Does Fiscal Policy Help Those Who Need It Most? Evidence from the US and the Eurozone

Abstract

With a notion that the eurozone does not have a fiscal federalist system, we analyze the effectiveness of fiscal policy at the country level of the eurozone and the state level of the US. Our empirical analysis shows that, not like the case of the US, the role of fiscal policy for counteracting regional income shocks is far weaker in low-income countries in the eurozone. In addition, the contagion effect of fiscal crises in the eurozone is stronger and this effect becomes stronger in the eurozone after the establishment of the European Union. Our finding provides empirical evidence for fiscal problems in a single-currency area without a fiscal federalist system.

Keywords:

• Fiscal federalist system
• eurozone
• US
• fiscal policy
• single currency

Acknowledgements

This work was supported by IT forum under the research project, “Future of European Economies after the Fiscal Crisis”. This work does not reflect any views of Korea Development Institute.

Notes

We modified the title of Hanson et al. (2006), who analyzed the asymmetric effects of monetary policy across the states in the US.

1. The European Central Bank (ECB) can play some role. However, it is still a difficult task to make well-coordinated action among member countries. The abrupt resignation of Germany's member of the ECB's board on 9 September 2011 is one example of disagreement over handling of the Eurozone debt crisis.

2. To qualify for EU funding, to receive additional support a country has to exceed the agreed-upon goal with the EU executive committee by over 5%, countries that have a budget deficit as a percent of GDP of over 3% over three consecutive years are presumed to have violated eurozone financial regulations and thus sanctions are discussed.

3. The reason for choosing these variables is related to the automatic stabilizer of fiscal policy. For example, it is well known that income tax revenue is procyclical and the government expenditure on unemployment insurance is countercyclical.

4. We also calculate the standard deviation and coefficient of variation. Since the correlation coefficients among three inequality measures are larger than 0.8 in all cases, we report the result based on the Entropy index only.

5. In general, when comparing classes with disparate income levels, a greater importance (i.e. a=0) is given to the change in income of the low-income class (Litchfield, 1999).

6. Eichengreen et al. (1996) use this equation to see the contagion effect of the exchange-rate crises and Bohn (1998) uses this in a model of fiscal crisis caused by the accumulated public debt.

7. Eurozone's high-income countries are in total of 14 countries including Belgium, Denmark, Germany, Ireland, France, Luxembourg, the Netherlands, Austria, Finland, Sweden, the UK, Iceland, Norway and Switzerland and low-income countries are in total of 12 countries including Bulgaria, the Czech Republic, Spain, Greece, Italy, Latvia, Hungary, Poland, Portugal, Rumania, Slovenia and Turkey. There are a total of 29 high-income US state governments which are Alaska, California, Colorado, Connecticut, Delaware, Columbia, Hawaii, Illinois, Iowa, Kansas, Maryland, Massachusetts, Minnesota, Nebraska, Nevada, New Hampshire, New Jersey, New York, North Carolina, North Dakota, Oregon, Pennsylvania, Rhode Island, South Dakota, Texas, Virginia, Washington, Wisconsin and Wyoming. There are 22 low-income state governments including Alabama, Arizona, Arkansas, Florida, Georgia, Idaho, Indiana, Kentucky, Louisiana, Maine, Michigan, Mississippi, Missouri, Montana, New Mexico, Ohio, Oklahoma, South Carolina, Tennessee, Utah, Vermont and West Virginia.

8. As mentioned previously, we do not report them here because the correlation coefficient of these measures with the Entropy index is higher than 0.8.

9. Duclos (2002) attempts factor decomposition of the formula, which can be stated as the following equation:where EI(0) _k is per capita fiscal revenue and expenditure of country or state government k and ν _k is the ratio of country or state government k population to the total population . λ _k indicates the ratio of the average of country or state government k's per capita fiscal revenue and expenditure to the average of the total population . This kind of decomposition makes it possible to analyze the inequality within so-called high- and low-income regions and between these two regions of each country or state government.

10. We add the lagged GDP growth rates, (Δln(GDP _i,t–1), Δln(GDP _i,t–2)) to account for policy lags when estimating equation (3).