November 1998

The taxi problem in Dublin is characterised by a secondary market for taxi plates. As demand for taxis has progressively increased in the city, rents accruing from owning a taxi plate have increased, driving up the price of taxi plates on the secondary market. The rents associated with owning a taxi plate are competed for in a secondary market where agents compete for the right to expropriate the rents. This paper advocates that any policy to increase taxi numbers in advance of deregulation must remove the transferability of taxi licences for the policy to be time consistent. A distinction, which we refer to as vertical differentiation, introduces a principal-agent problem, moral hazard, in an environment characterised by interminably long taxi queues and by cosy drivers, who have become urban sharecroppers.

1. General Introduction

1.1 Overview

The introduction of a limit on taxi plate numbers coupled with transferability (introduced in the 1978 Road Traffic (Public Service Vehicles) (Licensing) Regulations) has created a secondary market in taxi plates. As demand for taxis has increased in the intervening period, rents that accrue from owning a taxi plate have increased accordingly, a fact that has driven up the price of taxi plates on the secondary market. The current plate price is beyond the financial means of most people who would, in a free entry regime, opt to drive the taxis. This paper argues that the transferability of taxi licences brought about by the 1978 legislation, should be removed. The authors argue that a phased path to total entry deregulation, without removing transferability, is not sustainable due to problems of time consistency.

In addition, the paper draws particular attention to the distinction that has been created by the 1978 S.I. op. cit., a distinction between taxi owners and the taxi drivers. This distinction, which we refer to as vertical differentiation, introduces a principal-agent problem into the taxi market. The paper advances the idea that this secondary market, because of prohibitive entry costs, perpetuates this distinction between the taxi plate owner (the principal) and the driver (the agent), which neither benefits the public nor the average taxi driver.

There will be opposition to any proposed change in the licensing regime. Any attempt to ameliorate that (rent-seeking) opposition raises the possibility of compensation. The analysis in the paper briefly focuses on the question of whether the payment of compensation to incumbent taxi owners is efficient or not. We argue that there is no efficiency reason to award compensation to incumbent taxi plate owners - compensation rules are generally inefficient because of moral hazard problems. If the owners were to be compensated, they would have an incentive to overvalue their investment. Simply making taxi plates non-transferable (which is not the case at present) will ensure that the fixed (sunk) cost of entry into the Dublin taxi market is lowered, thus facilitating both ease of entry and exit.

Furthermore, the economic case for non-compensation is grounded in legal reasoning viz J. Costello in Hempenstall et al v The Minister for the Environment (1992):

1.2 Background

The Minister for the Environment is empowered under the Road Traffic Acts 1961 to regulate the operation of public service vehicles (PSV’s). In terms of the carriage of a small number of persons, PSVs can be broken down into hackneys, which, under current licensing arrangements, can only be hired on a private basis by phoning in advance or by going to the hackney outlet, and taxis proper which can by hired on a private and public basis. The legal basis for the taxi-hackney distinction is to be found the Road Traffic (Public Service Vehicle) Regulations 1963 (S.I. No. 191). In recent years, hackneys have become more substitutable for taxis as the hackney owners have made efforts to make their service more like that of taxis and more identifiable. This has been resisted by taxi groups and has been the subject matter of some High Court cases, notably, Hempenstall (1992) and O’Dwyer (1998).

Although Dublin taxis do face competition from hackneys in some segments of the market, taxis and hackneys are not in direct competition with each other in terms of public pickups either at taxi ranks or cruising. The taxi plate owners/drivers have expressed their concerns about encroachment into the protected segments of the market by hackneys, as witnessed by the recent protests by the taxi drivers’ federations. However, the impact of the increase in hackney licences since the early nineties has not, interestingly enough, had any long term effect on the value of taxi plates in the secondary market for plates. From this single observation we could infer that the recent developments in supply of hackneys has had little impact on the rents accruing to the taxi plate owners.

The price paid for a taxi plate in Dublin represents the present discounted value of the expected stream of annual profits for each taxi plate. But once an owner has paid this price, these profits, arguably, will not represent an economic profit, but a repayment for the up-front investment, with the owner earning a normal rate of return on this investment. The rent, however, associated with the supply restrictions in licences, has, for the most part, been captured by the initial owners of the taxi licences, although later owners have earned rents with changes in demand or institutional changes in the supply of licences. The allocative inefficiency in taxi licence supply has created a secondary market in taxi plates wherein rents accrue.

