Somebody may well haveÂ written on this already. If so, I apologise. But I haven’t read anything on the topic,Â so perhaps it’s new to you too.
Flipping through the HILDA dataset on the weekend (as you do), I stumbled upon an interesting variable. In 2004, the survey asked respondents who were unemployed or not in the labour force what was the lowest hourly wage they would work for. Below, I’ve just drawn a kernel density plot of those wages (omitting those aged under 21, who might earn youth wages).
The red line denotes the federal minimum wage prevailing at the time of the interview ($12.30 per hour). Pooling unemployed and not in the labour force (948 people),Â 24% of respondents said they would take $12 per hour or less – in other words, they effectively said that they would work for less than the minimum wage.
But for those not in the labour force, perhaps this was cheap talk? They weren’t actively looking for a job, and therefore the question might not have meant much. So the graph below restrictsÂ the sample to just the unemployed (237 respondents). Among this group, 23% of respondents effectively said they they would work for less than the minimum wage.
Of course, there’s a few problems with taking this kind of analysis to mean that abolishing the minimum wage would cut the unemployment rate by a quarter. Among them:
- Responses are coded in whole dollars, and quite a lot of those who cited a reservation wage below the minimum wage gave a number that was just below the minimum. For example, only 11% of the unemployed (13% including those not in the labour force) said that they would work for $10/hour or less.
- Economists prefer to trust revealed preference (what people do) than stated preference (what people say). Perhaps those who said that they would work for less than the minimum wage were saying what they thought the interviewer wanted to hear. Or they may just have made up a random number (this is what the unemployed guy who told the interviewer he wouldn’t work for less than $200/hour seems to have done). Or maybe it’s a spur of the moment thing, and if the putative subminimum wage workersÂ had done their sums, they would have realised that they couldn’t make ends meet on that kind of wage, and would not really have accepted a job that paid $10 per hour.
- It’s easy to flip things around. For example: “You mean to tell me that three-quarters of the unemployed wouldn’t take a job below the minimum wage?Â So who in their right mind would reduce it?”Â
- Employment isn’t all we care about. If employers have hiring power (what economists call monopsony power), then lowering the real minimum wage will reduce the wages of low-wage workers. This needs to be traded off against any potential employment gains.
- Even if a lower minimum wage would boost employment (which in my viewÂ must be right), we may regard very low wage work as unconscionable. In the same way that we set occupational health and safety standards, we may not want to live in a country that allows people to work for very low wages. For example,Â 1% of respondents said they would accept $5 per hour, which many might regard as exploitative inÂ modern-dayÂ Australia.
Anyhow, just thought I’d throw this out there. I’d be interested in your reactions.