Nominal GDP as an indicator

Many of my contrarian opinions stem from my focus on a single macroeconomic variable: the NGDP.

Take the recent period of high inflation. Nearly all economists believe that inflation was caused by a combination of supply and demand side shocks. In contrast, I believe that the high inflation was solely on the demand side, with supply shocks not playing a role at all, at least over the period 2019-2024 as a whole.

Consider some data from the past 4 1/4 years:

Under the 4% NGDP target, NGDP should have increased by 18.1% between 2019:Q4 and 2024:Q1. The actual increase was 29.0%.

At a PCE inflation target of 2%, prices should rise by 8.8% between January 2020 and April 2024. The actual increase was 17.8%.

Note that the NGDP increase of as much as 11% led to 9% above target inflation. That means supply shocks don’t explain any of this totally cumulative excess inflation. Yes, supply shocks clearly played a role during certain months in 2022. But those negative shocks were offset by positive supply shocks during other months. The supply side of the economy has been strong, with real GDP rising faster than expected, mainly due to immigration. Given the NGDP growth rate, we are fortunate that inflation was not even slightly higher. A positive supply shock (an increase in immigration) kept inflation at a level slightly lower than what would have been predicted from NGDP growth alone.

My contrarian views on the role of monetary policy in recent inflation reflect some similarly heterodox views on the Great Recession. I argued that the Great Recession was caused by tight money policies in 2008. Very few economists agree with me. When I claim that the Great Recession was caused by a large drop in the NGDP, people accuse me of using a tautology. According to them, a big drop in the NGDP is a recession. They confuse nominal and real GDP.

The last four and a quarter years clearly show that real and nominal GDP are not identical: a large excess of NGDP was reflected in excessive inflation, and not in very fast RBBP growth. So much for the ‘tautology’ theory.

Another complaint is that while declining NGDP was a problem in 2008, there was nothing the Fed could have done about it because we were stuck at the zero lower bound. But in 2008 we were not yet at the zero lower bound; the Fed pursued normal conventional monetary policy. In October 2008, they introduced IOR to prevent interest rates from falling, that is, to prevent the economy from overheating.

Why do my views differ so much from those of my colleagues? I see several factors.

1. If you weren’t expecting an increase in inflation, it’s natural to look for some unexpected factor to explain the result. Supply shocks are a convenient excuse, especially since they contributed to higher-than-normal inflation for a short period. But this is reasoned reasoning. Economists often overlook that the economy is also continually hit by positive supply shocks, such as the surge in immigration or the recovery of supply lines after Covid disruption subsided. They rightly saw negative supply shocks during certain months, but failed to see that the supply situation as a whole has been excellent over the past 4 1/4 years.

2. Most economists are relatively positive about the Fed’s monetary policy. So when the NGDP deviates dramatically from the 4% growth path, they are reluctant to blame monetary policy. That would almost be like blaming the economics profession for the policy disaster. It is much more satisfying to look for explanations that involve mysterious ‘exogenous shocks’.

3. The stance of monetary policy is often very different from what it looks like when we look at indicators such as interest rates. Rates fell in 2008, even as money became increasingly tight. Interest rates rose dramatically in 2022, even as monetary policy remained quite expansionary (although arguably slightly less than in 2021). If you misjudge the stance of monetary policy, you are much more likely to misdiagnose the cause of the recession or high inflation. This error is especially likely to occur when an exogenous factor (such as a housing market slump) causes a large change in the natural rate of interest, making the Fed’s policy rate a highly inaccurate indicator of the actual policy stance.

My focus on nominal GDP also explains why I’m not impressed by unconditional forecasts. I notice that many people who were right about the inflation of the early 1920s were wrong about the effects of the earlier quantitative easing programs under Bernanke. (And vice versa.) I’m much more impressed with it conditional predictions. What do you think would happen if the Fed allowed 29% NGDP growth in the 4 1/4 years after 2019:Q4? That’s the kind of question we need to think about.

While the NGDP is a useful indicator, inflation and interest rates are not. If you tell me that inflation is rising, I don’t know what that means for the economy, without knowing whether the increase is due to supply or demand shocks. If you tell me that interest rates will go lower, that means nothing unless I know whether the drop in interest rates is due to easy money or a weak economy.

Only NGDP provides an unequivocal indication of the current state of the economy. It doesn’t tell us everything we need to know, especially in the long run. But over the short to medium term, no other variable comes close as a way to understand current macroeconomic conditions.

There are times when economists are tempted to ignore the NGDP’s signals. Do not do that! Back on June 28, 2021, Jason Furman was interviewed by David Beckworth. Here’s Furman:

So I have some sympathy for the nominal GDP target. . . . If we had followed it now, we would have lifted interest already. And we will, with extreme probability, exceed the nominal GDP target we set.

So under you [Beckworth’s] within the framework you would have to compensate for that with a sustained period of lower nominal GDP growth than the trend. I don’t mean to bully you, this experience has destroyed anyone’s plans they wrote down before. It’s such a strange period. But to me that says, “I would like the Fed to take that into account, if the unemployment rate is still 5.5% a year from now, regardless of what happens to nominal GDP or prices.” an isolated problem and issue that they must take into account.” So I think everything should have a bipartisan mandate, but are you looking at nominal GDP and things like that, rather than inflation? Maybe.

Ouch! In June 2021, NGDP just returned to its pre-Covid trendline. In retrospect, this was the perfect time to tighten policies to prevent an NGDP overshoot. To his credit, Furman rightly suspected that tightening would be necessary to prevent an overshoot of the NGDP, but for other reasons he thought that was an unwise idea. He felt that the NGDP was sending a misleading signal that we should have looked at the unemployment rate (which is actually an unreliable indicator).

Looking back, we can clearly see that the NGDP signal was spot on and that Furman was wrong. It was time to tighten up.

Ignore NGDP at your peril.