« Kenneth Rogoff: Time for Change at the World Bank | Main | Keeping the Public Fully and Honestly Informed »

Thursday, April 19, 2007

Thomas Palley: The Case Against Inflation Targeting

I favor inflation targeting, or at least lean strongly in that direction. Thomas Palley does not, and he is by no means alone. Here are his reasons for opposing inflation targets. My rebuttal follows:

The Case Against Inflation Targeting, by Thomas Palley: A few months ago the Federal Reserve seemed to be inexorably moving toward adopting an inflation targeting policy regime. Fed Chairman Ben Bernanke is known to support such a framework having co-authored several articles and books on the subject. Moreover, his institutional hand had been strengthened by Frederic Mishkin’s appointment as Vice-Chairman of the Fed, Mishkin being one of Bernanke’s co-authors.

This situation has been transformed by Congressman Barney Frank’s assumption of the Chairmanship of the House Financial Services Committee. That is because Frank is known to be an inflation targeting skeptic, and his Chairmanship sets the stage for an important policy debate that also pertains to policy in Europe, Canada, and developing economies.

The economics profession is widely identified with supporting inflation targeting. However, the profession is also unusually monolithic, which makes it important that alternative positions get heard.

Former Federal Reserve Chairman Alan Greenspan long opposed inflation targeting on “process” grounds. Though known as an opponent of inflation, Greenspan’s view was that an explicit inflation target would be a millstone around the Fed’s neck. In particular, it would limit flexibility and discretion to adjust policy in response to unexpected circumstance. Additionally, inflation targeting could promote damaging mechanical policymaking, as happened with money supply targeting in the late 1970s. Lastly, it could provide an anvil on which financial markets could hammer and corner policy.

Greenspan’s opposition reflects his policymaker acumen. But there is a deeper case against inflation targeting that rests on economic theory and differences in the way the economy is understood.

Today’s economic consensus is rooted in Milton Friedman’s notion of a natural rate of unemployment that harkens back to classical economics, which ruled thinking in the Great Depression era. According to this classical view inflation and monetary policy have no lasting impacts on the unemployment rate or real wages, which are determined by labor supply and demand conditions that are independent of inflation. Moreover, inflation cannot affect economic growth, which is determined by labor force growth and the rate of technological advance that are again independent of inflation.

These propositions represent today’s consensus. But there is another view that holds there is a trade-off between inflation and unemployment. Slightly higher inflation can reduce unemployment because it greases the wheels of adjustment in labor markets. It is hard to lower wages because of trust and conflict issues. Consequently, instead of lowering wages in depressed sectors it is more efficient to raise wages and prices in the rest of the economy, thereby accomplishing the relative price adjustments needed to restore full employment.

Lower unemployment can in turn raise real wages because it gives workers more job options and increases their bargaining power. Finally, higher wages may raise growth by strengthening consumer demand conditions and stimulating investment – but this requires the economy to be “wage-led”. Alternatively, if the economy is “profit-led”, higher wages may lower growth by diminishing profitability and discouraging investment.

The important point is there are two competing coherent views of the economy, and they generate very different policy perspectives. According to the current consensus, monetary policy only affects inflation, which encourages very low inflation as the Fed’s goal. However, the minority post Keynesian view sees monetary policy as having lasting impacts on inflation, unemployment, real wages, and maybe growth. That makes inflation one concern among many.

This difference means that inflation targeting is an undesirable frame for public policy. Inflation is a “bad” in the sense that we would all prefer high employment without inflation. Consequently, if monetary policy is presented purely in terms of an inflation target, there will be a tendency to choose a low target. If the choice is presented as two versus three percent inflation, two percent is likely to win out and policy will also tend to privilege addressing inflation risks over employment risks.

However, if the reality is monetary policy impacts a quadruple consisting of inflation, unemployment, real wages, and growth, two percent inflation may be inferior to three percent inflation accompanied by higher real wages, lower unemployment, and possibly faster growth. That is the economic logic behind skepticism about inflation targeting as a policy framework.

As I noted above and many times previously (comments about Barney Frank in particular here), I don't agree. As for the specifics for this argument, the empirical evidence does not support the view that monetary policy has long-run impacts on real variables such as output growth. There are some results showing that increasing money growth (but not inflation) might increase output growth, and other results showing growth falls with inflation, but they are not very strong and it is generally agreed that money has no long-run impact.

Let me do this right. Here is a nice summary of empirical work in this area from page 1 of chapter 1 of Carl Walsh's Monetary Theory and Policy, 2nd. ed., The MIT Press, 2003:

1.2.1 Long-Run Relationships A nice summary of long-run monetary relationships is provided by McCandless and Weber (1995). They examine data covering a 30-year period from 110 countries using several definitions of money. By examining average rates of inflation, output growth, and the growth rates of various measures of money over a long period of time and for many different countries, McCandless and Weber provide evidence on relationships that are unlikely to be dependent on unique, country-specific events (such as the particular means employed to implement monetary policy) that might influence the actual evolution of money, prices, and output in a particular country. Based on their analysis, two primary conclusions emerge.

