Environmental business strategy under regulatory pressures : when tackling environmental issues increases business performance

(1)

Environmental business strategy under regulatory pressures:

When tackling environmental issues increases business

performance

Rick Halm

Student number: 10651098

July 15

th

, 2016

Bachelor thesis Business Administration

Supervisor: Pushpika Vishwanathan

(2)

Statement of Originality

This document is written by student, Rick Halm, who declares to take full responsibility for the contents of this document.

I declare that the text and the work presented in this document is original and that no sources other than those mentioned in the text and its references have been used in creating it.

The Faculty of Economics and Business is responsible solely for the supervision of completion of the work, not for the contents.

(3)

Abstract

A tradable permit system provides market incentives so that firms are motivated to tackle environmental issues; under which auctioning provides greater incentives than grandfathering. These statements rely mostly on theoretical argumentation and experimentation. This study empirically tests the extent of market incentives of the two instruments under a tradable permit system, namely; auctioning and grandfathering. Relying on the perspective of the Porter hypothesis, the study compares these market instruments within the power sector in the EU ETS, and tests whether Porter’s hypothesis holds under certain circumstances. The results show that grandfathering does not provide market incentives and auctioning has an initial negative effect on business performance. However, auctioning does provide market incentives and this initial negative effect can be offset by tackling environmental issues. This suggests that the Porter hypothesis holds and that environmental regulation enforced by auctioning can change a competitive environment which is only accessible for the truly sustainable firms.

(4)

1. Introduction

One of the global challenges the world faces today is the rising temperature, with carbon

emissions as its biggest cause. In June 2016 the WWF1, Sandbag, CAN2 Europe and HEAL3

published a report concerning 22.900 premature deaths caused by coal plants (Jones et al., 2016). And last year, the United Nations climate change conference (CoP21) discussed a worldwide approach to tackle these issues and asked businesses to take on its responsibility

to reduce its carbon emissions. On the 22nd of April 2016, more than 170 countries signed

the climate agreement aimed to cut carbon emissions (Falk, 2016). The question is, how do businesses respond to these regulatory pressures?

Profit maximizing firms are reluctant to change their logic towards a truly sustainable logic as it exposes the firm with risks that endanger the firms’ survivability; thus placing environmental awareness at the bottom of the strategic agenda (Sangle, 2010). Another view on the matter (controversial to classic economic theory) suggests that pollution shows inefficiency, and reducing pollution could enhance business performance. However, firms cannot anticipate all value maximizing opportunities. They rest on their current assumptions in which pollution reduction is seen as added costs and neglect tackling environmental issues. Market based regulatory pressures stimulates firms to view value maximizing from another angle and change current strategic logic to invest in environmental technologies (Porter & van der Linde, 1995). However, there is academically no consensus on the effect of different regulatory instruments on business strategy.

Figure 1 Different regulatory instruments (Keohane, Revesz, & Stavins, 1998)

1_{World Wildlife Fund}

2_{Climate Action Network Europe} 3_{Health and Environment Alliance}

Environmental regulation

Command and

Control Market Based

Pollution taxes Systems of tradable permits

Grandfathering

Auctioning ...

(6)

Within the environmental regulations, there are two types of regulatory mechanisms to tackle environmental issues through pollution abatement, see figure 1 for an overview. The first mechanism is command-and-control in which participating firms are forced to buy particular technology and have a performance standard with the maximum amount of pollution allowed. Within this regulation, there is no room for flexibility. The second regulatory mechanism is market-based. This mechanism aims to provide firms with a market incentive to reduce pollution. This regulation is mostly enforced with pollution taxes or systems of tradable permits. A system of tradable permits prices pollution

permits/allowances4 and allows firms to buy, sell and trade these permits. If a firm uses a

permit it is allowed to pollute a certain amount of emissions (Keohane, Revesz, & Stavins, 1998).

The system of tradable permits can again be divided in two instruments. First one is known as “grandfathering”, or free initial allocation based on existing levels of pollution. Thus allowing different sized firms to gain different amounts of permit allocation. Hereby, a polluting firm incurs costs only if they pollute more than the allocated permits freely received. The second instrument is the auctioning of permits. This instrument raises costs of firms for all of their emissions despite the size of the firm (Keohane, Revesz, & Stavins, 1998). However, there is no empirical evidence on how businesses respond to the two market incentive instruments; grandfathering and auctioning.

Theoretically, auctioning should provide more market incentives since firms incur costs for all of their emissions. However, grandfathering is typically preferred by the industry which makes it more politically feasible and is generally used. Therefore, data on auctioning is scarce and most research depends on theoretical argumentation or experiments without any systematic empirical evidence from the field (Delarue, Voorspools, and D’haeseleer 2008; Delarue, Ellerman, and D’haeseleer 2010; Grimm & Ilieva, 2013; Goeree et al., 2010). Whenever field data is available, it is hardly feasible to directly compare the success of the two different instruments since auctions cover only a part of the allocated permits and coexist with grandfathering (Grimm & Ilieva, 2013). This study seeks to fill this gap by empirically comparing the same firms that are affected by both grandfathering and auctioning in two distinguished periods.

The context of this paper is the European Union environmental regulation. The European Union set up a system of tradable permits namely, the European Union Emission Trading Scheme (EU ETS). Currently, the EU ETS is transitioning from a phase in which installations initially gained free permits/allowances (grandfathering) towards a phase in which all permits are fully auctioned. This process is progressively moving from zero percent

(7)

to a 70 percent auction of allowances over 2013 until 2020. However, the power sector is the only exception. This sector was able to push through the opportunity costs related to the allowances towards the end customer. This resulted in windfall profits for the power sector (Sijm, Neuhoff, & Chen, 2006). Therefore, from 2013 and onwards, the European Commission decided to auction 100 percent of the allowances in the power sector and prohibit any free allocation (grandfathering) to this sector. This transition from grandfathering to full auctioning in the power sector is an opportunity to compare the impact of two different regulatory market incentive instruments on business strategy and makes the EU ETS a perfect context for this study.

This study contributes to the environmental business strategy literature as it analyses and compares the impact of the two different market incentive instruments on business strategy. This paper starts with the basics of the social responsibility of a firm and the strategic logic to understand business assumptions towards regulatory pressures. This is followed by a discussion on strategy under environmental regulation which builds upon the popular Porter hypothesis and backs it up with additional literature. Continuing the discussion on the specific context of this study: the EU ETS and why this regulatory environment is a perfect context for this study. Finally, the results of the regression analyses are explained, ending with a discussion including the limitations of this study and suggestions for future research.

