“Endogenizing
Technological Change: Matching Empirical Evidence to Modeling Needs,”
forthcoming in Energy Economics (with
William A. Pizer).
Abstract
Given that technologies to significantly reduce fossil
fuel emissions are currently unavailable or only available at high cost,
technological change will be a key component of any long-term strategy to
reduce greenhouse gas emissions. In light of this, the amount of research on
the pace, direction, and benefits of environmentally-friendly technological
change has grown dramatically in recent years. This research includes
empirical work estimating the magnitude of these effects, and modeling
exercises designed to simulate the importance of endogenous technological
change in response to climate policy. Unfortunately, few attempts have been
made to connect these two streams of research. This paper attempts to bridge
that gap. We review both the empirical and modeling literature on
technological change. Our focus includes the research and development
process, learning by doing, the role of public versus private research, and
technology diffusion. Our goal is to provide an agenda for how both
empirical and modeling research in these areas can move forward in a
complementary fashion. In doing so, we discuss both how models used for
policy evaluation can better capture empirical phenomena, and how empirical
research can better address the needs of models used for policy evaluation.
"The Effects of
Innovation Policies on Private R&D Investment: A Cross-national Empirical Study,”
Economics of Innovation and New
Technology, June 2007, 16(4), 237–253 (with Yonghong Wu and Stuart
Bretschneider).
Abstract
This paper examines the effect of three major national
innovation policies (patent protection, research and development (R&D) tax
incentives, and government funding of business R&D) on business R&D
spending. Unlike previous work, we also consider the effect of openness to
international trade. We use data from nine OECD countries (Australia,
Canada, France, Germany, Italy, Japan, Spain, UK, and USA) in 1985-1995. Our
results show that all three innovation policies play a significant role in
stimulating business R&D. Enforcement of patent right matters most to
business R&D spending. In addition, R&D performed by the government has a
positive effect on business R&D, whereas R&D by the higher education sector
has a negative impact on business R&D. We also find modest empirical support
to the positive role of openness to international trade in business R&D
investment.
"They Don't Invent Them Like They
Used To: An Examination of Energy Patent Citations Over Time," Economics
of Innovation and New Technology, 15(8), November 2006, 753-776.
Abstract
This paper uses patent citation data to study flows of knowledge across time
and across institutions in the field of energy research. Popp (2002) finds
the level of energy-saving R&D depends not only on energy prices, but also
on the quality of the accumulated knowledge available to inventors. Patent
citations are used to represent this quality. This paper explores the
pattern of citations in these fields more carefully. I find evidence for
diminishing returns to research inputs, both across time and within a given
year. To check whether government R&D can help alleviate potential
diminishing returns, I pay special attention to citations to government
patents. Government patents filed in or after 1981 are more likely to be
cited. More importantly, descendants of these government patents are 30
percent more likely to be cited by subsequent patents. Earlier government
research was more applied in nature and is not cited more frequently.
Appendix: Patent classifications
used in this paper
"Innovation
in Climate Policy Models: Implementing Lessons from the Economics of R&D,"
Energy Economics,
28(5-6), November 2006, 596-609.
Abstract
Only recently
have economists considered the effect of induced innovation in climate
policy models. One reason is that, until recently, empirical evidence of
the magnitude of such effects was unavailable. Drawing on my experiences
with empirical studies on innovation and from modeling the climate change
problem, in this paper I present key lessons from the empirical literature
on innovation and environmental policy, and discuss how much of the
variation in results found in the modeling literature can be explained by
differences in implementing (or failing to implement) these lessons into
climate models. The paper concludes with a discussion of future research
needs, focusing on a framework for improving the modeling of technology
diffusion in climate change models.
"R&D Subsidies and Climate
Policy: Is There a 'Free Lunch'?",
Climatic Change,
77(3-4), August 2006, 311-341.
Abstract
Because of the long-term
nature of the climate problem, technological advances are often seen as
an important component of any solution. However, when considering the
potential for technology to help solve the climate problem, two market
failures exist which lead to underinvestment in climate-friendly R&D:
environmental externalities and the public goods nature of new
knowledge. As a result, government subsidies to climate-friendly R&D
projects are often proposed as part of a policy solution. Using the
ENTICE model, I analyze the effectiveness of such subsidies, both with
and without other climate policies, such as a carbon tax. While R&D
subsidies do lead to significant increases in climate-friendly R&D, this
R&D has little impact on the climate itself. Subsidies address the
problem of knowledge as a public good, but they do not address the
environmental externality, and thus offer no additional incentive to
adopt new technologies. Moreover, high opportunity costs to R&D limit
the potential role that subsidies can play. While R&D subsidies can
improve efficiency, policies that directly affect the environmental
externality have a much larger impact on both atmospheric temperature
and economic welfare.
