inamerrata

I have a few things I need to write, but am still a bit too sick with the flu to put together something novel, so instead I’m going to counter-blog Rob Collins recent claim that Money doesn’t matter. Rob’s thoughts are similar to ones I’ve had before, but I think they’re ultimately badly mistaken.

There’s three related, but very different, ways of thinking about money: as a store of value, as a medium of exchange, and as a unit of account. In normal times, dollars (or pounds or euros) work for all three things, so it’s easy to confuse them, but when you’re comparing different moneys some are better at one than another, and when a money starts failing, it will generally fail at each purpose at different rates.

Rob says “Money isn’t wealth” — but that’s wrong. In so far as money serves as a store of value, it is wealth. That’s why having a million dollars in your bank account makes you feel wealthy. The obvious failure mode for store of value is runaway inflation, and that quickly becomes a humanitarian disaster. Money can be one way to store value, but it isn’t the only way: you can store value by investing in artwork, buying property, building a company, or anything else that you expect to be able to sell at some later date. The main difference between those forms of investment versus money is that, ideally, monetary investments have low risk (perhaps the art you bought goes out of fashion and becomes worthless, or the company goes bankrupt, but your million dollars remains a million dollars), and low variance (you won’t make any huge profits, but you won’t make huge losses either). Unlike other assets, money also tends to be very fungible — if you earn $1000, you can spend $100 and have $900 left over; but if you have an artwork worth $1000 it’s a lot harder to sell one tenth of it.

Rob follows up by saying that money is “a thing you can exchange for other things”, which is true — money is a medium of exchange. Ideally it’s cheap and efficient, hard to counterfeit, and easy to verify. This is mostly a matter of technology: pretty gems are good at these things in some ways, coins and paper notes are good in others, cheques kind of work though they’re a bit to easy to counterfeit and a bit too hard to verify, and these days computer networks make credit card systems pretty effective. Ultimately a lot of modern systems have ended up as walled gardens though, and while they’re efficient, they aren’t cheap: whether you consider the 1% fees credit card companies charge, or the 2%-4% fees paypal charges, or the 30% fees from the Apple App Store or Google Play Stores, those are all a lot larger than how much you’d lose accepting a $50 note from someone directly. I have a lot of hope that Bitcoin’s Lightning Network will eventually have a huge impact here. Note that if money isn’t wealth — that is, it doesn’t manage to be a good store of value even in the short term, it’s not a good medium of exchange either: you can’t buy things with it because the people selling will have to immediately get rid of it or they’ll be making a loss; which is why currencies undergoing hyperinflation result in black markets where trade happens in stable currencies.

With modern technology and electronic derivatives, you could (in theory) probably avoid ever holding money. If you’re a potato farmer and someone wants to buy a potato from you, but you want to receive fertilizer for next season’s crop rather than paper money, the exchange could probably be fully automated by an online exchange so that you end up with an extra hundred grams of fertilizer in your next order, with all the details automatically worked out. If you did have such a system, you’d entirely avoid using money as a store of value (though you’d probably be using a credit account with your fertilizer supplier as a store of value), and you’d at least mostly avoid using money as a medium of exchange, but you’d probably still end up using money as a medium of account — that is you’d still be listing the price of potatoes in dollars.

A widely accepted unit of account is pretty important — you need it in order to make contracts work, and it makes comparing different trades much easier. Compare the question “should I sell four apples for three oranges, or two apples for ten strawberries?” with “should I sell four apples for $5, or two apples for $3” and “should I buy three oranges for $5 or ten strawberries for $3?” While I suppose it’s theoretically possible to do finance and economics without a common unit of account, it would be pretty difficult.

