Warning: Inflation and Deflation Are Toxic to Your Economic Health

WHEN YUGOSLAVIA dissolved into a bloody civil war in the 1990s, there were more than just ethnic and religious rivalries at work. Inflation, the continuous rise in the prices of almost everything, was also a factor. Thanks to an economic crisis in the early 1980s, prices in Yugoslavia were rising at annual rates of more than 1,200 percent by the late 1980s. Inflation helped dissolve the cohesion of Yugoslavia’s multi-ethnic middle class. Some people protected themselves by growing their own food or hoarding foreign currency. Others watched their incomes and savings get wiped out.

Throughout history, high inflation has often led to social upheaval. Hyperinflation, when prices rise by 50 percent per month, helped bring the Nazis to power in Germany and the communists in Russia and China, and topple both civilian and military governments in Argentina. These, of course, are extreme forms of the disease. But far more modest rates of inflation in the 1970s helped drive Labor from power in Britain and Jimmy Carter from the White House.

Why is inflation so destabilizing?

Prices are the market’s air supply; they signal surpluses and shortages and tell businesses and consumers when to produce more or consume less. Inflation contaminates this air supply. Suppose you are thinking of opening a new hotel in a city where rates are rising 10 percent, thinking that must be a sign of strong demand. But what if the cost of land is going up 12 percent, linen by 11 percent, chambermaid and doormen wages by 13 percent? The new hotel may actually lose money. Inflation makes it hard to interpret price signals.

Inflation also unsettles people because it arbitrarily punishes some people while it rewards others. A retiree who buys a government bond that pays 4 percent interest, only to see inflation jump to 5 percent, sees his purchasing power get clobbered. A home buyer lucky enough to lock in a mortgage at 5 percent and then sees home prices soar 50 percent scores a windfall.

Inflation is also a hidden tax. As wages rise to compensate for it, so does tax revenue, making it easier for the government to repay what it borrowed before inflation took off. In the process, however, it robs the currency in people’s wallets of purchasing power.

Another reason inflation is unsettling is that getting it back down is painful. Governments may resort to wage and price controls or other heavy-handed interventions to reduce inflation. Usually, though, it takes a recession to cure inflation—and that hurts everyone.

Goldman Sachs economists have shown that investors do best under low inflation (see Table 5.1). Under high inflation, only stocks gain, and not by much. Under hyperinflation, everything goes down.

Table 5.1 Investors Hate Inflation and Deflation

Source: Goldman Sachs Global Economics Paper #190, September, 2009.

Inflation’s dangers should not be overstated. It is hard to find evidence that steady inflation below 5 percent does much economic harm. The trouble is that as inflation rises it gets less predictable. This year 5 percent, next year who knows?

From Cigarettes to Aztecs

There are two competing schools of thought on the cause of inflation, and listening to their proponents is like listening to creationists and Darwinians argue over how life began. Monetarists blame the money supply, while neo-Keynesians blame a combination of excess spending and inflationary psychology. There’s truth to both schools.

Blame It on the Money Supply

Milton Friedman, the Nobel economist, said “Inflation is always and everywhere a monetary phenomenon.”

There are two competing schools of thought on the causes inflation. Listening to them is like listening to creationists and Darwinians argue over how life began.

Monetarism, as this brand of economics is called, blames inflation on too much money chasing too few goods. Intuitively, this makes sense. If you double the amount of money people spend on stuff, but leave unchanged the amount of stuff, prices will double.

