12 Things You May Not Know About the Federal Reserve

I just finished an excellent book — America’s Bank: The Epic Struggle to Create the Federal Reserve — by Roger Lowenstein, and while I enjoyed it purely in my avocation as an American history nerd, I also feel like I have a much better understanding of the inherent weirdness of the U.S. financial system.

Before reading this book, I think I assumed that the age of the Federal Reserve (it was founded in 1913) and the importance of its role in setting monetary policy meant that its current design and function were, primarily, the result of careful, purposeful construction, by people who knew what they were doing.

After reading this book (and falling down a deep rabbithole of additional Fed research … start with Peter Conti-Brown, if you’re interested in following me down), I no longer assume this.

Indeed, when U.S. Supreme Court Chief Justice John Roberts, writing for the majority in Trump v. Cook and defending Federal Reserve independence, cites the Fed’s “unique historical status and role,” I now question whether the Supreme Court (or anyone else!) really understands how much of the modern design and function of the Federal Reserve is the result of a series of bizarre accidents, near misses, tough compromises, and straight-up dumb decisions that have accumulated over 113 years.

So, as a public service, I offer today’s newsletter: 12 things you may not know about the Federal Reserve.

Before we get to my list, allow me to draw you a quick org chart, which I would like you to keep in your head as you read. The Fed has three primary components:

1. The Board of Governors. A federal agency in Washington, D.C., 7 members, nominated by the president and confirmed by the Senate.
2. The Regional Reserve Banks. These banks are located in Boston, New York, Philadelphia, Cleveland, Richmond, Atlanta, Chicago, St. Louis, Minneapolis, Kansas City, Dallas, and San Francisco.
3. The Federal Open Market Committee. This is the body that actually sets interest rates and it is made up of the 7 Washington governors plus 5 of the twelve regional bank presidents.

Got that?

OK, let’s go!

#1: It took 4 tries and 120 years to create a permanent central bank. 

The First Bank of the United States, chartered in 1791, was, according to Lowenstein, a capable central bank. It managed the young government’s debts, acted as its fiscal agent, and helped discipline a chaotic currency. Despite this, it died in 1811 when Congress declined to renew its charter by a single vote in each chamber.

The Second Bank of the United States, chartered in 1816, was arguably run even better. However, it died too when Andrew Jackson vetoed its recharter in 1832.

When the Whigs tried to charter a third national bank in 1841, President John Tyler vetoed it twice, once in August and again in September after Congress rewrote the bill to meet his objections. The second veto was incendiary enough that his entire cabinet (except Daniel Webster) resigned in protest and his own party formally expelled him.

After that, the country went on without a central bank for another 70 years — a genuine outlier among industrializing nations — until the Panic of 1907, which was a crisis so severe it had to be halted by J.P. Morgan personally corralling the other bankers into a rescue.

That fourth attempt — which came on the heels of the Panic of 1907 and that is chronicled in Lowenstein’s book — became the Federal Reserve we have today.

#2: The Fed wasn’t created to manage monetary policy.

The Fed’s most famous job, and the one that is most often cited as the reason for its independence, was not a job that it was originally created to do.

The stated purposes of the Federal Reserve, according to the Federal Reserve Act of 1913, were to act as a lender of last resort for banks and to create a national clearinghouse that could make the U.S. payments system work.

So, where did monetary policy come from?

Believe it or not, it was discovered by accident.

In the early 1920s, the individual Reserve Banks — which in that era operated with a lot of autonomy — started buying government securities. Not to influence the economy, but to earn income. In doing so, they noticed that when they bought securities in size, they pushed money into the banking system and credit conditions loosened across the whole country. When they sold, things tightened. They had stumbled onto open market operations as a byproduct of trying to balance their own books.

By 1923 they’d set up a committee to coordinate the buying, precisely because they’d realized it was an incredibly powerful economic policy lever, not just an income strategy.

#3: The Fed has become more centralized over time. But we still have the 12 regional reserve banks.

Why are there 12 regional reserve banks?

Well, originally it was part of the compromise that was needed to get the Federal Reserve created. The First and Second Banks of the United States were both killed because of political reasons, and those reasons were still very much in force in 1913. Populists (the political descendents of Andrew Jackson) and farmers living in western states did not want a single institution — read: New York, or, more precisely, Wall Street — to control central banking.

So the Federal Reserve Act created a system of 12 reserve banks, scattered around the country, and delegated most of the authority to them. In 1927, Congress rechartered the reserve banks and, finally learning a lesson from the past, made that rechartering permanent.

