Description
Lobbying (also lobby) is the act of attempting to influence decisions made by officials in the government, most often legislators or members of regulatory agencies.
Interest Group Lobbying and Corporate Strategy
Thomas P . Lyon and John W. Maxwell February 15, 2002
Abstract
We study corporate non-market strategies designed to influence the lobbying behavior of other special interest groups. We focus on conditions under which costly lobbying is an informative signal to policymakers about the true state of the world, and in which stringent policy is so costly to the firm that the firm is not viewed as a credible source of information. We study three corporate non-market strategies: 1) the ‘‘bearhug,’’ in which the firm subsidizes the lobbying activities of an interest group before the true state of the world is known, 2) ‘‘astroturfing,’’ in which the firm subsidizes the lobbying activities of a group with similar views after the state of the world is known, and 3) self-regulation, in which the firm voluntarily limits the potential social harm from its activities. All three of these strategies can be used to reduce the informativeness of lobbying. We identify conditions under which each strategy is profitable for the firm. All three strategies, however, reduce the welfare of the public decisionmaker.
1.
Introduction
The role of interest groups in politics has held a long-standing fascination for political economists. In the 1780s, James Madison famously warned of the power of ‘‘factions’’ in the Federalist, while nearly two hundred years later Mancur Olson and George Stigler elevated the study of interest group politics to an important field within economics.1 Early work theoretical work treated interest group ‘‘pressure’’ as a production function, smooth and twice continuously differentiable, but more recently theorists have been opening up the black box of political pressure to focus more specifically on campaign contributions or lobbying.2 Several recent contributions shed new light on the role of lobbying in conveying information to public decisionmakers.3 Even this recent work, however, typically does not distinguish firms from other special interest groups. We argue that in many lobbying situations, firms do indeed have preferences distinct from those of other groups. In particular, they often bear the costs of government policy but do not collect the benefits. This is especially true for policies dealing with externalities or the provision of public goods. In such circumstances, firms cannot credibly convey information because their powerful bias towards weak policies is common knowledge among decisionmakers. Nevertheless, firms
1 Olson (1965) elaborates a rational choice model of interest group action, while Stigler (1971) applies this approach to the study of regulation specifically. 2 Grossman and Helpman (2001) provide an excellent introduction to this literature. 3 See, for example, Lohmann (1993) and Krishna and Morgan (2001).
1
may be able to influence the lobbying behavior of other interest groups. The corporate strategies that accomplish this goal are the subject of our paper. One important example where firms may wish to influence lobbying behavior is in the siting of industrial facilities. Manufacturing plants, waste disposal facilities and electric generating units are likely to be met with a storm of local protesters organizing under the banner ‘‘Not In My Backyard’’ or NIMBY , for short. While protests may often present much noise with little signal, they also have the potential to provide decisionmakers with vital first-hand knowledge about how local communities will be affected by particular siting decisions. Corporate strategy has of necessity expanded to include techniques for responding to such protests. In this paper, we develop a formal interest-group model of corporate strategy in the lobbying process. We focus on three strategies for interacting with interest groups: the ‘‘bear hug’’ (which involves the corporate embrace, through subsidies, of other interest groups), ‘‘astroturf’’ (which involves targeted corporate subsides to groups supporting the company’s position), and self-regulation (which involves voluntary corporate actions that reduce the risk of social harm from the company’s operations). We provide conditions under which these strategies can be part of a Bayesian Nash equilibrium in the interestgroup game, and we also assess their effects on the welfare of the public decisionmaker. We coin the term ‘‘bear hug’’ to refer to a corporate strategy of embracing one’s opposition. This may involve providing financial support to a special interest group normally considered to have opposing views. For example, DeSimone and Popoff point out that ‘‘It is also important to recognize that there can be a disparity of resources and information between business stakeholder groups that makes trust difficult to develop. This may sometimes require action to redress the balance. Since the Brent Spar incident—when opposition prevented Shell from disposing of a large oil storage platform at sea—the company has made space available for environmental groups to explain their point of view in educational and other materials that it has prepared.’’4 The term ‘‘astroturf lobbying’’ was coined by Lloyd Bentsen, long-time Senator from Texas, to describe the artificial grass-roots campaigns that are created by public relations firms.5 This strategy differs from the ‘‘bear hug’’ in that it involves financial support of groups that support the company’s view. The creation of such groups is the business of public relations firms such as Davies Communications, whose advertising says ‘‘Traditional lobbying is no longer enough. Today numbers count. To win in the hearing room, you must reach out to create grassroots support. To outnumber your opponents, call thee leading grassroots public affairs communications specialists.’’6 Davies explains how his firm generates a ‘‘grassroots’’ letter-writing campaign techniques through the use of telephone banks: ‘‘We get them on the phone, and while we’re on the phone we say ‘Will you write a letter?’ ‘Sure.’ ‘Do you have time to write it?’ ‘Not really.’ ‘Could we write the
5 6 4 DeSimone and Popoff (2000), p. 165. Stauber and Rampton (1995), p. 79. Stauber and Rampton (1995), p. 90.
2
letter for you? I could put you on the phone right now with someone who could help you write a letter. Just hold, we have a writer standing by.’..If they’re close by we hand-deliver it. We hand-write it out on ‘little kitty cat stationery’ if it’s a little old lady. If it’s a business we take it over to be photocopied on someone’s letterhead. [We] use different stamps, different envelopes..Getting a pile of personalized letters that have a different look to them is what you want to strive for.’’7 Self-regulation is quite different from the other two strategies, in that it involves real changes in company operations that are designed to reduce the risks of social harm. If these actions are substantive enough, interest groups may decide that the further gains from lobbying are not enough to justify the costs, and they may eschew participation in the political process. Perhaps surprisingly, however, we show that self-regulation may also induce interest groups to participate more actively in the lobbying process, although the extra activity may not further the group’s goals in the long run. The remainder of the paper is organized as follows. Section 2 presents a simple model of the lobbying process. Section 3 studies the ‘‘bear hug’’ while section 4 considers ‘‘astroturf.’’ Section 5 addresses the effects of self regulation, and section 6 discusses extensions of our model to a setting with multiple interest groups. Section 6 concludes.
2.
A Simple Model of Lobbying
Our simple model of lobbying consists of two players, a government decision maker (DM), and a special interest group (SIG). In this version of the model the firm has no active role in the lobbying process . We assume that the firm proposes a new project that requires the approval of the decision maker.8 For example, the firm might propose a new manufacturing facility that requires a site license before it can be built, or the firm might propose to introduce a new pharmaceutical product that requires approval from the Food and Drug Administration before it can be produced. The proposed project may have social effects through a variety of mechanisms, e.g. it may create jobs in the local community, it may affect the environmental quality of the surrounding community, or it may affect the health and safety of that community. These effects can be characterized by a variable ? ? 0 and l = 0, i.e., the (positive biased) SIG knows the true state of the world and can costlessly communicate it to the DM. We examine the SIG’s incentives to report the state of the world when the DM believes the SIG’s announcement. Since the SIG always prefers a higher level of policy than the DM, it has no incentive to misreport when the state is ? = ?H . Misreporting may be desirable, however, if ? = ?L . In this case, the SIG misreports, i.e. reports ?H , if Thus, when ? = ?L , the SIG misreports if ?(? L ? ?L ? ? )2 < ?(? H ? ?L ? ? )2 ? > (?H ? ?L )/2. (1)
Consider a case where condition (1) holds. This implies that the SIG has a large degree of bias„ or alternatively, that the high and low states are relatively close together. Then the SIG will always report that ? = ? H , regardless of the actual state of the world. Assuming the DM knows ? , ? L , and ? H , he will recognize the SIG’s incentives, and hence will not update his prior based on the SIG’s report. Thus, the DM sets p = (? L + ?H )/2. Turn next to a case where lobbying is costly. Because the SIG is biased toward high levels of the policy, it is particularly concerned about the possibility that the DM sets p = ? L when the state is actually ?H . Thus, the SIG is strongly motivated to incur the cost of lobbying when the state is ?H , but may not find it worthwhile when ? = ? L . Under certain conditions, there exists an equilibrium in which the SIG only lobbies when ? = ? H , and the DM interprets the SIG’s lobbying as a statement that indeed ? = ? H . For this equilibrium 4
to exist, the SIG must prefer to refrain from lobbying when ? = ?L , i.e.,?(?L ? ?L ? ? )2 ? ?(? H ? ?L ? ? )2 ? l, or (2) l ? l ? (?H ? ? L )(2? ? (? H ? ?L )). Note that with some rearranging of terms, this expression reduces to (1) when l = 0. Thus, the larger is the lobbying cost l, the greater is the range of potential biases for which the SIG will report truthfully. Thus, costly lobby aids the SIG in truthful reporting by allowing it to express the intensity of its preferences. As we shall see in the subsequent section, this result gives rise to a somewhat unexpected corporate strategy aimed at undermining the SIG’s ability of express the intensity of its preferences. At the same time, the SIG must be willing to incur the lobbying cost when the state is high, i.e. ?(? H ? ?H ? ? )2 ? l ? ?(?L ? ? H ? ? )2 , which can be rewritten as (3) l ? l ? (?H ? ?L )(2? + ? H ? ?L ). If both (2) and (3) hold, then the equilibrium described above exists. Letting a = (?H ? ?L ), Figure 1 illustrates the values of l and a that give rise to truthful reporting by the SIG. The top line in the figure represent the combinations of l and a for which the SIG is just indifferent between lobbying when the true state of the world in ?H and not lobbying in that state. Above this line the SIG will choose not to incur the costs of lobbying even in the high state. The lower line traces out the combination of l and a for which the SIG is just indifferent between lobbying in the low state (and announcing ? H ) and not lobbying in the low state. For all combinations of l and a below the lower the SIG would strictly prefer to lobby in the low state (announcing ?H ). An example may help to fix ideas. Consider the siting of a new paper-making facility, which will release some volume of organochlorines into a river. The facility could use a number of alternative technologies for bleaching the pulp, which vary in their use of chlorine in the bleaching process and, thus, in the amount of organochlorines they release into the environment. The environmental organization Greenpeace has been highly critical of organochlorine releases, since they result in the presence of trace amounts of dioxins— known carcinogens—in the river downstream from the plant. Suppose condition (1) holds and lobbying is costless. In this case, Greenpeace will always participate in hearings about the plant, and it will argue that dioxins are highly toxic chemicals whose release should be avoided, regardless of the bleaching technology to be used and the quantity of releases involved. Since Greenpeace will always protest regardless of the firm’s technology, its actions convey little about the intensity of its concerns about the technology. If it is costly for Greenpeace to participate in the hearings, however, then the benefits of participation are small when dioxins are released in minute amounts, so Greenpeace will eschew participation. It will allocate its scarce lobbying resources to fighting the plant only when relatively large amounts of dioxins are likely to be released. Thus, when lobbying is costly and Greenpeace does show up to participate in the proceedings, this is credible evidence that the harm from the plant’s dioxin releases are likely to be large, i.e. the true state is ? H . In this case the decision makers can learn from the actions of Greenpeace.
5
3.
The ‘‘Bear Hug’’ as Corporate Strategy
In this section, we model the firm as an active player in the lobbying process. Let the firm’s objective function be F = ?? p2 , where ? > 0. The parameter ? can be interpreted as an efficiency parameter. Firm’s with large ? ’s tend to be less efficient at adapting to more stringent policies. The structure of the firm’s objective function indicates that the firm’s profits are strictly declining in the stringency of the DM’s policy. This might be the case, for example, for a proposed new manufacturing facility. The vast majority of the firm’s shareholders do not live in the local community, and hence are not directly affected by issues such as the availability of jobs within the community or environmental impacts of the plant. Assuming the DM is aware of the firm’s objectives, then it is easy to see that the firm is not a credible source of information regarding the state of the world: regardless of the true state, the firm has incentives to claim the state is ?L .9 As in the previous section, we assume there exists a positive biased SIG, for which conditions (1) through (3) hold. As shown in the preceding section, there exists an equilibrium in which the SIG’s lobbying activity fully reveals to the DM the true state of the world. In this section we explore the firm’s relationship with the SIG under these circumstances. In particular, we consider a strategy in which the firm subsidizes the SIG’s lobbying activities, and examine when such a strategy might be profitable. This question is addressed in the following proposition. Proposition 1 If conditions (1) through (3) hold, then the firm has incentives to subsidize the SIG’ s lobbying activity if l ? ? (?H ? ? L )2 /4. Proof. If the SIG’s lobbying is informative, then the firm’s expected payoff is E (F ) = 2 2 2 ??? 2 L /2 ? ??H /2 = ?? (? L + ?H )/2. Alternatively, if the DM sets a policy simply based on its prior, the firm’s payoff is F = ?? (?H + ? L )2 /4. Let the difference between these two 2 2 payoffs be denoted by ? = ?? (?H + ? L )2 /4 ? [?? (?2 L + ?H )/2] = ? (? H ? ? L ) /4 > 0. Thus, the firm is willing to spend up to this amount to subsidize the SIG’s lobbying activity. The intuition behind the proposition is straightforward. The firm’s profits are concave with respect to the stringency of policy. It faces very high costs from a policy of p = ?H , and thus has incentives to take action to avoid this outcome. By committing to subsidize the SIG, the firm effectively undermines the SIG’s credibility with the DM (the SIG can no longer show the intensity of its preferences), and reduces the DM to adopting the policy p = (?L + ? H )/2, it’s optimal when the state of the world is unknown. The strategy is thus a form of ‘‘signal jamming,’’ similar in spirit to the analysis of Fudenberg and Tirole (1986) in the context of predation. We use the term ‘‘bear hug’’ to refer to a strategy in which the firm embraces its opposition, clasping it so close as to smother it and reduce its effectiveness. The proposition shows that the firm can benefit from a policy of ‘‘bear hugging,’’ that is, undertaking actions that appear to be ‘‘socially responsible,’’ but that are simply part of a profit-maximizing strategy. Baron (2001) refers to such actions as
9 As a consequence it is suboptimal for the firm to lobby the DM directly. It may,however exert considerable influence over the actions of the actions of the other players.
