Recently on Open Democracy, William Davies described how the concept of fairness has been decoupled from issues of class and power in the UK. New Labour's innovation had been to find a way to pursue something called fairness, yet simultaneously to disavow any “old Labour” attack on class. Now, the Tories have been able to make the concept their own in defending their programme of austerity: the pain will be shared by all. Fairness has become precisely that which doesn't change the imbalances of society.
This generated much debate: Should we try to reclaim the concept? Or cede it and focus on concrete realities, not abstract notions? Yet the Tories have always taken a position on fairness. In denouncing the “politics of envy”, the right has always argued that the status quo is fair. Focusing on concrete experience will also only get us so far. We have to challenge the common sense of “meritocracy” and “equal opportunities”; not the prima facie demand for social mobility that everyone agrees with, but the hidden implication that “success” is necessarily earned, and that the poor are responsible for their own “failure”.
The pre-eminent intellectual justification of the liberal, meritocratic ideal comes from neoclassical economics. Labour, according to this paradigm, will be employed until its falling marginal productivity (the way it creates wealth) equals its rising marginal cost (the disutility suffered by people who work). Capital will similarly be employed until its falling marginal productivity equals its rising marginal cost (the disutility suffered by the capitalist when he foregoes consumption to invest). This determines the capital intensity of production, and also the distribution of income between wages and profit. Wages are determined by the marginal productivity of labour; profit or interest by the marginal productivity of capital. Both workers and capitalists are thus seen as creating wealth that returns to them as fair compensation for the costs they incur.
This is an attractively simple theory, and serves as a neoclassical meta-narrative, explaining – and justifying – why the world is as it is. It is a way of making sense of the world, and in a way satisfying to many. But like many a meta-narrative, it fails the text of falsifiability. The problem, identified by Joan Robinson, Piero Sraffa and others (See Cohen and Harcourt, Whatever happened to the Cambridge Capital Theory Controversies? ), is that you can't actually use this theory to predict the rate of profit. To do so you would need to quantify “capital” which, in the real world outside of single commodity models, can only be done by price, which itself depends on the rate of profit you want to determine. Sraffa demonstrated the possibility that rising interest rates can coincide with rising capital intensity and thus falling marginal capital productivity (“re-switching”), bringing into question the whole notion that the relative productivities of labour and capital determine their share of the product. The neoclassical picture of value creation and distribution requires a leap of faith into the untestable and logically questionable.
An alternative tradition is represented by Ricardo, Marx, Sraffa and contemporary post-Keynesianism. These all reject the idea that capital can be said to be productive, noting that capital is, itself, a product of labour. Marx saw capital as capturing a form of monopoly rent. Similarly, in explaining the rate of profit, most post-Keynesians would probably see a role for scarcity rent from the degrees of monopoly many firms and individuals may enjoy (e.g. from intellectual property), and for marginal rent from scarce natural resources such as land. Nonetheless, the Keynesian position also draws on the effects of liquidity preference (interest as compensation for the possibility of not having access to your money when you might happen to want it), on top of compensation for risk (i.e. for average losses), to explain positive returns on capital investment. An overview of post-Keynesian views of distribution, as well as of neo-classical attempts to explain pay inequality, is given by James Galbraith at the University of Texas Inequality Project.
It is also on the dubious foundations of marginal productivity that neoclassical theory seeks to explain today's huge inequalities of pay. “Wages” are said to be proportional to personal labour productivity, and the productivity of “human capital”. Again, this is not testable, and it is impossible to point to a mechanism by which it could actually happen in the real world. The problem is that there is no way to empirically compare the productivity of individuals performing different tasks. In the case of a firm, there exists no way to quantify the relative productivity of employees when they are members of an interdependent team. They are simply not separately productive; executives create no revenue without the people they manage and vice versa. The self-employed can point to the revenue they produce as a putative measure of their productivity, but in fact this is an interpretation dependent on the correctness of neoclassical theory, not proof of it. Marxian and Sraffian frameworks argue that the direction of causation in pricing is the reverse; that prices reflect embodied labour costs.
So how is it that an executive can earn more than 300x times the average wage of production workers? What explains bankers' bonuses? For post-Keynesianism, these are still unresolved questions. James Galbraith suggest that differential pay rates between different sectors are determined by different degrees of monopoly, with knowledge intensive sectors enjoying greater market power. This may be important, but it's not clear that it can explain extreme wage differentials, not least within a single firm. I believe a more complete explanation needs to take the following factors into consideration:
Contracts for the supply and payment of labour are not (pace the neoclassical imagination) based on an assessment of actual productivity. Instead, a commitment to pay is entered into before the labour is performed. Purchasing a service is typically the same. Such decisions are thus taken in the context of uncertainty. The uncertainty concerns the future impact of the labour, whether it will be positive, negative, and to what degree. Hiring decisions thus involve risk management, in the sense of coping with uncertainty.
Jobs and tasks vary in terms of potential for positive or negative impacts. Such influence depends on factors such as number of subordinates, amount of money managed, the degree to which the employee's expertise is relied upon by others, or the degree to which a service purchaser's own wealth or well-being is at state. Influence is not another way of saying productivity; it represents opportunity to enhance or diminish production efficiency, wealth and well-being.
From an employer's perspective, the greater the influence of a position, the more it matters to get a good candidate to fill it.
Uncertainty as to the future impact of an employee or service provider is inevitable, no matter how well qualified a candidate may be. For this reason, no level of qualification (or experience, etc.) can be regarded as "good enough". Hirers thus compete for relative quality ("as good a candidate as possible"), not absolute quality. This is important, because while the supply of absolutely qualified candidates may be demand elastic (more can be trained), the supply of relatively qualified candidates, or "best candidates", is perfectly inelastic.
These points, if correct, would allow us to draw the following conclusion. Under capitalist conditions, the output of labour, even after the subtraction of profit, is typically in excess of the amount required to pay all those in employment the minimum wage or the marginal disutility of labour. This excess takes the form of effective demand for relatively qualified candidates, which is captured by these candidates as a form of scarcity rent in proportion to the relative perceived influence of the jobs they are hired to do.
Paying rents – whether for land, for investment capital, or for relatively qualified candidates in an attempt to cope with the uncertainties and risks of hiring – is rational for individual economic agents, yet irrational for society as a whole. It is yet another prisoner's dilemma game: each firm or individual has to pay the market rates because all the others pay the market rates, but collectively we lose from doing so.
In the case of pay inequality, the hypothesis outlined above allows us to envisage a scenario in which the minimum wage is much higher than today, and pay inequality consequently much reduced, but in which the same people would still be in the same jobs, creating the same wealth. The same effect is predicted for higher employment implying a higher marginal disutility of labour. Such a flattening of the pay hierarchy would, as Abba Lerner indicated , imply a huge efficiency gain in using available resources to meet needs. If it were excepted that income inequalities are overwhelmingly the result of rent capture, not relative productivity, such a scenario would also surely be seen as more fair.