Our discussion of the causes and effects of inflation does not tell us much about the social problems that result from inflation. We turn to those problems now.

If you ask the average person why inflation is a social problem, he will probably answer that inflation makes him poorer. “Each year my boss gives me a raise, but prices go up and that takes some of my raise away from me.” The implicit assumption in this statement is that if there were no inflation, he would get the same raise and be able to buy more goods.

109

This complaint about inflation is a common fallacy. From Chapters 3, 7, and 8, we know that increases in the purchasing power of labour come from capital accumulation and technological progress. In particular, the real wage does not depend on how much money the government chooses to print. If the central bank reduces inflation by slowing the rate of money growth, workers will not see their real wage increasing more rapidly. Instead, when inflation slows, firms will increase the prices of their products less each year and, as a result, will give their workers smaller raises.

According to the classical theory of money, a change in the overall price level is like a change in the units of measurement. It is as if we switched from measuring distances in metres to measuring them in centimetres: numbers get larger, but nothing really changes. Imagine that tomorrow morning you wake up and find that, for some reason, all dollar figures in the economy have been multiplied by ten. The price of everything you buy has increased tenfold, but so has your wage and the value of your savings. What difference would such a price increase make in your life? All numbers would have an extra zero at the end, but nothing else would change. Your economic well-being depends on relative prices, not the overall price level.

Why, then, is a persistent increase in the price level a social problem? It turns out that the costs of inflation are subtle. Indeed, economists disagree about the size of the social costs. To the surprise of many laymen, some economists argue that the costs of inflation are small—at least for the moderate rates of inflation that most countries have experienced in recent years.⁷

Consider first the case of expected inflation. Suppose that every month the price level rose by 1 percent. What would be the social costs of such a steady and predictable 12 percent annual inflation?

One cost is the distortion of the inflation tax on the amount of money people hold. As we have already discussed, a higher inflation rate leads to a higher nominal interest rate, which in turn leads to lower real money balances. If people are to hold lower money balances on average, they must make more frequent trips to the bank to withdraw money—for example, they might withdraw $50 twice a week rather than $100 once a week. The inconvenience of reducing money holding is metaphorically called the shoeleather cost of inflation, because walking to the bank more often causes one’s shoes to wear out more quickly.

A second cost of inflation arises because high inflation induces firms to change their posted prices more often. Changing prices is sometimes costly: for example, it may require printing and distributing a new catalogue. These costs are called menu costs, because the higher the rate of inflation, the more often restaurants have to print new menus.

A third cost of inflation arises because firms facing menu costs change prices infrequently; therefore, the higher the rate of inflation, the greater the variability in relative prices. For example, suppose a firm issues a new catalogue every January. If there is no inflation, then the firm’s prices relative to the overall price level are constant over the year. Yet if inflation is 1 percent per month, then from the beginning to the end of the year the firm’s relative prices fall by 12 percent. Sales from this catalogue will tend to be low early in the year (when its prices are relatively high) and high later in the year (when its prices are relatively low). Hence, when inflation induces variability in relative prices, it leads to microeconomic inefficiencies in the allocation of resources.

111

A fourth cost of inflation results from the tax laws. Many provisions of the tax system do not take into account the effects of inflation. Inflation can alter an individual’s tax liability, often in ways that lawmakers did not intend.

One example of the failure of the tax system to deal with inflation is the tax treatment of capital gains. Suppose you buy some stock today and sell it a year from now at the same real price. It would seem reasonable for the government not to levy a tax, since you have earned no real income from this investment. Indeed, if there is no inflation, a zero tax liability would be the outcome. But suppose the rate of inflation is 12 percent and you initially paid $100 per share for the stock; for the real price to be the same a year later, you must sell the stock for $112 per share. In this case the personal income tax system, which ignores the effects of inflation, says that you have earned $12 per share in income, and the government taxes you on this capital gain. The problem, of course, is that the tax system measures income as the nominal rather than the real capital gain. In this example, and in many others, inflation distorts how taxes are levied.

A similar problem occurs with interest-income taxes. Because the Canadian tax system was designed for a zero-inflation environment, it does not work well when inflation occurs. It turns out that when inflation and the tax system interact, the result is a powerful disincentive to save and invest. And worse still, this problem occurs even for mild inflations and even when all individuals anticipate inflation perfectly. Let us see how this problem develops, by adding interest-income taxes to our discussion of the Fisher equation.

If t stands for the tax rate that an individual must pay on her interest income, the after-tax real yield is

After–tax r = i(1 – t) –π.

Consider what happens to this effective yield on savings if inflation rises from 0 percent to 10 percent. Assume that the nominal interest rate rises by the same 10 percentage points—enough to compensate lenders fully for the inflation, if it were not for the tax system. The revised Fisher equation makes clear that the after-tax real yield must fall by t times 10 percent in this case. If the individual’s marginal tax rate is 50 percent, the reduction in the real return to saving is a full tπ = 5 percentage points. Most economists believe that this represents a significant disincentive to save, and so inflation reduces capital accumulation and future living standards.

The problem is that the tax system taxes nominal interest income instead of real interest income. During an inflationary time, much of an individual’s nominal interest receipts are just a compensation for the fact that the loan’s principal value is shrinking. Since that “inflation premium” part of interest is not income at all, it should not be taxed. If the tax system were fully indexed, it would not be taxed. In that case, the after-tax real yield would be given by

After–tax r = (i –π) (1 – t).

112

In this case, as long as the nominal interest rate rises one-for-one with inflation (as we discussed in the nonindexed tax case), the after-tax real yield is not reduced at all by inflation. The disincentive for saving and investment is removed.

