Pay Statistics Tell the True Story
By Mark Wooden
The Australian Financial Review,
16 November 2006, p. 63
A few weeks ago the new Australian Fair Pay Commission (AFPC) raised the Federal Minimum Wage from $12.75 to $13.47 per hour, a rise of 5.6%.
This decision was received very differently by different sectors of the community. Employer groups were highly critical, claiming that the rise would be harmful for employment, while the ACTU was full of praise, believing that the decision vindicated their claims that wage rises for the low paid were necessary to protect their real living standards and would not be harmful for employment growth. So who has got it right?
The easiest claim to deal with is the often heard argument that minimum wage increases help the poor. The reality is that most of the poor are not employed and so, by definition, rises in the minimum wage can do little to help them.
The AFPC, however, appears to support the union view, stating in its decision that lower paid workers are relatively concentrated in lower income households. At best, this can only be described as an interesting interpretation of the evidence.
Data from the 2004 wave of the HILDA Survey, the same data source used in the research commissioned by the AFPC, reveals that the lowest paid are actually spread quite evenly throughout the income distribution. Indeed, the 20% of wage earners with the lowest hourly wage rates are actually underrepresented in the bottom part of the income distribution only 27 per cent of these employees are living in households in the bottom three deciles (i.e., the bottom 30 per cent) of the distribution.
The picture changes slightly if we focus both on adults (and thus exclude many children who are still living at home with their parents) and on the very low paid the bottom 10 per cent. The proportion of this group that are in the bottom three deciles of the income distribution now rises to almost 38 per cent, meaning they are slightly overrepresented in the bottom of the distribution. Much of this higher representation, however, can be explained by income support recipients who are able to work part-time while still receiving a government benefit. For most of these people the numbers of hours worked are relatively small. Further, beyond some very minimal income level, government income is withdrawn at the rate of either 50 or 60 cents for every extra dollar earned. Together, these two facts mean that minimum wage rises do relatively little to raise the incomes of these types of workers.
Even if low-wage workers were, at a single point in time, heavily concentrated in low-income households, it does not follow that such workers will necessarily benefit much from minimum wage rises. This all depends on how much they would have earned in the absence of minimum wage rises. We can never know this, but with the new longitudinal data HILDA Survey what we do now know is how much earnings for low-wage workers have risen in the recent past.
Specifically, the HILDA Survey tracks earnings growth for the same people over the period 2001 to 2004 and reveals an unexpected result. Earnings growth has been overwhelmingly concentrated among the persons earning the least at the start of the period. In fact, the median rate of growth in usual hourly earnings for employees who were among the bottom 10 per cent of wage earners in 2001 was a staggering 104 per cent. For the next 10 per cent the figure is much lower, but still a whopping 49 per cent. These rates of growth gradually fall until we reach the top of the earnings distribution, where earnings actually fell.
Provided they remain in employment, the vast majority of low-paid workers have experienced earnings increases that are well in excess of the rate of increase in the minimum wage. In fact, the HILDA Survey data suggest that 84 per cent of employees in the bottom decile of wage earners who were also in work three years later experienced earnings increases in excess of 13.2 per cent, and for most the rate of increase was well ahead of that. For most low-wage workers the minimum wage is simply irrelevant.
This, however, still leaves the unemployed and those low-wage workers who cycle between low-wage jobs and unemployment. For these groups the minimum wage may not be so irrelevant. The AFPC, however, (correctly) emphasizes that effects on aggregate employment depend on both demand and supply responses. But given our income support system, the incentive to work may not be great for some groups at the minimum wage. In other words, higher wages may be needed to induce some people to work. Nevertheless, this seems small consolation for those people (mostly young single people with few job skills) who want work but are unattractive to employers at the minimum wage.
MARK WOODEN is Professorial Research Fellow and Deputy Director, Melbourne Institute of Applied Economic and Social Research, The University of Melbourne.