Ranjit Dighe
March 6-10, 2000

Mon., March 6, 2000

* Today:
I. The unemployment-inflation tradeoff: the Phillips Curve
II. Inflation (begin)


We left off with unemployment, and a parting question: How could a recession help to reduce inflation? Before we deal with that question, which relates to the possible benefits that a recession might bring (despite the high unemployment that a recession also brings), lets look at the social consequences of unemployment and recessions:
-- Unemployment is probably the biggest problem that capitalism generates. Even in this long-term economic expansion, there are still six million unemployed people, and the average spell of unemployment last three to four months. Problems associated with unemployment:
---- Unemployed people and their families obviously suffer a big drop in their standard of living, since unemployment benefits aren't that generous and run out after a couple months. (A lot of people don't even apply for them.)
---- Unemployed people tend to become very depressed and frustrated. In recessions, when unemployment rises, a host of other social ills spread as well: rates of crime, divorce, suicide, alcoholism, drug abuse, spouse and child abuse all tend to rise along with unemployment.
------ Welfare dependency rises in times of recession, as many unemployed people drop out of the labor force altogether and many families break up. In the last recession (1990-92), which was relatively mild, the number of welfare recipients rose by almost three million. (That number fell by almost two million as the economy recovered from 1992-96.)

So recessions are bad. Nobody would disagree with that statement. And yet, most economists would tell you that the last two recessions in the U.S. (1980-82, 1990-92) were more or less planned: the Federal Reserve raised interest rates and tightened credit in order to deliberately slow down the economy. The Fed raised interest rates in both of those cases in order to reduce inflation. Which brings us back to our parting question:

Q: How could a recession help to reduce inflation?

A: In a recession, overall demand is low and jobs are scarce, therefore:

(1) companies will raise their prices more slowly (or perhaps not at all);
(2) workers will be less likely to demand higher wages.

-> There is a short-run tradeoff between inflation and unemployment: when one rises, the other will tend to fall.

-- [To underscore this point, we saw Case & Fair's Table 8.7, "Inflation Rates, 1974-1976 and 1980-1983," which showed how inflation fell sharply during and right after the severe recessions of 1974-75 and 1980-82. In both cases the Federal Reserve attempted to "break the back" of the inflation by raising interest rates and slowing down the economy, and in both cases it succeeded, though at the cost of having a recession.]

The PHILLIPS CURVE illustrates the tradeoff between inflation and unemployment (two "bads") or, by extension, between higher output and lower prices (two "goods").

-- [We saw two diagrams of the Phillips Curve: a generic one, and one for the 1960s that featured actual unemployment and inflation data for every year from 1960-69; refer to Case & Fair, Figures 15.5 and 15.6, p. 326.]

To repeat myself a bit:

Q: Why is there a tradeoff between inflation and unemployment? (Or, equivalently, why is there a tradeoff between low prices and real GDP growth? Or, why does the Phillips Curve slope downward?)

A: Two main reasons:

(1) It's easier for firms to raise prices when incomes (and hence product demand) are high. Sometimes they're just following supply and demand -- if demand increases and the firm faces diminishing returns to expanding production in the short run, then the new supply price of the firm's good will have to be higher than the old price. Firms in noncompetitive industries -- i.e., firms that have some price-setting ability will often take advantage of high product demand to raise their prices, because they know they can do that without losing a lot of customers.
(2) When firms want to expand production, workers tend to get paid higher wages. This is because expanding firms will require more labor, which often requires paying a higher wage to (a) attract more workers and/or (b) induce their existing workers to work overtime. Also, with tight labor markets, workers are harder to replace and thus are in a better position to demand higher wages. The higher wages are then passed onto consumers in the form of higher product prices.

