Directing Technical Change from Fossil-Fuel to Renewable Energy - - PowerPoint PPT Presentation

▶

Oct 04, 2022 177 likes •408 views

. . Directing Technical Change from Fossil-Fuel to Renewable Energy Innovation An Empirical Investigation Using Patent Data . . . . . Jolle Noailly a and Roger Smeets b a CPB Netherlands Bureau for Economic Policy Analysis b University of

SLIDE 1

. . . . . .

. . . . . . .

Directing Technical Change from Fossil-Fuel to Renewable Energy Innovation

An Empirical Investigation Using Patent Data Joëlle Noaillya and Roger Smeetsb

a CPB Netherlands Bureau for Economic Policy Analysis b University of Amsterdam, the Netherlands

University of Geneva January 24, 2012 J.Noailly@cpb.nl

Joëlle Noailly (CPB) Directing Technical Change in Energy Innovation UNIGE, January 24, 2012 1 / 18

SLIDE 2

. . . . . .

Outline

. . .

1

Motivation . . .

2

Objective . . .

3

Data . . .

4

Empirical strategy and Results . . .

5

Conclusions

Joëlle Noailly (CPB) Directing Technical Change in Energy Innovation UNIGE, January 24, 2012 2 / 18

SLIDE 3

. . . . . .

Motivation

Stern Review (2007):

global average temperatures >2oC by 2035, possibly >5oC by 2100. “Effective action (...) requires a widespread shift to new technology in key sectors such as electricity generation.” (Stern Review, 2007, p. 393)

Electricity generation (IEA, 2010):

41% of world CO2 emissions. 70% of world’s electricity based on fossil-fuels. 19% on renewables.

Fuel shares of world’s electricity generation

Source: IEA, 2010, Key World Energy statistics Joëlle Noailly (CPB) Directing Technical Change in Energy Innovation UNIGE, January 24, 2012 3 / 18

SLIDE 4

. . . . . .

Motivation

Key challenge is to tackle fossil fuels dependency ⇒ shift away from fossil fuels (FF) towards renewable (REN) energy

(Hoffert et al., Science, 2002). Innovation is key to reduce the costs of REN technologies.

Source: IEA, 2011, Projecting costs of electricity generation. Joëlle Noailly (CPB) Directing Technical Change in Energy Innovation UNIGE, January 24, 2012 4 / 18

SLIDE 5

. . . . . .

Motivation

How do we induce more innovation in renewable technologies? 2 market failures. (Jaffe et al., 2005)

Environmental externality Knowledge externatily Path-dependency in innovation firms that have innovated a lot in dirty technologies in the past will continue to do so (lock-in). Recent theoretical work by Acemoglu et al. (2011): Model of directed technical change → research is directed to most profitable sector Path-dependent innovation → past advances in dirty technologies make future clean innovation less profitable Optimal policy mix → carbon tax + directed research subsidies for the clean sector. Towards Green Industrial Policy?

Joëlle Noailly (CPB) Directing Technical Change in Energy Innovation UNIGE, January 24, 2012 5 / 18

SLIDE 6

. . . . . .

Motivation

Green Industrial Policy?

Can the government pick up winners? Edward Glaeser (New York Times, January 2011)

“Evergreen Solar’s move to China was supported by a $33 million loan from the Chinese government. Japan’s success in the 1980s was also attributed to its activist industrial policy, but subsequent research found that government subsidies backed losers more

ften than winners. For each effective government intervention,

there have been dozens, even hundreds, of failures, where public expenditures bore no fruit.”

Philippe Aghion (voxEU.org, November 2010)

“I think it was good that economists in the past 20 years have pointed to the downsides of the top down pick winner policy. Does it mean you should have no sectoral policy? I believe not because we know that under laissez faire if you don’t intervene, firms will innovate

dirty. So you have to step in and say we will induce you to innovate

clean.”

Joëlle Noailly (CPB) Directing Technical Change in Energy Innovation UNIGE, January 24, 2012 6 / 18

SLIDE 7

. . . . . .

Objective

Research question: What are the factors that induce firms to shift away from fossil fuels towards renewable innovation?

⇒ How important is path-dependency?

Design: Empirical analysis of the factors affecting the direction of innovation in the electricity generation sector.

Patents in REN and FF technologies for 7,000 European firms over 1978-2006 3 factors: fuel prices, market size and past knowledge stock (Acemoglu et al., 2009)

Key results:

Strong empirical evidence for path-dependency in innovation (especially for large firms with a long history of FF innovation) Gives support to Aghion’s view

Joëlle Noailly (CPB) Directing Technical Change in Energy Innovation UNIGE, January 24, 2012 7 / 18

SLIDE 8

. . . . . .

Objective

Literature

Empirical literature on the determinants of environmental innovations

(Popp, 2002; Johnstone et al, 2010; Hascic et al, 2009; Noailly, 2011)

Popp (2002)

US patent data in 11 energy technologies 1970-1994 Impact of past knowledge stock > energy prices

Aghion et al. (2011)

Compare factors affecting clean vs. dirty innovation Focus on automobile industry Path-dependency in innovation

Our contribution:

Focus on electricity generation Analysis for different types of firms (specialized and mixed firms)

Joëlle Noailly (CPB) Directing Technical Change in Energy Innovation UNIGE, January 24, 2012 8 / 18

SLIDE 9

. . . . . .