2. Existing literature

The following section gives an overview of the current academic discussion on environmental strategies. Starting with a discussion on social responsibility, a firms’ strategic logic, and when firms are motivated to tackle environmental issues. This also shows how firms initially respond to regulatory pressures. Finally, the Porter hypothesis is introduced, which is a contemporary view on how businesses respond to environmental regulations; which forms the perspective of this study.

2.1 The social responsibility of a firm

The following paragraph starts with a discussion the literature on the social responsibility of a firm and a firm’s preconceptions towards environmental issues. This section concludes that firms have a profit maximization strategy and environmental issues hardly register on the strategic agenda. Therefore, firms will only tackle environmental issues if it increases business performance. However, the current preconceptions assume that tackling environmental issues increases cost. Thus, the environmental issues can only be addressed through regulatory pressures.

(8)

Over the last decades, academics have discussed the social responsibility of a business. Earliest, Friedman (1970) discussed that the social responsibility of the firm is to maximize profit. He argues that the contract between the shareholders and a manager binds the manager to pursue the interests of the shareholder. By pursuing other means than increasing shareholder value is equal to taxing the shareholders. Also, whenever a business focuses on too many tasks, they become less efficient. And if a business pursues responsibilities beyond maximizing profit, it incurs costs, and will be wiped out in competition (Kolstad, 2007). Thus, a firm should only focus on maximizing profit.

On the other hand, the stakeholder management theory of Freeman states that firms should create value for all stakeholders because firms should understand the relationships between a business and the individuals who can affect or are affected. He argues that legally speaking, a legislation can give some rights to those groups that have a claim on the firm. Such as information on products ingredients and labor law protecting employees. Additionally, examples as externalities and moral hazards show that stakeholders can also influence the firm economically. Therefore, Freeman suggests that firms should pay attention to all stakeholders of the firm (Parmar et al. ,2010; Freeman, 2001)

The last and most contemporary view combines both views and states that creating value for stakeholders and shareholders go together in a systematic way. Creating value for stakeholders creates the most value for shareholders, and the other way around. Thus in this view, it doesn’t matter whether the firm pursues one or the other, managers will not have to face the dilemma to make a trade-off between social responsibility and profitability objectives since they have the same objectives (Kolstad, 2007). Current literature points out that in general, there is a positive relationship between corporate social performance and corporate financial performance (Margolis & Walsh, 2001). However, these two don’t always go together as for example customers are more sensitive to price than business ethics (Doane, 2005).

Narrowing the discussion towards corporate environment responsibility (CER) specifically; a sub-set of CSR, a firm will pursue in CER only if it increases profits (Kim & Statman, 2012). Academics attended to popularize ecological thinking that greenness equals profit (Garrod & Chadwick, 1996) and firms do proactively act on environmental issues when there is a positive relationship between environmental investment and business benefits (Sangle, 2010).

However, environmental management practices focus on the existing strategic paradigm that is originates in shareholder value maximization rather than environmentally oriented strategic focus (Garrod & Chadwick, 1996). Transforming towards an environmentally oriented strategy (Shevchenko, Lévesque and Pagell call this becoming truly sustainable) comes with many risks and puts the survival of the firm in danger. The

(9)

survivability depends on the ability to generate enough value when facing the uncertainty brought by the transition (York & Venkataraman, 2010). Today, oddly, an unsustainable firm, engaged in unethical behavior can still be deemed an exemplar in sustainability. Firms therefore keep their current strategic paradigm and solve stakeholder pressures by ‘compensation’ in which firms incrementally improve environmental/social performance without disrupting its logic used to create value. As long as the risks to become truly sustainable are higher than the costs of compensating, firms avoid becoming truly sustainable (Shevchenko, Lévesque, & Pagell, 2016). Kemp agrees and argues that firms in the capital-intensive fossil-fuel-based energy system have no interest in developing low carbon energy technologies. They will only move into the business of alternative energy technologies when fossil fuels are depleted or when the costs of extracting fossil fuels are becoming too high (1994). Consequently, in the current strategic paradigm, environmental issues hardly register on the strategic agenda of many firms whilst profitability does (Sangle, 2010).

However, Garrod and Chatwick find that firms are most motivated to tackle environmental issues whenever firms are under regulatory pressures (1996). More recently Berrone et al. also shows that greater environmental regulatory pressures positively influence firms’ propensity to engage in environmental innovation (2013). Even in a situation where “abundant evidence exists that there are highly profitable energy-saving opportunities, yet the technologies embodying these opportunities have not spread universally throughout the economy” (DeCanio, 1998). Therefore, incentives to adopt technologies that reduce pollution must come from environmental regulation (Popp, 2010). It seems that within environmental management practices, businesses are willing to go along with the Freeman’s stakeholder approach as long as Friedman’s shareholder approach agrees. Thus, firms’ environmental strategic focus is still to maximize profit and firms are motivated to tackle environmental issues under regulatory pressures.

2.2 Environmental strategies under regulatory pressures

The previous discussion shows how a profit maximizing firm responds to regulatory pressures. The traditional economic view based on neo-classical economic theory assumes that regulation in any way increases costs. For instance, regulations concerning technological standards restrict the amount of choices a firm has regarding technologies or inputs in the production process. Additionally, regulations concerning taxes and tradable permits “charge firms for emitting pollutants, a by-product of the production process that was previously free. These fees necessarily divert capital away from productive investments” (Ambec et al., 2013). As shown, in this static view, regulation limits the set of choices a firm has and is unable to pursue its most profitable choice. Therefore, costs have to rise. Due to

(10)

the rising costs, firms only comply with regulations if these pressures are compelling to do so.

In contrast with this view, Porter and van der Linde (1995) argue that; in reality, firms do not have access to all information and have limited time to analyze all choices. Additionally, the business environment is not static as the dynamics of competition puts pressure on firms to find innovative solutions. They propose a new perspective on business strategy under environmental regulation, called: The Porter hypothesis. This hypothesis states that properly designed regulation can trigger innovations that lower the total cost of a product or improve its value (Porter & van der Linde, 1995). They compose four arguments why this hypothesis should hold.

First, they argue that a regulation focused on information gathering (on a firm’s pollution) raises corporate awareness and compels firms to scrutinize their pollution. Then, whilst analyzing, the polluting activities reveal flaws in the production process and product design as it shows an incomplete use of resources. Due to the polluting activities, a firm must perform non-value-creating activities such as handling, storage and disposal of waste. Subsequently, reducing pollution is aligned with improving productivity and reduces the costs to perform these non-value-creating activities (Porter & van der Linde, 1995). Evidence from oil refineries suggests that in times with high environmental regulation the refineries did increase productivity due to the high regulatory costs. This shows that abatement costs may severely overstate the true cost of environmental regulation due to the unexpected productivity benefits (Berman & Bui, 2001).

