Phillips Curve
Adaptive vs. Rational Expectations
Policy Impotency Hypothesis
Ricardian Equivalence


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


Both models imply:
Y

Y
  (
P

EP
) agg. output natural rate of output a positive parameter expected price level actual price level

Other things equal, Y and P are positively related, so the SRAS curve is upward-sloping.


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 )

 An individual firm’s desired price is: p

P

(

Y ) where a > 0.
Suppose two types of firms:
firms with flexible prices, set prices as above
firms with sticky prices, must set their price before they know how P and Y will turn out: p

EP

(

EY )

p

EP

(

EY )

Assume sticky price firms expect that output will equal its natural rate. Then, p

EP

To derive the aggregate supply curve, first find an expression for the overall price level.
 s = fraction of firms with sticky prices.
Then, we can write the overall price level as…

P
 s EP ] ( 1 s P
 a Y

Y )] price set by sticky price firms price set by flexible price firms

Subtract (1
 s ) P from both sides: sP

[ ] ( 1 s a Y

Y )]

Divide both sides by s :
P

EP

( 1
 s s a
( Y

Y )

P

EP

( 1
 s a
( Y

Y ) s

High EP

High P
If firms expect high prices, then firms that must set prices in advance will set them high.
Other firms respond by setting high prices.

High Y

High P
When income is high, the demand for goods is high.
Firms with flexible prices set high prices.
The greater the fraction of flexible price firms, the smaller is s and the bigger is the effect of

Y on P .

P

EP

( 1
 s a
( Y

Y ) s

Finally, derive AS equation by solving for Y :
Y

Y
 
( P

EP ), where
  s
( 1
 s a

0

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 .

P
E
= 1
P
B
= 1
P
D
= 1

Suppose P rises but EP does not.

Supplier thinks her relative price has risen, so she produces more.

With many producers thinking this way,
Y will rise whenever P rises above EP .
P
C
= 1

P

EP
P

EP
P

EP
P LRAS
Y Y
 (
P

EP
)
Y
SRAS
Y
Both models of agg. supply imply the relationship summarized by the SRAS curve & equation.

Suppose a positive
AD shock moves output above its natural rate and
P above the level people had expected.
SRAS equation:
P
3

EP
3
P
P
2
P
1

EP
1
Over time,
EP
2

EP rises,
SRAS shifts up, and output returns to its natural rate.
Y Y
LRAS
Y
3
 Y
1

Y

Y
2
(
P

EP
)
SRAS
2
SRAS
1
AD
2
AD
1
Y

Inflation Rate
5.0
4.5
4.0
3.5
3.0
2.5
2.0
1.5
1.0
0.5
0.0
3.0
3.5
1969
1968
1967
1966
4.0
The 1960s
4.5
1965
5.0
1964
1960 1963
1962
5.5
1961
6.0
6.5
7.0
Unemployment Rate

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.

SRAS: Y

Y
 
( P

EP )
Phillips curve:
 
E
   ( u
 u n )  

SRAS curve:
Output is related to unexpected movements in the price level.

Phillips curve:
Unemployment is related to unexpected movements in the inflation rate.


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
  
 1
  ( u
 u n )  

  
 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.

 
E
   ( u
 u n )  
In the short run, policymakers face a tradeoff between
 and u .
E
 


1
The short-run
Phillips curve u u n

Inflation Rate
13.0
12.0
11.0
6.0
5.0
4.0
3.0
10.0
9.0
8.0
7.0
The 1970s
1970
1973
1974
1979
1972
1978
1971 1977
1976
1975
2.0
1.0
0.0
3.0
3.5
4.0
4.5
5.0
5.5
6.0
6.5
7.0
7.5
8.0
8.5
9.0
Unemployment Rate

Inflation Rate
15.0
14.0
The 1980s
1980
13.0
12.0
11.0
10.0
1981
9.0
8.0
7.0
6.0
1982
5.0
4.0
3.0
2.0
1989
1988
1987
1986
1984
1985
1983
1.0
0.0
3.0
3.5
4.0
4.5
5.0
5.5
6.0
6.5
7.0
7.5
8.0
8.5
9.0
9.5
10.0 10.5
Unemployment Rate

