Why should the Trump Administration issue 50-year bonds?
As the Trump economic team comes together, their economic vision is also coming together. In the last post, I laid out some overarching themes I am seeing from them on the hopes that reducing taxes and regulation will increase productive capital formation and long-term economic growth. You can put all of these ideas under the moniker of supply-side economics. I am also seeing a few individual ideas I wouldn’t put under that umbrella including the extension of the maturities of government bond issuance. Here are some thoughts on that issue.
Back in the 1990s, when Robert Rubin was Treasury Secretary, he initiated a program to shorten the average maturity of US government bond issuance as a measure to reduce interest costs for the government. He also introduced TIPS, inflation-protected securities, as an additional method of doing so. I want to focus on some of his comments from the 1996 TIPS announcement to get at the underlying logic he was employing. Rubin wrote:
The Treasury Department has through its history focused on the most cost-effective ways to finance the federal debt. If you recall, in 1993 Treasury changed the maturity mix of government securities, something that was initially looked on with some skepticism, but which since has won considerable praise and is saving the taxpayers $7 billion. And today, we are announcing the intention to issue inflation protected bonds as a further step in this direction, as well as a step we believe can help promote savings in the country.
Here’s what’s interesting about that statement; If you listen to the Treasury Secretary nominee Steven Mnuchin or likely Council of Economic Advisers Chair Larry Kudlow, they’re saying something that’s almost exactly the opposite. They are talking about extending maturities instead of shortening them. They say 50- or even 100-year bonds are coming under a Trump administration. Here’s Kudlow, in a post called Why Not 100-Year U.S. Treasury Bonds? for example:
The average duration of marketable Treasury bonds held by the public has been five years for quite some time. Almost incredibly, Treasury Department debt managers have not substantially lengthened the duration of bonds to take advantage of generationally low interest rates. Hard to figure.
Treasuries held in public hands have moved up from 32 percent of GDP back in 2008 to 74 percent today. Interest expense for fiscal 2016 is nearly $250 billion. So if Treasury debt managers had significantly lengthened their bond maturities, they would have saved taxpayers a bundle.
Now, with new economic-growth policies poised to drive up average Treasury rates to perhaps 6 percent, the Treasury folks better get moving fast to capture today’s historically low yields. Up to now they’ve been sleeping at the switch.
The key point? Start issuing much longer bond maturities. Much longer. If possible, the U.S. should experiment with 50-year debt issuance, and maybe go out as long as 100 years. And this better happen fast.
Who’s right, Rubin or Mnuchin? I’m firmly in Rubin’s camp here and let me tell you why.
Long-term interest rates are simply a series of short-term interest rates plus a term-premium. As such, long-term interest rates are almost always higher than short-term rates, making the yield curve upward sloping. That means for an issuer of debt securities, extending maturities increases the cost of issuance rather than decreasing it.
Now, the reason private debtors might forego lower interest payment and lengthen maturities is rollover risk – meaning they want to lock in borrowing rates without having to worry about rolling over their debt obligations in 30 days or even 2-years. If you recall from the financial crisis in 2008, banks that were excessively dependent on wholesale and short-term funding markets faced significant liquidity risks that threatened their solvency.
With sovereign currency issuers, there is no solvency risk. So rollover risk only presents them with a risk that interest rates will be higher at some point in the future. Sovereign currency borrowers won’t literally go bankrupt since they create the money in the currency area in which they are issuing bonds.
But what about the situation Kudlow describes where strong economic growth leads to rising rates and higher yields on government bond issuance? Well, that’s not an issue unless the Fed is setting real interest rates significantly higher than they should. Take this example. The Treasury could have issued 2-year paper with a coupon of 1% back in October. Or it could have issued a 30-year bond with a coupon of 2.875%. Now what Rubin is telling us is he would rather issue a lot more of the 1% notes than the 2.875% bonds. Fair enough. And notice that with core PCE inflation measures of 1.73, the real yields here are negative for the short-term paper at – 0.73%. In real terms, the Treasury is being paid 73 basis points. The real yield is 1.145% for the 30-year bond.
But what if the Trump Administration pulls all the right levers and rolls all the right dials and sustained real GDP growth soars to 4% by 2020? Suddenly the Fed has jacked up rates. Maybe interest rates are now 4% on the 2-year bond and 5.5% on the 30-year bond. With inflation at 2%, the Treasury could suddenly be paying 2% in real terms to issue two-year notes. That’s a lot more than the 1.145% it paid in real terms for 30-year paper four years before.
My view: the only two reasons monetarily sovereign public bond issuers should forego lower interest cost by lengthening maturities is to help funds manage asset/liability duration matching and to provide a real yield for investors in safe assets. That yield is normally more in the 2% range for long-term safe assets if you look at the data. In 2006, even 5-year paper came with a real yield of over 2%. But even if the Treasury pays 2% on two-year paper as I set out in my example, that is a scenario you actually want to have – strong economic growth, strong tax receipts, low unemployment. What is there to worry about?
Kudlow’s concern about locking in low yields doesn’t make any sense to me. There is no solvency constraint for sovereign currency issuers. When the economy is strong, tax receipts are high and interest costs recede as a concern. The only time interest payments are a political constraint is when the economy is doing poorly. And that’s a case in which shortening maturities makes more sense than lengthening them.