AD/AS Models and
Macro Policy Debates
Phillips Curve
Adaptive vs. Rational Expectations
Policy Impotency Hypothesis
Ricardian Equivalence
Introduction
 In previous models, we assumed the price level
P was “stuck” in the short run.
 This implies a horizontal SRAS curve.
 Now, we consider two prominent models of
aggregate supply in the short run:
 Sticky-price model
 Imperfect-information model
Introduction
 Both models imply:
Y  Y   (P  EP)
expected
price level
agg.
output
natural rate
of output
a positive
parameter
actual
price level
 Other things equal, Y and P are positively
related, so the SRAS curve is upward-sloping.
The sticky-price model
 Reasons for sticky prices:
 long-term contracts between firms and
customers
 menu costs
 firms not wishing to annoy customers with
frequent price changes
 Assumption:
 Firms set their own prices
(i.e., firms have some market power)
The imperfect-information model
Assumptions:
 All wages and prices are perfectly flexible,
so that all markets clear.
 Each supplier produces one good, consumes many goods.
 Each supplier knows the nominal price of the good she
produces, but does not know the overall price level.
 Supply of each good depends on its relative price:
the nominal price of the good divided by the overall price level.
 Supplier does not know price level at the time she
makes her production decision, so uses EP.
Lucas Island Metaphor
PB = 1
PA = 1
2
PD = 1
PE = 1
PC = 1
 Suppose P rises but EP
does not.
 Supplier A thinks her
relative price has risen,
so she produces more.
 With many producers
thinking this way,
Y will rise whenever P
rises above EP.
Summary & implications
P
LRAS
Y  Y   (P  EP)
P  EP
SRAS
P  EP
P  EP
Y
Y
Both models
of agg. supply
imply the
relationship
summarized
by the SRAS
curve &
equation.
Summary & implications
SRAS equation: Y  Y   (P  EP)
Suppose a positive
AD shock moves
SRAS2
P
LRAS
output above its
natural rate and
SRAS1
P above the level
people had
P3  EP3
expected.
P2
Over time,
EP2  P1  EP1
EP rises,
SRAS shifts up,
and output returns
to its natural rate.
AD2
AD1
Y
Y3  Y1  Y
Y2
Inflation Rate
5.0
The 1960s
1969
4.5
4.0
1968
3.5
1967
3.0
2.5
1966
2.0
1964
1.5
1961
1960
1965
1.0
1963
1962
0.5
0.0
3.0
3.5
4.0
4.5
5.0
5.5
6.0
6.5
7.0
Unemployment Rate
Inflation, Unemployment,
and the Phillips Curve
The Phillips curve states that  depends on
 expected inflation, E.
 cyclical unemployment: the deviation of the
actual rate of unemployment from the natural rate
 supply shocks,  (Greek letter “nu”)
  E   (u  u )  
n
where  > 0 is an exogenous constant.
Adaptive expectations
 Adaptive expectations: an approach that
assumes people form their expectations of future
inflation based on recently observed inflation.
 A simple version:
Expected inflation = last year’s actual inflation
E   1
 Then, P.C. becomes
n
   1   (u  u )  
Inflation inertia
   1   (u  u )  
n
In this form, the Phillips curve implies that
inflation has inertia:
 In the absence of supply shocks or
cyclical unemployment, inflation will
continue indefinitely at its current rate.
 Past inflation influences expectations of
current inflation, which in turn influences
the wages & prices that people set.
Two causes of rising & falling inflation
   1   (u  u )  
n
 cost-push inflation:
inflation resulting from supply shocks
Adverse supply shocks typically raise production
costs and induce firms to raise prices,
“pushing” inflation up.
 demand-pull inflation:
inflation resulting from demand shocks
Positive shocks to aggregate demand cause
unemployment to fall below its natural rate,
which “pulls” the inflation rate up.
Graphing the Phillips curve
In the short
run, policymakers
face a tradeoff
between  and u.