The second argument states that regulation reduces the uncertainty that environmental investments will be valuable. Therefore, regulation stimulates investment in the environment (Porter & van der Linde, 1995). Farzin and Kort agree with this argument and show that certainty of size and timing of environmental taxes increases environmental investments because firms are certain of the costs and can make the optimal investment. On the other hand, uncertainty delays such investments (2000). However, Hoffmann, Trautmann and Hamprecht discover underlying motivations to invest despite uncertain regulatory pressures. For instance, to secure competitive resources, to leverage complementary resource and alleviate institutional pressures. This suggests that firms invest in times of certainty and could, but not necessarily, postpone investments in an uncertain regulatory environment (2009). Therefore, current literature suggests that Porter’s argument holds and that environmental regulations increase certainty that environmental investments will be valuable in the future.

Third, environmental regulation provides time to transit towards environmental innovation and strategies. This regulated transition time ensures that competitors will not avoid environmental investments and gain market share (Porter & van der Linde, 1995).

(11)

Ahuja, Lampert and Tandon (2008) suggest that environmental innovation is risky, requires great financial commitment and returns on investments over the long term. The move towards an energy system based on renewables and low carbon emission technology is hindered by small-scale production. Also, the fact that so far they have benefited insufficiently from dynamic learning effects which are so important for energy technologies (Kemp, 1994). Therefore, regulation can give firms a safe haven and a buffer until these new technologies are proven and learning effects reduces production costs (Porter & van der Linde, 1995).

Last, as explained above, regulation is necessary to increase environmental quality. Innovation cannot always –especially on the short term– offset the compliance cost and, therefore, firms hesitate to invest in environmental innovation. However, Wagner (2013) shows that innovations triggered by regulation can improve the environmental performance of the affected product/process and lead to innovation offsets which exceed the cost of compliance and enhance competitiveness. However, on the short term, evidence suggests that the impact of environmental regulation on business performance is generally negative due to the high costs of technology investments (Ambec, Cohen, Elgie & Lanoie, 2013). In the long run however, Lenoie et al. find support in a longitudinal study of ten years that there is a positive relationship of environmental regulation and business performance (2008). Therefore, well-designed environmental regulation can trigger innovation that lower the total costs (Porter & van der Linde, 1995). The arguments of the Porter hypothesis are summarized in figure 2.

Figure 2 Main arguments of the Porter hypothesis (Porter & van der Linde, 1995)

So far this paper has discussed that firms’ strategic focus is profit maximization and are motivated to tackle environmental issues if they are compelled by regulatory pressures. However, the Porter hypothesis states that environmental regulation can create a market

incentive for firms to reduce pollution through environmental innovations. In this way, the

Porter hypothesis explains that instead of complying and incurring costs (as traditionally assumed), market incentive regulatory pressures give profit maximizing firms the possibility to follow their existing strategic logic and change their business strategy to reduce pollution.

Regulations lead to innovation; reduce costs; reduce pollution because:

- Regulation shifts focus on pollution and pollution reveals flaws

- Regulation reduces environmental uncertainty of environmental investments

- Regulation gives firms time to transform to a truly sustainable firm

(12)

However, Porter and van der Linde prescribe different steps to be taken by regulators for the effects of the Porter hypothesis to work.

2.3 Requirements for an effective environmental regulation

The last section explained the effect of environmental regulation on business strategy according to Porter and van der Linde. However, the Porter hypothesis states that it requires properly designed regulation in order to have the intended effect on business strategy. Porter and van der Linde require that regulators take three steps for the hypothesis to work. See figure 3 for an overview.

Figure 3 Porter’s requirements for an environmental regulation (Porter & van der Linde, 1995)

The first step is that it should set clear goals with flexible approaches. A successful environmental regulation focuses on pollution outcomes and not on specific technology as this hinders the innovative process. Flexibility transfers responsibility from engineers, typically in charge of environmental issues, to the management team, who treat emissions

allowances as financial assets (Ambec et al., 2013). Lanoie et al. (2011) show that

regulations focused on specific pollution outcomes are leading to more innovation than technological standards. These outcomes give a firm an incentive to seek out the means to reduce their environmental impacts, thus induces innovations. Testa et al. also shows that a flexible approach has a beneficial effect on not only innovations but also on investments in environmental technologies (2014). Additionally, flexible regulations can encourage product and process innovations to avoid pollution early in the value chain. Placing regulatory pressures as late as possible in the value chain would allow flexibility for innovation through all stages of the chain (Porter & van der Linde, 1995). Concluding, Porter and van der Linde require that an environmental regulation should be focused on the ends and not on the means; thus provide clear goals but allow firms to find their approach to meet the goals.

Second, an environmental regulation should seed and spread environmental innovation. It should include the use of market incentives such as pollution taxes or tradable permits. Setting emission caps/outcomes still fails to provide incentives to innovate beyond

A successful regulation consists of: 1. Clear goals with flexible approach

2. Regulation should seed and spread innovation through market incentives 3. Stable regulation through Industry participation

(13)

the allowed emissions. However, market incentives can encourage the innovation to introduce technologies that exceed current standards (Porter & van der Linde, 1995). Johnstone, Hascic and Popp (2009) explain that market incentive instruments increase the benefits of technological innovation and adoption compared to other regulation. They find an increased patenting applications for renewable energy technologies in times of national and international regulatory pressures and an increase of diffusion of inventions (Johnstone, Hascic & Popp, 2009; Calel & Dechezleprêtre, 2014). Current literature suggests that market incentives can achieve environmental goals while providing incentives for more efficient technology diffusion as firms are motivated to adopt more efficient technologies. This shows that a market incentive instrument seeds and spreads environmental innovations (Kerr and Newell, 2003).

Lastly, coordination of the regulation should not hinder innovation. Therefore, in setting environmental standards, industry participation is required from the beginning. An industry accepts the regulation if it is clear and certain and can then begin innovating to address the issues (Porter & van der Linde, 1995). Literature on manager-orientation shows that managers are focused on future policy developments. They imply that industry participation is crucial to create fair and equitable policies. Without this, the stability of the climate policy is at risk. In such a policy environment, firms will not participate since their managers understand that the future of this policy might be completely different with different implications for the strategy of the firm (Sarasini & Jacob, 2014). Thus, industry participation is required to ensure a stable regulatory environment through fair and equitable policies. All three requirements are summarized in figure 3.