An interesting aside to the issues raised in this paper was the attempt by hackney plate owners in the O’Dwyer case to create a secondary market in hackney plates (which existed during the moratorium on the issuing of hackney plates in 1991/1992 where hackney plates sold for IR£20,000). In this case J. Geoghegan found that;

2. Taxi Plates

There are various conditions which must be satisfied by the applicant for the taxi plate which mainly concern their suitability for taxi plate ownership. Taxi plates become transferable five years after being issued. The PSV office (located within Dublin Corporation) processes plate transfers in the same manner as it would an original application for a fee of £3,000. Under the present system, the holder nominates the person to whom the plate is transferred. At the end of 1997 there were 1,974 taxi plates and 3,000 hackney plates issued in the Dublin region. This is only marginally up from the figures in the early 1990’s as can be seen in Table 1 below:

Source: The 1991 figure comes from Should Irish Taxis Compete? by P. Massey in McNutt [ed.]: Perspectives on Competition Policy, 1994, CIEL, University of Ulster. The 1997 figure comes from an Article in the Irish Independent, 22 November 1997.

It is estimated in the Oscar Faber report that by the end of 1998 there will be 2,374 active taxis in the Dublin region. The turnover in taxi plates is estimated unofficially at about 2-3% per annum. It is generally accepted that taxi plates now trade in the region of £80,000, which gives an upper bound on the value of the total taxi plate stock of IR£158m approximately, based on 1997 taxi numbers. It is estimated that the monopoly profits arising from the present system is IR£30m (of which some IR£22m is estimated to be from the rent charged to cosies). It is important to bear in mind that these taxi plates are not, in the fullest sense, pure private property, as they will not, for example, be accepted as collateral by a bank.

This raises the issue brought up in Smythe as regards the nature of a taxi licence on asking "how is it that licences change hands outside the regulatory environment". Smythe goes further in his exposition of how the regulatory system (in the US) works:

2.1 PSV Driving Licences

PSV drivers must have a PSV licence which is issued by the Garda Commissioner. There are no quantitative controls on the number of PSV driving licences issued, although restrictions on drivers being available for 40 hours a week and not being engaged in a business or occupation which would be likely, in the opinion of the Commissioner, to impair their efficiency or otherwise be in conflict with their employment as a driver of taxi (or hackney), have recently changed. The taxi or hackney licence and licences to drive a taxi or hackney can be revoked at any time by the proper authorities if they have good cause.

The Fair Trade Commission (FTC) in its report, recommended that there be a gradual phasing out of quantitative limits on taxi plate numbers. The lifting of quantitative restrictions would not be accompanied by a change in qualitative restrictions. The FTC argued for the removal of the more restrictive regulations regarding the issuing of PSV driving licences. In particular, they pointed to the provision that the PSV driver be available for 40 hours a week. They also suggested that the Dublin taximeter area be extended to reflect the increase in population that had occurred in the previous two decades. Notwithstanding the FTC’s concern that the removal of price controls could lead to higher prices in the short term, they were of the opinion that price competition between taxis would be beneficial to consumers.

Kramer and Mellor (1995), in their study of the taxi market in Boston, recommended fundamental changes to the control imposed on Boston's taxi market. The hackney unit of Boston's police department would have to open entry into the taxi industry and regulate cabs in order to protect public safety by requiring a licence, doing a background check on drivers, a vehicle safety check and ensuring that the drivers had adequate liability insurance.

To allow market forces to work, it was recommended that the City Council should contemplate a change in rate structure along with opening entry to the industry. This would allow drivers to expand their base of customers by offering competitive fares, especially to the elderly - the fastest growing segment of society who rely heavily on taxicabs, but many of whom also have fixed incomes. In order to protect visitors to the city, as well as others who rely on taxi service and who may not be familiar with rates, a fare ceiling may be imposed to restrict any price gouging that may occur. They recommended that the price ceiling, below which competitive fares could be offered, would in all likelihood be what cabs must charge now.

