An Excessive Inflation Tax

Inflation, used as one tax among many, may be an efficient way of raising some of a government’s revenues. The effects of the inflation tax are like the effects of any tax: people respond by substituting away from the activity being taxed. When the government taxes cigarettes, people smoke less. When the government taxes income, people work less. When the government taxes the money people hold, people hold less money. These changes in behavior are the distortions caused by taxation. People substitute away from holding money in two ways: (1) during moderate inflations, people allocate more of their time to transactions; and (2) during high inflations, people may cease using money altogether.

During high inflations, the real value of money decreases quickly. So if you work and get paid in money, you had better go shopping quickly to make purchases. During hyperinflations, people may literally spend more time trying to get rid of their money than they do earning it in the first place. The same distortion applies, although less dramatically, in times of low to moderate inflation. People respond to inflation by carrying less cash, on average. To do so, they must spend more time standing in line in the bank and at automatic teller machines.

Imagine that ice cream were to be used as money. In a very cold climate, ice cream is just fine as a store of value. In a very hot climate, by contrast, ice cream is a bad store of value. You would probably want to get paid every day, and as soon as you received your ice cream, you would run to the store to buy other goods and services before your money melted. You and everyone else would spend much more time shopping and less time working. Melting ice cream, in this world, is like inflation.

There is a good reason why we do not use ice cream as a medium of exchange. Because it is such a bad store of value, people would quickly abandon it in terms of some other way of trading. During hyperinflations, this is exactly what we see: people substitute away from money completely and instead resort to barter trades. Often, some other commodity, such as cigarettes, starts being generally accepted as an alternative to money. But substitution away from money is costly to the economy. Money facilitates trade. It is generally easier to trade when everyone uses money rather than goods in exchange. When people respond to high inflation by eliminating money from trades, we are observing a distortion from the inflation tax.

Uncertainty and Real Interest Rates

It is the real interest rate that ultimately matters for saving and investment decisions. Yet loans are almost invariably quoted in nominal terms: a loan contract gives the borrower some money with a requirement to pay back that money plus interest in the future. The real and the nominal interest rates are linked by the Fisher equation:

real interest rate ≈ nominal interest rate − inflation rate.

To calculate the real interest rate you subtract the inflation rate from the nominal interest rate. So, for example, if the annual interest rate on a car loan is 12 percent and the current inflation rate is 4 percent, then the real interest rate on the car loan is 8 percent.

The Fisher equation glosses over an important point, however. Suppose you are thinking of taking out a loan this year, allowing you to borrow money now for repayment next year. The inflation rate that matters for this loan is the inflation between this year and next. At the time you sign the contract, you do not know what the inflation rate will be. You must make your decision about the loan without knowing for sure what the real interest rate will be. You have to make a guess:

expected real interest rate ≈ nominal interest rate − expected inflation rate.

Thus when a loan contract is signed, it is based on expectations of what will happen to prices in the future. If a borrower and lender would like to agree on a loan at a 4 percent real interest rate, but both expect 2 percent inflation, then they will agree on a 6 percent nominal interest rate.

What happens if the inflation rate turns out to be different from what the borrower and lender expected? Suppose the actual inflation rate turns out to be 4 percent. This means that the actual real interest rate, from the Fisher equation, is only 2 percent. This is good news for the borrower: he gets a loan at a lower rate than he expected. But it is bad news for the lender: she is repaid at a lower rate than she expected. The opposite is true if the inflation rate is lower than expected. Suppose the actual inflation rate is only 1 percent. Then the real interest rate is higher than anticipated—5 percent instead of 4 percent—which benefits the lender but is costly to the borrower.

Any divergence between actual and expected inflation therefore leads to a redistribution, either from the borrower to the lender or from the lender to the borrower. When inflation is higher than expected, the borrower is better off, and the lender is worse off. The opposite effects occur if inflation is lower than expected: the borrower loses, and the lender wins.

The possibility that the inflation rate will turn out to be unexpectedly high or unexpectedly low means that there is uncertainty whenever people sign loan contracts. A fixed nominal interest rate on a loan exposes both the borrower and the lender to the risk of inflation uncertainty. Uncertainty can prevent beneficial trades from taking place. Imagine that you were thinking of buying a used car, but you had to decide to buy without knowing whether the price was going to be $1,500 or $2,000. You might well decide not to buy in the face of this uncertainty. Similarly, people might sometimes decide not to sign loan contracts that would actually be beneficial to them.

The borrower and the lender could always change the form of their contract. Contracts do not have to specify nominal interest rates, and not all of them do. Some loans have interest rates that change with the actual inflation rate. In this way, borrowers and lenders can protect themselves from unexpected inflation. However, such contracts are unusual in practice and are most often seen in countries experiencing high and uncertain inflation. What should we conclude from the fact that loan contracts are rarely protected against inflation? Presumably one of two things is true: either such contracts are expensive to write or the benefit of these contracts is actually small.

Unexpected inflation can also have redistributive effects with other types of contracts. Labor contracts are an example. Although the worker and the firm ultimately care about real wages, most labor contracts are written in terms of nominal wages. That is, most labor arrangements are not indexed and thus leave the parties open to the effects of unanticipated inflation. So, for example, if inflation is higher than anticipated, then the real wage earned by the worker is lower than expected, which is a benefit to the firm.

Economies do respond to inflation, partly through the way in which people write contracts. In countries with high and volatile inflation, labor and other contracts generally provide some form of protection against inflation through indexation. For example, if you agree to a job that pays you $10 an hour this year, the nominal wage rate next year will change depending on inflation. If, for example, inflation was 20 percent this year, then under an indexed contract your nominal wage would automatically increase by 20 percent to $12. Under full indexation, the real wage you are paid is constant.