Inflation Rate
6.0
The 1990s
1990
1991
5.0
4.0
1993
3.0
2.0
1.0
1999 1998
1997
1996
1995
1994
1992
0.0
3.0
3.5
4.0
4.5
5.0
5.5
6.0
6.5
7.0
7.5
8.0
Unemployment Rate

Inflation Rate
5.0
4.5
4.0
3.5
3.0
2.5
2.0
1.5
1.0
0.5
0.0
3.0
3.5
2000
The 2000s
2006
2001
2005
2007
2004
2008
2003
2002
4.0
4.5
5.0
5.5
6.0
Unemployment Rate
6.5

Inflation Rate
15.0
14.0
13.0
12.0
11.0
10.0
5.0
4.0
3.0
2.0
9.0
8.0
7.0
6.0
1.0
0.0
3
1974
1979
1978
1976
1969
2006
1968
1967
2000
1966
2007
1999 1998
1965
1970
1991
1990 1971
1989
1988
2008
1972
1964
2004
2002
2003
1994
1987
1962
1961
1977
1986
1993
1985
1984
1992
4 5 6 7
1980
1981
8
1975
1982
1983
9 10 11
Unemployment Rate

 
E
   ( u
 u n )  
People adjust their expectations over time, so the tradeoff only holds in the short run.
E

2
 
E

1
 
E.g
., an increase in E
 shifts the short-run P.C. upward.
 u n u


To reduce inflation, policymakers can contract agg. demand, causing unemployment to rise above the natural rate.

The sacrifice ratio measures the percentage of a year’s real GDP that must be foregone to reduce inflation by 1 percentage point.

A typical estimate of the ratio is 5.


Example: To reduce inflation from 6 to 2 percent, must sacrifice 20 percent of one year’s GDP:
GDP loss = (inflation reduction) x (sacrifice ratio)
= 4 x 5

This loss could be incurred in one year or spread over several, e.g
., 5% loss for each of four years.

The cost of disinflation is lost GDP.
One could use Okun’s law to translate this cost into unemployment.

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.


Proponents of rational expectations believe that the sacrifice ratio may be very small:

Suppose u = u n 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 .


1981:

= 9.7%
1985:

= 3.0%
Total disinflation = 6.7% year u u n
1982 9.5% 6.0%
1983 9.5
6.0
1984
1985
7.4
7.1
6.0
6.0
u
 u n
3.5%
3.5
1.4
1.1
Total 9.5%


From previous slide: Inflation fell by 6.7%, total cyclical unemployment was 9.5%.
 Okun’s law:
1% of unemployment = 2% of lost output.

So, 9.5% cyclical unemployment
= 19.0% of a year’s real GDP.

Sacrifice ratio = (lost GDP)/(total disinflation)
= 19/6.7 = 2.8
percentage points of GDP were lost for each 1 percentage point reduction in inflation.

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).


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
July 1953
Aug 1957
April 1960
December 1969
November 1973
January 1980
July 1981
July 1990
March 2001 trough
May 1954
April 1958
February 1961
November 1970
March 1975
July 1980
November 1982
March 1991
November 2001 increase in # of unemployed persons
(millions)
2.11
2.27
1.21
2.01
3.58
1.68
4.08
1.67
1.50
Increase from 12/2007 thru 6/2009: 7.2 million!!!


Recessions cause economic hardship for millions of people.

The Employment Act of 1946:
“It is the continuing policy and responsibility of the
Federal Government to…promote full employment and production.”

The model of aggregate demand and supply
(Chaps. 9-13) shows how fiscal and monetary policy can respond to shocks and stabilize the economy.

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.

 definition: policies that stimulate or depress the economy when necessary without any deliberate policy change.

Designed to reduce the lags associated with stabilization policy.

Examples:
 income tax
 unemployment insurance
 welfare

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
15
10
5
0
-5
-10 source of LEI data:
The Conference Board
-15
1990 1992 1994 1996 1998 2000 2002
Leading Economic Indicators
Real GDP

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

Mistakes forecasting the 1982 recession

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.


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.


Prediction (based on past experience):
An increase in the money growth rate will reduce unemployment.

The Lucas critique points out that increasing the money growth rate may raise expected inflation, in which case unemployment would not necessarily fall.