Massey (1994) proffered many valid criticisms of the Inter-Departmental Review Group (IRG) which was set up in the early 1990s to examine legislation applying to the operation of small PSVs. The most pertinent of these is the fact that the IRG failed to recognise that competitive markets are superior to regulated ones in making the required adjustments to equate supply and demand. He had suggested that there be a gradual phasing out of quantitative limits on taxi plate numbers, followed (quite sensibly) by a lifting of the restrictions on price controls. He drew our attention to the case of Sweden in 1990, where price regulations were lifted at the same time as quantitative restrictions on the number of taxis were lifted, as an example of the perils of deregulating prices in a market which has quantitative restrictions on entry. In the year following de-regulation, the number of taxis increased by 26%. The gradual phasing out of quantitative restrictions would allow recent entrants to recoup their investment in taxi plates.

3. A Regulatory Framework

3.1 Taxi Plate Owners and Drivers

The holders of taxi plates in Dublin have many possible means of employing (accruing rents) the plate. Firstly, they can acquire a car, radio, insurance etc. and drive the taxi themselves. In this instance, the taxi plate owner takes the capital risk associated with the value of the plate and the risk associated with the income stream from fares. It may have been the case in the past that the holder of the taxi licence was also the driver, but in recent times this system has been undermined. This has occurred as the fixed costs of entry have increased; individuals who can now purchase a plate (and get a car on the road) will normally be people of significant means and this is correlated with a high opportunity cost of their time. If the taxi plate owner decides to drive, they face the decision as to whether or not to take on a cosy. In the Oscar Faber report, it is estimated that the increase in gross earnings from taking on a cosy is IR£8,024. Unless the taxi plate holder is very risk averse and has not recently made a large investment to acquire the plate, it makes economic sense to take on the cosy. The plate holder may decide not to drive the taxi if his opportunity cost is moderately high. In this case he can rent the plate out to two cosies (providing the vehicle, insurance etc.) or rent the plate to an intermediary.

In the American context, an article in the New York Times in 1992 recognised that taxi driving has been closed off (due to regulation) as the "poor-man’s gateway" to mainstream America. This development was noted in an article on the Boston taxicab market where it is stated:

One must focus on how the regulatory system for taxis in Dublin has increasingly turned taxi drivers into urban sharecroppers. One could argue that this is due to a bidding war for the monopoly rents that accrue to plate holders and the risk and informational difficulties associated with a plate owner-driver relationship. The regulation of taxi plate numbers in the face of increasing population and incomes creates a situation where demand exceeds supply.

The impact of regulation can be seen very starkly in the case of the city of Indianapolis before it de-regulated the taxi industry in 1994. There, a regulated taxi from the airport to the centre of the city was twice as expensive as an unregulated limousine which had a TV and VCR included, according to information provided by the Indianapolis mayor’s office. The rents associated with owning a taxi plate are then competed for in a secondary market where agents compete for the right to expropriate the rents. If the regulator does not react to the demand for more licences the inevitable outcome is a situation, like in present day Dublin, where the taxi plates trade for very high prices, and social surplus is transferred from the consumer to the taxi plate owners.

The large investment required to enter the regulated market induces incentives to vertically differentiate the industry into plate holders and their drivers. These plate holders have invested large sums to get the rents and will face financial losses if their monopoly position is undermined. So basic economic principles can explain how the regulatory environment has caused a secondary market to open up for taxi plates and how, over time, the price of these plates have risen to be substantial sums of money. These sums are beyond the means of most potential entrants. Further, given the nature of the licence, it is impossible to borrow money to finance the purchase of the plate. These facts ensure that, in all but the most unusual of cases, the present buyers of plates are individuals of sufficient means to raise £80,000 without recourse to credit markets.

3.2 Vertical Differentiation & Contracting: Urban Sharecroppers

Given the large sums of money involved in securing a taxi plate, it would be safe to assume (other than in the case of small co-operatives of drivers) that the person holding the plate has high opportunity costs. This then creates the need for plate owners to advertise for the services of a PSV licence-holding driver, known as a cosy. The relationship between the plate holder and the driver is a classic principal-agent relationship, with all the concomitant (asymmetric) informational difficulties.

As argued above, the prohibitive cost of plates in Dublin ensures that most new plate owners will not find it optimal to drive the taxi for themselves. They then have the choice to buy a car, insurance etc. themselves and search for cosies, or hire the plate out to an agent who will act as an intermediary. In either case, the holder of the plate is faced ultimately with three possible contracts that they could offer the potential driver:

It is our view that there are serious informational difficulties in contracts (ii) and (iii) which may rule out their attractiveness to the plate holder. This is because of the informational asymmetry normally referred to as moral hazard. Moral hazard refers to facets of a driver’s behaviour which cannot be monitored, without great cost, by the plate owner. These include the drivers level of effort and the actual fare income collected. Because of this, it is in the interest of the plate holder (the principal) to offer to the driver (the agent) the incentive-compatible contract, ie contract (i).

