DIMETIC Pécs 2009 The geography of innovation and growth: An introduction and overview by Attila Varga Department of Economics and Regional Studies Faculty

DIMETIC Pécs 2009

The geography of innovation and growth:

An introduction and overview

by

Attila Varga

Department of Economics and Regional StudiesFaculty 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?

I. Introduction

• 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. EGT: “Endogenous economic growth” models: endogenized technological change in growth theory (Romer 1986, 1990, Lucas 1986, Aghion and Howitt 1998, 2009)

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. GE: “Geographical economics” models:

- „New Economic Geography”(NEG): endogenized spatial economic structure in a general equilibrium model (Krugman 1991, Fujita, Krugman and Venables 1999, Fujita and Thisse 2002)- „Evolutionary Economic Geography” (EEG) (Boschma 2007)


In the New Economic Geography:

- 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, Anselin, Varga and Acs 1997)

- Empirical litarature: US, European, Asian analyses

- Common finding: much of knowledge flows in technological change are spatially bounded (though differences with respect to industry, stage of innovation, institutional proximity exist, related variety important)

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


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

• IS, GE, 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)

• EGT, IS, GE, 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)

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 (GE)

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

• Y = AKαLβ

• 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

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

: the “research productivity 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)


• 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


To test Eq.2 and Eq. 3:

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

– The dynamic cumulative process modeled econometrically (Varga, Pontikakis, Chorafakis 2009)

Table 1. Regression Results for Log (Patents) for 189 EU regions, 2000-2002 (N=567)

Model (1) (2) (3) (4) (5) (6) Estimation OLS OLS OLS OLS OLS 2SLS- Spatial

Lag (INV2) Constant W_Log(PAT) Log(GRD(-2)) Log(GRD(-2))*Log(δ(-2)) Log(GRD(-2))*Log(NETRD(-2)) Log(PSTCK(-2)) PAHTCORE

-1.6421*** (0.1776)

1.0822*** (0.0308)

-0.3107 (0.2316)

0.8453*** (0.0407)

0.3242*** (0.0389)

-0.5391* (0.2806)

0.9585*** (0.0886)

0.3222*** (0.0389)

-8.675E-05 (6.03E-05)

-1.7864*** (0.2381)

0.7142*** (0.0377)

0.2443*** (0.0351)

0.2502*** (0.0203)

-1.7227*** (0.2372)

0.6879*** (0.0384)

0.2136*** (0.0363)

0.2536*** (0.0202)

0.4814*** (0.1568)

-1.3881*** (0.2665) 0.0027** (0.0012)

0.6067*** (0.0483)

0.2429*** (0.0393)

0.2547*** (0.0214)

0.5007*** (0.1612)

R2-adj Log Likelihood

0.69 -885.30

0.72 -852.36

0.72 -851.32

0.78 -784.69

0.78 -779.98

0.78

Multicollinearity Condition Number F on pooling (time) F on slope homogeneity White test for heteroscedasticity LM-Err Neighb INV1 INV2 LM-Lag Neighb INV1 INV2

7

0.9071 0.4815

0.7529

111.78*** 252.17*** 215.12***

142.53*** 247.03*** 237.99***

10

0.6777 0.7613

1.0462

69.36*** 129.64*** 121.59***

100.88*** 159.07*** 148.93***

24

0.5644 0.5836

12.8409

66.85*** 117.26*** 114.45***

99.03*** 153.47*** 145.48***

13

0.8143 0.6485

3.6634

26.95*** 29.87*** 32.40***

24.99*** 28.16*** 31.42***

13

0.6425 0.4645

12.1852

23.46*** 26.13*** 29.24***

25.89*** 27.96*** 30.95***

Notes: Estimated standard errors are in parentheses; spatial weights matrices are row-standardized: Neigh is neighborhood contiguity matrix; INV1 is inverse distance matrix; INV2 is inverse distance squared matrix; W_Log(PAT) is the spatially lagged dependent variable where W stands for the weights matrix INV2. *** indicates significance at p < 0.01; ** indicates significance at p < 0.05; * indicates p < 0.1 . Endogenous variables in Model 6: W_Log(PAT) and Log(GRD(-2); Instruments: W_Log(PUB), Log(PUB), W_Log(GRD(-2)), PUBCORE. The Durbin-Wu_Hausman test for Log(GRD(-2))* Log(δ(-2)) does not reject exogeneity.

