I appreciate this opportunity to testify on the Federal Reserve’s conduct of monetary policy. The semi-annual monetary policy hearings for the Federal Reserve are a key element in the system of accountability you, the Congress, have established for how the Federal Reserve answers to the public for its conduct of monetary policy. As such, they are critical to maintaining the all-important degree of independence for monetary policy from short-term political interference you have granted the Federal Reserve; hearing from outside experts with divergent views on policy can only strengthen the hearings and therefore the public’s comfort with that independence—a subject I will return to at the end of my testimony.

Growth in the U.S. economy picked up in the second half of 2013 to more than 3 percent. Although some of this pick up reflected a one-time boost from an increase in inventory investment that probably now is in the process of reversing, the strengthening was also attributable to more robust growth in demands for goods and services from households and businesses. The overhangs of excess homes, autos, and other consumer durable goods that we entered the recession with have been worked off as the production of those items were cut way back over recent years. The debt that households incurred to buy these houses and durable goods has been worked down relative to income, and that plus rising wealth as low interest rates have boosted equity and housing prices have made households more willing to spend. Banks and other lenders have worked through bad loans and, with the encouragement of regulators, bolstered capital and liquidity, making them more willing and able to expand credit to support spending. Steady, albeit not spectacular, growth in jobs has reduced unemployment and added to incomes available for spending. Although some emerging market economies are struggling to maintain economic and financial stability, growth prospects in a number of advanced economies that are important US export markets have improved, and our current account deficit has declined relative to income. I expect these favorable developments to continue to support increases in spending in 2014.

To some extent, the rise in private spending in the second half of 2013 likely also reflected the ebbing of the restraining effect of the tax increases that occurred at the beginning of the year. And that points to another reason to be optimistic about 2014—federal government fiscal policy will be much less of a drag on growth. It was not only the tax increases a year ago, but also the sharp cutbacks in government spending that held back economic growth last year to the tune of 1-1/2 percentage points according to CBO estimates. With the agreement on spending you reached late last year, CBO estimates that fiscal policy will hold back growth only marginally in 2014. This should allow the underlying strengthening in the positions of the private sector to show through more convincingly and consistently into overall growth.

To be sure, some of the recent monthly data have not reflected this positive outlook. For example some surveys of manufacturers have suggested that production plans are being adjusted to deal with higher inventories. And the last two months’ reports on the labor markets over the turn of the year have suggested a slow down in hiring. But monthly data are highly volatile and subject to short-term influences, like weather. It’s my belief that the underlying fundamentals—including the continuation of a highly accommodative monetary policy—remain favorable for a bit faster growth than we have been accustomed to over most recent years. But the Federal Reserve should remain vigilant for additional indications that the expected strength in spending and hiring is not coming through.

Although growth has picked up, the US economy still is very far from where it can and should be. The unemployment rate at a little over 6-1/2 percent is still well above the 5-1/2 percent level that many economists estimate to be its sustainable level. And some of those who have dropped out of the labor force have done so because they became discouraged about finding work; one hopes that those folks will come back into the labor force as the job market strengthens further so that the unemployment rate understates the amount of labor available without adding to inflation pressures. And it’s not only labor that is underutilized; capacity utilization in US industries is a percentage point below its long-run average.

The slack in labor and capital use has resulted in very competitive conditions for businesses and workers, which have been reflected in very low inflation rates—well below the 2 per cent target set by the Federal Reserve. And cost pressures are also very damped, indicating that inflation will stay low for a while longer. Various measures of labor compensation show wages and compensation rising at a rate just above the rate of price increase and of productivity growth so that real wage have been stagnant and increases in unit labor costs of business also have been very low. We are in a risky zone for inflation: if inflation expectations start to decline, real short-term interest rates will rise hurting growth; and a downward surprise in demand could push us into or close to a destructive zone of deflation.

With unemployment of labor and capital too high and inflation too low, a highly accommodative stance of monetary policy would seem to be called for for some time to come. The Federal Reserve has put forward a 2 percent inflation target—in my view an appropriate interpretation of their price stability mandate—and they should do what they can to achieve it. They have also said that they believe that the unemployment rate can be reduced considerably further without endangering the inflation goal—and they should try to achieve that as well. Unemployed labor and capital are wasted resources that can be utilized to raise standards of living, especially, but not only, for the workers and business owners involved.

