A Tale of Two Market Failures : TECHNOLOGY AND ENVIRONMENTAL POLICY

October 2004 RFF DP 04-38


A Tale of Two
Market Failures
Technology and Environmental Policy

A d a m B . J a f f e , R i c h a r d G . N e w e l l , a n d R o b e r t N .
S t a v i n s


1616 P St. NW
DISCUSSION PAPER Washington, DC 20036
202-328-5000
www.rff.org

---

A TALE OF TWO MARKET FAILURES:
TECHNOLOGY AND ENVIRONMENTAL POLICY



Adam B. Jaffe
Brandeis University and National Bureau of Economic Research

Richard G. Newell
Resources for the Future

Robert N. Stavins
Harvard University and Resources for the Future



© 2004 Resources for the Future. All rights reserved. No portion of this paper may be reproduced without
permission of the authors.
Discussion papers are research materials circulated by their authors for purposes of information and
discussion. They have not necessarily undergone formal peer review or editorial treatment.



Corresponding Author:

Richard Newell
Resources for the Future
1616 P St NW
Washington, DC 20036
(202) 328-5111
newell@rff.org


Key words: technology, research and development, environment, externality, policy
JEL codes: O38; Q28; H23

---

A TALE OF TWO MARKET FAILURES:
TECHNOLOGY AND ENVIRONMENTAL POLICY


Abstract

Market failures associated with environmental pollution interact with market failures associated
with the innovation and diffusion of new technologies. These combined market failures provide a
strong rationale for a portfolio of public policies that foster emissions reduction as well as the
development and adoption of environmentally beneficial technology. Both theory and empirical
evidence suggest that the rate and direction of technological advance is influenced by market and
regulatory incentives, and can be cost-effectively harnessed through the use of economic-
incentive based policy. In the presence of weak or nonexistent environmental policies,
investments in the development and diffusion of new environmentally beneficial technologies
are very likely to be less than would be socially desirable. Positive knowledge and adoption
spillovers and information problems can further weaken innovation incentives. While
environmental technology policy is fraught with difficulties, a long-term view suggests a strategy
of experimenting with policy approaches and systematically evaluating their success.

1. INTRODUCTION
The influence of technological development on energy and environmental systems
permeates discussions of energy and environmental policy. New technology has been credited
with solving environmental problems by mitigating the effects of pollutants, and has been
maligned as a source of increased pollution. For modeling long-term environmental problems
such as global climate changes, the effects of technological change compounded over long time
horizons will likely be large. Thus, the single largest source of difference among modelers’
predictions of the cost of climate policy is often differences in assumptions about the future rate
and direction of technological change (Clark and Weyant 2002; Carraro, van der Zwaan, and
Gerlagh 2003; Energy Modeling Forum 1996).
But technological change does not exist in a vacuum. Environmental policy interventions,
such as carbon cap and trade systems and carbon taxes, generate incentives that will affect which
new technologies will be developed and how rapidly and deeply they will diffuse. The induced

1

---

effects of environmental policy on technology can therefore have substantial implications for the
normative analysis of policy. While researchers dispute the extent to which environmental
policy-induced technological change reduces the social cost of environmental compliance, there
is little dispute among economists that flexible, incentive-oriented policy approaches are more
likely to foster low-cost compliance paths than prescriptive regulatory approaches.1
The realization that the process of technological change is itself characterized by market
failures complicates policy analysis, and increases the likelihood that a portfolio of policies,
rather than policy directed at emissions reduction alone, will offer a more complete response to
environmental problems. The seeming intractability of some energy and environmental
problems, such as global climate change, combined with considerable uncertainties and the long
time frame over which their ultimate consequences will play out, may make the development and
deployment of new technologies attractive as a major policy response. That is, policies whose
purpose is generating technological change are likely to be important parts of the policy portfolio
for addressing certain environmental problems, in addition to the rules and regulations we
normally think of as environmental policies. Technology policy can be a costly approach,
however, if it is used as a substitute for, rather than complement, to environmental policy.
Environmental policy targeted directly at emissions (for example through an emissions tax or
cap-and-trade system) will still typically provide the most important single element of a cost-
effective environmental policy strategy.
This paper provides background for consideration of these issues. We begin by
discussing the key analytic issues that permeate policy discussions occurring at the nexus

1 For a detailed survey of the influence of environmental policies on innovation and diffusion see Jaffe, Newell and
Stavins (2003).

