Economics... again. (Sigh.)

In April, New Zealand’s Labour-led government raised the minimum wage from $16.50 to $17.70 per hour. Basic economic theory, which I have perforce become acquainted with over seven years of taking undergraduate notes in a far-flung range of subjects, would predict that the unemployment rate must rise proportionally. Being also acquainted with the methods by which economists arrive at basic economic theory, I am therefore entirely unsurprised to read that

New Zealand’s unemployment rate fell to 3.9 percent in the second quarter of 2019 from 4.2 percent in the previous period, compared to market expectations of 4.3 percent. That was the lowest jobless rate since the second quarter of 2008, when it was 3.8 percent. Unemployment rate in New Zealand averaged 5.99 percent from 1985 until 2019, reaching an all-time high of 11.20 percent in the third quarter of 1991 and a record low of 3.30 percent in the fourth quarter of 2007.

—Trading Economics

In case you’re wondering, yes, the second quarter of 2008 was towards the end of the Labour Party’s previous term in government, as was that record low in 2007, and 1991 was in the first term of the National government before them. Trading Economics has a graph of the New Zealand unemployment rate for the last thirty-three years, which I’ve taken the liberty of colour-coding rather crudely in Paint according to the leading party of government. (Note for Americans: in New Zealand as in most of the rest of the world, the left-leaning party is branded red and the conservative party blue.)

I don’t want to read more into this than is warranted. National took the reins of government from Labour twice during this period, and both times came shortly after worldwide financial crises, so I’d be cautious about blaming them for the big upticks in unemployment at those times. Crises aside, the longer-term trend is downward. What I do want to point out is that there is no sign of the upward drift that orthodox economic theory would lead us to expect during Labour’s tenures, despite the fact that they raise the minimum wage every year by much bigger increments than National does.

This semester I’m taking a first-year economics class, again. It’s struck me right from the start how much it’s re-treading ground I’ve been over and over since the first time round, seven years ago. If that doesn’t sound surprising, by contrast I’ve been assigned to papers in various health professions each semester since 2013, and every year there’s been a module on cancer, and every year there’s new insights about how cancer happens and what healthcare providers can do to fight it. That’s what progress looks like. That’s how you know a discipline is open to learning new truths.

In fairness, this semester’s lecturer has mentioned a few things which, in previous economics classes I’ve been to, have been skated over completely. He did take care to point out, for example, that sometimes people don’t buy certain goods or services not because they don’t want to pay the market price but because they can’t – the first time I’ve ever heard this distinction highlighted. (Unfortunately he seems to have missed some rather major implications of it, but we’ll get to that.) Also he’s promised an upcoming block of lectures on market failures, which up till now I haven’t heard economists talk about to students below third-year.

Still, as in previous years, the lecture material treats labour as just another good or service, which workers sell and employers buy. When you draw a supply-and-demand graph, the demand curve slopes downward, meaning that the more something costs, the fewer people will be willing to buy it and the less of it will get bought. Meanwhile the supply curve slopes upward, meaning the more it costs, the more people will be willing to sell it and the more of it will be available for sale. Where the two curves cross, the amount people are trying to buy equals the amount available for sale, and the price at which that happens is the market price or “equilibrium” price. They’re called “curves” but they’re usually graphed as straight diagonal lines. The slope of each curve is called its “elasticity”.

I’m not allowed to ask questions in lectures, and after any lecture there’s usually a few students wanting to talk to the lecturer and their practical need for information obviously takes precedence over my idle curiosity. But after the lecture introducing the concept of a supply curve, I had a rare opportunity for conversation. Rather than let out every expostulation I’ve had to swallow in an economics lecture, I just asked: Could there be such a thing as a negatively elastic supply curve (i.e. downward-sloping, i.e. the higher the price the less you sell)? The lecturer seemed blindsided by the question, so I clarified: Suppose you had to make a certain amount of revenue within a given time or go bankrupt, then if the price went down wouldn’t you have to sell more of the goods to meet your target? And what would the equilibrium price be? Well, he said, in that situation the concept of equilibrium wouldn’t really apply. But it was hard to believe such a situation could occur across a whole market.