2.4 Environmental strategy under different market incentive instruments

The second requirement, discussed in the previous paragraph, states that an environmental regulation should provide market incentives. A market incentive instruments provides financial incentives to participating firms to comply with regulatory pressures. However, different market incentive instruments can have different effects on business strategy. Porter and van der Linde suggest market incentives such as pollution taxes or a system of tradable permits (see figure 1 for an overview of different environmental regulations). As explained in the introduction, the system of tradable permits can be divided into two market incentive instruments; grandfathering and auctioning of emission allowances. Grandfathering is generally favored by participating firms and allocates permits at no cost. The amount of allocation is based upon past emissions levels. It is generally justified that if you make firms pay for permits, the prices will be passed through to the end user. The alternative market based instrument is to use regularly scheduled auctions in which none of the permits are

(14)

freely allocated among participating firms (Goeree et al., 2010). This paragraph discusses current literature on these two different instruments under tradable permits and constructs the three hypotheses of this study.

2.4.1 Relation between emission intensity and firm performance under tradable permit system

From the perspective of a profit maximizing firm, environmental regulation equals an increase in cost. Environmental regulation enforces with a command and control mechanism could require that certain technology (such as a particle filter) is adopted by polluting firms. The investment in these technologies increases total costs without the market incentive to

reduce emissions. Early theory5_{suggests that the tradable permit system provides an}

incentive for firms to equate the costs of investing in environmental technology with the costs of permits (Hahn & Stavins, 1992; Montgomery, 1972). From a cost minimizing perspective,

whenever reducing emissions intensity6_{is cheaper than buying permits; firms will buy low}

emission technology that reduces emission intensity. Hence, less permits have to be bought and increases business performance. Therefore, theoretically argued, the first hypothesis is:

Hypothesis 1: In general, under a tradable permit system, emission intensity is negatively related to business performance.

2.4.2 Relation between market incentive instruments and business performance

Under the tradable permits, a regulator can choose to enforce the environmental regulation through grandfathering or auctioning. As explained above, it is generally justified by regulators to use grandfathering because if you make firms pay for permits under auctioning, the prices will be passed on to the end user. Thus, regulators usually enforce the tradable permit system through grandfathering (Goeree et al., 2010). However, theoretically argued, whether these allowances are grandfathered or auctioned, emission allowances represent opportunity costs either way. Considering that companies can either use the allowances to cover their emissions or sell the allowances on the market (or avoid purchasing under auctioning). Therefore, theoretically, a company is expected to increase the production cost (due to the opportunity costs) and raise the prices of their products even when they are grandfathered (Sijm, Neuhoff & Chen, 2006; Schleich, Rogge & Betz, 2008).

5_{More recent theoretical argumentation focuses towards auctioning or grandfathering. This makes}

the more recent theoretical argumentation irrelevant to hypothesize the relationship under a tradable permit system in general.

6_{Emission intensity is the emissions divided by operating turnover. Thus lowering intensity means a}

(15)

As the tradable permits system increases opportunity costs, firms are able pass on the opportunity cost of allowances towards the end user. This means that under grandfathering, firms gain windfall profits as their permits are allocated at no cost. Under grandfathering, companies in the power sector were able to pass-through 60 to 100 percent of the permit price towards the end-customer, thereby increasing the energy price (Sijm, Neuhoff & Chen, 2006). Therefore, the effect of grandfathering is that companies gain windfall profits. Consequently, income from citizens (the end user) is transferred towards these polluting firms (Bruyn et al., 2010). However, under auctioning, the costs of permits (which have to be purchased contrary to grandfathering) transfers windfall profits from polluting firms to the government (Goeree et al., 2010). Concluding, the tradable permit system results in higher product prices as firms are able to pass on the opportunity costs. As permits are freely allocated, grandfathering results in windfall profits. In contrast to grandfathering, auctioning does not have windfall profits due to the costs of permits. Therefore, the second hypothesis is:

Hypothesis 2: Auctioning is negatively related to profit margin compared to grandfathering.

2.4.3 Moderation of the relation between emission intensity and business performance

Firstly, the adoption of more emission efficient technologies reduces the total cost of allowances that should be bought by the firm under auctioning or increases revenue from selling unused allowances under grandfathering. Theoretically, the effect on business performance is identical whether allowances are grandfathered or auctioned off. However, when diffusion of new technology is taken into account, this effect changes. This diffusion of technology decreases the total demand of allowances; therefore, market price would decrease. If allowances are allocated for free, the allowances will generate less revenue for the firms that sell their allowances, as a result of their emission reduction. Whenever allowances are auctioned off, the firm benefits in the long run, as the firm does not have to buy the allowances and does not lose revenue by a reduced allowance price. Therefore, auctioning instruments are thought to provide greater incentives for innovation, as they provide rewards for continuous improvement in environmental quality (Milliman & Prince, 1989).

Secondly, Procedures regarding grandfathering (free allowances allocation) are a hindrance to the incentive of reducing carbon emission intensity. Simply put, future allowances are allocated on the basis of current emission levels. Therefore, businesses do not have the incentive to reduce current emissions in order to get granted the same allocation in the future, and keep windfall profits. Similarly, free allocation to existing

(16)

participants is relatively generous, while allocations to new participants are stricter. This creates incentives for plant lifetime extension rather than plant modernization (Hepburn et al., 2006).

If closing an old highly polluting plant would not alter its free allocation rights, the incentives for replacement are identical under grandfathering or auctioning. However, under grandfathering, closing a polluting installation leads to the termination of allowance allocation the next year for that installation. Therefore, grandfathering does not provide sufficient incentives for fundamental changes in corporate climate innovation activities at a level adequate for long-term targets (Rogge, Scheider and Hoffmann, 2010). Hence, the incentives to replace old installations/reduce emissions are stronger under an auctioning scheme then they are under grandfathering (Schleich, Rogge & Betz, 2008).

This means that under auctioning, the tradable permit system provides a market incentive and the Porter’s hypothesis should hold. Thus, becoming more efficient and thus reducing emission intensity has a positive effect on business performance. On the other hand, grandfathering does not create market incentives and Porter’s hypothesis should not hold. Meaning that using grandfathering as regulatory instrument, reducing emissions

intensity has no relation or even positive relation7 (due to the loss of grandfathered permits)

with business performance. Therefore, under auctioning, emission intensity is negatively related to business performance whilst under grandfathering no relation exists. Thus, under auctioning there is a greater market incentive to reduce emission intensity and therefore third hypothesis is:

Hypothesis 3: The relation between emission intensity and business performance is negatively moderated by auctioning such that whenever a firm is affected by the auctioning instrument, a stronger negative relationship between emission intensity and profitability is observed.

Concluding, current literature based on theoretical argumentation and experimentation shows that a tradable permit system provides an incentive to reduce emission intensity. When looking specifically at the two market instruments of the tradable permit system, the direct relation between auctioning and business performance is negative, contrary to grandfathering. However, grandfathering does not create a market incentive for firms to reduce their emission intensity due to the free allowance allocation. On the other hand, auctioning provides market incentives for firms such that a reduction of emission intensity

7_{Meaning that if a firm reduces the amount of emissions per euro turnover, thus pollutes less,}

(17)

has a positive effect on business performance (thus are negatively related). The three hypotheses are visually summarized in a conceptual model, seen in figure 4.

