralised framework [Rodden
(2002)]. However, when macroeconomic instability predates
decentralisation, it is much more difficult to achieve macroeconomic
stability although not entirely impossible [Dillinger, et al. (2000)].
The possibility of soft budget constraint5 at the sub-national levels of
governments also makes it difficult to achieve macroeconomic stability
through decentralisation [Stein (1999); Bahl (1999)].
fiscal decentralisation may also have adverse consequences for
macroeconomic stability because decentralisation may be associated with
an increase in the degree of autonomy of the local governments. Ahmad,
et al. (2005) argue that macroeconomic stability or price stability for
an economy depends on the overall exposure to the risk. In this
situation, the critical element is the borrowing of all jurisdictions in
the country. Local governments have more authority to determine level of
expenses as well as to collect revenues in their jurisdictions under
decentralised set up. Hence, the central government has less control to
manage the fiscal activities of local governments which ultimately leads
to more macroeconomic instability.
Most of the criticism against decentralisation does not dismiss the
idea of decentralisation per se, but is rather meant to highlight the
need for augmenting the decentralisation process with sound
institutions. According to the critics, only when these institutions are
present does decentralisation bear the fruits that are promised by its
proponents. The benefits of decentralisation largely depend on
institutional arrangements that govern the design and implementation of
decentralisation [Iqbal, et al. (2012)]. A well-defined institutional
mechanism increases the accountability and transparency in the political
system and hence helps to reduce corruption. Leading to efficient
allocation of public resources [Iqbal, et al. (2012)]. Enikolopov and
Zhuravskaya (2007) argue that the success of fiscal decentralisation
depends on the quality of the political institutions in the country.
This study shows that fiscal decentralisation is more successful in
politically decentralised transition economies. They argue that the
positive contribution of fiscal decentralisation will be attenuated if
the country is plagued with a serious problem of corruption. On the
other hand, a country which is free from corruption will be able to reap
the benefits of fiscal decentralisation through macroeconomic stability.
More stable political system may accentuate the impact of fiscal
decentralisation on macroeconomic stability and vice versa [Enikolopov
and Zhuravskaya (2007)].
3. DATA, METHODOLOGY, AND ECONOMETRIC ISSUES
The empirical analysis is based on time series data from 1972 to
2010 for Pakistan. FD, the subject matter of this study, refers to the
devolution of policy responsibilities for public spending and revenue
collection from the central to the provincial governments. Data on
fiscal decentralisation variables is sourced from the Fifty Year Economy
of Pakistan and various annual reports published by the State Bank of
Pakistan. Three different indicators are constructed to measure the
level of FD.
* Revenue Decentralisation (RD): RD is measured as the ratio of the
provincial government's revenues (PR) to the total government
revenues (TR) (federal plus provincial) i.e. RD = PR/TR.
* Expenditure Decentralisation (ED): ED is defined as the ratio of
provincial government expenditures (PE) to the total government
expenditures (TE) (federal plus provincials) excluding the defence
expenditures (DE) and interest payments on debt (IE) since these
expenditures are mainly considered as part of the non-decentralised
government expenditures. ED = PE/TR - (DE + IE).
* Composite decentralisation (CD): Following Martinez-Vazquez and
Timofeev (2010), CD is measured by using both RD and ED. CD captures the
information in expenditure and revenue ratios. CD = RD/1-ED.
Figure 1 shows the trends in RD, ED and CD in Pakistan. Figure 1
shows that the share of provincial government revenue in total
government revenue ranges from 7 to 25 percent. The share of provincial
governments' revenue is 15 percent in total government revenue in
1980, thereafter showing an increasing trend to reach 23 percent in
1987. After this period, there is a decreasing trend in revenue
decentralisation and provincial revenue share in total government
revenue reaches 10 percent in 2010. Figure 1 shows that the share of
provincial government expenditure in total government expenditure ranges
from 34 to 69 percent during the last three decades. After reaching 50
percent in 1982, the share of provincial government expenditure shows a
decreasing trend reaching 39 percent in 1989. For most part of the
1990s, expenditure decentralisation shows an increasing trend. However,
after 1998 once again, provincial shares in total expenditures trend
downwards, declining from 55 percent in 1998 to 35 percent in 2010. The
trend shows that the 'Composite Decentralisation' measure
ranges from 14 to 49 percent.
