What Proportion of Wage Differentiation is Just?

Ciarán Ó Ceallaigh (Senior Sophister)

Should some people be paid more than others? This seems a question central to economics and yet it is not one easy to answer. Ciarán Ó Ceallaigh tackles the problem in this essay, examining the theoretical arguments in favour of wage differentiation and evaluating their practical relevance.


That there is variance in the amounts people are paid for supplying their labour is an obvious fact of economic life. This is not only true of market economies, but of all economies. There seems to be a consensus among economists that wage differentials are a good thing. If everyone was paid equal amounts, what incentive could be used to persuade some people to perform less pleasant tasks? Equal pay would effectively reward the lazy, as hard workers would receive no extra income in return for their efforts. Such objections would obviously make a critique of wage differentiation fruitless. However, the story does not end there. We must ask why people are paid different amounts. Having explained the main reasons for wage differentiation - and having worked towards a definition of 'just' - we can examine whether each of four reasons is a just reason for wage differentiation. We will then be much better placed to assess the ultimate justice of wage differentiation. It must be added that such an endeavour necessarily involves a degree of value judgement. As such, economics being a human science, we should not expect concrete conclusions. Such is the risk with going beyond observational discussion.


As an ambiguous moral value, the notion of justice has been under scrutiny for millennia. If we wish to arrive at a working definition, therefore, which we shall adopt as an assumption, we would be wise to narrow our search down to the question 'what is a just reason for wage differentiation?' It may be argued that the answer to this question involves two concepts, contribution and equality of opportunity.

The principle of contribution seems quite self-evident. Put simply one applies the maxim 'to each according to his contribution'. As such, it would be just for those who give more in terms of both production and utility to the employer and society respectively, to be paid more. This seems to be quite simple. However, according to John Rawls,

'[t]he marginal product of labour depends upon supply and demand. What an individual contributes by his work varies with the demand of firms for his skills, and this in turn varies with the demand for the products of firms. An individual's contribution is also affected by how many offer similar talents. There is no presumption, then, that following the precept of contribution leads to a just outcome unless the underlying market forces, and the availability of opportunities which they reflect, are appropriately regulated.'

What is Rawls saying here? It could be argued that he is saying that a wage system set according to contribution and efficiency may not be just unless such a competitive system allows for both choice of occupation and equality of opportunity.

So, how does our principle stand at this point? A difference in two wages is just if it is consistent with the principle of 'to each according to his contribution', allowing for certain demand and supply factors and for the principle of equality of opportunity. As 'contribution', in its relevant sense, is dependent on the scarcity of the relevant skill, the market dynamics must not be discounted. What, however, is equality of opportunity? Put simply, it means that '[I]n all sectors of society there should be roughly equal prospects of culture and achievement for everyone similarly motivated and endowed. The expectations of those with the same abilities and aspirations should not be affected by their social class.'

We have, it seems, arrived at something of a working definition. More coherent in its negative sense, an unjust reason for a wage difference is one that is in breach of the principle of 'to each according to his contribution' and/or the principle of equality of opportunity. This excludes any direct association of justice with efficiency. An efficient wage is to be found at the equilibrium point - the intersection of the demand and supply curves for labour. A just wage may or may not be at this point, though, as has been pointed out, the notion of 'contribution' is, to an extent, a market phenomenon.

It is probably regarded as quite unusual for an economist to write on justice- normally the reserve of the political theorist. Such a situation however, is merely unusual, not unknown. In reality, justice is an indefinable, even non-linguistic, concept. It is an emotional principle upon which the preservation of a community or society is based. The definition given above is quite narrow. In essence, we are comparing wage differentials to two basic assumptions economists make about society. We assume that it is fair, and we also assume that greater 'expenditure' of labour should, more or less, reap a greater return. It is not incorrect to describe such assumptions as 'justice'. Nor is it the full story of the moral philosopher and political theorist.