Data

Data: Patents

7,000 European firms over 1978-2006 Innovation measured by patents filed at European Patent Office + 17 national offices Sources: PATSTAT (EPO), matched at firm level using the HAN database (OECD). Patents identified using International Patent Classification codes (Lanzi et

al, 2010; Hascic et al, 2009)

REN: solar, wind, marine, hydro, geothermal, biomass and waste FF: steam engines plants, gas turbines plants, hot-gas, steam generation, burners, furnaces and ignition engines

Joëlle Noailly (CPB) Directing Technical Change in Energy Innovation UNIGE, January 24, 2012 9 / 18

SLIDE 10

. . . . . .

Data

Data: Patent trends

Most patenting activities take place in FF technologies Rapid increase of REN patents since mid 1990s

Number of patents per year

500 1000 1500 Number of patents 1980 1985 1990 1995 2000 2005 Year All patents REN patents FF patents

26,000 patents, 82% patents in FF technologies

Joëlle Noailly (CPB) Directing Technical Change in Energy Innovation UNIGE, January 24, 2012 10 / 18

SLIDE 11

. . . . . .

Data

Data: Patent trends

Most patenting activities take place in FF technologies Rapid increase of REN patents since mid 1990s

Patents per country

2,000 4,000 6,000 8,000 Number of patents DE FR CH GB IT SE FI NL AT DK BE LU NO ES IE GR PT REN patents FF patents Joëlle Noailly (CPB) Directing Technical Change in Energy Innovation UNIGE, January 24, 2012 10 / 18

SLIDE 12

. . . . . .

Data

Data: Patent trends

Most patenting activities take place in FF technologies Rapid increase of REN patents since mid 1990s

Patents per type of REN technology

50 100 150 200 Number of patents 1980 1985 1990 1995 2000 2005 Year Wind Solar Geo Marine Hydro Biomass & Waste Joëlle Noailly (CPB) Directing Technical Change in Energy Innovation UNIGE, January 24, 2012 10 / 18

SLIDE 13

. . . . . .

Data

Data: Patent trends

Most patenting activities take place in FF technologies Rapid increase of REN patents since mid 1990s

Patents per type of FF technology

100 200 300 400 Number of patents 1980 1985 1990 1995 2000 2005 Year Coal Engines Turbines Hotgas Steam Burners Furnaces Ignition Joëlle Noailly (CPB) Directing Technical Change in Energy Innovation UNIGE, January 24, 2012 10 / 18

SLIDE 14

. . . . . .

Data

Data: Mixed firms vs. specialized firms

Mixed firms (both REN and FF patents) → shift from FF to REN by

redirecting their innovation efforts

Specialized firms (only REN or FF patents) → shift from FF to REN mainly

by entry of new REN innovators Number of firms per type

100 200 300 400 Number of firms 1980 1985 1990 1995 2000 2005 Year REN firms FF firms Mixed firms

7,000 firms, 5% mixed, 30% REN firms, 65% FF firms

Joëlle Noailly (CPB) Directing Technical Change in Energy Innovation UNIGE, January 24, 2012 11 / 18

SLIDE 15

. . . . . .

Data

Data: Firms

Specialized REN firms account for 80% of all REN patents, but innovate

nly occasionally.

Mixed firms are large persistent innovators, mainly in FF technologies (50% of all FF patents)

Innovation frequency (years of innovation) Firmtype Mean

St. Dev.

Min. Med. Max. FF 1.8 2.1 1 1 28 REN 1.2 0.8 1 1 12 Mixed 6.2 5.9 1 4 29 Innovation in mixed firms... FF 5.2 6 1 3 29 REN 1.9 1.8 1 1 14

Mixed firms look for complementarities between REN and FF technologies (e.g. high correlation burners/biomass)

Joëlle Noailly (CPB) Directing Technical Change in Energy Innovation UNIGE, January 24, 2012 12 / 18

SLIDE 16

. . . . . .

Data

Data: Factors affecting innovation

Fossil fuel prices

Prices of oil for electricity generation in USD/toe per country (IEA, Energy Prices Database) Construction of firm-specific prices based on firms’ patent portfolio (weighted by countries in which patents have been filed)

Market size

Electricity output (in GWh) from renewable and fossil-fuel energy sources (IEA, Energy Statistics database) Construction of firm-specific market sizes based on firms’ patent portfolio (weighted by countries and technologies)

Knowledge stocks (path-dependency in innovation)

Cumulative number of FF (REN) patents over time at the firm level Knowledge becomes obsolete over time (depreciated by 15% annually).

Joëlle Noailly (CPB) Directing Technical Change in Energy Innovation UNIGE, January 24, 2012 13 / 18

SLIDE 17

. . . . . .

Empirical strategy and Results

Empirical strategy: Which factors affect the rate of REN and FF innovation?

