INDUCING ENVIRONMENTAL INNOVATION

Nick JohnstoneStructural Policy DivisionDirectorate for Science, Technology and IndustryOECD

Sustainable Prosperity Conference Ottawa

April 28&29, 2014

Source: OECD, Invention and Transfer of Environmental Technologies (2011) www.oecd.org/environment/innovation

19801982

19841986

19881990

19921994

19961998

20002002

20042006

20080

200

400

600

800

1000

1200

1400

0

20000

40000

60000

80000

100000

120000

140000

160000

Air pollution abatement (from sta-tionary sources)Water pollution abatementWaste managementTotal (right-axis)

Environment-Related Patents in OECD Countries(Number of patent applications - claimed priorities, worldwide)

CC Mitigation and Adaptation Technologies(Number of patent applications - claimed priorities, worldwide)

Source: Haščič, I. et al. (2010), “Climate Policy and Technological Innovation and Transfer: An Overview of Trends and Recent Empirical Results”, OECD Environment Working Papers, No. 30 http://dx.doi.org/10.1787/5km33bnggcd0-en

• Stringency/Ambition – how ambitious is the policy objective relative to “BAU”

• Predictability/Credibility – how certain and credible is the signal given by the policy

• Flexibility – how much space is provided to identify new technologies and methods

• Incidence – how closely does the measure target the environmental bad or good

• Depth – does the measure provide incentives over the range of all possible outcomes

4

Principles of environmental policy design in order to encourage 'green' innovation

Source: http://www.oecd-ilibrary.org/environment/environmental-policy-design-characteristics-and-technological-innovation_5kmjstwtqwhd-en

5

The Role of Policy Flexibility: The Effect on Patented Environmental Inventions

Note: Figure shows the estimated coefficients from negative binomial models of different characteristics of environmental policy framework (policy stringency, policy flexibility) in encouraging inventive activity in environmental technologies. Measured as the number of patent applications (claimed priorities) deposited during 1975-2006. Source: OECD (2011) Invention and Transfer of Environmental Technologies www.oecd.org/environment/innovation

Model 1 (No Year FE) Model 2 (Year FE)0

0.2

0.4

0.6

0.8

1

1.2

1.4

1.6

1.8

Stringency AloneStringency & Flexibility

6

The Role of Policy Predictability: Effect of Volatility in Public R&D on Inventive Activity

Note: Figure shows the estimated coefficients from negative binomial models. Figure shows the estimated response to a 1% increase in the level and volatility of public R&D in encouraging inventive activity in environmental technologies, measured as the number of patent applications (claimed priorities) deposited during 1975-2007 in a cross-section of OECD countries. Source: Kalamova, Johnstone and Hascic (2012) in V. Constantini and M. Mazzanti (eds.) The Dynamics of Environmental and Economic Systems (Springer, 2012).

R&D Volatility R&D Level

-0.40

-0.30

-0.20

-0.10

0.00

0.10

0.20

0.30

0.40

0.50

Model 1 (No FE)Model 2 (FE)

7

The Role of Policy Ambition: Effect of FITs on Cross-border Investment in Renewable Energy

Note: The table presents the coefficients of logistic regressions estimated at the deal level. Source: Criscuolo, Johnstone et al. “Renewable energy policies and cross-border investment: Evidence From M&A in solar and wind energy” OECD STI Working paper (forthcoming)

• Taxes

• can be ambitious (often depending upon recycling)

• direct incidence depending upon administrative costs of targeting

• flexible

• predictable and credible (depending upon ambition)

• deep incentives

• Technology-based standards

• can be ambitious (but often discriminatory for entrants)

• indirect incidence

• generally inflexible

• predictable and credible (at least in short-run)

• depth only to standard

8

A comparison of environmental taxes and technology-based standards: A caricature

Source: http://www.oecd-ilibrary.org/environment/environmental-policy-design-characteristics-and-technological-innovation_5kmjstwtqwhd-en

Prices spur innovationThe Effect of an Energy Tax on High-Value Patent Counts in Combustion Efficiency and Renewable Energy

Source: OECD Energy and Climate Policy and Innovation (2012). Based on estimation of sample of OECD economies over period 1978-2008. Results indicate that if oil price is approximately equal to prices reached in 2008 oil shock => switch from fossil fuel combustion efficiency innovation to renewable energy innovation. See also ENV WKP 45

