Latest / The Fiscal Policy Podcast with Fexingo: Government Budgets, Stimulus, and Public Spending / Why Governments Keep Borrowing at the Short End
Transcript
- Lucas: So the US Treasury borrows roughly 7 trillion dollars a year — every year — to roll over maturing debt and fund the deficit. And the single most consequential decision they make is not how much to borrow, but how long to borrow it for. Luna: Right — the maturity structure. Short-term bills versus long-term bonds. It sounds like a technical detail, but it determines how much interest taxpayers pay and how exposed the budget is to a sudden spike in rates. Lucas: Exactly. And the trend over the past two decades has been clear: governments have been borrowing shorter. The average maturity of US federal debt was around 70 months in 2001. Today it's under 50 months. That's a massive shift. Luna: Why? I mean, conventional wisdom says longer-term debt locks in low rates. If you think rates are going up, you'd want to issue 30-year bonds, not 2-year notes. Lucas: You would think. But the Treasury has a different incentive. They're judged on the cost of debt in the current fiscal year, not over the life of the loan. Short-term bills are cheaper right now — the 3-month Treasury bill yields maybe 4.5 percent, while the 10-year bond is closer to 4.8 or 5.0. So issuing short-term debt saves money in the budget today. Luna: But that's kicking the can. If rates stay high, you have to refinance at higher yields every few months. It's like taking out an adjustable-rate mortgage on a house you plan to keep for 30 years. Lucas: That's exactly the analogy. And it's not just the US. The UK's Debt Management Office faced a brutal version of this in September 2022. The so-called 'mini-budget' spooked markets, gilt yields shot up, and suddenly the cost of rolling over the UK's substantial stock of short-term debt became a crisis. The Bank of England had to step in and buy bonds to stabilize the market. That was a direct consequence of having too much debt maturing too soon. Luna: And the UK's average maturity at the time was around 12 years, which is actually longer than the US. But the problem was concentration — a huge chunk was due in a short window. Lucas: Right. So it's not just the average maturity; it's the distribution. The US Treasury has been issuing more in the 2- to 7-year range, and less in the 10- to 30-year range. The result is a wall of maturities coming due in the next few years. According to the Treasury's own data, about one-third of all marketable debt matures within one year. That's over 7 trillion dollars that needs to be refinanced every 12 months. Luna: That's enormous. And it creates what the economists call 'rollover risk'. If the market suddenly demands higher yields, the entire budget feels it almost immediately. Lucas: And the counterargument from the Treasury is that short-term debt is cheaper over time, historically. They have a point — since 1990, the average cost of 3-month bills has been lower than the average yield on 10-year bonds. So by issuing short, the government saves the taxpayer money. Luna: But that's a statistical argument that assumes the future looks like the past. If we enter a period of structurally higher rates — say, because of persistent inflation or larger deficits — the cost of rolling over short-term debt could spike. And then the savings disappear fast. Lucas: That's the tension. And it's not just a US story. Japan has the highest government debt to gdp ratio in the developed world, around 260 percent. But their average maturity is over 9 years, partly because they've been issuing 10-, 20-, and even 40-year bonds. They've deliberately locked in low rates for decades. The trade-off is that their interest cost is very stable, even though the debt stock is enormous. Luna: So they chose stability over cheapness. And it's worked — Japan's interest payments are only about 1 percent of GDP, compared to the US at roughly 2 percent. Lucas: And the interesting twist is that Japan's short-term rates are negative or near zero, so they could borrow even cheaper if they issued bills. But they don't. They prioritize reducing rollover risk. That's a deliberate policy choice, not a market constraint. Luna: It's a fascinating contrast. The US Treasury seems to optimize for lowest cost in the current year, while Japan optimizes for stability over decades. Lucas: And that brings us to the institutional incentives. The Treasury's borrowing decisions are made by the Debt Management Office, which is part of the Treasury. Their mandate is to 'borrow at the lowest cost over time'. But in practice, 'over time' gets interpreted as 'this year's budget'. There's a classic time-horizon problem. Luna: And no politician wants to explain a higher interest