So I came clean, or a little bit cleaner. Couldn’t it happen with labour in some markets, I asked? If you need to make a certain minimum amount each week in order to eat and pay the rent, then if wages go down aren’t you obliged to take on more hours? Well, you could argue that, he said. In fact, did I know there was a study on New York taxi drivers that found that was pretty much what they did? That they worked each day until they’d reached a particular total amount from their fares, working long hours when there weren’t many passengers and knocking off early on busy days? (I didn’t let on that that very study was one of the inspirations for my question.) It just went to show, he said, that people weren’t always rational – a rational taxi-driver would do the long hours on busy days, save up, and take the day off when business was slow.

I had another lecture to go to, so I didn’t stop to argue the rationality point. What I thought, though, was: Look, I’m trying to save money for an overseas holiday hopefully next year. You’re telling me that, if my wage were to go down for some reason, thus delaying my holiday, the rational response would be to take fewer work hours – thus delaying my holiday even longer. Rational? Really? What about if the stakes were not travel next year, but rent next fortnight and groceries tomorrow? In any situation where you have a limited time to make a minimum amount of money, doesn’t that flip the rationality of your choices?

The very next lecture I attended brought up the question of the minimum wage. If you have a mandatory minimum wage that is above the market wage, then according to orthodox economic theory it’s going to cause unemployment. Again, think of those criss-crossing supply and demand “curves”. The minimum wage is a horizontal line cutting across the graph above where those curves intersect. Because it’s higher than the market wage, demand is lower – fewer people are prepared to employ workers for whatever particular job the graph is about. But for the same reason, supply is higher – people are prepared to put more hours into that job, and some people now want to work in it who didn’t previously want to at all. The gap between the amount of work available and the amount of work sought is unemployment.

I’ve heard all that and had to type it down, gritting my teeth, before. But one thing that distinguishes this particular paper from other first- and second-year economics papers I’ve taken notes for is that the lecturers actually back up what they say with empirical evidence, albeit still fairly scanty evidence by the standards of science and health-science papers. First he showed us a graph with the average unemployment rates in states with minimum wages and states without minimum wages, respectively, and sure enough the states with minimum wages had higher unemployment. But, he said, we can’t stop there. There might be some other reason why the first line is higher. (For instance, unemployed people might vote in left-leaning governments who then institute minimum wages.) You have to hold everything else the same, so that the only thing that changes is the minimum wage, before you can be sure.

And then he showed us a study where they had done just that – watched the unemployment rate in two adjacent American states with very similar economies until just one of those states raised its minimum wage. Did that state’s unemployment rate go up as a result, compared to the other? It did not.

His hypothesis was that when their wage went up, employees felt more satisfied with their work and became more productive, as a result of which employers were able to afford to cover their increased employment costs. Well, sure, that’s possible. I can’t help feeling it’s a bit fortuitous, though. It’s not like any worker can keep on getting more and more productive indefinitely as a result of something so intangible as satisfaction. Seems a bit of a bummer, if you’re trying to capture a general cause-&-effect link that your theories say must be there, that it gets drowned out by what has to be a particular, local, context-sensitive circumstance.

At the very least, isn’t it worth considering the alternative hypothesis that the labour supply curve is negatively elastic? That most workers are like the New York taxi-drivers and me with my travel savings, and take more time off when they can meet their needs on fewer work hours? That the supply and demand curves on the labour graph slope in the same direction, so you wouldn’t expect the gap between them to get bigger as you move up the wage axis? And wouldn’t that also neatly explain the failure of New Zealand’s minimum wage and unemployment rate to correlate over time?

Presumably this condition wouldn’t go all the way down the graph. People can’t work twenty-four hours a day for free without sleep or lunch breaks. Eventually you would reach a point where, from working too long in too poor conditions on too little money to buy decent food, you’d be so unwell you’d collapse at your work-station. Below that point your personal labour supply curve would once again be positive – you wouldn’t be able to work any longer if your wage was cut again, no matter how much you might need the money.

The implications get a bit scary. Remember, when the supply curve slopes downward, there is no equilibrium. Remember also that what’s imposing this condition is the worker’s need to make a certain minimum income in order to survive; if they hold out for higher wages and shorter hours and every available employer says “no,” their children starve. Hence, if my hypothesis is correct, then absent a minimum wage (whether imposed by the law, by union agitation, or by competition with government welfare payments), the “market wage” will sit exactly at the workers’ point of collapse. If that seems improbable, I recommend a look into what labour conditions were like before we had unions and the welfare state, and still are like in low-wage countries that those institutions have not yet reached. Charles Dickens’ Oliver Twist is an emphatic but not exaggerated depiction.