6
to
‘‘strategic corporate social responsibility,’’ in contrast to corporate social responsibility that is altruistically motivated. Note that the firm’s incentives to engage in a ‘‘bear hug’’ are proportional to ? (?H ? ?L )2 . Hence, the value of this strategy grows as the gap between the high state and the low state grows. This is quite intuitive, since the bear hug can be seen as a form of insurance against costly stringent policies. Furthermore, ceteris paribus, less efficient firms have more incentive to adopt this strategy than do more efficient firms. There are several potential issues in assessing when the ‘‘bear hug’’ is a viable strategy. First, the strategy must apply to situations where the firm does not know the true state of the world at the time when the subsidy is granted. The reason for this restriction is as follows. If the firm knew the true state of the world was ?L , it would prefer that the conditions of truthful revelation held. These condition would require that no subsidy be given. The firm would then only be willing to subsidize the SIG if the true state were ?H . However, if the subsidy was state dependent then the DM could determine the true state simply by observing the subsidy payment. Since the firm’s aim in granting the subsidy to the SIG is to illustrate to the DM that the SIG did not incur the lobbying cost l it is pointless for the firm to hide its subsidy to the SIG.10 The bear hug strategy is then more likely to apply to situations with true scientific uncertainty or situations with a risk of accidents than to situations of pure adverse selection, where the firm holds inside information. The bear hug can thus be seen as a kind of insurance policy against worst-case policy outcomes. The second is that the SIG must be willing to accept the SIG’s support. This will only be true if the SIG prefers to costlessly obtain the mean policy outcome than incurring the lobbying cost l in the high state to credibly deliver the report ?H . Mathematically: The SIG must prefer ´ ³ ´ ³ 2 2 (4) .5 (? (?H ? ? L ) /2 ? ?H ? ? ) + .5 (? (? H ? ?L ) /2 ? ?L ? ? )
? ? ? l/2. (5) Expanding (4) and comparing it to (5) we see that the SIG is willing to accept the subsidy l if (6) l ? (?H ? ?L )2 /2 2 Comparing the condition (6) to the firm’s subsidy condition l ? ? (?H ? ?L ) /4, we are led immediately to the following corollary of proposition 1, which we state without proof. Corollary 2 For sufficiently large ? the bear hug exists as a viable corporate strategy. Figure 2 illustrates the existence of various bear hug equilibria. The curve lBH provides the locus of lobbying costs l for which the SIG is willing to accept the bear hug over the relevant range of a = (?H ? ?L ). The curve lf illustrates, over the relevant range of a, the maximum subsidy the firm is wiling to pay. Since lf exceeds lBH the SIG will be willing to accept the firm’s bear hug over the entire range of a. Note, however, that the firm will only engage in the bear hug for those value of a for which lBH > l. If this condition is violated the bear hug is not necessary as the SIG’s report lacks credibility.
10
This is not true if the SIG is negative biased, as we discuss in the subsequent section.
7
As with the firm, we have examined the SIG’s willingness to accept the firm’s subsidy from an ex ante perspective. If the SIG knew that the true state of the world was high the SIG would wish to reject the firm’s subsidy. Consequently, acceptance of the subsidy would reveal that the true state of the world was ? L and the bear hug strategy would fail. Finally, it is worth noting that the firm must be able to ensure that its subsidy is used to subsidize the SIG lobbying costs on the issue of concern. Thus, there may be difficulties implementing the bear hug strategy if the SIG operates in multiple policy arenas. Returning to our example, a general purpose gift to Greenpeace may simply go to subsidize the group’s fixed costs, but may not guarantee that extra funds are devoted to lobbying about dioxin. Thus, the firm may need to tie the gift to SIG activity in a particular issue area. This might be done by providing the SIG with a forum in which it can express its views. For example, in the paper industry case, Greenpeace could be invited to participate in a paper industry forum, at the industry’s expense, thereby targeting the support toward a particular issue. The effects of the ‘‘bear hug’’ on the DM’s expected utility are shown in the following corollary. Corollary 3 When conditions (1) through (3) hold, the DM’ s expected utility is reduced when the firm adopts the ‘ ‘bear hug’ ’ strategy. Proof. Without the subsidy, the DM’ s expected utility is E (G) = 0. The SIG can be relied upon to reveal the true state, and the DM can thus tailor policy perfectly to each state of the world. When the firm subsidizes the SIG, the DM’ s expected utility is E (GBH ) = 2 (1/2)[?((?L + ?H )/2 ? ?L )] + (1/2)[?((? L + ?H )/2 ? ? H )]2 = ?(?H ? ? L )2 /4 < 0. Hence the DM is worse off when the firm supports the SIG. The corollary shows that under conditions (1) through (3), the DM is strictly worse off when the firm provides financial support to the SIG. While signal-jamming is profitable for the firm, it is socially deleterious because it prevents the optimal matching of policy to circumstances. To this point we have treated the firm’s subsidy decision as a pure strategy: either it subsidizes the SIG or it does not. Recalling that the subsidization strategy involves the subsidy payment prior to the realization of the true state of the world, it might be more profitable for the firm to employ a mixed strategy, that is, to subsidize with probability ?.11 If the DM believes that the firm is pursuing such a strategy it’s decision problem is complicated considerably. If the SIG does not appear, matters remain simple. The DM knows that with a subsidy, the SIG would always appear. The failure to appear thus allows the DM to infer there was no subsidy and that the state is ? H . The DM then sets p = ?H . If the SIG appears, however, the DM is now uncertain whether the state is really ?L . The following proposition establishes the conditions under which randomized subsidization is an optimal strategy.
11
Alternatively the firm may employ one of the two pure strategies depending on its type (the size of its ? ), but the DM might face uncertainty over which type of firm it faces. See Aumann and Brandenburger (1995) for this alternative view of the mixed strategy equilibrium.
8
Proof. If the SIG fails to appear, then the DM will update his expectation of the true state of the world is ?H to unity and optimally set p = ?H . Given the SIG appears, however, the DM’s posterior belief of the true state of the world will be give by Pr(?L |X ) = Pr(X |? L ) Pr(?L )/ Pr(X ), where X is the event that the SIG appears before the DM. The DM’s prior is that Pr(?L ) = 1/2. In addition, he knows Pr(X |?L ) = 1, because even if there is not a subsidy then the SIG will appear. The DM thus needs to compute the unconditional probability that the SIG appears. Let Y denote the event that a subsidy was given, and Y 0 the event that a subsidy was not given. Again, according to Bayes’ Rule Pr(X ) = Pr(X/Y ) Pr(Y ) + Pr(X/Y ) Pr(Y ). The firm’s strategy is to set Pr(Y ) = ?, so Pr(Y 0 ) = 1 ? ? . It is clear that Pr(X/Y ) = 1, and Pr(X/Y 0 ) = 1/2. Thus, when there is a chance of a subsidy, Pr(X ) = ? + (1 ? ? )/2 = (1 + ?)/2. Finally, we can conclude that Pr(?L |X ) = (1)(1/2)/[(1 + ?)/2] = 1/(1 + ? ). Thus, if the SIG appears, the DM infers that there is a probability 1/(1 + ?) that the state is actually ? L . The DM wishes to maximize the expected value of G = ?(p ? ?)2 . Thus, if the SIG appears, the DM’s objective is to ¤ ? 1 £ ?(p ? ?L )2 + [?(p ? ?H )2 ]. max E (G) = p 1+? 1+? Taking the first-order condition and solving, we find that ? L + ?? H p? = . 1+? We must now check whether the firm would indeed find it profitable to pursue a strategy of randomization regarding its subsidy payment, given the DM’s behavior. Given the DM’s policy function the firm’s expected profits are ?? = ?[?? (p? )2 ? l] + (1 ? ?)[.5{?? (?H )2 } + .5{?? (p? )2 }]. Differentiating with respect to ? and solving yields ? p ? (? H ? ? L ) .5 ? ? 1. ? ˆ = ?1 + (? H ? ?L ) 2.0? l = l 2l Some tedious calculations show that expected profits are concave in ? . Hence the firm’s optimal strategy is ? ? 0 if l ? (?H ? ?L )2 /2 ? ? ˆ ? (0, 1) if l ? ((?H ? ?L )2 /8, (? H ? ?L )2 /2) ? = ? 1 if l ? (?H ? ?L )2 /8. This refines our earlier result, which showed that if l ? (?H ? ? L )2 /4 then subsidizing with certainty is more profitable than not subsidizing at all. The existence of this mixed strategy equilibrium is dependent on the assumption that the DM either knows or believes that the firm is engaging in mixing behavior 9
Proposition 4 Suppose conditions (1) through (3) hold. Then, if l ? ? (?H ? ?L )2 /2, the firm chooses not to subsidize the SIG, while if l ? ? (? H ? ?L )2 /8, then the firm subsidizes )2 /8 (? H ? ?L )2 /2), the firm randomizes, subsidizing with certainty. For l ? ¡ (? (?H ? ? L? ? ¢, ? ? with probability ? = (?H ? ?L ) ? / 2l ? 1.
4.
Astroturf
In the previous section we have shown how the firm may benefit, paradoxically, by supporting the lobbying activities of a group whose bias is opposite to that of the firm. In this section, we consider an alternative corporate strategy of Astroturf mentioned in the introduction. The strategy involves supporting an interest group whose bias is negative. Such a SIG is a natural ally of the firm since the SIG’s optimal policy outcome is more lenient than the DM’s optimal policy in all states of the world. As we have mentioned, our bear hug strategy is limited to situation in which subsidy payments are made prior to the revelation of the true state of the world to any player. Also, the firm must be able to ensure that the subsidy payment is aimed directly at the SIG’s lobbying costs. In many situations the firm may know the true state of the world prior to its permit application. The small, but growing literature on environmental justice clearly asserts that community characteristics are an important factor in the citing of industrial plants.12 Under the astroturf strategy the firm’s subsidy payment takes place ex post, that is, after it already knows the state of the world. In addition, in keeping with the connotation of astroturf, we assume the DM does not observe the firm’s subsidy to the SIG. Later we allow the DM to invest in auditing the SIG in order to determine whether or not the SIG has been subsidized. We begin by considering a single SIG with U = ?(p ? ? ? ? )2 ? l, where ? < 0. Suppose that lobbying is costless, (l = 0). Since the SIG always prefers a lower level of policy than the DM, it has no incentive to misreport when the state is ? = ?L . Misreporting may be desirable, however, if ? = ?H since the SIG may prefer to obtain a lower policy than ?H even in the high state. The SIG may place a greater value on the economic impacts of the firm’s facility than the community in general. The SIG will fail to report truthfully if Thus, the SIG will send a false report, when ? = ? H , if ?(?H ? ? H ? ? )2 < ?(? L ? ?H ? ? )2 .
(7) ? ? ?(?H ? ?L )/2. We can see from condition (7) that the SIG has an incentive to misreport if its bias is relatively large in absolute value. If condition (7) holds then the SIG will always report ? L and the DM’s optimal response to SIG’s report will be to set the average policy, since the SIG’s report carries no credible information. Similar to our result in the previous section, it is possible for the SIG to report credibly even when condition (7) holds if lobbying is costly. In this case the SIG only delivers a report when the state is low, since a policy mistake in this state is very costly to the SIG; if the state is high, however, the SIG may find it too costly to deliver a false report. As a result, the DM can infer that the state is low when the SIG appears, and high when it does not appear. For this equilibrium to exist, the SIG must prefer to refrain from lobbying when ? = ?H , i.e.,?(?H ? ?H ? ? )2 ? ?(?L ? ?H ? ? )2 ? l, or
12
See Taylor (1992) or Greer and Harding (1993).
10
l ? l ? ?(?H ? ?L )(2? + (?H ? ? L )). (8) recall that ? < 0. At the same time, the SIG must be willing to incur the lobbying cost when the state is low, i.e. ?(?L ? ?L ? ? )2 ? l ? ?(?H ? ? L ? ? )2 , which can be rewritten as (9) l ? l ? ?(?H ? ?L )(2? ? (?H ? ? L )). If both (8) and (9) hold, then the equilibrium described above exists. Figure 1 shows the region of truthful reporting in this case, where a = (? H ? ?L ). The question we want to investigate is: Can the firm raise its expected payoff relative to its payoff under truthful reporting? If the firm could once again subsidize the SIG openly prior to the revelation of the true state of the world it could obtain the average policy (which it would find preferable to its expected outcome under truthful reporting). That is, the firm could use the bear hug strategy on a SIG that is favorably biased towards the firm. Recall, however, that this requires payment prior to the firm knowing the state of the world. If it is common knowledge that the firm knows the state of the world the bear hug strategy would fail because it would be suboptimal for the firm to subsidize the SIG if the state of the world was low. Recall that under the conditions of astroturf the firm’s subsidy to the negatively biased SIG occurs ex post and is hidden from the DM.13 Although we assume that the DM cannot observe the firm’s subsidy payment, our discussion of astroturf in the introduction clearly indicated that policy makers are aware of the strategy. Thus, we assume that the DM can expend some resources in auditing the SIG’s actions in an attempt to determine whether a subsidy did in fact occur. Let ? denote the probability with which the DM conducts an audit, and ? denote the probability an audit, if conducted, generates conclusive information about whether a subsidy was conferred. The cost of maintaining an audit function is c(?), where c0 (?) > 0 and c00 (?) > 0. There are two possible types of equilibria, one in which astroturf does not occur and one in which it does.
4.1
No Astroturf
In this case, the DM knows that if the SIG appears the state is ?L , and if the SIG does not appear the state is ?H . To ensure these conditions hold however the DM must audit the SIG when it appears, in order to eliminate firm incentives for astroturf. Assuming the firm does not engage in astroturf, the DM can infer correctly the state of the world, and sets the optimal policy for each state. Let the DM’s equilibrium audit probability in this case be ?NA . Thus, the DM’s expected payoff is E (U ) = ?c(?NA )/2, since setting the correct policy generates an optimal utility of zero in both states. Conditional on the DM’s audit policy, the firm must prefer not to astroturf in state ?H . The firm’s profits if it does not astroturf are ?NA = ???2 H . If it were to astroturf, expected
13
Note that a policy of astroturfing a positive-biased SIG is not rational. The positive-biased SIG will always claim the state is high, and hence the firm does not want to subsidize it to appear before the DM and support efforts to make policy more stringent.
