A Review of Stephanie Kelton’s The Deficit Myth
[Review of Stephanie Kelton, The Deficit Myth: Modern Monetary Theory and the Birth of the People's Economy (New York: PublicAffairs, 2020).]
Listen to the Audio Mises Wire version of this book review.
I’ve got good news and bad news. The good news is that Stephanie Kelton—economics professor at Stony Brook and advisor to the 2016 Bernie Sanders campaign—has written a book on modern monetary theory (MMT) that is very readable and will strike many readers as persuasive and clever. The bad news is that Stephanie Kelton has written a book on MMT that is very readable and will strike many readers as persuasive and clever.
To illustrate the flavor of the book, we can review Kelton’s reminiscences of serving as chief economist for the Democratic staff on the US Senate Budget Committee. When she was first selected, journalists reported that Senator Sanders had hired a “deficit owl”—a new term Kelton had coined. Unlike a deficit hawk or a deficit dove, Kelton’s deficit owl was “a good mascot for MMT because people associate owls with wisdom and also because owls’ ability to rotate their heads nearly 360 degrees would allow them to look at deficits from a different perspective” (Kelton 2020, p. 76).
Soon after joining the Budget Committee, Kelton the deficit owl played a game with the staffers. She would first ask if they would wave a magic wand that had the power to eliminate the national debt. They all said yes. Then Kelton would ask, “Suppose that wand had the power to rid the world of US Treasuries. Would you wave it?” This question—even though it was equivalent to asking to wipe out the national debt—“drew puzzled looks, furrowed brows, and pensive expressions. Eventually, everyone would decide against waving the wand” (Kelton 2020, p. 77).
Such is the spirit of Kelton’s book, The Deficit Myth. She takes the reader down trains of thought that turn conventional wisdom about federal budget deficits on its head. Kelton makes absurd claims that the reader will think surely can’t be true…but then she seems to justify them by appealing to accounting tautologies. And because she uses apt analogies and relevant anecdotes, Kelton is able to keep the book moving despite its dry subject matter. She promises the reader that MMT opens up grand new possibilities for the federal government to help the unemployed, the uninsured, and even the planet itself…if we would only open our minds to a paradigm shift.
So why is this bad news? Because Kelton’s concrete policy proposals would be an absolute disaster. Her message can be boiled down to two sentences (and these are my words, not an exact quotation): Because the Federal Reserve has the legal ability to print an unlimited number of dollars, we should stop worrying about how the government will “pay for” the various spending programs the public desires. If they print too much money we will experience high inflation, but Uncle Sam doesn’t need to worry about “finding the money” the same way a household or business does.
This is an incredibly dangerous message to be injecting into the American discourse. If it were mere inflationism, we could hope that enough of the public and the policy wonks would rely on their common sense to reject it. Yet because Kelton dresses up her message with equations and thought experiments, she may end up convincing an alarming number of readers that MMT really can turn unaffordable government boondoggles into sensible investments, just by changing the way we think about them.
Precisely because Kelton’s book is so unexpectedly impressive, I would urge longstanding critics of MMT to resist the urge to dismiss it with ridicule. Although it’s fun to lambaste “magical monetary theory” on social media and to ask, “Why don’t you move to Zimbabwe?” such moves will only serve to enhance the credibility of MMT in the eyes of those who are receptive to it. Consequently, in this review I will craft a lengthy critique that takes Kelton quite seriously in order to show the readers just how wrong her message actually is, despite its apparent sophistication and even charm.
In her introductory chapter, Kelton lures the reader with the promise of MMT and also sheds light on her book title:
[W]hat if the federal budget is fundamentally different than your household budget? What if I showed you that the deficit bogeyman isn’t real? What if I could convince you that we can have an economy that puts people and planet first? That finding the money to do this is not the problem? (Kelton 2020, p. 2, bold added)
The first chapter of the book makes the fundamental distinction for MMT, between currency issuers and currency users. Our political discourse is plagued, according to Kelton, with the fallacy of treating currency issuers like Uncle Sam as if they were mere currency users, like you, me, and Walmart.
We mere currency users have to worry about financing our spending; we need to come up with the money—and this includes borrowing from others—before we can buy something. In complete contrast, a currency issuer has no such constraints, and needn’t worry about revenue when deciding which projects to fund.
