Inflation - Not so great expectations

Dec 3rd 2012, 15:09 by M.C.K. | WASHINGTON 

THOSE who criticise central banks for having acted with insufficient vigour generally argue that they have failed to talk a good game. Paul Krugman and Michael Woodford are among the best-known advocates of this view. The underlying theory is that the expectation of faster consumer price inflation causes prices to rise more rapidly as people attempt to offset the anticipated erosion of their purchasing power by spending more on goods and services. According to this model, central banks theoretically have the power to lower real borrowing costs and real debt burdens (and real wages) even when nominal interest rates have hit 0%, simply by talking convincingly.

The existence of this Jedi mind trick might be desirable given the economy’s weakness. However, it is possible that simply asserting a tolerance for faster inflation over a given period of time or, more radically, altering the inflation target, might actually be counterproductive if the Fed does not—or cannot—generate sufficient increases in wages and prices. If people's expectations of inflation overshoot what they actually end up enduring, the consequence could be a nasty recession.

Until the rich world was hit by “stagflation” in the 1970s, there had been a remarkably stable relationship between the level of joblessness and the speed of price increases. While many economists were befuddled by the apparent breakdown of the Phillips curve, the “rational expectations” school explained that the problem could be attributed to the fact that people are not stupid. Sudden alterations to millions of contracts can have large effects on economic activity, so faster inflation can stimulate real activity when it is unexpected. This is because it lowers real debt burdens and increases the nominal value of real assets. There is also some evidence to suggest that, when unemployment is high, businesses will respond to rumours of accelerating inflation by trying to hire relatively cheap workers before their competitors. Similarly, there is evidence that firms operating at full capacity tend to fire workers if they expect prices to fall. But once people come to expect rapid price increases they try to protect themselves by demanding higher wages, higher interest rates, higher commodity prices, and higher risk premiums. Eventually, the unemployment rate will be what it will be, irrespective of the inflation rate.

Before “eventually” is reached, however, there can be painful periods when expectations clash with reality. In the late 1970s, many believed that inflation would continue to get faster and faster. The cycle was not broken until Paul Volcker’s Fed brutally crushed the economy, which forced a slowdown in inflation and also a decline in its volatility. The recession was so severe in part because many households and businesses had assumed that the prevailing rate of price increases would persist. They were stuck with debts they could not repay and employees they could not afford. 

People’s expectations did eventually change, which in turn contributed to a boom in asset prices and the end of “stagflation”. (Ironically, real business investment did not grow any faster in the 1980s than it did in the 1970s, despite the massive decline in the cost of capital. Instead, the cheaper funding was used for acquisitions, equity buybacks, and takeovers, as well as real estate speculation in certain areas.) This shift can be attributed mostly to people’s experience of actual inflation, which had to have been determined by something other than those very expectations. At least in the 1980s, the Jedi mind trick was insufficient. Brute force was required. Less than what had been predicted by the standard Phillips curve models of the time; the slowdown in inflation was large relative to the magnitude of the recession. The pupils did learn. But it took a nasty rap on the knuckles to get their attention.

Those arguing that higher expectations of faster inflation are desirable therefore have a large burden of proof. To repeat, the question is not about whether faster inflation itself is desirable, but whether or not there might be unintended consequences of a sudden change in people’s beliefs about future price increases. Suppose that starting tomorrow, the Federal Reserve says that it will be okay with price increases at an annualised pace of around 5% instead of around 2%, either permanently or just for a few years. What might happen?

It is safe to say that the financial markets would react first. (Most workers would find it very difficult to renegotiate their wage contracts as long as the jobless rate remains so high. If businesses went on a hiring spree, which would take time, the bargaining power of labour would improve.) Investors would probably sell bonds and stocks and use the proceeds to buy commodities and land—the most obviously profitable strategy when you expect inflation to accelerate. This would immediately raise costs for businesses and households. Moreover, investors might reasonably conclude that any central bank that can adjust its inflation target once, even "temporarily," will do so again in the future. Investors cannot really account for this risk, so they will demand a higher real discount rate on all financial assets in compensation for this uncertainty. Households that cannot easily adjust their portfolios will suffer a nasty hit to their wealth.

It is worth noting that these developments could encourage commodity producers to expand their production, which, over time, would lower prices. Also, the increase in real interest rates would probably reduce industrial demand for commodities. It is difficult to say how these effects would net out, especially if there were uncertainty on the central bank's ability or commitment to deliver faster inflation. Similarly, higher land prices could redound to higher home values, which might offset losses on financial assets. Then again, the price of farmland continues to soar even as home prices remain far below peak.

It is difficult to say what would happen to the exchange vaue of the dollar in these circumstances, although a decline relative to America's trading partners would be bullish for the economy. Where would the commodity producers park their earnings? They could conceivably buy up all the stocks and bonds sold to purchase the commodities in the first place, in which case only the ownership of the assets has changed. Foreigners might even find dollar assets relatively attractive if real yields rise sufficiently, or they could be the ones doing the selling, in which case the dollar would fall. How do expectations of faster inflation in America affect the performance of other economies? (Thinking through this exercise is a good example of why it is so hard to consistently make money in markets without insider trading.)

Households and businesses might save relatively less and spend relatively more, which would lead to price increases, as theory suggests. But they might not. It will depend in large part on how they react to the changes in nominal interest rates. Plus, it is not obvious how the difference between 2% and 5% inflation should affect spending and saving. It is one thing to conclude that faster inflation means “spend more now!”. Figuring out the actual quantities is far harder. Rational people might not bother, which could have all sorts of unpredictable effects. Politicians might freak out in response to the government's higher interest expenses. These uncertainties could make investors even more nervous and further increase risk premiums.

Putting all of this together, it is not clear that it would be beneficial for central banks to express tolerance for faster inflation unless they can generate price increases of the proper speed and magnitude. Otherwise they might end up undermining their efforts by raising the real cost of capital. While by no means certain, this seems like enough of a possibility that central bankers should be very cautious about playing too much with the "expectations channel". The point is not that faster inflation itself is necessarily bad. After all, if people take home more money they will be able to repay their debts and fulfill their nominal savings objectives, which ought to cause a rebound in economic activity (and in prices). But there are multiple routes to achieve this desired end result. Something more direct would probably be more effective and less risky.

Source