Diapositive 1

1
Theories and Methods of the Business Cycle. Part
1 Dynamic Stochastic General Equilibrium
Models IV. A New Neoclassical Synthesis? Jean-O
livier HAIRAULT, Professeur à Paris I
Panthéon-Sorbonne et à lEcole dEconomie de
Paris (EEP)
2
1. Introduction

• The RBC theory is at odds with the neoclassical
  synthesis
• Real shocks vs. Monetary shocks
• Optimal vs. Suboptimal fluctuations
• RBC theory has not totally convinced that
  technology shocks can alone drive the business
  cycle. The internal mechanisms of the
  neoclassical growth model does not lead to
  fluctuations totally consistent with the stylized
  facts when calibration is seriously done.
• The correlation between hours and labor
  productivity is too high
• The response of output does not display a
  hump-shaped profile
• How to reconcile theory with the numerous
  empirical works which show the non-neutrality of
  money (in particular in the VAR framework)? Gali
  1989, Quarterly Journal of Economics

3
1. Introduction

• Along the development of RBC theory in the 80s,
  keynesianism was looking for micro-foundations
• How to get market failures when agents are
  assumed to optimize?
• Imperfect information, imperfect competition,
  strategic behaviors,New-Keynesianism based on
  real and nominal rigidities. Ball and Romer
  1990, Review of Economic Studies
• Non-walrasian features in the good market
  (monopolistic competition), the labor market
  (search frictions), the credit market (adverse
  selection and moral hazard)
• Mainly theoretical works as new-keynesianism has
  to address the Lucas critique.
• What a challenge to overcome RBC theory with
  their own methodology!
• The 90s was the decade during which a new
  neoclassical synthesis occurs intertemporal
  choices and strategic behaviors

4
2. Equilibrium unemployment

• Employment volatility represents 70 of total
  hours volatility
• Hours indivisibility à la Hansen 1985
• Do efficiency wages better explain the relative
  volatility of hours to labor productivity?
  Danthine and Donaldson 1990, European Economic
  Review. Employment is not volatile enough in
  their model
• Efficiency wages are rigid in the sense they do
  not clear the labor market, but they are elastic
  to technology shocks.
• Fluctuations generated by the model are
  essentially the same as those of the RBC
  canonical model.

5
2. Equilibrium unemployment

• Equilibrium unemployment, Pissarides 1990,
  Equilibrium unemployment theory, Basil Blackwell
• Hirings take time as there is imperfect
  information in the process of search search
  unemployment
• Search frictions could explain the persistence of
  fluctuations
• Merz 1995, Journal of Monetary Economics,
  Andolfatto 1996, American Economic Review

6
2. Equilibrium unemployment

• Hirings depend on vacancies (V) and unemployed
  people (U), but also the search intensity (e) of
  these latter.
• The participation rate is constant and exogenous.
• The probability to have a job is
• The probability to contact a worker
• There are externalities in the search process
• p depends positively on the number of vacancies
  (complementarity) and negatively on the number of
  unemployed workers (congestion)
• q depends negatively on the number of vacancies
  and positively on the number of unemployed
  workers (congestion)

7
2. Equilibrium unemployment

• The dynamics of employment depends on hirings (a
  combination of unemploment and vacancy) and on
  firings (a fixed proportion s of the employment
  stock)

8
2. Equilibrium unemployment

• Representative household (risk-sharing due to the
  large scale of the household)
• First order conditions are the same as those in
  the canonical RBC model, except there is no labor
  supply, no partipation decision and hours are
  negociated between households and firms (as wage)

9
2. Equilibrium unemployment

• The production function is traditional and
  combines total hours and capital.
• Firm labor demand is no more static, as they
  invest in vacancies. The firm program is now
  intertemporal.
• Labor demand is now determined by the
  intertemporal condition

10
2. Equilibrium unemployment

• Bargaining over hours and wage
• Nash criterion
• With
• is the bargaining power of firms. In Andolfatto
  1996, it is equal to the elasticity relative to
  vacancy in the matching function. In this case,
  the bargaining leads to the first best allocation
  as demonstrated by Hosios 1990, Review of
  Economic Studies.
• Given the existence of search frictions,
  fluctuations are not necessarily sub-optimal.
  This is why Andolfatto 1996 solves the planner
  program.

