C H A P T E R 2 5 Aggregate Demand and Supply Analysis 597
www.whitehouse.gov/fsbr /esbr.html
The White House sponsors an economic statistics briefing room that reports a wide variety of interesting data dealing with the state of the economy.
One group, led by Edward Prescott of the University of Minnesota, has developed a theory of aggregate economic fluctuations called real business cycle theory, in which aggregate supply (real) shocks do affect the natural rate level of output Yn. This theory views shocks to tastes (workersÕ willingness to work, for example) and technology (productivity) as the major driving forces behind short-run fluctuations in the business cycle, because these shocks lead to substantial short-run fluctuations in Yn. Shifts in the aggregate demand curve, perhaps as a result of changes in monetary policy, by contrast are not viewed as being particularly important to aggregate output fluctuations. Because real business cycle theory views most business cycle fluctuations as resulting from fluctuations in the natural rate level of output, it does not see much need for activist policy to eliminate high unemployment. Real business cycle theory is highly controversial and is the subject of intensive research.5
Another group of economists disagrees with the assumption that the natural rate level of output Yn is unaffected by aggregate demand shocks. These economists contend that the natural rate level of unemployment and output are subject to hysteresis, a departure from full employment levels as a result of past high unemployment.6 When unemployment rises because of a reduction of aggregate demand that shifts the AD curve inward, the natural rate of unemployment is viewed as rising above the full employment level. This could occur because the unemployed become discouraged and fail to look hard for work or because employers may be reluctant to hire workers who have been unemployed for a long time, seeing it as a signal that the worker is undesirable. The outcome is that the natural rate of unemployment shifts upward after unemployment has become high, and Yn falls below the full employment level. In this situation, the self-correcting mechanism will be able to return the economy only to the natural rate levels of output and unemployment, not to the full employment level. Only with expansionary policy to shift the aggregate demand curve to the right and raise aggregate output can the natural rate of unemployment be lowered (Yn raised) to the full employment level. Proponents of hysteresis are thus more likely to promote activist, expansionary policies to restore the economy to full employment.
Study Guide |
Aggregate supply and demand analysis are best learned by practicing applications. In |
|
|
this section, we have traced out what happens to aggregate output when there is an |
|
|
increase in the money supply or a negative supply shock. Make sure you can also |
|
|
draw the appropriate shifts in the aggregate demand and supply curves and analyze |
|
|
what happens when other variables such as taxes or the expected price level change. |
|
|
|
|
Aggregate demand and supply analysis yields the following conclusions (under the usual assumption that the natural rate level of output is unaffected by aggregate demand and supply shocks):
1. A shift in the aggregate demand curveÑwhich can be caused by changes in monetary policy (the money supply), fiscal policy (government spending or taxes),
5See Charles Plosser, ÒUnderstanding Real Business Cycles,Ó Journal of Economic Perspectives (1989): 51Ð77, for a nontechnical discussion of real business cycle theory.
6For a further discussion of hysteresis, see Olivier Blanchard and Lawrence Summers, ÒHysteresis in the European Unemployment Problem,Ó NBER Macroeconomics Annual, 1986, 1, ed. Stanley Fischer (Cambridge, Mass.: M.I.T. Press, 1986), pp. 15Ð78.
598 P A R T V I Monetary Theory
international trade (net exports), or Òanimal spiritsÓ (business and consumer optimism)Ñ affects output only in the short run and has no effect in the long run. Furthermore, the initial change in the price level is less than is achieved in the long run, when the aggregate supply curve has fully adjusted.
2.A shift in the aggregate supply curveÑwhich can be caused by changes in expected inflation, workersÕ attempts to push up real wages, or a supply shockÑ affects output and prices only in the short run and has no effect in the long run (holding the aggregate demand curve constant).
3.The economy has a self-correcting mechanism, which will return it to the natural rate levels of unemployment and aggregate output over time.
Application |
Explaining Past Business Cycle Episodes |
Vietnam War Buildup, 1964–1970
Negative Supply Shocks, 1973–1975 and 1978–1980
Aggregate supply and demand analysis is an extremely useful tool for analyzing aggregate economic activity; we will apply it to several business cycle episodes. To simplify our analysis, we always assume in all three examples that aggregate output is initially at the natural rate level.
