Dick Nichols
Federal government submissions to the last three safety net cases before the Australian Industrial Relations Commission (AIRC) have cited surveys here and overseas that typically claim that every 10% increase in the legal minimum wage reduces employment by between 1% and 6%. Is this really true?
At first glance the argument seems to fit the most basic concept of economics — if the price of a commodity (including labour power) rises, demand for it will decrease. Surely it's clear as day that if the bosses could pay $6 instead of $7.50 an hour for labour they would take on more workers. Except that it isn't.
Far from being a relation that's agreed on by all economists, the impact on employment of minimum wage movements is a hotbed of dispute. It involves controversy over effects within specific job markets ("microeconomics") as well as at the level of the economy as a whole ("macroeconomics").
The latest round in the dispute goes back to the early 1990s, when two US economists, David Card and Alan Krueger, found through a study of the New Jersey fast food industry that an increase in the legal minimum wage (from US$4.25 to US$5.05 an hour) led to an increase in employment in the industry.
On the basis of this and other studies, they concluded in their 1995 book Myth and Measurement — the New Economics of the Minimum Wage that within a certain range, an increase in the US minimum wage has no or a slightly positive impact on employment.
Two reasons
Card and Krueger gave two basic explanations for their findings. One is based on the assumption that in many job markets a single or only a few buyers of labour-power operate, i.e., the buyers have monopolistic power in the market.
For example, suppose there is in Deadend City only one pizza restaurant ("McPizza") to hire pizza workers. McPizza's monopolistic position enables it to drive wages down below what would be the case if it faced competition for workers from other pizza restaurants. If unemployed pizza workers don't like McPizza's wage offer of, say, $4 an hour, tough luck: They have nowhere else to go.
Suppose too that, if McPizza did face competition, it would have to offer pizza workers $5 an hour: the productivity of the workers would be still enough at this rate to make all pizza restaurant owners a nice profit. Yet, precisely because McPizza is the only game in town and can enforce a wage rate of $4 an hour, on top of this acceptable profit, it makes a super-profit of $1 an hour for every worker employed (and it probably rips off its customers as well).
Next, suppose that McPizza is expanding and needs to employ more workers. To attract them it will have to offer a higher wage, and this wage will have to be paid to all its workers, old and new. If a legal minimum wage were set at $4.50 an hour, McPizza would take on extra workers because it would still be making a super-profit (now of 50 cents instead of $1 an hour per worker).
Clearly, this is an argument for minimum wage regulation. If job markets are basically monopolistic, within a certain range both wages and employment can be increased by fixing a legal minimum wage (or through trade union action). The combined wage and job growth gets paid for out of the superprofits of the monopolistic employers.
Card and Krueger's second main argument is that an increase in the minimum wage "shocks" workers into greater commitment to their jobs, raising company productivity. If this "shock" is great enough, an increase in the minimum wage need not reduce employment.
Not surprisingly, given the corporate interests involved, both these arguments as well as Card and Krueger's statistical methods have come under heavy fire. Argument rages over whether and which labour markets really are subject to monopolistic practices (and how to prove that), over Card and Krueger's supposed lack of attention to the long-term impact of wage increases on job growth, and over the difficulties of weeding out other influences on employment besides wage movements.
With regard to the supposed "shock" impact of a minimum wage increase, a question frequently asked is why employers would be so short-sighted as not to pay higher wages themselves if these lead to higher productivity and profits.
The AIRC's last safety net decision looks at "the literature" of this controversy in some detail. It concludes that "substantial safety net adjustments may have some negative effects on employment in those sectors of the economy in which a high proportion of the workers are award-reliant [but] the material brought to the commission's attention does not establish an empirical basis for affording greater importance to concerns about employment effects than to other considerations to which we must have regard".
Predictably, however, the AIRC's outburst of theoretical rigour didn't translate into courageous practice. The country's low-paid workers were awarded a miserly $17 rise — even less than the $20 suggested by the ALP-governed states and territories!
Investment ignored
Whatever their conclusion, one disabling shortcoming of most studies of the "wage-employment relationship" is that they usually treat this connection in isolation from other economic influences, assuming "other things being equal" when they nearly always aren't.
Most importantly, they tend to assume that the rate of investment, the biggest driver of employment growth and one through which the wage-employment connection always operates, is "given". It's a fact of life basically determined by the capitalists' profit expectations and one not to be touched by any other social force.
The authors of study done by Monash University's Centre of Policy Studies to back up the federal government's argument in this year's minimum wage case admit that "despite increases in wage rates, employment can grow... [I]mprovements in productivity, increases in the terms of trade and reductions in required rates of return on capital all act to allow increased wage rates to be compatible with employment growth."
An historical examination of the wage-employment link reveals all the possible variants. A recent example is a 2003 World Bank study that showed that minimum wages increased sharply in Indonesia between 1990 and 1996. Yet, in the clothing, textile, footwear and leather industries, while there was some evidence of job loss for small domestic firms, there was none for large firms, foreign or domestic.
Why did this happen? For the same basic reason that real wages growth in the 1950s and '60s in Australia and other developed capitalist countries was accompanied by steady employment growth.
In the 1950-60s long boom, the rate of investment in new equipment (rate of scrapping of old equipment) was so high that rising real wages boosted employment; in the 1930s Great Depression, by contrast, it was so low that wage cuts simply reinforced the collapse in total demand, accelerating the downward spiral of job loss.
What about less extreme scenarios, such as the past 14 years in Australia? Here the underlying source of employment growth has been the continuing strength of business investment.
Did the wage restraint operating throughout this period (and since Labor's 1980s Accord) operate to increase employment even further? There's evidence that it did, but there's also a plausible case to be made that it could simultaneously have acted to cut potential job growth. That's because, in the face of wage cuts, employed workers maintained and improved their living standards by working extra hours. Given a decent pay increase or a serious union campaign for a shorter working week, those millions of overtime hours could have been worked by new workers.
All this proves is that once it is accepted that economic growth and jobs are dependent on private investment and "acceptable rates of return on capital", then we're trapped into focusing on wage rates as the main influence on jobs growth — the employers have got us thinking inside their square.
At bottom, the whole debate about the minimum wage ("either wages or jobs") is a smokescreen to cover the "race to the bottom" embodied in Prime Minister John Howard's anti-union plans. The union movement shouldn't accept these things as "given" either in theory or practice. The main driver of jobs is investment. If the labour movement wants to have a serious employment policy of its own it has no choice but to develop its own proposals for socially useful and environmentally sustainable investment.
A useful start would be serious union opposition to further privatisation of the country's remaining public assets — investment and jobs are far too important to be left to the corporate boardrooms.
[Dick Nichols is a member of the Socialist Alliance and managing editor of Seeing Red.]
From 91×ÔÅÄÂÛ̳ Weekly, June 22, 2005.
Visit the