Pounds of Salt: The Fed’s Epistemic Arrogance

Published on June 17, 2021

“The dots are not a great forecaster of future rate moves. And that’s not because—it’s just because it’s so highly uncertain. There is no great forecaster of the future—so dots to be taken with a big, big grain of salt.”

— Federal Reserve Chairman Jerome Powell, Press Conference on June 16, 2021

Once upon a time, Federal Reserve policy-setting meetings were not followed by press conferences or even explicit statements regarding changes in policy. Ben Bernanke began the practice of holding press conferences in April 2011. This followed a significant enhancement in policy setting disclosures announced in November 2007. If sunlight is the best disinfectant, the hope was that allowing market participants to gain better insight into the Fed’s thinking would have salutary effects on market function. There would be fewer instances of miscommunication leading to market volatility and, more importantly, less risk to the real economy since everyone would know what to expect in advance.

The level of disclosure and commentary from the Federal Reserve has gone into overdrive in recent years and especially over the past year as the economy suffered the effects of the COVID-19 pandemic. Aside from the Fed’s self-imposed “quiet period” in the days leading up to policy meetings, rarely a day goes by when a Fed official does not appear on financial television or makes a speech to some audience. Those unfortunate enough to pay attention to financial news during the trading day are presented with schedules of Fed officials appearing in various venues along with charts of the S&P 500 as traders react to the latest utterances.

The Federal Reserve appears more and more to be the genesis of market-moving news rather than a dispassionate observer of economic indicators and market sentiment. It is as if the Federal Reserve wishes to test its theories and trial balloons out on the market to see how it reacts and uses this information to set policy. Make no mistake, financial market participants have long been obsessed with Fed policy but prior to the new hyperactive communication strategy, much had to be inferred and the obsessing was concentrated during the periods surrounding policy meetings. Now, the obsession is constant. If the goal of the Fed since 2007 has been to reduce market volatility by being more transparent, has anyone bothered to study whether this goal has been achieved?

Epistemic Arrogance

Nassim Nicholas Taleb’s books are some of the few that I have re-read from cover to cover multiple times. While recently reading The Black Swan, I kept thinking about the Federal Reserve.

Taleb’s concept of epistemic arrogance involves the hubris we have regarding the limits of our knowledge. It is not that we know nothing or that our knowledge about the world does not increase over time. The trouble is that people tend to consistently overestimate what they know and, more importantly, what it is possible to know.

This problem appears to be endemic in human society. For example, Taleb cites a study where the researcher asked a room full of people to estimate a range of possible values for some statistic, such as the population of a country, with 98 percent confidence. The study was not attempting to test knowledge, but what people thought about the accuracy of their knowledge. One would expect the error rate from such an experiment to be around 2 percent, but in reality it is far higher:

“This experiment has been replicated dozens of times, across populations, professions, and cultures, and just about every empirical psychologist and decision theorist has tried it on his class to show his students the big problem of humankind: we are simply not wise enough to be trusted with knowledge. The intended 2 percent error rate usually turns out to be between 15 and 30 percent, depending on the population and the subject matter”

The black swan, p. 139-140

If the problem is severe in society at large, it is at epidemic levels in the world of finance and economics. It is trivially easy to create a set of assumptions about the world and build spreadsheets that use these assumptions to project, with seemingly precise accuracy, the state of a company or an economy years or decades into the future. The trouble is not only that we are likely wrong about the future but that small variances in key inputs can result in radically different outcomes in non-linear systems that compound over time.

A modern market economy is an extremely complex system and we only see the results retrospectively. If we do not replace epistemic arrogance with epistemic humility, we will pay the consequences. For market actors with money on the line, the consequences involve financial gain and loss. It is less clear what the consequences are for Federal Reserve officials.

Chairman Powell’s Attempt at Epistemic Humility

Am I beating a dead horse here? Isn’t the quote at the beginning of the article a sign that the Chairman of the Federal Reserve in fact does have a healthy level of epistemic humility? It is not my intent to vilify Jay Powell or to be unfair to the Federal Reserve, so let’s take an extended excerpt from the press conference transcript and read the full question and response that included the “grain of salt” quip:

QUESTION: Thanks, Chair Powell. Your economic projections today forecast 7 percent growth in 2022, unemployment at 4 ½ percent, and core inflation of 3 percent. If those conditions are achieved by the end of the year, would that constitute substantial further progress, in your mind?

And kind of more broadly, when you look at the sort of median forecast for interest rates in 2023 showing not one but two interest rate increases at the time, I mean, is this kind of—can you describe the sort of tone of the—of the discussion in the committee? And are we really moving towards sort of a post-pandemic stance? Is there greater confidence that, you know, the recovery will be a—you know, a full recovery sooner than expected?

