Long-run Implications of the Brazilian Capital Stock and Income Estimates

Abstract

The present study aims to analyze the empirical as well as theoretical implications related to the possible inconsistencies between the Brazilian capital stock estimate and its associated investment decision. The common practice of using the country’s accumulated (depreciated) fixed capital formation data as a proxy for the capital stock series generates a set of incompatible facts with dynamic models built on balanced growth and on aggregate production functions. Moreover, a related issue on the Brazilian capital income is considered in our analysis. According to the country’s National Accounts, the participation of capital income reaches about half of the aggregate income which is an unusual high share compared to international standards. It is shown that this problem can also be solved using alternative methods that lead to a more suitable capital stock series to be used in recursive equilibrium models. Finally, the long-run impacts of using the proposed capital stock series is studied using a modified basic growth model calibrated to reproduce some Brazilian empirical facts