Actually, the situation is a bit more nuanced. To truly reap the advantages unlocked by MMT, a government must enjoy monetary sovereignty. For this, being a currency issuer is a necessary but insufficient condition. There are two other conditions, as Kelton explains:
To take full advantage of the special powers that accrue to the currency issuer, countries need to do more than just grant themselves the exclusive right to issue the currency. It’s also important that they don’t promise to convert their currency into something they could run out of (e.g. gold or some other country’s currency). And they need to refrain from borrowing…in a currency that isn’t their own. When a country issues its own nonconvertible (fiat) currency and only borrows in its own currency, that country has attained monetary sovereignty. Countries with monetary sovereignty, then, don’t have to manage their budgets as a household would. They can use their currency-issuing capacity to pursue policies aimed a maintaining a full employment economy. (Kelton 2020, pp. 18–19, bold added)
Countries with a “high degree of monetary sovereignty” include “the US, Japan, the UK, Australia, Canada, and many more” (Kelton 2020, p. 19). (And notice that even these countries weren’t “sovereign” back in the days of the gold standard, because they had to be careful in issuing currency lest they run out of gold.) In contrast, countries like Greece and France today are not monetarily sovereign, because they no longer issue the drachma and franc but instead adopted the euro as their currency.
The insistence on countries issuing debt in their own currency helps to explain away awkward cases such as Venezuela, which is suffering from hyperinflation and yet has the ability to issue its own currency. The answer (from an MMT perspective) is that Venezuela had a large proportion of its foreign-held debt denominated in US dollars, rather than the bolivar, and hence the Venezuelan government couldn’t simply print its way out of the hole. In contrast, goes the MMT argument, the US government owes its debts in US dollars, and so never need worry about a fiscal crisis.
Yes, Kelton Knows about Inflation
At this stage of the argument, the obvious retort for any postpubescent reader will be, “But what about inflation?!” And here’s where the critic of MMT needs to be careful. Kelton repeatedly stresses throughout her book—and I’ve seen her do it in interviews and even on Twitter—that printing money is not a source of unlimited real wealth. She (and Warren Mosler too, as he explained when I interviewed him on my podcast) understands and warns her readers that if the federal government prints too many dollars in a vain attempt to fund too many programs, then the economy will hit its genuine resource constraint, resulting in rapidly rising prices. As Kelton puts it:
Can we just print our way to prosperity? Absolutely not! MMT is not a free lunch. There are very real limits, and failing to identify—and respect—those limits could bring great harm. MMT is about distinguishing the real limits from the self-imposed constraints that we have the power to change. (Kelton 2020, p. 37, bold added)
In other words, when someone like Alexandria Ocasio-Cortez proposes a Green New Deal, from an MMT perspective the relevant questions are not, “Can the Congress afford such an expensive project? Will it drown us in red ink? Are we saddling our grandchildren with a huge credit card bill?” Rather, the relevant questions are, “Is there enough slack in the economy to implement a Green New Deal without reducing other types of output? If we approve this spending, will the new demand largely absorb workers from the ranks of the unemployed? Or will it siphon workers away from existing jobs by bidding up wages?”
The Fundamental Problem with MMT
Now that we’ve set the table, we can succinctly state the fundamental problem with Kelton’s vision: regardless of what happens to the “price level,” monetary inflation transfers real resources away from the private sector and into the hands of political officials. If a government project is deemed unaffordable according to conventional accounting, then it should also be denied funding via the printing press.
What makes MMT “cool” is that it’s (allegedly) based on a fresh insight showing how all of the mainstream economists and bean counters are locked in old habits of thought. Why, these fuddy-duddies keep treating Uncle Sam like a giant corporation, which has to make ends meet and always has to satisfy the bottom line. In contrast, the MMTers understand that the feds can print as many dollars as they want. It’s not revenue but (price) inflation that limits the government’s spending capacity.
