Ranjit Dighe
March 13-17, 2000

Mon., March 13, 2000

* Today:
I.    The Phillips Curve and the NAIRU (finish)
II.   The Classical labor market
III.  Explanations of unemployment (begin)


Where we left off: the NAIRU (non-accelerating-inflation rate of unemployment). I said it seems to be around 3.5-4% right now, but nobody knows for sure. The only way to tell is to observe when unemployment is low and constant and the rate of inflation starts to accelerate, which happened in the late 1970s (when unemployment was just below 6%) and in the late 1980s (when unemployment was just below 5.5%). Lately, unemployment has been close to 4%, and there are no signs of higher inflation, let alone accelerating inflation. So it seems the NAIRU has fallen quite a bit in the past decade.
-- The temptation is to say the NAIRU is whatever the unemployment rate has been for the last few years, as long as we're not obviously in a recession. It was long thought to be around 6% (from about the mid-1970s to the mid-1990s).
-- More recently, many economists have said it's 3.5%, 4%, 4.5%, or 5%. In recent years, as unemployment has fallen all the way to 4.2% with few visible signs of higher inflation, it appears the NAIRU has fallen, too, perhaps to 4 or 4.5%. Yet economists in general and Federal Reserve policymakers in particular seem slow to adjust their estimates of the NAIRU; the last textbook I used estimates that the NAIRU is 5.5%, which to me seems way too high.
---- We saw a historic occasion in September 1996, when the most recent unemployment rate was 5.1%, the lowest in many years, and yet the Fed voted NOT to raise interest rates (which they would normally do in times of unusually low unemployment, in order to slow down the economy), in a move that surprised financial markets. (The Fed said they were reacting to indicators of a soon-to-come slowdown in the economy, not to the present. Still...) The Fed has repeated that non-move several times, which suggests that the Alan Greenspan and the other Fed governors also believe that the NAIRU is much lower now than it was 10 years ago.

Even when unemployment is higher than the NAIRU and the economy has some slack in it, the inflation-unemployment tradeoff is still present -- lower rates of unemployment will be associated with higher rates of inflation.
-- In the 1950s and 1960s, the Phillips Curve was extremely stable, in that the unemployment and inflation data fit a Phillips Curve extremely well. In periods of high unemployment (like recessions), inflation was low. And when the economy grew rapidly and unemployment fell, inflation rose. The economic policy debate of that time centered around where we should be on that curve. Conservatives favored reducing inflation (at the expense of higher unemployment); liberals favored reducing unemployment (at the expense of higher inflation). During the 1960s, as the economy underwent its longest-ever expansion, fueled largely by Vietnam War spending, we moved "up" the Phillips Curve toward lower unemployment and higher inflation.
---- [We saw an overhead of Case & Fair's Figure 15.6, "Unemployment and Inflation, 1960 to 1969" (p. 326).]
-- In the 1970s, the data did not fit a single Phillips Curve at all well. With the OPEC oil shocks, we saw inflation and unemployment rise at the same time, which had once been thought impossible. The term STAGFLATION was coined to refer to the combination of high unemployment (i.e., a stagnant economy) and high inflation at the same time.

--> What's happened in 1974-present: The Phillips Curve has SHIFTED a number of times.
---- The Phillips Curve shifts up when people come to expect a higher rate of inflation (expected rate of inflation rises; pe rises).
------ This will happen if inflation rates rise for a few years in a row, or if there is a big external spike in prices, from, say, a huge increase in oil prices.
---- The Phillips Curve shifts down when people come to expect a lower rate of inflation (pe falls).
------ This will happen if inflation rates fall for a few years in a row, on account of, say, a big recession induced by the Federal Reserve or because of a big drop in oil or import prices.

-- Apparently, the Phillips Curve shifted upward at least twice during the 1970s, first in 1973-74 with the first OPEC oil shock and then in 1979-80 with the second OPEC oil shock.
-- The Phillips Curve seems to have shifted downwards in the mid-1980s and during the 1990s, as people came to expect lower inflation rates and as oil prices fell.
---- [We saw an overhead of Case & Fair's Figure 15.7, "Unemployment and Inflation, 1970 to 1997" (p. 327). On that graph there is no discernible pattern of a tradeoff between inflation and unemployment. The reason is that the Phillips Curve shifted a number of times -- altogether, it shifted about 10 times!
------ I drew Phillips Curves on this graph, starting with one for 1970-73, which shifted way up in 1974 and did so again in 1975, then shifted down to a new position for 1976-79, shifted way up to a new position in 1980-83, shifted down for 1984-85, shifted down again for 1986-90, shifted up for 1991-93, and shifted down in 1994, and again for 1995-96, and shifted down yet again in 1997.
------ Except for the 1974 and 1979-80 OPEC oil shocks, most of the shifts seem to have been caused by changes in people's expectations of inflation. In 1976, 1984, and 1994, for example, the Phillips Curve seems to have shifted downward because the Fed's apparent willingness to induce recessions in order to tame inflation sent a strong signal that lowered people's inflationary expectations.]


