That a higher minimum wage can actually lower incomes is one of those things intensely debated. Absolutely every economist will agree that this can be true at some level of that minimum wage. If you make the minimum wage $100 an hour, you're going to see an awful lot of people earning absolutely nothing through having no job at all. There are perhaps a few (Nick Hanauer comes to mind) who think that this is nonsense.

The debate then becomes, well, at what wage does this become true? The argument then becomes intensely political, those to the Left of us arguing that it would take a much higher minimum wage to put people out of work, while those to the Right insist the minimum wage is already too high and is putting people out of a job. I am to the right of this point, despite the use of “us” there. But we’d all actually like proof.

At which point we have the University of Washington study into the Seattle minimum wage hike. This says that hourly wages have risen and hours worked fallen. This isn’t a surprise at all. But the hours have fallen more, meaning that weekly incomes have actually fallen among those very low-paid people that minimum wage hikes are supposedly helping. A more recent study from Berkeley tries to refute this but also agrees that it doesn’t have and therefore hasn’t looked at detailed data on hours worked. So, obviously enough, the Berkeley study is missing the claimed effect entirely.

Given the interest in this – both politically but also because we’d just like to darn well know, let’s get this right before we lower the incomes of the poor, perhaps – we should look for evidence elsewhere. Which gives us a study from my native Britain done by folks at my alma mater, the London School of Economics.

[Related: Amazon to set $15 minimum wage after Bernie Sanders attacks]

They point out that the young have suffered most grievously from the slow, insipid recovery from the Great Recession, weekly incomes for them being down by 16 percent in real terms. They also note that real hourly wages have risen strongly – there have been several minimum wage hikes over this period of time, quite substantial in real terms especially given generally falling wages across the economy.

But note that fact – hourly wages are up, weekly incomes down. There is only one explanation for this: hours worked have fallen. That is, it is entirely possible to have a minimum wage which is “too high.” For it is possible to have a minimum wage which reduces the incomes of the hard-working poor, which is not the point of what we’re trying to do at all. Sure, nearly all of us agree this is possible conceptually but we’re also being told that it is actually happening at the minimum wage levels we’ve got at present.

This is at wages rates of £7.38 (some $9.60) for 21 to 24 year olds, £5.90 ($7.70) for those 18 to 20. Britain’s a poorer country so these are actually higher in impact than the direct exchange rate conversion in their effects, so add another 15 to 20 percent perhaps.

No one empirical study is going to solve this or any other such question. But when the evidence starts to mount up we do have to start saying that we’re beginning to know. The old rule of thumb is that minimum wages at half of median wage or below (say, for full and part time/temporary wage rates the correct median to use, around $18 an hour for the U.S. now) don’t make much difference. So few people get paid such low wages that it just doesn’t appear in the numbers in any way we might meaningfully divine. Once we go over that 50 percent or so then we do start to see those disemployment effects. And the more numerical evidence we get from around the world the more this seems to be correct.

Sorry folks, substantial increases in the minimum wage just don’t raise low-end incomes.

Tim Worstall (@worstall) is a contributor to the Washington Examiner's Beltway Confidential blog. He is a senior fellow at the Adam Smith Institute. You can read all his pieces at The Continental Telegraph.