Figure 4 Conceptual model

3. Methodology

This section discusses the context of the study and methods used to answer the main question; whether grandfathering and auctioning provide market incentives for firms to reduce emission intensity. First, this section discusses the EU ETS as context of the study, explaining the history of the regulation, the specific sector of the study and whether the EU ETS in this sector succeeds in Porter’s requirements. This paragraph is finished with a discussion on why the EU ETS perfectly fits as context for this study. Finally, the section explains the research design and discusses the multiple datasets and data manipulations in order to increase the replicability of the study.

3.1 The context: European environmental regulation

The following section discusses the context of this research paper: EU ETS. The EU ETS was introduced in 2005 and consists of four phases, and is currently in its third phase. This section provides an overview of the three stages and explains the specific sector of this study.

3.1.1 History and present of the EU ETS

The European Union tackles climate change through the EU emissions trading system (EU ETS), which is on principle a tradable permit system. The EU sets an emissions cap on an installation/firm and the cap is reduced over time. Within this cap, a company can buy or receive (or trade) emission allowances; at the end of the year, a company must use the allowances to cover its years’ emissions. If a company reduces its emissions, it has to buy less allowances or sell the remaining (European commission, 2016a).

The first phase (2005-2007) was a pilot in which almost all allowances were given to businesses for free. The European Commission gave the allowances for free (grandfathering) to get businesses familiar with the regulation and participate in regulatory Emission Intensity Business Performance Instrument: Auctioning H3: - H2: - H1: -

(18)

updates. During this stage, the European Commission managed to set up a system of free trade in emission allowances and the infrastructure required to monitor the businesses. However, due to the absence of emission data before the year 2005, the free allowances exceeded the actual emissions. This created a demand and supply imbalance which resulted in a permit price of zero at the end of 2007 (European commission, 2016b).

With the emission data of the first phase, the EU lowered the cap on allowances in the second phase (2008-2012) and reduced the free allocation rate to 90 percent, auctioning the ten remaining percent. However, the economic crisis in 2008 led to emission reduction that was greater than expected which led to a large surplus of allowances, which affected the price of carbon throughout the phase (European Commission, 2016b). Still, academics point to attributable emission savings/reduction in phase II around the two and four percent per year (Lain et al. 2013).

The current phase, phase III (2013-2020) is characterized by the transition from free allocation to the auctioning of allowances. Most sectors will progressively move towards an auction, starting from 30 percent in 2013 and ending with for example 70 percent in the manufacturing industry. The power generation sector is the only sector which will auction 100 percent of the allowances 2013 on since this sector was able to push through 60 to 100 percent of the opportunity cost to the end user (European Commission, 2016a). Therefore, the power sector is the only sector within the EU ETS that can be used to compare the effect of grandfathering and auctioning on environmental business strategy.

3.1.2 The context: the power sector regulated by the EU ETS

The Porter hypothesis suggests that participating firms in an environmental regulation that enforces a market incentive, start innovating and that tackling environmental issues has a positive effect on business performance in the long term. However, allowances have been auctioned (the hypothesized market incentive instrument) in the power sector for three years. Lanoie et al. (2013) found a positive effect of innovation on business performance over ten years. This lag of innovation suggests that possibly there are no long term effects to be observed yet. However, the power sector has opportunities to reduce emission intensity in the short term.

In the power sector, due to the size of the energy system, even with the continued growth rates of renewable technologies, wind and solar power may only begin to replace conventional energy technologies well after 2020. Suggesting that a sector wide change can be observed in the long term (Jacobsson & Bergek, 2004). However, firm specifically, Popp suggest that profit maximizing firms will focus on technologies that generate short term profits under regulatory pressures, likely investments in wind energy, as this is the most cost-effective renewable option (2010). This suggests that the real impact of the

(19)

transformation of the whole energy sector will be seen after 2020. However, at firm level is seems that investments in clean-energy innovations within the power sector could quickly offset the compliance cost and increase business performance.

Additionally, energy generating firms have short term options to reduce their pollution namely; fuel switching. Power-generating plants can reduce emissions in the short term by switching fuels. An example is to replace a highly polluting fuel such as coal for a cleaner fuel such as natural gas or slowing down the generation of nuclear energy (Hintermann, Peterson & Rickels, 2016). Concluding, regulatory pressures have, in general, a negative effect on business performance in the short term. However, the energy generating firms are able to switch fuels in the short term and invest in technologies that drive short term profits. Therefore, within the power sector, the porter hypothesis should even hold in the short term. Still, the unobserved innovation offsets limit this study and is explained at the end of this paper.

3.2 EU ETS Scores on Porter’s requirements

This study tests the market incentive of two regulatory instruments based on the hypothesized effects of the Porter hypothesis. As market incentive is one of the three requirements that Porter and van der Linde (1995) prescribe. The context of this study; the power sector in the EU ETS should still succeed in remaining two requirements under both auctioning and grandfathering. Only then, the hypotheses can be tested. The scores of the power sector EU ETS on the prescribed requirements of Porter and van der Linde (1995) are represented in table 1.

Firstly, the EU ETS sets an emission cap/outcome goals without any requirements to invest in specific innovations thus providing flexibility. Thereby succeeding in the first requirement. The EU ETS started with a pilot that gave away free allowances (grandfathering) to participating firms in order to get the firms along from the beginning. The European Commission adopted critic from the first two phases in order to make the system more efficient and provide certainty in the third phase. Thereby succeeding in the third requirement; as the participating firms were included in the regulatory updates.

The difference between phase II and III is the reason why the EU ETS fits as the context for this study. According to the theory this suggest that in phase II characterized by grandfathering provides no market incentive, while phase III characterized by auctioning does. Previous studies could not directly compare the success of the two different instruments since auctions cover only a part of the allocated permits and coexist with grandfathering (Grimm & Ilieva, 2013). This context makes it possible to compare the effect of auctioning and grandfathering as the two phases make a distinction between the two applied instruments. Thereby, it is possible to compare the same firms in the same sector in

(20)

the same country and under the same regulation with one exception; auctioning or grandfathering.

