The state cannot function without money. If it wants to build roads, or send out stimulus cheques, or maintain the military, it needs dollars to pay for materials and salaries. To obtain dollars, the state can either levy taxes or borrow funds. As Margaret Thatcher said: “the state has no source of money, other than the money people earn themselves. If the state wishes to spend more it can only do so by borrowing your savings or by taxing you more.” When the state spends more than it taxes, the national debt increases. Eventually, this debt will have to be paid back: it represents a burden on future generations. It is therefore unsustainable to keep borrowing. When the deficit becomes too high, the government “runs out of money”, as Obama claimed had happened in 2009.
Or at least, that is how most people believe the economy works. It is based on an analogy with the household. I cannot keep spending money without increasing my income or borrowing the difference; neither can the state. I will eventually have to pay back my debts; so does the state. These principles seem reasonable for an individual, so why wouldn’t they apply to the nation as a whole?
But advocates of Modern Monetary Theory (MMT) claim that this is all wrong. There is an important difference between me and the state: I am a currency-user, while the state is a currency-issuer. Unlike any individual, household or local government, the federal state is the sole issuer of US dollars. Put simply, only the federal state can make dollars. Every dollar bill in people’s wallets was printed by the US, and every digital dollar in people’s bank accounts was created as an entry in some database by the US. If the government is literally the source of all money, then why would it need to tax or borrow before it can spend?
This is the insight that drives MMT. Although sometimes seen as a fringe movement within economics, it is quickly picking up steam. Stephanie Kelton, one of MMT’s main proponents, has written an introduction: The Deficit Myth. The book is aimed at a non-academic audience, and offers a dumbed-down account of the theory. Disclaimer: I am not an economist, so I cannot always judge Kelton’s arguments on their more technical merits. But the book is not aimed at economists, so I feel justified in reviewing it. The book roughly has two parts. In the first half, Kelton explains the descriptive side of MMT: how does the economy work, according to that theory? I found this part very interesting, as it changed my understanding of how the economy works. In the second half, Kelton addresses the prescriptive side: given MMT, which policies should we adopt? This part was a little less convincing. Although Kelton tries to emphasise that MMT itself is non-partisan, it still seemed like she was trying to use it as a justification for various policies she independently seemed to favour, such as universal healthcare. In this review, I will therefore mostly focus on the descriptive side, and only address the prescriptive side towards the end.
Before delving into the details of MMT, it is helpful to consider the historical background. MMT starts with a story about the origin of money. However, MMT itself is independent from the historical claim. Even if the historical claim is false, MMT might be true.
The historical claim exposes another myth: that money originated with barter. The standard story, which everyone has heard, is that people used to trade goods. But this was very cumbersome: if I have a cow, and want a sack of grain, then I have to find someone with a sack of grain who just happens to want a cow. If that person doesn’t exist, or if I cannot find him, then I can trade my cow for, say, a pair of chickens, in the hope that someone out there wants to trade a sack of grain for a pair of chickens. The introduction of money cut out the middle man. I can now ‘trade’ my cow for dollars, and use those dollars to buy whatever I want.
But advocates of MMT believe that this is not how money actually came into existence. The alternative history they tell is called ‘Chartalism’. In essence, money is a way for the state to direct economic activity. Suppose that the state wants to wage a war, so it needs cannons. But nobody wants to build cannons, because nobody wants to trade cannons for grain. The state could seize grain from the populace, and offer it to anyone in return for cannons. But Chartalists argue that the state instead employed a clever trick: they created a currency – call them Cannon Coins™ – and demanded that everyone pay their tax in Cannon Coins. Of course, no one has Cannon Coins™to begin with. But the state promises to pay everyone Cannon Coins™ in return for cannons. This incentivises the population to build cannons for the state, so they can pay their taxes. Not everyone is going to build cannons. The cannon builders will sell some of their Cannon Coins™ for goods such as grain. Cannon-builders also need to eat. Over time, Cannon Coins™ become a real currency.
In this scenario, it would be absurd to suggest that the state has to tax people to be able to spend Cannon Coins™. The state has literally created the concept of Cannon Coins™! It is true that cannon-builders only accept Cannon Coins™ because the state levies a Cannon Coin™ tax, but the sequence is reversed: the state can only tax people’s Cannon Coins™ after it has spent them on actual cannons. In Kelton’s words: “It’s not our tax money the government wants. It’s our time.”
