I can be reasonably certain of only one point: My economic forecast is highly likely to be wrong--but I don't know how. One implication of this pervasive uncertainty is that any statement about the future path of monetary policy must be conditional--dependent on the economy following the expected path. Although the FOMC has stated that the federal funds rate is likely to remain exceptionally low for an extended period, this statement explicitly depends on an economic outlook similar to the one I have given today. We cannot provide a precise timetable for when short-term interest rates will begin to return to normal because that depends on the evolution of actual and projected activity and inflation.

One implication of this pervasive uncertainty is that any statement about the future path of monetary policy must be conditional--dependent on the economy following the expected path. Although the FOMC has stated that the federal funds rate is likely to remain exceptionally low for an extended period, this statement explicitly depends on an economic outlook similar to the one I have given today. We cannot provide a precise timetable for when short-term interest rates will begin to return to normal because that depends on the evolution of actual and projected activity and inflation.

In my experience, these and other considerations put a premium on flexibility. The need to learn from and respond to news means that policy should have a substantial discretionary component. We have certainly needed to innovate over the past several years to contain the damage from unprecedented events in financial markets. But discretion has its limits as well. We must be able to explain and justify our actions within a coherent framework--even if the elements of that framework are adjusted from time to time as experience dictates. And to the extent that we can act predictably, households and businesses will be able to anticipate our actions, reinforcing their effects. Finally, we must not be flexible about our objectives. The goals of monetary policy--price stability and maximum employment--are stable and well known. The flexibility relates to the actions we take to get there.

As a result, policymakers must look to lessons that are common across alternative specifications and base policy on our current understanding of the most likely important structural factors.  Fortunately, I think that many of the models of nominal price and wage adjustment imply similar conclusions regarding the influence of commodity prices on the inflation outlook and the appropriate response of monetary policy.

Unfortunately, as noted above, empirical research suggests that no one measure of core inflation will work in all situations.  Therefore, central banks do not focus solely on core inflation; rather, they devote considerable resources to understanding inflation developments in an effort to distinguish signal from noise in the incoming data

The past decade has also witnessed significant progress in analyzing the policy implications of uncertainty regarding the structure of the economy...

Although Bayesian and robust-control methods provide insights into the nature of optimal policy, the corresponding policy recommendations can be complex and sensitive to the set of economic models being considered.  A promising alternative approach--reminiscent of the work that Bill Poole did in the 1960s--focuses on simple policy rules, such as the one proposed by John Taylor, and compares the performance of alternative rules across a range of possible models and sets of parameter values (Levin, Wieland, and Williams, 1999 and 2003).  That approach is motivated by the notion that the perfect should not be the enemy of the good; rather than trying to find policies that are optimal in the context of specific models, the central bank may be better served by adopting simple and predictable policies that produce reasonably good results in a variety of circumstances.

To live up to what is expected of us, we have to make considered judgments and not react to the latest data point or the "instant analysis" that is ubiquitous on the Internet or in the news media or among the countless financial analysts who pump out commentary like water from a fire hydrant.

The Board staff employs a variety of formal models, both structural and purely statistical, in its forecasting efforts. However, the forecasts of inflation (and of other key macroeconomic variables) that are provided to the Federal Open Market Committee are developed through an eclectic process that combines model-based projections, anecdotal and other "extra-model" information, and professional judgment. In short, for all the advances that have been made in modeling and statistical analysis, practical forecasting continues to involve art as well as science.

Nonetheless, during my tenure at the Fed, the FOMC has had to react to a number of important and difficult challenges: the Asian financial crisis, the Russian debt default, major movements in asset prices, the acceleration in productivity, Y2K, 9/11, and the risk of deflation. All of these issues generated policy questions that did not fit neatly into any familiar textbook framework. They exemplify how, when making tough decisions in unusual circumstances, it's important to follow sound policy-making principles:

• Look at a wide range of data and information, instead of one or two summary indicators;
• Use cogent economic theory to shape analysis;
• And respect the risks of undesirable outcomes for growth or inflation, even in environments that appear benign.

In particular, over the past few decades the natural unemployment rate and the path of potential output have apparently moved around quite substantially. If we do not recognize the potential for such shifts, they can pose serious pitfalls for the conduct of monetary policy...

To be sure, central banks need to form some views about the economy's potential to produce on a sustained basis. After all, as I have already noted, the amount of slack in the economy is a key determinant of inflation. But, rather than focusing on fixed estimates of potential output or the natural rate of unemployment, central banks should take an eclectic approach in assessing the overall balance of economic activity relative to productive capacity. In other words, in pursuing the dual mandate, the central bank should recognize that a wide variety of indicators drawn from labor, product, and financial markets provide information about the overall balance of supply and demand in the economy. In addition, central banks should use information from various price indicators to tell them whether the economy is overheating or running well below productive capacity.

Given this reality, policymakers are well advised to follow two principles familiar to navigators throughout the ages:  First, determine your position frequently. Second, use as many guides or landmarks as are available In the context of monetary policy, these principles suggest that policymakers should monitor bond yields carefully in judging the current state of the economy--but only in tandem with the signals from other important financial variables; direct readings on spending, production, and prices; and a goodly helping of qualitative information. Ultimately, a robust approach to policymaking requires the use of multiple sources of information and multiple methods of analysis, combined with frequent reality checks. By not tying policy to a small set of forecast indicators, we may sacrifice some degree of simplicity, but we are less likely to be misled when a favored variable behaves in an unusual manner.