  E   (u  un )  

1
The short-run
Phillips curve
E  
u
n
u
Inflation Rate
15.0
1980
14.0
13.0
1981
12.0
1975
11.0
10.0
1974
1979
9.0
1982
8.0
1978
7.0
1976
1970
6.0
1990 1971
5.0
1969
4.0
2006
2001
1973
1968
3.0
1967
2000
2.0
1966
19972005
1972
19961995
2007
1999 1998
1977
1986
1985
1993
2003
1994
1984
1983
1992
2004
1964
1965
1.0
1989
2008
1988
1991
1963
1987
1960
2002
1961
1962
0.0
3
4
5
6
7
8
9
10
11
Unemployment Rate
Shifting the Phillips curve
People adjust
their
expectations
over time,
so the tradeoff
only holds in
the short run.

  E   (u  un )  
E 2  
E1  
E.g., an increase
in E shifts the
short-run P.C.
upward.
u
n
u
Rational expectations
Ways of modeling the formation of expectations:
 adaptive expectations:
People base their expectations of future inflation
on recently observed inflation.
 rational expectations:
People base their expectations on all available
information, including information about current
and prospective future policies.
Painless disinflation?
 Proponents of rational expectations believe
that the sacrifice ratio may be very small:
 Suppose u = un and  = E = 6%,
and suppose the Fed announces that it will
do whatever is necessary to reduce inflation
from 6 to 2 percent as soon as possible.
 If the announcement is credible,
then E will fall, perhaps by the full 4 points.
 Then,  can fall without an increase in u.
The natural rate hypothesis
Our analysis of the costs of disinflation, and of
economic fluctuations in the preceding chapters,
is based on the natural rate hypothesis:
Changes in aggregate demand affect output
and employment only in the short run.
In the long run, the economy returns to
the levels of output, employment,
and unemployment described by
the classical model (Chaps. 3-8).
Both models of aggregate supply discussed in Chapter 13
imply that if the price level is higher than expected, then
output ______ natural rate of output.
a)
b)
c)
d)
Exceeds the
Falls below the
Equals the
Moves to a different
0%
a)
0%
0%
0%
b)
c)
d)
The classical dichotomy breaks down for a Phillips curve,
which shows the relationship between a nominal variable,
_____, and a real variable, _____.
a)
b)
c)
d)
Output; prices
Money; output
Inflation; unemployment
Unemployment; inflation
0%
a)
0%
0%
0%
b)
c)
d)
According to the natural rate hypothesis,
fluctuations in aggregate demand affect output in:
a) Both the short run and
b)
c)
d)
long run
Only in the short run
Only in the long run
In neither the short run
nor the long run
0%
a)
0%
0%
0%
b)
c)
d)
Stabilization Policy
 Should policy be active or passive?
 Should policy be by rule or discretion?
Growth rate of U.S. real GDP
Percent
change
from 4
quarters
earlier
10
8
6
Average
growth
rate
4
2
0
-2
-4
1970
1975
1980
1985
1990
1995
2000
2005
2010
Increase in unemployment during recessions
peak
trough
increase in # of
unemployed persons
(millions)
July 1953
May 1954
2.11
Aug 1957
April 1958
2.27
April 1960
February 1961
1.21
December 1969
November 1970
2.01
November 1973
March 1975
3.58
January 1980
July 1980
1.68
July 1981
November 1982
4.08
July 1990
March 1991
1.67
March 2001
November 2001
1.50
December 2007
June 2009
6.14
Arguments for active policy
 Recessions cause economic hardship for millions
of people
 The Employment Act of 1946
 AD/AS model shows how fiscal and monetary
policy can respond to shocks and stabilize the
economy
Arguments against active policy
Policies act with long & variable lags, including:
inside lag:
the time between the shock and the policy response.
 takes time to recognize shock
 takes time to implement policy,
especially fiscal policy
outside lag:
the time it takes for policy to affect economy.
If conditions change before policy’s impact is felt,
the policy may destabilize the economy.
Automatic stabilizers
 Designed to reduce the lags associated with
stabilization policy.
 Examples:
 income tax
 unemployment insurance
 welfare
Forecasting the macroeconomy
Because policies act with lags, policymakers must
predict future conditions.
Two ways economists generate forecasts:
 Leading economic indicators
data series that fluctuate in advance of the
economy
 Macroeconometric models
The LEI index and real GDP, 1990s
annual percentage change
15
10
5
0
-5
-10
-15
1990
source of LEI data:
The Conference Board
1992
1994
1996
1998
2000
Leading Economic Indicators
Real GDP
2002
Forecasting the macroeconomy
Because policies act with lags, policymakers must
predict future conditions.