Looking at recent history does not clearly answer
Question 1:
 It’s hard to identify shocks in the data.
 It’s hard to tell how outcomes would have been different had actual policies not been used.
The Great Moderation?


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.

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
 def: A 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.

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.

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.

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.

a.
Constant money supply growth rate
 Advocated by monetarists.
 Stabilizes aggregate demand only if velocity is stable.

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.

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.


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.

index of central bank independence

 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
Japan
Italy
Greece
Belgium
U.S.A.
France
Portugal
Germany
Canada
Austria
Gov Debt
(% of GDP)
173
113
101
92
73
73
71
65
63
63
Country
U.K.
Netherlands
Norway
Sweden
Spain
Finland
Ireland
Korea
Denmark
Australia
Gov Debt
(% of GDP)
59
55
46
45
44
40
33
33
28
14

1,2
1,0
WW2
0,8
0,6
Revolutionary
War
Civil War
0,4
WW1
Iraq
War
0,2
0,0
1790 1810 1830 1850 1870 1890 1910 1930 1950 1970 1990 2010

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

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 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…

Percent of U.S. population age 65+
Percent of pop.
23
20
17
14
11
8
5 actual projected

U.S. government spending on Medicare and
Social Security
Percent of GDP
8
6
4
2
0

CBO projected U.S. federal govt debt in two scenarios
300
250
200
150
100 pessimistic scenario
50 optimistic scenario
0
2005 2010 2015 2020 2025 2030 2035 2040 2045 2050

1.
Inflation
2.
Capital assets
3.
Uncounted liabilities
4.
The business cycle

MEASUREMENT PROBLEM 1:

Suppose the real debt is constant, which implies a zero real deficit.

In this case, the nominal debt D grows at the rate of inflation:

D / D =
 or

D =

D

The reported deficit (nominal) is

D even though the real deficit is zero.

Hence, should subtract

D from the reported deficit to correct for inflation.

MEASUREMENT PROBLEM 1:

Correcting the deficit for inflation can make a huge difference, especially when inflation is high.

Example: In 1979, nominal deficit = $28 billion inflation = 8.6% debt = $495 billion

D = 0.086

$495b = $43b real deficit = $28b

$43b = $15b surplus

MEASUREMENT PROBLEM 2:

Currently, deficit = change in debt

Better, capital budgeting: deficit = (change in debt)

(change in assets)

EX: Suppose govt sells an office building and uses the proceeds to pay down the debt.
 under current system, deficit would fall
 under capital budgeting, deficit unchanged, because fall in debt is offset by a fall in assets.

Problem w/ cap budgeting: Determining which govt expenditures count as capital expenditures.

MEASUREMENT PROBLEM 3:

Current measure of deficit omits important liabilities of the government:
 future pension payments owed to current govt workers
 future Social Security payments
 contingent liabilities, e.g
., covering federally insured deposits when banks fail
(Hard to attach a dollar value to contingent liabilities, due to inherent uncertainty.)

MEASUREMENT PROBLEM 4:

The deficit varies over the business cycle due to automatic stabilizers (unemployment insurance, the income tax system).

These are not measurement errors, but do make it harder to judge fiscal policy stance.

E.g.
, is an observed increase in deficit due to a downturn or an expansionary shift in fiscal policy?

MEASUREMENT PROBLEM 4:

Solution: cyclically adjusted budget deficit
(aka “full-employment deficit”) – based on estimates of what govt spending & revenues would be if economy were at the natural rates of output & unemployment.

3
2
1
0
-1
-2 actual cyclicallyadjusted
-3
-4
-5
-6
1960 1965 1970 1975 1980 1985 1990 1995 2000 2005 2010

Consider a tax cut with corresponding increase in the government debt.
Two viewpoints:
1.
Traditional view
2.
Ricardian view


Short run:

Y ,
 u

Long run:

Y and u back at their natural rates
 closed economy:
 r ,
 I
Crowding Out

 due to David Ricardo (1820), more recently advanced by Robert Barro

According to Ricardian equivalence , a debt-financed tax cut has no effect on consumption, national saving, the real interest rate, investment, net exports, or real GDP, even in the short run.


Consumers are forward-looking, know that a debt-financed tax cut today implies an increase in future taxes that is equal – in present value – to the tax cut.

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.

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.

