Um excelente artigo de Laurence Ball, mesmo no momento ideal. The case for 4% inflation, no Vox. Vale a pena ir ler tudo.
In five of the eight recessions since 1960, inflation began above 4%. With high inflation, nominal-interest rates were also high, so the Fed could cut them sharply without approaching zero. But what would have happened if inflation had started at 2%?
We can get an idea by examining real interest rates. If the nominal-interest rate, i, cannot fall below zero, then the real rate, r=i-π , cannot fall below -π . One way to interpret the danger of low inflation is that it raises the lower bound on the real interest rate.
If inflation is 2% when a recession begins, the bound on the real rate is -2% at that point. However, the recession is likely to push inflation down somewhat. In the three recessions that actually started with 2-3% inflation, the inflation rate fell to about 1% before the economy recovered. History suggests, therefore, that initial inflation of 2% will produce a bound of -1% on the real interest rate.
For the recessions that started with inflation above 4%, we can gauge the relevance of a real-interest-rate bound by examining the lowest value reached by the real rate during the recession and subsequent recovery. In two of the five cases – the recessions of 1973-75 and 1980 – the real rate fell below -4%. In these episodes, a lower bound of -1% would have severely distorted monetary policy. For the recession of 1969-70, the real rate fell to a minimum of -2.3%. For the recession of 1990-91, the minimum was -0.6%; this episode would have been a near-miss with a lower bound of -1%. Only in one case, the recession of 1981-82, was the minimum real rate above zero.
To summarise, history suggests that, with a 2% inflation target, the lower bound on interest rates is likely to bind in a large fraction of recessions.
Would 4% inflation hurt the economy? Economists have suggested various costs of inflation, such as variability in relative prices and distortions of the tax system. But research has not shown that these effects are quantitatively important for moderate inflation. As Krugman (1997) puts it: “one of the dirty little secrets of economic analysis is that even though inflation is universally regarded as a terrible scourge, efforts to measure its costs come up with embarrassingly small numbers”.
Some central bankers acknowledge that 4% inflation does not greatly harm the economy. Nonetheless, they oppose adoption of a 4% target because they think this action may actually cause inflation to rise above 4%, or at least create expectations of that outcome.
Bernanke (2010a), for example, asserts that “inflation would be higher and probably more volatile” with a 4% target and “inflation expectations would also likely become significantly less stable”. According to Bernanke: “The Fed, over a long period of time, has established a great deal of credibility in terms of keeping inflation low, around 2%… If we were to go to 4% and say we’re going to 4%, we would risk a lot of that hard-won credibility, because folks would say, well, if we go to 4%, why not go to 6%? It’d be very difficult to tie down expectations at 4%.” Mishkin (2011) makes a similar argument, asserting that “when inflation rises above the 3% level, it tends to keep on rising”.
We might call this view ‘the addictive theory of inflation’. Like an alcoholic’s first drink, 4% inflation may not do great harm by itself, but it is the first step in a dangerous process.
The rationale for this view is not clear. In other contexts, Bernanke and Mishkin argue that a central bank should determine its optimal policy, explain this policy to the public, and carry it out. Why can’t policymakers explain that the zero-bound problem makes 4% inflation desirable, raise inflation to 4%, and keep it there? Mishkin points to the 1960s, when inflation rose to 4% and the Fed let it keep rising, but why must policymakers repeat that mistake?
History does not suggest that it would be “difficult to tie down expectations” if inflation rises modestly. Inflation expectations, as measured by surveys, have generally followed actual inflation with a lag. They followed inflation up during the 1960s and 70s, and after that they followed inflation down. If inflation rises to 4%, it seems unlikely that expectations will overshoot this level.