However there’s a good reason why almost no other nation has had a decentralized governance structure for their central bank.

It doesn’t work very well!

Through the late 1920s and into the early 1930s, the twelve regional banks pursued their own credit policies, with no one clearly in charge. When the banking system began collapsing into the Great Depression, the Fed’s decentralized structure was catastrophically slow and ineffective at dealing with the crisis.

So the Banking Act of 1935 tried to fix it. Marriner Eccles, FDR’s Fed chairman, centralized much of the Fed’s authority in Washington, with the newly-renamed Board of Governors, which was given a majority of votes on the Federal Open Market Committee.

And yet! The 12 regional reserve banks, instead of being eliminated or relegated to simple branch offices, remained as they were (though with reduced power).

Why?

Because the senator who was the co-author of the original 1913 Act — Carter Glass — fought for them to remain. He regarded the regional structure of the Fed as his life’s work and he was able to help them retain 5 of the twelve FOMC seats, which they still hold today.

#4: The 12 reserve banks are federally chartered private corporations.

If we centralized power in Washington but kept the twelve regional reserve banks, you may be wondering: Who actually runs those banks, and how is power split between them and the Board?

Great questions!

The Board of Governors is fairly straightforward. 7 members, nominated by the president, confirmed by the Senate, serving staggered 14-year terms specifically designed so that no single president can pack it.

The 12 reserve banks are weirder. They are not government agencies. They are federally chartered private corporations, and their shareholders are the commercial banks in each district, which are required to buy stock in their regional reserve bank in return for a statutorily-defined dividend. Those member banks elect 6 of the 9 directors on each reserve bank’s board (though, since 2010, they don’t pick the bank presidents).

Among other things, what this structure means is that the Fed’s bank supervisors are frequently employees of these private corporations. The most vivid illustration: Greg Becker, the CEO of Silicon Valley Bank, sat on the board of the San Francisco Fed — one of his own supervisors — until almost the day his bank collapsed in 2023.

#5: Political independence is newer than you think.

For its first several decades, the Fed was not meaningfully independent of the U.S. Treasury Department. During and after World War II, it agreed to peg interest rates on government bonds, buying whatever it took to keep yields where the Treasury wanted them. Modern political independence dates specifically to the Treasury–Fed Accord of 1951, after a fight between Fed leadership and the Truman administration.

And even the norm of political independence is younger and more fragile than the Accord. After Truman, Presidents kept leaning on the Fed anyway. The most notorious example was Richard Nixon, who pressured his own appointee Arthur Burns into running easy money ahead of the 1972 election, an episode documented on the White House tapes and widely blamed for feeding inflation in the 1970s.

Real, meaningful political independence for the Fed is closer to a 40-year-old habit than a 200-year-old principle.

#6: The Fed has a ton of supervisory power, but doesn’t directly supervise very many banks.

The reason the Fed does bank supervision at all is because it’s the country’s lender of last resort. If your job is to lend to banks in a panic to keep the system from collapsing, you have a direct interest in knowing, ahead of time, which banks are sound enough to lend to. You don’t want to hand emergency cash to an institution that’s already a smoking ruin. So, originally, the Fed’s supervision of banks was a narrow and sensible requirement, a condition of membership for the state banks that joined the Federal Reserve System. Today, the Fed directly supervises roughly 700 state-chartered banks that opted into Fed membership.

However, indirectly, the Fed’s supervisory reach has expanded dramatically over the last 120 years. In 1956, the Bank Holding Company Act made the Fed the umbrella supervisor of bank holding companies — the corporate parents that own banks — which, as American banking reorganized itself under holding-company structures, quietly put nearly every large banking organization in the country under the Fed’s authority. In 2010, Dodd-Frank made the Fed the de facto systemic regulator, in charge of stress tests and the largest, most systemically important firms.

This expansive yet indirect supervisory arrangement has some weird effects. The Fed supervises banks to protect the Fed’s own functions — its liquidity backstop and its payment rails — not, fundamentally, to protect the banks’ customers. It cares about a bank to the degree that bank might someday need rescuing, or might threaten the integrity of the payment system.

However, if a state-chartered bank under the Fed’s direct supervision were to … oh, I don’t know … operate a large and staggeringly incompetent banking-as-a-service business, riddled with weak compliance and risk management controls, for years, would the Fed even notice? Would it take the very public collapse of that bank’s middleware partner — and the loss of tens of millions of its customers’ dollars — before the Fed acted?

This example is, of course, purely hypothetical.

#7: The Fed used to be a club. Then it became a tax.