MR. POWELL: Well, on your first question, the judgment of when we have arrived at substantial further progress is one that the committee will make, and it would not be appropriate for me to lay out particular numbers that do or do not—that do or do not qualify. That is, you know, the process that we’re beginning now. At the next meeting, we will begin meeting by meeting to assess that progress and talk about what we—what we think we’re seeing, and just do all of the things that you do to sort of clarify your thinking around the process of deciding whether and how to adjust the pace and composition of asset purchases.

In terms of the two hikes, so let me say a couple things first of all, not for the first time, about the—about the dot plot. These are, of course, individual projections. They’re not a committee forecast. They’re not a plan. And we did not actually have a discussion of whether liftoff is appropriate at any particular year because discussing liftoff now would be—would be highly premature. It wouldn’t make any sense.

In our—if you look at the transcripts from five years ago, you’ll see that sometimes people mention their rate path in their interventions. Often, they don’t.

And the last thing to say is the dots are not a great forecaster of future rate moves. And that’s not because—it’s just because it’s so highly uncertain. There is no great forecaster of the future—so dots to be taken with a big, big grain of salt.

However—so let me talk about this meeting. The committee spelled out, as you know, in our FOMC statements the conditions that it expects to see before an adjustment in the target range is made. And it’s outcome-based. It’s not time-based. And as I mentioned, it’s labor-market conditions consistent with maximum employment, inflation at 2 percent, and on track to exceed 2 percent. And the projections give some sense of how participants see the economy evolving in their most likely case.

And honestly, the main message I would take away from the SEP is that participants—many participants are more comfortable that the economic conditions in the committee’s forward guidance will be met somewhat sooner than previously anticipated. And that would be a welcome development. If such outcomes materialize, it means the economy will have made faster progress toward our goals.

So the other thing I’ll say is rate increases are really not at all the focus of the committee. The focus of the committee is the current state of the economy. But in terms of our tools, it’s about asset purchases. That’s what we’re thinking about. Liftoff is well into the future. The conditions for liftoff—we’re very far from maximum employment, for example. It’s a consideration for the future.

So the near-term thing is really—the real near-term discussion that we’ll begin is really about the path of asset purchases. And as I mentioned, we had a discussion about that today and expect to, at future meetings, continue to see—think about our progress.

WSJ: Fed Chief Powell’s Post meeting Press Conference, June 16, 2021

I suspect most readers know what Chairman Powell is referring to when he mentions the “dots”, but for the benefit of others, here is the latest “dot plot” released by the Fed. Each circle indicates the value of an individual committee member’s judgment of the midpoint of the appropriate target range for the federal funds rate at the end of the specified calendar year or in “the longer run”:

Source: Federal Reserve’s Summary of Economic Projections, June 16, 2021

Why does the Fed release this dot plot, along with the other assumptions in the Summary of Economic Projections document that is released after each policy meeting? The idea is that this information provides market participants with guidance regarding what interest rate policy will look like during the years in the graph as well as in the undefined “longer run”.

Since interest rates are a crucially important variable that impacts the valuation of all financial assets in the economy and also has major impacts on capital investment decisions, the Fed is hoping to give market participants the confidence to make decisions with greater insight into the likely progression of the Fed funds rate. The progression of the Fed funds rate over time also helps to guide market participants when it comes to the market clearing rate for other fixed income securities. For example, the two year treasury note yield is heavily influenced by what market participants think the Fed funds rate will be over the course of the next two years.

Those who review the Summary of Economic Projections document will see forecasts for many additional economic variables, most notably the expected rate of inflation, real GDP growth, and unemployment. Each member of the policy setting committee provides estimates for these variables and the committee as a whole presumably is informed by a composite of these individual estimates. The two-day meeting is ostensibly meant to refine these forecasts as the Fed’s army of staff economists furnish more detailed economic data and projections to committee members.

As Chairman Powell admits, the dot plot cannot really be used to estimate the future course of interest rate policy and indeed must be taken not only with big grains of salt, but with heaps of salt. Let’s see what the dot plot revealed after the December 15-16, 2020 policy meeting just six months ago:

Source: Federal Reserve’s Summary of Economic Projections, December 16, 2020

Clearly, the strength of the economy over the first half of 2021 caught the Fed by surprise. The Fed was certainly not alone. The degree to which the economy has come roaring back as COVID-19 vaccines have rolled out was not expected by most market participants. The strength of the economy and a resurgence in inflation has led committee participants to anticipate increases in the Fed funds rate much sooner than they expected just six months ago.