11
profits would be
NA ? = ?NA ? (???2 ? )(???2 (10) H ) + (1 ? ? L) ? l 2 NA NA ? )(???2 So a non-astroturf equilibrium requires ???H > ? ? (???2 H )+(1?? L )? 2 l. This can be rewritten as (1 ? ?NA ? )? (?2 H ? ?L ) ? l < 0. Thus, the DM must choose ?NA to make this inequality hold. This implies
l 1 ? (11) 2 . ? ? ? (? 2 H ? ?L) The DM must engage in this level of auditing to support the no-astroturf equilibrium.14 Note that the DM wants to minimize auditing expenses, consistent with preventing astroturf, so the firm’s profits in state ? H are simply ?NA = ?? 2 H. ?NA >
4.2
Astroturf Equilibrium
In an astroturf equilibrium the firm will find it profitable to subsidize the SIG in the high state even though it knows there is a possibility that its subsidy will be detected. In this case the SIG will report ?L to the DM in both states of the world. When the SIG appears, the DM audits with probability ?A . If the audit is conclusive, the DM knows for certain whether or not a subsidy was given. If the audit is not conclusive, then the DM sets the optimal policy given its uncertainty about the state, i.e., it sets p = (?H + ?L )/2. In the astroturf equilibrium the DM knows that the SIG is subsidized in the high state. If the audit is inconclusive then the presence of the SIG conveys no information and it is optimal to set the average policy. The DM’s expected utility is ?H + ? L 1 A [? ? (?0) + (1 ? ?A ? )(?( ? ?L )2 ] 2 2 ? H + ?L 1 ? ? H )2 ] ? c(?A ) + [?A ? (?0) + (1 ? ?A ? )(?( 2 2 (? H ? ? L )2 ? c(?A ) = ?(1 ? ?A ? ) 2 Optimizing with respect to ? yields E (U ) =
(12)
? ? E (U ) = (?H ? ? L )2 ? c0 (?A ) = 0. (13) ?? 2 Given our assumptions about c (?), it is easily checked that expected utility is concave in ?. Next we must check whether astroturf is profitable for the firm. If it chooses to astroturf, and ? = ?H , then
A ? A = ?A ? (???2 H ) + (1 ? ? ? )(?? (
14
?H + ?L 2 ) ) ? l. 2
< ?.
(14)
Since ?N A ? (0, 1) condition (11) requires that 0 <
l ? (? 2 ?? 2 ) H L
12
Alternatively, if it were not to engage in astroturf, and the state were ? = ?H , then the SIG would not appear and the DM would be able to infer the state. In this case, the firm 2 A would earn ? = ??? 2 H . Thus, for astroturf to be profitable, we require ? > ???H . That is
A ?A ? (???2 H ) + (1 ? ? ? )(?? ( which implies
?H + ?L 2 ) ) ? l + ??2 H > 0, 2
(15)
1 ? ?A ? ? (3? H + ?L )(?H ? ? L ) > l. (16) 4 This inequality must be consistent with conditions (8) and (9). The relevant condition here is (8). Conditions (16) and (8) require that 1 ? ?A ? ? (3?H + ? L )(?H ? ?L ) ? ?(?H ? ? L )(2? + (?H ? ?L )) 4 which can be rewritten as 4 (2? + (? H ? ?L )) . ??? (1 ? ?A ? ) (3?H + ?L ) Thus, we obtain the following lemma. (17)
(18)
Lemma 5 For sufficiently large ? the firm finds it profitable to engage in the astroturf strategy. Lemma 5 states that if the firm finds stringent policies sufficiently costs it will find it profitable to engage in the astroturf strategy, i.e., subsidize the SIG in the high state of the world even when it faces a positive probability of detection. In doing so the firm benefits from obtaining an average policy in the high state, although it does sacrifice the possibility of obtaining ?L when the state of the world is low. Finally, to determine whether an astroturf equilibrium will exist, we need to check whether the DM would prefer to deter astroturf and shift to the no-astroturf equilibrium. In the astroturf equilibrium, the firm engages in astroturf when ? = ?H , and the DM knows this. In response, the DM audits, but not frequently enough to make astroturf unprofitable. ?L )2 ? c(?A ). Recall that E (U NA ) = ?c(?NA )/2, while E (U A ) = ?(1 ? ?A ? ) (?H ? 2 Thus, for an astroturf equilibrium we require (? H ? ? L )2 ? c(?A ) > ?c(?NA )/2 (19) 2 The only way the astroturf equilibrium can be preferred is if ?NA is very high and/or the cost function is highly convex. Thus, we have the following proposition ? (1 ? ?A ? ) Proposition 6 An astroturf equilibrium exists when condition (19) and lemma 5 hold.
2 1 ? ? ?(?2l ??2 ) = [? (? 2 Recall that ?NA > ? H ? ?L ) ? l]/? . Thus, the size of ? is critical H L to determining which equilibrium can be supported. When ? is small, it becomes difficult for the DM to deter astroturf and ?NA becomes large. At the same time, in the astroturf
13
equilibrium, the marginal value of auditing declines so the DM audits less frequently. This increases the loss due to using a policy that doesn’t match the true state of the world, though it does decrease the DM’s expenditures on auditing. If c(?) is highly convex, then the increased costs of deterring astroturf will dominate, and the DM will be more likely to allow an astroturf equilibrium when ? is small.
5.
Self-Regulation
As noted in the previous section, astroturf is only possible with a SIG that has a negative bias. Thus, if the SIG which the firm wishes to influence has a positive bias, then a different strategy is necessary. While the bearhug can be used with either type of SIG, firms may not always be able to credibly commit to the use of this strategy. In this section, we study the possibility that the firm may be able to reduce the severity of the high state, i.e. to reduce ?H , through voluntary improvements made ex ante. This might be done, for example, through design measures for a new facility that reduce the impact of worst-case outcomes. Such actions, especially if they involve sunk costs, will readily be regarded as credible commitments. Our basic notion here is that if the difference between the high and low states is sufficiently small, then the SIG will have little motivation to show up before the DM. Hence, self-regulation by the firm may induce the SIG to eschew lobbying. What benefit does the firm obtain from doing this? As shown in Proposition 1, if the SIG does not appear and the DM must set policy based on uncertainty about the true state, the firm stands to benefit by an amount ? = ? (?H ? ?L )2 /4. Let us introduce the notation a = ?H ? ?L , and denote by a0 the initial gap between the states. We consider self-regulation as a voluntary reduction in a on the part of the firm, cutting it from a0 to a1 . Thus, if the firm reduces ?H from ? L + a0 to ?L + a1 , and this induces the SIG to eschew lobbying, then the DM sets policy at the expected level of the state of the world, i.e. at p = ?L + a1 /2, and the firm’s payoff is F SR = 2 ?? (?L + a1 /2)2 = ?? (?2 L + a1 ?L + a1 /4). If the firm took no action, and the SIG revealed the true state through its lobbying decisions, then the firm’s expected payoff would 2 2 2 be F 0 = ?? (?2 H + ?L )/2 = ?? (?L + a0 ?L + a0 /2). The net benefit to the firm is V SR 0 2 2 ? F = ? (a0 ? a1 )?L + ? (2a0 ? a1 )/4 > 0. ? (a) = F Recall that the payoff function for the firm is F = ?? p2 . How should we represent the cost of achieving a? If the firm were forced to comply with a policy of p = ?H , the cost difference between a0 to a1 is k(a1 ) = ?? (?L + a1 )2 ? (?? (?L + a0 )2 ) = ? (2?L + a0 + a1 )(a0 ? a1 ). In order to be consistent with this payoff function, we will assume that the cost to the firm of achieving such a reduction is k(a1 ) = ? (2?L + a0 + a1 )(a0 ? a1 ). Combining the benefits and the cost of voluntary action gives a net payoff to voluntary action of V (a1 ) = ?V (a1 ) ? k(a1 ) 2 = ? (a0 ? a1 )? L + ? (2a2 0 ? a1 )/4 ? ? (2?L + a0 + a1 )(a0 ? a1 ) 14
Differentiating with respect to a1 yields ? V/? a1 = ??L + 3a1 ? /2 > 0, which shows that the firm prefers the largest a1 (smallest amount of self-regulation) that is sufficient to cause the SIG to drop out. Setting V (a1 ) = 0 and solving, we find that the maximum a that q ¡ 2 ¢ 2? 2? the firm would be willing to undertake is a1 = ? 3 ?L + 3 ?L + 3?L a0 + 3a0 2 /2 ? 0. Note that the maximum amount of self-regulation the firm is willing to undertake is increasing in the initial ‘‘gap,’’ a0 , since ? a1 /? a0 = How will a voluntary action affect the decisions of the SIG? Recall that conditions (2) and (3) define when the SIG will find it worthwhile to appear before the DM. If l ? l = a(2? + a) then the SIG finds it worthwhile to show up when the state is ?H , while if l ?l= a(2? ? a) then the SIG does not show up when the state is ? L . Clearly self-regulation shifts l and l, and in the process may cause the SIG to change its lobbying behavior. This is perhaps most easily seen by reference to Figure 1, in which l and l divide the (a, l) space into three regions: 1) The region with l > l, in which the SIG never lobbies, 2) the region with l l > l, in which the SIG lobbies only if ? = ?H . Of course, only in the third region is lobbying activity actually informative to the DM. Figure 3 builds on Figure 1 but adds two shaded regions. In each of these regions, the initial point (a0 , l) is within horizontal distance a1 of either l or l, so selfregulation can profitably result in a westward move that causes the SIG to change behavior. In the shaded region close to l, self-regulation causes the SIG to abandon lobbying. In the shaded region close to l, self-regulation has the opposite effect: it induces the SIG to lobby in both states of the world. In either case, however, the SIG’s lobbying behavior becomes uninformative for the DM, which is profitable for the firm. We summarize these results in the following proposition. Proposition 7 There exist two regions in (a, l) space for which corporate self-regulation can profitably alter interest group lobbying behavior: a) if a0 (2? + a0 ) > l > a0 (2? ? a0 ) and l > (a0 ? a1 )(2? + a0 ? a1 ), then self-regulation induces the interest group to eschew lobbying, and b) if a0 (2? ? a0 ) < l < (a0 ? a1 )(2? ? a0 + a1 ), then self-regulation induces the interest group to become a pure advocate that lobbies regardless of the actual state of the world. In either case, the interest group’ s lobbying behavior becomes uninformative and profits rise. Proof. Begin with case (a). We require two conditions. First, a0 (2? +a0 ) > l > a0 (2? ?a0 ) s lobbying behavior is informative; ensures that the initial pair (a0 , l) is such that the SIG’ that is, it ensures that l ? (l, l). Second, l > (a0 ? a1 )(2? + a0 ? a1 ) ensures that after self-regulation, the SIG is in the region in which lobbying is never worthwhile; that is, after self-regulation, we have l > l. Now turn to case (b). The condition a0 (2? ? a0 ) < l ensures that the SIG does not always lobby at the initial pair (a0 l). The second condition l < (a0 ? a1 )(2? ? a0 + a1 ) ensures that after self-regulation, we have l a/2) that the SIG has an incentive to report falsely when ? = ?L . If it can thereby induce the DM to set a policy of p = ?H , it obtains payoff ?(a ? ? )2 ? l rather than the payoff of ?? 2 it would receive for truthful reporting that yields a policy of p = ?L . Indeed, if a = ?H ? ?L = ? , then the SIG’s most preferred situation is to report falsely when ? = ? L and thereby obtain policy p = ? L + ? . Note that for a > ? , the SIG’s payoff from false reporting rises when a is reduced closer to ? . This is because larger levels of a mean that a policy of p = ?H is really higher than the SIG would like if the state is really ? = ?L ; as a result, the incentive to lie is reduced. Self-regulation thus makes false reporting more advantageous for the SIG. It is also interesting to ask how the decisionmaker is affected by self regulation that renders lobbying uninformative. This is the subject of the following proposition.
Proposition 8 The public decisionmaker is worse off when the firm engages in self-regulation. Proof. Without self regulation, the DM’s expected payoff is simply G0 = 0, since policy can be tailored precisely to the state of the world ex post. With self regulation, the DM lacks information about the state and must set an ‘‘average’’ policy based on his beliefs. Suppose that the firm’s self-regulatory actions are observed by the DM. Then the DM sets p = ?L + a1 /2, which is equal to the expected value of the state. Now the DM’s expected payoff becomes Clearly GSR < G0 , and the DM is worse off as a result of the firm’s self-regulatory action. In this model, the DM is best able to maximize his objective function when he has full information about the state of the world. Then he can tailor policy to the specifics of the situation before him. Self-regulation is only undertaken by the firm if it will render the SIG’s lobbying uninformative. This deprives the DM of the information he desires, and as a result the DM is worse off. This result contrasts with that in earlier work, such as that of Lyon and Maxwell (2002), who show that the regulator benefits when industry selfregulation preempts the imposition of new regulations. The key difference is that Lyon and Maxwell (2002) study a model in which self-regulation does not affect the information flow to the regulator. It is worth noting that once lobbying is rendered uninformative, the DM does benefit from further corporate self-regulation, but that this never entirely compensates for the loss of information caused by the decision to self-regulate. 16 GSR = ?.5(?L + a1 /2 ? ?L )2 ? .5(?L + a1 /2 ? ?H )2 = ?a2 1 /2 < 0.
6.
Multiple Interest Groups
To this point, we have concentrated on the case of a single firm and a single interest group. In this section, we discuss how our results may be extended to cases with multiple interest groups. We follow the typology used by Grossman and Helpman (2001) to classify the structure of multiple SIG situations: 1) ‘‘Like bias’’ arises when all groups share the same direction of bias, but with different intensity; 2) ‘‘Opposite bias’’ arises when different groups are biased in opposite directions, and 3) ‘‘Unknown bias’’ arises when the groups receive imperfect signals regarding the state of the world. The first two cases, in contrast to the third, assume that both SIGs have perfect information regarding the state of the world. We consider these in turn, focusing on the case of two SIGs for simplicity.