For most of the Fed’s history, membership was a real thing with real costs and benefits. If a bank joined the Federal Reserve System, it got access to the discount window and the Fed’s payment services, but it was required to park a chunk of its deposits at the Fed as reserves, earning no interest.

In normal times, the cost of this deal was tolerable to most banks. However, when inflation spiked in the 1970s and, in response, interest rates climbed into the double digits, the opportunity cost of holding dead, zero-yield reserves became enormous — every dollar sitting at the Fed was a dollar not earning 15% out in the market. So banks did the rational thing: They quit. Through the 1970s, institutions fled Federal Reserve membership to escape the reserve requirement, dropping national charters or converting to state non-member status where the rules were lighter. This was a huge problem for the Fed because monetary policy was transmitted through member-bank reserves. As membership drained away, a shrinking slice of the banking system sat inside the Fed’s reserve regime, and the Fed watched its grip on the money supply erode in real time, in the middle of the worst inflation in modern American history.

The fix was the Monetary Control Act of 1980. Congress didn’t force banks back into membership. It made membership irrelevant by imposing uniform reserve requirements on every depository institution in the country, member or not, national or state, bank or thrift or credit union. Everyone, member and not, had to pay the tax to ensure the Fed’s ability to transmit monetary policy. In exchange, Congress opened access to the Fed’s services, including the payment rails and the discount window, to all depository institutions.

#8: Congress forgot to unbundle the membership benefits.

You hear the term “Fed master account” a lot these days (my friends and podcast co-hosts Jason Mikula and Kiah Haslett have written about them multiple times recently). An important thing to remember is that it’s not a new concept. Dating all the way back to 1913, when a bank joined the Fed, the concrete expression of that membership was an account at its reserve bank — the place its reserves lived, the way its checks cleared, and its standing to borrow at the discount window. One account, and through it flowed all three of the Fed’s functions: monetary policy (reserves), payments (clearing and settlement), and the lender-of-last-resort backstop (discount-window access). It was an all-or-nothing bundle, and it was available only to members.

When the Monetary Control Act dissolved the members/non-members distinction in 1980, it opened this account up to a far wider set of institutions. But — and this is the crucial part — nobody rethought what the account was. It stayed exactly what it had been in 1913: An indivisible, all-or-nothing bundle of reserves, payments, and backstop, handed out one applicant at a time at the discretion of the regional reserve banks.

The idea of a fully-bundled master account made perfect sense when the only applicants were ordinary Fed member banks. However, after 1980, it was theoretically available to a much stranger and broader universe of institutions.

For decades this didn’t cause much trouble, because the universe of applicants stayed conventional.

That’s no longer true.

#9: The Fed doesn’t control eligibility, but it does control access.

In our system, a state can charter a bank — states have been granting bank charters since long before the Fed existed. However, the Monetary Control Act reached across that line: It compelled every state-chartered depository institution to hold reserves at the Fed, whether it wanted to or not. That’s why Congress, in exchange, opened up access to Fed master accounts to all depository institutions, including state-chartered depository institutions. It was a give-back.

The problem is that Congress didn’t actually open up access. It opened up eligibility, but it left the discretion about who to grant that access to with the Fed’s reserve banks.

This has become a problem!

Because a “state depository institution” is defined by state law, a state can define it however it likes. So a state can charter a novel entity, call it a bank, and effectively hand it a claim on a Fed master account — without that entity ever submitting to the rigorous federal supervisory framework that traditional national and state-chartered, deposit insurance-holding banks endure. Wyoming did exactly this, creating “special-purpose depository institutions” more or less purpose-built for crypto firms.

But just because novel state-chartered banks, like Wyoming SPDIs, are eligible for master accounts doesn’t mean they are entitled to receive them. Custodia, a Wyoming SPDI, found this out the hard way. It applied for a master account, waited two years, and then was denied. The bank sued the Fed in response, but the courts sided with the Fed. In a pointed dissent in the Tenth Circuit decision, Judge Tymkovich warned that unreviewable discretion in this area “effectively handed the Reserve banks a veto over states’ chartering power.”

#10: The Fed’s payments mandate is to protect incumbents, not to encourage competition or innovation.

When Congress decided, in 1980, to open up access to the Fed’s payment services, it had two options.

Option 1 was to make the Fed’s payment services a true public utility; cheap or free, open to all, run to maximize competition in payments, even if that meant the central bank’s rails competed away the profits of the private banks and processors who made their living moving money.

Option 2 was to keep the Fed a disciplined market participant; require it to charge full freight for its services and stay in its lane, specifically so it wouldn’t undercut the private electronic payments industry, which was really starting to boom in 1980.