This story motivates MMT. I find MMT’s claims much more plausible when I think about the introduction of money in this way. Having said that, MMT is independent from the history of money. Even if the standard story about barter is true, MMT can still be right about the way money is used in contemporary society.
If MMT is correct, then the state does not need to tax or borrow to finance its spending. The argument that the deficit is unsustainable or that the government’s credit card has ‘maxed out’ is fallacious. But there is a more sophisticated argument against overspending: it leads to sky-high inflation!
The intuition here is that if the state literally prints money while the total output of the economy stays the same, every dollar becomes worth less in real terms. The same level of production is divided by a higher number of dollars. The simple fact that there is more money going around the economy doesn’t make us richer: our wealth is ultimately measured in terms of the riches we consume. This is called the monetary theory of inflation. Now, inflation wouldn’t be bad if wages also rose. However, wages are ‘sticky’ – they don’t rise with inflation. I think the idea here is that it is a big risk to give up your job and try to find a better-paying one. So, your employer can get away with not raising your wage, or raising it less than inflation, without running the immediate risk of you quitting. For you, the employee, this is bad news: your wage stays the same even though prices increase. Your real standard of living declines.
MMT has a difficult relation with inflation. On the one hand, Kelton is very honest that inflation is a real constraint, indeed the real constraint on the economy: “MMT is about distinguishing the real limits [i.e. inflation] from the self-imposed constraints that we have the power to change [i.e. keeping the deficit small]”. Indeed, for MMTers the main reason to levy taxes is to decrease inflation. The state doesn’t need the income from taxes to finance its spending. But when people have to pay part of their wages back in taxes, they will spend less on goods. This leads to a decrease in demand, which in turn leads to less inflation (recall: when demand goes up prices go up, and inflation just is increasing prices). Kelton, and other MMTers, fully agree that if the government would suddenly announce that all taxes are scrapped, inflation would go through the roof.
On the other hand, she often appears callous about the risk of inflation, advising that the government should spend much more than it currently does. There are several reasons for this. Firstly, Kelton notes that in the past few decades, inflation has consistently been lower than the target rate of 2%. Everyone agrees that it is good for the economy to have a low but non-zero level of inflation, partly to avoid the risk of deflation, which can lead to a recession. But over the past few decades inflation has been much lower than 2%, even though central banks have put an active effort into raising it, for example by setting zero or even negative interest rates. So, Kelton retorts, isn’t it a bit premature to worry about inflation being too high? (The recent high levels of inflation somewhat temper these claims.)
The second and more important reason is that MMTers don’t seem to believe that any amount of overspending immediately leads to higher inflation in a one-to-one ratio. This makes some sense. Suppose the state decides to print money to build a bridge. There is now more money in circulation, so each individual dollar is worth a little less. But the size of the economy increases as the bridge is used to transport goods, or for people to travel to work, or simply to visit friends and family. This means that each dollar is now worth a little more again. Perhaps the net result is still some inflation, but the effect is less pronounced.
Kelton here uses the metaphor of “running the economy hot”. For Kelton, the economy has a “real productive capacity”, a sort of limit on what we can collectively produce. The limit is set by the amount of people available to work, but also by the technology we have at our disposal and the material resources we possess. (Part of the book’s main flaws is that it remains unclear whether this is a limit on the size of the economy, or on the rate of growth.) If this capacity is underutilised, then there is ‘slack’. The advice of MMT is to utilise this capacity by spending more, thereby “running the economy hot”. Kelton claims that this does not increase inflation, since the additional spending also increases production, as when the government builds a bridge. But when the economy is already running hot, further spending leads to ‘overheating’. This does lead to inflation. The model Kelton proposes is thus highly non-linear: first, overspending does not increase inflation at all, but when the limit is hit it suddenly only leads to inflation.
I found this one of the weaker parts of the book. I can sort of see why sometimes increased spending does not lead to (much) additional inflation, because it has a positive effect on production. I can also see that when the economy is already ‘at capacity’ the same amount of spending does lead to inflation. Here is how Kelton explains it: “Once the economy exhausts its real productive capacity, the only way for the government to get the construction workers, architects and engineers, steel, concrete, paving trucks, cranes and so on that it needs is to bid them away from their current use. That bidding process pushes prices higher, giving rise to inflationary pressures.” But I thought that Kelton could have said much more on why and when there is slack versus why and when the economy is running (too) hot.