[FIGURE 1 OMITTED]
Inflation is measured as the growth rate of consumer price index
(CPI). The average inflation rate is 9.6 varying from 3.1 percent to 30
percent. Figure 2 shows the trend of inflation rate over the period
1992-2010. Figure shows that inflation rate touches the peak of 26.7
percent in 1974. After that inflation rate has declined from 26.7 in
1974 to 11.9 in 1980. During the decade of 1980s, the average inflation
rate was around 7 percent. While during the decade of 1990s, the average
inflation rate was around 10 percent. Inflation rate shows increasing
trend after 2004 and once again touches the peak of 20 percent in 2008.
[FIGURE 2 OMITTED]
The data on democracy is sourced from the Polity IV dataset. The
democracy index ranges from +10 (full democracy) to -10 (full
autocracy). The democracy index in the below Table indicates that the
average quality of the institutional framework is 0.85 ranging between
-7 to +8 in Pakistan. The control variables consist of physical capital
measured as capital stock per work, money supply measured as M2 to GDP
ratio, openness measured as trade as percent of GDP and tax to GDP
ratio. Data on these variables is taken from the Economic Survey of
Pakistan (various editions). The descriptive statistics are presented in
Table 1.
Following Neyapti (2004, 2010) the following model is proposed
which captures the links among FD, democratic institutions and
inflation:
[INF.sub.t] = [[delta].sub.0] + [[delta].sub.1] [FD.sub.t] +
[[delta].sub.2][INS.sub.t] + +[[delta].sub.3][FD.sub.t] * [INS.sub.t] +
[delta][X.sup.'.sub.t] + [[epsilon].sub.t]
Where INF is inflation rate, FD measures fiscal decentralisation,
INS represents democratic institutions, X is the vector of control
variables and e is the disturbance term. In this model, the interaction
term, FD * INS allows us to test the hypothesis of complementarity
between FD and democratic institutions. Based on this model, we aim to
empirically examine the following hypotheses:
(i) Fiscal decentralisation influences the inflation rate.
(ii) Fiscal decentralisation and democratic institutions are
complementary.
There are several studies that have used the Ordinary Least Squares
(OLS) estimation technique to empirically investigate the impact of FD.
However, a number of studies identify the possibility of reverse
causality and endogeneity among FD and other variables [see e.g. Zhang
and Zou (1998); Xie, el al. (1999); Lin and Liu (2000); Thiessen (2003);
Jin, el al. (2005); Iqbal, et al. (2012)]. OLS estimates become biased
and inconsistent in the presence of reverse causality and endogeneity.
Weuse instrumental variables approach based on the generalised method of
moments (GMM) to check endogeneity. The application of the generalised
method of moments (GMM) can be considered as an extension of the
instrumental variables (IV) estimation method. The main advantage of the
GMM estimation method is that the model need not be serially independent
and homoscedastic. Another benefit of the GMM estimation technique is
that it generates parameters through maximising the objective function,
which includes the moment restrictions in which correlation between the
lagged regressor and the error tenn is zero.
The standard approach to determine the stationarity of the time
series data is to check the existence of unit roots in the given series.
The most commonly employed test for unit root analysis is called
Augmented Dickey Fuller (ADF) test. We have used ADF tests to determine
the stationarity of series. The results of the ADF test are reported in
Table 2. The test statistics indicate that inflation, openness and M2 to
GDP ratio are stationary at level. While revenue decentralisation,
expenditure decentralisation, composite decentralisation, capital stock
per worker, tax to GDP ratio, democratic institutions and interaction
terms are non-stationary at level and become stationary at first
difference.