Table 2. Regression Results for Log (Publications) for 189 EU regions, 2000-2002 (N=567)

Model (1) (2) (3) (4) (5) (6) Estimation OLS OLS OLS OLS OLS 2SLS Spatial Lag

(Neigh) Heteroscedasticity

robust Constant W_Log(PUB) Log(GRD(-2)) Log(GRD(-2))*Log(δ(-2)) Log(GRD(-2))*Log(NETRD(-2)) Log(PSTCK(-2)) PUBCORE

1.4026*** (0.1298)

0.942*** (0.0225)

2.3886*** (0.1645)

0.445*** (0.0597)

0.0004*** (4.40E-05 )

2.196*** (0.202)

0.480*** (0.633) -0.0462 (0.0282) 0.0004**

* (4.40E-

05)

2.3395*** (0.1711)

0.4158*** (0.066)

0.0004*** (4.56E-05)

0.01758 (0.01689)

2.4568*** (0.1697)

0.4523*** (0.0602)

0.0004*** (4.68E-05)

0.2247** (0.1032)

2.6237*** (0.2127)

-0.0313*** (0.0021)

0.7083*** (0.1283)

0.0002*** (7.3225E-5)

0.1964* (0.1106)

R2-adj Log Likelihood

0.76 -707.30

0.79 -670.05

0.79 -668.70

0.79 -669.51

0.79 -667.89

0.79

Multicollinearity Condition Number F on pooling (time) F on slope homogeneity White test for heteroscedasticity LM-Err Neighb INV1 INV2 LM-Lag Neighb INV1 INV2

7

0.6694 0.2059

44.575***

0.7199 3.3586* 0.3687

12.214*** 1.6479

5.2928**

22

0.9269 0.357

77.378***

0.7727 2.5407 0.9367

3.0067* 0.0642 0.6649

23

0.6712 0.2752

84.013***

0.7518 1.8767 0.8782

2.4689 0.4640 0.1242

27

0.7141 0.2683

92.231***

0.9808 3.4006* 1.2604

4.2311** 0.061 1.9522

24

0.7055 0.2501

86.884***

0.5749 2.6595 1.020

3.7861* 0.0069 1.1352

Notes: Estimated standard errors are in parentheses; spatial weights matrices are row-standardized: Neigh is neighborhood contiguity matrix; INV1 is inverse distance matrix; INV2 is inverse distance squared matrix; *** indicates significance at p < 0.01; ** indicates significance at p < 0.05; * indicates p < 0.1. Endogenous variable in Model 6: W_Log(PUB). Instruments: W_Log(PAT), Log(PAT), W_Log(GRD(-2)), PATHCORE. The Durbin-Wu_Hausman test for Log(GRD(-2)) and Log(GRD(-2))* Log(NETRD(-2)) does not reject exogeneity.

Regional R&D productivity in competitive research

BETAPAT< -3 Std. Dev.-3.0 - -2.5 Std. Dev.-2.5 - -2.0 Std. Dev.-2.0 - -1.5 Std. Dev.-1.5 - -1.0 Std. Dev.-1.0 - -0.5 Std. Dev.-0.5 - 0.0 Std. Dev.Mean0.0 - 0.5 Std. Dev.0.5 - 1.0 Std. Dev.1.0 - 1.5 Std. Dev.1.5 - 2.0 Std. Dev.2.0 - 2.5 Std. Dev.

Regional R&D productivity in pre-competitive research

BETAPUB< -3 Std. Dev.-3.0 - -2.5 Std. Dev.-2.5 - -2.0 Std. Dev.-2.0 - -1.5 Std. Dev.-1.5 - -1.0 Std. Dev.-1.0 - -0.5 Std. Dev.-0.5 - 0.0 Std. Dev.Mean0.0 - 0.5 Std. Dev.0.5 - 1.0 Std. Dev.1.0 - 1.5 Std. Dev.

Table 3. Regression Results for (GRD2001-GRD1998) for EU regions (N=189)

Model (1) (2) (3) (4) Estimation OLS OLS OLS OLS-Heteroscedasticity

Robust (White) Constant BETAPAT1998 BETAPUB1998 PSTCK2001

-1423.44*** (196.344)

2345.18*** (301.971)

-1385.92*** (193.372)

1863.47*** (345.535)

364.843*** (131.183)

-550.111*** (149.521)

720.125*** (256.492) 190.336** (93.496)

360.982*** (26.322)

-550.111*** (143.413)

720.125*** (241.882)

190.336*** (69.895)

360.982*** (47.416)

R2-adj 0.24 0.27 0.63 0.63 White test for heteroscedasticity LM-Err Neighb INV1 INV2 LM-Lag Neighb INV1 INV2