This is a pretty key part and it’s where money matters a lot. If you have an employment contract saying you’ll be paid $5000 a month, then it’s pretty important what “$5000” is actually worth. If a few months down the track there’s a severe inflation event, and it’s only worth significantly less, then you’ve just had a severe pay cut (eg, the Argentinian Peso dropped from 5c USD in April to 2.5c USD in September). If you’ve got a well managed currency, that usually means low but positive inflation, so you’ll instead get a 2%-5% pay cut every year — which is considered desirable by economists as it provides an automatic way to devote less resources to less valuable jobs, without managers having to deliberately fire people, or directly cut peoples’ pay. Of course, people tend to be as smart as economists, and many workers expect automatic pay rises in line with inflation anyway.

Rob’s next bit is basically summarising the concept of sticky prices: if there’s suddenly more money to go around, the economy goes weird because people aren’t able to fix prices to match the new reality quickly, causing shortages if there’s more money before there’s higher prices, or gluts (and probably a recession) if there’s less money and people can’t afford to buy all the stuff that’s around — this is what happened in the global financial crisis in 2008/9, though I don’t think there’s really a consensus on whether the blame for less money going around should be put on the government via the Federal Reserve, or the banks, or some other combination of actors.

To summarise so far: money does matter a lot. Having a common unit of account so you can give things meaningful prices is essential, having a convenient store of value that you can use for large and small amounts, and being able to easily trade it for goods and services is a really big deal. Screwing it up hurts people directly, and can be really massively harmful. You could probably use something different for medium of exchange than method of account (eg, a lot of places accepting cryptocurrencies use the cryptocurrency as medium of exchange, but use regular dollars for both store of value and pricing); but without a store of value you don’t have a medium of exchange, and once you’ve got a method of account, having it also work as a store of value is probably too convenient to skip.

But all that said, money is just a tool — generally money isn’t what anyone wants, people want the things they can get with money. Rob phrases that as “resources and productivity”, which is fine; I think the economics jargon would be “real GDP” — ie, the actual stuff that goes into GDP, as opposed to the dollar figure you put on it. Things start going wonky quickly though, in particular with the phrase “If, given the people currently in our country, and what they are being paid to do today, we have enough resources, and enough labour-and-productivity to …” — this starts mixing up nominal and real terms: people expect to be paid in dollars, but resources and labour are real units. If you’re talking about allocating real resources rather than dollars, you need to balance that against paying people in real resources rather than dollars as well, because that’s what they’re going to buy with their nominal dollars.

Why does that matter? Ultimately, because it’s very easy to get the maths wrong and not have your model of the national economy balanced: you allocate some resources here, pay some money there, then forget that the people you paid will use that money to reallocate some resources. If the error’s large enough and systemic enough, you’ll get runaway inflation and all the problems that go with it.

Rob has a specific example here: an unemployed (but skilled) builder, and a homeless family (who need a house built). Why not put the two together, magic up some money to prime the system and build a house? Voila the builder has a job, and the family has a home and everyone is presumably better off. But you can do the same thing without money: give the homeless family a loaded gun and introduce them to the builder: the builder has a job, and the family get a home, and with any luck the bullet doesn’t even get used! The key problem was that we didn’t inspect the magic sufficiently: the builder doesn’t want a job, or even money, he wants the rewards that the job and the money obtain. But where do those rewards come from? Maybe we think the family will contribute to the economy once they have a roof over their heads — if so, we could commit to that: forget the gun, the family goes to a bank, demonstrates they’ll be able to earn an income in future, and takes out a loan, then goes to the builder and pays for their house, and then they get jobs and pay off their mortgage. But if the house doesn’t let the family get jobs and pay for the home, the things the builder buys with his pay have to come from somewhere, and the only way that can happen is by making everyone else in the country a little bit poorer. Do that enough, and everyone who can will move to a different country that doesn’t have that problem.