The first is that the correlation between inflation and the growth rate of the money supply is almost 1, varying between 0.92 and 0.96, depending on the definition of the money supply used. This strong positive relationship between inflation and money growth is consistent with many other studies based on smaller samples of countries and different time periods.[1] This correlation is normally taken to support one of the basic tenets of the quantity theory of money: a change in the growth tale of money induces "an equal change in the rate of price inflation" (Lucas 1981 p. 1005). This high correlation does not, however, have any implications for causality. If the countries in the sample had followed policies under which money-supply growth rates were exogenously determined, then the correlation could be taken as evidence that money growth causes inflation, with an almost one-to-one relationship between the two. An alternative possibility, equally consistent with the high correlation, is that other factors generate inflation, and central banks allow the growth rate of money to adjust. Any theoretical model not consistent with a roughly one-for-one long-run relationship between money growth and inflation, though, would need to be questioned.[2]

The appropriate interpretation of money-inflation correlations, both in terms of causality and in terms of tests of long-run relationships, also depends on the statistical properties of the underlying series. As Fischer and Seater (1993) note, one cannot ask how a permanent change in the growth rate of money affects inflation unless actual money growth has exhibited permanent shifts. They show how the order of integration of money and prices influences the testing of hypotheses about the long-run relationship between money growth and inflation. In a similar vein, McCallum (1984b) demonstrates that regression-based tests of long-run relationships in monetary economies may be misleading when expectational relationships are involved.

McCandless and Weber's second general conclusion is that there is no correlation between either inflation or money growth and the growth rate of real output. Thus, there are countries with low output growth and low money growth and inflation, and countries with low output growth and high money growth and inflation and countries with every other combination as well. This conclusion is not as robust as the money-growth-inflation one; McCandless and Weber report a positive correlation between real growth and money growth, but not inflation, for a subsample of OECD countries. Kormendi and Meguire (1984) for a sample of almost 50 countries and Geweke (1986) for the United States argue that the data reveal no long-run effect of money growth on real output growth. Barro (1995, 1996) reports a negative correlation between inflation and growth in a cross-country sample. Bullard and Keating (1995) examine the post-World War 11 data from 58 countries, concluding for the sample as a whole that the evidence that permanent shifts in inflation produce permanent effects on the level of output is weak, with some evidence of positive effects of inflation on output among low-inflation countries and zero or negative effects for higher-inflation countries.[3] Similarly, Boschen and Mills (1995b) conclude that permanent monetary shocks in the United States made no contribution to permanent shifts in GDP. Thus, there is somewhat greater uncertainty as to the relationship between inflation and real growth, and other measures of real economic activity such as unemployment, in the long run, but the general consensus is well summarized by the proposition, "about which there is now little disagreement.... that there is no long-run trade-off between the rate of inflation and the rate of unemployment" (Taylor 1996, p. 186).

Monetary economics is also concerned with the relationship between interest rates, inflation, and money. According to the Fisher equation, the nominal interest rate equals the real return plus the expected rate of inflation. If real returns are independent of inflation, then nominal interest rates should be positively related to expected inflation. This relationship is an implication of the theoretical models discussed throughout this book. In terms of long-run correlations, it suggests that the level of nominal interest rates should be positively correlated with average rates of inflation. Because average rates of inflation are positively correlated with average money growth rates, nominal interest rates and money growth rates should also be positively correlated. Monnet and Weber (2001) examine annual average interest rates and money growth rates over the period 1961 to 1998 for a sample of 31 countries. They find a correlation of 0.87 between money growth and long-term interest rates. For the developed countries, the correlation is somewhat smaller (0.70); for the developing countries, it is 0.84, although this falls to 0.66 when Venezuela is excluded.[4] This evidence is consistent with the Fisher equation.

Mishkin (1992) examines the interest rate-inflation correlation directly by testing for the presence of a long-run relationship between the two. Using U.S. data, Mishkin finds the evidence to be consistent with the Fisher relationship.

1. Examples include Lucas (1980b), Geweke (1986), and Rolnick and Weber (1994), among others. A nice graph of the close relationship between money growth and inflation for high-inflation countries is provided by Abel and Bernanke (1995, p. 242). Hall and Taylor provide a similar graph for the G-7 countries (Hall and Taylor 1997, p. 115). As will be noted, however, the interpretation of correlations between inflation and money growth can be problematic.

2. Haldane (1997) finds, however, that the money growth rate-inflation correlation is much less that '. among low-inflation countries.

3. Kormendi and Meguire (1985) report a statistically significant positive coefficient on average money growth in a cross-country regression for average real growth. This effect, however, is due to a single observation (Brazil), and the authors report that money growth becomes insignificant in their growth equation when Brazil is dropped from the sample. They do find a significant negative effect of monetary volatility on growth. 4. Venezuela's money growth rate averaged over 28°r., the highest among the countries in Monnet and Weber's sample.

4. Venezuela's money growth rate changed over 28%, the highest among the countries in Monnet and Weber's sample.

    Posted by on Thursday, April 19, 2007 at 11:11 AM in Economics, Inflation, Monetary Policy | Permalink  TrackBack (0)  Comments (67)


    TrackBack URL for this entry:

    Listed below are links to weblogs that reference Thomas Palley: The Case Against Inflation Targeting:


    Feed You can follow this conversation by subscribing to the comment feed for this post.