In its most basic form, this notion is uncontroversial, and economists of all stripes accept it. Let’s examine one example. In German prisoner-of-war camps, prisoners used cigarettes as currency, pricing bread, shirts, and chocolates in cigarettes. When Red Cross shipments arrived, suddenly everyone had more cigarettes to spend—and the prices of everything went up. As the cigarettes wore out or were smoked, prices started dropping again. Altering the supply of money has the same effect. A government normally finances its spending with taxes or by selling bonds to the public. Suppose, instead, it sells the bonds to the central bank, which pays for them by creating money out of thin air. This can produce hyperinflation, when prices rise 50 percent or more per month. In 2008 in Zimbabwe, prices were doubling roughly every day. Steven Hanke, a Johns Hopkins University economist, figures annual inflation equaled 89.7 sextillion percent (that’s 897 followed by 20 zeros). During such hyperinflations, people try to hold as little currency as possible. As soon as they’re paid, they spend the money before its value is wiped out. In many cases, people switch to foreign currency instead.

At the opposite extreme, fixing the money supply eradicates persistent inflation. That’s what happens when a country goes on the gold standard, which means that the currency is backed by gold. You could take your notes to a bank and receive their face value in gold. Prices could rise, but eventually, they’d fall again. When the United States was on the gold standard between 1790 and 1911, periods of inflation and deflation alternated; wholesale prices ended the period roughly where they began.

Under some conditions, though, the money supply can rise even with a gold standard in place. How? The supply of gold may increase. For example, when Spanish conquerors brought troves of Aztec and Inca treasures back to Europe, a century of mild inflation ensued. Or, another way this can happen is if the government allows the same amount of gold to back a larger amount of currency. Roman emperors and medieval kings routinely debased their coins—that is, they reduced their gold or silver content—to finance their wars. In 1933-1934, Franklin D. Roosevelt allowed gold to rise from $20.67 to $35 per troy ounce, a 41 percent devaluation, in a successful effort to end deflation.

Thus far, the link between the money supply and inflation is straightforward. It’s when you get to the case of a modern economy that the monetarist theory, while good in theory, proves almost useless in practice. Let’s examine why.

The central bank doesn’t control the entire money supply, only a narrow portion of it: specifically, the notes, coins, and reserves it supplies to commercial banks. (Reserves are cash that banks keep on deposit at the Fed to settle payments with each other, with the Treasury, or to exchange for currency to refill their ATMs.)

To understand why the link between money supply and inflation is muddy, imagine that the Fed distributes $1 trillion in freshly printed twenty-dollar bills to people on street corners. If they promptly rush home and stuff the money under their mattresses, what will happen to consumer spending and inflation? Zilch. For money to cause inflation, it must be lent and spent. Banks lend when they have healthy balance sheets and a lot of eager, creditworthy customers. Consumers spend when they feel confident about their jobs and salaries; they go to the ATM more often, run up bigger credit cards bills, remodel their homes, and buy faster cars.

Monetarists claim that growth in the money supply leads to more spending and more inflation. Actually, it’s the other way around. Every dollar consumers borrow or spend returns to the banking system and shows up in someone else’s checking or savings deposit or money market mutual fund, which are all part of the broader money supply (which has labels like M1, M2, and M3).

For this reason, the Fed doesn’t target the money supply. It uses its control of reserves only to ensure banks have enough cash to keep their ATMs full, and to control short-term interest rates. (I’ll explain how it does this in Chapter Ten.) Therefore, its influence over the broad money supply is indirect. If it raises interest rates, it will dampen spending and, eventually, the money supply. If, however, the economy is truly moribund, because no one wants to lend or borrow, the Fed can drive interest rates to zero and print gobs of money without causing broader measures of money and credit to grow. That’s what happened in 2009: The Fed had lowered rates to almost zero and increased reserves to banks by $1 trillion, yet total bank lending contracted.

The Other Side of the Story: Mind the Gap and Your Mind

So save some trouble and don’t preoccupy yourself with the money supply. For a more realistic picture of inflation—the neo-Keynesian picture—think instead of hotels in Scottsdale, Arizona. The supply of rooms is roughly the same all year, but there’s a lot more demand in January when the temperature averages 70 degrees than in July when it averages 100 degrees. The demand for rooms is higher in January, and so, not surprisingly, the hotel can charge a lot more than it would in July.