2

---

between technology and environmental policy. Section 3 discusses the possibilities for policies
designed to operate directly on technology to improve our ability to cope with environmental
problems. We offer concluding observations in Section 4.
2. KEY
ANALYTICAL
ISSUES
2.1. Fundamentals of Environmental Economics
Economic analysis of environmental policy is based on the idea that the potentially
harmful consequences of economic activities on the environment constitute an “externality,” an
economically significant effect of an activity, the consequences of which are borne (at least in
part) by a party or parties other than the party that controls the externality-producing activity. A
factory that pollutes the air, water, or land imposes a cost on society. The firm that owns the
factory has an economic incentive to use only as much labor or steel as it can productively
employ, because those inputs are costly to the firm. The cost to society of having some of its
labor and steel used up in a given factory is “internalized” by the firm, because it has to pay for
those inputs. But the firm does not have an economic incentive to minimize the “external” costs
of pollution.
Environmental policies attempt to equalize this imbalance by raising the incentive for a
firm to minimize these externalities. Policy choices accomplish this in one of two general
ways—either by internalizing the environmental costs so polluters make their own decisions
regarding their consumption of environmental inputs, or by imposing a limit on the level of
environmental pollution.
The cost of environmental policies could be in the form of decreased output of desired
products (for example, a scrubber on an electric power plant reduces its electricity production
from a given quantity of fuel), increased use of other variable inputs (for example, eliminating

3

---

certain gases from the waste stream in a smokestack may require more fuel to be burned),
purchase of specialized pollution control equipment (for example, catalytic converters on
automobiles), or substitution of inferior or more expensive products or production methods to
avoid pollution-causing products or methods (for example, less effective pesticides used when
DDT was banned).
In the short run, setting an efficient environmental policy requires a comparison of the
marginal cost of reducing pollution with the marginal benefit of a cleaner environment. All else
being equal, emissions of pollutants that are very harmful should be greatly restricted, because
the pollutants otherwise produce large marginal costs to society. But, all else being equal,
emissions of pollutants that are very costly to eliminate should be tolerated, because the marginal
cost of reducing them is high.
When technology enters the equation, the terms of the tradeoff between the marginal cost
of pollution control and its marginal social benefit is altered. In particular, technology
innovations—such as new pollution control equipment, cleaner production methods, or new
substitutes for environmentally harmful products—typically reduce the marginal cost of
achieving a given unit of pollution reduction. This means that a specified level of environmental
cleanup can be achieved at lower total cost to society, and it also means that a lower total level of
pollution can be attained more efficiently than would be expected if the cost of cleanup were
higher. Thus, in this simple static picture, technology improvements can be good for the
environment and good for the firm that must meet environmental mandates.
2.2. Fundamentals of the Economics of Technology
In this simple analytic scenario, the technology innovation results in greater overall social
benefit because the cost of reducing pollution has decreased and environmental health has

4

---

improved. If this were the end of this static story, than the only effect would be to convert the
analysis of environmental policy from a static cost/benefit tradeoff to a dynamic one. Policies to
reduce pollution have two effects, however—they reduce pollution today, and they also typically
change the incentives that firms face with regard to investing resources in developing new
technology for the future. In particular, when firms face an incentive to reduce their emissions,
this simultaneously creates an incentive for them to find ways to reduce pollution at lower cost.
The fact that the development of such technology will, over time, change the pollution
benefit/cost calculus means that choosing efficient environmental policy requires an analysis of
this dynamic interaction. The simple static model does not take into account the fact that new
technology is itself not free.
To reach the point where pollution is being reduced or some other benefit is realized, two
things must happen, both of which require the investment of resources. The first step—
innovation—involves scientific or engineering research to establish a new technical idea and to
develop that idea into a commercial product or process.2 The second step—adoption (or
diffusion)—is the process by which a new product or process gradually replaces older
technology throughout many firms and applications. Adoption is also costly, because firms must
learn about new technology, purchase new equipment, and adapt it to their particular
circumstances. If technological change is not free, can we expect Adam Smith’s “invisible hand”
to choose the right level of investment in both innovation and diffusion of new technology?

2 Schumpeter (1942) identified three steps in technological change: invention, innovation and diffusion. In the
Schumpeterian trichotomy, invention is the first technical development, and innovation the first commercial
introduction. For simplicity, we have collapsed these two steps into one and labeled it innovation.