Table 1 Regulatory requirements and the scores of the EU ETS

3.3 Research Design

3.3.1 Data collection and main variables

This paper combines different datasets to test three hypotheses. The first dataset includes the carbon emission data from power plants received from the EU ETS (CITL) database provided by the European Commission. This dataset shows the verified emissions (per

t/CO2) and allocated emissions of more than 13.000 installations across the European Union

from the years 2008 to 2015 (second and third phase). Important to note is that the firms that are included in the database are the firms that are still active and survived the past ten years. Since 2013, the power sector has to buy 100% of the allowances from auctions. However, eight countries got a temporary derogation from auctioning (European

commission, 2016c), therefore these countries are excluded from the dataset8_.

8_{These countries are Bulgaria, Cyprus, Czech Republic, Estonia, Hungary, Lithuania, Poland and}

Romania.

A successful regulation consists of:

1. Clear goals with flexible approach

2. Regulation should seed and spread innovation through market incentives

3. Industry participation from the beginning

Scores of the EU ETS on Porter’s requirements

1. Outcome focused without technological standards.

2. This requirement is tested in this study. Phase II characterized by

grandfathering whilst phase III characterized by auctioning. 3. Encouraged to participate from the

beginning and adopted critic from the first two phases.

(21)

In order to cross reference the financial data with the firms, the Bureau van Dijk

identification number (BvD number) of 93 percent of the firms9 were found. These numbers

were found using the installation identification number provided by the EU ETS database and using data manipulations similar to Jaraitė et al. (2013). Since many firms have multiple installations and financial performance is reported at the firm level, installations and its emissions from the same firm are combined. This results in a list of firms with their total emissions including all of their installations.

The EU ETS consists of all industries. Therefore, in order to detect the firms that generate electricity as core activity, the NACE Rev.2 codes (primary activities) of each firm are detected cross referencing the BvD numbers with the Orbis Database. NACE is the statistical classification of economic activities in the EU, regulated at the European Union level (Eurostat, 2008). The Orbis database consists of industry and financial data of listed and unlisted firms worldwide. Filtering the firms with electricity generation as core activities results in a list of 500 energy firms with in total 1147 power plants. Abrell, Faye and Zachman found 2101 firms and 3608 installations. However, they included all industries (2011). Also Berghmans and Alberola (2013) detected 1453 installations using the World Electric Power Plant database of Platts. However, they included the eight counties that got a temporary derogation from auctioning. Therefore, the sample of 1147 power plants and 500 energy firms appears to be in line with previous research.

To check whether the sample is representing the power sector, total freely allocated emissions are plotted in figure 5. It shows that there is a drastic reduction in 2013 in free allowances. The power sector is the only sector with such a reduction in freely allocated emissions (other industries received around 70 to 80 percent of the previous year), it verifies that this sample is representing the power sector. Additionally, the sample installations (7 percent of all installations) produce 32 percent of total emissions. This high proportion of emissions confirms that this is indeed the power sector as in 2006 the sector emitted 41 percent of the total emissions worldwide (International Energy Agency, 2008).

9_{Most of the missing BvD numbers were universities/hospitals/etc. which don’t have a BvD Number}

(22)

Figure 5 Freely allocated allowances of sample

Additionally, the sample of 500 firms is unevenly distributed between the EU as can be seen in figure 6. There seems to be a big difference between the amount of firms and the share in total emissions. For instance, Spain has many firms whilst less emissions, while Germany doesn’t have many firms and emits almost 45 percent of total emissions. It is important to note that these emissions are mostly emitted by a few countries from the total of 31 participating countries. Possible local regulations or subsidies in for example Germany might have a large impact on the analysis of the EU, and are controlled for.

Figure 6 Sample distribution

Financial data of the 500 energy firms are also obtained using the Orbis database. Around 80 percent of the financial data between 2008 and 2014 was available, between 350 and 450 firms are representing this year span. Unfortunately, only 16 percent of the financial data over 2015 was available, 83 firms are representing this year. Worth noting is that most

0 100 200 300 400 500 600 2008 2009 2010 2011 2012 2013 2014 2015 Mt C O 2 Year

Freely Allocated Allowances

0% 10% 20% 30% 40% 50% AT BE DE DK ES FI FR GBHR IT LU LV NL NO PT SE SI SK Sample Distribution

(23)

of these firms are subsidiaries of large conglomerates. Due to possible price arrangements, financial data could cloud the actual effect.

If this study used absolute emissions, a decrease in emissions could also mean that firms reduced production. However, the study tests whether there is a marketing incentive for firms to invest in low emission technologies. Therefore, a firm’s verified emissions are divided by the operating turnover (revenue generated only from its core operations) to calculate the emission intensity of a firm. Emission intensity measures the amount of emissions for each generated euro of operating turnover. Therefore, a decreasing emission intensity indicates that a firm can produce the same operating turnover with less emissions.

Additionally, to test business performance, profit margin is used as an indicator. This ensures that every firm is taken into account equally as the sample consists of many different sized firms with a high spread in turnover. Also, the emission intensity variable is

skewed as can be seen in appendix A, figure 1, the natural logarithm of the variable is

calculated, which results in an emission intensity that is more closely related to the normal distribution as can be seen in appendix A, figure 3.

There are however a few mistakes in the dataset. For instance, there are firms with minus two emissions, whilst the code of non-reported is minus one. Most of those mistakes are manually excluded from the sample. Another example of a mistake is a firm with two

euro operating revenue and 250.000 t/CO2 emissions, results in an emission intensity of

125.000, while the mean is 13.94. In a regression, these outliers have a strong effect (squared deviation from the mean) on the results. To exclude these outliers, z-scores are calculated, and 5 percent of the outer values (two tailed) are excluded from the dataset for both profit margin and the natural logarithm of energy intensity.

A dummy is created for the two phases where grandfathering (phase II) gains value zero and auctioning (phase III) the value one. Each firm has five grandfathering observations and three auctioning observations. To increase the reliability, the analysis analyzes grandfathering (phase II) against auctioning (phase III) without making a distinction between the years within the phases. This results in 2216 valid observations.

3.3.2 Control variables

Different variables should be included that are also positively or negatively related to profit margins and could also explain the relations that are tested. Firstly, the analysis is controlled for size. Large firms are able to gain economies of scale which could increase profit margins. Additionally, large companies have capital available to invest in new innovations which could increase profit margin, whilst small companies might have struggles to attract capital or large amounts of reserves. Controlling for size is in line with Kerr and Newell’s study (2003). Size

(24)

(𝑆𝑖𝑧𝑒_%) is measured using the natural logarithm of total assets, acquired from the Orbis database.

Secondly, economic growth of the different countries could have an effect on the profitability of the firms. A growing economy could mean an increase in demand for energy and increase profit margin, but also mean an increase in competition and lower profit margins. Controlling for economic growth is in line with the study of Abrell et al. (2011).

Economic growth data per country and year (𝐺𝑟𝑜𝑤𝑡ℎ%) is measured with GDP growth rates

(decimal rate) and is obtained from Eurostat; the statistical bureau of the European

commission10_.

