The powerful and prestigious Federal Reserve is having a tough year in 2022 in at least three ways:

• It has failed with inflation forecasting and performance;
• It has giant mark-to-market losses in its own investments and looming operating losses;
• It is under political pressure to do things it should not be doing and that should not be done at all.

Forecasting inflation

As everybody knows, the Fed’s overoptimistic inflation forecasts for the runaway inflation year of 2021 were deeply embarrassing. Then the Fed did it again for 2022, with another wide miss. In December 2021, it projected 2022 Personal Consumption Expenditures inflation at 2.6%, while the reality through June was 6.8%, with Consumer Price Index inflation much higher than that. It would be hard to give the Fed anything other than a failing grade in its supposed area of expertise.

The Fed’s interest rate forecast for 2022 was three federal funds target rate increases of 0.25%, so that its target rate would reach 0.9% by the end of 2022. It forecast the rate at 2% by the end of 2024. Instead, by July 2022, it already reached 2.5%.

In short, the Federal Reserve cannot reliably forecast economic outcomes, or what the results of its own actions will be, or even what its own actions will be. Of course, neither can anybody else.

It is essential to understand that we cannot expect any special economic or financial insight from the Federal Reserve. This is not because of any lack of intelligence or diligence, or not having enough computers or PhDs on the payroll, but of the fundamental and inevitable uncertainty of the economic and financial future. Like everybody else, the Fed has to make decisions in spite of this, so it will unavoidably make mistakes.

We should recall how Ben Bernanke, then Chairman of the Federal Reserve, accurately described his extended ‘QE’ strategy in 2012 as “a shot in the proverbial dark”. That was an honest admission, although unfortunately he admitted it only within the Fed, not to the public.

The Governor of the Reserve Bank of Australia (their central bank) described the bank’s recent inflation forecasting errors as “embarrassing”. Such a confession would also be becoming in the Federal Reserve, especially when the mistakes have been so obvious. The current fed funds rate of 2.5% may sound high today, if you have become used to short-term rates near zero and you have no financial memory. But it is historically low, and as many have pointed out, it is extremely low in real terms. Compared to CPI inflation of 8.5%, it is a real interest rate of negative 6%.

Savers will be glad to be able to have the available interest rates on their savings rise from 0.1% to over 2%, but they are still rapidly losing purchasing power and having their savings effectively expropriated by the government’s inflation.

Although in July and August 2022 (as I write), securities prices have rallied from their lows, the 2022 increases in interest rates have let substantial air out of the Everything Bubble in stocks, speculative stocks in particular, SPACs, bonds, houses, mortgages, and cryptocurrencies that the Fed and its fellow central banks so assiduously and so recklessly inflated.

A mark-to-market insolvent Fed

Nowhere are shrunken asset prices more apparent than in the Fed’s own hyper-leveraged balance sheet, which runs at a ratio of assets to equity of more than 200. As of 31 March 2022, the Fed disclosed, deep in its financial statement footnotes, a net mark-to-market (MTM) loss of $330bn on its investments. Since then, the interest rates on 5 and 10-year Treasury notes are up about an additional one-half percent. With an estimated duration of 5 years on the Fed’s $8tr of long-term fixed rate investments in Treasury and mortgage securities, this implies an additional loss of about $200bn in round numbers, bringing the Fed’s total MTM loss to over $500bn.

Compare this $500bn loss to the Fed’s total capital of $41bn. The loss is 12 times the Fed’s total capital, rendering the Fed technically insolvent on a mark-to-market basis. Does a MTM insolvency matter for a fiat currency-printing central bank? An interesting question – most economists argue such insolvency is not important, no matter how large. What do you think, candid Reader?

The Fed’s first defense of its huge MTM loss is that the loss is unrealised, so if it hangs on to the securities long enough it will eventually be paid at par. This would be a stronger argument in an unleveraged balance sheet, which did not have the Fed’s $5tr of floating rate liabilities. With the Fed’s leverage, however, the unrealized losses suggest that it has operating losses to come, if the higher short-term interest rates implied by current market prices come to pass.

The Fed’s second defense is that it has changed its accounting so that realized losses on securities or operating losses will not affect its reported retained earnings or capital. Instead, the resulting debits will be hidden in a dubious ‘deferred asset’ account. Just change the accounting! (This is exactly what the insolvent savings and loans did in the 1980s, with terrible consequences.)

What fun it is to imagine what any senior Federal Reserve examiner would tell a bank holding company whose MTM losses were 12 times its capital. And what any such examiner would say if the bank proposed to hide realised losses in a ‘deferred asset’ account instead of reducing its capital!