Everyone’s a Critic

Who am I to write an article criticizing expert economists who happened to be wrong about the course of economic growth and inflation over the past six months? Could I have done any better?

The answer is no!

The point of this article is not to criticize the Federal Reserve for being wrong about the course of economic activity during a pandemic that has no historical parallel in a modern economy but to question whether the Fed is suffering from epistemic arrogance. When questioned about this, Chairman Powell seems to express some humility but his organization seems addicted to continuing to produce overly precise forecasts that everyone should know will have no bearing on reality.

Why do they do this?

One reason is the institutional imperative that exists when you have armies of highly educated experts armed with PhDs who need to justify their existence. But it appears that the problem is more insidious than that. I will point to a comment at the end of Chairman Powell’s answer regarding “asset purchases”. What was he referring to?

The Federal Reserve will continue to increase its holdings of Treasury securities by at least $80 billion per month and of agency mortgage‑backed securities by at least $40 billion per month until substantial further progress has been made toward the Committee’s maximum employment and price stability goals. These asset purchases help foster smooth market functioning and accommodative financial conditions, thereby supporting the flow of credit to households and businesses.

FOMC Press release, June 16, 2021

During the depths of the pandemic crisis last year, the Federal Reserve embarked on an unprecedented effort to ensure that financial markets would not freeze up and part of the program involved massive purchases of treasury securities and mortgage-backed securities. Despite the fact that the economy has been booming in recent months and shows no signs of illiquidity, the Fed has insisted on continuing these asset purchases at a rate of $80 billion per month of treasury securities and $40 billion per month of mortgage-backed securities.

The Fed claims that it is doing this in order to “help foster smooth market functioning” which implies that the goal is to ensure that markets do not freeze up. However, anyone who has followed financial markets even superficially this year understands that the system is awash in liquidity. Rather than attempting to ensure liquidity, the Fed is a major actor in the treasury and mortgage-backed securities markets. There is no doubt that by doing so the Fed is attempting to influence longer term interest rates.

Warren Buffett has likened the effect of interest rates to “financial gravity” because the level of interest rates influences the valuation of all other financial assets. The Federal Reserve’s actions, whatever their motivation, has had some impact on asset prices over the past year. One can argue about the extent to which the Fed has caused asset prices to increase or whether we have asset price bubbles, but it is hard to argue that there has not been a directional impact.

But You Have to Have Models…

I am not advocating that the Federal Reserve fire its economists and give up on making economic projections entirely. However, I do question the utility of communicating these projections through countless speeches, policy statements, and press conferences that give the appearance of jawboning the market. For one thing, anyone who looks at the Fed’s track record can see that the projections have no particular credibility. We have more transparency than ever before, but more and more it looks like the “man behind the curtain” resembles the Wizard of Oz.

The Federal Open Market Committee is made up of twelve members who meet eight times per year and are supported by the army of economists I keep referring to. No matter how brilliant these minds are or how aligned their intentions are with fulfilling the mandates given to them by Congress, I will suggest that the forecasts of these individuals are next to useless when it comes to what they are trying to do which is “set market expectations”.

Rather than obsessing over setting market expectations, the Federal Reserve might want to invert their process and observe the market — a free and un-manipulated market. The Fed should stop interfering in the treasury and mortgage backed security markets — the argument of fostering liquidity and smooth market functioning is long past. Then, the Fed can observe the level of longer term rates that are being set by actual market participants with skin in the game.

Are market participants any better at forecasting than the Fed? Perhaps not individually, and maybe not even collectively. However, I would rather observe the level of interest rate and inflation expectations by looking at a free and un-manipulated yield curve and infer what thousands of market participants are thinking because they have their skin in the game. In free markets, participants who consistently have bad judgment will exit the game while more competent individuals take their place. There is no such self-correcting mechanism for the Federal Reserve, either among FOMC committee members or their staffs.

Risk takers with capital on the line will exit the game due to incompetence. Federal Reserve officials who appear on financial television constantly might exit their particular game after a period of time, but they will not suffer negative effects from their inaccurate views. Career academics, especially well-spoken ones of the type who appear on financial television, will always have career opportunities in the lucrative world of Wall Street. And if not, there are always speaking fees and board seats.

Epistemic humility is in short supply in general and I would suggest that despite Chairman Powell’s seeming humility, the actions of the entity he runs suffers from chronic epistemic arrogance and is in need of significant reform.

Pounds of Salt: The Fed’s Epistemic Arrogance