Foreign Direct Investment Flows and Economic Growth: An Autoregressive Distributed Lag Analysis (ARDL) for Chile, 1960-2020

Abstract

This paper utilizes the recent ARDL modelling technique to test whether foreign capital complements or substitutes for private capital and economic output (labor productivity) in the Chilean economy for the 1960-2020 period. That is, it evaluates whether the so-called “complementarity hypothesis” is refuted in the Chilean case. The discussion is motivated by presenting a modified production function which treats the foreign capital stock as an additional input and provides an economic rationale for its inclusion. The paper also addresses the issue of whether inward foreign direct investment diverts resources away from the financing of private investment spending via the repatriation of profits and dividends back to parent companies, often residing in developed countries. Several important findings were revealed. First, the Bounds test indicated that the included I(0) and I(1) variables in both the production (labor) productivity functions are cointegrated. Thus, there is a long-term and stable relationship between the variables over the period under review. The error-correction terms for both functions indicated that deviations away from the long-term output (labor) productivity relationship are corrected in subsequent periods. Second, the results suggest that both private and foreign capital have a positive (long-term) effect on both economic output and labor productivity, controlling for both quantitative and qualitative variables included in the empirical model, such as the terms of trade and dummy variables. Third, the model diagnostics for both the output and labor productivity equations indicated that the estimates do not exhibit serial correlation and heteroscedasticity, and that they are structurally stable over the period under review. Fourth, the estimates for the output and labor productivity equations which include the “net” foreign capital stock (viz., adjusted for the repatriation of profits), suggest that foreign capital continues to have a positive, albeit smaller, effect on economic output and labor productivity. The latter finding suggests that, although the impact of foreign capital in terms of financing may be diminished, it still may be providing other benefits or externalities in the form of the transfer of better technologies and managerial expertise. Fifth, Block-Granger “causality” tests find strong support for precedence from foreign and private domestic capital to output (labor productivity), but also for “reverse causality or precedence” from output (labor) productivity to foreign capital and from foreign capital to private domestic capital. This finding provides further support for the complementarity hypothesis.