In this, the principal can force the agent to take all the residual risk associated with the fare income and, in the face of a relatively large supply of drivers, force the driver to pay a rent which will leave him with his reservation wage (which in many instances may be the minimum wage). This contract has the benefit to the plate owner that he only has to face one rather than two uninsurable risks (the capital value of the plate and the fare income), and is incentive-compatible. The driver once he has paid his fee to drive the car, is left with the minimum wage.

In looking at the Dublin Corporation taxi register there were 1,974 taxi plates as at the 12 March 1997. We looked at the level of concentration in the ownership of taxi plates and found that there were only 69 obvious multiple owners who, in total, accounted for 193 plates. This left some 1,781 single plate owners. In essence, 3.7% of plate owners held some 9.8% of the plates. This measure, however, understates the level of power yielded by some individuals.

There is a vibrant market for individuals who rent plates and then act as an intermediary between the owner and the driver. This arrangement obviously benefits from some economies of scale in car maintenance, radio cost etc. It is estimated that one individual controls up to 45 plates in such a manner (some 2% of the total). Overall the ownership structure is flat with only a few outliers, in terms of holding over one or two plates. The Oscar Faber Report estimates that there are 1,500 cosies in Dublin. Based on its calculations of fare income inter alia, the Report estimates that cosies would have to work up to 70 hours a week in order to take home the average industrial wage. They report anecdotal evidence that these type of hours are being put in by cosies. The Oscar Faber report demonstrates that any current buyer of a taxi plate must hire a cosy to make the substantial investment economical. The Report also shows that for existing plate owners who did not outlay vast sums to secure a plate are foregoing up to IR£8,000 if they choose not to engage the services of a cosy. Indeed, on examining the Oscar Faber calculations, it is hard to see why the established plate owner does not hire out the plate completely, rather than drive the car himself (unless they have very low opportunity costs of their time).

From their report Oscar Faber show that, ignoring the cost of the plate, plate holders can earn an income of IR£27,383 per annum if they drive the car themselves and hire it out to a cosy. If insurance costs do not vary as between (a) having the plate owner and a cosy and (b) having only two cosies on the policy, the plate owner will receive an income of IR£15,344. This implicitly values the plate owner’s opportunity cost of time at IR£12,039 per annum. If the plate owner could earn more than this (on the basis of a 60-hour week) in any other activity, he should not drive the taxi himself, but engage cosies. The economics of this scenario demonstrates how the number of cosy drivers has risen and will continue to rise under the present regulatory regime.

A modern form of urban sharecropping, we contend, exists to a degree in the Dublin taxi market with this contractual arrangement. While there is as yet, no firm evidence as to the prevalence of this arrangement an article in the Irish Independent of 22 November 1997 by Jerome Reilly claims that the "ordinary driver...rents both his car and plate and...works a 50-hour week to make the average industrial wage." Whilst, in the normal landlord-sharecropper relationship the land to which the landlord had title was his exclusive property, the taxi plate is often the property of the taxi licensing authority. In an article in the Cincinnati Enquirer, it was stated that "city policies should not create a system where individuals or companies are making a profit by leasing city property." But this is precisely what has been allowed to happen in Dublin, if we strictly interpret the taxi plate owners as ‘holders’ rather than owners per se. However, as Kramer and Mellor (1995) point out, the regulatory authority must take some responsibility for creating the problem and should be willing to compensate individuals who would face financial loss. Kramer and Mellor argue that there is an argument for the regulator to buy back all or a portion of the value of existing plates. Of course, it is in the public’s interest that the regulator does not allow such entrenched positions to develop in the first instance and to facilitate ease of entry and exit into and out of the taxi market.

4. Well-Defined Property Rights

Well-defined property rights are necessary to achieve an efficient allocation of resources. Posner observes that

Additionally if a policy change provides access at a price less than an efficient entry price, the change will effect an uncompensated taking of the incumbent’s property right. If there was voluntary consent from incumbent taxi plate owners to guide the use of resources efficiently, then ownership costs would have to be borne by an entrant. An efficient entry price which reflects the ownership costs would therefore have to be computed - an administrative fee of IR£3000 plus the cost of specification of the car as a taxi, plus a non-refundable surety (a deterrent effect) in the event of a refusal to re-issue a licence. If new taxi plates were issued, both the entrant and incumbents would each have to incur such a cost of securing ownership rights in the licensed market and ownership costs (inclusive of deterrent costs) would deter Farrell’s dishonest entry in equilibrium. In addition, any proposed policy change in the regulatory framework must ensure efficient entry, a configuration with low entry costs and the dissipation of secondary market rents.