11
2. Equilibrium unemployment

• Stylized facts standart deviation in the first
  line, correlation with output in the second line
• Andolfatto 1996
• The wage is no longer equal to labor productivity
  and the labor share is variable and
  counter-cyclical. The variance of output is
  obtained without infinite labor elasticity, but
  results from its higher persistence.
• But the correlation between hours and wage is not
  replicated.
• Increasing the volatility of employment by giving
  more bargaining power to firms as their response
  to technology shocks will then be more elastic.

12
2. Equilibrium unemployment

• Productivity cycle and Beveridge curve
• Stylized facts
• Models predictions the productivity cycle is
  not replicated contrary to the Beveridge curve

13
3. Perfect Insurance

• Complete markets perfect risk sharing
• B is the amount of insurance whose price is tau.
• 1-alpha s for employed workers 1-alpha 1-p
  for unemployed workers
• Two insurance schemes profit
• The budgetary constraints

14
3. Perfect Insurance

• First-order conditions
• Risk-sharing condition
• For s n, u
• These conditions imply that the capital choices
  are identical whatever their employment status.
  For separable utility fonction, consumptions are
  the same, and the unemployed workers are better
  off. See Chéron and Langot 2002, Review of
  Economic Dynamics for a case where their welfare
  is lower than that of employed workers.

15
3. Perfect Insurance

• Given these optimal choices, the budgetary
  constraints can be rewritten as follows
• It can be noticed that
• The representative household program is then

16
4. Monopolistic competition and nominal rigidities

• Considering VAR studies (following the seminal
  article of Sims 1980, Econometrica), money
  supply shocks would impact the output. J. Gali
  1992, Quarterly Journal of Economics, Bec and
  Hairault 1993, Annales dEconomie et
  Statistiques 1993
• Taking into account money supply shock in DSGE
  models implies to have money demand theoretical
  foundations Why do we hold money ? Old issue in
  economics
• Money reduces transaction or search costs in the
  good market Kiyotaki and Wright 1989, Journal
  of Political Economy
• More tractable in DSGE models, cash-in-advance
  constraint can be added in the household program,
  Cooley and Hansen 1989, American Economic
  Review
• Without any nominal rigidities, this approach
  fails to generate a positive and strong response
  of output after a positive monetary (supply)
  shock. Only the inflation tax propagation
  mechanism which is a negative wealth effect
  positive response of labour supply and output,
  but very weak effects.

17
4. Monopolistic competition and nominal rigidities

• (Inflation tax ) nominal rigidities prices are
  rigid due to menu costs (New-Keynesian approach)
  (an alternative rigidity of the nominal wage due
  to the existence of wage contracts)
• This implies to consider imperfect competition in
  the good market monopolistic competition in the
  line of Blanchard and Kiyotacki 1897, American
  Economic Review
• Nominal rigidities monopolistic competition in
  DSGE Hairault and Portier 1993, European
  Economic Review

18
4. Monopolistic competition and nominal rigidities

• Firms have monopoly power due to good
  differentiation.
• Prices include a mark-up over the marginal cost.
• Firms maximize profit by taking into account the
  effect of prices on the good demand.
• The demand for good j is a proportion of the
  total demand which depends on the relative price
  of good j
• The total demand for good j is then

19
4. Monopolistic competition and nominal rigidities

• The nominal profit of firm j is
• Due to the presence of adjustment costs on
  prices, firm choices are now intertemporal

20
4. Monopolistic competition and nominal rigidities

• The first-order conditions are
• Prices are not equalized to marginal costs

21
4. Monopolistic competition and nominal rigidities

• Money is introduced into the utility function

22
4. Monopolistic competition and nominal rigidities

• The first-order conditions are
• The money supply process is

23
4. Monopolistic competition and nominal rigidities

• The Solow Residual is no longer a pure measure of
  technology. The factor elasticities are not
  consistently measured by factor shares in total
  revenues due to the existence of markups.
• The naive SR is contamined by money supply shocks
  as these latter make markups counter-cyclical in
  the business cycle
• After a money supply shock, firms want to
  increase prices in order to leave unchanged their
  mark-up. As there exist adjustment costs on
  prices, prices do not increase as much as
  invariant mark-up would imply. Markups are weaker
  than at the steady state.

