Unemployment, Inflation and Real Wages in Depression

Download Report

Transcript Unemployment, Inflation and Real Wages in Depression

Temin and Wigmore: The End of One Big Deflation
• US recovery from the Great Depression in the second
quarter of 1933—Roovelt’s Inauguration
– Sargent’s (1983) regime change paradigm
• Dollar devaluation was key to recovery...signaled change in
policy regime and reversed expectations of deflation
– Romer maintains that money growth mattered
• Fiscal stimulus was weak and statistically insignificant
Hoover: stuck in gold standard mindset
FDR: take action
• That action was devaluation...taken soon after inauguration
– Signaled the abandonment of the gold standard
 expansionary effects on American industry
• Stock market as index of expectations
– Stock prices rose because of expected inflation
“change in expectations, therefore, stimulated business investment
and expenditures on consumer durables, not consumption”
• Rise in the demand for automobiles encouraged a rise in auto
production, steel production, and industrial production
• Grain and cotton prices rose as the value of the dollar fell
 farmers had higher incomes
Temin and Wigmore conclude
• If Hoover had done what FDR did, the economy would have
recovered earlier
• But would economy recover on its own?
Natural rate hypothesis  Inherent stability
Review Phillips Curve/Natural Rate models
Bernanke and Parkinson: Unemployment, Inflation, and Wages in the
American Depression: Are There Lessons for Europe (AER, May 1989)
Puzzles: US in the 1930s/Europe in the 1980s/US today
• Persistence of high unemployment
– Wither self-correction? Is economy inherently stable?
• Insensitivity of inflation to high unemployment (a “floating NAIRU”?)
• Increasing real wage despite high unemployment (not now)
But note:
• High growth rates (in manufacturing sector) during the 1933-37 and
1938-40 recoveries
 strong self-correction consistent with natural rate
Error correction model (in logs) for manufacturing employment
1924:2 – 1941:4
Δnt = constant + a(L)Δnt + b(n*t-1 – nt-1) + c0(πt – πet ) + c0(πt-1 – πet-1 ) + et
In words
Manufacturing employment growth corrected for autocorrelation
responds to inflation surprises and closes the gap between “normal”
and actual manufacturing employment in prior quarter.
n*t = (Fraction of labor force employed in mfg in 1929:1)
x (Labor force in quarter t) = “normal” employment
πt – πet = inflation surprise
= residual when inflation estimated using lagged inflation
and commercial paper interest rates
Find
c > 0, consistent with Lucas-Rapping supply curve
b = .15, consistent with self-correction, homeostasis hypothesis
n* - n half-life = 3 quarters
Error correction Phillips Curve
Δnt = constant + a(L)Δnt + b(n*t-1 – nt-1) + c0(πt – πet ) + c0(πt-1 – πet-1 ) + et
Find
c > 0, consistent with Lucas-Rapping supply curve
b = .15, consistent with self-correction, homeostasis hypothesis
n* - n half-life = 3 quarters
Critique: Recovery was due to aggressive New Deal policies
Response: New Deal “cleared the way for recovery” but did not drive it
Critique: Other sectors may have been less resilient than manufacturing
Response: No data to test assertion
Bernanke and Parkinson conclude/suggest that the economy is
inherently stable
 self-correcting mechanisms work...not stuck in “trap”
• Since employment responded to inflation, monetary reflation—
increased money growth—may have assisted recovery (per Romer)
Puzzle: High and Increasing Real Wage in Depression
• Real wage increased not just when price level fell
New Deal transition to an “efficiency wage”?
• Shorter work-week (work sharing) but weekly reservation wage
• Strong unions reinforced by New Deal legislation
• Improved working conditions
• Higher wages
• Strong productivity growth
Bernanke and Parkinson
• Efficiency wage explains productivity growth  supply side growth
• High real wage spurred spending and output  demand side growth