6.1
Like Bias
We will label the two SIGs ‘‘radical’’ and ‘‘moderate,’’ with the former possessing a larger value of ? . The more radical group may be inclined to ‘‘babble,’’ i.e. to always claim that the state is high. This may be true even if lobbying is costly, if the radical group’s ? is high enough. The moderate group is more likely to reveal the true state. For example, it may only show up to lobby when the state is really high. If the firm prefers a policy set at the average level, then it prefers to mute the moderate group, since the radical group lacks credibility anyway. This can be accomplished by bearhugging the moderate group ex ante, if that group’s bias is great enough that it will always claim the state is high when lobbying is costless. (Alternatively, similar results can be achieved through astroturfing ex post, if the group has a negative bias.) Thus, this case differs little from the single SIG case analyzed above. Alternatively, if the radical SIG’s bias is not too great, then the DM could also rely on it to provide reliable information through costly lobbying.15 In this case, bearhugging (or astroturfing) the moderate SIG will not be sufficient to affect the DM’s decision. Instead, the firm must subsidize both SIGs. Again, however, this case differs only trivially from the case of a single SIG. A more substantial change occurs if the moderate group has a small bias. Now it may tell the truth in a world of costless lobbying. In this case, the firm doesn’t want to subsidize the moderate group. Instead, it prefers to raise the SIG’s costs of lobbying, perhaps by sponsoring scientific studies that call the extent of the problem into question. This may force the SIG to spend more on scientific work to establish credibly that there really are significant impacts from the issue. In this case, the moderate group might choose not to incur the cost of lobbying. Once again, though, this is a result that could also be found in the case of a single SIG. Overall, we conclude that the addition of a second SIG with like bias to that of the first SIG is unlikely to generate much additional insight. However, it is worth noting that if all groups must be subsidized, then the cost of any kind of subsidy strategy rises linearly with the number of SIGs. This is not true of the self-regulation strategy, however. A single
15
Krishna and Morgan (2001) show that in this case the DM rationally relies on only the more moderate of the SIGs to provide information.
17
voluntary improvement affects all SIGs at once. If the firm undertakes enough voluntary action to preempt the involvement of the most extreme group, then all other groups will be preempted as well. Thus, we hypothesize that self-regulation is likely to outperform subsidy strategies as the number of SIGs grows.
6.2
Opposite Bias
When the two SIGs are biased in opposite directions, matters become more interesting. At least two types of equilibria are possible: 1) DM ignores one SIG and simply relies on the other, and 2) Each SIG shows up in one state of the world. In particular, an equilibrium may exist in which the SIG with positive bias shows up in the high state, while the SIG with negative bias shows up in the low state.16 Recall from our earlier analysis that the bearhug can be applied to groups with either type of bias, but requires commitment ability and must be undertaken before the firm learns the state of the world. Astroturf does not require commitment ability, and can be undertaken ex post, but it can only be employed with groups having a negative bias. Self-regulation is undertaken ex ante, and will influence both types at once. Case 1 is similar to the case of like bias. If the firm successfully bearhugs the ‘‘active’’ SIG, then the inactive SIG may find it worthwhile to lobby, and the DM will find it worthwhile to pay attention to him. Thus, the firm may need to bearhug both of the SIGs. Alternatively, the firm may use self-regulation to influence both SIGs at once. If the SIG with positive bias is moderate while the SIG with negative bias is radical, then the firm may decide to use both self-regulation and astroturf. A moderate amount of self-regulation will be enough to preempt the lobbying activity of the moderate group, and astroturf could then be used to induce the radical SIG to become a pure advocate, thereby eliminating the informativeness of that group’s activity.17 Case 2 is not very different. If one group is silenced, then the initial equilibrium is destroyed. However, there is an alternative equilibrium in which the DM pays attention to only one of the SIGs, and this becomes the only equilibrium if one SIG is bearhugged. Hence, the firm must again silence both groups, either through bearhugs or self-regulation, if it wishes to be successful. The general point seems to be that the basic structure of our analysis remains valid in the presence of multiple SIGs, as long as those SIGs all possess full information regarding the state of the world. The main change from adding multiple groups is that self-regulation may become relatively more attractive as the number of SIGs rises.
6.3
Unknown Bias
Again, one group is assumed to be radical, and willing to lobby in both states of the world. However, the DM is assumed to be unable to distinguish one group from the other, hence can only make use of information regarding the number of firms that appear. Lohmann
16 17
See Grossman and Helpman (2001), chapter 5, for details. Note that self-regulation reduces the radical SIG’s incentive to lobby, and thus raises the firm’s cost of astroturfing. The two policies may thus be more in the nature of substitutes than of complements.
18
(1993) analyzes this setting in the context of N > 2 groups, but Grossman and Helpman (2001) show that her main insights can be derived in a model with just two groups. Consider the case of two groups with like biases. Lohmann emphasizes the case in which the more radical SIG always lobbies, regardless of the state. The DM does not know which group is the more biased, but can still use the extent of lobbying as a noisy signal regarding the state of the world. For example, the DM may conclude that the state is high if two SIGs lobby, and low if only one does.18 If the firm can identify the more moderate SIG, then it can subsidize the moderate group, just as in the case of known bias. If this is not possible, then the firm must subsidize both groups. Now consider the case of opposite bias. Suppose that the SIG with positive bias is the more radical one, and it plays the role of a pure advocate, that is, it always shows up to claim that the state is high. The more moderate SIG only lbbies when the state is low. Thus, the appearance of 1 SIG indicates that the state is high, while the appearance of 2 groups indicates the state is low, and the DM sets a low level of policy when both groups show up, but a high level when only one group appears. Once again, if the firm subsidizes the moderate group, then that group will always appear, and the DM must set policy without gaining any information from the SIGs.19 And if the firm cannot determine which group is which, then it must subsidize both. The case of unknown biases is more subtle than the first two cases we discussed, since the SIGs don’t know the state of the world for certain. Sometimes they will be wrong, and the DM must take this into account. Nevertheless, for our purposes, most of the qualitative features of the models seem basically the same. One new possibility may emerge in the case of N > 2 SIGs with imperfect information. If the groups move sequentially in presenting their information, the possibility of information cascades arises, as in Bikhchandani, Hirshleifer, and Welch (1992). In such a setting, there is a large premium to being the first SIG to lobby, since all the subsequent SIGs may be influenced by the actions of the first. There is also a large premium to the firm if it can influence the information revealed by the first SIG to lobby.
7.
Conclusions
In this paper, we have developed a model to explore how firms may influence the lobbying behavior of special interest groups. We built on the framework presented by Grossman and Helpman (2001), in which costly lobbying may convey information to a public decisionmaker. The basic idea is that when lobbying is costly, an interest group’s decision to lobby provides credible information about the strength of its preferences regarding a particular policy issue.
A failure to lobby by both firms is off-equilibrium path behavior. Grossman and Helpman (2001, p. 154) identify beliefs for the DM under which it infers the state is low when neither firm lobbies. 19 A question G&H don’t address is what the group says once it arrives. Presumably this is just a minor extension of the costless lobbying case we examined above, except that now the SIG is working off of some uncertainty regarding which is the true state. But it will still be the case that if the SIG would always claim ‘‘high’’ if lobbying were costless, then the bear hug will work.
18
19
We considered three corporate non-market strategies: 1) the ‘‘bearhug,’’ in which the firm subsidizes the lobbying activities of an interest group, 2) ‘‘astroturf,’’ in which the firm subsidizes the lobbying activities of a group with similar views, and 3) corporate social responsibility, in which the firm voluntarily limits the potential social harms from its activities. All three of these strategies can be used to reduce the informativeness of lobbying, which can be profitable for the firm if the costs of a complying with a stringent public policy are high. When compliance costs are convex, the firm gains if the public decisionmaker sets policy based on expected or average social harm, rather than face the risk that policy will be tailored to actual harm. The ‘‘bearhug’’ serves as a signal-jamming device that prevents the interest group from signalling the intensity of its views. One might expect that the group would be unwilling to accept a subsidy that reduces the credibility of the group’s statements. Nevertheless, we show that if lobbying is costly enough then it is optimal for the group to accept the firm’s embrace. It is important to note that this strategy may not be dynamically consistent for the firm: even though it raises expected profits ex ante, it is unprofitable ex post in some states of the world. Hence, the strategy is only feasible if the firm can commit to it when the firm does not yet know the true state of the world. This is particularly likely in situations of true scientific uncertainty, such as currently exists regarding the future impacts of global warming. In some situations, the firm is likely to know the true state of the world, especially if that state depends on characteristics of the firm’s technology or management processes. For example, the state of the world might be the level of health risk associated with the operation of a particular plant, which depends upon corporate decisions regarding technology and management. In such settings, the ‘‘bearhug’’ strategy cannot be implemented. However, ‘‘astroturf ’’ lobbying can be utilized by the firm when it already knows the state of the world. Astroturfing involves subsidy only in states of the world where the interest group would not otherwise lobby. We show that such subsidies will only be jointly incentive compatible for the firm and the interest group if the group’s political bias is toward relatively weak policies. For example, the group might represent local business organizations that stand to benefit if the firm builds a new plant in the area. We show that the decisionmaker has incentives to audit the relationship between the firm and the interest group for signs of astroturfing, but that in equilibrium astroturfing takes place nevertheless. The third strategy we study is self-regulation, namely, voluntary actions to reduce possibility of serious social harms. Such voluntary actions can change the lobbying incentives of interest groups, and may render them uninformative, which is profitable for the firm.20 Selfregulation has subtle effects in this model. The most intuitive effect is that self-regulation can preempt interest group lobbying, by reducing the payoff from lobbying relative to its costs. Another possibility, one that is less intuitive, is that self-regulation can strengthen the incentives of an positive-biased interest group to falsely report that the state is high when it is really low. An interest group with positive bias wants a policy greater than that justified by the true (low) state of the world, but it may not want the policy to be fully as stringent
20
Baron (2001) refers to such actions as ‘‘strategic’’ corporate social responsibility since it is motivated by profit-maximization rather than altruistic preferences.
20
as would be justified in the high state of the world. By bringing the high state closer to the low state, self-regulation makes it less costly for the interest group to endure the stringent policy, and makes it more attractive for the group to engage in lobbying in both states of the world. Under all three of these strategies, the public decisionmaker is made worse off. The key reason is that when the decisionmaker is fully informed, he can tailor policy precisely ex post to the particular state of the world. All three of the strategies we study here are designed to stem the flow of information, and while this increases profits it simultaneously renders the decisionmaker worse off. Further research is required to flesh out the full range of corporate non-market strategies for dealing with outside stakeholder groups. We have focused on the interactions between a single firm and a single interest group, but it would be interesting to explore further what happens when multiple interest groups are involved. Particularly promising is the setting in which interest groups are imperfectly informed about the state of the world.21 It would also be worthwhile to assess when combinations of strategies may be more profitable than any one strategy pursued singly. For example, there may be situations in which a combination of corporate social performance and astroturf lobbying is to be expected. Finally, empirical examination of these issues, perhaps through the use of case studies, would be valuable.
References
Aumman, R.J. and A. Brandenburger. 1995. ‘‘Epistemic Conditions for Nash Equilibrium,’’ . 2001. ‘‘Private Politics, Corporate Social Econometrica, 63: 1161-1180. Baron, David P Responsibility, and Integrated Strategy,’’ Journal of Economics and Management Strategy, 10: 7-46. Becker, Gary S. 1983 ‘‘A Theory of Competition among Pressure Groups for Political Influence,’’ Quarterly Journal of Economics, 98: 371-400. Bikhchandani, Sushil, David Hirshleifer, and Ivo Welch. 1992. ‘‘A Theory of Fads, Fashion, Custom, and Cultural Change as Informational Cascades,’’ Journal of Political Economy 100: 992-1026. Crocker, Keith J. and Sharon Tennyson. 1997. ‘‘Contracting with Costly State Falsification: Theory and Empirical REsults from Automobile Insurance,’’ mimeo, University of Michigan Business School. DeSimone, Livio D. and Frank Popoff. 2000. Eco-Efficiency: The Business Link to Sustainable Development. Cambridge, MA: The MIT Press. Fudenberg, Drew and Jean Tirole. 1986. ‘‘A ‘Signal-Jamming’ Theory of Predation,’’ The Rand Journal of Economcis, 17: 366-376. Greer, Marsha L. and Anna K. Harding. 1993. ‘‘The Health Impact of Hazardous Waste Sites on Minority Communities: Implications for Public Health and Environmental Health Professionals.’’ Journal of Environmental Health 55(77): 6-9. Grossman, Gene M. and Elhanan Helpman. 2001. Special Interest Politics. Cambridge,
21
If the groups move sequentially in presenting their information, the possibility of information cascades arises, as in Bikhchandani, Hirshleifer, and Welch (1992).
21
MA: The MIT Press. Khalil, Fahad. 1997. ‘‘Auditing without Commitment,’’ RAND Journal of Economics. 28: 629-640. Krishna, Vijay and John Morgan. 2001. ‘‘A Theory of Expertise,’’ Quarterly Journal of Economics, 116: 747-775. Lyon, Thomas P . and John W. Maxwell. 2002. ‘‘Self-Regulation, Taxation, and Public Voluntary Environmental Agreements,’’ Journal of Public Economics, forthcoming. Maxwell, John W., Thomas P . Lyon, and Steven C. Hackett. 2000. ‘‘Self-Regulation and Social Welfare: The Political Economy of Corporate Environmentalism,’’ Journal of Law and Economics. 43: 583-617. Olson, Mancur. 1965. The Logic of Collective Action. Cambridge, MA: Harvard University Press. Stauber, John and Sheldon Rampton. 1995. Toxic Sludge is Good for You: Lies, Damn Lies and the Public Relations Industry. Monroe, Maine: Common Courage Press. Stigler, George J. 1971. ‘‘The Economic Theory of Regulation,’’ Bell Journal of Economcs and Management Science, 2: 1-21. Taylor, D. 1992. ‘‘The Environmental Justice Movement.’’ EP A Journal 18(1): 24
22
doc_245035045.pdf
Lobbying (also lobby) is the act of attempting to influence decisions made by officials in the government, most often legislators or members of regulatory agencies.
