growth and the geography of innovation by attila varga

Growth and the geography of innovation

by

Attila Varga

Center for Research in Economic Policy (GKK)

and

Department of Economics

Faculty of Business and Economics

University of Pécs, Hungary

I. Introduction

• A-spatial mainstream economic theory

• K, L and A only? How about their spatial arrangements?

• Why should we care about space?- Transport costs (can be integrated relatively easily)- Agglomeration externalities (require a different approach)

• Policy relevance (EU)

Outline

• Introduction• Technological progress, spatial structure and

macroeconomic growth: An empirical modeling framework

• Integrating agglomeration effects to development policy modeling

• Concluding remarks

II. Technological progress, spatial structure and macroeconomic growth

Complex issue treated in four separate fields of economics:

A. EG: “Endogenous economic growth” models: endogenized technological change in growth theory (Romer 1986, 1990, Lucas 1986, Aghion and Howitt 1998)

in Romer (1990):- for-profit private R&D- knowledge spillovers are essential in growth- rate of technical change equals rate of per-capita growth on

the steady state- Simplistic explanation of technological progress, no

geography


B. IS: „Systems of innovation”literature: innovation is an interactive process among actors of the system (Lundval 1992, Nelson 1993)

actors of the IS:- innovating firms- suppliers, buyers- industrial research laboratories- public (university) research institutes- business services- “institutions”

level of innovation depends on:- the knowledge accumulated in the system- the interactions (knowledge flows) among the actors

- codified, non-codified (tacit) knowledge and the potential significance of spatial proximity- geography gets some focus, but IS does not say anything about growth


C. NEG: “New economic geography” models: endogenized spatial economic structure in a general equilibrium model (Krugman 1991, Fujita, Krugman and Venables 1999, Fujita and Thisse 2002)

- spatially extended Dixit-Stiglitz framework- increasing returns, monopolistic competition- spatial structure depends on some parameter conditions that determine the equilibrium level of centrifugal and centripetal forces- „cumulative causation”- C-P model by Krugman: still the point of departure- models quickly become complex: simulations if analytical solutions are not accessible

- Technological change not explained (not even included until very recently), the study of its relation to growth is a recent phenomenon


D. GI: The „Geography of innovation” literature: the study of the spatial extent of knowledge flows in innovation (Jaffe 1989, Jaffe, Trajtenberg and Henderson 1993, Audretsch and Feldman 1996, Anselin, Varga and Acs 1997)

- Empirical litarature: US, European, Asian analyses

- Common finding: much of knowledge flows in technological change are spatially bounded

- Not connected to growth and to the explanation of spatial economic structure


• IS, NEG, EGT, GI: complements to each other in growth explanation, no theoretical integration (Acs-Varga 2002)

• IS, NEG, EGT, GI: building blocks of a framework to shape empirical research (Varga 2006)

• Theoretical integration: endogenous growth and new economic geography (Baldwin and Forslid 2000, Fujita and Thisse 2002, Baldwin et al. 2003)

• EG, IS, NEG, GI: methodological problems in THEORETICAL integration (dramatically diverging initial assumptions, different theoretical structures, research methodologies)

• EMPIRICAL integration: very few work (Ciccone and Hall 1996, Varga and Schalk 2004, Acs and Varga 2004)

II. Technological progress, spatial structure and macroeconomic growth: An empirical modeling

framework

• Starting points („stylized facts”):

- Technological change is a collective process that depends on accumulated knowledge and interactions (IS)

- Technological change is the simple most important determinant of economic growth (EG)

- Codified and tacit knowledge: different channels of spillovers (GI)

- Centripetal and centrifugal forces shape geographical structure via cumulative processes (NEG)

- The resulting geographic structure is a determinant of the rate of growth (NEG)

• Y = AKαLβ (EG)

• The Romer (1990) equation as in Jones (1995)

dA = HA Aφ,

- HA: the number of researchers (“person-embodied”, knowledge component of knowledge production)

- A: the total stock of technological knowledge (codified knowledge component of knowledge production in books, patent documents etc.)- dA: the change in technological knowledge- : the “research productivity parameter” (0<<1)

φ: “codified knowledge spillovers parameter” - reflects spillovers with unlimited spatial accessibility

: the “research spillovers parameter”- reflects localized knowledge spillover effects (GI)- regional and urban economics and the new economic geography suggest: changes with geographic concentration of economic activities (depending on the balance between positive and negative agglomeration economies)

