Economists have been unable to agree on an appropriate model of employment determination in unionized firms. In some models, employment is determined unilaterally by the firm and bargaining covers only wages. In others, bargaining overs both wages and employment. In Discussion Paper No. 320, Research Fellows Stephen Nickell and Sushil Wadhwani depart from previous attempts to discriminate between these competing models by incorporating the implications of efficiency-wage models, where work effort and productivity in a particular firm are influenced by the wage level relative to wages in other firms.
They develop a general model of simultaneous wage and employment bargaining, which encompasses various existing models as special cases models where relative wages have no effect on work effort, efficient bargaining models, and `right to manage' models where employment is determined unilaterally by the firm. The authors find that the special cases of efficient bargaining and the right to manage imply identical restrictions and so cannot be distinguished in their general model.
In looking for an alternative approach to distinguishing between different models of employment determination, Nickell and Wadhwani focus on the traditional assumption in such models that the wage and level of employment are determined simultaneously. They argue that it is important to take account of the fact that, in UK industry at least, wages are set at discrete intervals while employment is adjusted continuously.
Nickell and Wadhwani therefore develop a sequential model in which unions and firms first bargain over the wage (taking the relevant exogenous variables as random), and then bargain over employment with the wage taken as given. In their empirical work, the authors use data on over 200 large UK manufacturing companies during 1972-82. The evidence on whether the level of employment in unionized firms is subject to bargaining is inconclusive, but there is evidence that employment adjustment is more sluggish in unionized firms.
The empirical results are consistent with the view that employment in a firm is negatively related to its wage. Some of the results also suggest that an increase in the outside wage raises employment by enough that, on aggregation, the `demand curve for labour' could be upward-sloping. This can happen if a rise in the outside wage increases employment in one firm, by reducing productivity in that firm, sufficiently to outweigh the employment-reducing effect of an increase in the firm's own wage. This result depends, however, on special assumptions about the effort and union utility functions.
The results also suggest that financial factors are important determinants of employment. It is conventional to assess creditworthiness on the basis of a firm's debt/equity ratio, so it is rational for a firm to focus on it when deciding employment, Nickell and Wadhwani argue. Stock market fluctuations also appear to influence the level of employment: if inflation depresses the equity market, then this may be a channel through which higher inflation can reduce employment

Employment Determination in British Industry: Investigations Using Micro-Data
Stephen Nickell and Sushil Wadhwani

Discussion Paper No. 320, June 1989 (AM)