Transcript Controversial Issues About the Recession and Recovery

Controversial Issues About
the Recession and Recovery
Robert J. Gordon
Northwestern University and NBER
CIRET Conference, New York City
October 14, 2010
The Plan: From Long-Run
to Short-Run to Policy



The reasoning behind my pessimistic long-term US
growth forecast, recently summarized in Business
Week
Graphs on dimensions of the weak labor market.
The labor market is in worse condition than the
product market.
New research on the “Demise of Okun’s Law” and
the near-term division of output growth between
productivity, hours, and employment growth
– Why is the labor market in such bad shape compared to
the product market?

The policy debate: what more can monetary and
fiscal policy do?
The Pessimistic Long-run
Conclusion





Comparing 2007-2027 forecasts with 19872007 actual:
Output growth will slow from 2.9 to 2.4
Output per capita growth will slow from
1.74 to 1.4
That is the slowest growth of income per
capita “since George Washington”
Compare to 2.16 1929-2007 or 2.02 18912007
Growth in MFP vs. Ypc by
Time Interval, 1891-2027
Growth in MFP and Real GDP per capita, selected intervals, 1891-2027
3
2.5
Percent per Year
2
MFP
GDP/Pop
1.5
1
0.5
0
1891-1928
1928-1950
1950-1972
1972-1987
1987-2007
2007-2027
Components of Growth in Y/H,
1987-2007 vs. 2007-27
Components of Growth of Labor Productivity, Two Intervals
2.5
Percent per Year
2
1.5
1987-2007
2007-2027
1
0.5
0
Output/Hour
Capital Deepening
Labor Quality
MFP
From Y/H to Y/N, the Role
of Falling LFPR
Components of Output Growth, Two Intervals
3.5
3
Percent per Year
2.5
2
1987-2007
2007-2027
1.5
1
0.5
0
Output
Output/Hour
Output/Person
Hours
Population
Possible Further Room for
Pessimism


These projections are based on the
historical record of growth between years of
“normal” utilization (1987, 2007)
No allowance here for long-run “tainting”
effects of the current abysmal economy
– Loss of skills and human capital
– Years of low investment will increase the age of
the capital stock and reduce the growth of both
capital quantity and capital quality
Policy Prescriptions for
Long-Run Growth Problem




Slowdown reflects aging of population and
stagnation of educational attainment
Solve the first by immigration, particularly of
high-skilled people
Work on the second by better governmentrun student loan programs and direct
measures to address the rising relative price
of college education (“higher education cost
disease”)
Stimulate demand to avert long-run supply
sclerosis
Next We’ll Look at Graphs of
Raw Numbers for Current
US Labor Market

Now We’re Looking at
– Magnitudes: How Severe Is This Episode?
– Timing: Do Labor Market Indicators Change at
the Same Time as Output (Real GDP)?
– Which Measures Are the Most Different from
1980-82?

We Consider 1980-82 as a Single Recession
and also as two back-to-back recessions
– (Jan-July 1980 and Jul 81 to Nov 82)
Output Gap vs. Gap in
Aggregate Hours of Work
Conclusion to this point

Comparing the 9.6 level of U rate now to 10.8 in
Nov & Dec 1982 is misleading
– U rate in July 81 or even Jan 80 started higher
– Overall increase in 2007-09 is greater
– Much more incidence this time of long-term
unemployment and forced part-time


The emergence of long-term unemployment: is the
US becoming more like Europe’s two decades 19852005?
Stylized fact: If the empl/pop ratio was the same
today as in 2000, there would be 14 million more
jobs (9 million from lower unemployment rate, 5
million from higher LFPR) “New Normal” for LFPR?
For U Rate?
Documenting and Explaining
the Change in Cyclical
Labor-market Behavior

Documenting
– A new approach to disentangling trends and
cycles

Use of “outside information” from inflation equation to
determine the unemployment rate gap
– A new approach to data


Total Economy not NFPB Sector
Conventional vs. Unconventional Measures
– Initial finding as reported before: hours gap >
output gap in 2008-09, the reverse of 1980-82
The Output Identity: Simple
Version and Conventional
Version
Y H E L
Y
   N
H E L N
P
Y
P
P
P
H
Y
H
E
E
L
 P P  H 
 N
H
E
E
L N
Conventional Compared to
Unconventional Identity
P
Y
P
P
P
H
Y
H
E
E
L
 P P  H 
 N
H
E
E
L N
I
H
H
Y
H
E
L
Y  H  H 
 N
L N
H
E
I
Kalman Trends, Conv vs.
Unconv Output
Aggregate Hours
Total Economy Labor
Productivity (Y/H)
Output Gap vs. Gap in
Aggregate Hours of Work
Close-up for Period since
1986
Regression Analysis

Changes in hours gap regressed on
–
–
–
–
Own lags
Current and lagged changes in output gap
Error-correction term (lagged level of hours gap)
“End of expansion” dummy variables
(capture overhiring end of expansion,
underhiring beginning of recovery)
“Early recovery productivity bubble”
Long-run Coefficients: The
“Demise of Okun’s Law”
Explanations Offered
in My Research