Appendix: Equations and calibration
of the ENTICE-BR model
"Comparison
of Climate Policies in the ENTICE-BR Model", Energy Journal, Special
Issue: Endogenous Technological Change and the Economics of Atmospheric
Stablilisation, 2006, 163-174.
Abstract
This paper uses the ENTICE-BR model to study the effects of various climate
stabilization policies. Because the ENTICE-BR model includes benefits from
reduced climate damages, it is possible to calculate the net economic impact
of each policy. In general, only the least restrictive concentration limit
is welfare enhancing. While the policies are welfare enhancing in
simulations using optimistic assumptions about the potential of the backstop
energy technology, such assumptions mean that the backstop is also used in
the no-policy base case, so that climate change itself is less of a problem.
Finally, assumptions about the nature of R&D markets are important. Removing
the assumption of partial crowding out from energy R&D nearly doubles the
gains from policy-induced energy R&D.
"The
Transition to Endogenous Technical Change in Climate-Economy Models: A Technical
Overview to the Innovation Modeling Comparison Project," Energy Journal,
Special Issue: Endogenous Technological Change and the Economics of Atmospheric
Stablilisation, 2006, 17-55 (with Jonathan Köhler, Michael Grubb, and Ottmar
Edenhofer).
Abstract
This paper assesses endogenous technical change (ETC) in climate-economy
models, using the models in the Innovation Modeling Comparison Project as a
representative cross-section. ETC is now a feature of most leading models.
Following the new endogenous growth literature and the application of
learning curves to the energy sector, there are two main concepts employed:
knowledge capital and learning curves. The common insight is that technical
change is driven by the development of knowledge capital and its
characteristics of being partly non-rival and partly non-excludable. There
are various different implementations of ETC. Recursive CGE models face
particular difficulties in incorporating ETC and increasing returns. The
main limitations of current models are: the lack of uncertainty analysis,
the limited representation of the diffusion of technology and the
homogeneous nature of agents in the models, including the lack of
representation of institutional structures in the innovation process.
"ENTICE-BR: The Effects
of Backstop Technology R&D on Climate Policy Models", Energy Economics,
March 2006, 28(2), 188-222.
Abstract
Recent attempts to endogenize technology in climate policy models have
produced mixed results. Models including alternative technologies
find large gains from induced technological change. However, technological
progress in these models comes through learning-by-doing, which ignores
the potential opportunity costs of technological change. Models using
R&D spending as the driver of technological change address this.
However, since these models typically include only a single representative
energy technology, substitution across technologies is not possible.
This paper addresses these shortcomings by including policy-induced energy
R&D in a model with a backstop energy technology. I show that,
while induced technological change is important, larger welfare gains come
from simply adding an alternative technology to the model. As in
models with a single technology, opportunity costs of research limit the
role induced innovation can play. Moreover, since the backstop technology
improves welfare even without climate policy, accurate policy analysis
depends on a carefully constructed baseline simulation.
"International Innovation and
Diffusion of Air Pollution Control Technologies: The Effects of NOX and SO2
Regulation in the US, Japan, and Germany", Journal of Environmental
Economics and Management, 51(1), 46-71, January 2006.
Abstract
Using patent data from the United States, Japan, and Germany, this paper
examines both the innovation and diffusion of air pollution control
equipment. Whereas the United States was an early adopter of stringent
sulfur dioxide (SO2) standards, both Japan and Germany introduced stringent
nitrogen dioxide (NOX) standards much earlier than the US. Nonetheless, in
both cases, tightened standards in the U.S. led to more domestic patenting,
but not more foreign patenting. Overall, the data suggest that inventors
respond to environmental regulatory pressure in their own country, but not
to foreign environmental regulations. Moreover, any technology transfer that
occurs appears to be indirect. Domestic innovation occurs even for
technologies that have already experienced significant innovative activity
abroad. Moreover, utilities in countries that adopt regulations later
nonetheless purchase pollution abatement equipment from domestic firms.