The third control variable is yearly average energy price per country. Logically, a higher energy price increases the profit margins per generated unit. This could be a result of

increasing energy demand or reduction in supply. Energy prices (𝐸𝑛𝑒𝑟𝑔𝑦%) are measured via

the yearly average energy prices and acquired from Eurostat11_.

Lastly, as explained above, the model is controlled for country. There are many reasons why different countries have different profit margins. For instance, the extend of a monopolistic power sector might differ per country or different local subsidies have an effect on profit margins in that country. With their research, Abrell et al. found that country specifics had impact on emission reduction, this explains that different country characteristics can have different effects on profit margins (2011). For each country a dummy variable is

created, in the regression equation the set of all dummies (𝐶𝑜𝑢𝑛𝑡𝑟𝑦%) is included. The

complete conceptual model including the control variables are visually represented in figure 7.

Figure 7 Complete conceptual model

10_{Dataset can be acquired from: http://ec.europa.eu/eurostat/web/products-datasets/-/tec00115} 11_{http://ec.europa.eu/eurostat/web/energy/data/database} Emission Intensity Business Performance Instrument: Auctioning Economic Growth Size _{Energy prices} H3: - H2: - H1: -

(25)

3.3.3 Regression model

To test these hypothesis, two regression models are analyzed. The first model tests the

direct effect of emissions intensity (𝐿𝑛𝐼𝑛𝑡𝑒𝑛𝑠𝑖𝑡𝑦%) and the dummy variable (𝐴𝑢𝑐𝑡𝑖𝑜𝑛%)

auctioning/grandfathering on the profit margin measured in percentages (𝜋%). The equation

is as follows:

𝜋% = 𝛼 + 𝛽>𝐿𝑛𝐼𝑛𝑡𝑒𝑛𝑠𝑖𝑡𝑦12_%+ 𝛽A𝐴𝑢𝑐𝑡𝑖𝑜𝑛%+ 𝛽B𝑆𝑖𝑧𝑒%+ 𝛽D𝐺𝑟𝑜𝑤𝑡ℎ%+ 𝛽E𝐸𝑛𝑒𝑟𝑔𝑦%+ 𝛽F𝐶𝑜𝑢𝑛𝑡𝑟𝑦%

+ 𝜀% 1

The second model tests the moderation effect 𝐿𝑛𝐼𝑛𝑡𝑒𝑛𝑠𝑖𝑡𝑦_%∗ 𝐴𝑢𝑐𝑡𝑖𝑜𝑛_% of the

dummy variable auctioning/grandfathering on the relationship between the emission intensity and the profit margin. Then, the equation is as follows:

𝜋% = 𝛼 + 𝛽>𝐿𝑛𝐼𝑛𝑡𝑒𝑛𝑠𝑖𝑡𝑦%+ 𝛽A𝐴𝑢𝑐𝑡𝑖𝑜𝑛%+ 𝛽B𝑆𝑖𝑧𝑒%+ 𝛽D𝐺𝑟𝑜𝑤𝑡ℎ%+ 𝛽E𝐸𝑛𝑒𝑟𝑔𝑦%+ 𝛽F𝐶𝑜𝑢𝑛𝑡𝑟𝑦%

+ 𝛽I 𝐿𝑛𝐼𝑛𝑡𝑒𝑛𝑠𝑖𝑡𝑦%∗ 𝐴𝑢𝑐𝑡𝑖𝑜𝑛% + 𝜀% (2)

4. Results

4.1 Descriptive statistics

The descriptive statistics of the two variables can be seen in table 2. By excluding 5 percent of outer values, both variables come closer to a normal distribution as can be seen in Appendix A table 1. The variable profit margin has a skewness of -0.269 and -0.802 for emission intensity and a kurtosis of 1.431 and 0.569 respectively; suggesting that both variables are closely related to the normal distribution. Including both phases, a total of 2216 observations of the natural logarithm of emission intensity with a mean of 0.8425 and a standard deviation of 1.4968 and profit margin with an average of 6.11 percent and a standard deviation of 12.75 percent. When dividing up the sample in grandfathering (phase II, 2008-2012) with a 1570 valid observations and auctioning (phase III, 2013-2015) with 646 valid observations, the results show that the means of the emission intensity and the profit margin during auctioning is lower compared to grandfathering. A reduction of profit margin during auctioning is expected as firms has to buy their emission permits. Whilst with grandfathering, they received these allowances for free. Thus auctioning raising costs and dropping profit margins with auctioning. Secondly, the results show a reduction in emission intensity during auctioning, meaning less emissions per euro operating turnover. This

12_{𝐿𝑛𝐼𝑛𝑡𝑒𝑛𝑠𝑖𝑡𝑦 = ln(} LMN%O%MP MQ%RR%STR

(26)

suggests that firms are reducing their emission intensity whenever they are affected by auctioning compared with grandfathering.

Moreover, table 2 also shows the correlations between the variables (dummy’s excluded). Firstly, there is generally no significant correlation between emission intensity and profit margin contrary to the first hypothesis. Secondly, economic growth is negatively correlated to profit margin and positively related to size; possibly explaining that in times of economic growth, firms are growing in size but competing for the growing demand thus squeezing their profit margins. Thirdly, economic growth and energy prices are also negatively correlated. This is in line with theoretically argumentation of van Zon & Yetkiner that explain that raising energy prices lower profitability of produced goods, thus profitability of research, and therefore a negative impact on economic growth (2003). Lastly, size is positively correlated with profit margin and negatively related to emission intensity. This could be explained that larger firms, with higher assets, have the ability to reduce emission intensity which results in a higher profit margin/business performance as expected.

4.2 Regression results

The full results of the two regression models can be seen in table 1 of Appendix B. To assure a clear overview, the results, excluding the country dummies, can be seen in table 3. The first and second model show that under a tradable permit system there is no significant relation between emission efficiency and business performance. With this unexpected result the first hypothesis remains unconfirmed. However, a tradable permit system has to be enforced by one of the two instruments; auctioning or grandfathering. By running a regression on the sample that only includes grandfathering observations, no significant relation is observed between emission intensity and profit margin (see table 2 in appendix B). Thus, the results show that a tradable permit system and more specifically grandfathering do not provide incentives to reduce emission intensity. This relation is elaborated upon in the discussion.

