P.P. McGuinness, “High wages do not help the young and the poor,” The Australian Financial Review, September 12, 1983, p. 3.

There is a strange reluctance among many members of the community, and indeed even among professional economists, to accept that there is a relationship between wage rates and levels of employment.

This is understandable enough in universities and similar institutions, where there is no direct relationship between levels of pay and numbers employed, since tax-financed educational institutions are not subject to the constraints of the market. However worthless their product, their incomes are not affected.

Moreover, the lack of connection between rates of pay and unemployment is a belief largely confined to Australia. Even here it is rapidly disappearing. A conference on inflation and unemployment held at Macquarie University last week brought together a number of distinguished economists from Australia and elsewhere, and all the papers accepted the reality of wage effects on employment.

This is not to say that wages are the only determining factor in employment, or even always the major factor. No one would assert that the present high unemployment in Australia was created solely by wage increases. But very few would assert that in the circumstances of declining per capita national income and declining international competitiveness, the sharp increases in real wages of 1981-82 had nothing to do with the increase in unemployment.

As well as the proposition that there is a causal relationship between real wages and unemployment levels for the economy as a whole, there are a number of related propositions. One is that minimum wages do the poor more harm than good. This is not of course an argument for making them poorer, but relates to whether relatively high minimum wages price unskilled labour out of the market.

The reductio ad absurdum of the argument that there is no connection between minimum wages and the employment of marginal workers is that if this were so, why not double minimum wages? If a hundred per cent increase in minimum wages must be assumed to have an adverse effect on unemployment, why not a 10 per cent increase?

But when it comes to analysing the impact of relative wages on different sectors of the labour force, there are great difficulties in deciding what the effects of levels of wages, and wage changes, are. This is partly because labour markets are very complex, and since the commodity is human beings, there is strong resistance to changes in customary practices. There is also rigid segmentation of some markets.

Thus, women quite correctly point out that there is much discrimination against women in employment in terms of traditional sex roles, so that women’s wages, which are generally lower than men’s, do not necessarily reflect any lower levels of value of the work done. The wages of nurses are a classic example of a rate of pay which was artificially depressed for many years by the fact that the work was almost entirely done by women, while the supply of female labour was maintained by the restricted range of alternative employment.

But this is not to say that there are not factors tending to depress women’s pay relative to men’s which would exist in a non-discriminatory, perfectly competitive labour market. The simple fact is that the majority of women bear children and have interrupted careers as a result. There is a strong social argument for compensating them for this economic disadvantage.

In a society where there is strong race prejudice, inequalities of pay may exist also because of market segmentation. There is considerable controversy in the United States about this, and the appropriate policies to deal with it. At least one black economist who is highly regarded in the profession, Professor Thomas Sowell, has concluded that measures of positive discrimination (“affirmative action”) do actual harm to the economic welfare of blacks.

In Australia, the effects of rates of pay for young workers on the employment of young people is a matter of controversy. The Bureau of Labor Market Research recently issued a report which pointed out that a connect between rates of pay and levels of youth unemployment could not be denied.

This was greeted by the Minister, Ralph Willis, as if it were a nasty smell, and a number of people have sought to discredit the report by saying that its conclusions were determined by the previous Government — an insult to the bureau and the many capable officials working in it.

It is not easy to analyse such an issue, however. It is complicated by supply-side effects (the fact that a rise in youth wages may draw more young workers into the labour market) and by the fact that, far from complete segmentation, the labour market is characterised by overlapping groups. Thus in some jobs there is competition between youths and married women.

A study just published by the Organisation for Economic Co-operation and Development (OECD Occasional Studies, No 45, June 1983), surveys the work which has been done on the effects of minimum wages on the youth labour market in North America and France, and allowing for differences shows that the Bureau of Labour Market Research study is certainly not outside the mainstream of serious economic examination of the problem.

The OECD survey of the North American studies on minimum wages draws two very interesting conclusions, with great significance for economic and social welfare policy. One is that the effect of minimum wages on youth employment is adverse, but not very much so.

The evidence is overwhelming that even though there is an offsetting supply effect on the youth labour market, rises in minimum wages do create unemployment, or, as one study puts it, “there is no convincing evidence to refute the prediction that minimum wages cause reductions of employment (for young workers at least).”

Incidentally, this is the proper way of framing the question. The presumption has to be that wages and unemployment are related and the job of the economist is either to quantify the relationship or investigate whether it can be refuted, and if so why. To argue that the assertion that there is no connection [that] has to be refuted is to stand the problem on its head.

But the North American studies do show that the net effect of minimum wages for young workers is quite small. In summary, “on the basis of these updated estimates it appears that on average a 10 per cent rise the minimum wage would result in a decline in teenage employment of about 1 per cent.”

Moreover, a substantial differential between youth minimum and adult minimum wages does not produce a substantial increase in youth employment while there is some less of adult jobs when young workers become cheaper and are used to replace adults.

So the American studies would not point to too high wages being the major cause of youth unemployment. However, they are one of the causes, and indicate that caution in increasing minimum wages is desirable.

But the other major conclusion of the OECD survey is even more important. It will be remembered that the major argument in favour of increasing young peoples’ wages is a welfare argument. But the American studies show that increases in minimum wages either have no effect in overcoming poverty for young workers, and the workforces generally, or actually make those on the minimum relatively worse off.

And this is even before any account is taken of increased unemployment as a result of increased minimum wages. One researcher “found that 23 per cent of adult low-wage workers were before his definition of the poverty line, but one-quarter of adult low-wage workers were in families with total incomes above the national median.” (The median being that figure which divides the range into two equal groups of equal numbers.)

“For teenagers,” the summary continues, “the median family income for low-wage workers was actually higher than that of high-wage workers. Thus much of the beneficial effects of a hike in the minimum wage would go to high-income families as well as low-income families so that the net effect on income redistribution would be relatively small.”

All this implies that if young workers do not have a satisfactory living standard, the best policy approach would be not to raise minimum wages, which would benefit many who do not need it and cause some increase in unemployment, but help them directory by way of direct income transfers.

That is, in general it would be more effective, less harmful, and cheaper to pay a social security supplement to the poor in employment, as well as the dole to those without employment rather than to increase minimum wages across the board for young people or adults.

For if there is one thing worse than having no social policy at all, it is having ill-considered social policies which actually do harm to the people you are trying to help.