What is the theoretical consensus and empirical support for the effect of regulatory structures on innovation

Mentor
Dr. Sarani Bhattacharya Department of Humanities and Social Sciences Indian Institute of Technology, Kanpur

Submitted by
Mohit Saluja(Y8295) Abhishek Saini(Y8030)

INTRODUCTION
As time passed however, it became clear that technical progress can create new or solve and reduce existing environmental problems. Parallel to increasing industrialisation, the development of modem technologies and the ever-more intensive use of the environment, new environmental and resource problems were created, since the environment was the supplier of non-renewable and renewable natural resources, the direct recipient of pollutants from the production process and the indirect recipient of pollutants generated by product distribution, use and disposal. Therefore "some people regard technological developments as one of the greatest threats to nature and the environment. Others, by contrast, see them as a possible salvation". If the latter are right, then technological progress must result in companies being able to generate innovations which make it possible to:

come up with new possibilities for the substitution of natural resources, reduce specific impacts on the environment, improve resource productivity or improve the reversibility of environmental damage.

Issues related to innovation research have been examined in a host of studies during the past few years. The emphasis was placed, amongst other things, on investigating innovation objectives or measuring innovation success. Little attention has been paid to the impact of environmental policy instruments on innovation. This article explores whether the effects of environmental policy instruments on innovation analysed in a merely theoretical manner differs from results of empirical studies. The paper begins with some basic definitions and theories. It subsequently analyzes the dynamic effects of environmental policy instruments. The next section presents some empirical studies firstly concerned with the influence of specific environmental instruments on innovations and secondly dealing with industry-specific effects of environmental policy. The paper is finally ended with some relevant conclusions.

BASIC DEFINITIONS AND THEORIES

There has been an ongoing debate from the very beginning between the policy makers and the economists, which is mainly centred around what is the most effective and desirable mechanism for achieving environmental protection objectives.

ECONOMISTS:
Most economic arguments support the increased use of market based approaches, such as emission charges and tradable permits. They are believed to lead to cost effective allocations among firms, as the firms will always try to minimize their total cost under the burden of achieving given levels of environmental protection. And secondly the MBIs also provide continuous dynamic incentives for adoption of environmentally superior technology, since under such a scheme it would always be in the firms interest to clean up more if sufficiently inexpensive clean-up technologies can be found out (INNOVATION).

Major Direct market based instruments Emission charges: Charged by authority based on the quality or quantity of pollutants. Tradable permits: Distribution of discharge rights to firms in form of permits. Nature of pollutant is important. Deposit Refund system: Return of recyclable goods by consumers in exchange of money.

POLICY-MAKERS
Although policymakers have been giving more importance to the market based approach recently, this group seems to favour more the conventional command and control approaches, such and performance and technology standards. It is often referred to as Technology Forcing, (i.e.) mandating performance levels that are not

currently viewed as technologically feasible or mandating technologies that are not fully developed. PERFORMANCE STANDARDS: Requirements such as firms do not emit more than specified amounts of pollutants per unit of their economic activity. TECHNOLOGY STANDARDS: Requirements such as any particular industrial equipment or a process should be employed by the firms.

Main problem with the Policy-Makers approach It does not lead to cost effective allocation among firms. Practically, it is nearly impossible to know that how much improvement over the existing technologies is feasible.

So, the standards set must either be made unambitious, or else they run the risk of ultimately being unachievable, which in turn may lead to a great political and economic disruption.

HOW CAN WE ACHIEVE WIDESPREAD BENEFITS FROM A NEW TECHNOLOGY:

INVENTION: The development of a new technological idea INNOVATION: The incorporation of that new idea into a commercial product or process and the first marketplace implementation thereof. DIFFUSION: The typically gradual process by which improved product or processes become widely used.

The above rates are mainly endogenously determined within the economic system, affected by the opportunities that the economy creates for firms and individuals to profit from investing in research, in commercial development, and in marketing and product development.

Governments often seek to influence each of these directly, by investment in public research, subsidies to research and technological development, dissemination of information, and other means, but it is inevitable that direct policy inducements will

have only modest effects on the overall incentives to engage in the activities that produce technological progress. Hence it is crucial that policies with large economic impactssuch as many of those designed to protect or enhance environmental qualitybe designed to foster rather than inhibit technological invention, innovation, and diffusion.