Congress chose option 2, which is why the Fed is legally required to charge for its payment services as if it were a private company, recovering all its costs plus an imputed markup (the “private-sector adjustment factor”). It’s why the Fed is legally prohibited from requiring banks to adopt its various payment services, which is why representatives of the Fed can be seen hawking FedNow at industry trade shows. And it’s why only depository institutions can hold a master account, thus forcing every non-bank that wants to touch the payments system to rent access through a bank.

This choice by Congress was highly intentional, and it explains why the Fed’s occasional attempts since 1980 to encourage competition and innovation in payments tend to result in confusion, rather than clarity. When Fed Governor Waller convened the Fed’s first-ever “Payments Innovation Conference,” he declared a new era, and pitched a “skinny” master account as a way to open the rails to new players and increase market competition and innovation. Barely two weeks later, he had to clarify his initial comments on his skinny master account idea, saying:

There’s a misunderstanding out there that just a fintech can show up and say, ‘Hey, I’d like a skinny master account.’ No. You’ve got to be an eligible depository institution … So it is technically, you have to have a bank charter. So if you’re not a bank, you don’t have a bank charter, you don’t have the right to ask for one at all.

The Fed wants to position its payment services as a public utility, intended to stimulate competition and innovation from fintech and crypto companies, but, legally, it has to price and sell its payment services only to banks and only to protect, rather than disrupt, payments incumbents.

#11: The U.S. has two lenders of last resort.

Here’s an irony that captures America’s schizophrenic relationship with central banking.

We spent 120 years refusing to build any permanent, centralized, government-backed liquidity backstop. Then, once we finally built one (the Fed), we almost immediately built a second one!

The second backstop is the Federal Home Loan Bank (FHLB) system, created in 1932 — and, tellingly, modeled on the Fed’s own design: a network of regional, cooperatively-owned banks. The FHLBs exist to fund housing lending, but in practice they’ve become what observers accurately call the “lender of next-to-last resort.”

Because borrowing from the FHLBs is easier and carries no stigma (tapping the Fed’s discount window is a signal of distress that can trigger the very run that troubled banks are trying to survive), banks in need of liquidity turn to them first. This — combined with the FHLBs’ “super-lien” status that puts them higher in the repayment hierarchy than other lenders, including the FDIC and the Fed itself — means that the FHLBs routinely extract a great deal of money from failing banks on the way down, while allowing the final failures to hit others.

This is exactly what happened with SVB. In its final days, SVB was leaning heavily on advances from the Federal Home Loan Bank of San Francisco, and only lunged for the Fed’s discount window at the very end — too late, in part because the FHLB doesn’t operate over weekends and SVB couldn’t reposition its collateral to the Fed in time to save itself. In the aftermath of SVB’s failure, the FDIC insurance fund took a roughly $16 billion hit, the San Francisco FHLB walked away unscathed, and the Fed’s discount window wasn’t tapped.

Bizarre.

#12: We’re now re-embracing a pre-1913 monetary system.

In June of this year, Congress passed the 21st Century ROAD to Housing Act. The bill, which became a law despite President Trump declining to sign it, includes a provision blocking the Federal Reserve from issuing a digital dollar through 2030. This makes official what has, for years, been the unofficial position of the U.S. government: We are not going to issue a central bank digital currency (CBDC), and will, instead, allow that product category to be filled by privately-issued, U.S. dollar-backed stablecoins.

Recall what the American monetary system looked like before 1913: A patchwork of private institutions issuing their own paper money, backed (ideally) by reserves, circulating as a means of payment. This fragmented, chaotic, privately-issued-money world was what the Federal Reserve was created to replace.

We spent more than a century consolidating money issuance into a coherent public framework. And now, on the digital frontier, we are voluntarily fragmenting it back out to private issuers. The logical move — if what you care about is coherent monetary control — would be a CBDC: Public digital money, issued by the same institution that runs monetary policy, keeping the tightest possible grip on the thing the Fed exists to manage. Whatever its flaws, a CBDC is the option that makes institutional sense. And we are pointedly not doing it.

This is remarkable to me. And very telling. The same American distaste for concentrated public monetary power that killed the First and Second Banks of the United States, that shattered the Fed into twelve pieces, that made the Fed a payments vendor forbidden to undercut private processors — that same instinct now expresses itself in a preference for letting private companies issue the digital dollar rather than our own central bank.

This is the history of the Federal Reserve in a nutshell: A deep American reluctance to let any single public institution hold too much power over money — a reluctance that has never once stopped us from creating the power, only from organizing it coherently.