The debate over the appropriate level for a central bank’s inflation objective reminds us of a 40-year-old Sherlock Holmes movie called “The Seven-Per-Cent Solution.” Convinced that Holmes’ addiction to cocaine (the solution in the title) had made him delusional, Watson took the master sleuth to Vienna to be treated by Sigmund Freud.
Has the 2% solution for inflation targeting in advanced economies made central bankers similarly delusional? Are they stubbornly attached to an outdated target? That argument gained ground in recent years as policymakers in Europe, Japan, and the United States struggled to stimulate weak economies and stabilize prices with policy interest rates stuck at the zero bound.
Our view is that if policymakers could start from scratch, they might well choose a somewhat higher inflation target. Their rationale would be to avoid having to lower the policy rate to zero again in a future recession. But the cost of changing the policy framework now would be a substantial loss of credibility, so there seems little chance for a new regime with a higher inflation target.
As a result, we will not be surprised if central banks again are compelled to use unconventional policy tools to battle a future downturn in the not too distant future. Put differently, occasional and persistent periods requiring unconventional monetary policy could very well be the new normal in low-inflation countries. This is an unattractive, but inevitable, consequence of the current framework for monetary policy.
Many volumes have been written praising the benefits of inflation targeting. Today, the number of explicit inflation targeting countries is up to 28. That doesn’t count the euro area, Japan and the United States – all of which have policy frameworks that incorporate key elements of inflation targeting. (You can find a list of the de jure inflation targeters here.) In fact, with India having recently adopted a form of inflation targeting, roughly two thirds of world GDP (measured at PPP) is currently produced in what we would call de facto inflation targeting jurisdictions.
Looking at the inflation targets themselves, the advanced economies have almost uniformly chosen an objective of 2%. Emerging market countries have picked numbers that are higher (like the new 4% central target in India). This difference is sensible, as advanced economies have systematically higher wages, and hence higher price levels than emerging economies. As the latter catch up with the former, they are likely to face a transitional period of higher inflation for nontraded goods and services (something known as the Balassa-Samuelson effect).
Returning to the advanced economies, the first countries to adopt explicit inflation targeting regimes in the early 1990s – New Zealand, Canada, the United Kingdom, and Sweden, in that order – picked a number close to 2% as their target.
Much of the discussion surrounding inflation targeting is focused on the idea that macroeconomic stability is enhanced when people’s inflation expectations are stable. But this logic only goes so far as it doesn’t differentiate between inflation targets of 1%, 2%, 5% or any other number for that matter.
So, why the 2% solution? After the experience of the 1970s and early 1980s, when inflation hit 20% in a number of countries, a broad consensus developed to restore price stability. Research on inflation measurement convinced many policymakers that commonly used consumer price indices overstate inflation by one percentage point or more. (In the United States, the 1996 Boskin Commission Report summarized this thinking. For Europe, see the papers here and here.) In other words, if the central bank could achieve measured inflation of about 1%, then true inflation would be close to zero. Not wanting to flirt routinely with deflation, the thinking went, it was safer to choose a slightly higher target.
Why allow any sustained rise in the true price level? The conventional argument is that modest inflation facilitates relative price and wage adjustments when it is costly to lower prices or wages in nominal terms. (While this view is controversial, just ask the Spaniards and Greeks who are trying to lower their prices and wages relative to those in Germany!). Research from the 1990s suggests that inflation of 3% or even a bit more would have such benefits.
Once a small group of central banks began to pick 2% as their objective, others soon followed. Then policymakers worked hard to establish that objective as credible and responsible, and to reinforce expectations that inflation over the long run would return to 2% even if there were occasional and temporary moves higher or lower. For example, after adopting inflation targeting in 1992, the Bank of England began an education program called “Target Two Point Zero” that aimed to persuade the population of the benefits of the new monetary regime. As a result, there is now an entire generation of U.K. citizens who grew up learning why 2% was the appropriate inflation target.
The biggest central banks were relative latecomers in this process. At the inception of the euro in 1999, the ECB’s Governing Council adopted a price stability target of close to, but slightly less than, 2% inflation as measured by the Harmonized Index of Consumer Prices. The U.S. Federal Open Market Committee adopted a 2% inflation objective in 2012, while the Bank of Japan raised its target to 2% from 1% only in 2013.
Why are some policy experts now questioning the 2% inflation objective? The key reason is the troubled experience that many countries have had recently with setting monetary policy at the zero lower bound (ZLB) for nominal interest rates. The idea is that the current policy regime – if credible – implies that the expected real interest rate (the nominal rate minus the expected inflation rate) cannot fall below -2% (that is, less than minus two percent). Policymakers used to think that this provided sufficient conventional policy headroom for stabilizing prices and economic activity, but they are no longer so sure.
