Taxi rank competition and regulation
Bart Voorn (10290680)
October 31, 2015
Abstract
While recent empirical research into the workings of taxi markets has brought evidence of price gouging and excessive entry at taxi ranks worldwide, models explaining these market imperfections still remain uncalibrated with empirical evidence of deregulation. This paper offers a framework for understanding observed ‘market failure’ at taxi ranks as a consequence of ‘first in, first out’ (FIFO) legislature. Our model suggests that, barring price and entry regulation, abandoning FIFO regulation is preferable. Otherwise, entry regulation is to be preferred over price regulation. Comparisons of successful deregulation (Ireland, New Zealand) with unsuccessful deregulation (Canada, Sweden, The Netherlands, UK, US) substantiate this claim.
Keywords: competition, excessive entry, FIFO, regulation, taxicabs, taxi ranks
Statement of Originality
This document is written by student Bart Voorn who declares to take full responsibility for the contents of this document.
I declare that the text and the work presented in this document is original and that no sources other than those mentioned in the text and its references
have been used in creating it.
The Faculty of Economics and Business at the University of Amsterdam
is responsible solely for the supervision of completion of the work, not for the contents.
Statement of Gratitude
I would like to thank Dr. Sander Onderstal for his support and supervision of this dissertation. This dissertation, and in particular the theoretical model,
was a product of much trial and error and Dr. Onderstal’s ideas helped make the end product possible.
Contents
1 Introduction 3
1.1 Background . . . 3
1.2 Contribution . . . 4
2 Literature review 6 2.1 Inhibitions of competition at taxi ranks . . . 6
2.2 Current accounts in perspective . . . 7
3 Theory 9 3.1 The model . . . 9
3.2 The FIFO environment . . . 10
3.3 Free choice . . . 11
3.4 Comparing the systems . . . 15
4 Empirical literature 17 4.1 Regulation across countries . . . 17
4.2 Deregulation without FIFO laws . . . 19
4.3 Deregulation while retaining FIFO laws . . . 20
4.4 Public versus private sector management . . . 22
5 Discussion and conclusion 23
1
Introduction
1.1 Background
Economists typically portray taxi ranks as a source of market failure in taxi industries. Empirical studies such as those by Teal and Berglund (1987)
and Schaller (2007) suggest that taxi ranks have inefficiently high fares and an oversupply of taxis. Economists have proposed various reasons why taxi
ranks function imperfectly, such as search costs and capacity constraints. However, these explanations are not wholly convincing: search costs at taxi
ranks tend to be low, and parking capacity often goes unused, particularly at airports. While there is empirical evidence of market failure at taxi ranks,
the current theories fall short in explaining this phenomenon.
Moreover, when examining the deregulation of taxi markets, economists
generally find ambiguous results. In some countries, such as in Ireland and New Zealand, deregulation is evaluated very positively, as it coincided with
large reductions in taxi fares for consumers (e.g. Barrett, 2010; Gaunt, 1996). In other places, such as in Sweden, the United Kingdom, and the
United States, deregulation did not precede a reduction in taxi fares, and its results are judged mostly negatively (e.g. Kang, 1998; Schaller, 2007).
The literature on taxi market economics has not yet presented convincing reasons as to why the outcomes of deregulation have diverged so strongly
across countries.
Why does competition at taxi ranks oftentimes fail? And what explains
diverging outcomes of taxi market deregulation across countries? As far as I am aware, economists have not provided convincing explanations for these
questions in the literature. Consequently, economists and policymakers have difficulty devising optimal regulation for the taxi industry, and excessive
entry and pricing continues to be a problem at many taxi ranks, hurting both taxi drivers and consumers.
