The Manhattan Institute’s Allison Schrager has a brave new article in the City Journal titled “Scrap the Dot Plot,” a directive aimed at the Federal Reserve. The chart Schrager has in mind is the Fed’s quarterly interest rate projection, the one that plots the Federal Open Market Committee members’ views on the appropriate target range for the federal funds rate.

Schrager argues that “All the dot plot really does is make plain that many economic variables are entirely unpredictable and that the Fed does not have a better crystal ball than anyone else.” Schrager is right and she deserves credit for having the courage to say it.

The bigger problem, though, is that even if the Fed kept the dot plot under wraps, the same problem would exist. The Fed has no forecasting superpowers. And that’s a problem for a few reasons.

First, the Fed’s policy stance depends on other economic forecasts, including those for GDP, inflation, unemployment, and potential output. More importantly, though, the Fed’s inability to outperform everyone else in forecasting these economic outcomes strongly suggests that it doesn’t control them.

On the surface, this statement shouldn’t be too controversial. If the Fed did precisely control inflation, unemployment, GDP, and interest rates, then it would simply conduct monetary policy to avoid recessions and inflation. Interest rates would not be 3 percentage points above the target the FOMC set in 2021. But, of course, we do still have recessions and the Fed has a hard time hitting its inflation and interest rate targets.

None of this is to say that the Fed doesn’t influence economic outcomes, but there’s huge gap between influence and control. And, of course, the Fed’s influence might not always be so helpful.

So even though few policymakers seem interested, it’s probably worth questioning how much monetary policy influences economic outcomes. And while people give the Fed an enormous amount of credit for “controlling” the economy, there’s mounting evidence to question this premise.

For instance, although recessions were more frequent in the pre-World War I era than after World War II, this comparison omits roughly 30 years that include the Great Depression. When the entire Fed period is compared to the full pre-Fed period, it looks like a wash.

Still, even when the interwar period is excluded, some evidence suggests that the average length of recessions, as well as the average time to recover from recessions, has been slightly longer in the post-World War II era than in the pre-Fed era.

Regarding inflation, the Fed has enjoyed a decent reputation largely because of the so-called Great Moderation, the period between the mid-1980s and the 2008 financial crisis. Again, though, the overall evidence suggests that this reputation is worth reconsidering.

For example, while variability in the rate of inflation declined in the post-World War II era, the average rate of inflation is much higher than it was before the founding of the Fed in 1913. Even focusing on just the post-World War II period, when inflation variability was cut approximately in half, it is not at all clear that the tradeoff was worthwhile because the average inflation rate was more than four times higher. (In fact, several academics (see here and here) have attributed this period more to simple good luck rather than good policy.)

So, using just a few basic statistics, the Fed’s track record on inflation and business cycles was questionable even before the inflation spike after the COVID-19 pandemic.

It is true, of course, that modern macroeconomic analysis has much more to offer than just these simple statistics. However, as my colleague Jai Kedia’s work has demonstrated, the most sophisticated tools available suggest that people are putting much too much faith in the Fed.

One of Jai’s papers, for instance, uses a monetary structural vector autoregression. Its results suggest that “monetary policy explains only a small fraction of the variability in inflation.” This finding holds using multiple inflation measures at several different time horizons and during multiple periods from 1960 through 2019.

The results suggest that most inflation variability is due to supply constraints, not monetary policy.

Because this SVAR approach is on the simpler side of modern macro analysis, it is open to many valid criticisms. To address such concerns, Jai used a benchmark model that is commonly used for such analyses, the Smets and Wouters (2007) framework. (Formally, this approach uses a dynamic stochastic general equilibrium model with Bayesian methods.)

These results confirm those found using the less sophisticated approach. Specifically, the results show that monetary policy is the least important driver of inflation, with supply factors (and often demand factors) explaining much more of the variability in inflation.

Combined, these studies, along with the basic statistics, make a strong case that the public places too much faith in the Fed’s ability to manage inflation and the broader economy. People give the institution way too much credit and, to be fair, too much blame.

Some may find this odd, but it’s good news that market forces have more to do with inflation than monetary policy. Policymakers should focus on ways to strengthen those forces, such as giving the Fed less authority and making it more transparent and predictable.

Giving the Fed more discretion to stabilize the economy all but ensures government officials will screw it up.