Is the US exorbitant privilege eroding?

U.S. Treasury debt has been a cornerstone of global financial markets for many decades because it is the closest thing the world has to a safe asset that can be easily and cheaply traded. It is this unique status of Treasury debt that has given the U.S. an important component of its exorbitant privilege: the ability to issue debt more cheaply than its peers in other advanced economies. However, with no end in sight to budget deficits and U.S. public debt projected to rise out of control, markets have begun to charge a risk premium relative to the rest of the G10. This post shows what has been happening to the U.S. convenience yield—the yield markets are willing to forgo in order to hold Treasuries—and how this advantage has eroded over time. This erosion is not new—it has been playing out steadily over the past decade—but it does mean that the exorbitant privilege associated with Treasury debt is eroding. That could bring higher borrowing costs for the government in the years ahead and—potentially—a weaker dollar.

What are convenience yields?

The convenience yield of U.S. Treasuries is the yield that markets are willing to forgo to hold a safe and liquid asset. There are two principal ways to measure the “specialness” of U.S. Treasury debt, which underpins its convenience yield. The first is to compare Treasury yields to yields on other U.S. assets of comparable maturity, which is typically debt of AAA-rated companies, i.e., the most creditworthy firms, or interest rate swaps, which price the path of Fed policy out into the future (in addition to term premia). The second is to compare Treasury yields to yields on government bonds in other advanced economies and—adjusting for the fact that foreign yields are denominated in currencies other than the dollar—looking at the spread between the U.S. and a trade-weighted foreign average.

Both approaches to calculating the convenience yield are conceptually similar. They compare the yield on Treasuries to some other asset—domestic or foreign—that does not have the specialness of U.S. government debt. It is the evolution over time of the spread between Treasury yields and yields on other assets that provides information on whether the specialness of U.S. government debt is declining in a relative sense.

Both approaches show that the specialness of U.S. government debt has fallen over time. This is not a recent phenomenon. It has played out steadily over the past decade and has been linked in academic research to increased U.S. debt issuance, which is eroding the specialness of Treasury securities. Of course, because both measures are relative, a declining convenience yield could mean either a loss of specialness of U.S. government debt or an increase in the specialness of U.S. corporate or foreign government bonds against which it is being compared. However, the recent resurgence of debt worries in Europe makes the latter explanation less likely, at least for foreign government bonds.

This post lays out the second approach, which calculates how much U.S. Treasury yields are above or below foreign government bond yields of similar maturity after these are hedged back into U.S. dollars using foreign exchange (FX) forwards. According to this approach, the U.S. lost its ability to issue debt more cheaply than its peers nearly a decade ago. Indeed, the cost of U.S. debt was steadily rising relative to its peers until 2022, after which it has stabilized. The most obvious reason for this is that the fiscal picture abroad has weakened faster than for the U.S. because central banks around the world have been raising interest rates to combat high inflation.

The U.S. convenience yield versus the G10

Figure 1 shows the difference between the 10-year U.S. Treasury yield and a trade-weighted average 10-year yield on government bonds in Germany (DE), Japan (JP), the UK (GB), Canada (CA), Switzerland (CH), Sweden (SE), and Australia (AU). To make the yields comparable, 10-year currency forwards are used to hedge yields on these government bonds into dollars (what this calculation essentially does is to sell the foreign currency exposure implicit in foreign bond holdings forward so that there is no FX risk). This differential is thus purely in dollar terms and therefore a good proxy for how “special” markets think Treasury debt is compared to the rest of the G10. When this difference is positive, it means the 10-year Treasury yield is above its trade-weighted average, i.e., the U.S. pays more to borrow than its peers, even adjusting for currency risk. Rather than enjoying a convenience yield, markets are demanding a premium to hold U.S. debt. If the difference is negative, the U.S. is able to issue 10-year debt more cheaply than its peers after adjusting for currency denomination differences.

Figure 1 shows that the U.S. used to issue debt on average over 10 basis points more cheaply than the trade-weighted foreign average, but this convenience yield eroded steadily and turned positive in 2018, meaning the U.S. began to pay a premium over its G10 peers. This erosion continued until 2022, after which the premium the U.S. pays stabilized around 15 basis points. This is likely because the rise in global interest rates since COVID-19 has shone a light on the precarious state of public finances globally, with high-debt countries in Europe looking especially vulnerable. This is helping the U.S. look better on a relative basis, but—of course—this does not mean borrowing costs in an absolute sense are falling. They are rising, but they are rising elsewhere too.

Figure 2 shows the two building blocks in the convenience yield calculation. The black line is the straightforward 10-year yield differential, i.e., it is the 10-year Treasury yield minus its trade-weighted foreign counterpart. The red line is what 10-year currency forwards price in by covered interest parity for this rate differential. Since U.S. yields tend to be higher than the trade-weighted foreign average, this means currency forwards price in dollar depreciation to offset higher yields. The convenience yield—the blue line, which is the same as in Figure 1—is the difference between these two, i.e., it is—in effect—a violation of the covered interest parity condition.

Figure 3 compares the annual average for the convenience yield in Figures 1 and 2 to the difference between U.S. public debt and a trade-weighted average for the same set of foreign countries. Two points stand out. First, the trend erosion in the convenience yield matches the steady rise in U.S. debt relative to its G10 peers. Second, the recent stabilization of the U.S. convenience yield—or premium rather—does not line up with an ongoing rise in U.S. debt relative to its peers. This is likely because countries like France are experiencing fiscal stress at lower debt than the U.S., which is helping the U.S. look more attractive on a relative basis.

Conclusion

The U.S. lost its ability to issue debt more cheaply than its peers nearly a decade ago. Indeed, the cost of U.S. debt was steadily rising relative to its peers until 2022, after which it has stabilized. The most obvious reason for this is that the fiscal picture abroad has weakened faster than in the U.S., which now—in a relative sense—helps the U.S. But, of course, it has not restored the borrowing-cost advantage the U.S. once enjoyed.

The big question, which we will address in a follow-on post, is what all this means for the status of the dollar as a reserve currency. The ability of the U.S. to issue debt more cheaply than its peers and the status of the dollar are obviously linked. The best way to think about the two is to recognize that the erosion of the U.S. convenience yield is still a relatively contained and small phenomenon. That may explain why U.S. reserve currency status has been relatively robust through lots of policy volatility in recent years. At the end of the day, the changes in the U.S. convenience yield are still small, but they reflect a longstanding erosion of the special borrowing-cost advantage the U.S. has historically enjoyed in Treasury markets.