Costco purchases less labour at a higher price than Walmart (evidence that a higher minimum wage means fewer workers hired)

See The Paul Krugman Puzzle With The Minimum Wage by Tim Worstall.

"I’ve mentioned before the headscratcher over what appears to be a complete about turn in the views of Paul Krugman concerning the minimum wage. 1990s Krugman pointed out that yes, if you raise the price of something then people are likely to purchase less of it and this applies to labour just as much as it does to anything else (OK, anything other than Giffen Goods, a special case and rare). He also used to point out that you can’t go around claiming that employers will benefit from lower staff turnover and so on. Because that requires paying more than alternative employers, not just everyone paying a higher wage. More recently he’s been arguing that actually, raising the minimum wage doesn’t create unemployment, that it does lower turnover and so on. It’s an entire turnaround in his views.

Today that headscratcher becomes even more difficult to decipher. For in one part of his column he says that higher wages don’t mean that people reduce the amount of labour that they purchase. And then in another part of the same column he says that they do. He just uses a different piece of economic jargon to make that second point which is why, perhaps, many people won’t get it.

So, here’s the higher wages don’t reduce employment:

And the overwhelming conclusion from all that evidence is that the effect you might expect to see — higher minimum wages leading to fewer jobs — is weak to nonexistent. Raising the minimum wage makes jobs better; it doesn’t seem to make them scarcer.

And again:

As a result, raising the minimum wage, while it makes labor more expensive, has offsetting benefits that tend to lower costs, limiting any adverse effect on jobs.

But then we get this:

Similar factors explain another puzzle about labor markets: the way different firms in what looks like the same business can pay very different wages. The classic comparison is between Walmart (with its low wages, low morale, and very high turnover) and Costco (which offers higher wages and better benefits, and makes up the difference with better productivity and worker loyalty).

But that explanation about Costco is exactly that Costco purchases less labour, albeit at a higher price. That’s what “better productivity” means. Either that they get the same amount of work done with less labour, or that they get more work done with the same labour. And either of those is equivalent to stating that higher labour costs reduce the amount of labour purchased for any given level of output.

Please do understand this. These two statements are equal “higher wages causes jobs losses” and “higher wages means increased labour productivity”. They mean the same thing.

I last looked at the Costco versus Walmart numbers a decade ago and I’m sure they’re somewhat different from what they were. But the basic fact was that Walmart hired about twice the hours of labour that Costco did for any given level of sales. There’s nothing either right or wrong about this, it’s just one company being an extensive user of labour the other an intensive. And Walmart most certainly could move over to being an intensive user of labour, like Costco. And thus pay the same sorts of wages and benefits as Costco. But to do so it would have to make its workers about twice as productive as they are now: OK, that can also be done. But note the end result of that. Assuming sales stay roughly where they are then Walmart would need only half the number of workers it has now. And yes, we would think that this is equal to higher wages having reduced the number of jobs.

All of which is what leaves us with this puzzle. Krugman’s far too good an economist not to get this equivalence between productivity and the amount of labour demanded. So why on earth has he put both statements into the same column?"