52.316***

0.113 0.009 0.089

0.096 2.696 0.595

57.884***

0.023 0.198 0.963

0.043 1.820 0.531

42.236***

0.068 1.148 0.942

0.103 1.996 1.989

Table 4. Regression Results for (HTEMP2001-HTEMP1998) for EU regions

(N=189) Model (1) (2) (3) (4) Estimation OLS OLS OLS ML – Spatial Error (INV2)

with Heteroscedasticity weights

Constant HTEMP1998 HTEMP1998*GRD1998 RDCORE LAMBDA

5399.78* (3032.61) 0.071*** (0.006)

8821.36*** (3314.62) 0.054*** (0.009)

3.788E-06** (1.582E-06)

9955.96*** (3267.78) 0.032*** (0.012)

5.043E-06*** (1.604E-06) 19896.5*** (6614.64)

11168.3*** (2879.48) 0.0262** (0.011)

5.624E-06*** (1.604E-06) 21321.1*** (6366.96) -0.0181**

(0.009) R2-adj 0.41 0.42 0.45 0.45 Multicollinearity Condition Number White test for heteroscedasticity LM-Err Neighb INV1 INV2 LM-Lag Neighb INV1 INV2

2

27.37***

0.922 0.052 1.008

2.181 0.479 4.000*

4

28.182***

0.164 0.023 3.263*

1.846 0.043

4.574**

6

34.522***

0.042 0.28

5.878**

1.916 0.645

4.316**


• Empirical integration of micro to macro (Eqs. 4-6): a real research challenge: needs an integrated macro-regional approach (back to this on Friday afternoon)

Research questions related to the empircal model: The structure of the week

– The geography of innovation: knowledge interactions in space• Stefano USAI• Francesco LISSONI

– How to explain the geographical structure of innovation and the resulting growth?

• EEG: Ron BOSCHMA, Giulio BOTTAZZI• NEG: Mark THISSEN

– How new product varieties emerge and how they are related to economic growth?

• Pier Paolo SAVIOTTI

– Empirical research methodology:• Frank van OORT• Attila VARGA

The structure of the weekMonday, June 29th 9.00 – 9.30: Welcome, practical and organizational issues – Ron BOSCHMA and Attila VARGA 9.30 – 10.30: The geography of innovation and growth: An introduction and overview – Attila VARGA 11.00 – 12.30: Knowledge spillovers in space – revisiting the issue with a novel OECD data set – Stefano USAI 14.00 – 16.00: PhD presentation n°1: Senior discussant: Attila VARGA 16.15 – 17.45: PhD presentation n°2: Senior discussant: Stefano USAI Tuesday, June 30th 9.00 - 10.30: Geography and growth: The perspective of the evolutionary economic geography – Ron BOSCHMA 11.00 - 12.30: Geography and growth: The perspective of the new economic geography – Harry GARRETSEN 14.00 - 16.00: PhD Presentation n° 3: Senior discussant: Ron BOSCHMA 16.15 - 17.45: PhD Presentation n° 4: Senior discussant: Harry GARRETSEN Wednesday, July 1st 9.00 – 10.30: New variety and economic growth - Pier Paolo SAVIOTTI 11.00- 12.30: Evolutionary modeling of location – Giulio BOTTAZZI 14.00-16.00: PhD Presentation n° 5: Senior discussant: Pier Paolo SAVIOTTI 16.15 - 17.45: PhD Presentation n° 6: Senior discussant: Giulio BOTTAZZI Thursday, July 2nd 9.00 – 10.30: Spatial spillovers or markets? – Francesco LISSONI 11.00- 12.30: Empirical methodologies to research agglomeration externalities – Frank van Oort 14.00-16.00: PhD Presentation n° 5: Senior discussant: Pier Paolo SAVIOTTI 16.15 - 17.45: PhD Presentation n° 6: Senior discussant: Giulio BOTTAZZI Friday, July 3rd 9.00 - 10.30: Applied Spatial Econometric Analysis I. – Julie Le Gallo 11.00 - 12.30: Applied Spatial Econometric Analysis II. - Julie Le Gallo 14.00-16.00: Spatial econometrics in practice – application in the field of the geography of innovation – Julie Le Gallo 16.15 – 16:45: Integrating the geography of innovation to policy modeling – Attila VARGA 17:00 – 17:45: Concluding session. A summary discussion – Ron BOSCHMA and Attila VARGA

Documents

DIMETIC Pécs 2009 The geography of innovation and growth: An introduction and overview by Attila Varga Department of Economics and Regional Studies Faculty