I hate to break it to Kelton and the other MMT gurus, but economists—particularly those in the free market tradition—have been teaching this for decades (and perhaps centuries). For example, here’s Murray Rothbard in his 1962 treatise, Man, Economy, and State:
At this time, let us emphasize the important point that government cannot be in any way a fountain of resources; all that it spends, all that it distributes in largesse, it must first acquire in revenue, i.e., it must first extract from the “private sector.” The great bulk of the revenues of government, the very nub of its power and its essence, is taxation, to which we turn in the next section. Another method is inflation, the creation of new money, which we shall discuss further below. A third method is borrowing from the public. (Rothbard 1962, pp. 913–14, bold added)
To repeat, this is standard fare in the lore of free market economics. After explaining that government spending programs merely return resources to the private sector that had previously been taken from it, the economist will inform the public that there are three methods by which this taking occurs: taxation, borrowing, and inflation. The economist will often add that government borrowing can be considered merely deferred taxation, while inflation is merely hidden taxation.
And it’s not merely that inflation is equivalent to taxation. Because it’s harder for the public to understand what’s happening when government money printing makes them poorer, there is a definite sense in which standard taxation is “honest” whereas inflation is insidious. This is why Ludwig von Mises considered inflationary finance to be “essentially antidemocratic”: the printing press allows the government to get away with spending that the public would never agree to explicitly pay for through straightforward tax hikes.
Kelton and other MMT theorists argue that inflation isn’t a problem right now in the US and other advanced economies and so we don’t need to be shy about cranking up the printing press. But whether or not the Consumer Price Index is rising at an “unacceptably” high rate, it is a simple fact that when the government prints an extra $1 million to finance spending, then prices (quoted in US dollars) are higher than they otherwise would have been, and people holding dollar-denominated assets are poorer than they otherwise would have been. Suppose that prices would have fallen in the absence of government money printing. In this case, everybody holding dollar assets would have seen their real wealth go up because of the price deflation. If the government merely prints enough new dollars to keep prices stable, it’s still the case that those original dollar holders end up poorer relative to what otherwise would have happened.
Now to be sure, Kelton and other MMT theorists would object at this point in my argument. They claim that if there is still some “slack” in the economy, in the sense of unemployed workers and factories operating below capacity, then a burst of monetary inflation can put those idle resources to work. Even though the rising prices lead to redistribution, if total output is higher, then per capita output must be higher too. So, on average, the people still benefit from the inflation, right?
On this score, we simply have a disagreement about how the economy works, and in this dispute I think the Austrians are right while the MMTers are wrong. According to Mises’s theory of the business cycle, “idle capacity” in the economy doesn’t just fall out of the sky, but is instead the result of the malinvestments made during the preceding boom. So if we follow Kelton’s advice and crank up the printing press in an attempt to put those unemployed resources back to work, it will simply set in motion another unsustainable boom/bust cycle. In any event, in the real world, government projects financed by inflation won’t merely draw on resources that are currently idle, but will also siphon at least some workers and raw materials out of other, private sector outlets, as I elaborate in this article.
In summary, the fundamental “insight” of MMT—namely, that governments issuing fiat currencies need only fear price inflation, not insolvency—is something that other economists have acknowledged for decades. Where the MMTers do say something different is when they claim that printing money only carries an opportunity cost when the economy is at full employment. But on this point, the MMTers—like their more orthodox cousins the Keynesians—are simply wrong.
Tough Questions for MMT
A standard rhetorical move is for proponents to claim that MMT isn’t ideological, but merely describes how a financial system based on fiat money actually works. (For example, this was the lead argument Mike Norman used when he and I were dueling with YouTube videos.) Yet since so much hinges on whether a government has “monetary sovereignty,” it’s amazing that the MMTers never seem to ask why some governments enjoy this status while others don’t.
For her part, Kelton criticizes certain nonmonetarily sovereign governments for particular actions, such as joining a currency union (Kelton 2020, p. 145), but she doesn’t ask the basic question: Once an MMT economist explains its benefits, why doesn’t every government on earth follow the criteria for becoming a monetary sovereign? Indeed, why don’t all of us as individuals issue our own paper notes—in my case, I’d print RPMs, which has a nice ring to it—and furthermore only borrow from lenders in our own personal currencies? That way, if you fell behind in your mortgage payments, you could simply print up more of your own personal notes to get current with the bank.