Nevertheless, because our tax system does not limit taxes on interest earnings to just real returns, one of the central costs of inflation is that it lowers the economy’s accumulation of capital, and so it reduces the standard of living for all members of future generations.

A fifth cost of inflation is the inconvenience of living in a world with a changing price level. Money is the ruler with which we measure economic transactions. When there is inflation, that ruler is changing in length. To continue the analogy, suppose that Parliament passed a law specifying that a metre would equal 100 centimetres in 2010, 95 centimetres in 2011, 90 centimetres in 2012, and so on. Although the law would result in no ambiguity, it would be highly inconvenient. When someone measured a distance in metres, it would be necessary to specify whether the measurement was in 2010 metres or 2011 metres; to compare distances measured in different years, one would need to make an “inflation” correction. Similarly, the changing value of the dollar requires that we correct for inflation when comparing dollar figures from different times.

For example, a changing price level complicates personal financial planning. One important decision that all households face is how much of their income to consume today and how much to save for retirement. A dollar saved today and invested at a fixed nominal interest rate will yield a fixed dollar amount in the future. Yet the real value of that dollar amount—which will determine the retiree’s living standard—depends on the future price level. Deciding how much to save would be much simpler if people could count on the price level in 30 years being similar to its level today.

Unexpected inflation has an effect that is more pernicious than any of the costs of steady, anticipated inflation: it arbitrarily redistributes wealth among individuals. You can see how this works by examining long-term loans. Loan agreements typically specify a nominal interest rate, which is based on the rate of inflation expected at the time of the agreement. If inflation turns out differently from what was expected, the ex post real return that the debtor pays to the creditor differs from what both parties anticipated. On the one hand, if inflation turns out to be higher than expected, the debtor wins and the creditor loses because the debtor repays the loan with less valuable dollars. On the other hand, if inflation turns out to be lower than expected, the creditor wins and the debtor loses because the repayment is worth more than the two parties anticipated.

Consider, for example, a person taking out a mortgage in 1960. At the time, a 30-year mortgage had an interest rate of about 6 percent per year. This rate was based on a low rate of expected inflation—inflation over the previous decade had averaged only 2.5 percent. The creditor probably expected to receive a real return of about 3.5 percent, and the debtor expected to pay this real return. In fact, over the life of the mortgage, the inflation rate averaged 5.5 percent, so the ex post real return was only 0.5 percent. This unanticipated inflation benefited the debtor at the expense of the creditor.

113

Unanticipated inflation also hurts individuals on fixed pensions. Workers and firms often agree on a fixed nominal pension when the worker retires (or even earlier). Since the pension is deferred earnings, the worker is essentially providing the firm a loan: the worker provides labour services to the firm while young but does not get fully paid until old age. Like any creditor, the worker is hurt when inflation is higher than anticipated. Like any debtor, the firm is hurt when inflation is lower than anticipated. The magnitudes involved in these arbitrary redistributions of income can be very large. For example, with inflation of just 5 percent per year, the purchasing power of money halves in value in fewer than 15 years. So in a period when inflation is underpredicted by 5 percent, pensioners’ real income is cut in half well before they are expected to die. This reduction in real income can be devastating.

These situations provide a clear argument against highly variable inflation. The more variable the rate of inflation, the greater the uncertainty that both debtors and creditors face. Since most people are risk averse—they dislike uncertainty—the unpredictability caused by highly variable inflation hurts almost everyone.

Given these effects of uncertain inflation, it is puzzling that nominal contracts are so prevalent. One might expect debtors and creditors to protect themselves from this uncertainty by writing contracts in real terms—that is, by indexing to some measure of the price level. In economies with extremely high and variable inflation, indexation is often widespread; sometimes this indexation takes the form of writing contracts using a more stable foreign currency. In economies with moderate inflation, such as Canada, indexation is less common. Yet even in Canada, some long-term obligations are indexed. For example, Canada Pension benefits for the elderly are adjusted annually in response to changes in the consumer price index, as is the basic personal exemption in the income tax system.

Finally, in thinking about the costs of inflation, it is important to note a widely documented but little understood fact: high inflation is variable inflation. That is, countries with high average inflation also tend to have inflation rates that change greatly from year to year. The implication is that if a country decides to pursue a high-inflation monetary policy, it will likely have to accept highly variable inflation as well. As we have just discussed, highly variable inflation increases uncertainty for both creditors and debtors by subjecting them to arbitrary and potentially large redistributions of wealth.

So far we have discussed the many costs of inflation. These costs lead many economists to conclude that monetary policymakers should aim for zero inflation. Yet there is another side to the story. Some economists believe that a little bit of inflation—say, 2 percent per year—can be a good thing.

The argument for moderate inflation starts with the observation that cuts in nominal wages are rare: firms are reluctant to cut their workers’ nominal wages, and workers are reluctant to accept such cuts. A 2 percent wage cut in a zero inflation world is, in real terms, the same as a 3 percent raise with 5 percent inflation, but workers do not always see it that way. The 2 percent wage cut may seem like an insult, whereas the 3 percent raise is, after all, still a raise. Empirical studies confirm that nominal wages rarely fall.

This finding suggests that some inflation may make labour markets work better. The supply and demand for different kinds of labour are always changing. Sometimes an increase in supply or decrease in demand leads to a fall in the equilibrium real wage for a group of workers. If nominal wages can’t be cut, then the only way to cut real wages is to allow inflation to do the job. Without inflation, the real wage will be stuck above the equilibrium level, resulting in higher unemployment.

115

For this reason, some economists argue that inflation “greases the wheels” of labour markets. Only a little inflation is needed: an inflation rate of 2 percent lets real wages fall by 2 percent per year, or 20 percent per decade, without cuts in nominal wages. Such automatic reductions in real wages are impossible with zero inflation.¹⁰