What’s Wrong With the Model?

I think that Joseph C. Sternberg misses some very important factors in his latest Wall Street Journal column critiquing the Federal Reserve:

Jerome Powell owes us an explanation. The Federal Reserve chairman this week confirmed what investors already had guessed: Surprisingly persistent inflation is dissuading the Fed from cutting its short-term policy rate as soon and perhaps as quickly as Wall Street had hoped.

It’s the right call. The Fed committed its worst error in 40 years when it acted far too slowly to tame inflation following the pandemic. Its institutional credibility—on which hangs a lot in a fiat-money system—now depends on Mr. Powell’s success in suppressing that inflation.

The problem is that the central bank keeps making new versions of the same old mistake, granted with a better policy outcome this time around. It’s still getting its inflation forecasts badly wrong and then acting on those forecasts in ways that exacerbate confusion in markets and the broader economy.

concluding:

While we wait, Mr. Powell needs to make policy today. What to do? Note here that one of the things that makes the Fed’s broken economic models so embarrassing is that the central bank keeps talking about them. Predictions are central to the Fed’s forward guidance—the press conferences, wordy policy statements and quarterly dot plots about future interest rates by which the Fed seeks to guide financial markets. Were it not for all this forward guidance, we wouldn’t know what the central bank’s models have been erroneously predicting in recent years.

Mr. Powell increasingly acts as if he understands this. One of his achievements over the past year has been to convince markets that concrete new data points such as the recent inflation uptick matter more to the Fed than the often bogus projections spit out by its computers. Yet the Powell Fed still relies on forward guidance to an unhealthy degree, a legacy of the Ben Bernanke and Janet Yellen eras. To adapt the old saw, perhaps if you don’t have something right to say, don’t say anything.

The first factor is that unlike the situation in, say, the early 1980s, monetary policy and fiscal policy are working at cross-purposes. That’s not unusual—it’s almost always the case. What was unusual is that in the 1980s that was not the case. The second factor is closely related to that one: the markets expected and expect a spending spree, i.e. monetary and fiscal policy will continue to work at cross-purposes.

The third factor is how different circumstances are now than they were in the 1980s, so different that I’m not sure how anyone could realistically expect Fed actions to have the same effects in the same timeframe as they did then. Just to cite one example of the differences in 1980 there were 192 banks with assets over $1 billion and more than 12,000 with assets less than $100 million. Now there are 250 with assets of $6 billion or more and about 4,500 banks in total. Banking has seen enormous consolidation over the last 40 years.

That isn’t the only difference. In 1980 the ratio of debt to GDP was about 30%; now it’s around 120%. Empirically, that has been demonstrated to make a significant difference in an economy’s performance. What effect does that level of debt overhang have on the Federal Reserve’s ability to control inflation with interest rates? We can guess but we don’t really know.

The last factor I want to mention is that inflation is a lagging indicator. Here’s a graph of the M2 money supply over time:

M0 is money in circulation plus commercial bank reserve balances. From the Richmond Federal Reserve:

M1 is defined as the sum of currency in circulation, demand deposits at commercial banks, and other liquid deposits; it is often referred to as “narrow money.” M2 is everything included in M1 plus savings accounts, time deposits (under $100,000), and retail money market funds. M3 is everything in M2 plus larger time deposits and institutional money market funds. (Because the cost of estimating M3 was thought to outweigh its value, the Fed stopped reporting it in 2006.)

What do you see when you look at the graph above keeping in mind that inflation is a lagging indicator? I conclude that the inflation that began to show up in the first quarter of 2021 was probably a consequence of the spending in excess of aggregate product in 2020. And we kept spending in excess of the increase in aggregate product. We’re still doing it.

I could go on listing factors. It’s a wonder that the Fed’s models reflect the behavior of the real economy at all.