4. RESULTS AND DISCUSSION
This study has estimated the impact of various forms of FD on
inflation. The Table 3 below shows the results. We observe a negative
and significant impact of RD on inflation implying that an increase in
RD leads to a lower level of inflation. RD helps to promote stability in
many ways. First, the higher the level of provincially owned revenues,
the less will be the dependence of provincial government on federal
revenue. Subsequently, the federal government can enhance its own
capacity by allocating more resources to public sector projects such as
power and infrastructure. Secondly, it provides more policy space to
central bank in controlling inflation. Shah (2005) argues that with
fiscal decentralisation the central bank will be more independent since
a decentralised system requires more clarified rules and regulations
under which a central bank operates. Also the central bank has more
flexibility to interact with different levels of government. The
combined effects of all these factors result in low inflation through
RD. The results show that ED has a negative but insignificant impact on
inflation. Martinez-Vazquez and McNab (2006) find similar results for
developing countries. The insignificant association between ED and
inflation may be due to weak institutional framework of the country.
Lack of economies of scale, absence of local accountability, lack of
institutional and administrative capacity and coordination problems are
the major factors that make expenditure decentralisation less effective
in controlling the inflation rate [Martinez-Vazquez and McNab (2006)].
Results also show that CD has a negative and significant impact on
inflation. This implies that ED becomes effective when it is
complemented with RD. Intuitively, provincial governments become more
responsive when their expenditure needs are met with their own revenues.
Various control variables are used in the analysis. Physical
capital has a negative and significant impact on inflation. This implies
increased investment in capital stock is associated with a decrease in
inflation rate. Investment in public infrastructure helps in many ways
to promote macroeconomic stability. For example, investment in roads,
electricity and other public amenities reduces the structural
bottlenecks hence reduces the macroeconomic instability. It is evident
that Pakistan has been facing supply-side constraints for the last few
years. These constraints impede the growth process and reduce
macroeconomic stability of the country. In this situation, public
investment for capacity building especially in the power sector is
required to maintain stability and growth. Private sector investment
typically reduces production costs hence relieving inflationary
pressure.
The money supply has a negative but insignificant relationship with
the rate of inflation. This indicates inflation is not primarily because
of money supply but it may be structural in nature and mainly attributed
to supply-side factors. Nasir and Malik (2011) also argue that inflation
in Pakistan is mainly supply side driven. Trade openness has a negative
and statistically significant impact on inflation. This result is in
line with Romer' view (1993), that inflation is lower in small open
economies. A number of other studies also show that trade openness is
negatively associated with inflation rate in Pakistan [Ashra (2002);
Gruben and McLeod (2004); Kim and Beladi (2005); Hanif and Batool
(2006); Mukhtar (2010)]. Openness enhances the efficiency and reduces
costs through change in composition of inputs procured internationally
and domestically, thus leading to lower inflation. Openness also affects
inflation through better allocation of resources and increased capacity
utilisation. Openness may also boost foreign investment which can
stimulate output and reduce the price level [Ashra (2002)]. Tax to GDP
ratio has a negative and statistically significant impact on inflation,
implying that higher the tax to GDP ratio; lower the level of inflation
in the country. Taxation generally reduces the level of income and with
lower level of income; demand for goods and services will decline that
will eventually lead to lower inflation.
To examine the role of democratic institutions, interactive terms
of democratic institutions and FD are added as additional explanatory
variables i.e. RD * INS, ED * INS and CD * INS. The results show that FD
becomes effective when interacted with democratic institutions (Table
4). All interactive terms have negative and significant impact on
inflation. This implies that FD and democratic institutions reinforce
each other.
Brambor, et al. (2006) and Iqbal, et al. (2012) show that it is
incorrect to include the interactive term simply due to the significance
of the coefficient of the interactive variable. The marginal effect of
FD on inflation should be observed by constructing confidence intervals
for the estimates of coefficients of FD and interactive term of FD and
INS. Figure 3 below is drawn on the basis of the coefficient estimates
of FD and also its interactive term with democratic institutions and
their variance-covariance terms. Figure 3 shows as the quality of
institutions improves, RD and CD exert increasingly negative and
significant impact on inflation. The impact of fiscal decentralisation
on inflation is very low when the quality of institutions is poor.
However, as the quality of institutions improves, the fiscal
decentralisation exerts a significant negative impact on inflation. The
institutional school of thought argues that the quality of institutions
increases the efficiency of the economic factors of production [North
(1981)]. It reduces the level of corruption and enhances the
accountability of the governments which lead to more stable
macroeconomic environment, including lower inflation.