AmericaÕs involvement in Vietnam began to escalate in the early 1960s, and after 1964, the United States was fighting a full-scale war. Beginning in 1965, the resulting increases in military expenditure raised government spending, while at the same time the Federal Reserve increased the rate of money growth in an attempt to keep interest rates from rising. What does aggregate supply and demand analysis suggest should have happened to aggregate output and the price level as a result of the Vietnam War buildup?
The rise in government spending and the higher rate of money growth would shift the aggregate demand curve to the right (shown in Figure 5). As a result, aggregate output would rise, unemployment would fall, and the price level would rise. Table 3 demonstrates that this is exactly what happened: The unemployment rate fell steadily from 1964 to 1969, remaining well below what economists now think was the natural rate of unemployment during that period (around 5%), and inflation began to rise. As Figure 5 predicts, unemployment would eventually begin to return to the natural rate level because of the economyÕs self-correcting mechanism. This is exactly what we saw occurring in 1970, when the inflation rate rose even higher and unemployment increased.
In 1973, the U.S. economy was hit by a series of negative supply shocks. As a result of the oil embargo stemming from the Arab-Israeli war of 1973, the Organization of Petroleum Exporting Countries (OPEC) was able to engineer a quadrupling of oil prices by restricting oil production. In addition, a series of crop failures throughout the world led to a sharp increase in food prices. Another factor was the termination of wage and price controls in 1973 and 1974, which led to a push by workers to obtain wage increases that had been prevented by the controls. The triple thrust of these events caused the aggregate supply curve to shift sharply leftward, and as the aggregate demand and
C H A P T E R 2 5 Aggregate Demand and Supply Analysis 599
Table 3 Unemployment and Inflation During the Vietnam War Buildup,
1964–1970
Year |
Unemployment Rate (%) |
Inflation (Year to Year) (%) |
1964 |
5.0 |
1.3 |
1965 |
4.4 |
1.6 |
1966 |
3.7 |
2.9 |
1967 |
3.7 |
3.1 |
1968 |
3.5 |
4.2 |
1969 |
3.4 |
5.5 |
1970 |
4.8 |
5.7 |
Source: Economic Report of the President.
Favorable Supply Shocks, 1995–1999
supply diagram in Figure 6 predicts, both the price level and unemployment began to rise dramatically (see Table 4).
The 1978Ð1980 period was almost an exact replay of the 1973Ð1975 period. By 1978, the economy had just about fully recovered from the 1973Ð1974 supply shocks, when poor harvests and a doubling of oil prices (as a result of the overthrow of the Shah of Iran) again led to another sharp leftward shift of the aggregate supply curve. The pattern predicted by Figure 6 played itself out againÑinflation and unemployment both shot upward (see Table 4).
In February 1994, the Federal Reserve began to raise interest rates, because it believed the economy would be reaching the natural rate of output and unemployment in 1995 and might become overheated thereafter. As we can see in Table 5, however, the economy continued to grow rapidly, with unem-
Table 4 Unemployment and Inflation During the Negative Supply Shock Periods, 1973–1975 and 1978–1980
|
Unemployment |
Inflation |
|
Unemployment |
Inflation |
Year |
Rate (%) |
(Year to Year) (%) |
Year |
Rate (%) |
(Year to Year) (%) |
1973 |
4.8 |
6.2 |
1978 |
6.0 |
7.6 |
1974 |
5.5 |
11.0 |
1979 |
5.8 |
11.3 |
1975 |
8.3 |
9.1 |
1980 |
7.0 |
13.5 |
Source: Economic Report of the President.
600 P A R T V I Monetary Theory
Table 5 Unemployment and Inflation During the Favorable Supply Shock
Period, 1995–2000
Year |
Unemployment Rate (%) |
Inflation (Year to Year) (%) |
1995 |
5.6 |
2.8 |
1996 |
5.4 |
3.0 |
1997 |
4.9 |
2.3 |
1998 |
4.5 |
1.6 |
1999 |
4.2 |
2.2 |
Source: Economic Report of the President; ftp://ftp.bls.gov/pub/special.requests/cpi/cpiai.txt.
ployment falling to below 5%, well below what many economists believed to be the natural rate level, and yet inflation continued to fall, declining to around 2%. Can aggregate demand and supply analysis explain what happened?
