Global Economy Chris Edmond Monetary Policy Revised: January 9, 2008 In most countries, central banks manage interest rates in an attempt to produce stable and predictable prices. In some countries they are also asked to moderate fluctuations in output and employment. We discuss these roles and summarize actual policy in terms of the popular Taylor rule. Roles of monetary policy It s hard to talk about what central banks should do before considering what they re capable of doing. There is clear evidence that they can control the price level and inflation, at least over periods of a year or more. There is also clear evidence that countries with high and variable inflation rates have poor macroeconomic performance, although the cause and effect are less clear. With these facts in mind, most countries charge their central banks with producing stable and predictable prices. In practice, this is typically understood to mean a stable inflation rate of 2-3% a year. In the US, the Federal Reserve Act asks the Fed to promote the goals of maximum employment, stable prices, and moderate long-term interest rates. The term maximum employment is interpreted to mean that the Fed should act to reduce the magnitude and duration of fluctuations in output and employment. What role does monetary policy play in these fluctuations? Expert opinion is currently that monetary policy has, at best, a modest impact. Certainly, the long-term growth rate of the economy does not depend on a country s monetary policy, but rather on its institutions and productivity. In the short run, expansionary monetary policy (low interest rates, high money growth) probably has a modest positive effect on employment and output, but the effect is almost certainly small. Worse, too much monetary expansion seems to lead not to higher output but to high inflation and lower output. Most experts suggest, therefore, that central banks (including the Fed) should emphasize price stability and give secondary importance to output and employment. Fed Chair Ben Bernanke puts it this way: The early 1960s [were] a period of what now appears to have been substantial overoptimism about the ability of [monetary] policymakers to fine-tune the economy. Contrary to the expectation of that era s economists and policymakers, the subsequent two
Monetary Policy 2 decades were characterized not by an efficiently managed, smoothly running economic machine but by high and variable inflation and an unstable real economy, culminating in the deep 1981-82 recession. Although a number of factors contributed to the poor economic performance of this period, I think most economists would agree that the deficiencies of... monetary policy including over-optimism about the ability of policy to fine-tune the economy... and insufficient appreciation of the costs of high inflation played a central role. Predictability Another hard-earned lesson of monetary policy is that it s important to be predictable. Firms and workers need to know what inflation is likely to be when they agree to wages. Investors and issuers of bonds need to know inflation to judge interest rates. Uncertainty about future policy makes it more difficult to judge future inflation in a world in which many prices are set in nominal terms. As Bernanke suggested, the unpredictability of policy in the 1970s was a factor in the poor macroeconomic performance of that decade. With this lesson in mind, most central banks adopted procedures over the last two decades that make policy more predictable. Several countries followed money growth rate rules, in which the target growth rate of the money supply was announced in advance. Many developing countries adopted fixed exchange rates against the dollar or euro, with the intention of making the value of their currencies predictable in terms of a more stable currency. The difficulty here is that fixed exchange rate systems sometimes collapse, making the currencies less predictable. Various means have been used to make exchange rate regimes more durable, with mixed success. The logic in all of these examples is that predictability of future policy allows firms and individuals to understand the impact of their economic decisions. Taylor s rule Stanford economist John Taylor suggested in 1993 an interest rate rule would provide a relatively simple summary of good monetary policy as practiced in developed countries at that time and now. The rule consists of the following equation: i t = r + π t + a 1 (π t π) + a 2 (y t y t ), (1) where i t is the short-term nominal interest rate, r is a target real interest rate, π t is the continuouslycompounded inflation rate, π is the target inflation rate, and y t y t is the deviation of log real GDP (y t ) from its trend (y t ). The parameters (a 1, a 2 ) indicate the sensitivity of the interest rate