In a properly functioning, competitive market, supply and demand should quickly reach an equilibrium point where quantity supplied and quantity demanded are equal. If the labor market worked that way, then there would be an equilibrium wage that cleared the market and there would be no unemployment. Let us examine what this ideal, competitive labor market would look like.
--[We saw a supply-and-demand diagram, titled "The Classical Labor Market." It was very similar to Case & Fair's Figure 15.1 (p. 319), though with just one demand curve.]
---- The price of labor is the wage, w.
---- The quantity of labor, L, can be thought of either as the number of workers or the total number of hours worked (by all workers combined). More generally, it is "units of labor."
---- The intersection of the SL (labor supply) and DL (labor demand) curves is the labor market equilibrium, with wage w* and employment or worker-hours level L*.
------> As long as the equilibrium wage w* prevails, then there is no surplus or shortage of labor, and hence no unemployment or vacancies.


In the classical labor market, there will never be any unemployment, because wages would always quickly adjust to any shocks to labor demand or labor supply. If the demand for labor falls (as it typically does during a recession), the equilibrium wage would fall right along with it. Employment would fall, but there would still be no unemployment, because fewer people are willing to work at that lower wage, so they drop out of the labor market altogether. Everyone who wants to work at the new, lower wage will still be employed.
-- [We saw an overhead of Case & Fair's Figure 15.1, "The Classical Labor Market" (p. 319).]


In the perfect world of the classical labor market, there is no unemployment. In the real world, of course, there always is some unemployment, and the question at hand is why. What causes unemployment?

Your textbook gives several different explanations of unemployment, all of which are variants of one basic idea:

STICKY WAGES -- the tendency of wages to remain "stuck" at levels above the market-clearing wage.
-- When the demand for labor falls (say, during a recession), wages typically do not fall from their previous level.
----> In the classical model of the labor market, that "downward wage rigidity" will cause unemployment.
------ [We saw a transparency of Case & Fair's Figure 15.2, "Sticky Wages" (p. 321). This is really a diagram of the classical model of the labor market with sticky wages.]


Wed., March 15, 2000

* Today: Explaining unemployment

I.   Wage stickiness, and explanations thereof
II.  Deficient demand


Where we left off: In the context of the classical labor market, there is one simple explanation for why unemployment exists: wages are set too high, and for some reason remain "stuck" at those high levels. In other words, we say that WAGES ARE STICKY.

Possible explanations of wage stickiness (downward wage rigidity):

* minimum wage laws -- The current minimum wage is $5.15 an hour; it sets a "price floor" for labor. These laws account for some unemployment, but only in markets for unskilled labor.
-- Teenagers, uneducated/unskilled adults are the most likely to be affected.
----> tradeoff: more jobs at lower wages, or fewer jobs at higher wages?
-- But the minimum wage ($5.15/hour) is much lower than the average wage, so clearly minimum wages can't explain the unemployment of, say, nuclear engineers or CPA's. (Clearly, if you can't get a job when you graduate, it won't be because of the minimum wage!)

* (explicit) contracts -- employment contracts that guarantee workers' wages, usually for a period of one to three years. Many workers have contracts, of 1+ years, specifying that they'll be paid a particular wage. If demand falls in mid-year and companies can't cut wages, then the alternative is to lay off some workers. Union contracts average 3 years in length.

* implicit contracts -- Unspoken commitments by firms not to cut their workers' wages. There is a social "taboo" against wage cutting, because it violates social norms and thus would antagonize workers and be bad for the firm's reputation. (Closely related to...)

* "efficiency wage" theory -- If the productivity of workers increases with the wage rate, then it may be most profitable for firms to pay a wage higher than the market-clearing wage.
-- "You get what you pay for"
-- A firm could attract enough workers at the market-clearing wage (w*), but it will likely get higher-quality, more motivated, more loyal workers if it pays a higher wage
----> The profit-maximizing wage, or the "efficiency wage," will likely be higher than w*.