Part of the answer here, which Kelton doesn’t discuss much but which is emphasised by other MMTers, is that it matters a lot what the state spends their money on. The monetary theory of inflation simply says that more money means more inflation. MMT predicts that money spent on parts of the economy which have significant slack has less of an inflationary effect than money spent on parts of the economy which are already running hot. For example, it seems that the market for smartphones is already running at capacity. If the government were to invest in tech further this would only drive up prizes, since the real bottleneck is the Chinese production of microchips. On the other hand, the market for green energy does not yet seem at full capacity. If the government were to invest in green energy solutions, this might lead to further innovations which could actually drive down prices. To put it slightly differently, MMT says that the government should look for current market failings and cleverly spend our way out of them.
So far, we have only considered spending and taxes. But it is not literally true that if the state spends more than it taxes, it has to print the difference. Typically, governments will borrow additional funds at the capital market. This works roughly as follows: the government announces that it wants to borrow dollar reserves. In return for these reserves, it gives out Treasury bonds with a certain interest rates. These are considered safe assets which retain their value, meanwhile paying out a nice interest rate. They are therefore popular with, for example, pension funds. But of course, the government eventually has to pay back these loans when the bonds mature. When that happens, it has to spend less, increase taxes or borrow again. Hence the idea that the government debt is a burden on future generations.
Kelton basically claims that borrowing on the financial market is equivalent to printing money. The Treasury bonds, she says, are akin to interest-bearing dollars. She calls them ‘yellow dollars’ to distinguish them from our standard green dollars. Of course, you can’t spend yellow dollars in a supermarket, but because they are considered so stable you can definitely use them for trade at financial markets or as part of an investment portfolio. Thinking about it this way, what happens when the government borrows dollar reserves is really that it creates new yellow dollars, and trades these one-for-one for green dollars that already exist. The net result is that there are now more dollars (red or yellow) in circulation.
This also means that it is very easy for the government to ‘pay off’ its debt: simply convert those yellow dollars back into green ones. Kelton claims that these days, the difference between dollar reserves and Treasury bonds is literally just that they sit in different bank accounts, only one of which has a non-zero interest rate. To pay off the national debt simply means transferring those dollars from one bank account to another! Therefore, those who claim that the deficit represents a burden on future generations are mistaken.
I find this story somewhat convincing, but I am still a bit confused about when we can call a financial asset such as a Treasury bond just a dollar by another name. If I give out personal bonds in return for dollars, have I increased the money stock? Perhaps not, because no one will ever accept a slip of paper which says “I, [name redacted], promise to pay you back 10 dollars by such-and-so date” as real currency. But even then, as long as the person whose money I borrowed trusts that I will pay them back, that person believes that the slip of paper is worth 10 dollars to them. They are then happy to spend a bit more, in the knowledge that they can always fall back on my debt to them. If that happens on a large scale, it would lead to an increase in demand the same way that government spending would. I don’t think this is a real objection against MMT, because the amount of money any individual lends is only a drop in the bucket compared to the amount of money the state borrows. But it makes me suspect that Kelton skips over some important differences between dollar reserves and Treasury bonds.
Finally, what are the policy consequences of MMT? Kelton takes the message that the deficit doesn’t matter to heart: the second half of The Deficit Myth contains a litany of suggestions for which programmes to fund, from universal health care to student debt relief. It is not hard to walk away with the idea that MMT is a left-winger’s dream. But it’s not that simple. MMT itself is a politically neutral theory, and not all of its implications are typically associated with the left (or the right, for that matter). In this section, I will discuss a few of the more surprising conclusions Kelton draws.
Firstly, MMT provides a strong argument against a wealth tax. Many on the left have argued for a tax on the wealth of the richest individuals. Biden’s ‘billionaire tax’ is only the most recent example. Part of the argument for such a tax is that it handily pays for costly government programmes, from universal health care to free tuition. But if MMT is true, the government does not need to find funds in order to finance its spending. We can have universal health care without a wealth tax, as long as inflation remains within acceptable limits.
The aim of taxes for MMT is to put a damper on inflation. But a wealth tax is spectacularly bad at this! The reason that taxes decrease inflation is that they stop people from spending their money on goods. In other words, they reduce demand. But rich people typically don’t spend their additional income, but either save or invest it. If you already have all the yachts, jets and tropical islands you need, the next bit of money you earn is just going into your investment fund. Therefore, rich people will still spend approximately as much after a wealth tax is instituted as before, so such a tax has little influence on inflation.
