Small truths about the minimum wage

Here’s another example of the “small truths, big errors” rightist rhetoric I wrote about recently: Sam Bowman says the minimum wage should be abolished to help the unemployed.
Now, unlike some on the left, I’m happy to concede that orthodox economics applies here, and that lower wages would create jobs. The question is: how many?
This in turn rests upon two questions:
1. How much would wages fall if the NMW were abolished? One possibility is that they’d drop a long way. There’s an excess supply of labour, and some workers would be willing to accept low wages, knowing that their incomes will be topped up by tax credits*.
On the other hand, though, the very fact that the NMW has existed has created a social norm that  £5.93ph is an acceptable minimum wage. Workers would feel insulted to be offered less than this. And employers would be loath to cut the wages of existing employees below this, fearing that it would lead to widespread shirking.
2. What is the price elasticity of demand for labour? One thing tells us that it can’t be very high - the introduction of the NMW in 1999 did not price many workers out of jobs. On the other hand, though, something else tells us that it might be high; the increased supply of labour created by immigration did not bid wages down much**.
Let’s stick some numbers on these competing alternatives. In the best case, wages fall 30% and the price elasticity of demand is 1.5. We know that minimum wage employment, at the last estimate, was 747,000. Multiplying through then gives us job creation of 0.3 x 1.5 x 747,000 = 336,150.
In the worst case, there’s a 10% wage fall and a 0.5 elasticity, giving us new jobs of 0.1 x 0.5 x 747,000 = 37,350.
With official unemployment at 2.45 million, this implies that abolishing the NMW would cut unemployment by between 1.5% and 13.7%. The mid-range of these estimates is only slightly greater than the sampling error (pdf) of the official measure of employment.
Now, you might object here that I’m ignoring the possibility that abolishing the minimum wage would also lead to wages falling for those paid more than the minimum, so more workers would be priced into jobs. However, the evidence suggests that the introduction of the NMW did not raise wages for workers outside the bottom decile (fig 2.15 here), so it’s hard to see why abolishing it would lead to lower wages.
The notion that abolishing the minimum wage will create jobs is, therefore, a small truth. To argue otherwise requires some heroic assumptions.
What’s the bigger truth that it’s hiding? Simple - that unemployment is a much bigger problem than can be solved by free market remedies alone.
* Deficit fetishists take note: abolishing the minimum wage would add to the deficit by increasing the tax credit bill.
** Illiberal rightists note: you can’t be an elasticity optimist when it comes to arguing for abolishing the minimum wage, but an elasticity pessimist when arguing against immigration.

Comments

[Gerard O'Neill]

The mistake that is always made by defenders of the minimum wage is that they think only of existing employers and their employees. The 'small truth' is that the minimum wage is a very welcome barrier to entry that prohibits new companies forming in, especially, labour intensive sectors.

By raising the cost of entry - along with all the other regulatory barriers etc facing anyone wishing to set up, say, a catering or hospitality business - this protects the profits of the incumbents. That's right: the minimum wage boosts the profits of established businesses through a powerful barrier to entry. An odd thing for someone on the Left to be advocating...

The other 'small truth': the people most adversely affected by the minimum wage are young, unskilled unemployed men in the main. By denying them access to the workforce through anti-competitive measures such as the minimum wage then the big truth becomes a generation denied the first critical step towards employment. It's the impact on youth unemployment that matters: not the overall unemployment situation.

I guess one person's 'small' is another person's 'big' when it comes to these things.

[Mark Wadsworth]

You're missing the obvious. It's the net wages that a worker cares about, not the gross. Even if they introduce the UC/SUT system touted by Iain Duncan Smith, that means a claimant only keeps £2.10 for each £6 per hour he earns (or whatever the NMW is going to be). Which is an improvement on the even smaller amounts he keeps under current rules.

If I were a low earner/welfare claimant, I would be indifferent between keeping 35% of £6 per hour or 50% of £4.20 per hour or indeed 70% of £3 per hour.

But to the potential employer this makes a HUGE difference, whether he pays £6 or £4.20 or £3.

We could do a compromise and e.g. reduce the NMW to £5 and the marginal withdrawal rate to 50% - hooray! Everybody wins! The worker is 40p per hour better off and the employer is £1 per hour better off - so more jobs become economically viable and people are more willing to work. It's the Laffer Curve in action, and all things being equal would reduce the net cash cost of welfare (UC paid out minus SUT tax paid).

The beauty of a 50% MDR is that the PAYE system is already geared up to deducting a flat 50% from wages with no personal allowance (inclusive of PAYE income tax and National Insurance), it's called a K-code.

What can possibly go wrong?

(Click my name to read all my blog posts that explain this in more detail).

[TheOneEyedMan]

If you got rid of the minimum wage we might see much more internships with travel and lunch stipends. By changing the structure of the pay we could see many jobs dropping below the hourly rate of the minimum wage but structured in a face saving way. A lot of desirable entry level jobs in the USA like fashion, music, and journalism are structured in this way.