Additionally, the results of the regressions show a significant negative relationship between auctioning and profit margin compared to grandfathering (p < 0.01). The model

Table 2. Means, standard deviations and correlations Variable Mean Mean (N = 1570) Grandfathering Mean (N = 646) Auctioning SD 1 2 3 4 1. Profit Margin 6.117 7.085326752 3.764664087 12.750 2. Emission Intensity 0.843 0.9252 0.6415 1.497 0.01 3. Economic Growth -0.002 -0.0046 0.0044 0.026 -.050* -0.026 4. Size 11.330 11.27 11.45 2.038 .049* -.071** .089** 5. Energy Prices 0.116 0.1146 0.1194 0.020 0.006 -0.011 -.052* .087** Note. N = 2216 * p < 0.05 ** p < 0.01

(27)

shows that whenever a firm is affected by an auctioning instrument, profit margin is two percent lower compared to grandfathering. As explained above, under auctioning firms in the power sector have to pay for 100 percent of their allowances, whilst under grandfathering, they got these allowances for free but could push through the opportunity cost to the end user. This increases costs and lowers the profit margin. This result was expected and confirms the second hypothesis.

The second model included the interaction effect between emission intensity and auctioning. The regression results show that the relation between emission intensity and profit margin are negatively moderated during auctioning (p < 0.05). Meaning that whenever a firm is affected by the auctioning instrument, a stronger (significant) negative relationship between emission intensity and profitability is observed, confirming the third hypothesis. This shows that under auctioning, low emission intensity is related to high profit margins.

The negative implications of auctioning on profit margin presented with the second hypothesis can be offset by this relation. The regression shows that reducing emission intensity with one percent has a positive effect of 0.88 percent on profit margin. The negative effect of auctioning on profit margin of two percent can be offset by a three percent decrease in emission intensity which increases profit margins by 2.64 percent. Thus, environmental regulation enforced with auctioning of tradable permits provides a market incentive to reduce emission intensity even with the negative effect of auctioning on business performance.

Additionally, most of the control variables are related to profit margin as expected. Economic growth is not significantly related to profit margin (p = 0.053). Additionally, the size of the company is positively related (p < 0.01) to profit margins. It could be explained that large firms are able to gain economies of scale and have the ability invest in low emission technologies. Lastly, energy prices (with mean of 0.116) are negatively related (p < 0.05) with profit margins. Showing that an increase of energy prices with one eurocent has a 0.67 percent negative relation with profit margin. A possible explanation is that an increase in energy prices is related to an increase in fuel (input) prices to generate power. However, they are not able to push all of the increasing fuel costs to the end user due to competition. Thus an increase in energy prices actually means an even higher increase in production costs and a reduction of profit margin. The standardized coefficients provide the ability to compare the variables and shows that the energy price is mostly related to profit margins, followed by size, the market incentive instrument; auctioning and the interaction effect. The explanations on the effects of control variables are not within the scope of this paper; and will not be further discussed. The whole model including the control variables are visually represented in figure 8.

(28)

Although the main hypothesis is confirmed, the R-squared of the first model is 4.1 percent and respectively 4.3 percent for the second model, indicating that this model explains 4.1 and respectively 4.3 percent of the variance in the variables. A higher percentage would indicate that this model explains more variance in business performance. However, the underlying reasons for business performance is a whole research field and includes many (unknown) variables. This paper is not trying to predict business performance, but analyzes the effect of emission reduction on business performance. The low R-squared does not necessarily indicate a low quality regression.

Figure 8 Conceptual model with regression coefficients (* p<0.05, ** p<0.01)

5. Discussion and Limitations

This study takes a closer look at market incentives under the tradable permit system and its two regulatory instruments; auctioning and grandfathering. As firms are assumed to be profit

Table 3. The impact of market incentive instruments on profit margin

Model 1 Model 2

Dependent variable: Profit Margin

Unstd Coeff Std. Coeff Unstd Coeff Std. Coeff Sig.

B Std. Error Beta Sig. B Std. Error Beta

(Constant) 21.755 8.006 0.007 21.532 7.998 0.007 Dummy Auctioning or grandfathering -2.523 0.646 -0.09 0.000 -1.944 0.691 -0.069 0.005 Emission Intensity -0.181 0.193 -0.021 0.347 0.14 0.237 0.016 0.554 Economic Growth -21.87 11.091 -0.044 0.049 -21.485 11.081 -0.044 0.053 Size 0.528 0.142 0.084 0.000 0.515 0.142 0.082 0.000 Energy Prices -76.281 31.541 -0.121 0.016 -67.521 31.735 -0.107 0.033

Emission Intensity x Auctioning -0.889 0.383 -0.064 0.020

R-squared 0.041 0.043 Emission Intensity Business Performance Instrument: Auctioning -0.064* -0.069** ns ns 0.082** -0.107* Economic Growth Size Energy prices

(29)

maximizing, a market incentive exists if there is a significant relationship between emission intensity and profit. The following discussion justifies the relationships found in this study and explains the contribution in the current literature. At the end of the discussion, limitations of the study and future recommendations are elaborated upon.

5.1 Elaboration on the study and its contributions

The first hypothesis remains unconfirmed as reducing emission intensity is unrelated to business performance under a tradable permit system. Although the early theory on the tradable permits system excluded auctioning or grandfathering, a tradable permit has to be enforced by either grandfathering or auctioning. The results show that the two regulatory instruments have different effects on profit margins. On the one hand, the regression on grandfathering sample finds no significant relation between emission intensity and business performance. On the other hand, the regression shows that under auctioning, this relationship exists. The insignificant relation of the tradable permit system is caused by the insignificant effect of grandfathering that reduces the significant effect of auctioning on business performance. Therefore, the relation between emission intensity and profit margin is non existing within the tradable permit system in general. This shows that the tradable permit system is not one system with two enforcing instruments, but two separate systems with different implications namely; auctioning and grandfathering of tradable permits.

Continuing, the regression on the grandfathering sample suggests that there is no

relation between emission intensity and profit margin if permits are allocated without cost. This suggests that firms with low emission intensities do not have significantly different profit margins than firms with high emission intensities. Theory suggest that a profit maximizing firm only voluntarily tackles environmental issues if it increases business performance. Therefore, a tradable permit system enforced by grandfathering does not encourage a firm to reduce its pollution since it does not increase business performance. This confirms the arguments of Rogge, Scheider and Hoffmann (2010) that suggest that grandfathering does not provide sufficient market incentives for fundamental changes in corporate climate innovation activities.

However, this result can also be interpreted that grandfathering of tradable permits can achieve pollution reduction without affecting business performance. This interpretation opens a discussion on ethical business practices. Environmental regulation enforced by grandfathering does not change the strategic logic as it has no effect on business strategy. However, it could renew the status quo in which firm’s no longer see emission reduction as costs; but as its social responsibility. It is possible that firms are reducing the pollution from their production processes out of ethical considerations. However, this subject is out of the

Environmental business strategy under regulatory pressures : when tackling environmental issues increases business performance