Fiscal shocks, inflation and the Lucas Critique

Why I'm skeptical of fiscal theories of inflation

In a recent paper, William Kinlaw, Mark Kritzman, Michael Metcalfe and David Turkington argued that fiscal stimulus was the major cause of the post-Covid inflation. This graph is from their paper:

On one level, I’m pleased with this finding. The authors find that supply shocks played only a minor role in the post-Covid inflation. I’ve argued that the rapid growth in nominal GDP during 2021-22 strongly suggests that the primary problem was excessive aggregate demand, not supply restrictions.

I am surprised that they found little evidence that money supply played an important role. The M2 money supply rose by 40% between February 2020 and February 2022, by far the fastest two-year growth in modern history:

To be clear, I don’t view M2 as the best indicator of the stance of monetary policy, rather I prefer NGDP growth. But I’m in the minority, and M2 is the much more conventional measure of monetary policy. Unfortunately, economists have never really solved the identification problem in macroeconomics, and hence even in 2026 we continue to debate the same issues that were being contested back in the 1960s—what is the relative importance of fiscal and monetary policy?

One problem is that there is no generally accepted definition of “causation”. Consider a bus that goes off a twisty mountain road with no guardrail. What caused the accident? One person could claim that the accident would not have occurred if the road had an adequate guardrail. Another person could claim that the accident was caused by a driver that was not sufficiently cautious. It is not obvious that either claim is “wrong”.

Was the 2021-22 inflation caused by excessive fiscal stimulus, or by the failure of the Fed to offset the stimulus with appropriate monetary policy? Perhaps both claims are true.

Let’s think about the fact that the 2020-21 fiscal stimulus had the sort of inflationary impact predicted by Keynesian models, but the 2013 fiscal austerity did not have the contractionary impact predicted by the very same models. Why is that?

The 2013 fiscal austerity did not occur during a period where policymakers viewed slower growth in NGDP as being desirable. Hence, the Fed offset the impact of the austerity with easier money, and the economy failed to slow as predicted. In contrast, the 2020-21 fiscal stimulus occurred in an environment where faster growth in NGDP was widely viewed as desirable, by both monetary and fiscal policymakers. In that environment, the Fed choose not to offset the expansionary impact of fiscal stimulus. In retrospect, both fiscal and monetary policymakers erred with excessively expansionary policies in 2021, but at the time the policy was viewed as appropriate.

If I am correct, then this suggests that the impact of fiscal policy will depend on the zeitgeist, the attitude of monetary and fiscal policymakers toward growth in nominal spending. The Fed will largely offset the impact of fiscal policy on nominal spending when the Fed doesn’t view that impact as being desirable. Historical studies of correlations between fiscal policy and inflation will not have reliable policy implications, for the same reason that historical studies of the correlation between inflation and unemployment from 1879 to 1968 had misleading policy implications for a world of unconstrained fiat money.

In 1976, Robert Lucas explained why the Phillips Curve was an unreliable guide to policymakers. Studies that found a negative correlation between inflation and unemployment were mostly looking at periods of time where authorities were not trying to manipulate the inflation rate to influence employment. Recall that between 1879 and 1968, the dollar was almost always fixed to gold at either $20.67/oz. or $35/oz. Under that policy regime, you cannot permanently lower unemployment with higher rates of steady state inflation. If temporary increases in inflation have an expansionary impact, it is largely because they are unanticipated. (As an aside, NGDP growth is a better variable than inflation when doing these sorts of Phillips Curve studies.)

Even if the 2020-21 fiscal stimulus did have an inflationary impact, that’s no reason to assume that a similar fiscal program adopted today would have the same sort of impact. Having seen what went wrong in 2021-22, today’s Fed would be far more likely to offset the impact of bigger budget deficits with tighter money, something they did not do in 2021.

Another recent study does find some evidence of causation running from fiscal shocks to inflation. Gabriel P. Fritsch and J. Zachary Mazlish’s new paper found strong evidence that positive fiscal shocks are inflationary. Here is the abstract:

We introduce a new methodology for identifying high-frequency fiscal shocks using Large Language Models. We apply this method to 1947-2025 US data. Our results show that the model successfully mimics a "professional forecaster" of the current and future US fiscal position, and is able to recover similar shocks to what have already been identified in the narrative fiscal shock literature. We then examine the effects of fiscal shocks on asset prices: in response to a 1pp shock to the present-value of the current and next ten-years deficits, ten-year Treasury yields rise more than 30bps, with real yields and break-even inflation expectations both contributing to the rise. The dollar appreciates significantly — as much as 4.8% — and the 2Y-10Y spread rises 16-24bps. Turning to macroeconomic outcomes, our fiscal shocks produce government spending multipliers in the 0.5-1 range. Tax shocks shows strong signs of anticipation, and using our data to account for anticipation, we find that output and consumption fall by more than 2% in anticipation of a 1% of GDP tax cut. The multiplier for an anticipated tax shock is 1.2, smaller than typical estimates.