Posed in this way, these questions have obvious answers. The reason Greece adopted the euro, and why Venezuela borrows so much in US dollar–denominated debt, and the reason I use dollars rather than conducting transactions in RPMs, is that the rest of the financial community is very leery of the Greek drachma, the Venezuelan bolivar, or the Murphyian RPM note. Consequently, the Greek and Venezuelan governments, as well as me personally, all subordinated our technical freedom to be “monetary sovereigns” and violated one or more of Kelton’s criteria.
In short, the reason most governments (including state governments in the US) in the world aren’t “monetary sovereigns” is that members of the financial community are worried that they would abuse a printing press. The Greek government knew its economy would receive more investment, and that it would be able to borrow on cheaper terms, if it abandoned the drachma and adopted the euro. The Venezuelan government knew it could obtain much larger “real” loans if they were denominated in a relatively hard currency like the USD rather than the Venezuelan currency, which could so readily be debased (as history has shown). And I personally can’t interest anybody in financial transactions involving my authentic RPM notes, and so, reluctantly, I have to join the dollar zone.
Now that we’ve covered this basic terrain, I have a follow-up question for the MMT camp: What would it take for a government to lose its monetary sovereignty? In other words, of those governments that are currently monetary sovereigns, what would have to happen in order for the governments to start borrowing in foreign currencies, or for them to tie their own currency to a redemption pledge, or even abandon their own currency and embrace one issued by a foreign entity?
Here again the answer is clear: a government that engaged too recklessly in monetary inflation—thus leading investors to shun that particular “sovereign” currency—would be forced to pursue one or more of these concessions in order to remain part of the global financial community. Ironically, current monetary sovereigns would run the risk of forfeiting their coveted status if they actually followed Stephanie Kelton’s policy advice.
MMT Is Actually Wrong about Money
For a framework that prides itself on neutrally describing the actual operation of money and banking since the world abandoned the gold standard, it’s awkward that MMT is simply wrong about money. In this section I’ll summarize three of the main errors Kelton makes about money.
Money Mistake #1: Contrary to MMT, the Treasury Needs Revenue before It Can Spend
A bedrock claim of the MMT camp is that unlike you, me, and Walmart, the US Treasury doesn’t need to have money before spending it. Here’s an example of Kelton laying out the MMT description of government financing:
Take military spending. In 2019, the House and Senate passed legislation that increased the military budget, approving $716 billion…There was no debate about how to pay for the spending…Instead, Congress committed to spending money it did not have. It can do that because of its special power over the US dollar. Once Congress authorizes the spending, agencies like the Department of Defense are given permission to enter into contracts with companies like Boeing, Lockheed Martin, and so on. To provision itself with F-35 fighters, the US Treasury instructs its bank, the Federal Reserve, to carry out the payment on its behalf. The Fed does this by marking up the numbers in Lockheed’s bank account. Congress doesn’t need to “find the money” to spend it. It needs to find the votes! Once it has the votes, it can authorize the spending. The rest is just accounting. As the checks go out, the Federal Reserve clears the payments by crediting the sellers’ account with the appropriate number of digital dollars, known as bank reserves. That’s why MMT sometimes describes the Fed as the scorekeeper for the dollar. The scorekeeper can’t run out of points. (Kelton 2020, p. 29, bold added)
For a more rigorous, technical treatment, the advanced readers can consult Kelton’s peer-reviewed journal article from the late 1990s on the same issues. Yet whether we rely on Kelton’s pop book or her technical article, the problem for the MMTers is still there: nothing in their description is unique to the US Treasury.
For example, when I write a personal check for $100 to Jim Smith, who also uses my bank, we could explain what happens like this: “Murphy instructed Bank of America to simply add 100 digital dollars to the account of Jim Smith.” Notice that this description is exactly the same thing that Kelton said about the Treasury buying military hardware in the block quotation above. (It’s true that Bank of America can’t create legal tender base money the way the Fed can; I plug that hole in the analogy a bit below with my Goldman Sachs example.)
Now of course, I can’t spend an unlimited amount of dollars, since I’m a currency user, not a monetary sovereign. In particular, if I “instruct” Bank of America to mark up Jim Smith’s checking account balance by more dollars than I have in my own checking account, the bank may ignore my instructions. Or, if my overdraft isn’t too large, the bank might go ahead and honor the transaction but then show that I have a negative balance (and charge me an insufficient funds fee on top of it).