5

---

The problem compounds, because independent of the externality associated with
pollution, innovation and diffusion are both characterized by externalities as well as other market
failures.
Knowledge Externalities. In the case of pollution as an externality, the polluter reaps the
benefits derived from polluting while imposing the pollution costs on others. The polluter
therefore lacks an incentive to reduce those costs. However, in the case of technology, the
problem is reversed. A firm that invests in or implements a new technology typically creates
benefits for others while incurring all the costs. The firm therefore lacks the incentive to increase
those benefits by investing in technology. Pollution creates a negative externality, and so the
invisible hand allows too much of it. Technology creates positive externalities, and so the
invisible hand produces too little of it.3
The positive externality of innovation comes from the public-good nature of new
knowledge—innovating firms cannot keep other firms from also benefiting from their new
knowledge and therefore cannot capture for themselves all the benefits of the innovation. In
addition, the process of competition will typically drive a firm to sell a new device at a price that
captures only a portion of its full value, which means that consumers also reap some of the
benefits from new technology. While patents and other institutions are employed to protect
firms’ investments in innovation, such protection is inherently imperfect. A successful innovator
will capture some rewards, but those rewards will always be only a fraction—and sometimes a

3 There is, however, an offsetting negative externality because R&D is a fixed cost that must, in equilibrium, be
financed by the stream of quasi-rents it produces. The entry of another R&D competitor, or an increase in the
R&D investment level of a competitor, reduces the expected quasi-rents earned by other R&D firms. This “rent-
stealing” effect (Mankiw and Whinston 1986) could, as a theoretical matter, lead to over-investment in R&D. The
empirical evidence suggests, however, that positive externalities associated with knowledge spillovers dominate
the rent-stealing effect, leading to social rates of return to R&D substantially in excess of the private rates of return
(Griliches 1992).

6

---

very small fraction—of the overall benefits to society of the innovation. Hence innovation
creates positive externalities in the form of “knowledge spillovers” for other firms, and spillovers
of value or consumer surplus for the users of the new technology.
Adoption Externalities. The environmental and knowledge externalities discussed above
have long been at the center of economic debates about technology policy. More recently, we
have come to understand some additional market failures that may operate in the adoption and
diffusion of new technology. For a number of reasons, the cost or value of a new technology
to one user may depend on how many other users have adopted the technology. In general,
users will be better off the more other people use the same technology. This benefit associated
with the overall scale of technology adoption has sometimes been referred to as “dynamic
increasing returns.”
Dynamic increasing returns can be generated by learning-by-using, learning-by-doing, or
network externalities.4 While the image of the world beating a path to the door of the successful
innovator may seem compelling, the diffusion of a new technology is typically gradual. It takes
time for potential users to learn of the new technology, try it, adapt it to their circumstances, and
become convinced of its superiority. An important mechanism in this learning process is the
observation of the adoption of the new technology by others. Hence the adopter of a new
technology creates a positive externality for others in the form of the generation of information
about the existence, characteristics, and success of the new technology. This phenomenon is
often called “learning-by-using.”
The supply-side counterpart, “learning-by-doing,” describes how production costs tend to
fall as manufacturers gain production experience. If this learning spills over to benefit other

4 See Jaffe, Newell, and Stavins (2003, pp. 491-494) for a review of the literature on dynamic increasing returns.

7

---

manufacturers without compensation it can represent an additional adoption externality. Finally,
network externalities exist if a product becomes technologically more valuable to an individual
user as other users adopt a compatible product (as with telephone and computer networks,
for example). These phenomena can be critical to understanding the existing technological
system, forecasting how that system might evolve, and predicting the potential effect of some
policy or event.
Incomplete Information. Both innovation and diffusion of new technology are
characterized by additional market failures related to incomplete information. While all
investment is characterized by uncertainty, the uncertainty associated with the returns to
investment in innovation is often particularly large. Further, information about the prospects for
success of given technology research investments is asymmetric, in the sense that the developer
of the technology is in a better position to assess its potential than outsiders. A firm attempting to
raise investment capital to fund the development of new technology will therefore find such
investors skeptical about promised returns, and likely to demand a premium for investment that
carries such risks. This likely imperfection in the market for capital for funding technology
development exacerbates the “spillover” problem and therefore contributes to our expectation
that the invisible hand encourages too little research and development.
In the context of environmental problems such as climate change, the huge uncertainties
surrounding the future impacts of climate change, the magnitude of the policy response, and thus
the likely returns to R&D investment, would seem to exacerbate this problem further. In the
extreme, for example, it is difficult to see how the technological solutions that would be required
to address the possibility of catastrophic effects of climate change would be provided for by the
market even if environmental policies sent appropriate signals about expected costs. In this

8

---

Document Outline

• ??
• ??
  • ??
  • ??
• ??
  • ??
  • ??
  • ??
• ??

---