The answer is yes. Two favorable supply shocks hit the economy in the late 1990s. First, changes in the health care industry, such as the movements to health maintenance organizations (HMOs), reduced medical care costs substantially relative to other goods and services. Second, the computer revolution finally began to have a favorable impact on productivity, raising the potential growth rate of the economy (which journalists have dubbed the Ònew economyÓ). The outcome was a rightward shift in the aggregate supply curve, producing the opposite result depicted in Figure 6: Aggregate output rose, and unemployment fell, while inflation also declined.
Summary
1.The aggregate demand curve indicates the quantity of aggregate output demanded at each price level, and it is downward-sloping. Monetarists view changes in the money supply as the primary source of shifts in the aggregate demand curve. Keynesians believe that not only are changes in the money supply important to shifts in the aggregate demand curve, but so are changes in fiscal policy (government spending and taxes), net exports, and the willingness of consumers and businesses to spend (Òanimal spiritsÓ).
2.In the short run, the aggregate supply curve slopes upward, because a rise in the price level raises the profit earned on each unit of production, and the quantity of output supplied rises. Four factors can cause the
aggregate supply curve to shift: tightness of the labor market as represented by unemployment relative to the natural rate, expectations of inflation, workersÕ attempts to push up their real wages, and supply shocks unrelated to wages that affect production costs.
3.Equilibrium in the short run occurs at the point where the aggregate demand curve intersects the aggregate supply curve. Although this is where the economy heads temporarily, it has a self-correcting mechanism, which leads it to settle permanently at the long-run equilibrium where aggregate output is at its natural rate level. Shifts in either the aggregate demand or the aggregate supply curve can produce changes in aggregate output and the price level.
C H A P T E R 2 5 Aggregate Demand and Supply Analysis 601
Key Terms
activists, p. 592 aggregate demand, p. 582
aggregate demand curve, p. 582 aggregate supply, p. 582 aggregate supply curve, p. 588 Òanimal spirits,Ó p. 586 complete crowding out, p. 587 consumer expenditure, p. 585 equation of exchange, p. 583 government spending, p. 585
QUIZ Questions and Problems
hysteresis, p. 597
Keynesians, p. 582
long-run aggregate supply curve, p. 590
modern quantity theory of money, p. 584
monetarists, p. 582
natural rate level of output, p. 590
natural rate of unemployment, p. 590
net exports, p. 585
nonaccelerating inflation rate of unemployment (NAIRU), p. 590
nonactivists, p. 592
partial crowding out, p. 587
planned investment spending, p. 585
real business cycle theory, p. 597
self-correcting mechanism, p. 591
supply shocks, p. 594
velocity of money, p. 582
Questions marked with an asterisk are answered at the end of the book in an appendix, ÒAnswers to Selected Questions and Problems.Ó
1.Given that a monetarist predicts velocity to be 5, graph the aggregate demand curve that results if the money supply is $400 billion. If the money supply falls to $50 billion, what happens to the position of the aggregate demand curve?
*2. Milton Friedman states, ÒMoney is all that matters to nominal income.Ó How is this statement built into the aggregate demand curve in the monetarist framework?
3.Suppose that government spending is raised at the same time that the money supply is lowered. What will happen to the position of the Keynesian aggregate demand curve? The monetarist aggregate demand curve?
*4. Why does the Keynesian aggregate demand curve shift when Òanimal spiritsÓ change, but the monetarist aggregate demand curve does not?
5.If the dollar increases in value relative to foreign currencies so that foreign goods become cheaper in the United States, what will happen to the position of the aggregate supply curve? The aggregate demand curve?
*6. ÒProfit-maximizing behavior on the part of firms explains why the aggregate supply curve is upwardsloping.Ó Is this statement true, false, or uncertain? Explain your answer.
7.If huge budget deficits cause the public to think that there will be higher inflation in the future, what is likely to happen to the aggregate supply curve when budget deficits rise?
*8. If a pill were invented that made workers twice as productive but their wages did not change, what would happen to the position of the aggregate supply curve?
9.When aggregate output is below the natural rate level, what will happen to the price level over time if the aggregate demand curve remains unchanged? Why?
*10. Show how aggregate supply and demand analysis can explain why both aggregate output and the price level fell sharply when investment spending collapsed during the Great Depression.
11.ÒAn important difference between monetarists and Keynesians rests on how long they think the long run actually is.Ó Is this statement true, false, or uncertain? Explain your answer.
Using Economic Analysis to Predict the Future
*12. Predict what will happen to aggregate output and the price level if the Federal Reserve increases the money supply at the same time that Congress implements an income tax cut.