However, patent citation data from the U.S. show that earlier foreign
patents are an important building block for NOX pollution control
innovations in the U.S., suggesting that American inventors build on
technological advances made in countries that adopted stringent regulation
earlier.
"Lessons From Patents: Using
Patents to Measure Technological Change: in Environmental Models",
Ecological Economics, 54(2-3), 209-226, August 2005.
Abstract
When studying solutions to long-term environmental problems such as climate
change, it is important to consider the role that technological change may
play. Nonetheless, to date few economic models of environmental policy
explicitly model the link between policy and technological change. There is
a growing body of evidence that the incentives offered by prices and
environmental regulations have a strong influence on both the creation and
adoption of new technologies. In several recent papers, I have used patent
data to examine the links between environmental policy and technological
change. In addition, I have used the results of this research to calibrate
the ENTICE model (for ENdogenous Technological change) of climate change,
which links energy-related R&D to changes in the price of carbon. Drawing on
my experiences from empirical studies on innovation and from modeling the
climate change problem, in this paper I review some of the key lessons from
recent empirical work using patents to study environmental innovation and
diffusion, and discuss its implications for modeling climate change policy.
I conclude by offering suggestions for future research.
"Uncertain
R&D and the Porter Hypothesis", Contributions to Economic Analysis
& Policy, 4(1), Article 6, 2005.
Abstract
Ever since Michael Porter proposed that environmental
regulations can improve competitiveness, much economic research has examined
the potential for such outcomes. Attempts to model Porter hypothesis
outcomes in a way consistent with neoclassical economics have focused on
things such as strategic relationships between firms, moral hazard problems,
and economies of scale. In this paper, I offer a simpler alternative. The
results of any R&D project are uncertain. Calibrating a simple model of
induced R&D with uncertainty so that the expected value of research is only
positive with environmental policy, I find that between 8 and 24 percent of
simulations result in cases where post-regulation profits are higher than
pre-regulation profits. This result is consistent both with Porter finding
specific cases with complete innovation offsets and with macro-level
findings that environmental policy is not costless. I conclude by discussing
the implication of these results for environmental policy and future
research.
"Institutions and Intellectual
Property: The Influence of Institutional Forces on University Patenting",
(with Yixin Dai and Stuart Bretschneider), Journal of Policy Analysis and Management,
24(3), Summer 2005, 579-598.
Abstract
Over the past 20 years, the number of patents
assigned to universities has increased dramatically. This increase coincided
with several policy initiatives, such as the Bayh-Dole Act of 1980, designed
to foster technology transfer between universities and the private sector.
This paper examines the effect of such policies using an institutional
framework, designed to illustrate how factors both from inside and outside
of academia influence the decision to patent university research. We find
passage of the Bayh-Dole Act spurred university patenting, but did not
induce additional applied research funding. Thus, Bayh-Dole fostered
technology transfer, but did not result in more applied research at
universities.
"Time In Purgatory:
Examining the Grant Lag for U.S. Patent Applications", (with Ted Juhl and
Daniel K.N. Johnson), Topics in Economic Analysis & Policy, 4(1), Article
29, 2004.
Abstract
As patent applications increase, and the range of patentable technologies
increases, the length of time it takes for an invention to go through the
examination process at the U.S. Patent Office has increased. Concerns over
the distributional effects of these changes have been expressed during
policy debates. We use data on U.S. patent applications and grants to ask
who is affected by longer grant lags. We augment this analysis with
interviews of patent examiners, leading to a better understanding of the
examination process. Our analysis finds that differences across technology
are most important. These differences do not erode over time, suggesting
that learning effects alone will not reduce grant lags. Inventor
characteristics have statistically significant effects, but the magnitudes
are small.
"ENTICE:
Endogenous Technological Change in the DICE Model of Global Warming",
Journal of Environmental Economics and Management, 48(1), July 2004,
742-768.
Abstract
Despite growing empirical evidence of the link between environmental
policy and innovation, most economic models of environmental policy treat
technology as exogenous. For long-term problems such as climate change,
this omission can be significant. In this paper, I modify the DICE
model of climate change (Nordhaus 1994, Nordhaus and Boyer 2000) to allow
for induced innovation in the energy sector. Ignoring induced technological
change overstates the welfare costs of an optimal carbon tax policy by
8.3 percent. However, cost-savings, rather than increased environmental
benefits, appear to drive the welfare gains, as the effect of induced innovation
on emissions and mean global temperature is small. Sensitivity analysis
shows that potential crowding out of other R&D and market failures
in the R&D sector are the most important limiting factors to the potential
of induced innovation. Differences in these key assumptions explain
much of the variation in the findings of other similar models.
