In the world of free capital mobility, capital flows from low-return to high-return locations. In theory, capital inflows are transformed to physical capital and increase output in the recipient countries. In the empirical literature, the effects of capital inflows on economic growth have been extensively investigated using cross-country growth regressions. However, these studies have been silent about the scale of the benefits from capital inflows.
This study proposes a new methodology to quantify the benefits from capital inflows. We derive the estimation equation from a small open-economy growth model with an incomplete asset market. There are two goods: consumption good and investment good, which is produced from the consumption good. The model has two key features. First, the investment good is assumed to be nontraded and produced from output, which is homogeneous across countries. Second, the country-specific price of investment is driven by exogenous shocks on productivity of the investment good sector. These two assumptions are motivated by the following stylized facts.
First, imported investment goods such as machinery occupy a small share in the aggregate investment expenditure. This suggests that prices of investment goods are largely driven by fluctuations of price of nontraded investment goods such as structure. Second, the price of investment goods relative to output is volatile particularly in developing countries. Third, fluctuations in the price of investment goods relative to output is largely driven by the price of investment goods, not the price of output.
With fluctuations in price of investment, the return to capital in our model must be adjusted by the price of investment goods relative to output. This adjustment was also proposed by Caselli and Feyrer (2007) and Hsieh and Klenow (2007), which study differences in price of investment across countries. In the long run, capital mobility reassures that the domestic and foreign returns are equalized up to the financial frictions. In the short run, productivity shocks in the domestic investment good sector cause the domestic return to fluctuates and thus create reallocation of capital. Our model gives three main predictions concerning the dynamics of current account.
The first is about the direction of capital flows. Regardless of the long-run level of capital stock, positive productivity shocks in the investment good sector raise the return to capital trigger capital inflows. For this reason, high-income countries can attract foreign capital and benefit as recipients of capital flows. Thus, an improvement in productivity in the investment good sector offers an explanation why capital flows from low-income to high-income countries as in Lucas (1990). To the contrary, negative productivity shocks in the investment good sector lowers the return to capital trigger capital outflows. Hence, we can exploit data on both capital inflows and outflows to estimate the model-based correlation between capital flows and domestic return. The existing empirical studies, however, exploit only information about capital inflows in their estimation.
The second finding is related to the scale of capital inflows. The correlation between the scale of capital inflows and domestic return is theoretically decreasing in the productivity in the investment good sector. This prediction results from the nontradedness of the investment good. To be specific, imported financial capital is subject to the domestic technology when it is transformed into domestic physical capital. Hence, an improvement in efficiency in producing investment good reduces the quantity of imported financial capital for a given level of domestic capital.
The last finding is about the scale of gains from capital inflows. The gain from capital inflows in terms of output per worker is increasing in productivity in the investment good sector. Again, this is because foreign financial capital is subject to the domestic technology in the investment good sector. Improvements in efficiency then increases the quantity of physical capital financed by capital inflows and the eventual output of final good.
In the empirical part, we employ a panel of 47 countries from 1970 to 2003 to construct time series of country returns. The adjustment for price of investment goods roughly doubles the standard deviation of returns. That allows us to exploit the volatility of return to estimate the correlation between net capital inflows and changes in return. The estimation also takes into account possible shifts in the world return. Then, we use the predicted scale of capital inflows to compute the output gains from capital flows. This is because the theory suggests that capital inflows contribute to output growth when they are driven by fluctuations in return to capital. In theory, capital outflows also benefit households in the source countries by allowing them to reap higher return than the domestic return and increase consumption as a result. However, our study focuses on the impact of capital flows in the recipient countries.
We find weak evidence that in the capital inflows respond to movements in returns in the short run. In particular, banking flows are found to be positively correlated with the domestic return and negatively correlated with the world return, as predicted by the model. 42 out of 47 countries in our sample reap positive gains from the capital inflows. However, the inflows raise output by less than 1 percent over the entire period. The quantitative impact is actually in line with the estimate by Caselli and Feyrer (2007). The results contrast with the finding in Henry (2003). For all countries, gains from capital inflows are short-lived and no countries reap gains in all years in which they import capital.
Our study does not take into account potential gains through other channels such as risk-sharing in Obstfeld (1994) and Athanasoulis and van Wincoop (2000). We also abstract from a possibility that capital inflows may raise productivity in the recipient countries. In this aspect, ours is closely related to the work by Gourinchas and Jeanne (2006), who theoretically investigate the effect of capital inflows on convergence of a small open economy. However, their work is abstract from the price of investment goods and found that capital flows substantially increase output although the welfare gains are quite small.
The paper is organized as follows. The model is in the next section. Section 3 discusses the empirical methodology and results. Section 4 concludes.
Download
PDF Ebook The Marginal Product of Capital,a Capital Flows and Convergence