1.2 Contribution
This dissertation investigates the proposition that both market failure at
taxi ranks and the diverging outcomes of deregulation can be explained by the ‘first in, first out’ (FIFO) laws that are imposed at taxi ranks in some
countries but not in others. FIFO laws, which force passengers to take the first taxi in line, are sometimes enforced at taxi ranks to ensure ‘public
order’ (LaCroix et al., 1986, Toner, 1992, Baanders and Canoy, 2010); that is, they are argued to prevent congestion and facilitate the in- and outflow
of taxis at ranks. Yet no studies have verified their necessity, and they are rarely employed at private ranks. Since they enhance cab drivers’ market
power (who no longer have to compete on prices as consumer choice is taken away), FIFO legislation may raise prices, reduce product variety, and entice
additional (excess) entry to the market, providing an alternative explanation of taxi rank ‘market failure’ and diverging outcomes of deregulation across
nations.
On the other hand, there may be a benefit to FIFO legislation that has
hitherto not been specified in the literature. FIFO laws encourage extra supply, and this increased entry may reduce waiting times for clients and
thus raise welfare. Consider, for instance, a small taxi rank where in a free choice equilibrium the entry of two firms leads both firms to make negative
profits, encouraging them both to enter only sometimes. It is plausible, or at least not inconceivable, that introducing FIFO legislation here would allow
the market to be sustainable for two firms, and that the reduced chance that no taxicab is available for customers may on the whole offset FIFO legislation’s anticompetitive effects.
The purpose of this paper is to answer the following research question:
“How does FIFO regulation affect the pricing and entry decisions of taxicab firms at taxi ranks?”
I answer this question through the use of two methods. First, I develop a model that captures both the positive and negative welfare effects of FIFO
regulation, and explore the theoretical effects of FIFO legislation on pricing and entry decisions of taxicab companies. Second, I test the hypotheses
derived from this model in an empirical section. While an international all-encompassing econometric study of taxi markets is impossible due to the
worldwide lack of data on annual taxi prices, a case study approach can at least descriptively test this paper’s thesis. I discuss the situation before
and after deregulation, both in countries that have FIFO regulation in place (Sweden, the United Kingdom, the United States) and in those where FIFO
laws have not been adopted (Ireland, New Zealand), particularly examining the outcomes of deregulation in changes in profits, taxi fares and consumer
satisfaction. In this empirical part, I also consider evidence from private taxi rank management as a proxy for what optimal regulation should look
like.
The remainder of this paper is organized as follows. The next section
reviews the literature and identifies evidence for market failure at taxi ranks, and critically engages with explanations of what caused these problems that
have hitherto been offered by economists. In section three, I construct a model of a taxi rank environment, and compare profits, consumer surplus, and welfare in this environment when FIFO laws are imposed and when they
are abandoned. Section four, next, dissects the empirical literature on taxi rank regulation and regulatory outcomes, and discusses empirical evidence
on how taxi ranks are managed in the private sector. Finally, section five discusses policy implications and concludes.
2
Literature review
What inhibits competition at taxi ranks and what explains it? This section reviews evidence for market failure at taxi ranks and engages with hitherto
proposed explanations. I conclude that current explanations fall short in explaining market failure at taxi ranks and propose a new hypothesis.
2.1 Inhibitions of competition at taxi ranks
Transport economists typically find that taxi ranks – queue areas where taxicabs congregate in anticipation of high passenger demand – are markets
that do not act in accordance with hallmark ideas of ‘perfect competition’. Oft-cited empirical studies, such as those by Teal and Berglund (1987) and
Schaller (2007), suggest that taxi ranks often have inefficiently high fares and an oversupply of taxis. Both consumers and producers are worse off
for it: consumers pay higher prices than they would need to, or opt for other modes of transport instead; producers find themselves in a cut-throat
economic environment with very small profit margins. Moreover, authors such as Baanders and Canoy (2010) note that collusive practices between
competitors and intimidation of customers and competitors often surface at taxi ranks.
Economists have proposed several hypotheses to explain the coincidence of excessive entry and price gouging at taxi ranks. Cairns and Liston-Heyes
(1996) construct a model explaining these as a function of search costs: when finding empty and desirable taxicabs takes time, firms have market power
and can price above the ‘perfect market’ price, which in turn entices excessive entry. Another explanation involves capacity constraints: Kang (1998) notes that taxi ranks, due to limited space, can only hold so many cabs, and thus
incumbents can price higher while competitors are incentivized to enter the market, and externalities such as congestion cause this entry to be ‘in excess’.