THEROITICAL ARGUEMENT:
Theoretical economic analyses have generally shown supported the notion that marketbased approaches provide the most long-term invention, innovation, and diffusion. This is because they provide continuous incentives for emissions reductions, since any reductions in emissions generate revenues or reduce costsin the form of permits that can be sold, subsidies that can be obtained, or taxes that can be avoided. In contrast, once a performance standard has been satisfied, there may be little benefit to developing and/or adopting even cleaner technology. In addition, regulated firms may fear that if they do develop a cleaner technology, the performance standard will be tightened. Finally, technology-based standards appear to perform worst in stimulating innovation, since by their very nature they constrain the technological choices available, and may thereby remove all incentives to develop new technologies that are environmentally beneficial. Thus, the theoretical arguments are relatively clear; only empirical analysis has been missing. In preparation for such an empirical investigation, we now turn to a general model of technology adoption decisions in the context of alternative forms of environmental regulation.

REVIEW OF LITERATURE
Dynamic-incentive effects of environmental policy are attracting an increasing research interest. Innovation is often the main response to environmental policies, and policy instruments may matter for induced innovation effects, but shared conclusions about various theoretical and empirical issues are still lacking.

An extensive review of theories and evidence on technological change and the environment, including the role of policy instruments, is presented in Jaffe (2003), who distinguish between analyses of induced innovation and evolutionary approaches. Requate(2005) gives account of the present state of theoretical research on the dynamic incentives of different policy instruments, in particular economic instruments(EIs). Recent contributions address the determinants of environmental innovation, including policy instruments, on econometric grounds (for example, Brunnermeier and Cohen, 2003; Jaffe and Palmer, 1997; Mazzanti and Zoboli, 2005). A growing stream of evolutionary minded applied research projects addresses technological and

organisational innovations associated to policy experiences, and in particular the role of institutional settings, observed industrial strategies, and policy-design approaches in influencing innovation (see, among others, Hemmelskamp , 2000; Klemmer, 1999; Kemp, 1997; Rennings , 2003).Most of available contributions do not go beyond a black-box representation of dynamic-efficiency mechanisms stimulated by policies and their instruments. The works on induced innovation based on neoclassical production functions and optimising behaviour on R&D investments, have difficulties in dealing with systemic uncertainties and on-the-path adjustments typically characterising agents behaviour in real innovation processes.

EMPERICAL ANALYSIS :
In this section empirical studies on the effects of environmental instruments will be discussed. The aim is to look what innovation effects different environmental instruments have in practice. There are only a limited number of empirical studies on the effects of environmental regulations on innovative behaviour. In the process we will firstly draw on studies concerned with the influence of individual regulations (taxes [pigouvian fees], standards, subsidies), and secondly on studies dealing with industryspecific effects (packaging, chemical, textile and foundry industry).

1. Emission levies: The effluent levy in Germany.

At an industry level FABER/STEPHAN (1987) conducted a case study in Germany examining the adaptation processes in a large chemical company in the field of prevention of water pollution. The company managed to drastically reduce the amount of effluents by cutting the production of effluent-intensive goods and establishing a closed cycle, amongst other things, as well as by introducing new production methods. Considerable influence on environmental protection measures in the field of effluents has been exerted in Germany by the Water Pollution Control Levy Act (Abwasserabgabengesetz, AbwAG). Effluent levies have been charged in Germany since 1981 and have been modified in four amendments. The adoption of the AbwAG coincided with an amendment of the Water Resources Management Act

