Skip Navigation Planning & Markets
Subscribe Submission Requirements Editorial Board Archive Links Search Home



II. Some Evidence Of The Leviathan's Effects

If Brennan and Buchanan's (1980), Jaber's (1994) and Lijeron's (1996) results possess general validity, it should be possible to find evidence indicating that more decentralized government revenues and expenditures lead to a smaller government sector and then to higher growth rates, as suggested by the Barro and Sala-i-Martin (1999, pp.38-39) cross-section and panel results.

Localization of government, and also partial localization of public finances, are major issues, particularly in developing countries. It will be interesting to compare developing countries with industrialized countries to determine what the impact of differing qualities of public financial decentralization on local government size and local growth might be. To analyze these questions, a comparative study was undertaken in two industrialized and two developing countries: Thailand and the Philippines and, Switzerland and Spain. To further stress the poverty angle, a relatively poor local community with a homogenous prevalent culture and little inequity was selected in each country: Tambon Huay Yai in Cholburi Province in Thailand; Barangay Labac in Cavite Province in Philippines; Zorita in the Province of Caceres, Region of Extremadura in Spain; and Schwende in Canton Appenzell Innerrhoden in Switzerland.

The four countries can be summarily characterized as follows:

Thailand is

  • more or less homogenous culturally and has a common language;
  • a constitutional monarchy with representative democratic system;
  • a strongly centralized governmental system and working on a process of decentralization; and
  • struggling to become a Newly Industrialized Country.

Switzerland is

  • home to four distinctly different cultures: German, French, Italian and Romansh;
  • a federal republic with a semi-direct democratic system;
  • an extremely decentralized governmental system; and
  • economically in a process of transforming into a post-industrial economy.

Spain is

  • a constitutional monarchy with representative democratic system;
  • a unitary, decentralized governmental apparatus; and
  • an historically forced homogenous culture derived from the politically dominant region of Castile, but contested by the other two major cultural regions: Cataluña and the Basque Territory. Today these two regions are "autonomous communities" where Catalan and Basque are official languages besides Spanish;
  • Economically Spain has transformed during the past 25 years from a poor agricultural economy into a very prominent and successful Newly Industrialized Country.

Philippines is:

  • ethnically and culturally extremely inhomogeneous with 45 distinctly different ethnographic groups; and
  • the only Southeast Asian country colonized by the West before having developed a central government. Consequently western political-administrative systems were introduced at an early stage. Historically, the Philippines was always a strongly regionalized country with a relatively weak central government thus always allowing a certain degree of regional and local self-rule of the elite classes.
  • After decades of struggle against dictatorial rule, the Philippines is today a unitary decentralized, presidential parliamentary system.
  • Economically the Philippines is strongly agriculture based belonging to the group of developing countries.

Table 1 below gives some evidence of the four countries' degree of decentralization of their public financial systems and of their GDP growth data.

TABLE 1: Public finance data for some selected countries

Country Government Expenditures of All Levels as % GDP 1997 Subnational Government Expenditures as % of Total Public Expenditures 1997 Subnational Government Revenues as % of Total Public Revenues 1997 Average Annual GDP Growth Rates 1990 - 1998
Thailand 20.7   9.6   5.5   7.4  
Philippines 14.2   6.5   4   3.3  
Spain 41.7   35   13.8   1.9  
Switzerland 36.9   49.3   35.5   0.4  
For comparison
Germany 45.8   37.8   28.8   1.6  
Sweden 59.8   36.2   31.4   1.2  
USA 33.0   46.4   32.9   2.9  
The Philippine expenditures data refer to 1980
Source: World Development Report 2000, World Bank, Washington D.C., 2000 and Bundesamt fuer Statistik, Neuchatel, 2000

With regard to the above data one can make the following observations:

  • Both, Thailand and the Philippines are very centralized in expenditures and revenue. The magnitude of their growth rates is due to their low initial GDP. Their growth figures are therefore not indicative of a superior performance relative to industrial country results.

  • Spain is decentralized in expenditures and centralized in revenues.

  • Germany, Switzerland, Sweden, and the USA are decentralized in expenditures and revenues.

