Reports circulated Monday that the US Treasury is considering a new type of debt. From Bloomberg:
The U.S., seeking to attract investors who might otherwise avoid Treasuries amid a $1.3 trillion budget deficit, is considering the sale of floating- rate notes in what would be its first new security since it began offering inflation-linked debt 14 years ago.
The Treasury Department said this month it asked Wall Street’s biggest bond dealers for recommendations on structuring securities with coupons that rise or fall with benchmark rates. Officials are scheduled to gather with the 22 primary dealers, who include Goldman Sachs Group Inc. and JPMorgan Chase & Co., on Oct. 28 as it decides whether to go further during their regular meeting that precedes each quarterly refunding.
I find this idea intriguing. On the surface, with US 10-year debt hovering around 2 percent, there seems to be plenty of demand to dry up whatever Treasury issues. And it looks like a wise move to lock in as much debt as possible at those low rates. That said, at some point in the future (hopefully) rates will rise, and it is easy to see an inflection point where investors, wary that monetary policy may shift abruptly, would be wary to add to their Treasury holdings. Having a floating rate product on hand could be important in such circumstances, preventing any abrupt funding shortfalls and rates spikes by offering investors a form of insurance against rising rates.
Daniel Indiviglio at the Atlantic offers up some potential problems with such a product. First, he correctly notes the desire to lock in today's low rates. Second:
Tying U.S. debt costs to how interest rates rise and fall could also be dangerous. For every $1 billion in floating-rate debt that the Treasury issues, a 0.5% (50 basis point) increase in interest rates would result in its debt costs rising by $5 million. You can imagine how quickly U.S. debt costs could rise if the government had something like $1 trillion in floating rate debt and interest rates jumped a few percent in a year. Suddenly, the government would have to issue tens of billions of dollars in additional debt just to keep up with rising interest rates.
As I noted, I think this product would be particularly useful in the expected transition to higher rates and would not expect it to be a primary funding mechanism. I can also see a benefit in a mechanism that explicitly penalizes the fiscal authority for overspending should actual fiscal constraints emerge in the future. Consider it something of an automatic stabilizer - more spending power automatically emerges when the economy dips and debt costs fall, and vice-versa. Indiviglio offers another potential problem:
And if you think that the government puts too much pressure on the Federal Reserve to act now, just wait until it directly controls a portion of the nation's interest costs. Floating-rate note would likely include a benchmark rate controlled by the Fed. So if the U.S. is ever struggling to meet its debt obligations, the Fed may feel obligated to keep interest rates lower than it otherwise would. If it increases rates, a payment shock could affect U.S. financial stability. Due to this, floating-rate debt could lead to higher inflation, since the Fed may feel pressured to leave interest rates lower for longer.
I thought the Federal Reserve would be a natural buyer of such debt. Consider the current (what I think overplayed) worry that the Federal Reserve is threatened by huge captial losses on its existing portfolio when interest rates rise. There is some concern that the Fed will resist raising rates to avoid the erosion of it capital base, as it would lose some of its independence if monetary policymakers needed a capital injection from the US Treasury. In addition, there is a related concern that should the Fed need to raise interest on reserves abruptly to control inflation, then the Fed's expenses will surge well above the interest paid on its Treasury portfolio. Yet again another trip for a Treasury bailout.
While I was not concerned much about these scenarios, note that they would both be eliminated if the Fed held a significant portion of floating rate debt in its portfolio. It would automatically create a revenue stream to support higher interest on reserves. Obviously, the risk of capital loss would recede. Moreover, a large Federal Reserve holding of such debt would protect the taxpayer as well - higher debt servicing cost would just flow right back to the US Treasury (after Federal Reserve expenses). Finally, note then-Federal Reserve Governor Ben Bernanke made a similar proposal in his famous 2003 Japanese monetary policy speech. Essentially, a floating rate portfolio eliminates some imagined or real constraints on monetary policy, reducing the risk that additional quantitative easing will turn inflationary.
In short, I think the US Treasury has good reason to consider adding floating rate debt - and should find a ready buyer not only in Wall Street, but just across town.