Estimate rate of innovation = firm-level patent count in REN or FF technologies. Count data Poisson estimation (Hausman et al, 1984)

E(Pijkt|Xijkt,ηi,υk,νt) = exp(β0i + β1 logpit−1 + β2j logMijt−1 + β3j logKSijt−1

+ ηi + υk + νt)

Zero-inflated Poisson estimation. Countries and years fixed effects, SE clustered at firm-level. Firms’ fixed effects captured by firms’ patent stock in the pre-sample period (1950-1978) (Blundel et al, 1995).

Robustness analysis → excluding single innovators, various time periods and technology-specific specifications

Joëlle Noailly (CPB) Directing Technical Change in Energy Innovation UNIGE, January 24, 2012 14 / 18

SLIDE 18

. . . . . .

Empirical strategy and Results

Main results

Dependent variable REN patents FF patents (1) (2)

spec. REN firms

spec FF firms FF price 0.983* 0.636* (0.356) (0.183) REN knowledge stock 0.761* (0.029) FF knowledge stock 0.663* (0.026) REN market size 0.085*** 0.002 (0.013) (0.014) FF market size

0.019

0.093*** (0.045) (0.014) Year fixed effects Yes Yes Country fixed effects Yes Yes Firm fixed effects Yes Yes Number of observations 50,070 117,175 Log-likelihood

10,393
34,952

Higher FF prices → increasing REN and FF (not always robust) patenting activities Evidence for path-dependency in the type

f innovation

Small positive effect of market size on innovation.

Joëlle Noailly (CPB) Directing Technical Change in Energy Innovation UNIGE, January 24, 2012 15 / 18

SLIDE 19

. . . . . .

Empirical strategy and Results

Main results

Dependent variable REN patents FF patents REN patents FF patents (1) (2) (3) (4)

spec. REN firms

spec FF firms mixed firms mixed firms FF price 0.983* 0.636* 0.316

0.321

(0.356) (0.183) (0.671) (0.823) REN knowledge stock 0.761* 0.258*

0.230***

(0.029) (0.087) (0.072) FF knowledge stock 0.663* 0.161 0.747* (0.026) (0.070) (0.069) REN market size 0.085* 0.002 0.031

0.014

(0.013) (0.014) (0.022) (0.019) FF market size

0.019

0.093***

0.096*

0.122 (0.045) (0.014) (0.051) (0.096) Year fixed effects Yes Yes Yes Yes Country fixed effects Yes Yes Yes Yes Firm fixed effects Yes Yes Yes Yes Number of observations 50,070 117,175 9,501 9,501 Log-likelihood

10,393
34,952
2,803
7,835

Joëlle Noailly (CPB) Directing Technical Change in Energy Innovation UNIGE, January 24, 2012 15 / 18

SLIDE 20

. . . . . .

Empirical strategy and Results

Additional results: Which factors induce a shift away from FF towards REN?

Mixed firms: shift from FF to REN by redirecting their innovation efforts (within-firm substitution)

Estimate relative innovation (ratio REN/FF patents) Panel subsample of mixed firms Dependent variable: log

(

PRENit PFFit

)

FE panel estimation, robust clustered SE

Dependent variable log(REN/FF) (1) FF price 0.091 (0.065) REN knowledge stock

0.017

(0.015) FF knowledge stock

0.191***

(0.041) REN market size

0.005

(0.007) FF market size 0.021 (0.029) Year fixed effects Yes Country fixed effects Yes Number of observations 9,501 Number of firms 361

Joëlle Noailly (CPB) Directing Technical Change in Energy Innovation UNIGE, January 24, 2012 16 / 18

SLIDE 21

. . . . . .

Empirical strategy and Results

Additional results: Which factors induce a shift away from FF to REN innovation?

Specialized firms: shift from FF to REN mainly by entry of new REN innovators

Estimate probability to file first innovation in REN (rather than in FF) Cross-section subsample of specialized firms innovating for the first time (no past knowledge stocks) Dependent variable: REN=1 (0) if the new innovator is a REN (FF) firm Estimation by probit model, robust clustered SE

Dependent variable first REN (1) FF price 0.374* (0.142) REN market size 0.230* (0.043) FF market size

0.458***

(0.110) Year fixed effects Yes Country fixed effects Yes Number of firms 5,907

Joëlle Noailly (CPB) Directing Technical Change in Energy Innovation UNIGE, January 24, 2012 17 / 18

SLIDE 22

. . . . . .

Conclusions and policy implications

Conclusions

Findings and policy implications REN innovation can be induced via an increase in FF (carbon) prices and REN market size Path-dependency in innovation → justifies R&D subsidies specifically targetted at REN innovation Focus on facilitating entry by new REN innovators. Green industrial policy? should not discourage entry and competition between firms (Aghion and Dewatripont, 2011). Future work Data on REN subsidies (feed-in tariffs) Knowledge spillovers across firms Why do FF firms start innovating in REN?

Joëlle Noailly (CPB) Directing Technical Change in Energy Innovation UNIGE, January 24, 2012 18 / 18