The Additionality of Public Support for Innovation: Evidence for Irish Manufacturing Firms

Nola Hewitt-Dundas (QUB)

Stephen Roper (ABS)

Paper presented to the Fifth Israeli/British & Irish Regional Science Workshop, Tel Aviv, Israel, 29th – 30th April 2007

Overview

• Economic & Policy Context

• Conceptual Basis

• Data source & research methods

• Discussion of Findings

• Key conclusions & policy implications

Regional & Policy Context

Manufacturing Output Growth in Ireland and Northern Ireland: 1991-2002 (1995=100)

50

70

90

110

130

150

170

190

210

230

250

1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002

1995=100

Northern Ireland

Ireland

Innovation Policy

• Interventionist Policy for R&D and Innovation

“Without state support and incentives the degree of investment

in technology will be less than is desirable from the point of

view of national economic development” (Culliton 1992 p.

55).

• EU Objective 1 status capability dev, infrastructure, R&D

• Innovation Support Programmes – focus on indigenously-owned firms

Ireland: Support for R&D and Innovation

1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 A. Shannon Development Near Market Support Product and Process Dev. R&D Capability RTI Measure I Measure III B. IDA Ireland* Near Market Support

R&D C. Udaras na Gaeltachta Near Market Support DEO27 [Technology Transfer] DEO 39 [Measure I, >50K] DEO 40 [Measure I, <50K] DEO47 [Measure III] DEO48 [Change] DEO49 [RTI] D. Enterprise Ireland** Near Market Support Measure 1 Measure 6 Research & Development RTI

Northern Ireland: Support for R&D and Innovation

1992 1993 1994 1995 1996 1997 1998 1999 2000 -1993 -1994 -1995 -1996 -1997 -1998 -1999 -2000 -2001 Near Market Support Product and Process Development Compete Pre-Competitive Schemes Science and Technology Programme START Technology Development Programme EU STRIDE Technology Use Measures Technology Audit Environmental Audit Environmental Management Scheme Innovation Audit Programme Other Support Schemes SMART Networking Programme Teaching Company Scheme/TCS IFI Science and Technology Programme/TICS

IFI RADIUS/RADIAN/RADIANE EU PRISMA

Business R&D support in Ireland and Northern Ireland, 1992-2001

0

5000000

10000000

15000000

20000000

25000000

30000000

35000000

40000000

1992 1993 1994 1995 1996 1997 1998 1999 2000 2001

Eur

o

NI Total Near MarketSupport €

Ireland Total NearMarket Support €

R&D and Innovation Grant Support €m pa

Implications for Research

• Differences in economic & policy environments Ireland & NI

• Targeting of Assistance

– Policy focus on indigenously-owned firms

– Policy focus on stimulating non-innovators to innovate - 1st Timers

Rationale for Interventionist policy

• Additionality & effect on business performance (Griliches 1995;

Mamuneas and Nadiri 1996)

• Reduce the cost of building up knowledge stocks (Trajtenberg 2000),

enhancing business performance (Klette and Johansen 1998), ability to

conduct future research (Mansfield and Switzer 1984, Luukkonen 2000)

• Development in Human Resources and innovation activity (Freel 2005)

• Absorptive capacity (Veugelers and Cassiman 1999, Cassiman and

Veugelers 2002)

• Reputational or ‘halo’ effects (Powell 1998)

• R&D cost savings through collaborative R&D (Irwin and Klenow 1996)

Focus of this paper - Project-level Additionality

… the decision of a firm either to abandon, go ahead with, or modify an innovation investment decision in light of the availability of public support.

– Extensive Additionality

– Incremental Additionality

– Radical Additionality

A Priori – where the population of firms comprises a mix of those undertaking no innovation, those undertaking incremental innovation and those undertaking radical innovation, public support for innovation should have positive extensive, incremental and radical additionality effects.

Public Support typically a proportion of total innovation cost

Data Source & Methods

• Innovation (knowledge) production function

• I = Innovation output

• X vector of plant level control variables

– knowledge sourcing

– Market position

– Resource base

• Z binary treatment variable

• Instrumental variables approach

zxI '

Irish Innovation Panel (IIP)

• 1991 – 2002 (analysis based on ‘94 to ‘02)

• 4 period postal survey

• Manufacturing plants >=10employees

• Avg response 34.5%

– c. 56% plants product innovators

– 12% sales new products

– 25% sales new and improved products

– c. 25% plants received public support for product dev.

Findings – Average Treatment Effects (ATEs)

Northern Ireland

Ireland All Plants

All Plants Extensive Additionality + (+) (+) Incremental Additionality + (+) + Radical Additionality + (+) + Indigenously-owned plants Extensive Additionality + + + Incremental Additionality + + + Radical Additionality (+) + +

Notes: + denotes significant positive effect; (+) denotes positive but insignificant effect.

Plant-level variables

All Firms

• Internal R&D

• Supply Chain Knowledge

Linkages

• Skill Levels

• Capital Intensity

• Production activity

Less important

• Scale, Ownership, Plant Vintage

Indigenous FIrms

• Internal R&D

• Supply-chain Knowledge

linkages

• Skill Levels

• Production activity

Less important

• Scale, Vintage, Multi-plant group

Policy Implications

• Positive effect of public support for product development

– Extensive, incremental & radical additionality effects

1 Grant assistance is effective either in isolation or as part of a package of innovation support measures

• Other factors

– In house R&D and supply chain knowledge linkages

2 Initiatives to strengthen internal R&D innovation and absorptive capacity

3 Initiatives to support knowledge linkages likely to also encourage innovation

– Organisational Context, skill base & capital investment

4 Measures to support skill development effectiveness of innovation through knowledge absorption & commercial exploitation