Latest / Monetary Policy Explained with Fexingo: Central Banks, Money Supply, and Interest Rates / How Central Banks Use the Discount Window
Transcript
- Lucas: If you ask most people what the Federal Reserve does, they'll say 'sets interest rates' — and they're right. But the Fed also runs a lending facility that's been around since 1914, and it's one of the most misunderstood tools in the central bank's toolkit. Luna: You're talking about the discount window. Most people have heard the term, but they probably don't know how it actually works. Lucas: Exactly. The discount window is basically the Fed's overnight lending desk for commercial banks. If a bank needs reserves at the end of the day — maybe they miscalculated their cash flows, or a large withdrawal caught them off guard — they can borrow directly from the Fed, usually just overnight. Luna: And the interest rate they pay is the discount rate. Which is set by the Fed, not the market. Lucas: Right. There are actually two tiers. Primary credit is for sound banks — no questions asked, you just borrow at the discount rate, which is typically set about 50 basis points above the top of the federal funds rate target range. Secondary credit is for banks that don't qualify for primary credit — maybe they're under financial stress — and that rate is another 50 basis points higher. Luna: So the discount window is supposed to be a safety valve. But for decades, banks were terrified to use it. Lucas: That's the stigma problem — and it's central to why the discount window is such a fascinating policy tool. Banks worried that if the market found out they were borrowing at the discount window, it would signal weakness. Other banks might stop lending to them, counterparties might pull back. So even healthy banks avoided it. Luna: The classic example is the 2008 financial crisis. In the weeks after Lehman collapsed, the discount window was barely used, even though banks were desperate for liquidity. Lucas: That's exactly right. The Fed actually cut the discount rate and extended the borrowing term to 30 days, but stigma was so powerful that banks still chose to borrow in the fed funds market at higher rates rather than touch the window. It took the creation of the Term Auction Facility — where banks could bid anonymously for funds — to get liquidity flowing. Luna: And then we had the Silicon Valley Bank collapse in March 2023. That was really the moment when the discount window became front-page news. Lucas: It's a great case study. In the days before SVB failed, they borrowed $140 billion from the discount window — a record. But it wasn't enough to stop the run. After that, the Fed created the Bank Term Funding Program as a new backstop, but they also redesigned the discount window itself. Luna: What did they change? Lucas: Two big things. First, they made the borrowing rate more attractive — they cut the spread from 50 basis points above the fed funds target to just a few basis points. Second, they changed the collateral rules, making it easier for banks to pledge a wider range of assets. The goal was to normalize discount window borrowing, to make it something banks used routinely, not just in emergencies. Luna: Has it worked? I mean, here we are in July 2026 — are banks actually using it more? Lucas: Yes and no. Average daily borrowing at the discount window has been hovering around $500 million per night in 2026. That's way up from the pre-2023 norm of almost zero — but it's a fraction of the $150 billion peak during the SVB crisis. So the stigma has eased, but it hasn't disappeared. Luna: Some banks still worry that if they're seen borrowing regularly, depositors might panic. Lucas: Exactly. And that's the tricky part — the Fed wants the discount window to be a routine liquidity management tool, like a credit line, but the reputational risk is real. The Fed has tried to address this by releasing aggregate data with a two-year lag, so individual banks' borrowing isn't identifiable. But market participants are good at sniffing things out. Luna: Before we go deeper on the design, I want to step back for a second. This kind of deep-dive into central bank plumbing is exactly what makes this show worth doing, and honestly, it takes a lot of work to put together every week. We keep the show ad-free because we want it to be purely about the economics, not about selling you something. Lucas: That's right. And a couple of dollars a month is genuinely what keeps these going — buy me a coffee dot com slash fexingo, if you've gotten something out of them. It makes a real difference. Luna: Okay, back to the discount window. Lucas, you mentioned the rate spread narrowing — how does that affect how banks think about borrowing? Lucas: It changes the cost-benefit calculation dramatically. Before 2023, if the fed funds rate was at 5 percent, the discount rate would be 5.5 percent. That half-point penalty was enough to make banks think twice — they'd rather scramble in the fed funds market for a cheaper rate. Now, with the spread at just a few basis points, the penalty is almost negligible. Luna: So if it's basically the same price, why not use it every day? Why is borrowing still only half a billion a night? Lucas: Part of it is habit — banks have internal policies that say 'don't use the discount window unless it's an emergency.' There's also the operational friction: you have to pre-position collateral with the Fed, which takes work. And yes, there's still a lingering stigma. But there's another angle: the discount window is meant to be a backstop, not the primary source of funding. Banks are supposed to get their daily reserves from the fed funds market or repo market. Luna: So the Fed doesn't actually want to replace the market — they just want the window to be there so that when the market seizes up, there's a safety valve that banks aren't afraid to use. Lucas: Exactly. It's a lender of last resort function. And the whole redesign is about making that safety valve credible. Because if banks won't borrow even in a crisis, the window is useless — and we saw what happens when the window fails: bank runs, contagion, systemic risk. Luna: Let's talk about the international side. Other central banks have similar facilities — the European Central Bank has its marginal lending facility, the Bank of Japan has the discount window. Are they facing the same stigma issues? Lucas: To varying degrees. The ECB's marginal lending facility has historically carried a stigma too, but it's less pronounced than in the US. In Europe, banks are more accustomed to borrowing from the central bank as a routine part of operations — partly because the ECB's refinancing operations are a regular feature of their monetary policy implementation. The Bank of England's operational standing facility is another example — they've pushed to destigmatize it, with some success. Luna: So maybe the real question is: will the Fed ever fully destigmatize the discount window? Or is it a tool that will always carry a certain reputational cost? Lucas: I think it'll always have some stigma, because the very fact that a bank needs to borrow from the central bank suggests that other lenders are unwilling to provide funds. That's an information signal you can't fully eliminate. But the goal is to reduce the stigma to a point where banks will use the window when they need to, without fearing a death spiral. Luna: And the SVB episode showed that when a bank waits too long to use the discount window, it's often too late. Lucas: Which is why the Fed is now encouraging banks to test their discount window access — to pre-position collateral and even do trial borrowings. It's like a fire drill. If you've never used the window, you're less likely to use it in a real crisis. So the more banks practice, the more natural it becomes. Luna: That seems like a smart approach. Do you think we'll see a day when discount window usage is routine, say, 5 or 10 billion dollars a night? Lucas: It's possible, but I don't think it's the Fed's goal. They want the market to function first. The discount window is insurance — you don't want to use it every day, but you want to know it's there and that you can use it without penalty. If borrowing becomes too routine, it might signal that something's broken in the interbank market. Luna: So the discount window is a bit like a spare tire — you're glad it's there, but you don't want to drive on it all the time. Lucas: That's a perfect analogy. And the Fed's job is to make sure the spare tire is in good condition, easy to access, and doesn't come with a flat fee that makes you avoid changing it. Luna: One last thing — how does the discount window interact with other tools like interest on reserves? Lucas: That's a great question. Interest on reserves, or IOR, is the rate the Fed pays banks on the reserves they hold at the central bank. That rate is usually set slightly below the discount rate, so there's no arbitrage — banks can't borrow cheap at the discount window and then earn a higher rate by parking that money at the Fed. The Fed keeps the discount rate as a ceiling in its rate corridor system, with IOR as the floor. So the discount window is the top of the corridor. Luna: Right, that corridor framework. The fed funds rate is supposed to trade between IOR and the discount rate. Lucas: Exactly. And when the discount window is functioning well, it ensures that the fed funds rate doesn't spike above the discount rate, because banks have a viable alternative at that rate. So it's not just a crisis tool — it's also a mechanism that helps the Fed control short-term interest rates in normal times. Luna: That's the part I think a lot of people miss. The discount window isn't just for emergencies — it's a structural part of how the Fed implements monetary policy. Lucas: It's a classic example of a tool that's most valuable when it's least used, but its existence shapes how the whole system works. So next time you hear 'discount window,' you'll know it's not just some obscure relic — it's a critical safety valve that's been redesigned for a new era of banking. Luna: And hopefully, if the next crisis comes, banks will actually use it.