Monetary Policy 3 to inflation and output and thus capture the two goals of monetary policy: stable prices and (possibly) reduced cyclical variation in output. This is a lot to swallow the first time you see it, so let s work our way though piece by piece as this is applied to US monetary policy. Nominal interest rate i t. Standard practice is to use the fed funds rate. In the US, commercial banks and other depository institutions have accounts (deposits) at the Federal Reserve that are referred to as fed funds. They trade these deposits among themselves in an overnight fed funds market. The Fed currently indicates its policy stance by setting an explicit target for the interest rate on these trades. For practical purposes, this rate is the very short end of the yield curve. Target real interest rate r. Experience suggests that the real fed funds rate (nominal rate minus inflation) has averaged about 2% over the last two decades, but it might very well move around over time, both over long periods of time (the 1980s had unusually high real interest rates) and over the business cycle. For practical purposes, we set r = 0.02 and use it as an anchor on the level of the fed funds rate. The first component of the target fed funds rate is thus the target real rate (2%) plus the current inflation rate, thus giving us a nominal interest rate. Inflation deviation (π t π). The next term is a reaction to the difference between current inflation and the target. If the target is 2% and actual inflation is 3%, then we increase the nominal fed funds rate by a 1. Typically a 1 > 0, meaning that we increase the interest rate in response to above-target inflation. Why? Because higher interest rates are associated slower money growth and therefore (eventually) lower inflation. Larger values of a 1 indicate more aggressive reactions to inflation. In some implementations, the current inflation rate is replaced by the expected inflation rate over the coming year. Output deviation (y t y t ). The final term is a reaction of the interest rate to deviations of output from trend. For example, if output is below trend, the Fed would reduce the interest rate in an attempt to push output up. Larger values of a 2 are associated with more aggressive reactions to output. The critical element here is the trend. We use a log-linear time trend for real GDP, estimated from the regression: y t = constant + γ t, where t is time (the date, measured in years) and γ is the estimated average continuouslycompounded annual growth rate. Notice however that this hides an important issue: if we thought that the economy s long-term growth rate had increased for some reason, we would
Monetary Policy 4 have to adjust the trend accordingly. In practice, it s not always clear whether a current increase in output is a change in the deviation from trend or an increase in the trend itself. (Think of the 1990s boom, for example). We can get a sense of how this works in practice by considering specific numbers. Taylor suggested parameter values of a 1 = a 2 = 0.5, giving equal weight to inflation and output deviations. With these values, Taylor has i t = 0.02 + π t + 1 2 (π t 0.02) + 1 2 (y t y t ) But we can also attempt to give values to the parameters of the rule by running a regression. We estimated the parameters this way. First we estimated the trend, finding that γ = 3%. Then we used the trend to estimate this variant of the Taylor rule: i t = (r a 1 π) + (1 + a 1 )π t + a 2 (y t y t ). Estimates of this linear regression over the period 1987-2004 imply a 1 = 0.26 and a 2 = 0.47. Figure 1 shows how the fitted values of the fed funds rate compare to the values set by the Fed over the same period. 10 United States! The Taylor Rule at work 8 Percent 6 4 2 0 1988q3 1992q3 1996q3 2000q3 2004q3 Taylor Rule Federal Fund Rate Figure 1: Fed Funds rate and Taylor rule approximation for the US, 1987-present.
