Bill Dudley and Jonathan Payne joined Markus’ Academy for a two-part conversation on US Treasury yields. Dudley is a Senior Advisor to Princeton’s Griswold Center and a former President of the Federal Reserve Bank of New York. Payne is an Assistant Professor at Princeton. The episodes are available on Spotify (Part 1, Part 2), and the slides are available here.
A summary in 5 bullets:
- There is no Phillips-style curve between debt-to-GDP ratios (safe asset scarcity) and the government funding advantage. The simple relationship disappears when adjusting our prior measures of the funding advantage for the tax treatment of the government’s long-term debt during the Great Inflation
- The expectations hypothesis holds in general throughout US history except in the period of 1965-1990, so that the risk premium on government debt was time-varying only then
- The US government’s funding advantage was largest in the 19th century during the national banking era, not after WWII or Bretton Woods
- Governments face financing trilemma, having to pick two among: (1) a large funding advantage, (2) a solvent banking sector, (3) a regime that inflates the debt away. The funding advantage is a reward for prudence
- The fact that Microsoft’s yields have widened with respect to other AAA corporates or sovereigns suggests growing default risks around AI, not that the government may be crowding out AI investment
Part 1: The Government Funding Advantage Is No Free Lunch
- https://youtu.be/SAUrOFLlK-k
- The first part covered the history of American debt starting in minute 21:09. Before then it started with a refresher on basic concepts to study yield curves.
Part 2: Canaries in the Treasury Coal Mine
- https://youtu.be/ZseTYM_A-60
- We present the notes of part 2 first, as it includes Bill and Jonathan’s positive notes.
Part 2: Canaries in the Treasury Coal Mine
[0:00] R* has drifted up
- After the financial crisis R* was near zero due to the damaged household and corporate balance sheets. September’s Fed Summary of Economic Projections forecasted a 3.2% long-run funds rate, implying a 1.2% R*
- It is driven by increasing demand for capital against the supply of savings: hyperscaler borrowing pulls on the pool, deficits and a falling household saving rate shrink it. Fiscal consolidation would lower R*, as in the mid-1990s
- Many argue policy needs to be more accommodative because R* is lower, but we have had three years at this funds rate while remaining at full employment
[8:15] The expectations hypothesis and the stock-bond correlation
- Classic contributions (Fama & Bliss, 1987; Backus et al., 1989, Cochrane & Piazzesi, 2005) argued the expectations hypothesis does not hold, so that the risk (term) premium is time-varying
- However, they all focused on the 30-year window from 1965. The hypothesis cannot be rejected at 10- and 15-year maturities until the 1960s. Hicks (1939) saw it hold
- If the nominal anchor is secure, borrowing costs will fall; lose it and a higher premium raises debt service costs, further raising the premium
- The stock-bond correlation was near zero in the 19th century, close to one in the 1970s when bonds did not hedge, sharply negative in the crisis, and creeping back up now. The two are linked – treasuries earn part of the advantage by acting as a hedge, so a positive correlation erodes it
[23:12] Is the government crowding out AI investment?
- Between January and July 2026, a spread opened within the AAA class: Treasuries to AAA corporates was flat at 43bp, while a spread of Microsoft to the rest of AAA corporates appeared
- That points to default risk in the AI-exposed borrowers rather than pure crowding out, under which every borrowing cost would rise together. The reassuring part is that the Treasury-to-corporate spread itself is not disappearing
[27:25] Hope is not a strategy
- The outlook: 6%-of-GDP deficits, the 2008-2022 debt rolling over at higher rates, and baby boomers retiring with Social Security exhausted by 2032. Many hope the AI boom will bail the US out, but “hope is not a strategy”
- History provides the positive note: let’s not forget how dire the fiscal situation was in the past. In 1812, the Civil War and both world wars it looked as though the US could not finance itself, and each time it built new institutions and achieved consolidation
- A low tax burden and weak income-tax compliance leave revenue to be raised
Part 1: The Government Funding Advantage Is No Free Lunch
[00:00] Today’s yields are not high by historical standards
- After the revolution the US borrowed at 8% against the UK’s 4%, Hamilton’s worry. They spiked in 1812, the Civil War and WWI, and the 1970s-80s inflation. WWII is the exception (Lehner et al., 2025)
[11:52] Fiscal-monetary interactions
- The Treasury owns the composition of the debt, the auction cycle and the buybacks; the Fed runs the auctions as its agent and manages the primary dealers
- QE and QT are not mirror images: QE arrives at size and speed, QT never completes, because banks’ demand for reserves ratchets up with regulation.
- Exchange rate policy is not the Fed’s: the Treasury decides what the dollar should be worth and the Fed intervenes on its behalf
[22:38] The history of America’s debt
- Hall & Sargent (2011) decomposed the post-war debt decline, attributing 20% of the decline to negative real returns, 40% to primary surpluses, and 40% to growth
- The lack of a yield spike during WWII reflects a sophistication of policy. The US learned to implement financial repression (including through yield curve control) to prevent a spike similar to those of other wars
- Debt management has been deliberately regular and predictable. Bessent is departing from this
[34:30] The national banking era was a stablecoin regime
- Through the 19th century Congress had to approve each individual bond issue, so the debt was long-dated
- To convince banks to hold it, the US allowed national banks to issue money if they backed it with long-term Treasuries. To a first approximation this is a stablecoin regime, and it explains why the government’s funding advantage was largest in the 19th century (and not after Bretton Woods)
- The establishment of the Fed ended this. National banks stopped being the money creators, short-term instruments became liquid, and the long-short spread re-emerged
- US and UK borrowing costs converged in the 1890s, before the dollar was international. Dollar privilege and an advantage over one’s own corporates are different things – many countries earn the second by organizing the domestic financial system to buy the government’s debt
[41:15] The funding advantage is a reward for prudence
- The standard scatter plot says the funding advantage was largest exactly when the government was inflating the debt away. However, this reflects mismeasurement rather than an underlying economic relationship because the sample includes flower bonds. These bonds were redeemable at par against estate taxes and so worth most when prices fell furthest below it; by the 1960s every Treasury beyond seven years was one (Lehner et al., 2025)
- Corrected, it collapsed to about zero through the 1970s. Investors did not pay up for a scarcer Treasury asset, they shifted to other assets (gold rose 2,000%)
- The mechanism is commitment. If you force the banks to hold the debt, you must also promise not to devalue it. If you do devalue the banks might collapse. After the Civil War the US repaid rather than default or inflate, returning to gold. Hence the trilemma
- Dodd-Frank presents the same financial dominance bargain: push the debt onto the banks, then protect them by not inflating. However, that protection can be costly: QE flooded the banks with deposits, Silicon Valley Bank put them into long Treasuries, and the tightening that followed left it with large losses