• Difficulty of targeting environmental ‘bad’ directly and excessive administrative costs – i.e. environmental policy and transaction costs

• Secondary ‘non-environmental’ market failures – i.e. information failures, split incentives, network externalities

• ‘Credibility’ of policy-induced price signals over the longer term may not be sufficient for risky investments

• Inertia in the market which can favour incumbent firms and technologies – “deadweight of past” may correlate with environment-intensity”

But – A Necessary but not Necessarily Sufficient Condition

The Need for a Mix of Policies: Sequencing and Complementarity in AFV Technologies

0

1

2

3

4

5

6

7

8

9

10

Electric Hybrid

Standards

Fuel prices

Fuel prices

StandardsPublic R&D Public R&D

Note: For ease of interpretation elasticities have been normalised such that effect of R&D=1. Unfilled bars indicate no statistical significance at 5% level.

Source: OECD (2011) Invention and Transfer of Environmental Technologies.

• To induce a 1% increase in electric vehicle innovations, the alternatives are:

– Increase R&D by 14% (i.e. $26 mln instead of $23 mln per year per country, on average)

– Increase fuel price by 63% (i.e. $1.30 instead of $0.80, on avg)

• To induce a 1% increase in hybrid vehicle innovations, the alternatives are:

– Increase R&D by 53% (i.e. $35 mln instead of $23 mln per year per country, on average)

– Increase fuel price by 5% (i.e. $0.84 instead of $0.80, on avg)

Policy Impacts and Distance from “Market”

Directing Change Without “Picking Winners”

• Since technology-neutral pricing of externality is not ‘sufficient’ = > necessity to be ‘prescriptive’ (at least to some extent) => main challenge for policy makers

• Some general principles:

• Support a ‘portfolio’ of technologies to diversify risk of getting it “wrong”

• Benefits of chosen portfolio should be robust with respect to information uncertainty (i.e. ancillary benefits)

• Identify “general purpose environmental technologies’ which complement a variety of emission-reducing strategies

Breadth of Sources of Environmental Innovation

“A backstop technology is defined as a new technology producing a close substitute to an exhaustible resource by using relatively abundant production inputs and rendering the reserves of the exhaustible resource obsolete when the average cost of production of the close substitute falls below the spot price of the exhaustible resource.”

15

Looking Ahead – Backstop/Breakthrough Technologies

Source: Dasgupta and Heal (1978)

Why Does it Matter?

16

Note: Assumptions concerning backstop technologies based on expenditures on technologies which are under research but not yet viable (e.g. advanced biofuels, nuclear and fuel cells)Source: OECD Economics of Climate Change Mitigation. Based on research undertaken by V. Bosetti et al. (2010) in http://www.feem.it/userfiles/attach/2010471754234NDL2010-042.pdf

Where Are they Going?

17

Note: All Claimed Priorities and Singulars. Extraction from PATSTAT May 2013.

1981

1983

1985

1987

1989

1991

1993

1995

1997

1999

2001

2003

2005

2007

2009

0

200

400

600

800

1000

1200

1400

NULEAR FUSIONCCS

Ideally need to identify trends in costs of ‘green’ technology and relative to ‘brown’ substitute. While imperfect - within patent data can draw upon citation, family, class data. – Emerging: An emerging technology is defined as one in which

prevailing classification system is initially inadequate to identify invention - ‘triangulate’ through use of co-classes until new class is created.

– Radical: A radical patent is a patent that cites previous patents belonging to technological classes other than the one(s) the patent itself belongs to. This suggests that the considered invention builds upon a different paradigm from the one it is applied to.

– Breakthrough: The notion of breakthrough patents captures the extent to which inventions serve as basis for future technological developments. Breakthrough inventions are defined as the top 1% cited patents in a particular year and technology field.

– General: The concept of generality reflects the importance of patents for later technological developments, and the number of fields – i.e. the range of industries .Based on the number and distribution of forward citations and the technological classes these citations belong to.

18

How to Identify?

Sources: DSTI/EAS/IND/WPIA(2012)1 and DSTI/EAS/IND/WPIA(2012)6

Thank You!