Is there a tax policy that can address simultaneously the following issues in a market: raise enough tax revenues; create few distortions; circumvent the excess entry of firms in the long-run; be voluntary in the sense that the firms may choose how to be taxed?
The goal of the current paper is to propose such a policy mechanism. The mechanism combines an excise with a tax-exemption auction: The government sets a tax rate τ per unit of firms' production, and also announces the auctioning of a number of tax exemptions. Namely, it invites the firms to bid in order to acquire, if they wish, the right to be excluded from paying τ. The firms that place the highest bids win this right and pay the government their bids (we assume a first-price auction); and all other firms remain subject to τ.
Via this mechanism each firm in the market essentially decides whether to transform a part of its marginal cost (i.e., the per unit tax) into a fixed cost (i.e., its bid in the tax exemption auction). If it does so, it may obtain a cost-efficiency advantage over its competitors in the product market. In this sense, the government acts as a "patent holder" who sells licenses to the use of a marginal cost-reducing "innovation" (the innovation being the avoidance of the per unit tax).
To fully capture the merits of the mechanism, which require competition among firms, we introduce it for markets operating as oligopolies, where firms' interactions are most prevalent. To boost competition for the exemptions, we assume that the government auctions-off a limited number of them. We show that the suggested mechanism has the desirable features spelled out earlier. Namely:
It can create more revenues for the government in comparison to the excise tax policy for any policy rate τ. This is due to the competition among the firms to acquire the exemptions and avoid competing with a marginal cost disadvantage. Among other things, this implies that if the government wants to collect revenues of magnitude, say, R(τ), it can do so by announcing a tax rate τ'<τ and by inviting the firms to bid for tax exemptions.
Setting lower tax rates, and also having some firms being tax-exempted, reduces distortions in the market and thus raises welfare. Moreover, announcing low(er) tax rates is often more attractive from a political view point.
In markets with free entry of firms, the mechanism induces lower equilibrium entry compared to the excise tax policy, thus correcting for the excess entry of firms often observed in imperfectly competitive markets.
The suggested tax scheme takes place within a voluntary mechanism, namely the firms choose whether to participate in the auction or not and hence choose how to be taxed.
Notice that there is an interesting relation of the current paper with the literature on tax buyouts. A tax buyout is a fiscal instrument according to which a citizen can choose to pay a fixed amount to the government in exchange for a reduction in their marginal tax rate (Del Negro et al. 2010; Goerke 2015). As the latter tax rate is distortionary with regards to labor supply decisions in the labor market, the tax buyout policy increases social welfare, without compromising on revenues, just as our mechanism does for the case of firms in product markets. The two mechanisms, therefore, complement one another.
Some comments on the applicability or the real-world relevance of the mechanism are in order. First, our paper does not suggest that a tax-exemption auction mechanism should be built for all markets (that fit our theoretical framework). This would be practically impossible. What we suggest is that a government may choose a small number of markets, that are well organized, with financially sound firms, etc, and therein run its auctions.
Secondly, in order to highlight the relevance of the mechanism for the real world we note that auctions are already used in a related policy issue in the field of international trade: the auctioning of import licenses. Via this policy the government of the country that imposes quotas on trade can uncover a part of the rents associated with the restriction. The mechanism has been used by countries like Australia, New Zealand, etc., (see Tan 2001) and its success depends on the structure of product markets (see Krishna 1988, 1991, 1993; Tan 2001). So, if an auction-backed government policy can work in international markets then a related mechanism could perhaps work in the current paper's context.
Thirdly, we refer to another real-world policy mechanism with a somehow similar structure: the auctioning of pollution permits. Through this mechanism firms purchase pollution permits which allow them to emit pollutants up to a certain degree. The auctions are designed so as to promote market efficiency and also tax revenue collection (see, for example, Cramton and Kerr 2002), i.e., goals our mechanism also sets. The success of the emission permit auctions hints that our mechanism could also work successfully in practice.
The paper connects to the literature on taxation in oligopoly. The roots of this literature go back to the mid 80's when economists started analyzing the design and optimality of various tax policies outside the two antipodean cases of perfect competition and monopoly. We refer the reader to Dierickx et al. (1988) for a review of the early literature on taxation under imperfect competition. The main issues examined in the literature include tax incidence, comparison between ad valorem and excise taxes, the comparison between ad valorem and excise taxation under uncertainty and/or asymmetric information, taxation in oligopolies under general equilibrium, taxation in vertical oligopolistic markets, etc. Not all aspects related to taxation in oligopoly are, of course, examined in the current paper. For example, the paper only focuses on excise and not on ad valorem taxation. Further, commodities in our model are not "sin" goods, so a potential increase in their consumption, due to our mechanism, is not a concern. Despite these restrictions though, our framework provides a first step towards introducing the tax mechanism and pointing out its merits.