[FIGURE 3 OMITTED]
5. CONCLUSION
In this study, the role of democratic institutions in modulating
the fiscal decentralisation-inflation nexus has been analysed using time
series data over the period 1972-2010. We have used the GMM estimation
procedure to estimate the model. The empirical analysis has shown that
revenue decentralisation has a significant and negative impact on
inflation rate while expenditure decentralisation has an insignificant
association with inflation. The expenditure decentralisation fails to
check inflation rate due to weak institutional framework of Pakistan
that leads to more corruption and less accountability when resources
through fiscal transfer are easily available to the provincial
governments. Composite decentralisation also has a negative association
with inflation. This implies that if Pakistan focuses simultaneously on
both types of fiscal decentralisation, then it helps in promoting
macroeconomic-stability in Pakistan. Further analysis has shown that
fiscal decentralisation becomes effective in controlling inflation when
complemented with democratic institutions. It is observed that
improvement in the quality of democratic institutions enhances the
ability of fiscal decentralisation to exert a moderating effect on
inflation.
The crux of the analysis is that institutions are indeed important
in realising the benefits of fiscal decentralisation. Strengthening
democratic institutions is a pre-requisite for achieving the goals of
fiscal decentralisation. Well defined and sound democratic institutions
make provincial as well as central governments accountable and
transparent in performance of their functions while remaining within
their jurisdictions.
REFERENCES
Ahmad, J., S. Devarajan, S. Khemani, and S. Shah (2005)
Decentralisation and Service Delivery. Washington, D. C. World Bank.
(World Bank Policy Research Working Paper Series No. 3603).
Arellano, M. and O. Bover (1995) Another Look at the Instrumental
Variables Estimation of Error-components Models. The Journal of
Econometrics 68, 29-51.
Arellano, M. and S. Bond (1991) Some Tests of Specification for
Panel Data: Monte Carlo Evidence and an Application to Employment
Equations. Review of Economic Studies 58, 277-297.
Ashra, S. (2002) Inflation and Openness: A Case Study of Selected
Developing Economies. Indian Council of Research on International
Economic Relations (ICRIER). (Working Paper No. 84).
Bahl, R. (1999) Fiscal Policy in China: Taxation and
Intergovernmental Fiscal Relations. The 1990 Institute.
Barro, R. J. (1996) Getting it Right: Markets and Choices in Free
Society. Cambridge, Massachusetts: MIT Press.
Brambor, T., W. R. Clark, and M. Golder (2006) Understanding
Interaction Models: Improving Empirical Analysis. Political Analysis 14,
63-82.
De Mello, L. R. (2000) Fiscal Decentralisation and
Intergovernmental Fiscal Relations: A Cross-Country Analysis. World
Development 28:2, 365-380.
Dillinger, W., G. Perry, and S. B. Webb (2000) Macroeconomic
Management in Decentralisation Democracies: The Quest for Hard Budget
Constraints in Latin America. In S. J. Burki and G. E. Perry (eds.)
Annual World Bank Conference on Development in Latin America and the
Caribbean, 1999: Decentralisation and Accountability of Public Sector.
Proceedings of a Conference held in Valdivia, Chile, 1999, Washington,
D. C. World Bank. 85-105.
Enikolopov and Zhuravskaya (2007) Decentralisation and Political
Institutions. Journal of Public Economics 91:11-12, 2261 -2290.
Gruben, W. C. and D. McLeod (2004) The Openness-Inflation Puzzle
Revisited. Applied Economics Letters 11:8, 465-468.
Hanif, M. N. and I. Batool (2006) Openness and Inflation: A Case
Study of Pakistan. Pakistan Business Review 7:4.
Iqbal, N., M. Din, and E. Ghani (2012) Fiscal Decentralisation and
Economic Growth: Role of Democratic Institutions. The Pakistan
Development Review 52:3, 176-196.
Jalal, A., M. Harun, and S. Mat (2012) Macroeconomic Instability
and Fiscal Decentralisation: An Empirical Analysis. Prague Economic
Papers 2, 150-165.
