國
立 政 治 大 學
‧
N a tio na
l C h engchi U ni ve rs it y
17
Lemma 2.3: In the case where the shareholders have a relatively large political influence, such that ˆ
f f f g g
, the equilibrium emission tax rate
decreases with . In the case where the environmentalists have a relatively large political influence, such that ˆ
f , the equilibrium emission tax rate increases with
.
By combining the above results with the fact that e t 0, we can pin down the sign of (2.17). The effect of on e can be either positive or negative, depending on the relative political influence of the lobbies. This result indicates that the audit policy does have an impact on e, although it works in an indirect way.
A more interesting implication of this result is that an increase in the probability of detection can enlarge the true amount of pollution emissions, in particular when the polluters have a relatively large political influence. As indicated in Lemma 2.3, when the shareholders are dominant, an increase in leads to a lower emission tax rate, and thus a larger amount of net emissions. This result is different from the conventional wisdom and the finding in Harford (1978, 1987) and Sandmo (2002).
This also indicates the possibility that policymaking will be misguided due to overlooking the political effect.
Since the audit policy has the real effect, it also affects social welfare. It is hardly surprising that the welfare effect of the audit policy is ambiguous. Here we focus on some polar cases to obtain definite results. We first consider the case where only the shareholders lobby. In this case, as indicated in Proposition 2.2, t is lower than to. In addition, Lemma 2.3 has shown that t decreases with in this case. The negative relationship between tand implies that an increase in causes the
‧ 國
立 政 治 大 學
‧
N a tio na
l C h engchi U ni ve rs it y
18
emission tax rate to deviate further from the optimal level. by assuming the social welfare function to be convex, the above analysis reveals that tightening the audit policy reduces social welfare. As a result, in this case the second-best audit policy that accommodates the influence of lobbying is the minimum value of .
We then turn to the case where only the environmentalists lobby. We have learned two facts in this situation: (i) tis higher than to, and (ii) tincreases with
. Combining these two facts reveals that an increase in causes t to rise further above to, thereby leading to a lower level of social welfare. Accordingly, in this case the second-best audit policy also requires that the minimum level of be set.
An increase in , however, can enhance social welfare in more general cases.
For example, it is possible that t is greater than to, and that it decreases with .11 In this case, an increase in reduces t, which narrows the gap between t and
to, and thus enhances efficiency.
The following proposition summarizes what we have found above:
Proposition 2.2: (i) An increase in the detection probability either raises or reduces the equilibrium emission tax rate, depending on the relative political influence of the lobbies; (ii) the welfare effect of tightening the audit policy is ambiguous, which implies that an increase in can reduce social welfare.
2.4.2 Penalties
We adopt the penalty function in a general functional form in the previous sections. In order to examine the effect of a shift in the penalty function, we specify the penalty
11 From (15) and Lemma 3, we know that this case may occur if ˆ
f and g is much larger than ˆ.
‧
where 0 denotes the shift parameter of the penalty function.
The effect of a shift in the penalty function on the net pollution emissions is:
* shareholders is so large that ˆ
f , then an increase in reduces t*. By contrast, increases pollution emissions. This result again shows that disregarding the political effect may lead the environmental policy to generate unintentional outcomes. On the other hand, if ˆ
f , then the net pollution emissions decrease with . In this case, raising the penalty achieves the intended goal.
We then move on to the welfare implication of a change in . Apparently, the
12 The derivation of (2.21) is presented in Appendix 2.3.
‧ 國
立 政 治 大 學
‧
N a tio na
l C h engchi U ni ve rs it y
20
welfare effect of a change in the penalty is ambiguous, and again we focus on some polar cases. In the case where only the shareholders lobby, t* is lower than to, and it decreases with . These two facts together imply that an increase in reduces social welfare. In the case where only the environmentalists lobby, t* is set at an excessively high level, and an increase in raises t*. These relationships also suggest that social welfare decreases with . Nevertheless, a stiffer penalty can be welfare-enhancing in other more general cases.
Proposition 2.3: (i) An increase in the penalty either raises or reduces the equilibrium emission tax rate, depending on the relative political influence of the interest groups; (ii) the welfare effect of a stiffer penalty is ambiguous, which implies that an increase in can reduce social welfare.
2.4.3 Discussion
The above analysis highlights the possible perverse effects arising from a tightening of the enforcement policy. This finding is in line with that of Kambhu (1989) and Heyes (2002).
Kambhu (1989) investigates the effect of imperfect regulatory enforcement on the compliance of polluters, when the polluters can take action to evade penalties for non-compliance. He shows that a marginal increase in the penalty may cause compliance to fall, because a stiffer penalty increases the polluters’ incentive to intensify their challenge to enforcement.
Heyes (2002) constructs a two-stage model of regulatory enforcement. An inspection in the first stage gives rise to a noisy signal of the inspected firm’s true amount of pollution. Only if the signal exceeds the “trigger” will the inspected firm be audited. Heyes (2002) finds that a tighter trigger may cause the performance of
‧ 國
立 政 治 大 學
‧
N a tio na
l C h engchi U ni ve rs it y
21
serious violators to decline, and lead to more pollution emissions. The intuition behind this result is that a tighter trigger enlarges the gap between the trigger and the true level of emissions, and thus reduces the marginal benefit of abatement.
Although our finding is similar in spirit to the conclusions of these two papers and others, this present chapter takes the political aspect into consideration, which has been overlooked in the above literature. Moreover, these papers are positive models, but we provide some normative implications in addition to positive issues.
2.5 Concluding remarks
In this chapter, we investigate how the influence of lobbying shapes environmental policy in the presence of imperfect compliance. We also examine the effects of enforcement policy on the stringency of environmental policy. Our major finding is that tightening the enforcement policy can increase pollution emissions and reduce social welfare. This result is quite different from the finding in Becker (1968), who claims that the optimal fine should be set at the maximal level.
In order to obtain tractable results, we construct a simple setting. Nevertheless, doing so overlooks several issues, including the monitoring costs, and the interaction between the intertemporal self-reporting decision and the detection probability.
Further extension could give rise to important and interesting insights.