(Wasserhaushaltsgesetz, WHG) providing that minimum standards in accordance with the generally accepted state-of-the-art technology must be met when discharges are authorized (MEYER-RENSCHHAUSEN 1990). In an ex-ante empirical study prior to the effective date of the Water Pollution Control Levy Act, companies and municipalities were asked about their reactions (cf. EWRINGMANN 1980). While making the qualification that apart from instruments of anti-water pollution policy, other factors, too, influenced the adaptive behaviour of the sample, the study came to the conclusion that the mere announcement of the laws had triggered off responses. The effluent levy was a special case in this connection, since it was a new instrument and consequently it was difficult to obtain information on the available technologies. There was uncertainty concerning the cost burden small and medium-sized enterprises in particular had to expect. Nonetheless it became evident that the companies surveyed were improving effluent treatment in the run-up phase. Some companies changed their product range or externalized areas of production. The fact that on the whole the effluent levy has had a positive effect is also stressed by FABER (1989). However, FABER (1989) criticize the combination of the effluent levy and the juridical instrument of the Water Resources Management Act, since they feel that taking into account state-of-the-art technology waters down the actual economic effect of the effluent levy. This effect was further intensified by the recent amendments (GAWELIEWRINGMANN 1993). In her criticism JASS (1990) even attributes merely

enforcement-enhancing functions to the effluent levy and puts the actual effects on innovation down to requirements. She finds that in particular in the area of residual pollution, no reduction in emissions was accomplished, i.e. that the effluent levy failed to have any dynamic innovative effects (cf. also MEYER-RENSCHHAUSEN 1990).

2. An empirical analysis of an environmental law:

The German Toxic Substances Control Act The German Toxic Substances Control Act covers dissemination of new chemical compounds outside the company premises, protection of the workforce during production, use of hazardous substances and instructions. The Association of the Chemical Industry (VCI) and BASF PLC cite this act as an example of the existing excessive regulations, which to their mind prevent innovations in the chemical industry (VCI 1993; BASF 1994). A study on the impact of the Toxic Substances Control Act conducted by STAUDT (1993) came to the conclusion that with regard to the innovation activities the act has lead to R&D projects on new substances being discontinued, to them being shifted abroad and to an increased use of traditional substances, Staudt's study furthermore concluded that compliance, depending on company-specific factors, entails delays and cost increases for the companies, amongst other things due to belated market entry. Thus, the companies, with a view to reducing costs and time, chiefly respond by improving their internal organisational structures. The direction of the attempts at innovation is not solely determined by the Toxic Substances Control Act, rather it depends on the interrelationship of various parameters. However, one has to question the general validity of these findings. GLOEDE (1994), for example, criticizes the fact that the study is based on technical interviews and case studies which are not sufficiently representative. The delays in market entry discovered by Staudt are deemed insignificant by Gloede in view of the overall period needed for new developments. At the same time Gloede emphasizes that many substances on the list of traditional substances are more like new substances, because not much is known about their properties and the substances in question have hardly been marketed to date.

3. Subsidies: Lessons learnt from the Danish "Clean Technology Development Programme"
GEORGE (1992) analysed how environmental innovations take place when polluters, their suppliers or consultants are engaged in the development processes initiated through the Danish Clean Technology Development Programme. Therefore the five cases contained in the study focus on the impact of subsidies on the development of clean technologies. The Danish Clean Technology Development Programme (launched in 1986 with a three-year term) supported surveys on clean technologies and their potential use in different industries, the construction of a prototype of a computerbased information system covering clean technologies, as well as development and implementation projects. GEORGE concluded that most of the clean technology solutions resulting from the payment of subsidies were process-oriented. The programme is considered a success, for in most projects substantial improvements in terms of environmental protection were made without placing any financial burden on the companies. Some companies even managed to cut costs by substituting inputs. Some of the results were patented or they defined clean technologies as technologies which "seek to prevent pollution by input- substitution, process changes, encouraging recycling, lengthening product durability and developing cleaner consumer products" introduced in the market. The success is largely put down to- the fact that-firstly only projects aiming at solving specific environmental problems were supported and secondly the eligibility criteria took into consideration that often it is not just the polluter who is the innovator, but that solutions are found through cooperation between polluters, their customers, their suppliers and consultants.

