The squabbling over whether Modern Monetary Theory is a Copernican revolution for public policy or a load of nonsense hasn’t been very enlightening. The prevailing sentiment in both camps—“You just don’t get it” is a declaration of faith that hardly inspires confidence in economics as a science.

As it happens, I’m more open to MMT than you might expect, given my slavishly orthodox leanings. I think it asks good questions and proposes some clarifying thought experiments. I also think it ends up in entirely the wrong place.

Many years ago, at the height of policy makers’ interest in “monetarism,” I worked in the U.K. Treasury, helping produce short-term forecasts of the money supply. This exercise didn’t use the Treasury’s economic forecasting model but instead estimated financial flows using sectoral balance sheets, flows of funds and accounting identities. Everything added up; each sector’s surplus was another’s deficit. It was much like doing MMT, before the term existed.

This flow-of-funds perspective jarred violently with the standard textbook treatment. Our tables made it clear, for instance, that the monetary base (currency plus central-bank reserves) wasn’t a policy variable, as textbooks back then asked you to believe; it moved around to reflect the public sector’s deficit, among other things. And the spreadsheets showed how government spending could be self-financing—which is MMT’s central claim. Together with the central bank, the Treasury can create new money by the very act of spending it.

Accounting identities can’t be wrong. So MMT rightly asks, why is public borrowing a problem? Come to think of it, why does the government need to borrow (or collect taxes, for that matter) in the first place? The basic answer is it doesn’t need to. The issuer of a currency can issue as much as it likes. The only constraint, says MMT, is the capacity of the economy to meet demand. If the government spends too much, the excess of demand will cause inflation.

It all leads inexorably to the view that it is not possible for the government to spend “more than it can afford.” There’s no limit to what it can “afford.” The only concern is that its spending might turn out to be inflationary.

This is illuminating—and, up to this point, not really at odds with orthodox macroeconomics. Mainstream economists know that central banks can create money, and that’s why a household budget is a bad analogy for the government’s budget. On this, the two sides agree.

Yet mainstream economists also concern themselves with the sustainability of government finances. MMTers mock such talk—but they’re mistaken for two reasons. The first is that the risk of inflation is itself a constraint on unbounded government spending. The other is that one of the main purposes of a central bank—indeed, it is part of the U.S. Federal Reserve’s mandate—is to control inflation. MMTers ignore this fact, then fail to reckon with the implications.

The Fed is about to raise interest rates to curb inflation, which continues to surge. Using monetary policy to restrain demand means the Fed can’t permanently monetize government spending. This puts fiscal sustainability back on the table. Policy makers now have to ask: What is the outlook for growth, interest rates, deficits and debt? Suppose the Fed succeeds in containing inflation but the government maintains its spending. Will higher interest rates cause public debt to grow faster than the economy—faster than might be sustainable?

Granted, for MMT this question doesn’t arise—but not because of secret truths encoded in accounting identities. The reason is that MMT assumes that the Fed will play no role in fighting inflation. The MMT prescription for monetary policy is “leave the policy rate at zero and issue money as required.” So fiscal policy has to do everything. It has to realize the MMTers’ expansive ambitions for public spending—guaranteed employment, Green New Deal, universal health care, retirement security, tuition-free college, student debt forgiveness and every other highly affordable thing you can think of—and contain the resulting inflationary pressure.

How does MMT propose to do this? Perhaps in due course by raising taxes, though MMTers have mixed feelings about this (to acknowledge the need for tax increases is to suggest that not everything can be afforded). Maybe with direct controls on prices, wages, profits and credit. Perhaps by examining spending minutely, program by program, favoring outlays that mitigate inflationary pressure (say by boosting labor supply) and avoiding those that worsen it (for instance by adding to demand in overstretched sectors).

This is a tall order for a Congress that generally struggles to do anything, never mind execute a demanding new fiscal policy with foresight and technocratic exactitude.  

Mainstream economists have long advocated a division of labor between fiscal policy (subject to swirling political pressures, biased by short-term calculation toward inflationary excess) and monetary policy (delegated to independent central banks, charged with anchoring expected inflation). They had reason to. In removing the Fed from the picture, MMT ignores lessons learned over the course of decades.

One more thing: MMTers are right to say they’re proposing a radical change of regime. MMT neuters the institution tasked with controlling inflation, demands unimaginable levels of wisdom and competence in Congress, advocates more government spending, minimizes the costs of inflation, and views talk of fiscal discipline as gibberish.

If MMTers were ever put in charge, inflation expectations would surge before they even made a single policy announcement. Thus the only constraint on spending they acknowledge would tighten at once, throttling their ambitions and hobbling their program at the outset.

But, as I say, their ideas are really interesting.

Clive Crook is a Bloomberg Opinion columnist and writes editorials on economics, finance and politics. He was chief Washington commentator for the Financial Times, a correspondent and editor for the Economist and a senior editor at the Atlantic.