MMT and Inflation: What Modern Monetary Theory Actually Says

MMT is treated either as the doctrine of unlimited money printing or as a self-evident truth about monetary sovereignty. Both readings miss the structure of what its proponents actually claim — and what its critics actually contest.

Modern Monetary Theory rests on three load-bearing claims about monetary sovereignty, the relationship between taxes and currency, and a job guarantee as price anchor. The mainstream critique does not deny the accounting identities; it disputes the causal relationships and the policy implications drawn from them.

Reading MMT through its slogans — “deficits don’t matter,” “we can always print money” — distorts what Mitchell, Wray and Kelton actually wrote. Reading the mainstream rejection through Twitter polemics distorts what Mankiw, Summers and Galí actually argued. This article reconstructs both sides without picking one, and isolates the single point on which both camps converge: inflation is the ultimate binding constraint.

The three load-bearing pillars of MMT

The first pillar is the source of most popular misreadings. The MMT claim is that a sovereign monetary issuer is not financially constrained the way a household or a corporate borrower is, because it cannot involuntarily default on liabilities denominated in the currency it issues. The framework does not claim there is no constraint — it claims the binding constraint is real-resource availability and inflation, not solvency. A detailed treatment can be found in our extended treatment of inflation regime architecture. Wray (2015, “Modern Money Theory: A Primer on Macroeconomics for Sovereign Monetary Systems”) is explicit: “the question is never whether the government can afford to spend, but whether the spending will trigger inflation.”

The second pillar inverts the textbook story about money. Rather than money emerging from barter as a medium of exchange, the chartalist account holds that the state imposes a tax liability denominated in a unit of account it controls, then issues tokens that discharge the liability. Citizens demand the tokens to pay the tax. Tymoigne and Wray (2013) trace the historical evidence for this view through colonial currency arrangements and the late-nineteenth-century gold-standard transitions. The implication for inflation analysis is that the demand for money is anchored by fiscal mechanisms rather than by money-supply targeting.

The third pillar — the Job Guarantee — is the one most explicitly inflation-relevant. Kelton (2020, “The Deficit Myth”) argues that traditional NAIRU-based stabilisation uses unemployment as the inflation buffer, while a Job Guarantee uses public employment at a fixed wage as the buffer. When the private sector overheats and bids workers out of the public scheme, inflation pressure rises and the public-employment pool shrinks. When the private sector contracts, displaced workers absorb into the scheme at the fixed wage, anchoring the bottom of the labour market.

The mainstream critique: identities versus causality

Mankiw (2020, “A Skeptic’s Guide to Modern Monetary Theory”, AEA Papers and Proceedings) provided the cleanest statement of the orthodox objection. He accepted the accounting identities — a sovereign issuer cannot involuntarily default in its own currency — but contested the causal mapping from accounting to policy. The accounting identity that “government spending equals taxes plus borrowing plus money creation” is true but does not establish that money creation is a free policy lever. The expectations channel, the central bank’s reaction function, and the term-structure of fiscal credibility all matter for whether deficit-financed spending translates to inflation.

Summers (2019, op-eds in Washington Post and Bloomberg) framed the objection in stronger terms, calling MMT “voodoo economics” and warning that the framework underestimated the inflation risk of large fiscal expansions. His prediction in early 2021 that the American Rescue Plan would produce sustained above-target inflation was directionally vindicated by the 2021-2022 inflation surge, although the relative weight of demand stimulus, supply disruption and energy shock in the inflation print remains contested. Inflation and the State isolates the mechanical channels through which deficit financing transmits to prices.

Galí and Roeger (2024, ECB working paper) ran the comparison most carefully. They formalised the MMT framework inside a New Keynesian DSGE model and showed that, once inflation is treated as the binding constraint, MMT prescriptions converge with optimal Ramsey policy under reasonable parameterisations — meaning the substantive policy disagreement is smaller than the rhetorical disagreement. Where MMT and the New Keynesian framework genuinely differ is on the optimal monetary-fiscal mix during disinflations and on the role of the central bank’s independence as a credibility device.

The convergence point: inflation as ultimate constraint

The 2021-2022 U.S. inflation episode functioned as an unintentional test of both frameworks. The American Rescue Plan injected roughly USD 1.9 trillion of stimulus into an economy with closing output gaps; CPI inflation peaked at 9.1% in June 2022. MMT proponents argued the inflation signal validated their inflation-as-constraint emphasis (Wray 2022 commentary, Tcherneva 2022) — the policy mistake was insufficient real-resource scaling, not the principle of fiscal expansion. Mainstream economists argued the same data demonstrated the inadequacy of MMT’s anti-inflation toolkit: when inflation rose, the Fed had to deliver a 525-basis-point tightening cycle that no MMT-aligned policy framework had a clean answer to. Inflation and bonds traces the bond-market response that any framework must contend with.

What MMT does not say

The framework is sometimes accused of denying that bond markets matter. The accusation misreads the position. MMT proponents argue that for a monetary sovereign, bond issuance is a monetary-policy operation (draining excess reserves to maintain interest-rate targets) rather than a fiscing-funding operation. The policy implication is not that bond markets do not matter, but that they matter through different channels than household-budgeting analogies suggest. Whether one accepts the reframing depends on whether one views the central bank’s reaction function as a binding constraint or as a policy choice.

Critics also accuse MMT of ignoring exchange-rate constraints. The MMT response is that the framework applies most cleanly to large open economies issuing reserve currencies (U.S., Japan, U.K., Eurozone in part); for small open economies with foreign-currency-denominated liabilities, the constraint set is materially different, as Mitchell-Wray-Watts 2019 chapter 23 explicitly acknowledges. Negative real rates and their consequences documents one of the conditions under which the constraint set has been notably loose in recent decades. Eichengreen (2000, “Capital Flows and Crises”) provides the canonical reading on how the constraint tightens for non-reserve-currency issuers — Argentina’s repeated currency crises, the 1997 Asian episode, and the 1998 Russian default all illustrate the asymmetry.

Where this leaves the analytical comparison

MMT and the mainstream agree that inflation is the binding constraint on fiscal expansion. They disagree about (i) whether sovereign solvency is a meaningful prior constraint or a non-issue for currency issuers; (ii) whether the central bank’s reaction function should be treated as exogenous or as a policy variable; (iii) whether a Job Guarantee is a superior automatic stabiliser to NAIRU-based monetary policy. Each of these is a genuine policy question, separable from the rhetorical packaging that surrounds the MMT debate. Inflation and savings and core vs headline inflation document the empirical inflation framework that any policy claim — MMT or mainstream — must reckon with.

For the broader fiscal-monetary framework that situates this debate, see the next satellite on fiscal policy and inflation, the cluster pillar at the complete inflation guide, the underlying regime mapping at the inflation regimes pillar, and the empirical context in breakeven inflation rates.

Last updated — 7 May 2026