The only difference between my situation and the US Treasury’s is that I actually have bounced checks and online payments before, whereas the US Treasury hasn’t. Indeed, Kelton’s own journal article shows that the Treasury consistently maintained (as of the time of her research) a checking account balance of around $5 billion and that the daily closing amount never dipped much below this level (Kelton 1998, p. 11, figure 4).
Indeed, the Treasury itself sure acts as if it needed revenue before it can spend. That’s why the Treasury secretary engages in all sorts of fancy maneuvers—such as postponing contributions to government employees’ retirement plans—whenever there’s a debt ceiling standoff and Uncle Sam hits a cash crunch.
The MMTers take it for granted that if the Treasury ever actually tried to spend more than it contained in its Fed checking account balance, the Fed would honor the request. Maybe it would, and maybe it wouldn’t; CNBC’s John Carney (who moderated the debate at Columbia University between MMT godfather Warren Mosler and yours truly) thinks it’s an open question in terms of the actual legal requirements, though Carney believes that in practice the Fed would go ahead and cash the check.
Yet, to reiterate, so far the Treasury has never tried to spend money that it didn’t already have sitting in its checking account. The MMT camp would have you believe that there is something special occurring day in and day out when it comes to Treasury spending, but they are simply mistaken: so far, at least, the Treasury has never dared the Fed by overdrawing its account.
Indeed, Kelton herself in her technical article from the late 1990s implicitly gives away the game when she defends the MMT worldview in this fashion:
[S]ince the government’s balance sheet can be considered on a consolidated basis, given by the sum of the Treasury’s and Federal Reserve’s balance sheets with offsetting assets and liabilities simply canceling one another out…the sale of bonds by the Treasury to the Fed is simply an internal accounting operation, providing the government with a self-constructed spendable balance. Although self-imposed constraints may prevent the Treasury from creating all of its deposits in this way, there is no real limit on its ability to do so. (Kelton 1998, p. 16, italics in original)
What Kelton writes here is true, but by the same token we can consider the Federal Reserve and Goldman Sachs balance sheets on a consolidated basis. If we do that, then Goldman Sachs can now spend an infinite amount of money. Sure, its accountants might still construct profit and loss statements and warn about bad investments, but these are self-imposed constraints; so long as the Fed in practice will honor any check Goldman Sachs writes, then all overdrafts are automatically covered by an internal loan from the Fed to the investment bank. The only reason this wouldn’t work is if the Fed actually stood up to Goldman and said no. But that’s exactly what the situation is with respect to the Treasury too.
Whenever I argue the merits of MMT, I debate whether or not to bring up this particular quibble, because wondering whether the Fed would actually cover a Treasury overdraft doesn’t get to the essence of what’s wrong with MMT. I’m actually sympathetic to the MMT claims that the Fed would be obligated to backstop the Treasury in all circumstances; it would be very naïve to think that the Fed actually enjoys “independence” from the federal government that grants the central bank its power. Furthermore, I believe that the various rounds of quantitative easing (QE) during the Obama years weren’t merely driven by a desire to minimize the output gap, but instead were necessary to help monetize the boatloads of new federal debt being issued. (Of course Trump and Powell are performing a similar dance.)
Even so, I think it’s important for the public to realize that the heroes of MMT are misleading them when they claim there is something unique to Uncle Sam in the way he interacts with his banker. So far, this is technically not the case. Even when the Fed has clearly been monetizing new debt issuance—such as during the world wars—all of the players involved technically have gone through the motions of having the Treasury first float bonds in order to fill its coffers with borrowed funds and only then spending the money. The innocent reader wouldn’t know this if he or she relied on the standard MMT accounts of how the world works.