II. Technological progress, spatial structure and macroeconomic growth: An empirical modeling

framework

II. Technological progress, spatial structure and macroeconomic growth: An empirical modeling framework

Eq.1 Regional knowledge production:Kr = K (RDr, URDr, Zr)

A cumulative process described by Eqs. 2 and 3 (dynamic agglomeration effects:

Eq.2 (Static) agglomeration effect in R&D effectiveness: ∂Kr/∂RDr = f (RDr, URDr, Zr)

Eq.3 R&D location: dRDr = R(∂Kr/∂RDr)

Eq.4 Geography and : = (GSTR(HA))

Eq.5 dA = HA Aφ

Eq.6 dy/y = H(dA, ZN)

Empirical research on geography, technology and growth: 1986-2004

0

10

20

30

40

50

60

1986

1987

1988

1989

1990

1991

1992

1993

1994

1995

1996

1997

1998

1999

2000

2001

2002

2003

2004

papers

1986-2004: 253 papers on the geography of knowledge spillovers

journal articles: 175

books, book chapters, working papers: 78


• To test Eq.1: most of the empirical models are based on the

„knowledge production framework”

log (K) = + log(R) + log(U) + log(Z) +

• The KPF framework to study localized knowledge spillovers

USA: Jaffe 1989 Acs, Audretsch and Feldman 1991

Anselin, Varga and Acs, 1997 Varga 1998

Feldman and Audretsch 1999 Acs, Anselin and Varga 2002

EU: Moreno-Serrano, Paci, Usai 2005Italy: Audretsch and Vivarelly 1994, Capello 2001

France: Autant-Bernard 1999 Austria: Fischer and Varga 2003 Germany: Fritsch 2002


• Eq.2 and Eq. 3:

– empirical studies test the effects SEPARATELY (Jaffe 1989, Bania et al 1992, Anselin, Varga, Acs 1997a,b, Varga 2000, 2001)

– The dynamic cumulative process is not modeled empirically

• Empirical integration of micro to macro (Eqs. 4-6) is also missing

III. Integrating agglomeration effects to development policy modeling

• Knowledge-based development policies (R&D promotion, infrastructure investments, education support etc.)

• Modeling the effect of geography on policy effectiveness - three steps:1. modeling static agglomeration effects generated by the spatial distribution of the instruments2. modeling dynamic agglomeration effects of policy intervention: “cumulative causation” – induced technological change3. modeling the resulting macroeconomic effects

• In most of the current policy analysis models: no geography incorporated

III. A key issue in development policy modelling: integrating the spatial dimension of technological change

• The GMR Hungary model:

- integrates all the above three aspects

- developed for ex-ante CSF intervention analysis for the Hungarian government (planning period 2007-13)

- result of on international collaboration with German, Dutch and Japanese institutes

- both macro and regional aspects are estimated

IV. Outline of the GMR model

• CSF instruments targeting technology development:

– Infrastructure investments– Education/training support– R&D promotion


• GMR consists of three sub-models:

- the TFP sub-model (static agglomeration effects)

- the spatial computable general equilibrium (SCGE) sub-model (dynamic agglomeartion effects)

- a complete macroeconomic model (the effects of geography on macroeconomic variables)

The function of the TFP sub-model

• To generate STATIC TFP changes as a result of CSF interventions (direct short-run CSF-effect)

• NOT for forecasting but for impact analysis

Main characteristics of the TFP sub-model

• TFP equation:

- estimates the effects of geographically differently located knowledge sources (local, national, international)

- estimates the effects of CSF-instruments (infra, edu)

• Time-space data

The TFP equationThe estimated regional model of technological change

TFPGR = α0 + α1KNAT + α2RD+ α3 KIMP + α4INFRAINV + α5HUMCAPINV + ε,

TFPGR: the annual rate of growth of Total Factor Productivity (TFP),

KNAT: domestically available technological knowledge accessible with no geographical restrictions (measured by stock of patents),

RD: private and public regional R&D,

KIMP: imported technologies (measured by FDI),

INFRAINV: investment in physical infrastructure,

HUMCAPINV: investment in human capital,

region i and time t

α1 estimates domestic knowledge effects

α2 estimates localized (regional) knowledge effects

α3 estimates international knowledge effects

Table 1: Pooled FGLS estimation results for TFP growth rates for 20 Hungarian counties, 1999 – 2003

Note: estimated standard errors are in parentheses

Model 1

Model 2

Model 3

Model 4

Model 5

-2.5434*** -2.2152*** -2.2452*** -2.2000 -2.2904*** C

(0.2989) (0.2779) (0.2831) (0.2760) (0.2389)

