In praise of an explicit number for inflation

Many central banks throughout the world have adopted an explicit, numerical objective for inflation, commonly referred to as an inflation target. The US Federal Reserve is currently not one of them, but it is discussing this possibility. In the current circumstances with the economy in freefall, is it crazy for the Fed to move in this direction? Is an increased commitment to stabilising inflation the right thing to do when we are in the throes of a financial crisis and the economy is doing so poorly?

My answer is absolutely yes. Adopting an explicit, numerical inflation objective is exactly what is needed right now to help the US economy to recover.

The usual argument for establishing a transparent and credible commitment to a specific numerical inflation objective is that it provides a firm anchor for long-run inflation expectations, thereby directly contributing to the objective of low and stable inflation. Adoption of an explicit, numerical inflation objective has been successful in other countries in keeping inflation from going too high. However, not as well recognised is that an inflation target can help prevent inflation from falling too low. At this critical juncture that benefit can have enormous value.

Until recently, inflation risks were on the upside. But the contractionary shock from the severe disruptions in the financial markets that we have been experiencing lately has shifted the economic landscape completely. Not only has the economy entered a deep recession, but inflation has plummeted. Consumer price index inflation on a year-on-year basis has fallen below 2 per cent, which I would argue is below what would be a sensible inflation objective consistent with price stability; on a three-month basis it has been falling at a negative 10 per cent rate. Core measures of inflation, which strip out food and energy prices and so potentially offer more accurate guides to underlying inflation, also indicate the risks to inflation are on the downside. Core CPI inflation on a year-over-year basis is about 2 per cent, but during the past three months has been below 0.5 per cent.

The danger right now is not that inflation expectations will be too high, but rather that they become unanchored in the negative direction. Indeed, inflation expectations – whether measured by consumer surveys, surveys of professional forecasters or the difference between interest rates on non-indexed Treasury securities and Treasury indexed bonds – all point to a sharp decline. Longer-run inflation expectations have only recently come down to levels consistent with price stability, but if they fall further they could lead to two dangerous consequences.

First, there would be an increased likelihood inflation would become persistently negative: that is deflation. Experiences of deflation in the Great Depression and the “lost decade” in Japan suggest it causes great hardship. Second, with the federal funds rate near zero and therefore unable to go lower, persistent deflation would raise the effective cost of borrowing to households and business because it would mean the interest rate adjusted for changes in the prices of real goods and services would rise. Despite an interest rate of zero, monetary policy would become highly contractionary.

How would adopting an explicit numerical inflation objective help? First, a commitment by the Federal Reserve to keep the inflation rate near an explicit objective, say 2 per cent, over a longer-term horizon would provide more incentives for the Fed – both because it would want to stick to its word and because it would subject it to more public scrutiny – to make monetary policy sufficiently expansionary in the future. Research has shown a lack of such commitment was one reason why unconventional monetary policy actions such as quantitative easing by the Bank of Japan were ineffective to promote economic recovery.

Second, when the financial system starts to recover, to keep future inflation under control the Federal Reserve will need to drain the massive amounts of liquidity it has pushed into the financial system during the past year and a half. A commitment to an explicit numerical inflation objective would encourage the Fed to explain to the public how this liquidity will be removed and subject the Fed to public pressure if it was not taking the necessary steps to make this happen.

Critics of inflation targeting fear adoption of an explicit numerical inflation objective might lead to too little focus on stabilising economic activity. Experience with central banks that have adopted explicit numerical inflation objectives shows that to do so still leaves plenty of flexibility to react to economic downturns. (That is why the term “inflation targeting” is somewhat of a misnomer because it does not mean the central bank should try to hit the target over a fixed horizon.) In addition, an explicit numerical inflation objective by the Federal Reserve would be adopted only if it was consistent with the dual mandate specified by past Congressional legislation of both price stability and stability of economic activity.


Originally published at the FT and reproduced here with the author’s permission.

One Response to "In praise of an explicit number for inflation"

  1. CookieM   January 15, 2009 at 5:31 am

    Unbelievable. The cult of inflation-targetting, formalized or not, was surely what us into this mess, encouraged by naratives of Great Globalizations and Moderations. If there is a lesson to be drawn, its that policymakers must think more broadly about setting rates. Who needs an MPC if policy is to be made a function of a single number?