Interest Group Lobbying and Corporate Strategy
Thomas P . Lyon and John W. Maxwell February 15, 2002
Abstract
We study corporate non-market strategies designed to influence the lobbying behavior of other special interest groups. We focus on conditions under which costly lobbying is an informative signal to policymakers about the true state of the world, and in which stringent policy is so costly to the firm that the firm is not viewed as a credible source of information. We study three corporate non-market strategies: 1) the ‘‘bearhug,’’ in which the firm subsidizes the lobbying activities of an interest group before the true state of the world is known, 2) ‘‘astroturfing,’’ in which the firm subsidizes the lobbying activities of a group with similar views after the state of the world is known, and 3) self-regulation, in which the firm voluntarily limits the potential social harm from its activities. All three of these strategies can be used to reduce the informativeness of lobbying. We identify conditions under which each strategy is profitable for the firm. All three strategies, however, reduce the welfare of the public decisionmaker.
1.
Introduction
The role of interest groups in politics has held a long-standing fascination for political economists. In the 1780s, James Madison famously warned of the power of ‘‘factions’’ in the Federalist, while nearly two hundred years later Mancur Olson and George Stigler elevated the study of interest group politics to an important field within economics.1 Early work theoretical work treated interest group ‘‘pressure’’ as a production function, smooth and twice continuously differentiable, but more recently theorists have been opening up the black box of political pressure to focus more specifically on campaign contributions or lobbying.2 Several recent contributions shed new light on the role of lobbying in conveying information to public decisionmakers.3 Even this recent work, however, typically does not distinguish firms from other special interest groups. We argue that in many lobbying situations, firms do indeed have preferences distinct from those of other groups. In particular, they often bear the costs of government policy but do not collect the benefits. This is especially true for policies dealing with externalities or the provision of public goods. In such circumstances, firms cannot credibly convey information because their powerful bias towards weak policies is common knowledge among decisionmakers. Nevertheless, firms
1 Olson (1965) elaborates a rational choice model of interest group action, while Stigler (1971) applies this approach to the study of regulation specifically. 2 Grossman and Helpman (2001) provide an excellent introduction to this literature. 3 See, for example, Lohmann (1993) and Krishna and Morgan (2001).
1
may be able to influence the lobbying behavior of other interest groups. The corporate strategies that accomplish this goal are the subject of our paper. One important example where firms may wish to influence lobbying behavior is in the siting of industrial facilities. Manufacturing plants, waste disposal facilities and electric generating units are likely to be met with a storm of local protesters organizing under the banner ‘‘Not In My Backyard’’ or NIMBY , for short. While protests may often present much noise with little signal, they also have the potential to provide decisionmakers with vital first-hand knowledge about how local communities will be affected by particular siting decisions. Corporate strategy has of necessity expanded to include techniques for responding to such protests. In this paper, we develop a formal interest-group model of corporate strategy in the lobbying process. We focus on three strategies for interacting with interest groups: the ‘‘bear hug’’ (which involves the corporate embrace, through subsidies, of other interest groups), ‘‘astroturf’’ (which involves targeted corporate subsides to groups supporting the company’s position), and self-regulation (which involves voluntary corporate actions that reduce the risk of social harm from the company’s operations). We provide conditions under which these strategies can be part of a Bayesian Nash equilibrium in the interestgroup game, and we also assess their effects on the welfare of the public decisionmaker. We coin the term ‘‘bear hug’’ to refer to a corporate strategy of embracing one’s opposition. This may involve providing financial support to a special interest group normally considered to have opposing views. For example, DeSimone and Popoff point out that ‘‘It is also important to recognize that there can be a disparity of resources and information between business stakeholder groups that makes trust difficult to develop. This may sometimes require action to redress the balance. Since the Brent Spar incident—when opposition prevented Shell from disposing of a large oil storage platform at sea—the company has made space available for environmental groups to explain their point of view in educational and other materials that it has prepared.’’4 The term ‘‘astroturf lobbying’’ was coined by Lloyd Bentsen, long-time Senator from Texas, to describe the artificial grass-roots campaigns that are created by public relations firms.5 This strategy differs from the ‘‘bear hug’’ in that it involves financial support of groups that support the company’s view. The creation of such groups is the business of public relations firms such as Davies Communications, whose advertising says ‘‘Traditional lobbying is no longer enough. Today numbers count. To win in the hearing room, you must reach out to create grassroots support. To outnumber your opponents, call thee leading grassroots public affairs communications specialists.’’6 Davies explains how his firm generates a ‘‘grassroots’’ letter-writing campaign techniques through the use of telephone banks: ‘‘We get them on the phone, and while we’re on the phone we say ‘Will you write a letter?’ ‘Sure.’ ‘Do you have time to write it?’ ‘Not really.’ ‘Could we write the
5 6 4 DeSimone and Popoff (2000), p. 165. Stauber and Rampton (1995), p. 79. Stauber and Rampton (1995), p. 90.
2
letter for you? I could put you on the phone right now with someone who could help you write a letter. Just hold, we have a writer standing by.’..If they’re close by we hand-deliver it. We hand-write it out on ‘little kitty cat stationery’ if it’s a little old lady. If it’s a business we take it over to be photocopied on someone’s letterhead. [We] use different stamps, different envelopes..Getting a pile of personalized letters that have a different look to them is what you want to strive for.’’7 Self-regulation is quite different from the other two strategies, in that it involves real changes in company operations that are designed to reduce the risks of social harm. If these actions are substantive enough, interest groups may decide that the further gains from lobbying are not enough to justify the costs, and they may eschew participation in the political process. Perhaps surprisingly, however, we show that self-regulation may also induce interest groups to participate more actively in the lobbying process, although the extra activity may not further the group’s goals in the long run. The remainder of the paper is organized as follows. Section 2 presents a simple model of the lobbying process. Section 3 studies the ‘‘bear hug’’ while section 4 considers ‘‘astroturf.’’ Section 5 addresses the effects of self regulation, and section 6 discusses extensions of our model to a setting with multiple interest groups. Section 6 concludes.
2.
A Simple Model of Lobbying
Our simple model of lobbying consists of two players, a government decision maker (DM), and a special interest group (SIG). In this version of the model the firm has no active role in the lobbying process . We assume that the firm proposes a new project that requires the approval of the decision maker.8 For example, the firm might propose a new manufacturing facility that requires a site license before it can be built, or the firm might propose to introduce a new pharmaceutical product that requires approval from the Food and Drug Administration before it can be produced. The proposed project may have social effects through a variety of mechanisms, e.g. it may create jobs in the local community, it may affect the environmental quality of the surrounding community, or it may affect the health and safety of that community. These effects can be characterized by a variable ? ? 0 and l = 0, i.e., the (positive biased) SIG knows the true state of the world and can costlessly communicate it to the DM. We examine the SIG’s incentives to report the state of the world when the DM believes the SIG’s announcement. Since the SIG always prefers a higher level of policy than the DM, it has no incentive to misreport when the state is ? = ?H . Misreporting may be desirable, however, if ? = ?L . In this case, the SIG misreports, i.e. reports ?H , if Thus, when ? = ?L , the SIG misreports if ?(? L ? ?L ? ? )2 < ?(? H ? ?L ? ? )2 ? > (?H ? ?L )/2. (1)
Consider a case where condition (1) holds. This implies that the SIG has a large degree of bias„ or alternatively, that the high and low states are relatively close together. Then the SIG will always report that ? = ? H , regardless of the actual state of the world. Assuming the DM knows ? , ? L , and ? H , he will recognize the SIG’s incentives, and hence will not update his prior based on the SIG’s report. Thus, the DM sets p = (? L + ?H )/2. Turn next to a case where lobbying is costly. Because the SIG is biased toward high levels of the policy, it is particularly concerned about the possibility that the DM sets p = ? L when the state is actually ?H . Thus, the SIG is strongly motivated to incur the cost of lobbying when the state is ?H , but may not find it worthwhile when ? = ? L . Under certain conditions, there exists an equilibrium in which the SIG only lobbies when ? = ? H , and the DM interprets the SIG’s lobbying as a statement that indeed ? = ? H . For this equilibrium 4
to exist, the SIG must prefer to refrain from lobbying when ? = ?L , i.e.,?(?L ? ?L ? ? )2 ? ?(? H ? ?L ? ? )2 ? l, or (2) l ? l ? (?H ? ? L )(2? ? (? H ? ?L )). Note that with some rearranging of terms, this expression reduces to (1) when l = 0. Thus, the larger is the lobbying cost l, the greater is the range of potential biases for which the SIG will report truthfully. Thus, costly lobby aids the SIG in truthful reporting by allowing it to express the intensity of its preferences. As we shall see in the subsequent section, this result gives rise to a somewhat unexpected corporate strategy aimed at undermining the SIG’s ability of express the intensity of its preferences. At the same time, the SIG must be willing to incur the lobbying cost when the state is high, i.e. ?(? H ? ?H ? ? )2 ? l ? ?(?L ? ? H ? ? )2 , which can be rewritten as (3) l ? l ? (?H ? ?L )(2? + ? H ? ?L ). If both (2) and (3) hold, then the equilibrium described above exists. Letting a = (?H ? ?L ), Figure 1 illustrates the values of l and a that give rise to truthful reporting by the SIG. The top line in the figure represent the combinations of l and a for which the SIG is just indifferent between lobbying when the true state of the world in ?H and not lobbying in that state. Above this line the SIG will choose not to incur the costs of lobbying even in the high state. The lower line traces out the combination of l and a for which the SIG is just indifferent between lobbying in the low state (and announcing ? H ) and not lobbying in the low state. For all combinations of l and a below the lower the SIG would strictly prefer to lobby in the low state (announcing ?H ). An example may help to fix ideas. Consider the siting of a new paper-making facility, which will release some volume of organochlorines into a river. The facility could use a number of alternative technologies for bleaching the pulp, which vary in their use of chlorine in the bleaching process and, thus, in the amount of organochlorines they release into the environment. The environmental organization Greenpeace has been highly critical of organochlorine releases, since they result in the presence of trace amounts of dioxins— known carcinogens—in the river downstream from the plant. Suppose condition (1) holds and lobbying is costless. In this case, Greenpeace will always participate in hearings about the plant, and it will argue that dioxins are highly toxic chemicals whose release should be avoided, regardless of the bleaching technology to be used and the quantity of releases involved. Since Greenpeace will always protest regardless of the firm’s technology, its actions convey little about the intensity of its concerns about the technology. If it is costly for Greenpeace to participate in the hearings, however, then the benefits of participation are small when dioxins are released in minute amounts, so Greenpeace will eschew participation. It will allocate its scarce lobbying resources to fighting the plant only when relatively large amounts of dioxins are likely to be released. Thus, when lobbying is costly and Greenpeace does show up to participate in the proceedings, this is credible evidence that the harm from the plant’s dioxin releases are likely to be large, i.e. the true state is ? H . In this case the decision makers can learn from the actions of Greenpeace.
5
3.
The ‘‘Bear Hug’’ as Corporate Strategy
In this section, we model the firm as an active player in the lobbying process. Let the firm’s objective function be F = ?? p2 , where ? > 0. The parameter ? can be interpreted as an efficiency parameter. Firm’s with large ? ’s tend to be less efficient at adapting to more stringent policies. The structure of the firm’s objective function indicates that the firm’s profits are strictly declining in the stringency of the DM’s policy. This might be the case, for example, for a proposed new manufacturing facility. The vast majority of the firm’s shareholders do not live in the local community, and hence are not directly affected by issues such as the availability of jobs within the community or environmental impacts of the plant. Assuming the DM is aware of the firm’s objectives, then it is easy to see that the firm is not a credible source of information regarding the state of the world: regardless of the true state, the firm has incentives to claim the state is ?L .9 As in the previous section, we assume there exists a positive biased SIG, for which conditions (1) through (3) hold. As shown in the preceding section, there exists an equilibrium in which the SIG’s lobbying activity fully reveals to the DM the true state of the world. In this section we explore the firm’s relationship with the SIG under these circumstances. In particular, we consider a strategy in which the firm subsidizes the SIG’s lobbying activities, and examine when such a strategy might be profitable. This question is addressed in the following proposition. Proposition 1 If conditions (1) through (3) hold, then the firm has incentives to subsidize the SIG’ s lobbying activity if l ? ? (?H ? ? L )2 /4. Proof. If the SIG’s lobbying is informative, then the firm’s expected payoff is E (F ) = 2 2 2 ??? 2 L /2 ? ??H /2 = ?? (? L + ?H )/2. Alternatively, if the DM sets a policy simply based on its prior, the firm’s payoff is F = ?? (?H + ? L )2 /4. Let the difference between these two 2 2 payoffs be denoted by ? = ?? (?H + ? L )2 /4 ? [?? (?2 L + ?H )/2] = ? (? H ? ? L ) /4 > 0. Thus, the firm is willing to spend up to this amount to subsidize the SIG’s lobbying activity. The intuition behind the proposition is straightforward. The firm’s profits are concave with respect to the stringency of policy. It faces very high costs from a policy of p = ?H , and thus has incentives to take action to avoid this outcome. By committing to subsidize the SIG, the firm effectively undermines the SIG’s credibility with the DM (the SIG can no longer show the intensity of its preferences), and reduces the DM to adopting the policy p = (?L + ? H )/2, it’s optimal when the state of the world is unknown. The strategy is thus a form of ‘‘signal jamming,’’ similar in spirit to the analysis of Fudenberg and Tirole (1986) in the context of predation. We use the term ‘‘bear hug’’ to refer to a strategy in which the firm embraces its opposition, clasping it so close as to smother it and reduce its effectiveness. The proposition shows that the firm can benefit from a policy of ‘‘bear hugging,’’ that is, undertaking actions that appear to be ‘‘socially responsible,’’ but that are simply part of a profit-maximizing strategy. Baron (2001) refers to such actions as
9 As a consequence it is suboptimal for the firm to lobby the DM directly. It may,however exert considerable influence over the actions of the actions of the other players.