"Pollution Control Innovations
and the Clean Air Act of 1990", Journal of Policy Analysis and Management,
22(4), Fall 2003, 641-660.
Abstract
One advantage often cited for market-based environmental policies is
that they are more likely to promote technological innovation than command
and control regulations. This paper uses patent data to study innovation
in flue gas desulfurization units (“scrubbers”) across these policy regimes.
Using plant level data to study the effect of these patents on pollution
control, I find the effect of these patents changed after passage of the
1990 Clean Air Act, which instituted a market for sulfur dioxide (SO2)
permits. Previous regulations requiring plants constructed before
1990 to install scrubbers created incentives for innovation that lowered
the costs of operating scrubbers, but did little to improve the environmental
effectiveness of the technology. In comparison, innovations occurring
since 1990 do serve to improve the removal efficiency of scrubbers.
"Forced
Out of the Closet: The Impact of the American Inventors Protection Act on the
Timing of Patent Disclosure," (with Daniel Johnson), RAND Journal
of Economics, 34(1), Spring 2003, pp. 96-112.
"Induced Innovation and
Energy Prices," American Economic Review, 92(1), March 2002,
pp. 160-180.
Abstract
I use U.S. patent data from 1970 to 1994 to estimate the effect of
energy prices on energy-efficient innovations. Using patent citations to
construct a measure of the usefulness of the existing base of scientific
knowledge, I consider the effect of both demand-side factors, which spur
innovative activity by increasing the value of new innovations, and supply-side
factors, such as scientific advancements that make new innovations possible.
I find that both energy prices and the quality of existing knowledge have
strongly significant positive effects on innovation. Furthermore, I show
that omitting the quality of knowledge adversely affects the estimation
results. (JEL O31, Q40, Q42)
"Altruism and the Demand for Environmental
Quality," Land Economics, 77(3), August 2001, pp. 339-349.
Abstract
This paper asks whether individuals consider the value future generations
will receive from environmental quality when deciding what level of environmental
protection to provide. Using data on life expectancy, I develop two
tests for altruism towards future generations. One, a test for strong
altruism, asks whether individual motives are purely altruistic when deciding
to provide environmental quality. The second, a test for weak altruism,
combines an individual’s concern for both self-interest and the interest
of future generations. Using data from a Washington Post survey on
environmental attitudes to implement the test, I find evidence of weak
altruism.
Data appendix to "Altruism and
the Demand for Environmental Quality"
"The
Effect of New Technology on Energy Consumption," Resource and Energy
Economics, 23(3), July 2001, pp. 215-239.
Abstract
This paper uses patent data to estimate the effect of new technologies
on energy consumption. Matching energy patent counts to the industries
using these patents, I create stocks of energy knowledge for 13 industries.
Including the stocks in restricted variable cost functions, I estimate
the median present value of long-run savings from a new patent to be over
14.5 million dollars. Combining these results with estimates of price-induced
innovation, I conclude that two-thirds of the change in energy consumption
with respect to a price change is due to simple price-induced factor substitution,
while the remaining third results from induced innovation.
Data appendix to "The Effect of New Technology
on Energy Consumption"
Results appendix to "The Effect
of New Technology on Energy Consumption"
"What is the Value of Scientific Knowledge? An Application to Global
Warming Using the PRICE Model," with William Nordhaus, The Energy Journal,
18(1), January 1997, pp. 1-45.
Abstract
Governments must cope with the enormous
uncertainties about both future climate change as well as the costs and benefits
of slowing climate change. This study analyses the value of improved information
about a variety of geophysical and economic processes. The value of information
is estimated using the "PRICE model" which is a probabilistic extension of
earlier models of the economics of global warming. The study uses five different
approaches to estimating the value of information about all uncertain parameters
and about individual parameters. It is estimated that the value of early
information is between $1 and $2 billion for each year that resolution of
uncertainty is moved toward the present. We estimate that the most important
uncertain variables are the damages of climate change and the costs of reducing
greenhouse gas emissions. Resolving the uncertainties about these two parameters
would contribute 75 percent of the value of improved knowledge.
“Renewable
Energy Policies and Technological Innovation: Evidence Based on Patent Counts,”
January 2008, NBER Working Paper #13760, under review at
Environmental and Resource Economics
(with Nick Johnstone and Ivan Hascic).
