The estimation results of the system (5) for the eight economic sectors are presented in . Despite the fact that the periods are different and that the FDI figures are not fully comparable, the results confirm most of the findings of the aggregate model in . First, the coefficient of the linear FDI effect is positive and highly significant in the output equation. In general, the coefficients of the FID#sector interaction variables (not shown due to space limitations) are not significant individually at the 5% level and are not jointly significant either. This means that we fail to reject differences in the effects of FDI on output across sectors, or, put differently, that FDI is equally pro-growth in all sectorsIndore Stock. Second, the coefficient of the linear GDP effect remains positive in the FDI equation and is now even larger (around 4.3)Bangalore Investment. Nevertheless, now it is only marginally significant (at the 10% level)Surat Investment. Again, the coefficients of the output#sector interaction variables (not shown) are not significant, indicating that we cannot discard that FDI is equally responsive to growth in all sectors. Regarding the FDI equation, it is worth noting that most reported coefficients are not significant, thus leaving most of the explanatory power to sector and year dummy variables and their interactions (not shown). This means that FDI in Chile concentrates on specific sectors (such as mining) and seems to be highly sensitive to industries’ cycles. One clear example is the “copper super cycle”, which took place more or less during the estimation period. The copper super cycle was initially characterized by a sharp increase in international copper prices, leading to sizable inflows of foreign investments in the Chilean mining sector (). A third important result is that the output variable remains positively related to the demand for both unskilled (column ) and skilled labor (column ), with estimated coefficients that remain highly significant. The output#sector interaction variables (not shown) indicate that the effects of output expansion are particularly strong in the agriculture-forestry-fishing (AFF) sector. On the contrary, this effect is the lowest in the mining sector, where the estimated interaction coefficient is the most negative (and statistically significant). This latter result is not surprising given the relatively high capital intensity of this industry in Chile, at least compared to other sectors (). Finally, except for skilled labor, factor demands relate in general negatively to the factor’s price, as the theory indicates. The cross-price elasticities are somewhat variable across sectors, suggesting sector-specific substitution/complementarity relationships among production factors. The prevalent pattern is, however, negative cross-price elasticities, confirming factors’ complementarity found with the aggregate model.
Article Address： http://lseshc.com/Stock/138.html
Article Source：Admin88
Notice：Please indicate the source of the article in the form of a link。