To summarize:
Q: Why don't wages fall to clear the labor market and eliminate unemployment?
So far, three competing answers:
(1) government intervention in labor markets (minimum wages)
(2) contracts, explicit or implicit
(3) efficiency wages
BUT... all of those explanations seem to assume that falling wages WILL clear the labor market, and hence implicitly assume that involuntary unemployment is solely the result of high wages.
Is it necessarily true that falling wages WILL clear the labor market and eliminate all unemployment? This brings us to another explanation of unemployment, one that I find more convincing, at least for the case of cyclical unemployment (in recessions and depressions).


[NOTE: This does not appear in Case & Fair's textbook at all. I may put together a handout on this topic, complete with graphs, for after spring break, but I'm not making any promises.]

This theory is mainly due to John Maynard Keynes, the founding father of macroeconomics.

Background: The classical labor market assumes perfect competition in the labor market and in the product market.
-- Under perfect competition, every firm faces a flat demand curve for its products and can sell all of its output at the market price P*.
-- The conventional formulation of the labor demand curve [drawn in class] goes with the competitive assumption that the firm faces a flat demand curve and can sell all of its output at the market price [drawn in class]. Because of diminishing returns, the extra output of additional unit of labor (the marginal product of labor, MPL) is smaller than the output generated by the previous unit of labor. Just as a profit-maximizing firm produces up to the point where MC=P*, so does a profit-maximizing firm hire labor up to the point where the revenue generated by an additional unit of labor (the marginal revenue product of labor, MRPL) is equal to the wage. The labor demand curve is actually a combination of all the points that satisfy the equation w=MRPL.

PROBLEM: The classical labor market is NOT a realistic depiction of how labor and product markets operate, especially in recessions.
-- Instead, firms are often sales-constrained or demand-constrained: They know, or believe, that they can only sell a certain amount of output, q. (We say that the firm has a SALES CONSTRAINT at q.) The demand curve facing the firm ends at q. [drawn in class]
--> Then the firm only needs to hire as many workers as are necessary to produce q.
---- Let's call this maximum number of necessary workers L.
---- If the firm can't sell any output beyond q, then it makes no sense to hire any more than L workers. The marginal revenue product (MRPL) of an additional worker is ZERO, if the firm can't sell any of the extra output that the worker produces.
------> The firm now has an EFFECTIVE LABOR DEMAND curve that is VERTICAL at wage levels below the current wage (w0). It is still downward-sloping for wage levels above w0 -- higher wages would still reduce employment. But lower wages would not increase employment, since the firm cannot sell any more output than it is currently selling. [drawn in class]
--------> RESULT: Cutting the wage below w0 (the current wage) will NOT increase employment. There's no point in hiring extra workers if you can't sell the extra output they would generate.

The sales constraint, and the deficient-demand model of unemployment, could apply to the economy in general, as in a recession or depression, or it could be specific to a particular firm or industry, as in structural unemployment.


Fri., March 17, 2000


* Today: Money and monetary policy (begin)
I.  What is money?
II. Monetary policy: An introduction

* As promised last time, here is a pithy definition of the deficient-demand theory of unemployment:
Defn. DEFICIENT-DEMAND THEORY OF UNEMPLOYMENT: when demand for goods and services is lacking, then firms are sales-constrained, and they will hire only as many workers as they need to produce up to that sales constraint. Lower wages will not induce firms to hire more workers.



Before we go any further, let us recall that the macroeconomist's usage of the term "money" refers not to wealth or income (as in "I have a lot of money" or "I make a lot of money") but to the following definition:
-- MONEY - the medium of exchange; anything that is generally accepted as payment; "liquidity."

The money supply is not the same thing as GDP. Students sometimes confuse the money supply with GDP, but they are two very different things. GDP is the economy's total output of final goods and services in a given year; the money supply is, roughly speaking, the total amount of cash plus bank accounts in the economy at a given point in time. GDP is a "flow" measure, in that it measures the output of something over a particular interval; the money supply is a "stock" measure, in that it measures the quantity of money at a particular point in time.

Again, money is defined as anything that is generally accepted as payment or, in a word, as liquidity. Since having liquidity is a much smaller priority for most people than earning a high income or becoming wealthy is, we need to figure out what it is that money is good for. Economists have identified three main functions of money:

1. Medium of exchange -- you can use it to buy whatever you want. Money makes it much easier for people to exchange the goods and services they produce for the goods and services that they want. Money "greases the wheels of commerce," by making it much easier for people to exchange the goods and services they produce for the goods and services that they want. Having a generally accepted currency eliminates the need for barter (trading goods and services for each other), and makes the volume of transactions a lot larger than it would otherwise be.