Lowering unemployment and raising inflation will require more spending relative to the economy’s capacity to produce. Faster increases in spending will directly employ greater proportions of the economy’s capital and labor resources, and indirectly raise inflation by reducing those margins of underutilized capital and labor that have been putting downward pressure on prices. The Federal Reserve can stimulate spending only by making financial conditions very easy—by its influence on interest rates and through interest rates on asset prices including the prices of equity and other forms of wealth, and on the dollar’s exchange rate to help our exporters and import competing industries. The Fed has been using two techniques to lower longer-term interest rates since short-term rates hit zero in late 2008. One has been the purchase of long-term securities—so-called QE; the other has been guidance on the circumstances in which short-term interest rates will be raised with that guidance along with the forecasts of the participants at the FOMC indicating that rates will be very low for a while longer, even as the economy approaches closer to its output potential.

A wide variety of studies have indicated that these techniques have been successful in easing financial conditions—lowering long-term rates, raising asset prices, and probably helping to keep the dollar from rising further in a troubled global economy. Logic, experience over very long periods, and observation of recent data would suggest that these steps have helped the US economy. Housing, auto sales, exports, consumption generally are stronger than they would have been if the Fed had sat on its hands in recent years. It’s hard to say with confidence how much good these unconventional polices have done; yes, its disappointing that they haven’t been more effective, though a variety of developments like very tight fiscal policy and problems overseas might explain some of the short fall; and it’s very hard to prove the counterfactual—it would have been worse—convincingly. But we have only to observe the slowdown in the improvement in the housing sector after rates rose last summer when markets anticipated a slowdown in QE to see some evidence of the effects on spending of unconventional policies.

With growth looking better and the unemployment rate having fallen to just about a point over its longer-term rate, the Federal Reserve has decided that it can dial back its security purchases, ultimately ending QE and capping its portfolio. It’s doing this gradually because any faster decrease in purchases would limit its balance sheet more than market participants expect, and that would raise interest rates at a time when the pace of expansion does not seem sufficiently robust to be immune from Fed tightening surprises, risking continued high unemployment and low inflation. And gradual reductions in purchases would give it more opportunities to pause if the very recent data do indeed portend a slower growth path than has been predicted.

But the Federal Reserve has chosen also to strengthen its articulation of its intent to keep short-term rates close to zero until the economy is stronger, and the unemployment rate and inflation closer to their objectives of maximum employment and stable prices. This seems about the right policy mix to me, especially given the very low inflation, which has not rebounded the way the Fed thought it would.

Even if the economy evolves as expected and the balance sheet stops growing near the end of this year, the Federal Reserve faces considerable challenges in the execution of monetary policy. The most important such challenge will be deciding when to begin raising interest rates and at what pace they should rise. Raise them too soon or too steeply and growth will soften and inflation remain too low. Raise them too late or too slowly and the economy would over shoot its long-run potential and if it overshoots too much or for too long inflation will settle above its 2 percent target and inflation expectations would begin to rise. In my view, the more serious mistake would be to raise them too soon or by too much. We know how to deal with extra inflation that would accompany the too-late mistake—the Federal Reserve can raise interest rates without limit. But as we’ve seen in recent years, correcting for persistent low growth and high unemployment and dangerously low inflation that would result from the too-early error is very difficult, especially when interest rates are already close to zero and the fiscal authorities are focused on deficit reduction.

Unfortunately there are no reliable formulas for making this decision. We are in uncharted waters with respect to economic circumstances and policy responses. When the economy behaves in unprecedented ways, policy must respond in unprecedented ways—and the financial crisis, the resulting great recession and sluggish recovery were unprecedented in post-war US economic history. The Federal Reserve has responded by holding rates at zero for more than five years and by trying to make up for its inability to reduce them below zero by unconventional policy actions to reduce longer-term rates, with some, but limited, success. That implies a need to hold rates lower for longer than might be implied by conventional policy rules to make up for the constraint of the zero lower bound. Moreover there’s lively discussion going on now among economists as to whether even “normal” interest rates will be lower than we are used to for a while because the potential growth of economic activity may have been negatively affected by the cutbacks in capital spending in recent years, by the costs of requiring a safer financial system backed by more equity capital, and by longer-term downtrends in productivity growth and in labor force participation as the population ages. In these circumstances there is no substitute for judgment and flexibility in the conduct of policy.