The “Disposable Worker” Hypothesis
Similar sources as rising US inequality
Increased market power of managers and
highly paid professionals
– Increased share of executive incomes coming
from stock options

Reduced market power of workers due to:
– Declining unions, declining real minimum wage,
low-skilled immigration, and imports
Implications for the
Unemployment Rate




Translate forecast growth in hours into the
hours gap
Regression coefficients imply roughly 60%
of an improvement in hours gap flows into
employment rate gap
Optimistic path, output gap shrinks 0.8
points per year (3.3 vs. 2.5)
Pessimistic path, output gap stays where it
is forever (2.5 percent growth forever)
Implied Unemployment
Paths
Unemployment Rate: Actual and Projected
Optimistic and Pessimistic Outlooks
12.00%
10.00%
8.00%
6.00%
Optimistic Projection
4.00%
Pessimistic Projection
Actual
2.00%
0.00%
1986
1992
1998
2004
Year
2010
2016
Reasons This May Be
Too Optimistic

Why does pessimistic U Rate decline even though output gap
is fixed?
– Error-correction term implies mean reversion both of negative
hours gap and positive productivity gap

Division of predict real GDP growth between productivity and
hours: mean reversion may be delayed
– Much publicized “reluctance to hire”
– Matched with “reluctance to invest”
– Firms are not expecting the 3.3 growth scenario


If the 3.3 growth rate actually happened, short-term outcome
would be more productivity growth and less hours and
employment growth
But will that output scenario happen?
Reasons To Be Skeptical
of Optimistic Scenario



Reinhart-Rogoff. Recoveries after
financial crises are slower than after
garden-variety recessions
Ineffectiveness of monetary policy
Political paralysis just as Obama
stimulus is about to be withdrawn
The Fed is Out of
Ammunition


Textbook IS-LM model still taught in
intermediate undergrad macro
Monetary policy is ineffective if:
– Horizontal LM curve
– Vertical IS curve


The Fed now is plagued by both
Why should QE2 work since QE1
didn’t?
The Fed Can’t Control the
Cost of Business Borrowing
Fundamental Causes of
Weak Recovery (Vertical IS)

Consumption
– Collapse of Household Net Worth
– Record-high indebtedness

Residential Construction
–
–
–
–
–
Foreclosures and Under-water Mortgages
People walk away from under-water
Their credit is tainted for years
Their houses add to supply but not to demand
My mortgage broker’s story, 3 vs. 80
Consumption Problem:
Household Balance Sheet
The Twin Peaks of Household Net Worth
Percentage of Disposable Personal Income
800
750
700
650
Total Assets
600
550
500
Net Worth
450
400
100
50
0
Total Liabilities
-50
-100
-150
1970
1975
1980
1985
1990
1995
Sources : Federal Reserve Board Flow of Funds Accounts and Bureau of Economic Analysis NIPA Tables . Details in Appendix C-4.
2000
2005
2010
Housing Starts Used to be a
Leading Indicator, but Not
Any More
Quarterly Housing Starts, 1970-2010
2500
Quarterly Housing Starts (thousands)
2000
1500
1000
500
0
1970
1975
1980
U.S. Census Bureau Manufacturing, Mining and Construction Statistics
1985
1990
1995
2000
2005
2010
Where is Fiscal Policy?



Krugman NYT Monday: it wasn’t tried.
Look at government employment
excluding census workers
The Obama stimulus was too small
and too much was wasted on tax cuts
and capital-intensive infrastructure
spending
What Ended the Great Depression?
Chart Extends 1929-41 Quarterly
30
25
Spending/GDP Ratio
20
Transfer Payments/GDP
15
10
Federal Spending/GDP
5
State and Local Spending/GDP
0
1929
1931
1933
1935
1937
1939
1941
How Does the Obama
Stimulus Measure Up?
30
25
Spending/GDP Ratio
Transfer Payments/GDP
20
Federal Spending/GDP
15
10
State and Local Spending/GDP
5
0
1980
1985
1990
1995
2000
2005
2010
The Current Debate,
What About Inside vs.
Outside Govt Debt?





Monetary Policy is parralized but still can
support fiscal expansion
A fiscal stimulus does not raise outside debt
(“held by the public”) if the Fed buys the
bonds
This is the classic Milton Friedman
“heliocopter drop” of money
In today’s terminology, QE2 supports a
fiscal stimulus
But where is the political will?
Conclusions





A Real GDP path of 3.3 brings the U rate down to 6
percent by early 2014
Reasons this may be too optimistic
Pessimistic path leaves U rate at 8 percent in 2016,
Europe déjà vu 1985-2005
Model has mean-reversion built in, underhiring in
2009-10 will be reversed, productivity bubble will
be reversed.
Basic problems: Reinhart-Rogoff, ineffective
monetary policy, lack of political will on fiscal policy,
about to become much worse after the elections