24
4. Monopolistic competition and nominal rigidities

• Stylized facts
• Models predictions

25
5. Financial imperfections

• Limited participation and money supply shocks
• Fuerst 1992, Journal of Monetary Economics,
  Christiano and Eichenbaum 1992, American
  Economic Review
• The nominal and real interest rates decline after
  a positive monetary shock in empirical studies

26
5. Financial imperfections

• The nominal and real interest rates decline after
  a positive monetary shock since the supply of
  deposits is pre-determined
• Firms rent wages

R
Dd, Ds
27
5. Financial imperfections

• Financial Accelerator, Carlstrom and Fuerst
  1997, American Economic Review, Bernanke,
  Gertler and Gilchrist, Handbook of
  Macroeconomics
• Entrepreneurs have access to a risky project.
  Lenders have no information about the ex-post
  return. They have to pay a verification cost.
• The optimal contract is then a debt contract. The
  debtor interest rate depends on the
  entrepreneurs wealth.
• Serial correlation of output growth depends on
  the agency costs

28
6. Sunspot and fluctuations

• Self-fulfilling propheties, Farmer and Guo
  1994, Journal of Economic Theory
• Animal spirits at the heart of the business cycle
• Extrinsic shocks vs. Shocks on fundamentals
• Indeterminacy of the equilibrium too many
  eigenvalues inferior to 1 (more than the number
  of pre-determined variables).
• Sunspot equilibria can arise in this case.
• This approach must be distinguished from news
  about the future of some fundamentals, see
  Hairault, Langot and Portier 1997, Journal of
  Economic, Dynamics and Control, Beaudry and
  Portier 2006, American Economic Review

29
6. Sunspot and fluctuations

• Final good is produced by using intermediated
  inputs
• Production of these inputs under increasing
  return to scale and monopolistic competition
• The reduced form of the model is
• with

30
6. Sunspot and fluctuations

• If the labor elasticity is small enough relative
  to a, ie the labor demande curve is increasing
  with a slop superior to that of the labor supply
  curve, the eigenvalues are strictly inferior to
  one and sunspot equilibria exist

31
7. Business cycle costs

• Stabilization of business cycle? What does it
  mean in the DSGE framework?
• Welfare criterion must be considered, and not
  volatility criterion, especially the output
  volatility
• Business cycle costs are very small in terms of
  stationary consumption eliminating all
  fluctuations is equivalent in welfare units to
  0.008 of the steady state consumption, Lucas
  1987, Models of Business Cycles, Basil
  Blackwell.
• This implies that the distorsions introduced by
  the stabilization policy must be very small too.
• Harberger triangles could be much more important
  than Okun gaps, Greenwood and Huffman 1991,
  Journal of Monetary Economics.
• I argue in the end that, based on what we know
  now, it is unrealistic to hope for gains larger
  than a tenth of a percent from better
  countercyclical policy Lucas 2003, American
  Economic Review

32
7. Business cycle costs

• Reis 2007

33
7. Business cycle costs
34
7. Business cycle costs
35
7. Business cycle costs
36
8. Okun Gaps and Harberger triangles
37
8. Okun Gaps and Harberger triangles
38
8. Okun Gaps and Harberger triangles
39
8. Okun Gaps and Harberger triangles
40
8. Okun Gaps and Harberger triangles
41
8. Okun Gaps and Harberger triangles
42
8. Okun Gaps and Harberger triangles