Endogenous international technology spillovers and biased technical change
FvT July 2000

Contents of this file:
1. General
2. Experiments
3. Suggestions for further experiments
4. Aggregation used 


1. GENERAL
------------------
This version implements the GTAP add-on for endogenous technology spillovers
as documented in the GTAP technical paper TP15:                      

 Hans van Meijl & Frank van Tongeren, "Endogenous international             
 technology spillovers and biased technical change in the GTAP model"      

 For applications see: 
  Hans van Meijl & Frank van Tongeren (1998),  "Trade, technology spillovers and food production in China",              
  Weltwirtschaftliches Archiv, 134 (3).             
  Hans van Meijl & Frank van Tongeren (1999), "Endogenous international technology spillovers and biased technical      change in agriculture", Economic Systems Research, 11 (1).                          

NOTE: this implementation uses the new version 5 data format and model pre-release 5.1. Results may therefore differ slightly from those reported in TP15.


The experiments included here allow you to implement endogenous technology spillovers
of two kinds: - spillovers that affect TFP in specific sectors of 
                destination regions [ao(i,s)]    
              - spillovers that are input augmenting in specific sectors of
                destination regions  [af(i,j,s)], and being biased towards primary factors (land, labor) 

The model code and variable naming has changed since version 5. 
The new tech change equations allow for shocks that are sector-, region-, or factor specific:
(see also help file for help on closure changes)

EQUATION                                                                                                                                   SHOCKED 
                                                                                                                                                      in this application
----------------------------------------------------------------------------------------------------------------------------------------------------------
Equation AOWORLD
# sector/region specific average rate of output augmenting tech change #
(all,j,PROD_COMM)(all,r,REG)                                    ao(j,r) = aosec(j) + aoreg(r)                            + aoall(j,r);
                                                                                                                                                                 ^^^^^^^^^
Equation AFWORLD
# sector/region specific average rate of intermediates augmenting tech change #
(all,i,TRAD_COMM)(all,j,PROD_COMM)(all,r,REG)     af(i,j,r) = afcom(i) + afsec(j) + afreg(r)           + afall(i,j,r);
                                                                                                                                                                ^^^^^^^^^^
                                                                                                                                                       ALTERED
                                                                                                                                                       in this application
----------------------------------------------------------------------------------------------------------------------------------------------------------
Equation AFEWORLD
# sector/region specific average rate of prim. factor i augmenting tech change #
(all,i,ENDW_COMM)(all,j,PROD_COMM)(all,r,REG)     afe(i,j,r) = afecom(i) + afesec(j) + afereg(r)  + afeall(i,j,r);
                                                                                                                                                                ^^^^^^^^^^^

The core spillover equation is:
                a(i,s) = E(r,s)^[1 - d(r,s)]) * a(i,r)                                                                                (1)
where:
a(i,s)      : a technical change shifter in sector i, region s (destination) and region r (source of innovation)
E(r,s)    :  some measure of knowledge embodied in trade flows
d(r,s)     : some measure of absorption capacity in the receiving country



2. EXPERIMENTS
------------------------
NOTE - NOTE -NOTE -NOTE -NOTE -NOTE -NOTE -NOTE -NOTE -NOTE -NOTE -NOTE -NOTE 
Because of the necessity to distinguish source- and destination regions for innovations, as well as source- and affected sectors of innovations, there are additional sets definitions in the parameter files. In adition, there are a number of additional parameters which are specific to the appication and the experiments.
Please take care to use the correct parameter files
NOTE - NOTE -NOTE -NOTE -NOTE -NOTE -NOTE -NOTE -NOTE -NOTE -NOTE -NOTE -NOTE 

E1: illustrates the working of the spillover equation, by simulating a Hicks-neutral TFP shock in the transports equipmnt.
and machinery sector in North America with spillovers to other regions.
In this case we consider knowledge embodied in final products (e.g. machines), which is transmited using the 
Coe, Helpman & Hoffmaister  embodiment index, which is simply an export share:
E(r,s) = VXMD(i, r, s) / sum(k,SPILL_DEST, VXMD(i,r,k))

(see section 3 in TP15)

E1a: linear spillovers, aoall(i,s) = E(s,r) * aoall(i,r)
         this is achieved by setting the binary switch ABSFLEX = 0 in the 
        parameter file E1a.prm

E1b: spillovers with differences in human capital determining the absorption capacity
        d(s,r) = min[1, hs/hr] 
        where hs and hr denote average years of schooling in respectively the destination and source region 
        and
        aoall(i,s) = E(s,r)^[1-d(r,s)] * aoall(i,r)                                                                                      (2)
       parameter file E1b.prm                         

E1c: as E1b, but the absorption index determind by human capital AND structural similarity
       d(s,r) = min[1, hs/hr] * exp[-abs[ls - lr]/dmax]                                                                            (3)
       where ls and lr denote land/labour ratios in respectivel the destination and source region 
       and dmax denatoes the largest observed absolute difference between any two regions.
       parameter file E1c.prm

--------------------------------------------
E2: this illustrates the idea that knowledge is embodied in traded intermediate inputs, which may lead to factor biased
      technical change.
     The basic mechanism is:
      input augmenting TC:   afall(j,i,r) -> spillover -> afall(j,i,s) -> productivity change in using j in production of i in s
      factor bias                 :   afeall(f,i,r) = beta * afall(j,i,r) -> spillover -> afeall(f,i,s) = beta * afall(j,i,s) 
                                                                                                -> productivity change in using factor f in production                                                                                                                    of i in s

The embodiment index is  more complicated than in experiment 1,  and approximates the amount 
of innovative inputs used in activity i sourced from the innovating region r by: 

(total firm imports of innovative input by region s from region r) * (activity i's share of total imports into
                                                                                                      s of innovative input)
See TP15, section 4.3 for a complete derivation.

