Dollars, deficits and debt

David Zetland, The one-handed economist

The US dollar (USD) is the world’s currency of exchange, debt investment, and reserve currency. These different roles are often merged in our heads because they’ve been synonymous for decades, but the wheels are falling off and those roles — I think — will unbundle. I’m going to explain, but feel free to correct me and/or point out missing bits.

The USD will function as a currency of exchange for a long time to come. It’s handy for two people with different non-USD currencies to be able to say “send me x dollars” because of the USD’s “reference” role. It doesn’t matter if the exchange rate is up or down (the € bought less than $1.12 for the past 3 years, now it’s around $1.18), it’s still well known.

Dollar-denominated debt (bonds) will still be attractive investments due to their enormous supply ($37 trillion, or $108,000 per person), liquid markets, and variations in maturity (anything from 4-week “T-bills” to 2-year “notes” to 30-year “bonds”). The interest rates on debt vary by maturity, but the price of debt depends on “interest rates” (I’ll define in a second), which are becoming very relevant, especially since 32% of debt is maturing in less than a year (needs to be re-issued as new debt at current interest rates). The other 68% matures in 2-30 years.

I always need to stop and think about how the price of debt in the market responds to a change in interest rates. In short, higher interest rates mean a lower price. I’ll give an example in a second.

Interest rates (IRs) depend on the supply of lending and demand for borrowing. If more people want to borrow than lend, then rates go up (for all maturities). The Federal Reserve has a big influence on short-term IRs because it participates (“open market operations”) on the supply or demand side of the market as it tries to maintain its “Federal Funds target rate” (FFTR). The FFTR is now 4.25-4.5%. Many banks and investors use the FF to set their own IRs (for mortgages, for example), and the FFTR affects longer-term IRs because they are connected.

For example: In 2015, new 30-year bonds paid 2.3-3% interest when they were sold on the market at “par”. Thus, a $1000 30-year bond paying 3% sells for $1000, pays $30 per year (the “coupon”) for 30 years, and is then redeemed for $1000. At issue, the “face rate” — which never changes — is the same as the “yield,” which depends on market IRs. If IRs rise, then the prices of previously issued bonds fall, to equate the yields on older bonds to the yields on new debt. Today, the 30-year rate is 4.86%. That means that the price of 2015-vintage bonds must be lower so that someone who buys them gets a return of 4.86%. Nobody will pay “par” on a 3% (old) bond when they can pay par to get a 4.86% (new) bond! Perplexity says that those 2015 bonds are now trading at 70-75 cents on the dollar, which seems aggressive, but not if you’re looking at “sub-market coupons” for another 20 years!

All this matters because the supply of investments competes for the demand of investors. Investors will always be happy to buy US debt, but they are not going to give away their money. They will demand “market yields,” which means that the cost to the US government (and citizens) rises as IRs rise.

I was going to ignore “risk” as a factor affecting market prices for US government debt, which is often called “risk-free” due to (a) the US issuing its own currency (so it can always print money to pay its debts, unlike non-USD issuing countries that need to buy USD to pay their USD debts) and (b) America’s long history of “honoring its debts.” That risk is now becoming relevant, since Prez Cheeto — a serial bankrupt himself — has cast doubt on his willingness to honor those obligations. Madison is turning in his grave. 

So, to recap, the USD is a handy reference for exchange rates (like the metric system), and US debt is a pretty good investment.

But is the USD a good reserve currency? Should you keep your savings in USD, if you want it to hold its value of time? The answer — as you saw in the example above — is no when the US government is spending more than it makes (the 2024 deficit was 6.3% of GDP and 22% of total spending; since 1990, deficits have ranged from a surplus of -2.3% in 2000 to 14.7% in 2020) AND when there’s no viable promise that it will reduce its debt — meaning that yields will be rising far into the future. “Cheeto’s BBB” (budget busting billionaire bailout) adds $3 trillion to debt by 2035, when total debt (due to other sources of overspending) will rise to $54 trillion, taking (external) debt-to-GDP from 100% to 124%. NB: Most sources say the current debt-to-GDP ratio is around 100%, but they often exclude “internal debt”, e.g., money the government owes social security. If you include that debt, then the ratio is more like 125% now, and it will rise further!)

Economists get nervous — in terms of fiscal stability — when debt-to-GDP is above 100% and deficits are larger than the growth in GDP. Given that US is over the 100% barrier and 6% deficits are higher than 2% growth rates/projections, they are nervous. 

Something is going to break, and that’s why the USD is no-longer a good store of value.