2. Unit of account -- with a standard monetary unit, like the dollar, we can easily price individual goods and services, putting a single price on each item instead of having to compute a different exchange price for every different pair of commodities (e.g., 1 cup of coffee = 2 newspapers = 6 minutes of office work as a temp = 3 minutes of my teaching services).

3. Store of value -- money has some use as an asset, because it holds its nominal value over time and, unlike stocks or bonds, its value does not fluctuate from day to day. Unlike stocks or bonds, there is no risk that dollars will suddenly become worthless. In sum, money is a virtually riskless asset. It is also a very liquid (convertible into cash; spendable) asset, which is another desirable quality.

At various times and places, many different things have been used as money. Long ago, most types of money were:
COMMODITY MONEY: items used as money that also have (intrinsic) value in some other use.
-- Ex.: gold coins

Today, almost every country's currency is instead:
FIAT MONEY: items designated as money that are intrinsically worthless.
-- Ex.: U.S. paper money - only other value is as paper
-- Fiat money has value only because the government declares it to be legal tender and because people believe it has value.
---- LEGAL TENDER: money that a government has required to be accepted as payment for debts.


There are two main approaches that economic policymakers can use to influence the level of real GDP, unemployment, and inflation:

(1) FISCAL POLICY: the spending and taxing policies used by the government to influence the economy.
-- has not really been used since the early 1980s.

(2) MONETARY POLICY: the steps that the Federal Reserve takes to influence the economy through changes in the supply of money and credit (which usually take the form of changes in interest rates).
-- has become more important than ever before. Today, "the Federal Reserve runs the economy."

In most countries, including the U.S., monetary policy is carried out by a central bank, which has the power to change the supply of money and credit through its actions. In the U.S., that central bank is not one single bank but a whole network called the FEDERAL RESERVE SYSTEM (a.k.a. the Federal Reserve, or the Fed). At the head of the Federal Reserve is Alan Greenspan, who has been its Chairman since 1987.

Despite the confusion between the word "Fed" (nickname for Federal Reserve) and "Feds" (the federal government), the Fed is not part of the government. It is what we'd call a "quango" -- a quasi-non-governmental organization.
-- The Fed is technically an independent, private agency, but--
---- Its most powerful members are appointed by the President and confirmed by the Senate (much like Supreme Court Justices).
---- The chairperson of the Fed serves a four-year term, which means that each President gets the chance to appoint a Fed chair of his choosing.
------ Alan Greenspan was originally appointed by Ronald Reagan in 1987 and was reappointed by George Bush and Bill Clinton.
---- Also, the Fed, unlike a truly private organization, must return 100% of any excess profits it makes to the government (although it does not get its funding from the government).

The Federal Reserve System consists of:
* 12 regional Federal Reserve Banks (FRB's, or Fed banks).
-- The New York Fed, located in the Wall Street financial district, is by far the most powerful. Most major U.S. financial centers have a Fed bank -- these include Chicago, San Francisco, Boston, Atlanta, Dallas, and St. Louis.
* about 8000 commercial banks. All banks are required to set a certain fraction of their deposits aside as reserves, to be held as cash at the bank or at the nearest regional Fed bank. The Fed manipulates the levels of bank reserves in order to control the money supply and interest rates.
* the Federal Open Market Committee (FOMC) -- the group that really controls monetary policy.
-- The FOMC has 12 members. They are:
---- the president of the New York Fed (who serves as Vice-Chairman)
---- four presidents from the other 11 regional Fed Banks, who serve on the FOMC on a rotating basis;
---- the seven members of the Federal Reserve Board of Governors.
------ The seven Fed Governors are economists who serve 14-year terms and are headquartered in Washington, DC. Since they are the most permanent and prestigious members of the FOMC, they are also the most powerful.
-- The FOMC meets every six weeks to plot the course of monetary policy. It usually does this by adjusting two interest rates:
---- DISCOUNT RATE: the interest rate at which the regional Fed banks loan money to commercial banks (currently 5.25%);
---- FEDERAL FUNDS RATE: the interest rate at which banks loan reserves to each other, on an overnight basis (currently about 5.75%).
------ The Fed does not set the federal funds rate directly, but rather sets a target for it, and then takes steps to keep the federal funds rate close to that target. It does this by manipulating the supply of bank reserves so as to affect the federal funds rate by affecting supply and demand in that market.