Was the 2% solution a delusion? Is there now a new, more powerful, case for a higher inflation objective? A controversial IMF staff position note published in 2010 argues just that. The authors suggest that a 4% inflation target might make sense.
To illustrate the additional policy headroom associated with a higher inflation target, we can do a calculation based on a Taylor rule. A version of this simple policy guide can be stated as follows: set the policy rate equal to the sum of: (1) the neutral real interest rate (sometimes called the “natural rate” as described in this recent post); (2) the inflation target; (3) one half of the inflation gap (the difference between current inflation and the target); and (4) -1 times the unemployment gap (the difference between the current unemployment rate and the frictional or natural rate of unemployment).
In the long run, if monetary policy is credible, inflation will be at its target of 2% and unemployment will equal the natural rate, so that the last two terms in the Taylor rule will be equal to zero. If, as Taylor assumed, the neutral real rate is 2%, then the rule tells us that the policy rate in the long run would be set to 4%. Starting from this long-run position, if the unemployment rate were to rise to a level that is 4 percentage points above the natural rate, the Taylor rule rate implies that policymakers would hit the zero bound.
How frequently might this happen? The following chart plots the unemployment gap using the Congressional Budget Office estimate of the natural rate of unemployment. There are two episodes when the unemployment gap exceeded 4% – namely, 1982-83 and 2009-10. Put differently, in some portion of 4 out of the past 55 years (or almost 5% of the quarters since 1960), the unemployment rate has been more than 4 percentage points above the estimate of the natural rate.
Raising the inflation target by even one percentage point to 3% appears to eliminate this problem. If the inflation target is 3%, then the policy rate in the long run would be 5% (rather than 4%). Even in the Great Recession, the U.S. unemployment gap did not reach 5%, although it was briefly close.
U.S. unemployment gap, 1960 to 2014
Note: Shaded areas denote recessions. Source: Civilian unemployment rate (UNRATE) less the natural rate of unemployment (NROU) from FRED.
Several criticisms can be leveled at this very simple illustration. First, as our colleagues on the U.S. Monetary Policy Forum recently explained, the neutral real rate may only be 1%, rather than 2%. If they are correct, then we should be concerned by unemployment gaps as small as 3%. In that case, the recession that ended in 1975, when the gap went to 2.7%, would be added to list of worrisome episodes. Moreover, the number of quarters in the 55-year period with an unemployment gap at or above 3% rises to nearly 9% of the total.
Another possibility is that – when the unemployment rate begins to rise in some future cycle – inflation could be below the 2% target, so the actual Taylor rule rate could start below 4 percent. For example, had a big negative demand shock hit in the U.S. economy in 1998 when the inflation rate was less than 1%, the Taylor-rule-implied policy rate might have hit the zero bound a decade earlier than it did.
So, the risks of hitting the lower bound, combined with the view that a small amount of inflation facilitates relative wage and price adjustment, lead us to the conclusion that if policymakers were to start inflation targeting today – from scratch – some might well choose a target above 2%. Numbers like 3% or even 4% are plausible.
At the same time, several arguments have been leveled against this revisionist view. The first and oldest goes back to Milton Friedman’s famous point that the optimal return on money should be equal to that on other risk-free assets – or roughly, the neutral real rate. This implies a steady deflation (rather than inflation) of 1% to 2%.
Second, past experience suggests that the level of inflation and the volatility of inflation tend to rise together. If that holds in the future, a higher inflation target would be accompanied by less predictable inflation fluctuations, adding to the uncertainty facing households and businesses.
Third, and more recently, academics have argued that the zero lower bound itself is an artifact of the way in which we have organized our monetary system. In theory, if we could get rid of cash and all cash equivalents, and instead use an electronic money as a flexible unit of account, then a central bank could set the policy interest rate significantly below zero. However, compared to eliminating cash, it would almost certainly be simpler to aim at a slightly higher inflation rate. (We have discussed elsewhere the serious questions that this e-money approach raises about both policy and privacy. See our posts on implementing negative nominal interest rates and on eliminating paper currency.)
Finally, as we discussed in an earlier post, the move to a common inflation target across countries is a useful form of international monetary policy coordination. It is highly doubtful that the leading central banks would simultaneously agree to shift to a new common target in a relatively narrow time frame.
Our bottom line: If policymakers had it to do over again, they could very well opt for a higher inflation target. But, given where we stand today, such a fundamental change in the policy framework, both nationally and internationally, is very unlikely. Rather, the belief that academic economists and central bankers can convince their elected officials and publics that 3% or 4% – rather than 2% – is the solution may be the real delusion.