Q: Why does the Phillips Curve "curve" (i.e., why does it get flatter as you follow it from left to right?)
A: The economy has an "inflationary bias" -- it's much easier for prices and wages to go up than to go down. Companies are reluctant to cut the prices of their products, even in periods of slack demand; and workers fiercely resist any attempts to cut their nominal wages. So periods of rapid economic growth and low unemployment will generate many price and wage increases, and hence be very inflationary; by contrast, even a severe economic recession, while it will reduce the rate of growth of prices, will probably be unable to wipe out inflation altogether. A certain amount of "inertial inflation" is simply built into the system.

In the 1970s, inflation and unemployment frequently increased at the same time, something that had never really happened before. There was no obvious tradeoff between inflation and unemployment, which led many people to charge that the whole concept of the Phillips Curve was pure bunk.
--> What's more likely is that the Phillips Curve can shift up or down, and did so during the 1970s. In particular, the Phillips Curve shifted up at least twice during the 1970s. (It also seems to have shifted down in the mid-1980s and again in the 1990s.)
---- Economist Arthur Okun said it best: "The Phillips curve has become an unidentified flying object."

Today, the Phillips Curve is somewhat controversial -- it is not popular (which may be why your textbook doesn't introduce it until Chapter 15), but most economists do agree that there is some tradeoff between unemployment and inflation. Recessions cause firms to moderate their price increases and workers to moderate their wage demands. The Fed's Board of Governors seem to operate under the assumption that their policies will affect both unemployment and inflation. It is almost universally accepted that the Fed's tight-money policies in 1979-82 were responsible both for lowering inflation and for producing the 1981-82 recession. To quote Case & Fair:
"There is a tradeoff between inflation and unemployment, but other factors besides unemployment affect inflation. Policy involves much more than simply choosing a point along a nice, smooth [Phillips] curve."


* Defns.:

INFLATION: an increase in the overall price level.

DEFLATION: a decrease in the overall price level.
-- a major deflation has not occurred in this country since the Great Depression of the 1930s. A very minor one did occur in 1954, when the consumer price index fell 0.4% (the inflation rate was -0.4%).

DISINFLATION: a decline in the rate of inflation
-- recent ex.: early 1980s: inflation reduced from > 10% in 1979-81 to about 4% in early 1980s
-- The costs of a deliberate disinflation tend to be high, because the only sure way to reduce inflation is by creating a recession.

-- [We saw a graph of the leading price index (the Consumer Price Index, or CPI) over time, which showed that the CPI has increased steadily over the past 50 years. In other words, there has been inflation in virtually every year. Alas, this graph does not appear in Case & Fair's textbook.]

-- [We also saw an overhead of Case & Fair's Table 8.9, "The CPI, 1950-1997" (p. 166). That table shows the inflation rate -- that is, the "percent change in the CPI" -- for each of those years. The inflation rate was positive in all of those years but one (1955, when prices fell a tiny 0.4%).]

-- [We finished with an overhead of Case & Fair's Table 8.10, showing "Inflation and Unemployment in 1997" in other countries. The low inflation rates that the U.S. currently enjoys are far from a worldwide phenomenon. Bulgaria, for example, was experiencing a hyperinflation in 1997, with an inflation rate of 1,268%. So were the Congo and Rumania, with inflation rates of 544% and 175%. Turkey, Hungary, and several Latin American countries (Venezuela, Mexico, Colombia) also experienced very high inflation.]


Wed., March 8, 2000

* Today: Inflation (continued)
I. The inflation rate and the price level
II. Costs of inflation

I. INFLATION (continued)

Recall: INFLATION: a persistent increase in the overall price level.

There are different measures of inflation that correspond to different price levels.
-- The most commonly used price level in calculating the rate of inflation is the consumer price index (CPI), which measures the cost of purchasing a "basket" of certain consumer goods (including basic items like food, clothing, rent, and transportation).
-- The producer price index, formerly called the wholesale price index, measures the cost of industrial commodities at the wholesale level; hence it measures the prices that companies receive for the goods they produce, as opposed to the retail prices that we pay.
-- The broadest measure of the price level is the GDP price index, or GDP implicit price deflator, which measures the cost of purchasing all of the goods and services in GDP; so it includes the prices not just of consumer goods but also of investment goods, government-purchased goods and services, exports, and imports.