4.1 Ease of Entry Conditions

The thrust of the argument, presented earlier, points to a potential time consistency problem in any proposal which suggests increasing the number of plates under the present system with transferability. For example, if the regulator were simply to double the amount of plates and retain the present licensing system, not only the core problem of an inefficient allocation of taxis remain but an inter-generational effect would emerge, exacerbating the problem in the secondary market.

To ensure ease of entry, the (high) sunk costs of entry must be removed. It is sub-optimal to have high sunk costs in a market where potential entrants are prepared to enter in reaction to (excess) demand for taxi services. A lowering of the sunk costs can be facilitated by liberalising the current licensing system. To ensure that entrenched dominance does not again creep into the system, it is important, not just to increase taxi numbers, but to ensure that we are not just forestalling dealing with the problem, by making the taxi plates non-transferable. This would vertically re-integrate the system to ensure that the holder of the plate is also a driver of the car.

Though the stipulation that the plate owner be a driver is not a necessary condition for efficiency, it could be argued that if the costs of moral hazard (in terms of the insurance fees) are abated and an improved incentive structure put in place, this would be it a sufficient condition for efficient service. In this manner, in recognition of the fact that the conditions attaching to plates are a matter of public policy, making plates non-transferable will ensure that the fixed costs of entry into the taxi market are such that they facilitate both entry and exit.

Even in an efficient entry regime, the market will be characterised by the existence of cosies. In this instance, there would be some people who would not have access to the capital to put a car on the road and pay any administrative fee etc., or there may be people who choose to remain agents due to their own preferences. The crucial difference being that the vast majority of the current cosies would now be in a position to put their own taxi on the road and work on their own behalf. While it is not our intention to second-guess the market on liberalisation, it does appear that any degree of monopsony power currently enjoyed now by plate owners will be diluted. It is our belief that, in a liberalised regime, the contracts offered to cosies would have to be more attractive than those currently on offer, as the opportunity cost of cosies would rise.

The system cannot be fully liberalised as there are matters such as public safety that necessitate a degree of qualitative regulation. This can be readily accomplished in terms of regular checks and yearly licence renewals. This will militate against hit-and-run entry, which is a problem in an industry with low reputation effects (low probability of hiring the same taxi driver twice). There is also some necessity for some price controls in the form of reasonable price ceilings and the display of fares. In the movement from the regulated system to the liberalised system, given the large investment of the incumbent plate owners, there are clearly going to be financial losses. Investments in taxi plates were made with full knowledge of the nature of the plates yet were done under the auspices of the PSV licensing authorities who must take some responsibility for the current situation.

In the taxi environment, consumers ‘experience’ different taxis and hackneys, queuing, reliability of service inter alia. Farrell (1986) has argued that

"in an experience goods industry, an entrant who could make positive profits by providing a better deal to buyers than do incumbents, may cheat buyers by providing goods of low quality to make even greater profits".

Entry conditions are important for market performance. In particular, if the market is providing less consumer surplus (prices are higher than competitive prices), an entrant can make a positive profit by offering the consumer lower prices. Of course, if buyers have rational expectations, they will be unwilling to buy from the entrant. As a result, moral hazard in a seller’s choice of the quality of an experience good can lead to a barrier to entry. Farrell continues to argue that if hit-and-run entry occurred, the threat of such entry may not discipline the pricing strategies of incumbents. Furthermore, if a new licensing regime is presented as a mechanism for the delivery of honest service and quality that would otherwise be absent, licensing then proffers society a premium for such honesty.

4.2 Just Compensation

The two conditions viz. well defined property rights and efficient entry, are interrelated, arguably, since property rights help to ensure that market exchange is voluntary. All parties may agree that the policy change has to conform to a Pareto improvement criterion (it is just). However it is imperative that the losses associated with the policy change be identified as true opportunity costs in order to avoid moral hazard: if potential losers know with certainty that they will receive compensation, the policy change will induce moral hazard (under any hypothetical bargain, neither party would reveal the ‘true’ preferences under voluntary exchange) on the part of incumbent taxi owners. Alternatively, if taxi owners know that their losses from a policy change will go uncompensated, potential losers will use their real resources to resist the Pareto improving policies. Potential winners may opt to do likewise in defending the change. Consequently incumbent monopoly profits are dissipated by rent-seeking expenditures.