The same is true for the economy as a whole: When demand for all goods and services runs ahead of the supply (i.e., potential output), inflation rises. When demand falls short of potential, inflation falls. When unemployment is below its natural rate, workers are better able to win raises. This relationship was captured by Alban William Phillips, a New Zealand economist. The Phillips Curve, which shows an inverse relationship between unemployment and inflation, is at the heart of the neo-Keynesian inflation model.

A hotel whose occupancy suddenly rises to 95 percent from 80 percent will eventually add rooms, but will first simply charge more. Similarly, if occupancy falls, the hotel may eventually close. But first, it will simply charge less. The difference between actual gross domestic product (GDP) and potential GDP is the output gap, which you could think of as a vacancy rate for the entire economy. Inflation always falls after recessions because the output gap is so large: hotels and office buildings are empty, factories are idle, and the unemployed are everywhere.

The surest sign the economy has exceeded its capacity is if firms are paying higher wages to attract qualified workers. Inflation needs a wage- price spiral; if wages don’t rise, there’s no spiral.

Like the natural rate of unemployment, potential output is a slippery thing to measure. It’s easy to tell if a hotel, factory, or power plant is at full capacity. But what about a law firm or an Internet dating service? Potential also changes. In the early 1970s, high oil prices rendered a lot of existing factories obsolete; this reduced potential. In the late 1990s, companies found they could use the computers and the Internet to boost production with fewer workers. For example, airlines replaced reservation agents with web sites. This boosted potential.

Globalization has also relaxed the constraints on capacity. A company that interprets X-rays can’t charge as much if a competitor offers to do it using Indian radiologists at a fraction of what U.S. radiologists cost.

It is difficult to know when the economy has exceeded its capacity, but there are telltale signs. The surest sign is if firms are paying higher wages to attract qualified workers. Inflation needs a wage-price spiral; if wages don’t rise, there’s no spiral.

An economy with a large output gap can grow rapidly with little threat of inflation, just as a near-empty hotel that manages to boost occupancy to 50 percent is still in no position to raise room rates. But once the output gap has been closed, the economy can only grow in line with the labor force and productivity. For the United States, that means a growth rate between 2.25 percent and 2.75 percent.

Strange as it may seem, inflation also depends a lot on what people think it will be. Suppose an employer and its union sit down to hammer out a new contract. If both parties agree inflation will be 2 percent, they will quickly agree to a 2 percent cost-of-living increase and the firm will plan on setting prices to cover those costs. If every firm in the country and its employees do the same thing, inflation will settle at 2 percent. Thus, expectations of inflation can be self-fulfilling.

Rapidly shifting expectations lead to quicker changes in inflation. If a jump in oil prices suddenly boosts the cost of living, firms and workers quickly raise wages and prices to compensate, and a wage-price spiral ensues. This means that the tradeoff between inflation and unemployment in the Phillips Curve is only temporary. Pushing the economy past its potential can drive down unemployment for a little while, but as inflation picks up, so will workers’ wage demands, and unemployment will return to where it was.

On the other hand, if people have gotten used to 2 percent inflation year in and year out, they might endure a jump in oil prices without expecting wages to automatically follow. Well-anchored expectations can keep inflation steady even when the economy is above or below potential.

Even Worse than Inflation

Inflation is a familiar scourge. Deflation, when prices are falling, is rarer and, potentially, nastier. This may seem odd. Shouldn’t we be happy if the prices we pay go down year after year? Well, it’s sort of like weight loss. What’s the reason for it: you’re eating better and exercising more (good), or starving to death (bad)?

Good deflation occurs when workers and companies become more productive and learn to make things at a lower cost. Intel, for example, is continuously cutting the price of computer chips because it keeps finding new, cheaper ways to make them. Intel’s profits and its employees’ salaries still go up. If you multiply that across the entire economy, it’s possible for prices to fall across the board even as incomes rise.

