Do minimum wages cause unemployment?
The Productivity Commission intends to find out. In the words of its chief, Peter Harris, it wants to know “whether or not there is an impact from the minimum wage on employment – we will try and prove up that, or determine if it is a myth.”
This is quite ambitious.
The minimum wage is one of the most contentious issues in economics. This issue has been banged around since at least the 18th century. The broader question of whether imposing price floors reduces supply is centuries older again.
But it is obviously true that any sufficiently large minimum wage above the market price will lock workers out of the workforce.
Imagine we doubled the minimum wage from its current $16.87 an hour to $34. Employers would shrink their workforces and only hire people whose productivity could justify the new cost.
Still some doubts? Imagine tripling the minimum wage. Quadrupling it. Make it $168.70 an hour. Of course people would lose jobs.
Labour markets are markets. They are governed by the impersonal, amoral forces of supply and demand.
Yet our Fair Work Commission thinks “modest minimum wage adjustments” have a “small, or zero, effect on employment”.
Small or zero? How small is small? How modest is modest? It is obviously true that as a minimum wage increase approaches zero its unemployment cost will approach zero as well.
In Fairfax papers on Saturday, the economics writer Peter Martin argued there was little evidence that the minimum wage costs jobs.
Martin cited the most famous paper on the minimum wage in the last few decades – a 1993 study by the economists David Card and Alan Krueger. Card and Krueger looked at a minimum wage increase in New Jersey in 1992 and found a) the minimum wage didn’t cause unemployment, b) it actually increased employment, and c) it increased it by a lot.
Card and Krueger’s paper has become one of the most influential papers in modern economics. But it’s not the only study done on minimum wages. There’s much evidence that points the other way.
For instance, this paper from 2014 found that American minimum wage increases during the late 2000s increased the unemployment-to-population ratio by 0.7 percentage points.
A 2012 study that looked at 33 different countries between 1971 and 2009 found raising minimum wages “reduce employment levels amongst young people and those at the margins of work”.
This 2011 paper finds that minimum wages cause employers to favour young workers from more privileged households than less privileged ones.
This study concludes that the minimum wage hurts job growth over time, a burden that falls most on young workers and low-wage industries.
And in a 2003 paper, Australian economist-turned-politician Andrew Leigh also found small but real unemployment costs of the minimum wage.
We could go on, but ideally Drum columns should not just be lists of journal articles.
It’s true that for all the studies that find the minimum wage causes unemployment in the short or medium term, there are some studies that disagree. This is not a surprise, for a few reasons.
First, much of the research has been done in the United States, which has famously low minimum wages. American minimum wages are probably very close to the wages that would prevail in the open market, so they can’t distort employment all that much.
Second, when looking at minimum wage increases, we’re talking about very small changes to prices in very complex systems. Disentangling what policy change causes what variation in employment – particularly over the course of years, when there can be lags and broader economic changes – is incredibly difficult. Measurement is hard. Determining cause and effect is even harder. Welcome to economics.
And third, the cost of minimum wage increases might not show up in reported employment or wage data, but still could be worn by employees nonetheless.
For instance, employers might reduce conditions to compensate. They might save on training. They might spend less on heating the workplace. They might reduce non-monetary benefits. These costs are hard to measure, but they’re very real. (This paper from the US-based National Centre for Policy Analysis details those non-monetary costs.)
Despite these challenges, surveying the broad evidence in their book Minimum Wages, the economists David Neumark and William L. Wascher conclude that minimum wages are a “relatively ineffective social policy for aiding the poor”:
They entail disemployment effects that are felt most heavily by low-skilled workers. They discourage human capital formation. They lead to price increases on products frequently consumed by low-income families. And, on balance, they seem to do little, if anything, to raise the incomes of poor and near-poor families, and more likely have adverse effects on these families.
Of course, it is possible to accept that minimum wages cause unemployment at some margin but still support them, under the belief that the social security net should catch people who are kicked out of the workforce as a result.
But as I argued in the The Drum last month, our actually-existing safety net is a hotchpotch of paternalism and bureaucratic restriction.
Imagine how bad it will be if the Abbott Government legislates its no-welfare-for-six-months policy. Young workers unable to find work at the minimum wage will also be ineligible for the dole. This is a recipe for destitution.
It’s true that minimum wages are popular. So were housing rent controls and trade protectionism.
One day, hopefully, the Australian public will realise that by preventing the most vulnerable Australians from getting a foothold in the labour market, the minimum wage is creating the very poverty trap it is supposed to alleviate.