Bankers' Pay and the Evolving Structure of US Banking

We study the evolution of pay in US bank holding companies since 1986 using a structural model of the banking firm. The model incorporates a strong complementarity between capital provided by shareholders and bankers' talent and its non-contractible effort. Bankers' pay is given as solution to the second-best problem of maximizing payoff to shareholders subject to the bankers' participation and incentive compatibility constraints. Market equilibrium is found as an assignment model in which managers with different levels of talent are matched in rank order with banks of different capital. We set out the main empirical characteristics of the US banking sector in both cross-section and time series focussing on the consolidation of the banking sector and on three characteristics of bankers' pay: labor's share of bank value-added, the level of bankers' pay and its sensitivity to bank performance. We then calibrate the structural model that we have introduced to see if it can reproduce the empirical characteristics that we have found. We find that three major changes in banking regulation have shaped bankers' pay in the last three decades. First, the removal of obstacles to interstate banking which set-off consolidation process that is still on-going., Second, since the Gramm, Leach, Bliley Act the freedom to combine credit intermediation with activities generating non-interest income has driven a trend toward higher pay and higher incentive pay in banks aiming for higher shares of non-interest income. Finally, the mass of tougher regulations brought on by the financial crisis has had the effect of imposing an implicit tax on size and complexity which in turn has moderated the trend toward higher and more sensitive pay in large, complex banks. Indirectly this has given an opening for smaller banks to compete for some of the business outside of standard credit intermediation. But in so-doing, this has resulted in an increase of their pay levels and pay sensitivity. We find some evidence of a decline in average talent in the sector and that the trend toward high average pay has been driven in large part by the increase in managers' options outside banking. Overall, after controlling for the hypothesised regulatory tax on large banks we find a secular trend toward a decline of labor's share brought-on by a continuing process of consolidation in the US banking sector. Finally we find that although pay levels have risen significantly in three decades the premium received over fair pay in our model is rather small.