Basic Foundations

It is quite obvious that if every person and job were the same - if workers and jobs were homogeneous - wage levels would be uniform. It is also quite obvious that this is not the case. Jobs and workers are heterogeneous. The heterogeneous nature of jobs leads us to one reason for wage differential. If we think of a wage as compensation for a worker, i.e., for the loss of leisure time, it seems quite reasonable that some jobs require greater compensation. For example, some jobs may '...involve disagreeable working conditions - dust, noise, extremes of heat or cold, risk of injury, heavy physical effort - while others have light, clean, and pleasant surroundings.'

So some wage differentials may be a result of the work environment. The premium paid over the homogenous wage due to work conditions is known as the 'compensatory wage differential'. Is the unpleasant nature of a work environment a just reason for paying different wages? It seems that it is. By no means does it breach the principle of equality of opportunity. It could also be argued that, by performing an unpleasant task, a worker is contributing relatively more. As such, it seems that an employer would be justified in paying a compensating wage differential. The compensating wage differential, however, in no way accounts for wage differentials on the whole. Such differentials may generally be concentrated amongst unskilled workers. Even there, the evidence is not concrete:

'...studies generally find, for example, that wages are one-half to 2 percent higher for workers in manufacturing industries with the average risk of job fatalities (about 1 in 10,000 per year) than for comparable workers in industries with half that level of risk. If the death risk data are collected and correlated with wages by occupation instead of industry, however, the theory finds less support.'

As such, it would be best to look elsewhere.

Human Capital

According to P. T. Bauer, '...people are not equally motivated or endowed in their economic attitudes and attributes, and this explains substantial economic differences between them in open societies.' While not completely untrue, such reasoning could be seen as quite crude. There are many more concrete reasons for wage differentiation. Marvin Kostersof the American Enterprise Institute in Washington D.C. attributes roughly half of the variation in earnings to educational level and work experience.

In Ireland, the Economic and Social Research Institute stated, in a report of the 11th January, 1995, that a person with no qualifications is five times more likely to be 'poor' than a person with a Leaving Certificate. It also revealed that the risk of poverty was twenty times greater for the unskilled worker than for the professional or managerial worker. In the basic model, human capital is built up through a procedure of investment: '[w]hen a person makes a current expenditure on education or training, it is anticipated that one's knowledge and skills and, therefore, future earnings will enhanced.'

To put it simply, one incurs costs in the hope of reaping future benefits. Such hopes have a solid basis. For example, in the United States wage increases for college graduates out-paced those of high school graduates by over 15% between 1979 and 1988. The same is true of on-the-job training. One incurs a short-term opportunity cost, in the form of a wage reduction, thus opening up the opportunity to earn high wages in the future. Such a system is true more often of general, as opposed to specific, training. General training gives the worker skills that may be applied outside the relevant firm. As such benefits accrue not only to the employer, but to the employee as well. Wage reduction, therefore, is a means of sharing the costs between the two beneficiaries. Specific training, on the other hand, leaves one with skills that are useful solely to the employer. As this prevents the worker from gaining a return, in the form of higher wages, from the training, the employer should logically bear the brunt of costs, i.e., should not reduce wages.

While based on a very simple principle, human capital theory is highly relevant to the explanation of wage differentials. However, when applying the principle of justice to human capital theory, it may be wiser to examine what McConnell and Brue call a 'modified human capital model'. Having established that greater schooling, etc., leads to higher wages - and, by implication, a greater divergence in schooling levels leads to greater wage differentiation - we should ask what causes some people to invest more in human capital than others.

'Many economists believe that we can better understand why the earnings distribution is skewed rightward by expanding the human capital model - that is, by taking a multifactor approach to earnings distribution - to include additional factors such as (1) ability, (2) family background.'

Put simply, ability is the inherent capacity to utilise one's human capital. It is the ''power to do'. Economists who emphasise the significance of ability believe that differences in the attainment of human capital and the returns accrued can be reduced to differences in ability. So, for example,

'[p]eople who possess higher intelligence are more likely to choose to attend college than those with less intelligence. Even if these highly intelligent people did not go to college, they could be expected to have larger earnings than less intelligent people who did not attend college. In other words, if we could somehow control for the skills and knowledge gained during college, this high ability group still would have substantially higher earnings than their less able counterparts. Consequently, according to this view, much of the inequality of earnings normally attributed to differences in education and training is actually the result of differences in ability.'