4. The effects environmental regulations have on innovation in the foundry and textile industry
In a written survey of companies belonging to the German gray cast iron foundry industry the link between environmental protection and innovation was examined (THEIBEN 1987). The foundry industry is among the contracting industries in Germany

and produces in a relatively environmentally-intensive manner. To start with the study analysed the company features which are believed to influence innovative behaviour. They are: size of the company, location, number of R&D staff, share of university graduates, R&D expenditure, production technology, water requirements and effluents as well as environmental regulations. Among the relevant environmental regulations are environmental requirements concerning effluent temperature, noise emission and air pollution control in accordance with the Technical Instructions for Air. Large companies were not found to be harder hit by regulations. However, smaller enterprises tend to spend a larger proportion of their turnover on environmental protection. Companies examined primarily chose end-of-pipe-solutions as environmental protection measures, dust extraction plants for example. On the whole the production process in the gray cast iron foundry companies has become more productive, more energy efficient and more environmentally-friendly over the past years. The study is restricted to descriptive analyses of the above mentioned factors influencing innovative behaviour. Links between the factors are not established. The extent to which innovations are a consequence of the environmental regulation intensity in this industry is not a central theme. MAAS (1987) conducted a written survey along the same lines in the German textile industry. This industry is equally faced with requirements to reduce effluents, outgoing air or noise. The most important innovations in the textile industry aimed at automating production. The measures primarily served to rationalize, however, almost automatically they resulted in an improvement of the environment due to a more efficient use of resources. Environmental innovations were primarily implemented because of environmental regulations concerning effluents, but also concerning noise and air. The measures adopted included both integrated and end-of-pipe-measures. For example, a reduction in air pollution was achieved by limiting the use of fuel oil in favour of gas or by means of automatic apportioning of dyes which reduces water pollution caused by dyes.

General Model of Environmental Regulation and Technology Adoption

We begin with a generic pollution abatement technology choice problem, where a firm needs to choose both whether or not to adopt some environmental protection technology and the time of any such adoption. While the most general model would allow the policy instruments to affect the fundamental nature of the technological choices that firms make, we model the more tractable problem of the decision of whether and when to adopt some particular technology that the firm knows to be available. We assume the firm seeks to minimize the present discounted value of the sum of the following streams of costs: the costs prior to adoption of normal variable costs plus any payments of Pigouvian pollution taxes (where the vector of input quantities and the level of pollutant emissions are both functions of input prices and technology); the costs subsequent to technology adoption of variable costs plus Pigouvian tax payments; the costs of adoption (including any government subsidy); the implicit costs of violating any performance standard and/or technology standard prior to adoption; and the implicit costs of violating any binding performance standard subsequent to adoption.4 Thus, the firm chooses the time of technology adoption, T min PV (T)= 0T [P.X(P,I0)+Z.E (P,I0)].e-rt dt + T [P.X(P,I1)+Z.E(P,I1)].e-rt dt + CT.e-rt + 0T[1.Dtu..F[E(P,I0)+ 2.Dts] . e-rtdt + T [1. Dtu . F{E(P,I1)}] . e-rt dt where T -time (t) of adoption of pollution-control technology (T >= 0) P- vector of input prices. I -indicator variable for technology adoption, where no technology is represented by the value l0= 0 and the presence of technology is represented by I1= 1 ------------- (1)

X(.) -vector of input quantities as a function of prices and technology Z- environmental (Pigouvian emission) tax E(.) -optimal pollutant emissions, a function of input prices and technology

e- base of natural logarithms 1- implicit cost of violating a performance standard 2- implicit cost of violating a technology standard Dtu- dummy variable for existence of a (uniform) performance standard in year t Dts -dummy variable for existence of a technology standard in year t

F[.]- probability of a sanction for violating a performance standard, a function of level of pollutant emissions CT -cost of technology adoption, including any government subsidies.

We thus provide a means in Eq. (1) for comparing in a single economic context the impacts of two types of price instrumentstaxes and subsidiesand quantity instruments. We do this by positing that there is an implicit cost, 2, of violating an existing technology standard (where this is specified as a parameter to be estimated econometrically) and an implicit cost, 1, of violating a performance standard. In the case of this second implicit cost, we posit that the probability of a sanction being imposed is a function of the level of (excess) pollutant emissions (whether or not a technology standard is also in place and being obeyed or not). So, a necessary condition for optimal adoption at time T is [P. { X(P,I1)-X(P,I0) } + Z. {E(P,I1) - E(P,I0) } ]. e-rt +[ 1. Dtu. {F. E(P,I1) F.E(P,I0) } + 2.Dts-r.CT +dCT/dT ] . e-rt >=0 which yields P. { X(P,I1)-X(P,I0) } + Z. {E(P,I1) - E(P,I0) } + 1. Dtu. {F. E(P,I1) F.E(P,I0) } + 2.Dts >= [r.CTdCT/dT] This equation tells us that a firm will adopt the pollution-control technology at time T if operating cost savings plus savings from avoided emission taxes plus any avoided penalties for not adopting the technology or exceeding a performance standard are greater than adoption costs (including any subsidy) minus the time rate of change of (subsidized) adoption costs. The left-hand side of the equation essentially says that higher avoided costs