Money Mistake #2: Contrary to MMT, Taxes Don’t Prop Up (Most) Currencies
Another central mistake in the MMT approach is its theory of the origin and value of money. (If you want to see the Austrian view, consult my article on the contributions of Menger and Mises.) To set the stage, here is Kelton explaining how Warren Mosler stumbled upon the worldview that would eventually be dubbed modern monetary theory:
Mosler is considered the father of MMT because he brought these ideas to a handful of us in the 1990s. He says…it just struck him after his years of experience working in financial markets. He was used to thinking in terms of debits and credits because he had been trading financial instruments and watching funds transfer between bank accounts. One day, he started to think about where all those dollars must have originally come from. It occurred to him that before the government could subtract (debit) any dollars away from us, it must first add (credit) them. He reasoned that spending must have come first, otherwise where would anyone have gotten the dollars they needed to pay the tax? (Kelton 2020, p. 24)
This MMT understanding ties in with its view of the origin of money and how taxes give money its value. Kelton explains by continuing to summarize what she learned from Mosler:
[A] currency-issuing government wants something real, not something monetary. It’s not our tax money the government wants. It’s our time. To get us to produce things for the state, the government invents taxes…This isn’t the explanation you’ll find in most economics textbooks, where a superficial story about money being invented to overcome the inefficiencies associated with bartering…is preferred. In that story, money is just a convenient device that sprang up organically as a way to make trade more efficient. Although students are taught that barter was once omnipresent, a sort of natural state of being, scholars of the ancient world have found little evidence that societies were ever organized around barter exchange.
MMT rejects the ahistorical barter narrative, drawing instead on an extensive body of scholarship known as chartalism, which shows that taxes were the vehicle that allowed ancient rulers and early nation-states to introduce their own currencies, which only later circulated as a medium of exchange among private individuals. From inception, the tax liability creates people looking for paid work…in the government’s currency. The government…then spends its currency into existence, giving people access to the tokens they need to settle their obligations to the state. Obviously, no one can pay the tax until the government first supplies its tokens. As a simple point of logic, Mosler explained that most of us had the sequencing wrong. Taxpayers weren’t funding the government; the government was funding the taxpayers. (Kelton 2020, pp. 26–27, bold added)
I have included these lengthy quotations to be sure the reader understands the superficial appeal of MMT. Isn’t that intriguing—Mosler argues that the government funds the taxpayers! And when you think through his simple point about debits and credits, it seems that he isn’t just probably correct, but that he must be correct.
Again, it’s a tidy little demonstration. The only problem is that it’s demonstrably false. It is simply not true that dollars were invented when some autocratic ruler out of the blue imposed taxes on a subject population, payable only in this new unit called “dollar.” The MMT explanation of where money comes from doesn’t apply to the dollar, the euro, the yen, the pound…Come to think of it, I don’t believe the MMT explanation applies even to a single currency issued by a monetary sovereign. All of the countries that currently enjoy monetary sovereignty have built their economic strength and goodwill with investors by relying on a history of hard money.
In a review of Kelton’s book, I’m not going to delve into the problems with the alleged anthropological evidence that purportedly shows that ancient civilizations used money that was invented by political fiat, rather than money that emerged spontaneously from trade in commodities. For that topic, I refer the interested reader to my review of David Graeber’s book.
Yet let me mention before leaving this subsection that the MMT story at best only explains why a currency has a nonzero value; it doesn’t explain the actual amount of its purchasing power. For example, if the IRS declares that every US citizen must pay $1,000 in a poll tax each year, then it’s true, US citizens will need to obtain the requisite number of dollars. But they could do so whether the average wage rate were $10 per hour or $10,000 per hour, and whether a loaf of bread cost $1 or $1,000.
Furthermore, other things equal, if the government lowers tax rates, then it strengthens the currency. That’s surely part of the reason that the US dollar rose some 50 percent against other currencies after the tax rate reductions in the early Reagan years. So the MMT claim that taxes are necessary, not to raise revenue (we have a printing press for that), but to prop up the value of the currency, is at best seriously misleading.