The most important way the Federal Reserve can reduce uncertainty is by achieving its Congressional mandates for employment and prices. Households and businesses in planning for the future care far more about their prospective income, sales, and the rate of inflation than they do about the size of the Fed’s portfolio or the level of interest rates. If it takes unconventional and hard-to-predict changes in the Fed’s instruments to achieve less uncertainty about variables that matter, my guess is that the public would make that trade. That’s not to argue that the Fed should deliberately behave in unpredictable ways. Rather, it should be as predictable as possible, but it and we as outside observers should recognize the limits of predictability under current circumstances. Unexpected things happen and the economy evolves in unexpected ways, reflecting in part our very limited understanding of economic relationships. Policymakers must be prepared to respond.

And that brings me to the second challenge in the years ahead—communication about policy. Having short-term rates at the zero lower bound heightens the importance of clear communication about policy. In these circumstances, influencing expectations about future interest rates and future inflation and economic activity are among the few ways the Federal Reserve has to accomplish its objectives. But communication must recognize the inherent uncertainty in policymaking, and I think that’s where the Fed got in trouble last summer; its communication left too strong an impression that it was committed to reducing its purchases beginning in the fall absent a major change in the outlook, and it tied the decision to wind down purchases in part to evolution of the unemployment rate, which is an ambiguous and increasingly difficult to understand metric for the amount of slack in the labor market and economy.

In its interest rate guidance, the Federal Reserve has said it would hold interest rates at zero until well past the time that the unemployment rate fell to 6-1/2 percent. With the unemployment rate rapidly approaching that level, the Federal Reserve will need to explain what it will be looking at to judge when it is appropriate to raise rates and how fast to raise them when it begins.

The Federal Reserve should continue to work on communicating clearly—but everyone should recognize its limits. The Federal Reserve can’t promise more certainty than consistent with a highly uncertain environment. It needs to retain flexibility to react to the unexpected. We need to recognize the limits too and not be surprised when the Federal Reserve changes course because things aren’t working out the way they thought they would. Our economy is a complex mechanism, whose state is not readily summarized in one or two variables and policy needs to react to the whole array of indicators pointing to the evolution of economic activity and prices. This complexity presents challenges for communication and guidance about how interest rates might evolve, but it is a reality.

The third key challenge associated in part with monetary policy is maintaining financial stability. Unconventional monetary policy, by driving down yields on safe assets, does encourage people to take more risks they might otherwise have done. In part this may simply overcome the very natural sharp rise in risk aversion that followed the severe financial crisis. But unconventional policies can create unusual asset price configurations, especially in bond markets, which are transmitted to other markets. These distortions, to the extent they are distortions, have been necessary to deal with the more serious distortion in our economy—the unemployment of labor and capital and the risk of deflation.

The issue is what problems might ensue as security purchases come to an end and interest rates are subsequently raised. The Federal Reserve is clearly monitoring these risks closely and using a variety of methods of discovering and dealing with potential sources of instability. It is including sharp increases in interest rates in its stress tests of the large banks; it is working with other bank regulators on increasing supervisory oversight where slippage in credit standards have been identified; and it is working with other regulators on FSOC to strengthen the financial system and make it more resilient to unexpected developments. It considers this approach to safeguarding financial stability to be superior to one in which interest rates are raised under current circumstances, which might discourage some kinds of risk taking but would also keep unemployment high and elevate the risk of deflation. I agree with this approach. As the Federal Reserve recognizes, however, if risks build despite these efforts, a policy adjustment might become necessary.

Finally, I want to return to a subject I raised in my opening paragraph. In my view, independence from short-term political interference in how the Federal Reserve calibrates its instruments will be critical for preserving price stability. Congress has set the overall goals for the Federal Reserve and it should hold the Fed accountable for achieving those goals. The Federal Reserve should be required to explain how its policy actions will lead to achieving those objectives and if they do not succeed, why they haven’t. If alternative strategies for meeting legislative objectives have been suggested and seem promising, you should ask the Federal Reserve why it has rejected these alternatives. It should also be required to discuss any adverse side effects of its actions—e.g. for financial stability—and how it would mitigate those side effects.

And Congress should periodically revisit whether it has set the appropriate goals for the Federal Reserve and whether the structure of the Fed is best suited for meeting those goals. This committee’s intention to examine many aspects of the Federal Reserve this year is appropriate and welcome. I hope this will be a nonpartisan examination, in which experts with a wide variety of views are heard and in which this committee keeps an open mind while evidence is collected.

But we need to be very careful to safeguard the arms-length relationship of the Federal Reserve to the political process when it comes to setting the instruments of policy. Much evidence over time and across countries strongly indicates that leaving the setting of policy to technical experts with some separation from day to day political pressures produces much better outcomes than when elected officials, whose focus is on the next election cycle, can influence how policy is conducted in pursuit of agreed goals.