Bad deflation occurs when spending collapses and companies have to cut their prices to prop up sales, just as hotels cut their rates when tourist traffic dries up. Once people expect falling prices, they hold off on purchases. Workers initially resist pay cuts, so employers must lay some off to cope with falling prices. Eventually, fear of unemployment persuades workers to accept lower pay. Prices and wages follow each other down, the mirror image of an inflationary wage-price spiral. We saw this happen between 1929 and 1933 in the United States when prices fell 7 percent per year. Japan has endured a milder form of this bad deflation since the late 1990s.

If prices and wages are falling at the same rate, is anyone the worse for it? After all, paychecks are smaller, but purchasing power is the same because prices have fallen. The problem is that debt is fixed so as incomes and prices fall, the burden of debts rises. Homeowners slash spending to keep up with their mortgage payments. Or worse, the homeowner goes into foreclosure because the home may not be worth enough to repay the loan. The bank fails, deepening the economic stress. “The more the economic boat tips, the more it tends to tip,” wrote Irving Fisher, the American economist who labeled this phenomenon debt-deflation in 1933.

Deflation can be harder to cure than inflation. Faced with inflation, a central bank that wants to can generally raise interest rates as high as needed. Faced with recession, it can stimulate spending and restore growth by lowering its interest rate below the inflation rate, making the real cost of borrowing negative. Clearly, that’s impossible when inflation itself is negative, since the central bank can’t lower interest rates below zero: During deflation the real interest rate will always be positive. (In Chapter Ten, I’ll describe other tools the central bank can use if it’s already lowered short-term rates to zero.)

The People’s Choice

In the wake of the Great Recession, a weird schizophrenia has overcome the economics fraternity. It was nicely encapsulated in a country-and-western You Tube ditty¹ by Merle Hazard, the pseudonym of Jon Shayne, a money manager:

Inflation or deflation, tell me if you can:
will we become Zimbabwe or will we be Japan?

In the long run, inflation is a political choice.

So, which will it be? Probably, neither. Yet there are risks on both sides. The Great Recession left so much unused economic capacity that inflation, already low, could cross the line into deflation.

In the long run, though, inflation is a political choice. When society won’t pay the taxes necessary to meet its own demands to create jobs, provide social programs, or fight wars, government has to borrow and may pressure the central bank to keep interest rates low to help with that borrowing. That would eventually lead to inflation. In the extreme, the government may simply order the central bank to print the money, which can lead to hyperinflation.

It sounds tempting, but don’t assume politicians will succumb. Voters hate inflation. In the 1970s, people consistently rated inflation a bigger concern than unemployment in Gallup polls. In a 1996 study, Robert Shiller, a Yale University economist, found that if forced to choose, Americans, Germans, and Brazilians all preferred higher unemployment to higher inflation. Thus, if inflation rises, politicians will eventually be forced to tame it or find a central banker who will. Jimmy Carter and Ronald Reagan, for instance, stood by as the Fed induced two savage recessions to break the back of inflation.

Into the Weeds

When the Bureau of Labor Statistics (BLS) was created in the late nineteenth century, the cost of living was one of the first things it tried to measure. Today, the consumer price index (CPI) is the economic statistic that most affects Americans’ daily lives since it is used to calculate cost-of-living adjustments. Once a month, BLS statisticians and contractors fan out across the country, and visit thousands of businesses to collect prices on over 80,000 items in 200 categories from new cars to funerals. It uses regular surveys of consumers’ spending habits to assign a weight to each category in the index—32 percent for shelter, 0.3 percent for sugar and candy. The 12-month percentage change in the CPI is the most common measure of inflation.

Fresh food and energy account for many of the monthly swings in the CPI. Because an increase one month is often undone a few months later, economists regularly exclude food and energy. The remainder, or core inflation, provides a more stable picture of underlying inflation. This picture will be misleading, though, if energy and food costs march steadily higher (or lower) over time instead of reverting to their old levels.

The CPI isn’t flawless. Consumers are constantly shifting to stores that have cheaper prices—to Wal-Mart, for example, from pricey department stores, and to cheaper products, such as Internet telephone calls instead of landline calls. The CPI tries to capture these changes by surveying consumers’ spending habits every two years, but in between, it may slightly overstate inflation.