A great deal of research has been performed into ability. Scholars have attempted to compare ability to earnings, holding such things as education equal. However most have come to the conclusion that the problem of 'ability bias' is small. It could be argued that the primary factor in family background is motivation. A child from a more wealthy family is much more likely to demand higher education than children from less wealthy families. A major aspect of the intergenerational poverty trap seems to be the lack of motivation to acquire human capital on the part of children who come from less wealthy backgrounds. Such people also don't share the financial advantages or contacts of the wealthy. It may even be the case that malnutrition (as opposed to undernutrition) may affect the ability of poor people for upward mobility. As such, earnings differentials may be a result of human capital gained through amongst other an inter-related mixture of ability and family background.

So, in this light, is human capital a just reason for wage differentiation? Being more of a sociological than an economic phenomenon, this question is very difficult to answer. However, as human capital differences account for in or around half of wage differences, it should be attempted. Ostensibly, it seems, rewarding people for having greater human capital is quite just. It is most definitely not in breach of the principle of contribution. People with higher skills are more productive. It does not seem to contravene equality of opportunity either as, at its basic level, everyone has the chance to increase their stock of human capital. However, if the above is true, and family background has an effect on acquisition of human capital, it is in breach of equality of opportunity. It means that, in part at least, one may have a greater wage by virtue of the fact that one was born into a wealthy family. The question then becomes one of degree. To what extent is family background relevant? If it is the greater reason for differences, then human capital is an unjust reason for paying differing wages. On the other hand, however, financial constraints are reduced by state intervention in the form of grants, scholarships, and student loans, all of which reduce the supply price for this group.

In Ireland the cost impact of education is lessened by a structure of grants and subsidies. In the Programme for Competitiveness and Work, the Social Partners promised '...a focusing of resources towards the disadvantaged', in terms of education. Nevertheless, costs still seem to be high, and the wealthy also appear to have an advantage. So, again, is human capital a just reason for wage differentiation? All that can be said is that it is probably more just than unjust. As an investment, one should reap a return. However, it may be the case that there may be inequalities in who gets to make the investment. It seems that to know for sure we would have to answer some virtually unanswerable questions. What proportion of those with little in terms of human capital are like that due to lack of ability, and what proportion due to lack of motivation? Of these, what proportion lack the ability or motivation due to underprivileged backgrounds?


Discrimination may be given two definitions. Firstly, if one person is paid less than another person of equal ability for doing the same job, that person is discriminated against. In the second place, if a group in society, howsoever they may be defined, receives a per capita income that is smaller than that of a majority group, usually white males, we question whether that group is discriminated against. For example, in 1990, according to the American Bureau of the Census, white females earned 65 percent of the income of white males, while black females earned only 58 percent. This is despite the fact that '...the wage differential between continuously employed men and women is closing over time for almost every race-schooling group.' In order to discern the just of discrimination, we shall have to examine two aspects of the phenomenon: wage discrimination, i.e., what has been cited above, and human capital discrimination.

It is quite obvious that women are paid less than men. However, is this a result of unjust prejudice? Partly, but that is not the whole story. As a result of the traditional status of women in society, family obligations, while not dissuading women from working, 'may influence their choice of occupation and hours'. This narrowing of women's occupational horizons seems to lead to the distribution of the occupations of women being skewed towards those which are low paid. The narrowing of horizons, however, is not the only reason. Women seem to be endowed with less human capital than men: '[t]he advantage which white males have enjoyed compared to white females and blacks in obtaining a college education has been magnified through the greater access these white males have had to postmarked jobs training which has increased their productivity and earnings.'