INFLATION RATE: the annual percent change in the overall price level (usually in the CPI).
-- The (CPI) inflation rate averaged 3.9% from 1982-1990 and is currently about 2%. It has fallen fairly steadily since the early 1990s.
-- Ex.: If the CPI was 152.4 in 1995 and 156.9 in 1996 (meaning that prices in 1995 and 1996 were 152.4% and 156.9% as high as prices were in the base year, which was 1982-84), then:
1996 inflation rate = {(1996 price level)/(1995 price level)} - 1 * 100%
                            = {156.9/152.4 - 1} * 100%
                            = {1.0295 - 1} * 100%
                            = .0295 * 100%
                            = 2.95%

How to construct a price index:

A price index measures the cost of buying a certain "basket" of goods, so one must:
(1) total up the dollar cost of buying given quantities of all of the items in the basket, for each of the years we are looking at.
Then, so that we may more easily compare price levels for different years, we index those cost totals to the however much it costs to buy that basket of goods in the base year, which is whatever year we choose to be our basis of comparison. Thus, we must
(2) choose a base year. Then, for each year, compute the price index by dividing the total cost of the basket of goods in that year by the total cost of that basket of goods in the base year, and then multiply by 100. So the price index for the base year is always 100, because we're dividing a number by itself and then multiplying by 100. If the same basket of goods costs 4% more (i.e., 104% as much) in the next year, then the next year's price index is 104.

price index for year t = 100 * (cost total for year t)/(cost total for base year)

Example of constructing a price index and calculating the inflation rate from it:

Imagine the same small island economy that we used in our nominal-vs.-real GDP example from earlier. In 1995, which we will use as our base year, the average person there consumed just three commodities -- beer, pretzels, and bicycles -- in the following quantities:

Commodity p q p * q
Case of beer $20 1     $20
Bag of pretzels $1 20 +  $20
Bicycle $200 1 +$200

-- Since 1995 is the base year, the price index for 1995 is: {$240/$240} * 100 = 100

-- In 1996, the price of beer was still $20, and the price of a bag of pretzels rose to $1.50 and the price of a bike rose to $210. Now that same basket of goods costs a bit more:



p * q
Case of beer $20 1     $20
Bag of pretzels $1.50 20 +  $30
Bicycle $210 1 +$210

-- In 1996, the price index was: {$260/$240} * 100 = 1.0833 * 100 = 108.33
-- The 1996 inflation rate was just the difference between the two, or 8.33%. To verify:
    1996 inflation rate = {(108.33/100) - 1} * 100
                                = (1.0833 - 1) * 100
                                = .0833 * 100
                                = 8.33%


Inflation and unemployment are perhaps the two most visible macroeconomic problems. The costs of unemployment are clear enough -- lost output, a lower standard of living, the despondency of joblessness, etc. -- and arguably are a lot more severe than the costs of inflation. When unemployment rates rise, suicide, crime, and divorce rates rise, too. People don't kill themselves over inflation, by contrast.
-- "Inflation, like every teenager, is grossly misunderstood, and this gross misunderstanding blows the political importance of inflation out of all proportion to its economic importance" (-- economist Alan Blinder).
-- Yet the same Alan Blinder has said that the 1970s proved that the U.S. economy does not function well under high inflation. Why not? We shall see...

Q: What are the costs of inflation?