0.0002*** 0.0002*** 0.0002*** 0.0002*** 0.0002*** KNAT (-2)

(.68E-05) (2.49E-05) (6.08E-06) (2.47E-05) (2.16E-05)

4.62E-06** 1.35E-05** 1.15E-0.5** 6.30E-06* RD(-3)

(1.87E-06) (1.55E-07) (5.56E-06) (3.49E-07)

0.0772* 0.0763** KIMP(-3)

(0.0402) (0.0345)

2.22E-06*** d(INFRA(-1))

(7.17E-07)

6.66E-06*** d(HUMCAP(-1))

(1.88E-06)

-0.0511*** DUM99

(0.0062)

-0.2326 -0.1900 -0.01076 BPDUM

(0.1408) (0.1441) (0.1216)

Weighted Statistics

R2-adj 0.31 0.40 0.42 0.42 0.64

F-statistic 54.02 34.19 24.20 19.40 26.10

Prob (F-statistic) 0.00 0.00 0.00 0.00 0.00

Durbin-Watson stat 1.90 1.67 1.66 1.71 2.17

N

Unweighted Statistics

100 100 100 100 100

R2-adj 0.14 0.24 0.25 0.27 0.34

The function of the SCGE sub-model

• To generate DYNAMIC TFP changes that incorporate the effects of agglomeration externalities on labor-capital migration (induced long-run CSF effect)

• Agglomeration effects depends on:

- centripetal forces: local knowledge (TFP)

- centrifugal forces: transport cost, congestion• To calculate the spatial distribution of L, I, Y, w by

sectors for the period of simulation

The SCGE sub-model

• Adaptation of RAEM-Light (Koike, Thissen 2005)

• C-D production function, cost minimization, utility maximization, interregional trade, migration

• Equilibrium: - short run (regional equilibrium)- long run (interregional equilibrium)

Main characteristics of the SCGE sub-model

• NOT for historical forecasting• The aim: to study the spatial effects of shocks

(CSF intervention)• Without interventions: it represents full

spatial equilibrium - regional and interregional (no migration)

• Shock: interrupts the state of equilibrium, the model describes the gradual process towards full spatial equilibrium

The function of the MACRO sub-model

• Based on dynamic TFP values: the resulting effects on macro variables

The characteristics of the MACRO sub-model

• Complete macro model (supply, demand, income distribution) – the EcoRET model (Schalk, Varga 2004)

• C-D production technology, cost minimization• Supply and demand side effects of CSF • A-spatial model• Describes the effects of exogenous technological

change• Baseline: TFP growth without CSF interventions• Policy simulations: describe the effects of CSF-

induced TFP changes on macro variables

Table 2: The main TFP-related equations in the macroeconomic sub-model

Variable name Variable description

Equation

ETB Employment of the business sector

ETB=ETB(-1)*EXP(-0.634761 +LOG(GDPBV/GDPBV(-1)) -0.1*(LOG(ETB(-1)/GDPBV(-1)) +(1-XTAU)*LOG((WSSE/XTAU) /(UCC/(1-XTAU)))+XTAU*LOG(ELEFFU)) -0.029658*DUMMY95)

IPV Private total fixed capital formation, volume

IPV=IPV(-1)*EXP(-0.436159-0.126956*(LOG(IPV(-1) /GDPBV(-1))-(1./0.1) *LOG(GDPBV/GDPBV(-1)) -XTAU*LOG((WSSE/XTAU)/(UCC/(1.- XTAU)))+XTAU*LOG(ELEFFU)))

GDPBV Gross domestic product, business sector, volume, factor cost

GDPBV=FGDPBV*EXP(1.917987 -0.343706*LOG(CKL/PGDPB) +0.735110*LOG(FDDV/FGDPBV))

WSSE Compensation rate of the business sector

WSSE=EXP(-0.350168+0.804890*LOG(PCP) +LOG(PROD)-0.005358*UNR(-1))

UCC User cost of capital UCC=PIT*(IRL-DPGDP+10)

CKL Unit capital-labor costs CKL=EXP(XTAU*(LOG(WSSE/XTAU) -LOG(ELEFFU)) +(1.-XTAU)*LOG(UCC/(1.-XTAU)))

LF Labor force LF=POPT*(0.023508 +(LF(-1)/ POPT(-1))+0.381196 *LOG(ETB/ETB(-1))-0.00297249*UNR(-1) -0.0008826*TIME)