Money Mistake #3: MMT Confuses Debt with Money
Amazingly, even though their system claims to explain how money works, the MMTers apparently don’t know the simple difference between money and debt. Here’s Kelton trying to defuse hysteria over the national debt:
The truth is, we’re fine. The debt clock on West 43rd Street simply displays a historical record of how many dollars the federal government has added to people’s pockets without subtracting (taxing) them away. Those dollars are being saved in the form of US Treasuries. If you’re lucky enough to own some, congratulations! They’re part of your wealth. While others may refer to it as a debt clock, it’s really a US dollar savings clock. (Kelton 2020, pp. 78–79, bold added)
The part I’ve put in bold in the quotation above is simply wrong. And I don’t mean, “It’s wrong according to Austrian economics but right according to MMT.” No, even in the MMT framework, Kelton’s claim about the national debt is wrong. The outstanding federal debt would only correspond to “how many dollars [have been] added to people’s pockets without subtracting…them away” to the extent that the Federal Reserve had monetized the debt by taking the Treasury securities onto its own balance sheet. But to the extent that some of the outstanding Treasury debt is currently held by individuals and entities that aren’t the Federal Reserve, Kelton’s statement is simply wrong.
In the MMT framework, federal government spending creates new dollars, while taxing destroys them. Since a federal budget deficit refers to a situation where Uncle Sam spends more than he taxes, it’s understandable why Kelton concluded that the federal debt—which reflects the cumulative history of the net budget deficits and surpluses over time—is equal to the net number of dollars that Uncle Sam “spent into existence.”
But to repeat, this is wrong. Kelton forgot that when the Treasury floats new bonds, that action (in the MMT framework) also destroys dollars by removing them from the hands of the public. So if all of the outstanding Treasury debt were held by the public (or foreign central banks), then the cumulative federal budget deficits wouldn’t correspond to any net dollar creation, even in the MMT framework.
Stay with me; we have one more step: in the MMT framework (and the Austrian framework too, for that matter), when the Federal Reserve buys outstanding Treasury securities in the secondary market and takes them onto its balance sheet, this creates new dollars. Therefore, to the extent that the outstanding Treasury securities are sitting on the Fed’s balance sheet, then that portion of the national debt would correspond to “how many dollars [have been] added to people’s pockets without subtracting…them away.”
Does the reader see how cumbersome the MMT framework is? It led its chief proponent to make an elementary mistake in her attempt to explain the basics to the public. In contrast, coming from an orthodox background, I immediately knew Kelton’s claim was wrong, because borrowing money per se doesn’t create money. This is true whether corporations do it or whether Uncle Sam does it. (Just imagine $1 billion in actual currency and that the Treasury keeps issuing new $1 billion bonds to keep borrowing that same pile of green pieces of paper to continually respend them. This procedure would run up the national debt as much as we want, but at any moment there would still be the same $1 billion in currency.)
To drive home just how confused Kelton is on the difference between US Treasurys and US dollars, later in the book she writes, “Heck, I don’t even think we should be referring to the sale of US Treasuries as borrowing or labeling the securities themselves as the national debt. It just confuses the issue and causes unnecessary grief” (Kelton 2020, p. 81).
Here’s another way I can demonstrate that Kelton’s discussion is obviously missing something: if Kelton were right and the US national debt were a tally of how many dollars on net the government has “spent into existence,” then when Andrew Jackson paid off the national debt, the American people would have had no money—the last dollar would have been destroyed. And yet even Kelton doesn’t claim that dollars were temporarily banished from planet Earth, she merely claims that Jackson’s policy caused a depression. (For the Austrian take on this historical episode, see this article.)
For an even starker illustration of the MMT confusion between debt and money, consider Kelton’s approving quotations of a thought experiment from Eric Lonergan, who asked, “What if Japan monetized 100% of outstanding JGBs [Japanese government bonds]?” That is, What if the Bank of Japan issued new money in order to buy up every last Japanese government bond on earth? Lonergan argues that “nothing would change,” because the private sector’s wealth would be the same; the BOJ will have engaged in a mere asset swap. In fact, because their interest income would now be lower while their wealth would be the same, people in the private sector would spend less after the total debt monetization (!), according to Lonergan.
In response to these claims, I make a simple point: you can’t spend Japanese government bonds in the grocery store. That’s why money and debt are different things. If Lonergan were correct, then we could also go the other way: specifically, if the Japanese government issued enough bonds to absorb every last yen on planet Earth, then apparently Lonergan would have to say that aggregate demand measured in yen would go through the roof. Yet how could it, if nobody held any yen anymore? Remember, you can’t pay your rent or buy groceries with government bonds.
Do Government Deficits = Private Savings?