(1) Arbitrary redistribution of income
(a) People whose incomes do not increase at the same rate as the price level will lose purchasing power -- i.e., their real incomes will fall. Inflations are rarely balanced -- real wages go up for some people, down for others. People's whose incomes are not "indexed" to inflation (some pensioners, welfare recipients, people with multi-year fixed-wage contracts) will see their real incomes fall.
-- Historically, wage increases have kept pace with price increases, so that real wages on average have risen or stayed the same, despite inflation.
-- Still, since inflations are rarely balanced, some people's real incomes do fall. If inflation causes a lot of people's real incomes to fall, then complaints about inflation will tend to be widespread.
(b) Inflation redistributes income from creditors to debtors (i.e., from savers and lenders to borrowers)
-- Defn.: REAL INTEREST RATE. The real, or inflation-adjusted, interest rate is computed as:
              real interest rate = interest rate - inflation rate
---- An unanticipated inflation lowers the real interest rate on existing loans. Ex.: If you borrow money at 10% today, when inflation is 2%, then you're probably expecting to pay a real interest rate of 8% (10% - 2%). If inflation then rises to 10%, then the real interest rate is 0%, so you're in effect getting an interest-free loan. That's great news for you, but bad news for the bank. Thus inflation is good for borrowers and bad for lenders.
---- Note: an unanticipated deflation has just the opposite effect: if you borrow money, prices fall (and your salary probably does, too), but your debt and interest payments are still the same, so the burden of your debt is greater.
---- Note also: if the inflation is anticipated, lenders can just raise their nominal interest rates so that the real interest rate is unchanged. When the inflation rate is high, (nominal) interest rates are high, too.
---- Aside: "Redistribution of income" is typically taken to mean "from rich to poor"-- e.g., tax the rich, feed the poor. That's not precisely what "redistribution of income" means in this case, yet some people say "inflation is good because it helps poor people." This is partly true in the sense that poor people tend to be debtors, while rich people tend to be creditors. But the poorest of the poor typically live on fixed incomes (e.g., welfare) or receive wages (e.g., the minimum wage) that are not indexed to inflation.

(2) Inefficient allocation of resources
-- Inflation robs people of information about relative prices, thereby making it harder for them to get the best deals or make the best decisions.
---- Ex.: You go to the supermarket and all prices have doubled from last week --> you might think, "I'm never shopping here again!" You drive to another supermarket and find that prices have doubled there, too. Disgusted, you drive to a third supermarket, only to find that they too are charging twice as much as last week. By now you realize that prices have doubled everywhere, but in the meantime you've wasted quite a bit of time looking around for a cheaper supermarket.
---- Without knowing the relative prices of the goods you might consume, it's easy to make bad decisions, and very hard to maximize your consumer satisfaction according to the Econ 101 principle of buying more and more of whatever goods give you the most marginal utility per price, until the marginal utility per price of every good you consume is the same. That's hard to do when those prices keep changing.
-- People waste more time making trips to the bank, because the nominal interest rate goes up as a result of a higher inflation rate. When your bank is paying higher rates of interest, you carry less cash, since cash pays no interest, and leave more money in the bank. The only way to do this without cutting back on your spending is to make smaller but more frequent withdrawals from your bank account -- say, withdrawing $20 once a day instead of $140 once a week. Making all those extra trips to the bank tends to be time-consuming and very inconvenient. Economists refer to those extra trips to the bank as the "shoe-leather costs of inflation" (since you might literally wear out the soles of your shoes more quickly if you have to make all these extra trips to the bank).

(3) Increased investment risk and (possibly) slower long-term economic growth
-- When unanticipated inflation occurs regularly, the degree of risk associated with new investments increases. Uncertainty about future inflation may inhibit people from investing in capital and long-term projects. Lower investment lowers the long-term rate of economic growth.


Fri., March 10, 2000


* Today:
I. Costs of inflation (finish)
II. Causes of inflation
III. Hyperinflation
IV. The Phillips Curve and the NAIRU

I. COSTS OF INFLATION (finish) -- [for these notes, I grouped them in with the previous day's lecture, to keep that unit together.]


Q: What causes inflation?
A: There are two main types of inflation:

(1) Demand-pull inflation (the not-so-bad kind)-- inflation that is initiated by an increase in demand-driven economic growth. Demand-pull inflation can be caused by any of the following: increased government spending, a "consumption binge" in which households try to consume a lot more, an "investment binge" in which firms rush to build new plant and equipment, or a big increase in the supply of money and credit that causes people and firms to borrow more money to finance new consumption or investment.
-- Exs.: the inflation of the late 1960s was fueled by the government's high levels of spending on the Vietnam War; the rapid inflation of 1977-79 was fueled by the Federal Reserve's rapid expansion of bank credit.