13.Suppose that the public believes that a newly announced anti-inflation program will work and so lowers its expectations of future inflation. What will
2 Appendix to Chapter 25
and wages decline over time. Hence the Phillips curve supports the view of aggregate supply in Chapter 24 that when the labor market is slack, production costs will fall and the aggregate supply curve will shift to the right.2
Figure 1 shows what the Phillips curve relationship looks like for the United States. As we can see from panel (a), the relationship works well until 1969 and seems to indicate an apparent trade-off between unemployment and wage inflation: If the public wants to have a lower unemployment rate, it can ÒbuyÓ this by accepting a higher rate of wage inflation.
In 1967, however, Milton Friedman pointed out a severe flaw in the Phillips curve analysis: It left out an important factor that affects wage changes: workersÕ expectations of inflation.3 Friedman noted that firms and workers are concerned with
Annual |
|
|
|
|
|
|
|
|
|
Annual |
|
|
|
Phillips Curve |
|
Phillips Curve |
Rate of |
|
|
|
|
|
|
|
|
|
Rate of |
|
|
|
Early 1970's |
|
Mid-to-late 1970's |
Wage |
|
|
|
|
|
|
|
|
|
Wage |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
Change |
|
|
|
|
|
|
|
|
|
Change |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
Phillips Curve |
(%) 8 |
|
|
|
|
|
|
|
|
|
(%) 8 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
51 |
|
|
|
|
|
|
|
|
51 |
|
|
|
79 |
81 |
|
|
Early 1980's |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
77 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
48 |
|
|
72 |
|
|
|
|
|
|
|
|
48 |
|
|
|
|
|
|
|
|
|
|
|
|
78 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
78 |
|
|
6 |
|
|
69 |
|
|
|
|
|
|
|
6 |
|
|
|
69 |
|
73 |
74 |
80 |
|
|
|
|
|
50 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
68 |
|
|
|
|
|
|
|
|
|
|
|
90 |
50 |
71 |
|
75 |
|
|
|
|
|
|
|
|
|
|
|
53 68 55 |
|
|
50 |
|
|
|
|
|
53 |
55 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
59 |
|
|
|
|
|
|
|
|
59 |
|
|
83 |
|
|
|
|
|
|
|
|
|
|
|
91 |
|
|
|
|
|
4 |
|
|
|
|
|
|
|
|
4 |
|
|
|
|
|
|
|
|
|
|
82 |
|
|
52 |
56 |
65 |
62 |
|
|
|
|
|
|
52 |
5665 70 |
62 |
|
84 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
57 |
|
|
|
|
|
|
|
96 |
|
66 |
89 |
|
|
|
|
|
|
|
|
66 |
|
|
|
|
|
|
|
92 |
|
53 |
|
|
|
|
|
|
|
6463 |
|
|
|
|
|
|
|
|
|
57 |
6463 |
|
|
|
|
2 |
|
|
67 |
|
61 |
|
|
|
2 |
|
95 |
|
93 |
67 |
|
87 61 |
|
|
|
|
|
|
|
60 |
49 |
58 |
|
|
|
|
|
94 |
97 |
|
|
60 |
49 |
58 |
|
|
|
|
Phillips Curve, |
|
|
|
|
|
|
|
|
99 |
|
|
85 |
|
Phillips Curve |
|
|
|
|
|
|
|
|
Phillips Curve, |
98 |
|
|
|
|
86 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1948–1969 |
|
1948–1969 |
|
|
54 |
|
|
|
|
|
|
|
|
|
|
54 |
|
|
|
|
|
|
|
|
|
|
|
1992–2002 |
|
|
|
|
|
|
|
|
|
1985–1991 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
0 |
1 |
2 |
3 |
4 |
5 |
6 |
7 |
8 |
9 |
10 |
0 |
1 |
2 |
3 |
|
4 |
|
|
5 |
6 |
7 |
8 |
9 |
10 |
|
|
|
|
Annual Unemloyment Rate (%) |
|
|
|
|
|
Annual Unemloyment Rate (%) |
(a) Phillips Curve, 1948–1969 |
|
|
|
|
(b) Phillips Curve, 1948–2002 |
|
|
|
|
F I G U R E 1 Phillips Curve in the United States
Although the Phillips curve relationship worked fairly well from 1948 to 1969, after this period it appeared to shift upward, as is clear from panel (b). Looking at the whole period after World War II, there is no apparent trade-off between unemployment and inflation.