I like the approach they use, even though their findings conflict with my “monetary dominance” view of macroeconomics. By looking at market responses to policy surprises, they are able to address the identification problem that has made it so difficult to establish causality. Assuming the findings hold up in future research, this study seems to clearly indicate that fiscal stimulus has a positive impact on inflation.

Nonetheless, there is an important difference between statistical significance and economic significance. For instance, look at the biggest fiscal policy shocks during the period since WWII, from the Fritsch and Mazlish paper:

Notice that the election of Reagan was by far the largest positive shock, twice the size of the second largest (Trump’s first election.) And yet during the Reagan administration, we saw the largest disinflation of my lifetime. Did the disinflation occur because markets misjudged Reagan’s fiscal policy? Not at all—budget deficits did increase sharply during the 1980s due to a combination of much higher military spending and sharply lower tax rates. The positive fiscal shock that was predicted after Reagan was elected did in fact occur.

And yet inflation fell sharply during the 1980s, even as many Keynesian economists predicted that inflation would increase. The explanation is simple; Reagan’s fiscal expansion was more than offset by a much tighter monetary policy, which brought inflation down from the double digits of 1979-81 to approximately 4% during 1982-89. Two things can both be true:

A. Reagan’s fiscal stimulus boosted inflation, other things equal.

B. Monetary policy was by far the dominant factor determining the path of inflation during the 1980s.

Also notice that during the 1970s, the two largest fiscal shocks were both contractionary—the Nixon budget and the Ford budget. And yet any story of the 1970s will focus on the extremely high (and rising) inflation of the period, far worse than the recent post-Covid inflation. Even if the two budget shocks did have a contractionary impact at the margin, the effects were completely overwhelmed by the extremely expansionary Fed monetary policy of 1965-81. Again, statistical significance does not always imply economic significance.

In a world where the Fed targets inflation at 2% but also cares about unemployment, supply shocks may have a temporary impact on the CPI. But non-monetary demand shocks should be fully offset by monetary policy, at least if the Fed is doing its job. Any (undesirable) impact of (demand-side) fiscal policy on inflation would result from the Fed failing to properly do its job. This means that any model showing how fiscal policy affects inflation will implicitly be a model of monetary policy failure.

I have no problem with studies that show that fiscal policy action X is correlated with monetary policy failure Y. Just don’t expect that study to provide reliable policy guidance to future fiscal policymakers.

PS. The Fritsch and Mazlish paper makes this claim:

Ten days after a shock that increases the present discounted-value of the current and next ten-year’s expected deficits over GDP by 1pp, the dollar appreciates 4.8%, ten-year nominal Treasury yields rise 46 basis points (bps), and ten year real yields are 34bps higher.

That implies a 12-basis point rise in inflation expectations, or a 20-basis point increase from the Reagan shock (which was 1.7 pp). Actual inflation fell by roughly 800 basis points.

PPS. Off topic, I was amused to see this headline in the OC Register:

Why do 31% of Americans want a housing crash?

And this data:

The survey found that 37% of renters who were rooting for a crash said a price collapse would improve their odds of buying a home. Just 12% of owners felt the same way.

Conversely, 39% of owners hoping for a housing crash wanted the lower property taxes a drop would create, compared with 15% of renters.

Of course, a housing crash caused by lower demand does not make housing more “affordable”, as we saw when a deep recession and tighter lending standards caused less housing to be built in 2008. On the other hand, a fall in house prices due to more supply does make housing more affordable, as we recently saw in Austin, Texas.

In other words, NRFPC

Speaking of housing, recent attempts to ban corporate ownership of housing are just one more piece of evidence that we are living in a new dark age of economics. They turned a YIMBY bill into a NIMBY bill. Both the left and the right have completely lost touch with reality. As I keep saying, “affordability” isn’t about prices, it’s about output. The way things are going, it may take decades to get back to the sensible neoliberalism of the 1990s.