Abstract
This paper examines the effect of environmental policies on technological
innovation in the specific case of renewable energy. The analysis is
conducted using patent data on a panel of 25 countries over the period
1978-2003. It is found that public policy plays a significant role in
determining patent applications. Different types of policy instruments are
effective for different renewable energy sources.
“Information Disclosure Policy: Do States’
Data Processing Efforts Help More than the Information Disclosure Itself?”
October 2007, (with Hyunhoe Bae and Peter Wilcoxen).
Abstract
Although the disclosure of Toxic Release Inventory (TRI) is expected to
reduce toxic risks by lowering transaction costs for collective action by
the public, the massiveness and complexity of the raw TRI data might result
in information overload, whereby the information recipient can face
difficulties in extracting the relevant information and interpreting it
appropriately. This study examined whether the states’ data processing
efforts played a more critical role for achieving a designated policy goal
than states’ simple data dissemination efforts, by mitigating the
information processing burden of the public or information recipients. The
estimation results show that the states’ TRI data dissemination efforts were
effective at lowering the total quantity of toxic release but were not
effective at lowering the actual health risk due to toxic release, which is
the true policy goal. States’ data processing efforts, in contrast, had a
significant impact on lowering health risk
level due to toxic release. This implies that the states’ data processing
efforts were critical in achieving the designated policy goal.
“Policy
vs. Consumer Pressure: Innovation and Diffusion of Alternative Bleaching
Technologies in the Pulp Industry,” September 2007, NBER Working Paper
#13439, (with
Tamara Hafner and Nick Johnstone).
Abstract
In the late 1980s and early
1990s, concern over dioxin in both paper products and wastewater led to the
development of techniques that reduced the use of chlorine in the pulp
industry. Both regulatory and
consumer pressure motivated this change.
We use patent data to examine the evolution of two completing
bleaching technologies in five major paper-producing countries, both of
which reduce the use of chlorine in the pulping process.
By the end of the 1990s, nearly all pulp production in these
countries used one of these technologies.
Unlike other papers using patents to study environmentally-friendly
innovation, we focus on a process innovation, rather than on end-of-the-pipe
solutions to pollution.
Moreover, while previous studies emphasize the importance of regulation for
inducing innovation, here we find substantial innovation occurring before
regulations were in place.
Instead, pressure from consumers to reduce the chlorine content of paper
drives the first round of innovation.
However, while some companies choose to adopt these technologies in
response to consumer pressure, not all firms will differentiate their
product in this way. Thus,
governments need to regulate if their goal is broad diffusion of the
environmental technology.
"Exploring Links Between
Innovation and Diffusion: Adoption of NOX Control Technologies at
U.S. Coal-fired Power Plants," NBER Working Paper #12119, March 2006.
Abstract
While many studies have looked at innovation and adoption of technologies
separately, the two processes are linked. Advances (and expected advances)
in a single technology should affect both its adoption rate and the adoption
of alternative technologies. Moreover, advances made abroad may affect
adoption differently than improvements developed domestically. This paper
combines plant-level data on U.S. coal-fired electric power plants with
patent data pertaining to NOX pollution control techniques to
study these links. I show that technological advances, particularly those
made abroad, are important for the adoption of newer post-combustion
treatment technologies, but have little effect on the adoption of older
combustion modification techniques. Moreover, I provide evidence that
adaptive R&D by U.S. firms is necessary before foreign innovations are
adopted in the U.S. Expectations of future technological advances delay
adoption. Nonetheless, as in other studies of environmental technologies,
the effect of other explanatory variables is dominated by the effect of
environmental regulations, demonstrating that the mere presence of
environmental technologies is not enough to encourage its usage.
"Knowledge Spillovers in Interdependent
Economies" (with Yonghong Wu and Stuart Bretschneider), May 2001.
Note: because of size constraints, Table 4 is not included.
Abstract
In this paper, we improve upon Coe and Helpman’s model of international
R&D spillovers, using seemingly unrelated regression (SUR) to include
interdependence among national economies and allow for variations in coefficients
across countries. We find that the impact of knowledge spillovers on national
productivity is context dependent: positive in some cases while negative
in others. From our interpretation, the results suggest that both beneficial
and competitive effects from foreign knowledge spillovers are important.
We view the most important contribution of our work as simply providing
evidence of this variation, and suggesting directions for future research
to explain this phenomenon.