Docstoc

Discussants Comments on Interdependenies of US manufacturing

Document Sample
Discussants Comments on Interdependenies of US manufacturing Powered By Docstoc
					       Discussant’s Comments on
Interdependencies of U.S. Manufacturing
 Sectoral TFP’s: A Spatial VAR Approach
            by Klaus Neusser

            Hanna-Leena Männistö
               Bank of Finland
  CEPR-EABCN Workshop, Helsinki 28-29 Nov 2005
                            Motivation


  What causes business cycles?

  The author expects to find joint evolution of growth
     rates of sectoral TFP 's due to
  - Marshallian externalities or spillovers
  - Lack of substitutability in each sector's inputs
  - Aggregation




                                                         2
Osastolyhenne/päiväys
                        Modelling strategy


  NBER-CES Manuf. Industry Database, 458 sectors

  The effect a sector has on another depends on the
     "economic distance" between them
  - input factor shares (avg 1958-1996)
  - input-output tables (1987)

  Two specifications, SVARX and ARSX
  • autocorrelation and lagged effect across sectors
  • exogenous aggregate variables to count for the
    cyclical variation in capacity utilization
  • Marshallian externalities quantified directly      3
Osastolyhenne/päiväys
                             Findings


  Growth rate of TFP in a sector is influenced by neither
    lagged TFP growth rates in other sectors nor by
    growth rates of capital intensities in other sector.

  Suggests
  • Absence of mechanisms that could generate
    business cycles from non-aggregate shocks
  • Comovements of sectoral measures of economic
    activity not due to complementarities among
    sectoral technology shocks


                                                            4
Osastolyhenne/päiväys
                           Comments


  Empirical findings seem to be against author’s
    motivation and against discussant’s intuition
  How to reconcile them, say, with literature on
    technology diffusion?

  Eaton & Kortum, IER 40(3), 1999
     Innovative countries are the most productive, but
     there innovations also drive growth elsewhere.

  Maliranta, J Industry, Competition and Trade 5(1) 2005
   R&D, International trade and creative destruction –
    empirical findings from Finnish manuf. industries
                                                           5
Osastolyhenne/päiväys
                          Comments



  Measuremet problems more than likely

  1. TFP
  • micro level measurement demanding,
     competition/market structure

  2. Distance
  • not time-varying
  • the two rather similar?
  • aggregates dimensions a lot

                                         6
Osastolyhenne/päiväys
                           Suggestions


  Check robustness cross different types of measures

  1. TFP
  • value added rather than gross output
  • control for competition/market structure

  2. Distance
  • something simpler, eg based on statistical
     classification (distance could be number of knots in
     the classification tree from one sector to another)


                                                            7
Osastolyhenne/päiväys

				
DOCUMENT INFO
Shared By:
Categories:
Tags:
Stats:
views:4
posted:10/6/2012
language:Latin
pages:7