Federal Reserve Chair Janet Yellen testifies before a Joint Economic Committee hearing on Capitol Hill last week. (Mark Wilson/Getty Images)

Lawrence Summers is a professor at and past president of Harvard University. He was treasury secretary from 1999 to 2001 and an economic adviser to President Obama from 2009 through 2010.

While debates about the role of secular stagnation in current economic conditions continue to rage, there is now almost universal acceptance of a crucial part of the argument — that the “neutral interest rate” has declined substantially and is likely to be lower in the future than in the past throughout the industrial world because of growing saving relative to investment. The idea that real interest rates — that is, interest rates adjusted for inflation — will be lower going forward is reflected in the pronouncements of policymakers such as Federal Reserve Chair Janet Yellen, the medium-term forecasts of both official agencies like the Congressional Budget Office and the International Monetary Fund, and the pricing of government bonds whose payments are tied to inflation.

This is important progress and has contributed to more prudent monetary policies than would have been pursued otherwise and the avoidance of a deflationary psychology taking hold, particularly in Europe and Japan. Despite having adjusted their views, policymakers still overestimate the extent to which neutral real interest rates — the level consistent with full employment and stable inflation — will rise.

Neutral real interest rates may well rise over the next few years should the U.S. economy create jobs at a rapid pace and the effects of the financial crisis diminish. This is what many expect, though the fact that an imminent move toward historically normal interest has been widely expected for the past six years should invite skepticism.

A number of considerations make me doubt the U.S. economy’s capacity to absorb significant increases in real rates over the next few years. First, real rates were trending down for 20 years before the crisis began and have continued down since. Second, there is a significant risk that as the rest of the world and especially emerging markets struggle, there will be substantial inflows of capital into the United States, leading to downward pressure on rates and upward pressure on the dollar, which in turn would reduce demand for traded U.S. goods. Third, the increases in demand achieved through low rates in recent years have come from pulling demand forward in time, leading to lower levels of demand in the future; for example, lower rates have accelerated purchases of cars and other consumer durables and created apparent increases in wealth as asset prices inflated. Fourth, profit rates are starting to turn down, and regulatory pressures are increasingly inhibiting lending to small and medium-size businesses. Fifth, it may be that inflation mismeasurement is increasing, as the share in the economy of hard-to-measure items such as health care rises. If so, apparent neutral real interest rates will decline even if there is no change in properly measured real rates.

All of this leaves me far from confident that there is substantial scope for tightening interest rates in the United States — and probably even less scope in other parts of the industrialized world. The fact that central banks in Europe, Sweden, Israel and a number of other countries where rates had been zero found themselves reversing course after raising rates adds to the cause for concern.

But there is a more profound worry. U.S. and international experience suggests that once a recovery is mature, the odds that it will end within two years are about half and that it will end in less than three years are over two-thirds. Because normal growth is now below 2 percent rather than near 3 percent, as has been the case historically, the risk may even be greater now. While the risk of recession may seem remote given recent growth, it bears emphasizing that since World War II, no postwar recession has been predicted a year in advance by the Fed, the White House or the consensus forecast.

Historical experience suggests that when recession comes it is necessary to cut interest rates by more than 300 basis points. I agree with the market that the Fed likely will not be able to raise rates by 100 basis points a year without threatening to undermine the recovery. But even if this were possible, the chances are very high that recession will come before there is room to cut rates by enough to offset it. The knowledge that this is the case must surely reduce confidence and inhibit demand.

Central bankers bravely assert that they can always use unconventional tools. But there may be less in the cupboard than they suppose. The efficacy of further quantitative easing in an environment of well-functioning markets and already very low medium-term rates is highly questionable. There are severe limits on how negative rates can become. A central bank that is forced back to the zero lower bound is not likely to have great credibility if it engages in forward guidance.

The Fed will in all likelihood lift rates this month. Markets will focus on the pace of the Fed’s tightening. I hope and expect that their response will involve no great turbulence. But the unresolved question that will hang over the economy is how policy can delay and ultimately contain the next recession. It demands urgent attention from fiscal as well as monetary policymakers.