قیمت در مقایسه با مقادیر: دارایی های عمومی و انتخاب ابزار نظارتی

Since Montgomery [1972. Markets in licenses and efficient pollution control programs. Journal of Economic Theory 5, 395–418] and Weitzman [1974. Prices vs. quantities. Review of Economic Studies 41, 477–491], it has generally been assumed that, in a full-information setting, tradable quotas and taxes are equivalent regulatory instruments. We demonstrate that, if the government is movitated by public-finance concerns, this equivalence holds only if quotas are short lived. Moreover, the government prefers quotas with the shortest possible lifetime, or, what is equivalent in our setting, taxes.

In this paper we revisit the problem considered by Weitzman (1974) and others of whether a certain economic activity should be regulated by a price instrument (i.e. a tax) or a quantity instrument (i.e. a quota or license).1 We demonstrate that, when regulation is by tradable quotas allocated by auction and the government is motivated by public-finance concerns, the regulatory outcome depends on the lifetime, or duration, of quotas. Moreover, the government prefers quotas with the shortest possible duration, or, what is equivalent in our setting, taxes. This is true even if information is perfect. More specifically, we demonstrate that longer-term quotas create an intertemporal ‘inconsistency’ problem, in the sense that the government would over time increase the number of quotas beyond the globally efficient level. When regulation is by taxes, no similar problem arises. Essentially, this is related to the insight that a monopolistic producer of a durable good prefers leasing to avoid the problems of intertemporal credibility associated with selling, see Coase (1972), Bulow (1982) and von der Fehr and Kühn (1994). Here, the public-finance-oriented government corresponds to the monopolist, tax regulation corresponds to leasing, while regulation by long-lived quotas corresponds to selling a durable good. Taxes and quotas are equivalent policy instruments only if quotas have short duration, which amounts to leasing. The interaction of public-finance and regulatory considerations arises in many different contexts. City planning involves weighing income-generating use of land (such as selling building plots or collecting taxes from commercial properties) against other uses (such as for recreational purposes). In allocation of spectrum, revenues from the auctioning of rights are considered along with the their content (such as duration, coverage and conditions on technology). In fisheries, an important issue has been the distribution of resource rent between public and private parties.2 Overall, there seems to be an increasing tendency to use auctions to distribute rights, e.g. in public transportation. For the sake of concreteness, we formulate our analysis in the context of environmental regulation. Public-finance concerns have been an important issue in the debate on environmental policy.3 When taxes are compared to quotas that are distributed gratis (or ‘grandfathered’)—which have often been the two alternatives considered in practice—taxes are clearly preferred from a public-finance perspective (Hoel, 1998). However, if public finance matters,4 then it is more natural to consider quotas that are auctioned rather than grandfathered.5 In this case it is not obvious that there is a difference between the two types of instrument; indeed, Montgomery (1972) showed that in a static, full-information setting taxes and auctioned quotas achieve the same outcome. The thrust of this paper is that this result does not carry over to a dynamic setting; here, under quota regulation, the outcome, and hence the performance relative to tax regulation, depends on the duration of quotas. This finding suggests that quota duration needs to be explicitly considered in policy analyses, an issue that has typically been ignored. Fundamentally, our findings stem from different public-finance properties of taxes and tradable quotas. If a tax regime is in place, a relaxation of regulation leads to a gain from a larger tax base (a quantity effect) which is, however, counteracted by the lower tax rate (a price effect). If a system of quotas is in place, a corresponding relaxation of regulation (i.e. the sale of additional quotas) results in a larger increase in revenues, since the price effect (i.e. fall in price on outstanding quotas) is born by initial quota holders.6 Conversely, tightening regulation involves a higher cost in the quota regime than in the tax regime, as the reduction in tax base is compensated for by a higher tax rate.7 Research on the choice of regulatory instruments has tended to rely on a modelling framework in which these different public-finance aspects of taxes and quotas are irrelevant. The literature following Weitzman (1974) has mostly been concerned with how the choice of instrument affects allocative efficiency in the market in question; see e.g. Stavins (1996) and Hoel (1998). The partial-equilibrium approach employed in these analyses has consequently disregarded implications of regulatory policies in other sectors of the economy; in particular, public-finance considerations, such as using proceeds from regulation to reduce distortionary taxation, have not been explicitly taken into account. Related problems have been studied in a general equilibrium setting where efficiency in a particular regulated market is balanced against effects in other sectors. In that case public-finance aspects enter naturally, for example, through the possibility of reaping a ‘double dividend’ from the collection of environmental fees through simultaneous reduction of distortionary taxes; see e.g. Bovenberg and de Mooij (1994) and Goulder et al. (1999). These models are not, however, concerned with the comparison of taxes and quotas and are typically static with full equivalence in the implementation of price and quantity controls. Implicitly, in such models, quotas have a lifetime equal to the length of the decision period and taxes are set for one period at a time. Public-finance concerns are important in the literature on optimal incentive regulation (see e.g. Laffont and Tirole, 1993). There, the source of the regulatory problem is private information and incentives that are not in line with social objectives. If public finance is not an issue the problem of regulation in some sense becomes trivial since the regulator may elicit desired behavior through use of arbitrarily large economic incentives (see e.g. Loeb and Magat, 1979). Public finance therefore has been an integral part of this theory. In the next section we introduce a multi-period version of the static set-up of Weitzman and others (see e.g. Baumol and Oates, 1988). In the following two sections, we consider the implications of the different public-finance characteristics of taxes and quotas. Regulatory policy may be adjusted in each period by setting the tax rate or by trading quotas. If quotas are long-lived firms typically hold a positive amount of quotas at the time of policy revision. The government then has incentives to relax policy—sell more quotas—in excess of what it would have done had there been no pre-existing quotas. Agents will, however, rationally expect this behavior and compensate for it by a lower valuation when quotas are sold. The consequence is a slacker policy without net revenue gains. Long-lived quotas therefore create a problem of time inconsistency: public-finance concerns lead the government to issue more quotas over time even if there are no changes in fundamental market conditions.8 If the government were able to commit to a quota policy then public-finance incentives as such would not lead to intertemporal adjustments in quota allocation. In the absence of such commitment possibilities, the time-inconsistency problem induces the government to reduce the lifetime of quotas or choose taxes, which do not involve such inconsistency problems. In Section 5, we discuss some of our modelling assumptions and the relationship between our results and the literature on durable-goods monopoly. The final section concludes.