[jaffa]

The main reason behind the introduction of the minimum wage was to protect the state from poor employers. Remember the burger flippers paid one pound an hour and clocked off when the cafe was empty but not allowed to go home. I recall a factory in wales paying 1.99 per hour (the state picking up the rest of the required living wage) whilst the multi millionaire owner swanned around in the south of France drawing a salary equivalent to over 1000 pounds per hour. So it really is not about workers but taxpayers.

[Curt Doolittle]

Good points in the comments above. But something troubles me.

How do we know that the opposite isn't true: that employers in at least some sectors, artificially reduce wages by forming a wage-price cartel using state-sponsored price fixing of minimum wages? Wages are also a form of data, and employees can rate a company based on it's wage offerings. Also, in many sectors, 'labor' in the sense of pure physical expenditure, is irrelevant, and other skills are not (literacy, attractiveness, manners) but those valuable skills are not paid for in wages because all related businesses can claim minimum wage barriers, and if they compete at all, do so by trading entirely on environment rather than wages?

It just appears that a cartel effect is in place, at least in the upper two quintiles. Has anyone studied that distortion? There are plenty of people who will work at certain companies simply in exchange for the environment (at a discount) rather than in sectors where the cartel effect drives down wages (less comfortable environments). In other words, they're compensated partly with education, partly with environment, partly with lifestyle.

Mexican "illegal" construction day labor on the west coast generally costs no less than $14 per hour. And if the buyer wants people who know concrete, it's as much as $20. Waitresses in popular restaurants can, and often do, earn hundreds of dollars per day - often more than in their 'day jobs'. Janitorial labor in office buildings costs well above minimum wage, even in our current economy. Retail clerks, ie: jobs with comfortable social, clean environments that require unskilled, easy labor, seem vulnerable to low wages. But thats the discount the employee accepts for the ability to work in such an environment.

I don't have a stake in this argument, it's not my area of expertise, but I'm not entirely confident that the cartel-effect created by the minimum wage (at least in some sectors) is not driving down wages in at least some areas of the economy. And I don't see any literature that grasps that much of the minimum wage economy is indeed the 'on-the-job-training in exchange for rate discount' ecosystem. Or 'I'm husband shopping with this job". Or "every other job is real work, and this makes my parents happy". Or "I'm doing this job that's easy while I"m in college." All of these are not meaningful if considered on wages as compensation alone.

What might be more interesting is experimenting with minimum wages in certain sectors in order to drive talent and 'social status' into them. I don't think its important for us to subsidize retail and fast food clerks. It's more important to drive more people back into skilled and semi-skilled labor.

For example, (to be a bit crass) if we're talking about single moms flipping burgers and waiting tables, or performing office work, (and we're accepting the fact that we aren't educating men and women on mate selection, and we're saying that it's acceptable for women to have children that they can't support, and therefore that it's acceptable for them to export their preference for childbearing, and poor choice of husband onto others) then why don't we simply wage-match people with such problems rather than distort the entire pricing structure?

I'm not advocating this but just wondering if there is any work in that field that isn't riddled with the typical errors so endemic in quantitative probabilism (rather than evidence) preferred in the field.

I'm sure there is quite a bit of data out there. I just am not sure that any paper that I've read on this topic relies on anything other than errors-of-aggregation, unsupportable probabilism, or questionable a priori logic, all of which run counter to observation.

Again, this isn't my field, but I don't see the cartel-effect or these other issues addressed in the literature that I've spent time on.

[Neil Harding]

At the end of the day, there are social costs to lower wages and lower benefits - as they say in Africa 'while my neighbour starves my chickens are not safe'. Everybody, even Tories pay when crime and alienation increase. It is better to pay people a living wage and those at the top not quite so much. This is one experiment that has never been tried, a shame.

[Shuggy]

"abolishing the minimum wage would add to the deficit by increasing the tax credit bill."

Yup - and these tax credits would effectively represent a subsidy for firms paying low wages. The right would never approve of such welfare dependency, surely?

Is it not the case that the demand for low paying jobs is particularly price inelastic? You can't out-source your need for a haircut, for example.

[Business directory]

Businesses should remember: without the good salary there is no good work:)

[michelmas]

Abolish the minimum wage? We can do better than that: bring back slavery!

[Christopher Carr]

Most of you have summed up this problem without even realising it. Modern governments take all their information from very intelligent people like yourselves who can examine demand elasticity, barriers to entry and all other associated factors and whilst you do this completely forget that you are talking about real people with real lives and responsibilities.

£5.93 ph is a pitiful amount of money to expect someone to live off.

However most of you will never have to support yourself let alone a family on such money as many of you will go on to well-paid stable jobs.

When you do have to make NMW calculations in your future careers just be sure to go to your local coffee shop or fast food outlet and ask one of the staff how they cope day to day.

Cut NMW?

For shame gentlemen, for shame.