Jin, IT, Y. Quian, and B. Weingast (2005) Regional Decentralisation
and Fiscal Incentives: Federalism, Chinese Style. Journal of Public
Economics 89:9-10, 1719-1742.
Kim, M. and M. Beladi (2005) Is Free Trade Deflationary? Economics
Letters 89:3, 343-349.
King, D. and Y. Ma (2001) Fiscal Decentralisation, Central Bank
Independence and Inflation. Economics Letters 12, 95-98.
Komai, J. (1979) Resource-constrained versus Demand-constrained
System. Econometrica 47:4, 801-819. Lin, J. Y. and Z. Liu (2000) Fiscal
Decentralisation and Economic Growth in China. Economic Development and
Cultural Change 49:1, 1-21.
Martinez-Vazquez, J. and R. M. McNab (2003) Fiscal Decentralisation
and Economic Growth. World Development 39, 1597-1661.
Martinez-Vazquez, J. and R. M. McNab (2006) Fiscal
Decentralisation, Macrostability and Growth. Hacienda Publica
Espanola/Revista de Economia Publica 179:4, 25-49.
Martinez-Vazquez, J. and A. Timofeev (2010) Decentralisation
Measures Revisited. International Studies Programme, Andrew Yong School
of Policy Studies. (Working Paper 09-13).
Mukhtar, T. (2010) Does Trade Openness Reduce Inflation? Empirical
Evidence from Pakistan. The Lahore Journal of Economics 15:2, 35-50.
Nasir, M. and W. S. Malik (2011) The Contemporaneous Correlation of
Structural Shocks and Inflation--Output Variability in Pakistan. The
Pakistan Development Review 50, 145-162.
Neyapti, B. (2004) Fiscal Decentralisation, Central Bank
Independence and Inflation: A Panel Investigation. Economics Letters 82,
227-230.
Neyapti, B. (2010) Fiscal Decentralisation and Deficit:
International Evidence. European Journal of Political Economy 26,
155-166.
North, D. C. (1981) Structure and Change in Economic History. New
York: Norton and Co.
Oates, W. E. (1993) Fiscal Decentralisation and Economic
Development. National Tax Journal 46:2, 237-243.
Riker, W. (1987) The Development of American Federalism, Boston,
M.A.: Kluwer.
Rodden, J. (2002) The Dilemma of Fiscal Federalism: Grants and
Fiscal Performance around the World. American Journal of Political
Science 46:3, 670-687.
Rodden, J. and E. Wibbels (2002) Beyond the Fiction of Federalism:
Macroeconomic Management in Multitiered Systems. World Politics 54,
494-531.
Romer, D. (1993) Openness and Inflation: Theory and Evidence.
Quarterly Journal of Economics 108, 869-903.
Shah, A. (1994) The Reform of Intergovernmental Fiscal Relations.
Washington, D. C. (World Bank Policy Research Working Paper Series No.
726).
Shah, A. (1999) Fiscal Federalism and Macroeconomic Governance: For
Bette for Worse. In K. Fukasaku and R. Jr, De Mello Luiz (eds.) Fiscal
Decentralisation in Emerging Economies: Governance Issues. Washington,
D. C. Organisation for Economic Co-operation and Development. 37-54.
Shah, A. (2005) Fiscal Federalism and Fiscal Performance.
Washington, D. C. (World Bank Policy Research Working Paper Series No.
3786).
Shah, A. (2006) Fiscal Decentralisation and Macroeconomic
Management. International Tax and Public Finance 13:4, 437-462.
Stein, E. (1999) Fiscal Decentralisation and Government Size in
Latin America. Journal of Applied Economics 2:2, 357-391.
Thiessen, U. (2003) Fiscal Decentralisation and Economic Growth in
High Income OECD Countries. Fiscal Studies 24, 237-274.
Thornton, J. (2007) Further Evidence on Revenue Decentralisation
and Inflation. Economic Letter 95, 140-145.
Treisman, D. (2000) Decentralisation and Inflation: Commitment,
Collective Action, or Continuity. American Political Science Review
94:4, 837-857.