(2) Cost-push, or supply-side, inflation (the worse kind) -- inflation that is initiated by an increase in input or materials costs (independent of demand shifts). For example, a big increase in wages (say, because of stronger unions) would raise production costs and cause firms to pass on those higher costs in the form of higher prices. So would an increase in energy costs, since energy is necessary for all types of industrial production.
-- Exs.: the OPEC (Oil Producers' Exporting Cartel) oil shocks of 1973-74 and 1979-80 caused big increases in the inflation rate because oil is a vital input in production and because gasoline is a big part of consumers' budgets.
-- Closely related is "inertial" inflation, occurs when firms raise their prices because they expect other prices to go up, too.

Expectations of higher inflation in the future tend to become self-fulfilling.


Defn. HYPERINFLATION: a very rapid rise in the price level, of at least 100% and usually by more than 1000%.

The United States has never experienced a hyperinflation in its modern history, but it did during the Revolutionary War, when the value of a "Continental" dollar at war's end was only one-hundredth of its earlier value. Also, the Confederacy experienced a hyperinflation of its currency during the Civil War.

The most famous hyperinflation occurred in Germany during the 1920s and led to a total breakdown of German society that helped pave the way for Adolf Hitler's rise to power.
-- After Germany's defeat in World War I in the late 1910s, the country faced massive war-related debts and a huge bill for reparations payments to the Allies. Unable to raise such huge sums through taxes, the German Weimar Republic attempted to pay its bills the old-fashioned way -- by printing money.
-- When a government tries to finance its spending by printing money, the typical result is a massive inflation. Germany's was all the more so. To quote economist Campbell McConnell:
"During 1922 the German price level went up 5,470 percent. In 1923, the situation worsened; the German price level rose 1,300,000,000,000 [1.3 trillion] times. By October of 1923, the postage on the lightest letter sent from Germany to the United States was 200,000 marks. Butter cost 1.5 million marks per pound, meat 2 million marks, a loaf of bread 200,000 marks, and an egg 60,000 marks. Prices increased so rapidly that waiters changed the prices on the menu several times during the course of a lunch. Sometimes customers had to pay double the price listed on the menu when they ordered."


Recall: there is a short-term tradeoff between inflation and unemployment. We'd like both to be lower, but in the short term:

Inflation will tend to not just rise but to accelerate (keep on rising higher and higher) if the unemployment rate drops below a certain level. We call that minimum sustainable rate of unemployment the non-accelerating-inflation rate of unemployment (NAIRU).
-- (For future reference: The level of GDP that corresponds to the NAIRU is called potential GDP.)

Defn. NAIRU: the lowest level of unemployment that the economy can sustain without causing inflation to accelerate; the Non-Accelerating-Inflation Rate of Unemployment.
-- The NAIRU is sometimes called the "target" rate of unemployment, since it tends to be the rate that government and Federal Reserve policymakers try to aim for. We'd like unemployment to be low and, for most of us, a little inflation in exchange for low unemployment is acceptable, but accelerating inflation is out of the question.
-- Case & Fair's book refers to the NAIRU by the name "natural rate of unemployment," which is a usage I believe should be avoided at all costs. The word "natural" falsely implies that this certain level of unemployment is all voluntary and even desirable, which of course it isn't, and the book concedes this. So why go on using the term "natural rate of unemployment"? Another alternative is the "target rate of unemployment." Still, "non-accelerating-inflation rate of unemployment" is more accurate, and "NAIRU" (pronounced like "Nehru") is easy enough to pronounce.
-- When unemployment falls below the NAIRU and inflation accelerates, the Phillips Curve will shift upward, causing the inflation-unemployment tradeoff to worsen.

Exactly how high is the NAIRU? Nobody knows for sure. My best guess is about 4%, perhaps a bit lower (3.5%?).