فرانچایزینگ (امتیاز) بزرگراه و ارزش واقعی املاک

It has become increasingly common worldwide to auction the construction and operation of new highways to the bidder that charges the lowest toll. The resulting highway franchises often entail large increases in the value of adjoining land developments. We build a model to assess the welfare implications of allowing large developers to participate in these auctions. Developers bid more aggressively than independent construction companies because lower tolls increase the value of their land holdings. Therefore developer participation unambiguously increases welfare, yet this increase is not necessarily monotonic in the number of developers participating. Welfare also increases when large developers can bid jointly.

Highways leading to new land developments have been traditionally financed with gen- eral public funds. In principle, the outlays can be recovered by property taxation, as the benefits wrought by the new highways become capitalized in property values. Yet practical limitations of property taxation have increasingly led to private financing of roads in many countries. 1 The private solution that has become increasingly popular is to auction so-called build- operate-and-transfer (BOT) contracts to the firm bidding the lowest toll. Under such a contract, a private firm builds and finances the road and then collects tolls for a long period (usually between 10 and 30 years). 2 When the franchise ends the road is transferred to the state. In this paper we build a model to assess how the participation of large developers in the auction of the road that will increase the value of their land holdings impacts on social wel- fare. A recent example motivating the issues we consider is the Radial Nororiente project which joins Santiago, the capital of Chile, with the adjacent Chicureo Valley, which is ex- pected to expand in coming years. In 1999 the Chilean government decided to franchise a US$170MM road. 3 , 4 The idea of using open auctions instead of regulation—competition for the field as a substitute for competition in the field—goes back to Chadwick [1] and was popularized by Demsetz [2]. The claim is that competition in the auction will eliminate market power and yield a toll equal to average cost. 5 When applied in the setting considered in this paper, the Chadwick–Demsetz analysis assumes that franchise holders are firms that specialize in building roads. When land developers can participate in the auction, several issues in competition policy arise, as illustrated by the Chicureo project. 6 Some critics claimed that developers should be excluded from the auction, because their land interests gave them an “undue” competitive advantage. Others argued that developers’ common interest would encourage them to collude in the auction. Finally, it was suggested that developers would price discriminate, charging low tolls to buyers of their land and high tolls to other users. The basic economics of developer participation is best appreciated in the case where a developer is allowed to build the highway and set its toll at will (the laissez-faire case). Since tolls that are higher than marginal cost reduce road usage below the socially optimal level, they create a distortion which, in a competitive land market, reduces the value of land one-by-one. This fact creates the basic tradeoff facing a developer that owns the road and is free to choose the toll: on the one hand, she would like to charge a toll equal to marginal cost to those who buy her land; on the other hand, she would like to charge the monopoly toll to the remaining users. Consequently, her profit-maximizing toll will lie between the monopoly toll (if she owns no land) and a zero toll (if she owns all land and there are no through drivers). 7 Therefore, a developer who owns sufficient land would set tolls below average cost, as the losses she makes in the highway business are more than made up by the higher land prices. This is welfare-increasing, because tolls are closer to our assumption of zero marginal cost in the absence of operation and maintenance costs. Consider now an auction. Competition forces independent construction companies to bid a toll equal to average cost. If only one large developer participates in the auction, he will bid more aggressively, choosing his laissez faire toll. As a result, developer participa- tion yields tolls closer to zero and higher overall welfare. When two large developers participate in the auction, it often happens that each devel- oper is better off when the other one builds the road and charges his laissez-faire toll. We show formally that the auction outcome has two Nash equilibria when land holdings by both developers are not too dissimilar. The larger developer builds the road in one equi- librium, the smaller developer in the other one. In both cases, the franchise holder charges his laissez-faire toll. It follows that having two large developers participate in the auction may lead to higher tolls, and lower welfare, than when only one (the larger) developer par- ticipates. Developer participation unambiguously increases welfare, yet welfare does not necessarily increase monotonically with the number of developers. Another interesting result obtains once we allow developers to bid jointly at no cost. Tolls are then reduced and welfare increases, because landholders maximize joint profits by asking for the same toll as a single developer who owns their combined land shares. Our paper is related to the literature on franchise bidding pioneered by Chadwick [1] and Demsetz [2] (see also Stigler [21], Posner [17], Riordan and Sappington [18], Chapter 9 in Spulber [20], Chapters 7 and 8 in Laffont and Tirole [15], Harstad and Crew [11] and Engel et al. [4,6]). We add to this literature by exploring the consequences of including bidders whose downstream profits increase with lower tolls. Somewhat less related is the limited literature on auctions of objects with externalities. For example, Jehiel et al. [13,14] solve a mechanism design problem in which the auctioned object causes an identity-dependent externality on bidders that loose in the auction. We differ from them in that we model the origin of the externality—the winning toll affects the welfare of all landowners—which allows us to assess the welfare impact of alternative policies. The rest of the paper is organized as follows. Section 2 describes the model. Section 3 analyzes the consequences of letting one landowner build the road and giving him a free hand to choose tolls (‘laissez faire’). Section 4 examines the bidding behavior and outcome of the auction. Section 5 explores some extensions and applications. Section 6 concludes. Two appendices with formal proofs follow.

Highways, and more generally infrastructure projects, change the value of land, because their benefits are capitalized into its price. This paper has examined the strategies of large real estate developers and how these strategies affect social welfare. Our results depend on the fraction of the land that is owned by the largest landowners, and on the possibility of discrimination between different users of the road. Since it is always in the interest of the real estate developer to charge a zero toll on buyers of her land, she always prefers to discriminate in tolls. If she is not allowed to discriminate, and this rule is enforceable we show that welfare is maximized when large landowners are allowed to combine and bid jointly (i.e. they are allowed to collude), and that competition may lower welfare. On the other hand, when discrimination is possible, competition among small landowners leads to higher welfare than under uniform tolls. In the light of this analysis, it is interesting to examine the aftermath of the auction for the road to Chicureo, which we described in the Introduction. As predicted by our model, the largest landowners formed a group to present a single bid. 22 In the end, however, no one showed up for the auction. The landowners complained that contingent subsidies against losses in the highway project were too small, making the franchise unprofitable. Our analysis, however, suggests that profitability of the highway itself is not a true measure of the overall private value of the project for large developers. There are two possible explanations. Since large landowners internalize most of the social benefits of the highway, building the highway might not have been socially desirable. If this were the case, the fact that there was no participation was welcome. More plausibly however, the large landowners were lobbying for a larger government handout, and were willing to wait given the then low prices for real estate, due to an economic slowdown at the time which made waiting costless. A subsidy in this case would be a pure wealth transfer with no allocation effects. 23 Finally it is worth considering whether the issues considered in this paper are quantita- tively relevant. Consider the case in which there are 6000 plots of land, and assume that families make three trips a day in the equilibrium and that the cost of the road is $170 million. 24 If there is no toll discrimination, no collusion and there are no other users for the road, small landowners would have to finance the road out of tolls, which implies a toll of $2.59. If we assume linear demand, we can calculate the benefits from having a single landowner as compared to dispersed landowners, by measuring the effect of reducing tolls to zero. The increase in welfare depends on the toll at which plot owners stop using the road (i.e., the vertical intercept). It varies between 29 and 91% of the construction cost when the intercept varies between $7 and $4.