---

If this blog actually had any readers, and if any of them followed the economics label, they would find that my other big problem with undergraduate economics is its definition of the word “welfare” and its conflation of “willingness to pay” with ability to pay. Yes, these two things together constitute one single problem.

Let’s start with “willingness to pay”. It means what it looks like it means, i.e. how much any individual consumer is prepared to pay for a good or service. If the price is lower than that amount, they’ll buy it; if it’s higher, they won’t. Either way, they end up enjoying what they value more – the good or service if they buy it, the money if they don’t. That way, the people who value it most will be the ones who get it. If the supply of gluten-free food runs short, for example, and the suppliers put the price up, then presumably the people with coeliac disease will pay the higher price while the fad dieters won’t, and so the people who really need it will be the ones who get it.

Already you’ll have seen the problem, and as I say this lecturer (for the first time in my economics note-taking experience) has alluded to it. Sometimes the people who would be most willing to pay a high price, on account of needing the product the most, are not in a position to be able to pay the high price. He’s even stated outright that the same dollar makes more difference to a poor person than it does to a rich person. But although he’s acknowledged both these things, he doesn’t seem to have spotted the implications for the concept of “welfare”.

“Willingness to pay” is measured in dollars. Subtract the actual price you paid from your “willingness to pay”, and the result is the “surplus value”, i.e. how much better off you are for having made the purchase. The seller also enjoys surplus value, calculated by subtracting their “willingness to sell” from the price they actually get, i.e. how much better off they are for having made the sale. And in economics, “welfare” equals the total surplus value from all transactions across the economy. “Welfare” in this sense is what economists seek to maximize.

Now I don’t know about you, but when I hear the word “welfare” I instinctively think of something a bit more closely connected to human wellbeing. Let’s suppose a government is choosing between two different economic policies. One would improve “welfare”, in the technical sense, by one million dollars, all of which would be enjoyed by one single billionaire. The other would improve “welfare” by one hundred thousand dollars, which would be shared evenly among one hundred homeless people. Which policy is better? By the technical definition of “welfare” it has to be the first policy; the second one creates only 10% as much surplus value and thus results in a 90% “deadweight loss”. But I think it’s pretty clear that giving a hundred homeless people enough money to rent a house for a month or two and buy job-interview suits and make a start on their dental hygiene improves human wellbeing vastly more than nudging up the number in a billionaire’s bank account by an increment he’ll barely notice.

This much of course I’ve complained about many times before. I even know the economic jargon to put it in: instead of maximizing aggregate surplus value from transactions, we should be trying to maximize the aggregate utility of surplus value from transactions. But recently our lecturer said something with implications that, once you think in terms of “ability to pay” rather than “willingness to pay”, are frankly rather horrifying, and I really hope he just hadn’t thought through those implications.

The topic at hand was price ceilings. When the government imposes a price ceiling on a good or service which is lower than the market price, this is bad because more people want to buy the thing and fewer people can sell the thing and so you end up with a shortage. That, again, I’d heard several times before. But, said the lecturer, there’s another bad consequence as well. Some of the people who now succeed in buying the thing before it runs out will be people whose “willingness to pay” is below the market price – who wouldn’t have bothered shopping for it before the government implemented the price ceiling. And some of the people who miss out on the thing because of the shortage will be people whose “willingness to pay” was much higher. So the thing will have been taken away from people who really value it and provided to people who don’t, and that’s a problem.

Which is a good and reasonable objection as long as you can be absolutely sure that it really is willingness to pay, and not ability to pay, that determines demand in this market. If not, well, run through that last paragraph substituting “ability” for “willingness”. Some of the people who buy the thing before it runs out will be people who can’t afford the market price. Some of the people who miss out will be people whose available funds were much higher. Basically the argument boils down to “Poor people are benefiting from this thing instead of rich people, and that’s bad.”

So basically I’ll award economics half marks for effort, but it still really has a long way to go before it can claim the status either of a science or of a boon to society. To achieve either, it must subject its core theories to the possibility of falsification by empirical facts. Maybe if I keep saying this over and over, an idle Google search will bring it to the eyes of someone who can actually do something about it.