bill today for a potential crisis 10 years from now. It's a classic agency problem. Lucas: Look, if walking through the economy with us has made something click — if you find yourself noticing the incentives behind a government bond auction — that's because a handful of listeners chip in monthly through buy me a coffee dot com slash fexingo. It's what funds making these episodes, and it keeps us independent. No ads, just people who care about public finance. Luna: Yeah, it's a small group, but it makes a difference. We're grateful for that. Lucas: So back to the maturity question. There have been proposals to reform how the Treasury manages debt. One idea is to set a minimum average maturity — say, 5 years — to prevent the drift toward shorter borrowing. Another is to issue more ultra-long bonds, like 50-year or even 100-year bonds, which some countries like Austria and Belgium have done. Luna: But those ultra-long bonds have limited demand. Most institutional investors want maturities that match their liabilities — insurance companies and pension funds are natural buyers of 30-year bonds, but 100-year bonds are more speculative. Lucas: True. The US did issue a 30-year bond again in 2006 after a four-year hiatus, and it's been a mainstay since. But issuing longer would require building a new investor base. The Treasury has actually studied it — in 2015 they released a report on potential benefits of a 50-year bond. They concluded it could lower borrowing costs by maybe 5 to 10 basis points, but the risk of illiquidity was a concern. Luna: So it's a trade-off. More liquidity in the 2- to 10-year part of the curve, less in the long end. But the bigger risk, I think, is the macroeconomic one: if the government is constantly refinancing short-term debt, it's more vulnerable to a sudden loss of confidence. Lucas: And that's not just theoretical. Look at emerging markets. Countries that borrow heavily in short-term foreign-currency debt have historically been the ones that experience sudden stops and crises. The classic example is Mexico in 1994 — they had a lot of short-term dollar-linked debt, and when the peso devalued, the government couldn't roll it over. That was the 'Tequila Crisis'. Luna: So the principles apply from Mexico to the US to Japan. The maturity structure of government debt is a fundamental risk management decision. Lucas: And it's one that doesn't get nearly enough public scrutiny. The Treasury Borrowing Advisory Committee — a group of market participants that advises the Treasury — meets quarterly, but their discussions are mostly behind closed doors. The average voter has no idea that their government's debt manager is making a bet on future interest rates every single day. Luna: So should there be more transparency? Maybe a publicly announced target for average maturity, like the Bank of England's 2009 framework? Lucas: The UK actually had a target for net debt maturity until 2015, but it was abandoned. The US has no explicit target. The Treasury's own Office of Debt Management has recommended a more structured approach, but it's never been implemented. It feels like one of those issues that only gets attention during a crisis. Luna: And by then it's too late. You can't extend maturity in a panic — the market will demand a premium for exactly the risk you're trying to avoid. Lucas: Exactly. So the question is: given that the US has a large deficit and rising debt to gdp, should the Treasury actively lengthen the maturity of its debt? Some economists argue yes — that it's a form of insurance. The cost is a few basis points higher today, but it reduces the chance of a funding crisis tomorrow. Luna: And what's the counterargument? That it's expensive and unnecessary, because the US borrows in its own currency and can always print more? Lucas: That's the modern monetary theory view — if you borrow in your own currency, you can't be forced into default. But even if that's true for solvency, it's not true for liquidity. In 2020, the Treasury market experienced severe dysfunction — the 'dash for cash' — and the Fed had to intervene massively. That was partly a maturity issue: as investors sold short-term Treasury bills for cash, the market seized up. Luna: So the maturity structure matters even for the world's safest borrower. And it's a choice, not a given. Lucas: Right. And that's the point I want to leave listeners with: every time the Treasury auctions a 3-month bill instead of a 10-year note, they're making a judgment about the future. It's a judgment that affects your taxes, your borrowing costs, and the stability of the whole financial system. And it's happening in a room most people have never heard of. Luna: It's the quietest big decision in government finance. And now you know to watch for it.