However, differences in human capital stocks, no matter how much a result of prejudice, are not sufficient to explain discrimination. We must look at segregation between careers as regards women: highly paying professional careers can be over 90% male, while low paid clerical careers can be over 90% female. Here, both men and women have invested in their stocks of human capital, but the women have, or have been, focused towards careers where the return on human capital is relatively low. How much of this is a result of preferences and how much is a result of discrimination, it is impossible to say. However, the problem is serious enough for some analysts to suggest programmes of 'affirmative action', either give the disadvantaged a foothold in a previously discriminatory environment or explicitly promote such groups through the use of quotas. Though difficult to quantify, discrimination does exist as a reason for wage differentiation. As such it could be said to be totally unjust. It is a violation of both the principle of equality of opportunity and the principle of contribution. As such, many would argue that such intervention in the labour market as quoted above may be justified. That, however, is for policy-makers to decide.

Labour Unions

Among economists, it is generally agreed that labour unions increase wage differentials between members and non-members. If a union negotiates a wage increase for its members, the employer's demand for labour will fall. When this happens, workers who would have been demanded by the employer under the old wage, swell the ranks of jobs-seekers, thus putting downward pressure on non-union wages. So, when union wages rise, according to the theory, non-union wages fall. Indeed '...between 1979 and 1984 union wages rose 5.1 percent more than non-union wages in manufacturing and 5.6 percent more in non-manufacturing industries.'

An interesting aspect of this is the public service trade union movement. Here, pay increases have more far reaching effects. In a report by George Lee, it was shown that wage levels since 1987 in the Irish public service have '...far outstripped both average industrial earnings and inflation over the period.' That is, wage levels increased by just under 7 percent while inflation lay at just over 3.5 percent per annum. Since it is very difficult to perform rationalisation in the public sector, demand for labour cannot fall. The wage increases, therefore, must be paid. However, a greater differentiation in wages does not come about in this direct manner only. As the wages must be (eventually) paid for out of taxes, taxes must increase to pay more. This means that net incomes will fall, thus increasing the wage differentiation. If taxes are not increased, public services must be cut back, thus, in theory at least, decreasing the welfare of the people. Although, where the effects of unions are concerned, there is much evidence that the tendency for earnings distributions to widen is offset by other factors.

Let us examine the justice of union-related wage differentials. It is quite obvious that the equality of opportunity principle is not breached, there are few restrictions on who can join a union. However, what about the contribution principle? Do union members contribute sufficient amounts to justify higher wage increases? Probably not. They may contribute more, often being skilled, but their contribution might not rise sufficiently to justify larger increases in wages. So, we may be able to say that, it is just that they receive higher wages, but unjust that they receive higher wage rises.


So, is wage differentiation just? We have discovered that stock of human capital and compensatory payments seem at least to be more just reasons than not for wage differentiation. On the other hand, discrimination is in most cases unjust. Union intervention is unfortunately probably too complex to assess. As such, it seems that wage differentiation is in general just, given that stock of human capital and compensatory payments probably make up half or more of the reasons. However, if it is just, it is not perfect. Many might recommend care to be taken through minimising unjust wage differences. The crucial point is to find a balance between intervention and interference. We must, it is argued, maximise the advantage s of the market, but minimise the inequities created by the people in the market. In short, we must moderate ourselves to searching for a golden mean.


Rawls, John A Theory of Justice, Oxford University Press, Oxford, 1973

Reynolds, Lloyd G., et al Labor Economics and Labor Relations, Prentice-Hall, London, 1991

Ehrenberg, Ronald G. and Smith, Robert S. Modern Labor Economics, Harper Collins, New York, 1994

Bauer, P. T.Equality, the Third World and Economic Delusion, Methuen, London, 1981

Kosters, Marvin H. 'Schooling, Work Experience, and Wage Trends', in American Economic Association Papers and Proceedings (Vol. 80, No. 2, May 1990), American Economic Association, Princeton (NJ), 1990

McConnell-Campbell, R. and Brue, Stanley L.Contemporary Labor Economics, McGraw-Hill, London, 1989

Programme for Competitiveness and Work, Government Publications, Dublin, 1994

Light, Audrey 'Gender Differences in Wages and Job Turnover Among Continuously Employed Workers', in American Economic Association Papers and Proceedings (Vol. 80, No. 2, May 1990), American Economic Association, Princeton (NJ), 1990

McClelland, Peter DThe American Search for Economic Justice, Basil Blackwell, Oxford, 1990