Since Montgomery (1972) and Weitzman (1974), it has generally been assumed that, in a full-information setting, tradable quotas and taxes are equivalent regulatory instruments. We demonstrate that, unless the government is indifferent about revenues, the equivalence result holds only if the duration of quotas is no longer than one decision period. A government will choose quotas with the shortest duration possible. Longer-term quotas create an intertemporal ‘inconsistency’ problem, in the sense that the government would over time increase the number of quotas beyond the globally efficient level. When regulation is by taxes, no similar problem arises. This corresponds to the result that a monopolist would prefer to lease rather than to sell durable goods. In the literature on environmental regulation taxes and grandfathered quotas are often compared as alternatives and in that case taxes are obviously preferred from a public-finance perspective. The thrust of this paper is that even if quotas are auctioned under perfect information there remains a disadvantage which makes long-run quotas less attractive to policymakers. In our setup there are no costs related to shortening the duration of quotas. Clearly such costs may exist, for example related to the frequency of regulatory intervention. Longer-term quotas may reduce the need for costly transactions and provide better predictability about regulatory policy. Therefore, if public-finance considerations lead policy makers to shorten quota duration (or regulate by taxes) real costs are involved. This would be analogous to costs arising from planned obsolesence in durable-goods models where a producer may intentionally shorten the lifetime of goods in order to commit to higher future prices. We have cast our model in the setting of environmental regulation, but similar concerns would seem to arise in many other contexts. The crucial characteristics for the public-finance argument to arise are that the government regulates the overall level of economic activity and can choose the duration of rights to this activity. As pointed out in the Introduction such activities would include land use, exploitation of renewable natural resources and public transportation.