The second group of observations is that of non-ferrous metal and primary products listed in the CIS 333, 334 and 335 categories (primary ferrous metal metals, secondary metal metal metals and rolls and non-ferrous designs). These non-ferrous metal production facilities are included as a control group. The level of commercial defence of these products is relatively low compared to the steel industry (i.e. the ferrous metals industry), but it has similar product categories and production processes such as CIS 331 steel mills. A monitoring group is particularly important with respect to the trade policy of the VRA, where virtually all categories of steel have been taken into account in an AIT, which poses a problem of identifying separately the effects of theVRA on steel mills of an un observable annual effect, if one considers only steel mills. While the use of the complete CM as a steering group can provide an interesting robustness check, data availability and input difficulties make this impossible. Roeger, Werner. (1995) “Can imperfect competition explain the difference between primary and dual product measures? Estimates for U.S. Manufacturing, Journal of Political Economy, Vol. The results provide even clearer evidence of the non-equivalent market power effects of tariffs and REAs than in the pooled sample. First, there is still no sign of positive market effects through tariff protection. Second, the significant effects of the market strength of mandatory ARVs in 1987 are clearly for steel production facilities (integrated and mini-mills) and not for steel processors. This is consistent with the idea that SAs increase the cost of major steel products purchased by steel processors and reduce the impact of ARVs on their final product.
An interesting result is that there is evidence of significant market power effects for both 1972 and 1987 VR, with very similar sizes in integrated installations and mini-mills. Finally, we find that these results are very similar in the FE and FD specifications for instrumental variables. The main assumptions in this paper relate to the different effects that trade defence policies imposed on foreign competitors in our sample facilities may have on the price capacity of a facility above marginal costs. To examine it empirically, we are resiencing the MCP to a number of product-specific (k-indexed) trade policies. In light of our previous discussions, we expect quotas and AD duties to significantly increase the MCP, while tariffs and CVDs have an insignificant impact.