Alternative development policy models

The paper has as major objective to compare the relative performance of two alternative development policy models with respect to investment promotion. These models are: (i) promotion of externally generated investments and (ii) promotion of locally supplied businesses. They are alternative development models that many governments, at all levels, from local to national, face on their policies. More concretely, the paper develops an econometric method to compare the relevance of these two alternative models on economic growth and test their relative impact using Brazilian data. As a byproduct, the method also allows to test the hypothesis that development policies can have impact on growth performance.

The paper also brings a review of the recent literature on the subject, with a major focus on the role of promotion of multinational investments, which is a particular case of promotion of externally generated investments, when country borders are the spatial limits. As there is still controversy on their impact on growth, the paper also shed some lights on this particular subject, although the empirical tests presented use data to a particular country.

The method relies on a two steps procedure, which ends up with a standard cross section growth regression, which explains relative spatial per capita GDP performance by particular features of the spatial units. In a first step of the method, two variables to capture the intensity of each of these policies are generated as residuals in equations determining: (i) the share of registered workers on total active population, and (ii) the share of employers on total active population. It is argued in the paper why these residuals so strongly correlated with the strength of the two types of policies previously mentioned. In the second step, these two residuals are included in the growth regressions. The method is applied to Brazilian county data with growth defined to the period 2000 to 2009 and the other variables all determined at the year 2000.

The logic of the paper is exploratory, so that there was no a priori expected hypothesis on the relative power of these alternative models to push local development.  Results indicate that promotion of locally supplied businesses had a positive and statistically significant impact on growth, while no definitive conclusion about promotion of externally generated investments emerges, although there are stronger evidence that it harms growth, as pointed by some literature on multinationals and growth. Together, these outcomes reject the hypotheses that development policies do not have any impact on economic growth and support the hypotheses that promotion of locally supplied businesses can boost per capita GDP growth.