Here is a shorthand way to think about the dynamics of how Fed operating losses would arise from their balance sheet: The Fed has about $8tr in long-term, fixed-rate assets. It has about $3tr in non-interest-bearing liabilities and capital. Thus, it has a net position of $5tr of fixed rate assets funded by floating rate liabilities. (In other words, inside the Fed is the financial equivalent of a giant 1980s savings and loan.)

Given this position, it is easy to see that pro forma, for each 1% rise in short-term interest rates, the Fed’s annual earnings will be reduced by about $50bn. What short-term interest rate would it take to wipe out the Fed’s profits? The answer is 2.7%. If their deposits and repo borrowings cost 2.7%, the Fed’s profits and its contribution to the US Treasury will be zero. If they cost more than 2.7%, as is called for in the Fed’s own projections, the Fed starts making operating losses.

How big might these losses be? In the Fed staff’s own recent projections, in its most likely case, the projected operating losses add up to $60bn. This is 150% of the Fed’s total capital. In the pessimistic case, losses total $180bn, over four times its capital, and the Fed makes no payments to the Treasury until 2030.

In such cases, should the Fed’s shareholders, who are the commercial banks, be treated like normal shareholders and have their dividends cut? Or might, as is clearly provided in the Federal Reserve Act, the shareholders be assessed for a share of the losses? These outcomes would certainly be embarrassing for the Fed and would be resisted.

The central bank of Switzerland, the Swiss National Bank (SNB), did pass on its dividends in 2013, after suffering losses. The SNB Chairman gave a speech at the time, saying in effect, “Sorry! But that’s the way it is.” Under its chartering act, the SNB – completely unlike the Fed – must mark its investment portfolio to market in its official profit and loss statement. Accordingly, in 2022 so far, the SNB has reported a net loss of $31bn for the first quarter and a net loss of $91bn for the first six months of this year.

The Swiss are a serious people, and also serious, it seems, when it comes to central bank accounting and dividends.

In the U.S., the Federal Reserve Bank of Atlanta did pass its dividend once, in 1915. As we learn from the Bank’s own history, “Like many a struggling business, it suspended its dividend that year.” Could it happen again? If the losses are big and continuing, should it?

A third Fed defense is that its “mandate is neither to make profits nor to avoid losses.” On the contrary, the Fed is clearly structured to make seigniorage profits for the government from its currency monopoly. While not intended by the Federal Reserve Act to be a profit maximizer, it was also not intended to run large losses or to run with negative capital. Should we worry about the Fed’s financial issues, or should we say, “Pay no attention to the negative capital behind the curtain!”

What the Fed can and can’t do well

The Federal Reserve is also suffering from a push from the current administration and the Democratic majority in the House of Representatives to take on a politicized agenda, which it probably can’t do well and more importantly, should not do at all.

This would include having the Fed practice racial preferences, that is, racial discrimination, and as the Wall Street Journal editors wrote, “Such racial favoritism almost certainly violates the Constitution. So does the [House] bill’s requirement that public companies disclose the racial, gender identity and sexual orientation of directors and executives.” The bill would “politicize monetary policy and financial regulation.” A lot of bad ideas.

Moreover, the Federal Reserve already has more mandates than it can accomplish, and its mandates should be reduced, not increased.

As has been so vividly demonstrated in 2021 and 2022, the Fed cannot accurately forecast economic outcomes, and cannot know what the results of its own actions, or its “shots in the proverbial dark,” will be.

Meanwhile, it is painfully failing to provide its statutory mandate of stable prices. Note that the statute directs the goal of ‘stable prices’, not the much more waffly term the Fed has adopted, ‘price stability’. It has defined for itself that ‘price stability’ means perpetual inflation at 2% per year.

The Fed cannot ‘manage the economy’. No one can.

And the Fed is obviously unable to guarantee financial stability. No one can do that, either. Moreover, by trying to promote stability, it can cause instability, an ironic Minskian result—Hyman Minsky was the insightful theorist of financial fragility who inspired the slogan, ‘stability creates instability’.

There are two things the Fed demonstrably does very well.

The first is financing the government. Financing the government of which it is a part is the real first mandate of all central banks, especially, but not only, during wars, going back to the foundation of the Bank of England in 1694. As the history of the Atlanta Fed puts it so clearly:

During the war [World War I], the Fed was introduced to a role that would become familiar…as the captive finance company of a US Treasury with huge financing needs and a compelling desire for low rates.