Two ways economists generate forecasts:
 Leading economic indicators
data series that fluctuate in advance of the
economy
 Macroeconometric models
Large-scale models with estimated parameters
that can be used to forecast the response of
endogenous variables to shocks and policies
Unemployment rate
Mistakes forecasting the 1982 recession
Forecasting the macroeconomy
Because policies act with lags, policymakers must
predict future conditions.
The preceding slides show that the
forecasts are often wrong.
This is one reason why some
economists oppose policy activism.
The Lucas critique
 Due to Robert Lucas
who won Nobel Prize in 1995 for rational
expectations.
 Forecasting the effects of policy changes has
often been done using models estimated with
historical data.
 Lucas pointed out that such predictions would not
be valid if the policy change alters expectations in
a way that changes the fundamental relationships
between variables.
Question 2:
Should policy be conducted by
rule or discretion?
Rules and discretion:
Basic concepts
 Policy conducted by rule:
Policymakers announce in advance how
policy will respond in various situations,
and commit themselves to following through.
 Policy conducted by discretion:
As events occur and circumstances change,
policymakers use their judgment and apply
whatever policies seem appropriate at the time.
Arguments for rules
1. Distrust of policymakers and the political process
 misinformed politicians
 politicians’ interests sometimes not the same as the
interests of society
2. The time inconsistency of discretionary policy
 Scenario in which policymakers have an incentive to

renege on a previously announced policy once
others have acted on that announcement
Destroys policymakers’ credibility, thereby reducing
effectiveness of their policies.
Examples of time inconsistency
1. To encourage investment,
govt announces it will not tax income from capital.
But once the factories are built,
govt reneges in order to raise more tax revenue.
Examples of time inconsistency
2. To reduce expected inflation,
the central bank announces it will tighten
monetary policy.
But faced with high unemployment,
the central bank may be tempted to cut interest
rates.
Examples of time inconsistency
3. Aid is given to poor countries contingent on fiscal
reforms.
The reforms do not occur, but aid is given
anyway, because the donor countries do not want
the poor countries’ citizens to starve.
Monetary policy rules
a. Constant money supply growth rate
 Advocated by monetarists.
 Stabilizes aggregate demand only if velocity
is stable.
Monetary policy rules
a. Constant money supply growth rate
b. Target growth rate of nominal GDP
 Automatically increase money growth
whenever nominal GDP grows slower than
targeted; decrease money growth when
nominal GDP growth exceeds target.
Monetary policy rules
a. Constant money supply growth rate
b. Target growth rate of nominal GDP
c. Target the inflation rate
 Automatically reduce money growth whenever
inflation rises above the target rate.
 Many countries’ central banks now practice
inflation targeting, but allow themselves a little
discretion.
Central bank independence
 A policy rule announced by central bank will
work only if the announcement is credible.
 Credibility depends in part on degree of
independence of central bank.
average inflation
Inflation and central bank independence
index of central bank independence
Government Debt and Budget Deficits
 The size of the U.S. government’s debt, and
how it compares to that of other countries.
 Problems with measuring the budget deficit.
 How does government debt affect the economy?
Deficit = G – T
Gov’t Debt = Σ Deficits
Indebtedness of the world’s governments
Country
Gov Debt
(% of GDP)
Country
Gov Debt
(% of GDP)
Japan
173
U.K.
59
Italy
113
Netherlands
55
Greece
101
Norway
46
Belgium
92
Sweden
45
U.S.A.
