The recent global debate over the pay of bankers has raised issues that are basic to economic theory, and to moral and political philosophy, simultaneously. One question concerns what level of pay is necessary in order to attract people to do a certain job, and to do it effectively. Another question concerns what level of pay would be appropriate, fair, or just to pay people to do that job. Both are entailed in the current debate on the pay of bankers (particularly investment bankers).
The G-20, and more recently the European Parliament, as well as individual countries (especially Germany, France and the UK) have promoted the institution of new norms to govern the pay of bankers, requiring for example that their pay should be spread over a number of years, or be provided in part in the form of shares or other instruments, the return of which will depend on the success of the bankers’ strategies over a longer period than previously. There are incentive-based arguments for these reforms, which suggest that altering pay packages in such a way will reduce the propensity of bankers to engage in inappropriate risk-taking (socializing risk while privatizing benefits). However, there are also, implicit in the political discussion, concerns about fairness. Financial sector rewards have been unusually large relative to those prevailing in other sectors of the economy, even for people with otherwise apparently comparable qualifications.
The spokespersons of the banking industry claim that very high rewards for bankers are necessary in order to attract them to do the work which is required. Implicit here is the idea that the skills which are required to perform this work are based in naturally occurring talents which are sparsely distributed, are costly to acquire or are greatly onerous and unpleasant to employ. For one or another of these reasons, in the absence of suitably high rewards, individual banking firms are likely to lose the workers they need to function — if the rewards they offer are not sufficiently high. By extension, such firms are likely to shrink or go bankrupt, or to be forced to move to other more permissive jurisdictions, if they wish to avoid such a fate. Countries introducing legislation forcing the industry to dampen wages may similarly experience a painful and perverse outcome, losing revenues and employment and the economic benefit of financial intermediation services. Is this a reasonable portrait of the likely consequences of the proposed reforms to the banking industry?
In the labor market described by standard economics textbooks, workers are a homogeneous and interchangeable ‘factor of production’ which competes on wages to be employed. Similarly, there is intense wage-based competition by profit-maximizing firms, within a given industry as well as across different industries, to attract workers. In such an environment, workers are paid what they are because that is what is necessary to attract them, and because it is profitable to do so at that wage (in view of their incremental contribution to output). A wage ceiling which makes it impossible for a specific firm or industry to attract workers will (if there are no reasons to assign social valuations to the goods produced which are lower, or to the inputs used which are higher, than the market prices for them) create a ‘wedge’ which obstructs efficient decisions and reduce the economic output of the society. This is the conceptual picture which defenders of current compensation practices in the banking industry have in mind.
The standard portrait does not seem to provide a persuasive portrait of wage determination in the global banking industry for a number of reasons. First, wage differentials have been and remain very large, between the financial industry and other industries, including many which appear to employ persons with broadly comparable skills (such as MBAs from prestigious business schools). Second, wage differentials are so large that it seems highly unlikely that they are needed on the observed scale to bring about participation or to elicit effort. Other industries also involve long work hours and employ highly talented and educated persons. Indeed, many such industries appear to employ skills (e.g. engineering) which require even more specialized training and arguably involve greater rarity than the “generalist” pool from which finance professionals have been heavily drawn.
There are other ways of thinking about wage determination which may be pertinent to the case of the banking industry. The financial industry appears to benefit from large “rents”, or supernormal returns. These may result from barriers to entry based on the need for large initial pools of capital, the hoarding of specialized knowledge generated from past experience or position of centrality in the market, and other factors. In such an industry, wages will typically include a component of rent sharing. The sharing of the spoils in order to maintain an internal “moral economy” which is conducive to success reflects the desirability to the firm of avoiding the disruptions to production which result from employee turnover or non-cooperation. In large partnership-based firms such as Goldman Sachs, the workers, or at least the elite among them, are prominently and explicitly among the rentiers.
In an industry in which rent sharing is present, individual firms may well face the necessity to compete fiercely for specialized workers with other firms in the industry. Competition within an industry and the presence of rent sharing in an industry are entirely possible to reconcile. It can be true at one and the same time that global investment banks compete fiercely for available workers, and that those workers are collectively paid more than they need to be in order to bring about their participation in the industry and to encourage them to work effectively. It follows that an individual firm may have great difficulty in introducing wage restraints in such an industry, but the industry as a whole may be able to do so without loss to its productive capabilities. It is for exactly this reason that recent moves to bring about wage restraint in the financial industry, as an object of national and indeed global public policy, may be justified.
Global investment banking is an industry in which pure shareholders often (as in the case of Goldman Sachs) play a smaller role in decision-making than the elite worker-proprietors who run the banks and whose wages account for its primary costs. It is hardly surprising that the bankers who are the primary beneficiaries of this system should vociferously claim that wage restraints will destroy their ability to “create value”. It is an empirical question how far relative wages can be lowered without causing a reallocation of workers from banking to other industries. A coordinated lowering of relative wages need not, moreover, have any substantial impact on the competitiveness of individual firms or the viability of specific geographical locations of the industry (especially if, in light of the globalized nature of the banking industry, such restraints are globally coordinated). The goal of such a policy, once the fever pitch has been lowered, is to turn banking into an “ordinary” industry.
Insofar as the financial sector has grown “too large”, or its products are in part socially harmful, public policies which incrementally direct valuable resources from it toward other sectors may create rather than destroy social value. However, there are more direct and effective means than wage restraints of establishing a healthier relationship between the financial and the real economy. There is a respectable case for wage restraint, but if it is conceived of as mere ‘leveling down’ it attacks the symptom rather than the cause. Changing the qualitative structure of bankers’ incentives so that they avoid certain actions and engage in others is more promising. There is, however, no substitute for addressing the true root cause: supernormal returns to financial speculation. Far better to lower the spigot than to skim the froth.
Sanjay G. Reddy, Associate Professor of Economics, The New School for Social Research