6
to
‘‘strategic corporate social responsibility,’’ in contrast to corporate social responsibility that is altruistically motivated. Note that the firm’s incentives to engage in a ‘‘bear hug’’ are proportional to ? (?H ? ?L )2 . Hence, the value of this strategy grows as the gap between the high state and the low state grows. This is quite intuitive, since the bear hug can be seen as a form of insurance against costly stringent policies. Furthermore, ceteris paribus, less efficient firms have more incentive to adopt this strategy than do more efficient firms. There are several potential issues in assessing when the ‘‘bear hug’’ is a viable strategy. First, the strategy must apply to situations where the firm does not know the true state of the world at the time when the subsidy is granted. The reason for this restriction is as follows. If the firm knew the true state of the world was ?L , it would prefer that the conditions of truthful revelation held. These condition would require that no subsidy be given. The firm would then only be willing to subsidize the SIG if the true state were ?H . However, if the subsidy was state dependent then the DM could determine the true state simply by observing the subsidy payment. Since the firm’s aim in granting the subsidy to the SIG is to illustrate to the DM that the SIG did not incur the lobbying cost l it is pointless for the firm to hide its subsidy to the SIG.10 The bear hug strategy is then more likely to apply to situations with true scientific uncertainty or situations with a risk of accidents than to situations of pure adverse selection, where the firm holds inside information. The bear hug can thus be seen as a kind of insurance policy against worst-case policy outcomes. The second is that the SIG must be willing to accept the SIG’s support. This will only be true if the SIG prefers to costlessly obtain the mean policy outcome than incurring the lobbying cost l in the high state to credibly deliver the report ?H . Mathematically: The SIG must prefer ´ ³ ´ ³ 2 2 (4) .5 (? (?H ? ? L ) /2 ? ?H ? ? ) + .5 (? (? H ? ?L ) /2 ? ?L ? ? )
? ? ? l/2. (5) Expanding (4) and comparing it to (5) we see that the SIG is willing to accept the subsidy l if (6) l ? (?H ? ?L )2 /2 2 Comparing the condition (6) to the firm’s subsidy condition l ? ? (?H ? ?L ) /4, we are led immediately to the following corollary of proposition 1, which we state without proof. Corollary 2 For sufficiently large ? the bear hug exists as a viable corporate strategy. Figure 2 illustrates the existence of various bear hug equilibria. The curve lBH provides the locus of lobbying costs l for which the SIG is willing to accept the bear hug over the relevant range of a = (?H ? ?L ). The curve lf illustrates, over the relevant range of a, the maximum subsidy the firm is wiling to pay. Since lf exceeds lBH the SIG will be willing to accept the firm’s bear hug over the entire range of a. Note, however, that the firm will only engage in the bear hug for those value of a for which lBH > l. If this condition is violated the bear hug is not necessary as the SIG’s report lacks credibility.
10
This is not true if the SIG is negative biased, as we discuss in the subsequent section.
7
As with the firm, we have examined the SIG’s willingness to accept the firm’s subsidy from an ex ante perspective. If the SIG knew that the true state of the world was high the SIG would wish to reject the firm’s subsidy. Consequently, acceptance of the subsidy would reveal that the true state of the world was ? L and the bear hug strategy would fail. Finally, it is worth noting that the firm must be able to ensure that its subsidy is used to subsidize the SIG lobbying costs on the issue of concern. Thus, there may be difficulties implementing the bear hug strategy if the SIG operates in multiple policy arenas. Returning to our example, a general purpose gift to Greenpeace may simply go to subsidize the group’s fixed costs, but may not guarantee that extra funds are devoted to lobbying about dioxin. Thus, the firm may need to tie the gift to SIG activity in a particular issue area. This might be done by providing the SIG with a forum in which it can express its views. For example, in the paper industry case, Greenpeace could be invited to participate in a paper industry forum, at the industry’s expense, thereby targeting the support toward a particular issue. The effects of the ‘‘bear hug’’ on the DM’s expected utility are shown in the following corollary. Corollary 3 When conditions (1) through (3) hold, the DM’ s expected utility is reduced when the firm adopts the ‘ ‘bear hug’ ’ strategy. Proof. Without the subsidy, the DM’ s expected utility is E (G) = 0. The SIG can be relied upon to reveal the true state, and the DM can thus tailor policy perfectly to each state of the world. When the firm subsidizes the SIG, the DM’ s expected utility is E (GBH ) = 2 (1/2)[?((?L + ?H )/2 ? ?L )] + (1/2)[?((? L + ?H )/2 ? ? H )]2 = ?(?H ? ? L )2 /4 < 0. Hence the DM is worse off when the firm supports the SIG. The corollary shows that under conditions (1) through (3), the DM is strictly worse off when the firm provides financial support to the SIG. While signal-jamming is profitable for the firm, it is socially deleterious because it prevents the optimal matching of policy to circumstances. To this point we have treated the firm’s subsidy decision as a pure strategy: either it subsidizes the SIG or it does not. Recalling that the subsidization strategy involves the subsidy payment prior to the realization of the true state of the world, it might be more profitable for the firm to employ a mixed strategy, that is, to subsidize with probability ?.11 If the DM believes that the firm is pursuing such a strategy it’s decision problem is complicated considerably. If the SIG does not appear, matters remain simple. The DM knows that with a subsidy, the SIG would always appear. The failure to appear thus allows the DM to infer there was no subsidy and that the state is ? H . The DM then sets p = ?H . If the SIG appears, however, the DM is now uncertain whether the state is really ?L . The following proposition establishes the conditions under which randomized subsidization is an optimal strategy.
11
Alternatively the firm may employ one of the two pure strategies depending on its type (the size of its ? ), but the DM might face uncertainty over which type of firm it faces. See Aumann and Brandenburger (1995) for this alternative view of the mixed strategy equilibrium.
8
Proof. If the SIG fails to appear, then the DM will update his expectation of the true state of the world is ?H to unity and optimally set p = ?H . Given the SIG appears, however, the DM’s posterior belief of the true state of the world will be give by Pr(?L |X ) = Pr(X |? L ) Pr(?L )/ Pr(X ), where X is the event that the SIG appears before the DM. The DM’s prior is that Pr(?L ) = 1/2. In addition, he knows Pr(X |?L ) = 1, because even if there is not a subsidy then the SIG will appear. The DM thus needs to compute the unconditional probability that the SIG appears. Let Y denote the event that a subsidy was given, and Y 0 the event that a subsidy was not given. Again, according to Bayes’ Rule Pr(X ) = Pr(X/Y ) Pr(Y ) + Pr(X/Y ) Pr(Y ). The firm’s strategy is to set Pr(Y ) = ?, so Pr(Y 0 ) = 1 ? ? . It is clear that Pr(X/Y ) = 1, and Pr(X/Y 0 ) = 1/2. Thus, when there is a chance of a subsidy, Pr(X ) = ? + (1 ? ? )/2 = (1 + ?)/2. Finally, we can conclude that Pr(?L |X ) = (1)(1/2)/[(1 + ?)/2] = 1/(1 + ? ). Thus, if the SIG appears, the DM infers that there is a probability 1/(1 + ?) that the state is actually ? L . The DM wishes to maximize the expected value of G = ?(p ? ?)2 . Thus, if the SIG appears, the DM’s objective is to ¤ ? 1 £ ?(p ? ?L )2 + [?(p ? ?H )2 ]. max E (G) = p 1+? 1+? Taking the first-order condition and solving, we find that ? L + ?? H p? = . 1+? We must now check whether the firm would indeed find it profitable to pursue a strategy of randomization regarding its subsidy payment, given the DM’s behavior. Given the DM’s policy function the firm’s expected profits are ?? = ?[?? (p? )2 ? l] + (1 ? ?)[.5{?? (?H )2 } + .5{?? (p? )2 }]. Differentiating with respect to ? and solving yields ? p ? (? H ? ? L ) .5 ? ? 1. ? ˆ = ?1 + (? H ? ?L ) 2.0? l = l 2l Some tedious calculations show that expected profits are concave in ? . Hence the firm’s optimal strategy is ? ? 0 if l ? (?H ? ?L )2 /2 ? ? ˆ ? (0, 1) if l ? ((?H ? ?L )2 /8, (? H ? ?L )2 /2) ? = ? 1 if l ? (?H ? ?L )2 /8. This refines our earlier result, which showed that if l ? (?H ? ? L )2 /4 then subsidizing with certainty is more profitable than not subsidizing at all. The existence of this mixed strategy equilibrium is dependent on the assumption that the DM either knows or believes that the firm is engaging in mixing behavior 9
Proposition 4 Suppose conditions (1) through (3) hold. Then, if l ? ? (?H ? ?L )2 /2, the firm chooses not to subsidize the SIG, while if l ? ? (? H ? ?L )2 /8, then the firm subsidizes )2 /8 (? H ? ?L )2 /2), the firm randomizes, subsidizing with certainty. For l ? ¡ (? (?H ? ? L? ? ¢, ? ? with probability ? = (?H ? ?L ) ? / 2l ? 1.
4.
Astroturf
In the previous section we have shown how the firm may benefit, paradoxically, by supporting the lobbying activities of a group whose bias is opposite to that of the firm. In this section, we consider an alternative corporate strategy of Astroturf mentioned in the introduction. The strategy involves supporting an interest group whose bias is negative. Such a SIG is a natural ally of the firm since the SIG’s optimal policy outcome is more lenient than the DM’s optimal policy in all states of the world. As we have mentioned, our bear hug strategy is limited to situation in which subsidy payments are made prior to the revelation of the true state of the world to any player. Also, the firm must be able to ensure that the subsidy payment is aimed directly at the SIG’s lobbying costs. In many situations the firm may know the true state of the world prior to its permit application. The small, but growing literature on environmental justice clearly asserts that community characteristics are an important factor in the citing of industrial plants.12 Under the astroturf strategy the firm’s subsidy payment takes place ex post, that is, after it already knows the state of the world. In addition, in keeping with the connotation of astroturf, we assume the DM does not observe the firm’s subsidy to the SIG. Later we allow the DM to invest in auditing the SIG in order to determine whether or not the SIG has been subsidized. We begin by considering a single SIG with U = ?(p ? ? ? ? )2 ? l, where ? < 0. Suppose that lobbying is costless, (l = 0). Since the SIG always prefers a lower level of policy than the DM, it has no incentive to misreport when the state is ? = ?L . Misreporting may be desirable, however, if ? = ?H since the SIG may prefer to obtain a lower policy than ?H even in the high state. The SIG may place a greater value on the economic impacts of the firm’s facility than the community in general. The SIG will fail to report truthfully if Thus, the SIG will send a false report, when ? = ? H , if ?(?H ? ? H ? ? )2 < ?(? L ? ?H ? ? )2 .
(7) ? ? ?(?H ? ?L )/2. We can see from condition (7) that the SIG has an incentive to misreport if its bias is relatively large in absolute value. If condition (7) holds then the SIG will always report ? L and the DM’s optimal response to SIG’s report will be to set the average policy, since the SIG’s report carries no credible information. Similar to our result in the previous section, it is possible for the SIG to report credibly even when condition (7) holds if lobbying is costly. In this case the SIG only delivers a report when the state is low, since a policy mistake in this state is very costly to the SIG; if the state is high, however, the SIG may find it too costly to deliver a false report. As a result, the DM can infer that the state is low when the SIG appears, and high when it does not appear. For this equilibrium to exist, the SIG must prefer to refrain from lobbying when ? = ?H , i.e.,?(?H ? ?H ? ? )2 ? ?(?L ? ?H ? ? )2 ? l, or
12
See Taylor (1992) or Greer and Harding (1993).
10
l ? l ? ?(?H ? ?L )(2? + (?H ? ? L )). (8) recall that ? < 0. At the same time, the SIG must be willing to incur the lobbying cost when the state is low, i.e. ?(?L ? ?L ? ? )2 ? l ? ?(?H ? ? L ? ? )2 , which can be rewritten as (9) l ? l ? ?(?H ? ?L )(2? ? (?H ? ? L )). If both (8) and (9) hold, then the equilibrium described above exists. Figure 1 shows the region of truthful reporting in this case, where a = (? H ? ?L ). The question we want to investigate is: Can the firm raise its expected payoff relative to its payoff under truthful reporting? If the firm could once again subsidize the SIG openly prior to the revelation of the true state of the world it could obtain the average policy (which it would find preferable to its expected outcome under truthful reporting). That is, the firm could use the bear hug strategy on a SIG that is favorably biased towards the firm. Recall, however, that this requires payment prior to the firm knowing the state of the world. If it is common knowledge that the firm knows the state of the world the bear hug strategy would fail because it would be suboptimal for the firm to subsidize the SIG if the state of the world was low. Recall that under the conditions of astroturf the firm’s subsidy to the negatively biased SIG occurs ex post and is hidden from the DM.13 Although we assume that the DM cannot observe the firm’s subsidy payment, our discussion of astroturf in the introduction clearly indicated that policy makers are aware of the strategy. Thus, we assume that the DM can expend some resources in auditing the SIG’s actions in an attempt to determine whether a subsidy did in fact occur. Let ? denote the probability with which the DM conducts an audit, and ? denote the probability an audit, if conducted, generates conclusive information about whether a subsidy was conferred. The cost of maintaining an audit function is c(?), where c0 (?) > 0 and c00 (?) > 0. There are two possible types of equilibria, one in which astroturf does not occur and one in which it does.
4.1
No Astroturf
In this case, the DM knows that if the SIG appears the state is ?L , and if the SIG does not appear the state is ?H . To ensure these conditions hold however the DM must audit the SIG when it appears, in order to eliminate firm incentives for astroturf. Assuming the firm does not engage in astroturf, the DM can infer correctly the state of the world, and sets the optimal policy for each state. Let the DM’s equilibrium audit probability in this case be ?NA . Thus, the DM’s expected payoff is E (U ) = ?c(?NA )/2, since setting the correct policy generates an optimal utility of zero in both states. Conditional on the DM’s audit policy, the firm must prefer not to astroturf in state ?H . The firm’s profits if it does not astroturf are ?NA = ???2 H . If it were to astroturf, expected
13
Note that a policy of astroturfing a positive-biased SIG is not rational. The positive-biased SIG will always claim the state is high, and hence the firm does not want to subsidize it to appear before the DM and support efforts to make policy more stringent.
