As the Mercatus Center’s Scott Sumner often says, one ought never to reason from a price change. Interest rates, like other prices, can change for all sorts of reasons; the implications of the change generally depend on the particular reason for such a change.
Consequently, there’s no simple answer to the question, “If the interest rate on bellwether bonds, such as a 30-year Treasury bond, increases by 200 basis points, will the average US citizen (or the US Treasury, or both), be better off or worse off?” The most correct answer is “It depends.”
Any nominal interest rate reflects two predominant influences: the state of economic productivity and the expected future rate of inflation. Of these influences, the inflation rate is far more variable. Most of the decline in nominal interest rates since the 1980s reflects a corresponding decline in inflation.
Lately, however, both productivity and inflation have been subdued. Inflation has been hovering around 1 percent, while annual total factor productivity growth has been bouncing between 0 percent and 0.6 percent. In light of such figures, today’s remarkably low T-bill yield of just under 2.34 percent is hardly surprising. Still, it’s disturbing to realize that this value reflects market makers’ opinion that current low rates of inflation and productivity growth are likely to persist for some time.
To say that long-term Treasury bill rates mainly reflect the course of economic productivity and inflation doesn’t mean that those rates don’t themselves depend on monetary policy. Monetary policy is, of course, an important determinant of the inflation rate and of the public’s inflation expectations. In the long run, a looser monetary policy stance means higher inflation and therefore higher nominal interest rates, ceteris paribus. In the short run, however, looser policy can, and often does, lower both nominal and real interest rates. Its ability to do so—especially its ability to lower long-run rates—will be limited to the extent that it results in relatively rapid, upward adjustment in inflation expectations.
In any event, monetary easing alone can’t reduce rates for long, though it may appear to do so when it happens to coincide with a decline in either productivity growth or inflation expectations. It follows that, despite popular opinions to the contrary, easy money hasn’t had much—if anything at all—to do with the low rates that have prevailed since 2009. Had monetary policy really been easy all this time, spending growth and inflation would not have remained so subdued.
The more complicated truth is that, although the Fed has added trillions to the monetary base, the demand for both cash reserves and other relatively safe assets has also grown proportionately. That growth is in part a result of other Fed policies, including the decision to reward banks for holding reserves; the adoption and enforcement of Basel III’s Liquidity Coverage Ratio; and the more stringent regulation, if not outright prevention, of many once-conventional (and mostly prudent) kinds of bank lending. These and other measures have served to “shunt” available bank funds into a relatively limited set of markets, contributing to the “easiness” of money in those markets, while making it scarce elsewhere. Bubbles, perhaps; but no suds.
Having considered why rates are so low to begin with, it’s evident that they might rise in the near future owing to either an increase in the expected rate of inflation or an increase in the growth rate of productivity. Rates might increase by 200 basis points because the expected inflation rate increases by 200 basis points, with no change in the productivity growth rate; because the rate of productivity growth increases by 200 basis points, with no change in the expected rate of inflation; or because the two rates change by other values that sum to 200 basis points.
Some of those possibilities are of course less likely than others. For several decades now the growth rate of total factor productivity has seldom been as high or higher than 2 percent, or 200 basis points, and more recent experience suggests that it’s likely to remain well below that level for some time.
Notice that none of these possibilities depend on the Fed’s tightening its policy stance. On the contrary, whatever the more immediate interest rate effects might be, such tightening would almost certainly lead to reduced levels of both actual and expected inflation. Though the effects of Fed tightening on productivity are less predictable, those are also more likely to be negative than positive.
To come finally to the question that forms the topic of this colloquium, it should be clear by now that the general economic implications of an increase in long-term interest rates will depend on the underlying cause of the increase. An increase based on more rapid productivity growth should be a cause of celebration, for the simple reason that such productivity growth is desirable in itself. Higher interest rates will mean higher costs of borrowing, but those costs will be higher because there are more opportunities to use funds productively and because people can afford to bear the higher rates.
A substantial increase in rates based mainly or entirely on higher inflation is, in contrast, likely to do more harm than good. Even those experts who favor a rate of inflation close to 2 percent doubt that still higher rates are desirable. Because it tends to distort relative prices, inflation at such rates is likely to undermine both productivity and overall economic prosperity.
To the extent that it hasn’t been fully anticipated (and it is clear that markets today are not anticipating any substantial rise in inflation), a higher rate of inflation will also tend to reward debtors at the expense of creditors. In particular, it will reduce the government’s real debt burden at the expense of those who own non-indexed Treasury securities. The government might, therefore, benefit from an inflation-based increase in long-term interest rates, even though such an increase would make things worse for the average Joe.