CPV Private final consumption expenditure, volume

CPV=YDRH*EXP(0.067187 +0.745648*LOG(CPV(-1)/YDRH(-1)) -0.588445*LOG(YDRH/YDRH(-1))- 0.005051*IRL+0.02838*DUMMY94)

PGDPB Gross domestic product, business sector, deflator

PGDPB=PGDPB(-1)*EXP(-1.907540 +0.331580*LOG(CKL)-0.347119 *LOG(PGDPB(-1)))

Regional and national level short run and long run effects of TFP changes induced by TFP-related CSF interventions

1. Intervention in any region increases regional TFP level in the mth sector (static agglomeration effect)

2. Short run effect: - price of the good decreases

- decreasing demand for both L and K (assuming output unchanged) - increasing regional and interregional demand for the good that increases demand for L and K- increased regional demand increases utility levels of consumers in the region

3. Long run effects: increasing utility levels induces labor migration into the region followed by capital migration

- resulting in a further increase in TFP (dynamic agglomeration effect)

- and finally a changed spatial economic structure

4. Macroeconomic variables reflect the long run equilibrium TFP level resulting from dynamic agglomeration effects

Regional and national level short run and long run effects of TFP changes

induced by TFP-related CSF interventions

SCGE sub-model(regional model)

Macroeconomic sub-model (demand, supply, income distribution)

TFP sub-model(Regional model)

Effects on spatial structure

Economic policy instruments: infrastructure, R&D and education

Macroeconomic effects

Static TFP changes

Dynamic TFP changes

Does geography matter in public policy?

Allocation of CSF support in Mill. 2004 HUF

CSF intervention 2007 2008 2009 2010 2011 2012 2013 Total

Investment 215735,2 196295,8 174740,8 165896 171666 177560,2 183545,2 1285439 Research and Development 74407,67 68089,27 61005,64 58472,44 61170,81 63279,05 65419,73 451844,6 Infastructure 354249,1 415866,1 483112,9 527311 544220,1 562892,1 581442,1 3469093

Education and training 124741,4 111177,6 96604,36 88378,25 87454,07 90409,37 93410,13 692175,2

Elasticity of GDP growth rate changes with respect to CSF spending by geographic concentration of funds (Budapest)

0.00

2.00

4.00

6.00

8.00

10.00

12.00

2009 2010 2011 2012 2013

5%

10%

15%

20%

25%

30%

35%

40%

45%

50%

Elasticity of GDP growth rate changes with respect to CSF spending by geographic concentration of funds

(Veszprém, Komárom, Győr-Sopron, Vas, Fejér)

0.00

1.00

2.00

3.00

4.00

5.00

6.00

7.00

2009 2010 2011 2012 2013

25.0%

30.0%

35.0%

40.0%

45.0%

50.0%

55.0%

60.0%

65.0%

70.0%

Elasticity of GDP growth rate changes with respect to CSF spending by geographic concentration of funds (East of Danube)

0.00

1.00

2.00

3.00

4.00

5.00

6.00

2009 2010 2011 2012 2013

45%

50%

55%

60%

65%

70%

75%

80%

85%

90%

Elasticity of GDP growth rate changes with respect to spatial concentration of CSF spending by geographic concentration of funds (Budapest)

0.00

0.20

0.40

0.60

0.80

1.00

1.20

1.40

1.60

1.80

10% 15% 20% 25% 30% 35% 40% 45% 50%

2008

2009

2010

2011

2012

2013

Elasticity of GDP growth rate changes with respect to spatial concentration of CSF spending by geographic concentration of funds

(Veszprém, Komárom, Győr-Sopron, Vas, Fejér)

-0.20

-0.10

0.00

0.10

0.20

0.30

0.40

0.50

30.0% 35.0% 40.0% 45.0% 50.0% 55.0% 60.0% 65.0% 75.0%

2008

2009

2010

2011

2012

2013

Elasticity of GDP growth rate changes with respect to spatial concentration of CSF spending by geographic concentration of funds (East of Danube)

-0.20

-0.10

0.00

0.10

0.20

0.30

0.40

0.50

50% 55% 60% 65% 70% 75% 80% 85% 90%

2008

2009

2010

2011

2012

2013

Conluding remarks

• Growth and the geography of innovation: theoretical versus empirical integration

• Geographic effects in policy modelling: the GMR model

• Results show that agglomeration effects are important factors in macroeconomic performance and neglecting them in development policy analyses could result in misleading expectations as to how a particular mixture of policies affect the economy.

growth and the geography of innovation by attila varga

Documents