When judging the importance of barriers to entry it is imperative to adopt a criterion of ownership of real resources which encourages investment in the market and militates against hit-and-run entry. Any judgement on barriers must take cognisance of the property rights equilibrium configuration implied by entry. If a Paretian standard is applied to the Irish taxi market it raises the related issue of just compensation for incumbents in any proposed policy change. To recast ownership rights as a basis for compensation in taxi market liberalisation, we would have to ask: how does a barrier which originates in, and can be latterly justified by, the rights system, affect consumer welfare generally?

Market imperfections like the current taxi licensing regime are inherent in any assignment of property rights and do give rise to, and do, perpetuate X-inefficiencies (long queues, for example) in the market. In this paper, we acknowledge that the licensor has ownership rights (a set of well-defined property rights) and that compensation may be an issue. However, compensation rules are generally inefficient because of moral hazard problems. Any adjustment of rights including a removal of the transferability right, and the elimination of the "saleable market" (the secondary market) for taxi plates have a well established legal basis, viz Hempenstall and O’Dwyer.

5. Concluding Thoughts

If the market were liberalised in terms of entry, as advocated by the authors, the secondary market for plates would disappear. The value of existing licences would only be worth whatever administrative charges plus were levied by the licensing authorities. From the perspective of consumers, this must be seen as the second best solution, when coupled with quality and fare regulation (in the form of maximum fares allowable being displayed prominently in taxis). From the perspective of recent plate purchasers, complete and immediate entry deregulation would represent substantial financial losses.

The Oscar Faber Report recommends that there be a 10-year transition period in which 350 new plates would be issued each year in the interim. This transition would be followed by complete entry deregulation. In our view this policy is not time consistent. The recent introduction of 400 new plates has not had an adverse impact upon the prices the people are willing to pay for plates. Investors may continue to support the current equilibrium in the belief that if the market value of plates does not approach the free entry level in the 10 year run up to full deregulation, the authorities will then be forced to renege on their promise to fully deregulate entry. If the Oscar Faber reasoning does not support free entry now, it would not support free entry in 7-10 years time, if the market price of plates has not fallen to free entry levels. In this manner, phased free entry can be time inconsistent. In comparison, the report of the Dublin Taxi Forum recommends an annual increase of 200 taxis (all wheelchair accessible) per annum to 2002 and has no proposals for any increases beyond that time. This proposal does not fundamentally alter the current situation and does not envisage complete deregulation as the end point of the process. The impact on the secondary market for taxi plates would be minimal and this will ensure that the debate about taxis in Dublin will run, and run well into the next century.

However, time consistency problems associated with phased deregulation could be solved if the transferability allowed under the 1978 S.I. is revoked. This would have the immediate effect of closing down the secondary market in plates and the investors would not be able to speculate as to whether the authorities would fulfil their stated intention to fully deregulate in 10 years. This time consistency problem remains for all phased solutions once the secondary market remains open for taxi plates. In the run-up to the date of revocation of transferability, the market value of taxi plates would converge to the present value of rents that new entrants can hope to make in the transitional period. Recent entrants can choose to take their profits to sell the plate at the new market value or to keep the plate and take the protected profits in the period up to full deregulation.

If the licensing authorities choose to deregulate in a phased fashion or indeed to just increase numbers of plates in accordance with their view on the numbers required, it is imperative that they take steps to close down the secondary market in taxi plates. As argued in the paper, dissipation of secondary market rents is necessary for efficient entry. The incumbent rent-seekers will militate against attempts to liberalise the taxi entry regime. The authors would prefer a swift move to complete entry liberalisation, but in the absence of such a move, the revocation of the right to transfer the taxi plate is a necessary and inevitable step for any sustainable long term market configuration with entry liberalisation. A more efficient entry regime would, in our opinion, abate the principal-agent problem and allow urban sharecroppers to reap the benefits of ownership. It would also create an avenue for people of limited means to enjoy the benefits of an enterprise culture and the free market, while increasing consumer surplus for taxi users.

Adverts