This statement is remarkably candid: “the captive finance company of [the] US Treasury.” True historically, true now, and why central banks are so valuable to governments.

The second thing the Fed does well is emergency funding in a crisis by creating new money as needed. This was its original principal purpose as expressed by the Federal Reserve Act of 1913: “to furnish an elastic currency,” as they called it then. This it can do with great success in financial crises, as shown most recently in 2020, and of course during wars, although not without subsequent costs.

However, the Fed is less good at turning off the emergency actions when the crisis is over. Its most egregious blunder in this respect in recent years was its continuing to stoke runaway house price inflation by buying hundreds of billions of dollars of mortgage securities, continuing up to the first quarter of 2022. This has severe inflationary consequences as the cost of shelter drives up the CPI and erodes the purchasing power of households. Moreover, it now appears the piper of house price inflation is exacting its payment, as higher mortgage rates are resulting in falling sales and by some accounts, the beginning of a housing recession.

Among the notions for expanding the Fed’s mandates, the worst of all is to turn the Fed into a government lending bank, which would allocate credit and make loans to constituencies favored by various politicians. As William McChesney Martin, the Fed Chairman 1951-1970, so rightly said when this perpetual bad idea was pushed by politicians in his day, it would “violate a fundamental principle of sound monetary policy, in that it would attempt to use the credit-creating powers of the central bank to subsidize programs benefitting special sectors of the economy.”

It is a natural and permanent temptation of politicians to want to do just that—to use the money printing power of the central bank to give money to their political supporters without the need for legislative approval or appropriation, and to surreptitiously finance it by imposing an inflation tax without legislation.

One way to achieve this worst outcome would be to have the Fed issue a ‘central bank digital currency’ (CBDC) that allows everybody to have a deposit account with the Fed, which might then become a deposit monopolist as well as a currency monopolist.

Should that happen, the Fed would by definition have to have assets to employ its vastly expanded deposit liabilities. What assets would those be? Well, loans and securities. The Fed would become a government lending bank.

The global experience with such government banks is that they naturally lend based on politics, which is exactly what the politicians want, with an inevitable bad ending.

On top of that, with a CBDC in our times of Big Data, the Fed could and probably in time would choose to know everybody’s personal financial business. This could and perhaps would be used to create an oppressive “social credit system” on the model of China which could control credit allocation, loans, and payments. Given the urge to power of any government and of its bureaucratic agencies, that outcome is certainly not beyond imagining, and is, in my view, likely.

The current push to expand the Fed’s mandates is consistent with Shull’s Paradox, which states that the more blunders the Fed makes, the more powers and prestige it gets. But we should be reducing the Fed’s powers and mandates, not increasing them. Specifically:

• The Fed should not hold mortgage securities or mortgages of any kind. It should take its mortgage portfolio not just to a smaller size, but to zero. Zero was just where it was from 1914 to 2008 and where it should return.
• The Fed should not engage in subsidizing political constituencies and the proposed politicized agenda should be scrapped.
• Congress should definitively take away the Fed’s odd notion that the Fed can by itself, without Congressional approval, set a national inflation target and thereby commit the country and the world to perpetual inflation.
• Congress should repeat its instruction to the Fed to pursue stable prices. As Paul Volcker wrote in his autobiography, “In the United States, we have had decades of good growth without inflation,” and “The real danger comes from encouraging or inadvertently tolerating rising inflation and its close cousin of extreme speculation and risk taking.”
• The Fed should be required to have sound accounting for its own financial statements, with no hiding losses in the former savings and loan style allowed. This requires taking away from the Fed the power to set its own accounting standards, which nobody else has.
• The Fed should be prohibited from buying TIPS (Treasury Inflation-Protected Securities), because this allows it to manipulate apparent market inflation expectations.
• The funding of the Consumer Financial Protection Bureau expenses out of Fed profits should be terminated. This is an indefensible use of the Fed to take away the power of the purse from Congress and to subsidize a political constituency. It would be especially appropriate to end this payment if the Fed is making big losses.
• Finally, and most important of all, we must understand the inherent limitations of what the Federal Reserve can know and do. There is no mystique. We must expect it to make mistakes, and sometimes blunders, just like everybody else.

Knowing this, perhaps the 2020s will give us the opportunity to reverse Shull’s Paradox.

This paper is based on the author’s remarks at the American Enterprise Institute conference, ‘Is It Time to Rethink the Federal Reserve?’, 26 July 2022. It was originally published by The American Institute for Economic Research and is reprinted here with permission.