We assume further that the union recognizes the possibility that price level increases may offset wage rate increases, and it does not entirely disregard the effect of price increases arising from its own wage increases upon the ``real'' wage rate. For internal political reasons, the union asks for (and accepts) increases in the basic wage rate, and would vigorously oppose a reduction in this rate, but the adjustment of the basic wage rate upwards is essentially up to the discretion of the companies of the industry.

Changes in the basic wage rate are cost raising, and they constitute an argument for raising prices. However, it is not known to either the union or the public precisely how much of a cost increase is caused by a given change in the basic wage rate, although the companies are presumed to have reliable estimates of this magnitude.

In this model, then, the industry is presumed to realize that they could successfully resist a change in the basic wage rate, but since such a change is the only effective means to raising prices they may, in circumstances to be spelled out in Part 2, below, find it to their advantage to allow the wage rise. Thus, for non negative changes in the basic wage rate, the industry becomes the active wage setter, since any increase in the basic wage rate can occur only by reason of industry acquiescence. The presumption in the literature would appear to be that the basic wage rate would be unchanged in this case, on the grounds that it is ``clearly'' not in the interest of the industry to raise wages gratuitously. From this presumption it is an easy step to the conclusion that any observed increases in the basic wage rate must be due to union behavior different and more aggressive than assumed in our model. It is this conclusion that we challenge; we do so by disproving the presumption on which it is based.