Source: Economic Report of the President. http://w3.access.gpo.gov/usbudget/
2 Because workers normally become more productive over time as a result of new technology and increases in physical capital, their real wages grow over time even when the economy is at the natural rate of unemployment. To reflect this, the Phillips curve should include a term that reflects the growth in real wages due to higher worker productivity. We have left this term out of the equation in the text, because higher productivity that results in higher real wages will not cause the aggregate supply curve to shift. If, for example, workers become 3 percent more productive every year and their real wages grow at 3 percent per year, the effective cost of workers to the firm (called unit labor costs) remains unchanged, and the aggregate supply curve does not shift. Thus the w/w term in the Phillips curve is more accurately thought of as the change in the unit labor costs.
3 This criticism of the Phillips curve was outlined in Milton FriedmanÕs famous presidential address to the American economic Association: Milton Friedman, ÒThe Role of Monetary Policy,Ó American Economic Review 58 (1968): 1Ð17.
Aggregate Supply and The Phillips Curve |
3 |
real wages, not nominal wages; they are concerned with the wage adjusted for any expected increase in the price levelÑthat is, they look at the rate of change of wages minus expected inflation. When unemployment is high relative to the natural rate, real (not nominal) wages should fall ( w/w e 0); when unemployment is low relative to the natural state, real wages should rise ( w/w e 0). The Phillips curve relationship thus needs to be modified by replacing w/w by w/w e. This results in an expectations-augmented Phillips curve, expressed as:
w/w e h(U U |
) or |
w/w h(U U ) e |
n |
|
n |
The expectations-augmented Phillips curve implies that as expected inflation rises, nominal wages will be increased to prevent real wages from falling, and the Phillips curve will shift upward. The resulting rise in production costs will then shift the aggregate supply curve leftward. The conclusion from FriedmanÕs modification of the Phillips curve is therefore that the higher inflation is expected to be, the larger the leftward shift in the aggregate supply curve; this conclusion is built into the analysis of the aggregate supply curve in the chapter.
FriedmanÕs modifications of the Phillips curve analysis was remarkably clairvoyant: As inflation increased in the late 1960s, the Phillips curve did indeed begin to shift upward, as we can see from panel (b). An important feature of panel (b) is that a trade-off between unemployment and wage inflation is no longer apparent; there is no clear-cut relationship between unemployment and wage inflationÑa high rate of wage inflation does not mean that unemployment is low, nor does a low rate of wage inflation mean that unemployment is high. This is exactly what the expectations-aug- mented Phillips curve predicts: A rate of unemployment permanently below the natural rate of unemployment cannot be ÒboughtÓ by accepting a higher rate of inflation because no long-run trade-off between unemployment and wage inflation exists.4
A further refinement of the concept of aggregate supply came from research by Milton Friedman, Edmund Phelps, and Robert Lucas, who explored the implications of the expectations-augmented Phillips curve for the behavior of unemployment. Solving the expectations-augmented Phillips curve for U leads to the following expression:
U Un ( w/w e)/h
Because wage inflation and price inflation are closely tied to each other, can be substituted for w/w in this expression to obtain:
U Un ( e)/h
4 This prediction can be derived from the expectations-augmented Phillips curve as follows. When wage inflation is held at a constant level, inflation and expected inflation will eventually equal wage inflation. Thus in the long run, e w/w. Substituting the long-run value of e into the expectations-augmented Phillips curve gives:
w/w h(U Un) w/w
Subtracting w/w from both sides of the equation gives 0 h(U Un), which implies that U Un. This tells us that in the long run, for any level of wage inflation, unemployment will settle to its natural rate level; hence the long-run Phillips curve is vertical, and there is no long-run trade-off between unemployment and wage inflation.
4 Appendix to Chapter 25
This expression, often referred to as Lucas supply function, indicates that deviations of unemployment and aggregate output from the natural rate levels respond to unanticipated inflation (actual inflation minus expected inflation, e). When inflation is greater than anticipated, unemployment will be below the natural rate (and aggregate output above the natural rate). When inflation is below its anticipated value, unemployment will rise above the natural rate level. The conclusion from this view of aggregate supply is that only unanticipated policy can cause deviations from the natural rate of unemployment and output. The implications of this view are explored in detail in Chapter 28.