Surprise!

Why don’t central banks level with us?

Inflation, once the scourge of the West, has become a fabled gift. In the 1970s and early 1980s, rapid rises in average prices – partly due to oil shortages -- troubled consumers in Europe and the United States. But today, central banks yearn for a bit of inflation that would make their job easier. The global economy has not fully recovered from the financial crunch of 2008, and the central banks want wriggle room in order to prod demand by cutting interest rates, since this would encourage people to borrow and spend. To be pared, the market rates must be above zero. A modicum of inflation – say, 2% per year -- will do the trick: Lenders will raise their interest rates so that they can cover the higher prices when they are finally paid back.

That’s the situation in the West. But in the Third World, inflation is a little too healthy. In Kazakhstan, it has exceeded 7% since August 2015, when the National Bank began targeting the interest rate on overnight loans between commercial banks (the “base rate”) in order to manage the tenge supply. Our average rate of consumer inflation for that period is above 14%. At the moment, inflation rates elsewhere in Central Asia range from 4.1% in Kyrgyzstan to 9% in Tajikistan.

Inflation this brisk confuses us. It hits different markets at different times, so we aren’t sure how to interpret it. Has the price of a cup of coffee risen because people demand more coffee, or because it is a harbinger of inflation that will eventually infect the whole economy? If the cafés take the first view, they will extend their hours. So will other industries as they respond to surges in their own output prices. But eventually, people will figure out that prices are rising because of an inflated money supply, not because of new demand for particular products. The cafés will cut their hours and lay off waiters, and recession will ensue.

Central banks in poorer countries can avoid such confoundment by telling us the rate of inflation. The National Bank of Kazakhstan posts monthly rates on its Web page, but not all central banks in Central Asia are so diligent.  Like the Kazakhstani bank, the National Bank of the Kyrgyz Republic posts the actual and the target rate of inflation; and the National Bank of Tajikistan reports the actual rate of inflation but not a target rate.  But the Central Bank of the Republic of Uzbekistan does not post inflation rates in English or Russian in an easily accessible site.  The worst performer in the region (you won’t be astonished to hear this) is the Central Bank of Turkmenistan, which does not publish inflation rates in English or Russian – or, for that matter, statistics in general – in a site that is easy to reach.  So in Ashgabat, the average price that people expect may be miles from the price that they suddenly observe.

Monetary hijinks

This may tempt central banks into tomfoolery. When the price surprise is large, people may interpret it as a big rise in demand for their output, so they will produce a lot more. To spur the economy in this way, the central bank may secretly print a lot more money. This unexpectedly raises average prices, since more som than before chase every pizza and SUV. However, if the bank tries this chicanery too often, people will catch on and ignore the price surprises.

On the other hand, a more honest central bank will win the public’s trust. So when it does try the occasional price surprise, people will interpret it as mainly a rise in demand for their own work, and they will expand accordingly.

This suggests a negative relationship across countries between variation in output and variation in average prices. An honest central bank (think New Zealand) will keep the expected average price close to the actual average price, so the variation -- the price surprise -- is small. When the bank does try a surprise, people will respond, so the variation in output will be large. Now consider a dishonest central bank. It is always trying surprises, so its variation in price is large. Burned twice, people will usually ignore it (think Zimbabwe), so the resulting variation in output is small.

The Nobel laureate Robert Lucas found evidence of this negative relationship in a statistical study in 1973. Across countries, fluctuations in output were small when demand shocks were large, which would include shocks due to fluctuations in the money supply. In 1988, Laurence Ball, Gregory Mankiw and David Romer extended the result to a 43-country dataset.

That sounds like solid evidence for the surprise model, but a few holes exist. For example, announced changes in central bank policy affect the economy in the real world, notes Romer.  The simplest Lucas model says that only price surprises should matter.    

In short, price surprises may influence output, but not too often. Not even a central bank can fool most of the people most of the time, so it might as well publish what it knows. –Leon Taylor tayloralmaty@gmail.com             

Notes

A simple form of the Lucas supply function is Q = a [P – E(P)], where Q is the log of output, P is the price level, E(P) is the public expectation of the price level, and a is a parameter.  Thus Var(Q) = a^2 Var[P – E(P)]: Given a, the variance of output relates positively to the variance of the unexpected price change, P – E(P).

But a, the response parameter, may vary from one country to another, because it depends on economic history. In a country with large and frequent price surprises, the public may eventually learn to ignore them; a there is small.  But in a country where price surprises are rarely large, the public may interpret a big one as mainly due to changes in demand and so would adjust output; a there is large.

In sum, across countries, frequent and large price surprises lead to small changes in output, and infrequent price surprises lead to large changes in output.      

References

Laurence Ball, N. Gregory Mankiw and David Romer. The New Keynesian economics and the output-inflation tradeoff. Brookings Papers on Economic Activity 1: 1-65. 1988.

Robert Lucas. Some international evidence on output-inflation tradeoffs. American Economic Review 63: 326-334. 1973.

David Romer. Advanced macroeconomics. Third edition. Boston: McGraw-Hill Irwin. 2006.