Loans are a serious answer to the problem in general: if the family is going to be able to work and pay for the house eventually, the problem isn’t one of money, it’s one of risk: whoever currently owns the land, or the building supplies, or whatever doesn’t want to take the risk they’ll never see anything for letting the house get built. But once you have someone with founds who is willing to take the risk, things can start happening without any change in government policies. Loaning directly to the family isn’t the only way; you could build a set of units on spec, and run a charity that finds disadvantaged families, and sets them up, and maybe provide them with training or administrative support to help them get into the workforce, at which point they can pay you back and you can either turn a profit, or help the next disadvantaged family; or maybe both.

Rob then asks himself a bunch of questions, which I’ll answer too:

• What about the foreign account deficit? (It doesn’t matter in the first place, unless perhaps you’re anti-immigrant, and don’t want foreigners buying property)
• What about the fact that lots of land is already owned by someone? (There’s enough land in Australia outside of Sydney/Melbourne that this isn’t an issue; I don’t have any idea what it’s like in NZ, but see Tokyo for ways of housing people on very little land if it is a problem)
• How do we fairly get the family the house they deserve? (They don’t deserve a house; if they want a nice house, they should work and save for it. If they’re going through hard times, and just need a roof over their heads, that’s easily and cheaply done, and doesn’t need a lot of space)
• Won’t some people just ride on the coat-tails of others? (Yes, of course they will. That’s why you target the assistance to help them survive and get back on their feet, and if they want to get whatever it is they think they deserve, they can work for it, like everyone else)
• Isn’t this going to require taking things other people have already earnt? (Generally, no: people almost always buy houses with loans, for instance, rather than being given them for free, or buying them outright; there might be a need to raises taxes, but not to fundamentally change them, though there might be other reasons why larger reform is worthwhile)

This brings us back to the claim Rob makes at the start of his blog: that the whole “government cannot pay for healthcare” thing is nonsense. It’s not nonsense: at the extreme, government can’t pay for enough healthcare for everyone to live to 120 while feeling like they’re 30. Even paying enough for everyone to have the best possible medical care isn’t feasible: even if NZ has a uniform health care system with 100% of its economy devoted to caring for the sick and disabled, there’s going to be a specialist facility somewhere overseas that does a better job. If there isn’t a uniform healthcare system (and there won’t be, even if only due to some doctors/nurses being individually more talented), there’ll also be better and worse places to go in NZ. The reason we have worrying fiscal crises in healthcare and aged support isn’t just a matter of money that can be changed with inflation, it’s that the real economic resources we’re expecting to have don’t align with the promises we’re already making. Those resources are usually expressed in dollar terms, but that’s because having a unit of account makes talking about these things easier: we don’t have to explicitly say “we’ll need x surgeons and y administrators and z MRI machines and w beds” but instead can just collect it all and say “we’ll need x billion dollars”, and leave out a whole mass of complexity, while still being reasonably accurate.

(Similar with “education” — there are limits to how well you can educate everyone, and there’s a trade off between how many resources you might want to put into educating people versus how many resources other people would prefer. In a democracy, that’s just something that’s going to get debated. As far as land goes, on the other hand, I don’t think there’s a fundamental limit to the government taking control over land it controls, though at least in Australia I believe that’s generally considered to be against the vibe of the constitution. If you want to fairly compensate land holders for taking their land, that goes back to budget negotiations and government priorities, and doesn’t seem very interesting in the abstract)

Probably the worst part of Rob’s blog is this though: “We get 10% less things done. Big deal.” Getting 10% less things done is a disaster, for comparison the Great Recession in the US had a GDP drop of less than half that, at -4.2% between 2007Q4 and 2009Q2, and the Great Depression was supposedly about -15% between 1929 and 1932. Also, saying “we’d want 90% of folk not working” is pretty much saying “90% of folk have nothing of value to contribute to anyone else”, because if they did, they could do that, be paid for it, and voila, they’re working. That simply doesn’t seem plausible to me, and I think things would get pretty ugly if it ended up that way despite it’s implausibility.

(Aside: for someone who’s against carbs, “potato farmer” as the go to example seems an interesting choice… )