Monetary Policy 5 Similar rules seem to approximate monetary policy in other time periods and countries. In the US, there seems to be a sharp change in policy in 1979, when Paul Volcker became chairman of the Federal Reserve. Estimates for earlier periods suggest much weaker responses to inflation: estimates of a 1 based on pre-1979 data are typically negative, implying that when confronted with inflation above target π, the Fed was raising the nominal interest rate less than one-for-one. That is, the Fed was letting the real interest rate r t = i t π t fall, adding further stimulus to the economy when it ought to have been trying to keep inflation under control. More active responses to inflation since 1979 have led to lower inflation. Some also attribute to monetary policy part of the decline in output volatility we ve seen since the mid-1980s. In Europe and Japan, estimated rules are similar, again suggesting that current monetary policy places greater weight on inflation. The funds rate and other market rates [Adapted from comments by Ben Bernanke, March 30, 2005]. So far we have discussed how the Federal Reserve manages the federal funds rate. How does the Fed influence other interest rates? The interest rates most closely linked to the funds rate are those prevailing in short-term money markets. The federal funds market is tied particularly closely to the so-called Eurodollar market. Eurodollar deposits are dollar-denominated deposits held at non-us banks. Regulatory and technological developments have made these deposits easily tradable even by US banks and by many nonbank institutions not eligible to participate in the fed funds market. Because large banks can trade in either the federal funds market or the Eurodollar market and because fed funds and Eurodollars are easily substitutable forms in which to hold short-term liquidity, overnight Eurodollar interest rates line up closely with the funds rate, even within the day. Were that not the case, then banks could profit by borrowing in the cheaper market and lending in the market in which the rate is higher. As a consequence of this potential arbitrage, the FOMC s target for the funds rate effectively determines other very short-term rates as well. How does the Fed (try to) influence longer-term rates such as mortgage rates and corporate bond rates? For the time being, let s focus on the relationship between the FOMC s actions and the yields on longer-term Treasury securities, such as five-year or ten-year Treasury notes. If monetary policy can affect Treasury yields, then it can also affect other yields. For example, mortgage rates are closely linked to long-term Treasury yields, the spread between the two rates being explained largely by factors such as the risk of default on mortgage loans and the so-called prepayment risk (that is, the risk that homeowners will choose to pay off their mortgages early). Any longterm interest rate, such as a five-year or ten-year Treasury rate, is the sum of two components: a weighted average of expected short-term interest rates and a term premium. Thus monetary policy
Monetary Policy 6 has a chance to affect Treasury yields through its effect on the expected future path of short-term interest rates. In other words, to affect long-term rates, the FOMC must somehow signal to the financial markets its plans for setting future short-term rates. How can this be done? The most direct method is through talk. The FOMC s post-meeting statement, the minutes released three weeks after the meeting, and speeches and congressional testimony by the Chairman and other Federal Reserve officials all provide information to the markets and the public about the nearterm economic outlook, the risks to that outlook, and the appropriate course for monetary policy. With the aid of this information, financial market participants make estimates of the likely future path of short-term interest rates, which in turn helps them to price longer-term bonds. FOMC talk probably has the greatest influence on expectations of short-term rates a year or so into the future, as beyond that point the FOMC has very little, if any, advantage over market participants in forecasting the economy or even its own policy actions. Influencing policy expectations for the more distant future may be more difficult. However, the FOMC has two general ways to help financial market participants divine the long-run course of policy. First, to the extent practical, the FOMC strives to be consistent in how it responds to particular configurations of economic conditions and transparent in explaining the reasons for its response. By building a consistent track record, the FOMC increases its own predictability as well as public confidence in its policies. Second, more generally, comments by FOMC officials about the Committee s general policy framework help the public deduce how policy is likely to respond to future economic circumstances. Importantly, FOMC members do not have to guess about the effects of their words and actions on the public s expectations of future policy actions. Information about those expectations is revealed in a number of ways in financial markets. For example, market prices on actively-traded futures contracts on the funds rate or on the Eurodollar rate tell us a great deal about the funds rate that market participants expect to prevail at various dates in the future. To conclude, the FOMC controls very short-term interest rates fairly directly. However, the Committee s control over longer-term yields and over the prices of long-lived financial assets depends crucially on its ability to influence market expectations about the likely future course of policy. In the past decade or so, the Federal Reserve has become substantially more transparent and open in its communication with the public. Executive summary 1. Monetary policy is typically charged with price stability. In some countries, output growth and stability may also be goals. 2. One of the key features of monetary policy is its predictability. 3. Taylor s rule relates a central bank s choice of short-term interest rate to deviations of inflation from a target value and of output from trend. Estimates suggest that central banks
Monetary Policy 7 in developed countries now place greater weight on inflation than output, perhaps because their ability to affect output is limited. 4. The impact of monetary policy on long-term rates depends in part on the central bank s ability to communicate its long-term intentions. Further reading For more on the art of monetary policy: Fed Governor Bernanke s Remarks on US monetary policy, February 2003; the source of the quotation and a great overview of US policy since the 1960s and our changing views of the impact of monetary policy on the economy. His Implementing Monetary Policy is the source of the discussion in the last section. The Bank for International Settlements links to central banks. John Taylor s policy rule web site. c 2008 NYU Stern School of Business