11
profits would be
NA ? = ?NA ? (???2 ? )(???2 (10) H ) + (1 ? ? L) ? l 2 NA NA ? )(???2 So a non-astroturf equilibrium requires ???H > ? ? (???2 H )+(1?? L )? 2 l. This can be rewritten as (1 ? ?NA ? )? (?2 H ? ?L ) ? l < 0. Thus, the DM must choose ?NA to make this inequality hold. This implies
l 1 ? (11) 2 . ? ? ? (? 2 H ? ?L) The DM must engage in this level of auditing to support the no-astroturf equilibrium.14 Note that the DM wants to minimize auditing expenses, consistent with preventing astroturf, so the firm’s profits in state ? H are simply ?NA = ?? 2 H. ?NA >
4.2
Astroturf Equilibrium
In an astroturf equilibrium the firm will find it profitable to subsidize the SIG in the high state even though it knows there is a possibility that its subsidy will be detected. In this case the SIG will report ?L to the DM in both states of the world. When the SIG appears, the DM audits with probability ?A . If the audit is conclusive, the DM knows for certain whether or not a subsidy was given. If the audit is not conclusive, then the DM sets the optimal policy given its uncertainty about the state, i.e., it sets p = (?H + ?L )/2. In the astroturf equilibrium the DM knows that the SIG is subsidized in the high state. If the audit is inconclusive then the presence of the SIG conveys no information and it is optimal to set the average policy. The DM’s expected utility is ?H + ? L 1 A [? ? (?0) + (1 ? ?A ? )(?( ? ?L )2 ] 2 2 ? H + ?L 1 ? ? H )2 ] ? c(?A ) + [?A ? (?0) + (1 ? ?A ? )(?( 2 2 (? H ? ? L )2 ? c(?A ) = ?(1 ? ?A ? ) 2 Optimizing with respect to ? yields E (U ) =
(12)
? ? E (U ) = (?H ? ? L )2 ? c0 (?A ) = 0. (13) ?? 2 Given our assumptions about c (?), it is easily checked that expected utility is concave in ?. Next we must check whether astroturf is profitable for the firm. If it chooses to astroturf, and ? = ?H , then
A ? A = ?A ? (???2 H ) + (1 ? ? ? )(?? (
14
?H + ?L 2 ) ) ? l. 2
< ?.
(14)
Since ?N A ? (0, 1) condition (11) requires that 0 <
l ? (? 2 ?? 2 ) H L
12
Alternatively, if it were not to engage in astroturf, and the state were ? = ?H , then the SIG would not appear and the DM would be able to infer the state. In this case, the firm 2 A would earn ? = ??? 2 H . Thus, for astroturf to be profitable, we require ? > ???H . That is
A ?A ? (???2 H ) + (1 ? ? ? )(?? ( which implies
?H + ?L 2 ) ) ? l + ??2 H > 0, 2
(15)
1 ? ?A ? ? (3? H + ?L )(?H ? ? L ) > l. (16) 4 This inequality must be consistent with conditions (8) and (9). The relevant condition here is (8). Conditions (16) and (8) require that 1 ? ?A ? ? (3?H + ? L )(?H ? ?L ) ? ?(?H ? ? L )(2? + (?H ? ?L )) 4 which can be rewritten as 4 (2? + (? H ? ?L )) . ??? (1 ? ?A ? ) (3?H + ?L ) Thus, we obtain the following lemma. (17)
(18)
Lemma 5 For sufficiently large ? the firm finds it profitable to engage in the astroturf strategy. Lemma 5 states that if the firm finds stringent policies sufficiently costs it will find it profitable to engage in the astroturf strategy, i.e., subsidize the SIG in the high state of the world even when it faces a positive probability of detection. In doing so the firm benefits from obtaining an average policy in the high state, although it does sacrifice the possibility of obtaining ?L when the state of the world is low. Finally, to determine whether an astroturf equilibrium will exist, we need to check whether the DM would prefer to deter astroturf and shift to the no-astroturf equilibrium. In the astroturf equilibrium, the firm engages in astroturf when ? = ?H , and the DM knows this. In response, the DM audits, but not frequently enough to make astroturf unprofitable. ?L )2 ? c(?A ). Recall that E (U NA ) = ?c(?NA )/2, while E (U A ) = ?(1 ? ?A ? ) (?H ? 2 Thus, for an astroturf equilibrium we require (? H ? ? L )2 ? c(?A ) > ?c(?NA )/2 (19) 2 The only way the astroturf equilibrium can be preferred is if ?NA is very high and/or the cost function is highly convex. Thus, we have the following proposition ? (1 ? ?A ? ) Proposition 6 An astroturf equilibrium exists when condition (19) and lemma 5 hold.
2 1 ? ? ?(?2l ??2 ) = [? (? 2 Recall that ?NA > ? H ? ?L ) ? l]/? . Thus, the size of ? is critical H L to determining which equilibrium can be supported. When ? is small, it becomes difficult for the DM to deter astroturf and ?NA becomes large. At the same time, in the astroturf
13
equilibrium, the marginal value of auditing declines so the DM audits less frequently. This increases the loss due to using a policy that doesn’t match the true state of the world, though it does decrease the DM’s expenditures on auditing. If c(?) is highly convex, then the increased costs of deterring astroturf will dominate, and the DM will be more likely to allow an astroturf equilibrium when ? is small.
5.
Self-Regulation
As noted in the previous section, astroturf is only possible with a SIG that has a negative bias. Thus, if the SIG which the firm wishes to influence has a positive bias, then a different strategy is necessary. While the bearhug can be used with either type of SIG, firms may not always be able to credibly commit to the use of this strategy. In this section, we study the possibility that the firm may be able to reduce the severity of the high state, i.e. to reduce ?H , through voluntary improvements made ex ante. This might be done, for example, through design measures for a new facility that reduce the impact of worst-case outcomes. Such actions, especially if they involve sunk costs, will readily be regarded as credible commitments. Our basic notion here is that if the difference between the high and low states is sufficiently small, then the SIG will have little motivation to show up before the DM. Hence, self-regulation by the firm may induce the SIG to eschew lobbying. What benefit does the firm obtain from doing this? As shown in Proposition 1, if the SIG does not appear and the DM must set policy based on uncertainty about the true state, the firm stands to benefit by an amount ? = ? (?H ? ?L )2 /4. Let us introduce the notation a = ?H ? ?L , and denote by a0 the initial gap between the states. We consider self-regulation as a voluntary reduction in a on the part of the firm, cutting it from a0 to a1 . Thus, if the firm reduces ?H from ? L + a0 to ?L + a1 , and this induces the SIG to eschew lobbying, then the DM sets policy at the expected level of the state of the world, i.e. at p = ?L + a1 /2, and the firm’s payoff is F SR = 2 ?? (?L + a1 /2)2 = ?? (?2 L + a1 ?L + a1 /4). If the firm took no action, and the SIG revealed the true state through its lobbying decisions, then the firm’s expected payoff would 2 2 2 be F 0 = ?? (?2 H + ?L )/2 = ?? (?L + a0 ?L + a0 /2). The net benefit to the firm is V SR 0 2 2 ? F = ? (a0 ? a1 )?L + ? (2a0 ? a1 )/4 > 0. ? (a) = F Recall that the payoff function for the firm is F = ?? p2 . How should we represent the cost of achieving a? If the firm were forced to comply with a policy of p = ?H , the cost difference between a0 to a1 is k(a1 ) = ?? (?L + a1 )2 ? (?? (?L + a0 )2 ) = ? (2?L + a0 + a1 )(a0 ? a1 ). In order to be consistent with this payoff function, we will assume that the cost to the firm of achieving such a reduction is k(a1 ) = ? (2?L + a0 + a1 )(a0 ? a1 ). Combining the benefits and the cost of voluntary action gives a net payoff to voluntary action of V (a1 ) = ?V (a1 ) ? k(a1 ) 2 = ? (a0 ? a1 )? L + ? (2a2 0 ? a1 )/4 ? ? (2?L + a0 + a1 )(a0 ? a1 ) 14
Differentiating with respect to a1 yields ? V/? a1 = ??L + 3a1 ? /2 > 0, which shows that the firm prefers the largest a1 (smallest amount of self-regulation) that is sufficient to cause the SIG to drop out. Setting V (a1 ) = 0 and solving, we find that the maximum a that q ¡ 2 ¢ 2? 2? the firm would be willing to undertake is a1 = ? 3 ?L + 3 ?L + 3?L a0 + 3a0 2 /2 ? 0. Note that the maximum amount of self-regulation the firm is willing to undertake is increasing in the initial ‘‘gap,’’ a0 , since ? a1 /? a0 = How will a voluntary action affect the decisions of the SIG? Recall that conditions (2) and (3) define when the SIG will find it worthwhile to appear before the DM. If l ? l = a(2? + a) then the SIG finds it worthwhile to show up when the state is ?H , while if l ?l= a(2? ? a) then the SIG does not show up when the state is ? L . Clearly self-regulation shifts l and l, and in the process may cause the SIG to change its lobbying behavior. This is perhaps most easily seen by reference to Figure 1, in which l and l divide the (a, l) space into three regions: 1) The region with l > l, in which the SIG never lobbies, 2) the region with l l > l, in which the SIG lobbies only if ? = ?H . Of course, only in the third region is lobbying activity actually informative to the DM. Figure 3 builds on Figure 1 but adds two shaded regions. In each of these regions, the initial point (a0 , l) is within horizontal distance a1 of either l or l, so selfregulation can profitably result in a westward move that causes the SIG to change behavior. In the shaded region close to l, self-regulation causes the SIG to abandon lobbying. In the shaded region close to l, self-regulation has the opposite effect: it induces the SIG to lobby in both states of the world. In either case, however, the SIG’s lobbying behavior becomes uninformative for the DM, which is profitable for the firm. We summarize these results in the following proposition. Proposition 7 There exist two regions in (a, l) space for which corporate self-regulation can profitably alter interest group lobbying behavior: a) if a0 (2? + a0 ) > l > a0 (2? ? a0 ) and l > (a0 ? a1 )(2? + a0 ? a1 ), then self-regulation induces the interest group to eschew lobbying, and b) if a0 (2? ? a0 ) < l < (a0 ? a1 )(2? ? a0 + a1 ), then self-regulation induces the interest group to become a pure advocate that lobbies regardless of the actual state of the world. In either case, the interest group’ s lobbying behavior becomes uninformative and profits rise. Proof. Begin with case (a). We require two conditions. First, a0 (2? +a0 ) > l > a0 (2? ?a0 ) s lobbying behavior is informative; ensures that the initial pair (a0 , l) is such that the SIG’ that is, it ensures that l ? (l, l). Second, l > (a0 ? a1 )(2? + a0 ? a1 ) ensures that after self-regulation, the SIG is in the region in which lobbying is never worthwhile; that is, after self-regulation, we have l > l. Now turn to case (b). The condition a0 (2? ? a0 ) < l ensures that the SIG does not always lobby at the initial pair (a0 l). The second condition l < (a0 ? a1 )(2? ? a0 + a1 ) ensures that after self-regulation, we have l a/2) that the SIG has an incentive to report falsely when ? = ?L . If it can thereby induce the DM to set a policy of p = ?H , it obtains payoff ?(a ? ? )2 ? l rather than the payoff of ?? 2 it would receive for truthful reporting that yields a policy of p = ?L . Indeed, if a = ?H ? ?L = ? , then the SIG’s most preferred situation is to report falsely when ? = ? L and thereby obtain policy p = ? L + ? . Note that for a > ? , the SIG’s payoff from false reporting rises when a is reduced closer to ? . This is because larger levels of a mean that a policy of p = ?H is really higher than the SIG would like if the state is really ? = ?L ; as a result, the incentive to lie is reduced. Self-regulation thus makes false reporting more advantageous for the SIG. It is also interesting to ask how the decisionmaker is affected by self regulation that renders lobbying uninformative. This is the subject of the following proposition.
Proposition 8 The public decisionmaker is worse off when the firm engages in self-regulation. Proof. Without self regulation, the DM’s expected payoff is simply G0 = 0, since policy can be tailored precisely to the state of the world ex post. With self regulation, the DM lacks information about the state and must set an ‘‘average’’ policy based on his beliefs. Suppose that the firm’s self-regulatory actions are observed by the DM. Then the DM sets p = ?L + a1 /2, which is equal to the expected value of the state. Now the DM’s expected payoff becomes Clearly GSR < G0 , and the DM is worse off as a result of the firm’s self-regulatory action. In this model, the DM is best able to maximize his objective function when he has full information about the state of the world. Then he can tailor policy to the specifics of the situation before him. Self-regulation is only undertaken by the firm if it will render the SIG’s lobbying uninformative. This deprives the DM of the information he desires, and as a result the DM is worse off. This result contrasts with that in earlier work, such as that of Lyon and Maxwell (2002), who show that the regulator benefits when industry selfregulation preempts the imposition of new regulations. The key difference is that Lyon and Maxwell (2002) study a model in which self-regulation does not affect the information flow to the regulator. It is worth noting that once lobbying is rendered uninformative, the DM does benefit from further corporate self-regulation, but that this never entirely compensates for the loss of information caused by the decision to self-regulate. 16 GSR = ?.5(?L + a1 /2 ? ?L )2 ? .5(?L + a1 /2 ? ?H )2 = ?a2 1 /2 < 0.
6.
Multiple Interest Groups
To this point, we have concentrated on the case of a single firm and a single interest group. In this section, we discuss how our results may be extended to cases with multiple interest groups. We follow the typology used by Grossman and Helpman (2001) to classify the structure of multiple SIG situations: 1) ‘‘Like bias’’ arises when all groups share the same direction of bias, but with different intensity; 2) ‘‘Opposite bias’’ arises when different groups are biased in opposite directions, and 3) ‘‘Unknown bias’’ arises when the groups receive imperfect signals regarding the state of the world. The first two cases, in contrast to the third, assume that both SIGs have perfect information regarding the state of the world. We consider these in turn, focusing on the case of two SIGs for simplicity.