Table 1 does not indicate an obvious relation between sub-national fiscal behavior and GDP growth rates. Therefore, a more extensive correlation analysis of 21 industrialized and 18 developing countries for which complete comparable data could be found in the World Bank's World Development Report 1999/2000 was carried out. Results are shown in Table 2, below. The two-country classifications are necessary as growth rates in developing countries are generally higher than those of industrialized countries due to the base effect of low initial GDP. This would bias the correlation results.

Tables 2 and 3 below show the correlation results between GDP growth rates and fiscal decentralization. The analysis of the industrialized countries shown in Table 2 below reveals a statistically significant negative correlation at the 99% confidence level between sub-national expenditure ratios (-0.593), sub-national revenue ratios (-0.672) and central government expenditures.

The correlation between central government expenditures and GDP growth rates is weakly negative, but statistically not significant.

TABLE 2: 21 Industrialized Countries: Correlation of Average GDP Growth Rates 1990 -1998 and Fiscal Decentralization

Variables Real Annual GDP Growth Rate 1990-1998 Sub-national Government Expenditures as % Total Public Expenditures 1997 Sub-national Government Revenues as % Total Public Revenues 1997 Total Central Government Expenditures as % GDP 1997
Real Annual GDP
Growth Rate 1990-1998
Pearson Correlation
Sig (2-tailed)
N
1.000
.
21
-.105
.652
21
-.320
.157
21
-.046
.884
21
Subnational Government Expeditures as % Total Expeditures 1997 Pearson Correlation
Sig (2-tailed)
N
-.105
.652
21
1.000
.
21
.829XX
.000
21
-.593XX
.005
21
Subnational Government Revenues as % Total Public Revenues 1997 Pearson Correlation
Sig (2-tailed)
N
-.320
.157
21
.829XX
.000
21
1.000
.
21
-.672XX
.001
21
Total Cental Government Expenditures as % GDP 1997 Pearson Correlation
Sig (2-tailed)
N
-.046
.844
21
-.593XX
.005
21
-.672XX
.001
21
1.000
.
21
XX. Correlation is significant at the 0.01 level (2-tailed).
Source: World Development Report 1999/2000, Word Bank, Tables A1, 11, 14

TABLE 3: 18 Developing Countries: Correlation of Average GDP Growth Rates 1990 -1998 Fiscal Decentralization

Variables  Real Annual GDP Growth Rate 1990-1998 Sub-national Government Expenditures as % Total Public Expenditures 1997 Sub-national Government Revenues as % Total Public Revenues 1997 Total Central Government Expenditures as % GDP 1997
Real Annual GDP
Growth Rate 1990-1998
Pearson Correlation
Sig (2-tailed)
N
1.000
.
18
-.091
.718
18
-.035
.891
18
-.354
.150
18
Subnational Government Expeditures as % Total Expeditures 1997 Pearson Correlation
Sig (2-tailed)
N
-.091
.718
18
1.000
.
18
.752XX
.000
18
-.107
.673
18
Subnational Government Revenues as % Total Public Revenues 1997 Pearson Correlation
Sig (2-tailed)
N
-.035
.891
18
.752XX
.000
18
1.000
.
18
-.254
.327
18
Total Cental Government Expenditures as % GDP 1997 Pearson Correlation
Sig (2-tailed)
N
-.354
.150
18
-.107
.673
18
-.254
.327
18
1.000
.
18
XX. Correlation is significant at the 0.01 level (2-tailed).
Source: World Development Report 1999/2000, Word Bank, Tables A1, 11, 14

The analysis of the 18 developing countries reveals the same correlation pattern, though the coefficients are statistically not significant. With some caution one could therefore make the prediction that, more decentralized public expenditures and revenues lead to lower central government expenditures and thus to higher growth rates.

Barro and Sala-I-Martin (1995), in what follows called BSM, and of Barro (1999, pp. 38-39) both found that the level of government consumption has a negative (-12.9%, BSM; -11.0%, Barro) effect on the per capita growth rate. Assuming that their findings, which relate to the national level, are also valid for the local level, the above sub-national fiscal data become important local growth determinants.

page 31

IndexContinue

USC Seal


Main Page | Subscribe | Submission Requirements | Editorial Board | Archive | Links

PLANNING & MARKETS
http://www-pam.usc.edu/
ISSN 1548-6036

Copyright 1999-2000
University of Southern California
Los Angeles, California 90089-0626
USA