Relative Returns to Policy Reform : Evidence from Controlled Cross-Country Regressions

de Castro, Alexandre Samy ; Goldin, Ian ; Pereira da Silva, Luiz A.

World Bank

关键词: AGRICULTURAL EXPORTS;    AGRICULTURAL PRODUCTS;    AGRICULTURE;    ANDEAN PACT;    ANTIDUMPING;   

DOI  :  10.1596/1813-9450-2898
RP-ID  :  WPS2898

学科分类：社会科学、人文和艺术（综合）

来源: World Bank Open Knowledge Repository

【 摘 要 】

The authors aim at contributing tounderstand the dispersion of returns from policy reformsusing cross-country regressions. The authors compare the"before reform" with "after reform" GDPgrowth outcome of countries that undertookimport-liberalization and fiscal policy reforms. They surveya large sample (about 54) of developing countries over theperiod 1980-99. The benefits of openness to trade and fiscalprudence have been extensively identified in the growthliterature, but the evidence from simple cross-sectionanalysis can sometimes be inconclusive and remainsvulnerable to criticism on estimation techniques, such asidentification, endogeneity, multi-colinearity, and thequality of the data. The authors use a different analyticalframework that establishes additional controls. First, theyconstruct a counterfactual control group. These arecountries that-under specific thresholds-did not introducepolicy reforms under scrutiny. Second, the authors also tryto use the most appropriate variable of policy reform, forexample, exogenous changes in import-tariffs instead of theendogenous sum of all trade flows. Third, the authors try tobase the before-after reform comparison on the most accuratedate for the beginning of a policy reform period (instead ofcomparing averages over fixed intervals of time). Once thesecontrols are set, they explain the difference betweenaverage GDP growth rates during the country-specific postand the pre-reform periods, relative to the average GDPgrowth of the relevant control group. The explanatoryvariables in the regressions include the standardgrowth-regression controls. The results are the following:1) With a better measurement and timing of the policyreforms, the growth effect (the "returns onreform") is generally smaller than in previous papers.2) There is evidence of contingent relationships betweenpolicy and growth, corresponding to the country's size,its export profile, and its governance. 2) Within the groupof policy reformers, some countries have exhibited arelatively weaker growth response. Overall, the findingssuggest that more accurate measurement and definition of thetiming of reforms does not strengthen the significance ofthe effects of reforms on GDP growth. In fact, the effectsare weaker than indicated in most cross-section studies.This suggests that the policy implications to be derivedfrom these relationships should be treated with even morecaution than previously thought.