Tsebelis, G. (1995) Decision Making in Political Systems: Veto
Players in Presedentialism, Multicameralism and Multipartyism. British
Journal of Political Science 25:3, 289-325.
Wasylenko, M. (1987) Fiscal Decentralisation and Economic
Development. Public Budgeting and Finance 7:4, 57-71.
Wibbels, E. (2000) Federalism and the Politics of Macroeconomic
Policy and Performance. American Journal of Political Science 44:4,
687-702.
Xie, D., H. Zou, and H. Davoodi (1999) Fiscal Decentralisation and
Economic Growth in the United States. Journal of Urban Economics 45,
228-239.
Zhang, T. and H. Zou (1998) Fiscal Decentralisation, Public
Spending and Economic Growth in China. Journal of Public Economics 67:2,
221-240.
(1) Oates's "Theorem of Decentralisation" postulates
that: "For a public good--the consumption of which is defined over
geographical subsets of the total population, and for which the costs of
providing each level of output of the good in each jurisdiction are the
same for the central or the respective local government--it will always
be more efficient (or at least as efficient) for local governments to
provide the Pareto-efficient levels of output for their respective
jurisdictions than for the central government to provide any specified
and uniform level of output across all jurisdictions" [Oates
(1993)].
(2) Martinez-Vazquez and McNab (2003) mention five different
channels through which fiscal decentralisation may have influence on
economic growth including (i) consumer efficiency, (ii) producer
efficiency, (iii) the geographical distribution of resources, (iv)
macroeconomic stability, (v) corruption and captures by elites. However,
our focus in this dissertation is only to analyse macroeconomic
stability channel.
(3) See for example Treisman (2000), King and Ma (2001), Neyapti
(2004), Martinez-Vazquez and McNab (2006), Shah (2006) and Thornton
(2007).
(4) The empirical support for this continuity hypothesis is found
by Treisman (2000).
(5) The idea of soft budget constraint (SBC) is introduced by
Kornai (1979) to analyse the behaviour of slate owned firms. The SBC is
used in decentralisation system to refer to lower level governments that
look to a higher level government to recover or bailout their excessive
deficits. The term bailout implies to the additional funding that the
higher level government provides to the lower level governments when it
would otherwise be unable to service its obligations. On the other hand,
hard budget constraint (HBC) implies that lower level governments have
to face the full costs of their expenditure decisions. The soft-budget
constraint problem refers to the fact that federal transfers to
subnational governments are based on ex post financial needs and not, as
it should be, on ex-ante characteristics of the recipient states [Rodden
(2002)].
Author's Note: This paper is heavily drawn from my PhD Thesis
title "Fiscal Decentralisation, Macroeconomic Stability and
Economic Growth". I am grateful for the comments received from the
seminar participants al PIDE and anonymous referees.
Nasir Iqbal <nasir@pide.org.pk> is Staff Economist at the
Pakistan Institute of Development Economics (PIDE), Islamabad.
Table 1
Descriptive Statistics
Variables Obs. Mean Std. Dev Min Max
Revenue 39 0.130 0.041 0.071 0.221
Decentralisation (RD)
Expenditure 39 0.465 0.067 0.336 0.686
Decentralisation (ED)
Composite 39 0.247 0.089 0.129 0.494
Decentralisation (CD)
Inflation (INF) 39 9.587 5.748 03.10 30.00
Capital Stock per Worker 39 75273 16727 42950 95884
(CS/W)
Openness (OPN) 39 0.338 0.037 0.273 0.432
Tax to GDP Ratio (T/GDP) 39 0.123 0.015 0.095 0.145
M2 to GDP Ratio (M2/GDP) 39 0.403 0.039 0.297 0.469
Democratic Institution 39 0.846 6.745 -7.000 8.000
(INS)
Source: Author's own calculation.
