Robin Hanson has a very interesting blog post discussing the fact that authorities do not update their forecasts as frequently as an optimal forecaster would:

The best estimates of a maximally accurate source would be very frequently updated and follow a random walk, which implies a large amount of backtracking. And authoritative sources like WHO are often said to be our most accurate sources. Even so, such sources do not tend to act this way. They instead update their estimates rarely, and are especially reluctant to issue estimates that seem to backtrack. 

The random walk observation reminded me of a column I wrote for The Hill back in 2018, criticizing the way the Fed conducts monetary policy:

In December 2015, the Fed raised its interest rate target for the first time in more than a decade. At the time, Fed officials signaled another four rate increases for 2016 in order to prevent the economy from overheating. Unfortunately, the economy quickly slowed, and the Fed would not raise rates again until the following December. Indeed by early 2016, it was clear that the previous rate increase should not have occurred.

So why didn’t officials immediately admit their mistake and cut rates back to zero in early 2016? It turns out that the Federal Open Market Committee (FOMC) has been exceedingly reluctant to quickly reverse course after setting out on a new policy track.

This is very different from how asset markets work. If stocks fall on Tuesday due to worries about a trade war, the market is not at all reluctant to shoot right back up again on Wednesday, if new information makes that scenario less likely. Unlike Fed policymakers, markets don’t have egos and they don’t worry about “credibility.”

Then I suggested a reform to make Fed policy more accurate:

Given modern technology, there is no reason why the Fed can’t adjust its settings far more frequently. FOMC members could, for example, email their preferred interest rate target to Chairman Powell each business day, and the actual Fed target could be set at the median vote.

Instead of quarter-point increments, FOMC members could select an interest rate target to the nearest “basis point”—which is 1/100th of 1 percent. Daily movements would look more like a financial asset price, moving up and down each day in a sort of “random walk” as new information about the economy comes in.

Toward the end of the article, I cited an example of what can go wrong when Fed officials are reluctant to shift course:

Consider the following quotation from the minutes of a November 1937 Fed meeting. In this meeting, Fed officials considered reversing their earlier decision to raise reserve requirements — a decision now blamed for helping to trigger a severe double-dip depression in late 1937, throwing millions out of work:

We all know how it developed. There was a feeling last spring that things were going pretty fast… If action is taken now it will be rationalized that, in the event of recovery, the action was what was needed and the System [Fed] was the cause of the downturn. It makes a bad record and confused thinking. … I would rather not muddy the record with action that might be misinterpreted.

Here we see a Fed official arguing that a reversal of the previous mistake would be taken as evidence of Fed guilt — especially if it was successful!