Early 1980s:
Reagan tax cuts increased deficit.
National saving fell, real interest rate rose
1992:
Income tax withholding reduced to stimulate economy.
 This delayed taxes but didn’t make consumers better off.

Almost half of consumers increased consumption.


Proponents of R.E. argue that the Reagan tax cuts did not provide a fair test of R.E.

Consumers may have expected the debt to be repaid with future spending cuts instead of future tax hikes.

Private saving may have fallen for reasons other than the tax cut, such as optimism about the economy.

Because the data is subject to different interpretations, both views of govt debt survive.

OTHER PERSPECTIVES:

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:
“Fiscal policy is not made 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
 since future taxpayers cannot participate in the decision process, their interests may not be taken into account

This is another reason for the proposals for a balanced budget amendment

OTHER PERSPECTIVES:

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)

In the sticky-price model, the relationship between output and the price level depends on: a) The target real wage rate b) The target nominal wage rate c) The proportion of firms with flexible prices d) The implicit agreements between workers and firms
31%
54%
15%
0% b) a) c) d)

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) Exceeds the b) Falls below the c) Equals the d) Moves to a different
54%
38% a) b)
8% c)
0% d)

The classical dichotomy breaks down for a Phillips curve, which shows the relationship between a nominal variable,
_____, and a real variable, _____.
a) Output; prices b) Money; output c) Inflation; unemployment d) Unemployment; inflation
85%
15%
0% a)
0% b) c) d)

According to the natural rate hypothesis, fluctuations in aggregate demand affect output in: a) Both the short run and long run b) Only in the short run c) Only in the long run d) In neither the short run nor the long run
100%
0% a) b)
0% c)
0% d)

The time between a shock to the economy and the policy actions responding to that shock is called the: a) Automatic stabilizer b) Time inconsistency of policy c) Inside lag d) Outside lag
62%
23%
15%
0% a) b) c) d)

The fact that traditional methods of policy evaluation do not take into account the impact of policy on expectations is known as: a) The political business cycle b) The Lucas critique c) Okun’s Law d) Stabilization policy
69%
15%
8% 8% a) b) c) d)

Policy is conducted by rule if policymakers: a) Announce in advance how policy will respond to various situations and commit themselves to following through on this announcement b) Are free to size up the situation case by case and choose whatever policy seems appropriate at the time c) Set policy according to election results d) Manipulate policy to ensure both low inflation and unemployment on election day
100% a)
0% b)
0% c)
0% d)

A monetary policy rule that targets nominal GDP would
_____ money growth when nominal GDP rises above the target and ______ money growth when nominal GDP falls below the target.
100% a) Reduce; raise b) Raise; reduce c) Reduce; reduce d) Raise; raise a)
0% b)
0% c)
0% d)

The amount by which government spending exceeds government revenues is called the _____, and the accumulation of past government borrowing is called the
____.
92% a) Deficit; debt b) Debt; deficit c) Devaluation; deflation d) Deflation; devaluation
8% a) b)
0% c)
0% d)

Assume that the nominal interest rate is 11 percent, the inflation rate is 8 percent, and government debt at the beginning of the year equals $4 trillion. By how much is the government budget deficit overstated as a result of inflation?
62% a) $0.12 trillion b) $0.32 trillion c) $0.44 trillion d) $0.80 trillion
38% a) b)
0% c)
0% d)

The debt of the US government is underreported in the view of many economists because are excluded except : all of the following liabilities a) Future pensions of government employees b) Debt owed to foreigners c) Future Social Security benefits d) Government guarantees of student loans
77%
23%
0% a) b) c)
0% d)

According to the traditional viewpoint, a tax cut without a cut in government spending: a) Stimulates consumer spending and reduces national saving b) Stimulates consumer spending and increases national saving c) Has no effect on consumer spending but reduces national saving d) Has no effect on consumer spending but reduces private saving
92% a)
8% b)
0% c)
0% d)

According to the theory of Ricardian equivalence, if consumers are forward-looking, they will view a tax cut that has no plans to reduce government spending as ______, so their consumption will ______.
a) Additional disposable income; increase b) Additional disposable income; remain unchanged c) A rescheduling of taxes into the future; increase d) A rescheduling of taxes into the future; remain unchanged
0% a)
8% b)
0% c)
92% d)