Table 2
Unit Root Test (ADF Test)
Level
Variables No Trend With Trend Result
Revenue Decentralisation (RD) -2.13 -3.24 NS
Expenditure Decentralisation (ED) -1.72 -2.48 NS
Composite Decentralisation (CD) -1.69 -3.41 NS
Inflation (INF) -4.02 -3.62 S
Capital Stock Per Worker (CS/W) -2.81 -1.62 NS
Openness (OPN) -2.93 -3.56 S
Tax to GDP Ratio (T/GDP) -1.32 -2.02 NS
M2 to GDP Ratio (M2/GDP) -2.95 -4.58 S
Democratic Institution (INS) -1.97 -1.91 NS
INS*RD -1.75 -1.73 NS
INS * ED -1.91 -1.86 NS
INS * CD -1.71 -1.69 NS
First Difference
Variables No Trend With Trend Result
Revenue Decentralisation (RD) -4.63 -4.56 S
Expenditure Decentralisation (ED) -7.19 -7.02 S
Composite Decentralisation (CD) -5.49 -5.43 S
Inflation (INF)
Capital Stock Per Worker (CS/W) -1.44 -3.83 S
Openness (OPN)
Tax to GDP Ratio (T/GDP) -5.12 -5.71 S
M2 to GDP Ratio (M2/GDP)
Democratic Institution (INS) -5.71 -5.76 S
INS*RD -5.47 -5.51 S
INS * ED -5.58 -5.62 S
INS * CD -5.40 -5.44 S
Note: 5 percent critical value is -2.87 for the case of no-
trend, and -3.42 when a trend is included. AIC is used for
lag selection. S stand for stationary series and NS stand
for non-stationary series.
Table 3
The GMM Estimates, Dependent Variable is INF
Variables (1) (2) (3)
RD -1 175 ***
(0.409)
ED -2.337
(4.467)
ED -1.227 ***
(0.427)
Tax/GDP -1.395 ** -0.973 -1.514 **
(0.670) (1.104) (0.677)
Openness -1.070 *** -1.082 ** -1.023 ***
(0.212) (0.540) (0.222)
M2/GDP -1.423 -1.403 -1.255
(1.092) (2.054) (1.121)
Physical Capital -1.202 *** -0.803 -1.269 ***
(0.356) (0.512) (0.367)
Constant 29 24 *** 17.76 * 31.39 ***
(9.699) (10.22) (10.10)
Observations 37 37 37
R-squared 0.628 0.382 0.616
Wald Chi2 Test 98.89 41.37 94.72
Normality Test 1.16(0.56) 2.19(0.33) 1.63 (0.44)
End. Test P.V. 0.0509 0.0303 0.0380
0I Test P.V. 0.6376 0.8147 0.5857
D. W. Test Value 1.96 2.06 1.95
Note: Robust standard errors in parentheses; *** p<0.01,
** p<0.05, * p<0.1; The STATA v12 has been used
for estimation by using 'ivregress-GMM' command.
Table 4
The GMM Estimates, Dependent Variable is INF with
INS Included in the Model
Variables (1) (2) (3)
RD -0.551
(0.391)
ED -3.184
(4.383)
ED -0.655 *
(0.353)
RD * INS -0.150 *
(0.0894)
ED * INS -0.588 **
(0.292)
CD * INS -0.200 *
(0.113)
INS 0.0155 0.0160 0.0170 *
(0.0101) (0.0167) (0.00973)
Tax/GDP -1 789 *** -1.894 * -1.932 ***
(0.486) (1.011) (0.464)
Openness -0.710 *** -0.323 -0.647 ***
(0.161) (0.561) (0.155)
M2/GDP -0.206 1.417 0.00270
(0.729) (1.901) (0.730)
Physical Capital -1.058 *** -1.112 ** -1.160 ***
(0.259) (0.467) (0.236)
Constant 26.91 *** 26.95 *** 29.82 ***
(6.875) (8.718) (6.406)
Observations 37 37 37
R-squared 0.772 0.556 0.770
Wald Chi2 Test 201.7 151.6 211.1
Normality Test 2.10(0.35) 2.51 (0.23) 1.65 (0.44)
End. Test P.V. 0.0808 0.0374 0.0496
01 Test P.V. 0.5620 0.4219 0.5443
D. W. Test Value 2.16 1.81 2.20
Note: Robust standard errors in parentheses; *** p<0.01, **
p<0.05, * p<0.1; The STATA v12 has been used for estimation
by using 'ivregress-GMM' command.