73
Spain
44
France
73
Finland
40
Portugal
71
Ireland
33
Germany
65
Korea
33
Canada
63
Denmark
28
Austria
63
Australia
14
Ratio of U.S. govt debt to GDP
1.2
1.0
WW2
0.8
0.6
Revolutionary
War
Civil War
Iraq
War
WW1
0.4
0.2
0.0
1790 1810 1830 1850 1870 1890 1910 1930 1950 1970 1990 2010
The U.S. experience in recent years
Early 1980s through early 1990s
 debt-GDP ratio: 25.5% in 1980, 48.9% in 1993
 due to Reagan tax cuts, increases in defense
spending & entitlements
Early 1990s through 2000
 $290b deficit in 1992, $236b surplus in 2000
 debt-GDP ratio fell to 32.5% in 2000
 due to rapid growth, stock market boom, tax
hikes
The U.S. experience in recent years
Early 2000s
 the return of huge deficits, due to Bush tax cuts,
2001 recession, Medicare expansion, Iraq war
The 2008-2009 recession
 fall in tax revenues
 huge spending increases (bailouts of financial
institutions and auto industry, stimulus package)
The troubling long-term fiscal outlook
 The U.S. population is aging.
 Health care costs are rising.
 Spending on entitlements like
Social Security and Medicare is
growing.
 Deficits and the debt are projected
to significantly increase…
U.S. population age 65+,
as percent of population age 20-64
40
Percent
of pop. 35
age
20-64 30
actual projected
25
20
15
10
2000
2005
2010
2015
2020
2025
2030
2035
U.S. government spending on Medicare
and Social Security, 1950-2009
Percent 8
of GDP
6
4
2
0
1950
1960
1970
1980
1990
2000
2010
Projected U.S. federal govt debt in two
scenarios, 2000-2035
200
“Alternative fiscal scenario”
incorporates widely-expected
changes to current law, such as
extension of Bush tax cuts
180
Percent of GDP
160
140
120
100
80
60
40
“Extended baseline scenario” –
assumes no changes to current law
Actual
20
0
2000
2005
2010
2015
2020
2025
2030
2035
Problems measuring the deficit
1. Inflation
2. Capital assets
3. Uncounted liabilities
4. The business cycle
Is the govt debt really a problem?
Consider a tax cut with corresponding increase
in the government debt.
Two viewpoints:
1. Traditional view
2. Ricardian view
The traditional view
 Short run: Y, u
 Long run:
 Y and u back at their natural rates
 closed economy: r, I
Crowding Out
The Ricardian view
 due to David Ricardo (1820),
more recently advanced by Robert Barro
 According to Ricardian equivalence,
a debt-financed tax cut has no effect on
C, S, r, I, and Y, even in the short run.
The logic of Ricardian Equivalence
 Consumers are forward-looking
 The tax cut does not make consumers better off, so
they do not increase consumption spending
Instead, they save the full tax cut in order to repay the
future tax liability
 Result: Private saving rises by the amount public
saving falls, leaving national saving unchanged
Problem Set #21
Problems with Ricardian Equivalence
 Myopia: Not all consumers think so far ahead,
some see the tax cut as a windfall.
 Borrowing constraints: Some consumers
cannot borrow enough to achieve their optimal
consumption, so they spend a tax cut.
 Future generations: If consumers expect that
the burden of repaying a tax cut will fall on future
generations, then a tax cut now makes them feel
better off, so they increase spending.
Evidence For/Against RE
 Reagan tax cuts (1980s)
 National saving fell, real interest rate rose
 Private saving may have fallen for reasons other than the tax
cut, such as optimism about the economy
 Consumers may have expected the debt to be repaid with
future spending cuts instead of future tax hikes
 Bush I withholding experiment (1992)
 Income tax withholding reduced to stimulate economy
 This delayed taxes but didn’t make consumers better off
 Almost half of consumers increased consumption
Because the data is subject to different
interpretations, both views of govt debt survive
OTHER PERSPECTIVES: Balanced budgets
vs. optimal fiscal policy
 Some politicians have proposed amending the
U.S. Constitution to require balanced federal
govt budget every year.
 Many economists reject this proposal, arguing
that deficit should be used to:
 stabilize output & employment
 smooth taxes in the face of fluctuating income
 redistribute income across generations when
appropriate
OTHER PERSPECTIVES:
Debt and politics
“Fiscal policy is made not by angels…”
– Greg Mankiw, p.487
 Some do not trust policymakers with deficit spending.
They argue that:
 policymakers do not worry about true costs of their
spending, since burden falls on future taxpayers
OTHER PERSPECTIVES:
Fiscal effects on monetary policy
 Govt deficits may be financed by printing money
 A high govt debt may be an incentive for
policymakers to create inflation (to reduce real
value of debt at expense of bond holders)