The CPI’s measure of home ownership is also controversial. It’s not a measure of home prices. Rather, it’s a measure of what a homeowner would pay to rent the same house. The two prices usually move together, but not always. Between 1998 and 2007, home prices rose 84 percent but because rents were much less buoyant, the CPI recorded only a 38 percent increase in the cost of owning a home.

There are other inflation measures, including:

• PCE Index. An important but little-known alternative to the CPI is the price index of personal consumption expenditures, or PCE index, which the Bureau of Economic Analysis uses to calculate GDP. The Federal Reserve’s forecasts are based on the PCE index rather than the CPI. The PCE is based on what businesses actually sell rather than what consumers say they buy (which may be flawed). As a result, it assigns less importance to housing than the CPI, and more to medical care. The PCE index has quirks, too—it puts a price on things that have no price, like Sunday mass and no-fee checking accounts.

• GDP Deflator. The GDP deflator measures prices paid by all sectors of the economy: businesses, government, foreign buyers of exports, as well as consumers. It’s used to calculate how much of an increase in nominal GDP is due to inflation and how much is real activity.

• Producer Price Index. The producer price index (PPI) measures the prices that sellers receive, rather than what consumers pay. While it tracks prices for some services, like transportation and health, its main attraction is its index of finished goods, which excludes services and intermediate goods like the rubber and steel that eventually go into cars. Because it excludes services, the PPI is a much narrower measure of inflation than the CPI, and it is much more volatile.

• Import Price Index. The import price index tracks what we spend on imported goods, and thus signals inflationary or deflationary pressure from abroad or from the dollar’s exchange value.

Gold and commodity prices are much better measures of the fear of inflation than as predictors of inflation.

• Inflation Expectations. These expectations can be monitored through surveys. Each month, the Thomson Reuters/University of Michigan Surveys of Consumers asks consumers what they expect inflation to be over the next year, and the next 5 to 10 years. Treasury inflation-protected securities (TIPS) provide a minute-by-minute measure of investors’ inflation expectations. If a TIPS bond yields 3 percent and a regular bond yields 5 percent, the difference, 2 percent, is the expected inflation rate. Be cautious with this because numerous technical factors push these yields around.

• Gold and Commodity Prices. Many investors look to gold and commodity prices for early warning signs of inflation and deflation. These prices are much better measures of the fear of inflation than as predictors of inflation. That’s partly because so many other things affect them. Gold responds to global unrest, the demand for jewelry, and the dollar. Commodity prices respond to the strength of the global economy, strikes, and bad weather.

• Wages and Labor Costs. Hourly and weekly wages can be tracked each month in the BLS’s payroll survey that I discussed in Chapter Four. The quarterly employment cost index is more comprehensive because it also includes benefits and bonuses. Benefits for health care, pensions, and payroll taxes are now almost 20 percent of compensation, up from 5 percent in the 1940s. Still, to determine if rising wages are inflationary, you have to compare them to productivity. If a painter’s salary doubles because he can now paint twice as much with a paint sprayer, his salary per square foot has not risen at all. Labor costs, adjusted for productivity, are measured through unit labor costs, which the BLS reports quarterly along with productivity.

The Bottom Line

• High inflation is destabilizing and corrosive; deflation can be destructive. The best inflation is not too high nor too low: from 1 percent to 3 percent seems about right.

• The money supply is a lousy guide to where inflation is going. Better, instead, to monitor how far the economy is operating from its capacity. For example, if unemployment is 5 percent, it doesn’t have much spare capacity left. Wages are the best evidence of an economy running out of capacity. If wages aren’t rising, a wage-price spiral can’t happen.

• Inflation is more likely to rise if people expect it to, because they’ll adjust their wage and price behavior accordingly. Stable inflation expectations are a bulwark against both inflation and deflation.