This is not to say that there is no argument for a wealth tax. Kelton herself favours a wealth tax in order to reduce inequality. You might want to avoid inequality either because you believe it is bad in itself, or because, as Kelton argues, it leads to power imbalances. If a select few individuals own most of the world’s assets, then those individuals also have a lot of power and influence in politics compared to the average worker. Nevertheless, the argument that wealth inequality is bad because that wealth is better put to work elsewhere fails when we adopt MMT.
Secondly, Kelton argues that MMT supports the idea of a ‘universal job guarantee’. It is unclear whether this is a prediction of MMT, or just an idea that many MMTers are on board with. Recall that according to MMT, spending does not lead to inflation when there is slack in the relevant sector of the economy. I found it hard to think of areas where there clearly is such slack: most companies already seem to be running a pretty tight ship. But Kelton sees slack everywhere, namely in the millions of unemployed people who want to work but cannot find a job. These people are unused resources in the economy. If all these people would work, we would collectively be better off. Therefore, Kelton proposes that the government literally promises to offer a job to everyone who wants one. These would typically be jobs in the ‘care economy’, such as health care or care for the elderly. Of course, the government would also provide the necessary training, free of charge. Since there is high demand in these areas, the additional workforce would not crowd out any already existing workers. But we would all benefit from better care.
I am a bit sceptical of this idea. It seems that Kelton simply ignores the downstream effect of a job guarantee. Even if it does not lead to inflation in the health care sector, it would probably lead to inflation elsewhere. After all, all those people who now find themselves with disposable income are going to buy food, watch a movie, get their car repaired. If those industries are already ‘running hot’, then the increased demand will lead to inflation there. Elsewhere in the book, Kelton seems aware that there is a trade-off between unemployment and inflation, as predicted by the monetary theory of inflation she critiques. But when she discusses the job guarantee plan she suddenly claims that we can employ everyone without any effect on inflation. I fail to see how that is an implication of MMT.
Finally, Kelton never mentions it but certain readers of this blog might have wondered: doesn’t MMT provide a nice argument in favour of a universal basic income (UBI)? I take it that the main case against UBI is simply that it would cost so much that the state could never sustainably afford it. But if MMT is right, then worries about the size of the national deficit are overblown; what matters are the real productive resources in the economy. And it does seem that a universal basic income would unleash some of those resources, as it allows people to continue education or retrain professionally, for instance. Should advocates of UBI rejoice? Unfortunately, I don’t think MMT really supports UBI, for the same reasons I don’t think it supports a universal job guarantee. The people who receive a UBI – on most versions, almost everyone – will want to spend that money on various goods and services, thereby increasing demand. This will lead to higher inflation, which in turn means that people’s basic income becomes worth less in real terms. This seems to induce an unsustainable vicious circle.
That may seem a pessimistic note to end on. As Kelton admits, MMT does not provide a free lunch – but does it provide anything at all? I do believe that it is a real advance in our understanding of the economy to realise that deficits themselves don’t matter. I’m also happy to buy Kelton’s argument that issuing additional currency does not always and immediately lead to inflation, since spending can also often increase productivity. So, after having read Kelton’s book I am probably slightly more in favour of higher government spending than I already was, and slightly less worried by rising deficits or government borrowing.
I also think it’s an important insight that inflation is always a local phenomenon. Some sectors of the economy are running hot; others have significant slack. If we would take MMT’s policy prescriptions to the extreme, that would essentially lead to a planned economy. The government would further stimulate sectors that don’t fully utilise the productive resources available, while it would carefully use taxes to attenuate demand in sectors that are already at capacity. But that approach is a good deal subtler than a universal job guarantee or a universal basic income, and probably a good deal less popular too. Short of a planned economy, the main lesson of MMT is that the government should spend more on those sectors of the economy which currently seem undervalued. We could think of the state as a smart investor: ‘buy’ industries which could produce more than is currently believed, and ‘sell’ industries which are right at their peak.
Overall, I thought The Deficit Myth was a good introduction to MMT, and from my non-expert perspective I find the general line of thought highly plausible. Kelton’s book is definitely aimed at a general audience, so don’t expect to come away with an understanding of the nitty-gritty of the theory. But if you would like to know more about MMT and are not already an economist, then The Deficit Myth is as good a starting point as any.