6.1
Like Bias
We will label the two SIGs ‘‘radical’’ and ‘‘moderate,’’ with the former possessing a larger value of ? . The more radical group may be inclined to ‘‘babble,’’ i.e. to always claim that the state is high. This may be true even if lobbying is costly, if the radical group’s ? is high enough. The moderate group is more likely to reveal the true state. For example, it may only show up to lobby when the state is really high. If the firm prefers a policy set at the average level, then it prefers to mute the moderate group, since the radical group lacks credibility anyway. This can be accomplished by bearhugging the moderate group ex ante, if that group’s bias is great enough that it will always claim the state is high when lobbying is costless. (Alternatively, similar results can be achieved through astroturfing ex post, if the group has a negative bias.) Thus, this case differs little from the single SIG case analyzed above. Alternatively, if the radical SIG’s bias is not too great, then the DM could also rely on it to provide reliable information through costly lobbying.15 In this case, bearhugging (or astroturfing) the moderate SIG will not be sufficient to affect the DM’s decision. Instead, the firm must subsidize both SIGs. Again, however, this case differs only trivially from the case of a single SIG. A more substantial change occurs if the moderate group has a small bias. Now it may tell the truth in a world of costless lobbying. In this case, the firm doesn’t want to subsidize the moderate group. Instead, it prefers to raise the SIG’s costs of lobbying, perhaps by sponsoring scientific studies that call the extent of the problem into question. This may force the SIG to spend more on scientific work to establish credibly that there really are significant impacts from the issue. In this case, the moderate group might choose not to incur the cost of lobbying. Once again, though, this is a result that could also be found in the case of a single SIG. Overall, we conclude that the addition of a second SIG with like bias to that of the first SIG is unlikely to generate much additional insight. However, it is worth noting that if all groups must be subsidized, then the cost of any kind of subsidy strategy rises linearly with the number of SIGs. This is not true of the self-regulation strategy, however. A single
15
Krishna and Morgan (2001) show that in this case the DM rationally relies on only the more moderate of the SIGs to provide information.
17
voluntary improvement affects all SIGs at once. If the firm undertakes enough voluntary action to preempt the involvement of the most extreme group, then all other groups will be preempted as well. Thus, we hypothesize that self-regulation is likely to outperform subsidy strategies as the number of SIGs grows.
6.2
Opposite Bias
When the two SIGs are biased in opposite directions, matters become more interesting. At least two types of equilibria are possible: 1) DM ignores one SIG and simply relies on the other, and 2) Each SIG shows up in one state of the world. In particular, an equilibrium may exist in which the SIG with positive bias shows up in the high state, while the SIG with negative bias shows up in the low state.16 Recall from our earlier analysis that the bearhug can be applied to groups with either type of bias, but requires commitment ability and must be undertaken before the firm learns the state of the world. Astroturf does not require commitment ability, and can be undertaken ex post, but it can only be employed with groups having a negative bias. Self-regulation is undertaken ex ante, and will influence both types at once. Case 1 is similar to the case of like bias. If the firm successfully bearhugs the ‘‘active’’ SIG, then the inactive SIG may find it worthwhile to lobby, and the DM will find it worthwhile to pay attention to him. Thus, the firm may need to bearhug both of the SIGs. Alternatively, the firm may use self-regulation to influence both SIGs at once. If the SIG with positive bias is moderate while the SIG with negative bias is radical, then the firm may decide to use both self-regulation and astroturf. A moderate amount of self-regulation will be enough to preempt the lobbying activity of the moderate group, and astroturf could then be used to induce the radical SIG to become a pure advocate, thereby eliminating the informativeness of that group’s activity.17 Case 2 is not very different. If one group is silenced, then the initial equilibrium is destroyed. However, there is an alternative equilibrium in which the DM pays attention to only one of the SIGs, and this becomes the only equilibrium if one SIG is bearhugged. Hence, the firm must again silence both groups, either through bearhugs or self-regulation, if it wishes to be successful. The general point seems to be that the basic structure of our analysis remains valid in the presence of multiple SIGs, as long as those SIGs all possess full information regarding the state of the world. The main change from adding multiple groups is that self-regulation may become relatively more attractive as the number of SIGs rises.
6.3
Unknown Bias
Again, one group is assumed to be radical, and willing to lobby in both states of the world. However, the DM is assumed to be unable to distinguish one group from the other, hence can only make use of information regarding the number of firms that appear. Lohmann
16 17
See Grossman and Helpman (2001), chapter 5, for details. Note that self-regulation reduces the radical SIG’s incentive to lobby, and thus raises the firm’s cost of astroturfing. The two policies may thus be more in the nature of substitutes than of complements.
18
(1993) analyzes this setting in the context of N > 2 groups, but Grossman and Helpman (2001) show that her main insights can be derived in a model with just two groups. Consider the case of two groups with like biases. Lohmann emphasizes the case in which the more radical SIG always lobbies, regardless of the state. The DM does not know which group is the more biased, but can still use the extent of lobbying as a noisy signal regarding the state of the world. For example, the DM may conclude that the state is high if two SIGs lobby, and low if only one does.18 If the firm can identify the more moderate SIG, then it can subsidize the moderate group, just as in the case of known bias. If this is not possible, then the firm must subsidize both groups. Now consider the case of opposite bias. Suppose that the SIG with positive bias is the more radical one, and it plays the role of a pure advocate, that is, it always shows up to claim that the state is high. The more moderate SIG only lbbies when the state is low. Thus, the appearance of 1 SIG indicates that the state is high, while the appearance of 2 groups indicates the state is low, and the DM sets a low level of policy when both groups show up, but a high level when only one group appears. Once again, if the firm subsidizes the moderate group, then that group will always appear, and the DM must set policy without gaining any information from the SIGs.19 And if the firm cannot determine which group is which, then it must subsidize both. The case of unknown biases is more subtle than the first two cases we discussed, since the SIGs don’t know the state of the world for certain. Sometimes they will be wrong, and the DM must take this into account. Nevertheless, for our purposes, most of the qualitative features of the models seem basically the same. One new possibility may emerge in the case of N > 2 SIGs with imperfect information. If the groups move sequentially in presenting their information, the possibility of information cascades arises, as in Bikhchandani, Hirshleifer, and Welch (1992). In such a setting, there is a large premium to being the first SIG to lobby, since all the subsequent SIGs may be influenced by the actions of the first. There is also a large premium to the firm if it can influence the information revealed by the first SIG to lobby.
7.
Conclusions
In this paper, we have developed a model to explore how firms may influence the lobbying behavior of special interest groups. We built on the framework presented by Grossman and Helpman (2001), in which costly lobbying may convey information to a public decisionmaker. The basic idea is that when lobbying is costly, an interest group’s decision to lobby provides credible information about the strength of its preferences regarding a particular policy issue.
A failure to lobby by both firms is off-equilibrium path behavior. Grossman and Helpman (2001, p. 154) identify beliefs for the DM under which it infers the state is low when neither firm lobbies. 19 A question G&H don’t address is what the group says once it arrives. Presumably this is just a minor extension of the costless lobbying case we examined above, except that now the SIG is working off of some uncertainty regarding which is the true state. But it will still be the case that if the SIG would always claim ‘‘high’’ if lobbying were costless, then the bear hug will work.
18
19
We considered three corporate non-market strategies: 1) the ‘‘bearhug,’’ in which the firm subsidizes the lobbying activities of an interest group, 2) ‘‘astroturf,’’ in which the firm subsidizes the lobbying activities of a group with similar views, and 3) corporate social responsibility, in which the firm voluntarily limits the potential social harms from its activities. All three of these strategies can be used to reduce the informativeness of lobbying, which can be profitable for the firm if the costs of a complying with a stringent public policy are high. When compliance costs are convex, the firm gains if the public decisionmaker sets policy based on expected or average social harm, rather than face the risk that policy will be tailored to actual harm. The ‘‘bearhug’’ serves as a signal-jamming device that prevents the interest group from signalling the intensity of its views. One might expect that the group would be unwilling to accept a subsidy that reduces the credibility of the group’s statements. Nevertheless, we show that if lobbying is costly enough then it is optimal for the group to accept the firm’s embrace. It is important to note that this strategy may not be dynamically consistent for the firm: even though it raises expected profits ex ante, it is unprofitable ex post in some states of the world. Hence, the strategy is only feasible if the firm can commit to it when the firm does not yet know the true state of the world. This is particularly likely in situations of true scientific uncertainty, such as currently exists regarding the future impacts of global warming. In some situations, the firm is likely to know the true state of the world, especially if that state depends on characteristics of the firm’s technology or management processes. For example, the state of the world might be the level of health risk associated with the operation of a particular plant, which depends upon corporate decisions regarding technology and management. In such settings, the ‘‘bearhug’’ strategy cannot be implemented. However, ‘‘astroturf ’’ lobbying can be utilized by the firm when it already knows the state of the world. Astroturfing involves subsidy only in states of the world where the interest group would not otherwise lobby. We show that such subsidies will only be jointly incentive compatible for the firm and the interest group if the group’s political bias is toward relatively weak policies. For example, the group might represent local business organizations that stand to benefit if the firm builds a new plant in the area. We show that the decisionmaker has incentives to audit the relationship between the firm and the interest group for signs of astroturfing, but that in equilibrium astroturfing takes place nevertheless. The third strategy we study is self-regulation, namely, voluntary actions to reduce possibility of serious social harms. Such voluntary actions can change the lobbying incentives of interest groups, and may render them uninformative, which is profitable for the firm.20 Selfregulation has subtle effects in this model. The most intuitive effect is that self-regulation can preempt interest group lobbying, by reducing the payoff from lobbying relative to its costs. Another possibility, one that is less intuitive, is that self-regulation can strengthen the incentives of an positive-biased interest group to falsely report that the state is high when it is really low. An interest group with positive bias wants a policy greater than that justified by the true (low) state of the world, but it may not want the policy to be fully as stringent
20
Baron (2001) refers to such actions as ‘‘strategic’’ corporate social responsibility since it is motivated by profit-maximization rather than altruistic preferences.
20
as would be justified in the high state of the world. By bringing the high state closer to the low state, self-regulation makes it less costly for the interest group to endure the stringent policy, and makes it more attractive for the group to engage in lobbying in both states of the world. Under all three of these strategies, the public decisionmaker is made worse off. The key reason is that when the decisionmaker is fully informed, he can tailor policy precisely ex post to the particular state of the world. All three of the strategies we study here are designed to stem the flow of information, and while this increases profits it simultaneously renders the decisionmaker worse off. Further research is required to flesh out the full range of corporate non-market strategies for dealing with outside stakeholder groups. We have focused on the interactions between a single firm and a single interest group, but it would be interesting to explore further what happens when multiple interest groups are involved. Particularly promising is the setting in which interest groups are imperfectly informed about the state of the world.21 It would also be worthwhile to assess when combinations of strategies may be more profitable than any one strategy pursued singly. For example, there may be situations in which a combination of corporate social performance and astroturf lobbying is to be expected. Finally, empirical examination of these issues, perhaps through the use of case studies, would be valuable.
References
Aumman, R.J. and A. Brandenburger. 1995. ‘‘Epistemic Conditions for Nash Equilibrium,’’ . 2001. ‘‘Private Politics, Corporate Social Econometrica, 63: 1161-1180. Baron, David P Responsibility, and Integrated Strategy,’’ Journal of Economics and Management Strategy, 10: 7-46. Becker, Gary S. 1983 ‘‘A Theory of Competition among Pressure Groups for Political Influence,’’ Quarterly Journal of Economics, 98: 371-400. Bikhchandani, Sushil, David Hirshleifer, and Ivo Welch. 1992. ‘‘A Theory of Fads, Fashion, Custom, and Cultural Change as Informational Cascades,’’ Journal of Political Economy 100: 992-1026. Crocker, Keith J. and Sharon Tennyson. 1997. ‘‘Contracting with Costly State Falsification: Theory and Empirical REsults from Automobile Insurance,’’ mimeo, University of Michigan Business School. DeSimone, Livio D. and Frank Popoff. 2000. Eco-Efficiency: The Business Link to Sustainable Development. Cambridge, MA: The MIT Press. Fudenberg, Drew and Jean Tirole. 1986. ‘‘A ‘Signal-Jamming’ Theory of Predation,’’ The Rand Journal of Economcis, 17: 366-376. Greer, Marsha L. and Anna K. Harding. 1993. ‘‘The Health Impact of Hazardous Waste Sites on Minority Communities: Implications for Public Health and Environmental Health Professionals.’’ Journal of Environmental Health 55(77): 6-9. Grossman, Gene M. and Elhanan Helpman. 2001. Special Interest Politics. Cambridge,
21
If the groups move sequentially in presenting their information, the possibility of information cascades arises, as in Bikhchandani, Hirshleifer, and Welch (1992).
21
MA: The MIT Press. Khalil, Fahad. 1997. ‘‘Auditing without Commitment,’’ RAND Journal of Economics. 28: 629-640. Krishna, Vijay and John Morgan. 2001. ‘‘A Theory of Expertise,’’ Quarterly Journal of Economics, 116: 747-775. Lyon, Thomas P . and John W. Maxwell. 2002. ‘‘Self-Regulation, Taxation, and Public Voluntary Environmental Agreements,’’ Journal of Public Economics, forthcoming. Maxwell, John W., Thomas P . Lyon, and Steven C. Hackett. 2000. ‘‘Self-Regulation and Social Welfare: The Political Economy of Corporate Environmentalism,’’ Journal of Law and Economics. 43: 583-617. Olson, Mancur. 1965. The Logic of Collective Action. Cambridge, MA: Harvard University Press. Stauber, John and Sheldon Rampton. 1995. Toxic Sludge is Good for You: Lies, Damn Lies and the Public Relations Industry. Monroe, Maine: Common Courage Press. Stigler, George J. 1971. ‘‘The Economic Theory of Regulation,’’ Bell Journal of Economcs and Management Science, 2: 1-21. Taylor, D. 1992. ‘‘The Environmental Justice Movement.’’ EP A Journal 18(1): 24
22
doc_245035045.pdf