Transcript Diapositive 1

Theories and Methods of the Business
Cycle.
Part 1: Dynamic Stochastic General
Equilibrium Models
IV. Can RBC be saved?
Jean-Olivier HAIRAULT, Professeur à Paris I
Panthéon-Sorbonne et à l’Ecole d’Economie de Paris
(EEP)
IV. 1. Extending the RBC approach
 Two kinds of extensions: solving empirical puzzles and
investigating other issues.
 Some criticisms have been adressed into the RBC theory,
ie. models where productivity shocks are the main
perturbation of optimal fluctuations.
 The canonical model has been extended to international
fluctuations and to the study of particular historical
episodes like the 1930’s crisis.
IV. 2. Public spending shocks
w
E’
E
E’’
Ns, Nd
IV. 2. Public spending shocks
 Government shocks have wealth effects which alters labor
supply. The correlation between wages and hours is then
negative in case of government shocks.
 Taking into account both shocks leads to decrease this
correlation.
E’’
E
E’
IV. 2. Public spending shocks
IV. 2. Public spending shocks
 Business cycles properties with public expenditures shocks
 From kprdea2.m to kprdea2G.m
 The correlation between hours and wages is decreased but
remains too high. Public expenditures shocks have a
weaker impact on aggregates than technology shocks.
 Artificially decreasing the volatility of the technology shocks
could lead to a negative correlation.
IV. 2. Public spending shocks
 Only Technology Shocks N
 standard deviation 0.0081
Y
0.0142
C
0.0050
I
0.0523
P
0.0066
 relative st. dev.
0.5667
1.0000
0.3480
3.6733
0.4655
 Correlation
0.9745
1.0000
0.8910
0.9864
0.9623
0.6924
0.7972
0.6770
0.6924
 serial corr.
 correlation N-W
0.6750
0.8770
 +public spending shocks
 standard deviation 0.0069
0.0134
0.0062
0.0486
0.0074
 relative st. dev.
0.5187
1.0000
0.4634
3.6346
0.5519
 Correlation
0.9290
1.0000
0.8569
0.9816
0.9380
 serial correlation
0.6808
0.6920
0.7715
0.6798
0.6920
 correlation N-W
0.7441
IV. 3.Back to the Lucas critique
IV. 3.Back to the Lucas critique
IV. 3.Back to the Lucas critique
 RBC methodology allows us to take into account all the
implications of a change in public policy or in the environment of
private agents.
 The dynamics derive from the interaction between the
environment (shocks, public policies) and fundamentals
(Technology, preferences).
 Which public spending rule does better stabilize the economy?
 Volatility ? Which volatility?
 Welfare? What are the market failures in our economy?
 What are the size of business cycle costs? More on this issue in
course 6.
IV. 4. The indivisible labor hypothesis
IV. 4. The indivisible labor hypothesis
IV. 4. The indivisible labor hypothesis
IV. 4. The indivisible labor hypothesis
IV. 4. The indivisible labor hypothesis
Ns
Nd
IV. 4. The indivisible labor hypothesis
 Eta=1
N
 standard deviation 0.0081
Y
0.0142
C
0.0050
I
0.0523
P
0.0066
 relative st. dev.
0.5667
1.0000
0.3480
3.6733
0.4655
 Correlation
0.9745
1.0000
0.8910
0.9864
0.9623
0.6924
0.7972
0.6770
0.6924
 serial corr.
 correlation N-W
0.6750
0.8770
 Infinite elasticity
 standard deviation 0.0124
 relative st. Dev.
0.7270
0.0170
1.0000
0.0057
0.3329
0.0638
3.7442
0.0057
0.3329
 Correlation
0.9750
1.0000
0.8759
0.9865
0.8759
 serial correlation
0.6706
0.6883
0.8034
0.6727
0.6883
 correlation N-W
0.7472
IV. 5. Capital utilization
IV. 5. Capital utilization
IV. 5. Capital utilization
IV. 5. Capital utilization
IV. 5. Capital utilization
IV. 6. Labor Hoarding
 Taking into account the quality or intensity of worked hours
 labor intensity increases (decreases) in expansions (contractions).
 Firms prefer to adjust labor intensity rather employment.
Individual hours are assumed to be indivisible and employment
are submitted to adjustment costs. Employment gets
predetermined.
 Productivity cycle: labor productivity measured as
output/employment leads employment in the data.
 Correlation lagged and contemporanous productivityemployment: .11 0.37 0.47 0.47 0.34
IV. 6. Labor Hoarding
 Y = A F(K, X(eH))
 The Solow Residual is no longer a pure measure of
technology: variations of e must be eliminated.
 Naive Solow residual overestimates the standart deviation
of the technology innovation by 35%, as estimated by
Burnside, Christiano and Rebelo.
 Technology shocks are not volatile enough to replicate the
variance of output. Only 58%.
 Adding another real shock: public spending.
IV. 7. International Business Cycles
 Backus, Kehoe and Kydland [1992], Journal of Political Economy,
Baxter [1995], Hanbook of International Economics
 More failures than in closed economies due to capital
international mobility and portfolio diversification…but at the
origin of the new open economy macroeconomics, Obstfeld and
Rogoff [1995]. I will present some international RBC models in the
open macroeconomics course…
IV. 7. International Business Cycles
 After a technology shock abroad, aggregates increase as in the closed RBC
economy.
 As the capital return is higher abroad, domestic households invest abroad:
the capital stock decreases in the domestic country, and so the labor
demand decreases too. This explains why the model predicts a negative
correlation between production factors across countries.
 Output across countries is then less correlated in the model than in the
data.
 As there is perfect risk-sharing (complete markets), the marginal utility of
consumption is equalized across countries: the wealth distribution does
not vary over the business cycle. In case of positive shock abroad, the
domestic receives a payment provided by a state-contingent security.
 Consumption across countries is then more correlated in the model than in
the data.
IV. 8. The Great Depression
 Special issue of the Review of Economic Dynamics in 2002
investigating the 1930’s crisis in the main developed
countries using the RBC model
 P. Beaudry and F. Portier [2002], RED, The French
Depression in the 1930’s
IV. 8. The Great Depression
IV. 8. The Great Depression
IV. 8. The Great Depression
IV. 8. The Great Depression
IV. 8. The Great Depression
IV. 8. The Great Depression
IV. 9. Business Cycles Accounting
IV. 9. Business Cycles Accounting
IV. 9. Business Cycles Accounting