(due to technology adoption) can encourage adoption. The first term on the right-hand side of the equation indicates that higher adoption costs and higher interest rates discourage adoption (and that government subsidies can encourage adoption). Finally, the presence of the last termthe time derivative of adoption costindicates that adoption is discouraged by expectations of decreased (effective) costs of adoption in the future. Thus, even if the sum of current savings in operating costs, avoided emission taxes, and avoided regulatory penalties is greater than (the annual annuity of) adoption costs, it can pay to wait if those adoption costs are expected to fall over time at a sufficiently rapid rate.

CONCLUSION
The technological know-how a company has gained over time determines the future choice and application of processes and products. A fundamental change in the direction of innovation efforts entails at least a partial loss of acquired knowledge on the part of the company. That is why innovations as a rule build on experiences gained in the past, resulting in continuous and successive improvements of an existing product or a production process. Radical innovations" on the other hand, which make it possible to gain a lead in terms of know-how in new areas of technology, are rarely attempted.

By shaping the parameters external to companies, the government can exert a significant influence on the direction of innovation. One important influence on innovation processes are environmental policy measures. However, the impact of environmental measures on a company's innovative behaviour has hitherto received little attention in the research community and In the political debate. Most theoretical studies comparing innovative effects of individual environmental regulations come to the conclusion that direct requirements provide little incentives for dynamic effects and that emission taxes and permits are better instruments to promote innovations.

However, the empirical studies show that the dynamic effects of environmental policy instruments in practice partly differ from the ideal instruments analyzed in theoretical studies.

One reason could be that the real design of environmental instruments is influenced by the environmental policy process. The way environmental regulations are worded and introduced is usually determined by the interaction between the legislator, federations of business enterprises, trade unions and the public and their corresponding interests. This potential discrepancy between theoretical assumptions andpractical

implementation regarding the design of environmental regulations became evident in the case of the German effluent levy. Another reason is that the environmental policy instruments merely constitute one of many innovation-relevant determinants. Modem theoretical as well as empirical innovation research makes distinctions between numerous economic, social, legal or technological factors influencing the scope and the direction of R&D activities as well as the generation of innovations, their market launch and diffusion. Such factors are: company size and market structure, demand pull, technological opportunities, appropriable conditions (like patents), certain company characteristics and legal and administrative framework under which environmental policy can also be subsumed. The impact of an environmental policy instrument on innovations strongly depends on the influence of these factors. Consequently, studies on the innovative effects of environmental policy instruments should take these innovation-relevant factors into account and examine them closely.

References
Adam B. Jafee and Robert N. Stavins (1995). "Dynamic Incentives of environmental regulations : The effects of alternative policy instruments on technology diffusion" Journal of Environmental Economics and management 29, S-43-S-63 (1995). Aparna Sawhney. A Review of Market Based Instruments for Pollution Control : Implications for India No.2 ,1997. Karen Palmer and Adam B. Jaffe . Environmental Regulation and Innovation : A Panel Data Study, NBER working paper series, Working Paper 5545, April 1996. Jens Hemmelskamp. Environmental Policy Instruments and their Effects on Innovation Discussion Paper No. 96-22, June 1996. BROCKMANN, K.L./ HEMMELSKAMP, J./ HOHMEYER, O. (1996): Certified Tropical Timber and Consumer Behaviour - The Impact of a Certification Scheme for Tropical Timber from Sustainable Forest Management on German Demand. DUCHIN, F./ LANGE, G.-M. (1994): The Future of the Environment. Ecological Economics and Technological Change. New-York. WILLIAMS, H. E./ MEDHURST, J./ DREW, K. (1993): Corporate Strategies for a Sustainable Future, In: FISCHER, K./ SCHOT, J. : Environmental Strategies for Industry. Washington. 117-146.