E2a: Chemicals augmenting shock in grains in EUR, no land factor bias with full spillovers (equation 2 and 3), but no
         land factor bias, betaland = 0 in all regions
        shock:  Shock afall("crp","gro","EUR") = uniform 10;
        parameter file E2a.prm

E2b: As E2a, but no full land factor bias, betaland = 1 in all regions
        shock:  Shock afall("crp","gro","EUR") = uniform 10;
        parameter file E2b.prm

E2c: As E2a, but with North America (NAM) the source region for the innovation
        betaland = 0 in all regions
        Shock afall("crp","gro","NAM") = uniform 10;
        parameter file E2c.prm

3. SUGGESTIONS for further experiments:
----------------------------------------------------------
- Based on E2, simulate innovation in Machinery (trm), with labour bias.
- experiment with different source regions of innovations
- use SSA to analyse impact of size of tech change shocks and degree of factor bias
- compare welfare effects of embodied spillover experiment with trade liberalization effects, or combine both

4. AGGREGATION used 
---------------------------
The database is from GTAP version 3
The sectoral aggregatioon is as follows:
 gro         & grain crops
 ngc         & non grain crops
 lst         & livestock
 fof         & forestry and fisheries
 min         & mining
 pcf         & processed food
 opa         & other processed agriculture
 tex         & textiles and wearing
 crp         & chemicals rubbers and plastics
 omf         & other manufacturing
 trm         & transport industries and machinery
 svc	     & services
= = = = = = = = = = = = = = = = = = = = = = = = = = = = = =
!
! 2. COMMODITY AGGREGATION MAPPING
! | Commodity                       new sector
= = = = = = = = = = = = = = = = = = = = = = = = = = = = = =
 pdr, paddy rice                     &  gro
 wht, wheat                          &  gro
 gro, grains                         &  gro
 ngc, non grain crops                &  ngc
 wol, wool                           &  lst
 olp, other livestock                &  lst
 for, forestry                       &  fof
 fsh, fisheries                      &  fof
 col, coal                           &  min
 oil, oil                            &  min
 gas, gas                            &  min
 omn, other minerals                 &  min
 pcr, processed rice                 &  pcf
 met, meat products                  &  pcf
 mil, milk products                  &  pcf
 ofp, other food products            &  pcf
 b_t, beverages and tobacco          &  opa
 tex, textiles                       &  tex
 wap, wearing apparels               &  tex
 lea, leather etc                    &  tex
 lum, lumber                         &  opa
 ppp, pulp paper etc                 &  opa
 p_c, petroleum and coal             &  omf
 crp, chemicals rubbers and plastics &  crp
 nmm, nonmetallic minerals           &  omf
 i_s, primary ferrous metals         &  omf
 nfm, nonferrous metals              &  omf
 fmp, fabricated metal products      &  omf
 trn, transport industries           &  trm
 ome, machinery and equipment        &  trm
 omf, other manufacturing            &  omf
 egw, electricity water and gas      &  svc
 cns, construction                   &  svc
 t_t, trade and transport            &  svc
 osp, other services (private)       &  svc
 osg, other services (govt)          &  svc
 dwe, ownership of dwellings         &  svc


The regional aggregation is as follows:
!          GTAP REGION                | Aggregated  Region
= = = = = = = = = = = = = = = = = = = = = = = = = = = = = =
 AUS    Australia                     &  AUS
 NZL    New Zealand                   &  AUS
 JPN    Japan                         &  JAN
 KOR    Republic of Korea             &  JAN
 IDN    Indonesia                     &  RAS
 MYS    Malaysia                      &  RAS
 PHL    Philippines                   &  RAS
 SGP    Singapore                     &  JAN
 THA    Thailand                      &  RAS
 CHN    China                         &  CHN
 HKG    Hong Kong                     &  JAN
 TWN    Taiwan                        &  JAN
 IDI    India                         &  RAS
 RAS    Rest of South Asia            &  RAS
 CAN    Canada                        &  NAM
 USA    United States of America      &  NAM
 MEX    Mexico                        &  SAM
 CAM    Central America and Caribbean &  SAM
 ARG    Argentina                     &  ARG
 BRA    Brazil                        &  SAM
 CHL    Chile                         &  SAM
 RSM    Rest of South America         &  SAM
 E_U    European Union 12             &  EUR
 EU3    Austria Finland and Sweden    &  EUR
 EFT    European Free Trade Area      &  EUR
 CEA    Central European Associates   &  ROW
 FSU    Former Soviet Union           &  ROW
 MEA    Middle East and North Africa  &  SAM
 SSA    Sub Saharan Africa            &  ROW
 ROW    Rest of World                 &  ROW


Note: ALTERTAX is not available.