In chapter 4, Kelton lays out the MMT case that government deficits, far from “crowding out” private sector saving, actually are the sole source of net private assets. Using simple accounting tautologies, Kelton seems to demonstrate that the only way the nongovernment sector can run a fiscal surplus is if the government sector runs a fiscal deficit.
Going the other way, when the government is “responsible” by running a budget surplus and starts paying down its debt, by sheer accounting we see that this must be reducing net financial assets held by the private sector. (This is why it should come as no surprise, Kelton argues, that every major government surplus led to a bad recession [Kelton 2020, p. 96].)
In the present review, I won’t carefully review and critique this particular argument, as I’ve done so in this article. Suffice it to say that you could replace “government” in the MMT argument with any other entity and achieve the same outcome. For example, if Google borrows $10 million by issuing corporate bonds and then it spends the money, then the net financial assets held by The World Except Google go up by precisely $10 million. (Or rather, the way you define terms in order to make these claims true is the same way Kelton gets the MMT claims about Uncle Sam to go through.) So did I just prove something really important about Google’s finances?
Obviously something is screwy here. Using standard definitions, people in the private sector can save, and even accumulate net financial wealth, without considering the government sector at all. (I spell all of this out in this article.) For example, Robinson Crusoe on his deserted island can “save” out of his coconut income in order to finance his investment of future labor hours into a boat and net. Even if we insist on a modern financial context, individuals can acquire shares of equity in new corporations, thus acquiring assets that don’t correspond to a “debit” of anyone else.
It is a contrived and seriously misleading use of terminology when MMT proponents argue that government deficits are a source of financial wealth for the private sector. Forget the accounting and look at the big picture: even if the central bank creates a new $10 million and simply hands it to Jim Smith for free, it hasn’t made the community $10 million richer—except in the nominal sense in which we could all be “millionaires” with this practice. Mere money creation doesn’t make any more houses or cars or acres of arable farmland available. Printing new money doesn’t make the community richer. At best it’s a wash with redistribution, and in fact in practice it makes the community poorer by distorting the ability of prices to guide economic decisions.
The MMT Job Guarantee
The last item I wish to discuss is the MMT job guarantee. Strictly speaking, this proposal is distinct from the general MMT framework, but in practice I believe every major MMT theorist endorses some version of it.
Under Kelton’s proposal, the federal government would have a standing offer to employ any worker at $15 per hour (Kelton 2020, p. 68). This would set a floor against all other jobs; Kelton likens it to the Federal Reserve setting the federal funds rate, which then becomes the base rate for every other interest rate in the economy.
Kelton argues that her proposal would eliminate the unnecessary slack in our economic system, where millions of workers languish in involuntary unemployment. Furthermore, she claims that her job guarantee would raise the long-term productivity of the workforce and even help people find better private sector job placement. This is because currently “Employers just don’t want to take a chance on hiring someone who has no recent employment record” (ibid., p. 68).
There are several problems with this proposal. First of all, why does Kelton assume that it would only draw workers out of the ranks of the unemployed? For example, suppose Kelton set the pay at $100 per hour. Surely even she could see the problem here, right? Workers would be siphoned out of productive private sector employment and into the government realm, providing dubious service at best at the direction of political officials.
Second, why would employers be keen on hiring someone who has spent, say, the last three years working in the guaranteed job sector? These would be, by design, the cushiest jobs in America. Kelton admits this when she says that the base wage rate would be the floor for all other jobs.
Looking at it another way, it’s not really a job guarantee if it’s difficult to maintain the position. In other words, if the people running the federal jobs program are allowed to fire employees who show up drunk or who are simply awful workers, then it’s no longer a guarantee.
Stephanie Kelton’s new book The Deficit Myth does a very good job explaining MMT to new readers. I must admit that I was pleasantly surprised at how many different topics Kelton could discuss from a new view, in a manner that was simultaneously absurd and yet apparently compelling.
The problem is that Kelton’s fun book is utterly wrong. The boring suits with their standard accounting are correct: it actually costs something when the government spends money. The fact that since 1971 we have had an unfettered printing press doesn’t give us more options, it merely gives the Fed greater license to cause boom/bust cycles and redistribute wealth to politically connected insiders.