A Service of zbw Leibniz-Informationszentrum Wirtschaft Leibniz Information Centre for Economics Globalization and developing countries -a shrinking tax base? Globalization and Developing Countries -a Shrinking Tax Base ?

Standard-Nutzungsbedingungen: Die Dokumente auf EconStor dürfen zu eigenen wissenschaftlichen Zwecken und zum Privatgebrauch gespeichert und kopiert werden. Sie dürfen die Dokumente nicht für öffentliche oder kommerzielle Zwecke vervielfältigen, öffentlich ausstellen, öffentlich zugänglich machen, vertreiben oder anderweitig nutzen. Sofern die Verfasser die Dokumente unter Open-Content-Lizenzen (insbesondere CC-Lizenzen) zur Verfügung gestellt haben sollten, gelten abweichend von diesen Nutzungsbedingungen die in der dort genannten Lizenz gewährten Nutzungsrechte. This paper evaluates the impact of globalization on the tax bases of countries at varying stages of development. We see globalization as a process that induces countries to embrace greater trade and financial integration, and macro stabilization. This in turn should shift their tax base from "easy to collect" taxes [tariff, seigniorage, etc.] towards "hard to collect" taxes [VAT, income tax, etc.]. We confirm this prediction -the revenue/GDP ratio of the "easy to collect" taxes declined by about 12% in developing countries between the early 1980s and the late 1990s, while the revenue/GDP of the "hard to collect" taxes increased by 16%. The relatively small initial base of "hard to collect" taxes in developing countries implied a net 2% drop in total tax revenue/GDP. Applying panel regressions and controlling for structural factors, we find that trade openness and financial integration have a positive relationship with "hard to collect" taxes, and negative relationship with the "easy to collect" taxes. Fiscal revenue from financial repression has also decreased, further reinforcing these results. The high income and the middle income countries managed to more than compensate for the revenue decline of the "easy to collect" taxes, increasing the total tax/GDP. In contrast, the upper and low income developing countries experienced sizeable drop in the tax/GDP. We also identify strong fiscal convergence during 1980s -1990s: the coefficient of variation of tax revenue/GDP measures across countries declined by about 40% for seigniorage, about 40% for tariff, and about 4% for the "hard to collect" taxes. We confirm the robustness of the main results to IV methodology, where trade globalization is inferred from applying the gravity methodology. These results are consistent with the notion that improving the performance of the "hard to collect" taxes is more challenging than reducing the use of "easy to collect" sources of revenue. 2 The factors explaining globalization include faster, more reliable, and cheaper delivery of goods and services across borders, and advances in information technology that allow cheaper fragmentation of production via FDI and outsourcing, etc. Terms of use: Documents in EconStor may 3 base of developing countries, and the degree to which these countries managed to switch from the traditional "easy to collect" taxes to new, "hard to collect" taxes. We trace factors explaining the heterogeneity of the negative fiscal shock induced by globalization and thus explaining the depth of the fiscal adjustment accomplished already by countries at varying levels of development. . 5 While the drop in "easy to collect" effective tax rates is more pronounced for high income countries, the greater initial base of "hard to collect" taxes in these countries relative to developing countries implies a net increase in total taxes of 6% in the first group, and a drop of 2% in the second. The position of the 4 South Asian countries is in between the high income and the developing countries: the large increase in the revenue from hard to collect taxes more than offset the drop in the revenue from the "easy to collect" taxes, increasing total tax revenue by 4%. These figures mask significant heterogeneity of the adjustment across countries, an issue that will be investigated in this paper using regression analyses. 3 The Appendix B provides the list of countries in our study, and definitions of the various regions. The choice of these countries is dictated by data availability over the period 1980-1999. 4 We compare a 5-year average separated by ten years due to two reasons. First, a 5-year average should smooth variations associated with business cycles. Second, the cumulative effects associated with globalization should be more visible by comparing two sub-periods separated by longer time. The country coverage in the figures is dictated by data availability. 5 Limited data availability about tax revenue from financial repression prevents us from adding this implicit tax to the figures dealing with "easy to collect" taxes. Hence, these figures tend to understate the shrinking of "easy to collect" taxes. In section 2 we conduct a regression analysis for the subset of countries where we managed to collect data about tax revenue from financial repression. 4 (measured horizontally) and in the late 1990s (measured vertically). The further away is a country from the 45-degree ray, the greater is the change over time in that country. increased. The average total tax revenue/GDP declined from 0.182 to 0.176. The average revenue/GDP of "easy to collect" taxes declined during the period from 0.060 to 0.046 [see Further insight is gained from cross-country regressions as well as a detailed panel study explaining the patterns of taxes over time. Both are described in the next section