Indeed, economists since Shreiber (1975) have emphasized that entry
restrictions allow internalization of these externalities at taxi ranks. Many economists, however, note that such entry regulation should be paired with
price regulation in order to prevent the abuse of market power (e.g. Teal & Berglund, 1987). This was the academic case for regulating the taxi
industry, and most countries today have both price and entry regulation in place for their major taxi ranks, such as those near airports and railway
stations (Kang, 1998). According to Schaller (2007), this case has mostly brought about policy success: those countries that opt for combinations
of price and entry regulation generally manage to reduce the oversupply of taxis at taxi ranks, to the benefit of welfare. Meanwhile, the extent to which
regulators succeed to keep fares low vastly differs between countries, where places such as Sweden have managed to keep fares relatively low, while fares
remain high in countries such as France (Kang, 1998).
2.2 Current accounts in perspective
Despite successes in mitigating excessive entry, transport economists are certainly not uniformly in favor of these regulations. For instance, Teal
and Berglund (1987) emphasize that these restrictions come at a cost: they create black markets, deter innovation, and may lead to rent seeking and
regulatory capture. The counterargument to this has been that deregulation also comes at a cost: that of rising entry without rising prices. Indeed, the
recent experience in countries such as the Netherlands (Bakker, 2007) and the United States (Schaller, 2007) has been just that. Yet there have also
been examples of countries in which deregulation has not led to excessive entry and pricing. The literature has, for instance, lauded the results of
entry and price deregulation in Ireland (Barrett, 2003) and New Zealand (Gaunt, 1996), where fares have decreased by margins of up to 30 per cent
However, the strongly juxtaposed experiences of deregulation between
these countries on the one hand and the Netherlands and the United States on the other hand have scantily been scrutinized in the literature. While
some authors, most renownedly Kang (1998), have compared regulatory approaches and outcomes across countries, their objective has generally not
been to discern what caused the discrepancy between countries, but rather to provide data for the economic literature at large. This leaves us with an
important question: why was deregulation successful in countries such as Ireland and New Zealand, but not in countries such as the Netherlands and
the United States? For both economists and regulators, an answer to this question would surely be beneficial.
Moreover, the empirical literature sheds doubt on theoretical accounts as to why taxi ranks experience both oversupply and overly high prices.
While differences may exist between different countries in the extent to which search costs or capacity constraints are present, it seems improbable
that these differences are large enough to explain the variation in regulatory outcomes (cf. Small & Verhoef, 2007). While Kang’s case that capacity
constraints could raise prices by limiting supply is true, in practice, major taxi ranks rarely exhibit capacity constraints. At airports, there are major
incentives for government agencies or airport owners to ensure sufficient space for passenger transportation, and ranks are rarely fully occupied. In addition, the search cost model of Cairns and Liston-Heyes (1996) may not
apply to taxi ranks.1 Search costs at taxi ranks are typically slim, since firms can easily advertise prices, and are often even forced to do so. In fact,
Williams (1980) argues that a major reason for taxi ranks to develop in the first place is to reduce search costs for passengers. Thus, these accounts of
excessive pricing and entry at taxi ranks might not suffice.
1
Note that Cairns and Liston-Heyes (1996) do not themselves apply their model to taxi ranks but rather to the dispatch market. The theory that search costs are relevant at taxi ranks is instead advanced by Kang (1998).
3
Theory
In this section, I present the theoretical model. Subsection 3.1 introduces the model and explains the key assumptions. Subsection 3.2 considers what
happens when FIFO regulation is introduced, and subsection 3.3 considers what happens when such regulation is abandoned. Finally, subsection 3.4
compares the two regulatory environments and 3.5 offers extensions.
3.1 The model
The model below illustrates depicts a simple case: that of two taxi firms i and
j that independently decide whether or not to pick up a single customer at a rank. Competition in the model works as in a Salop circle with perimeter 1,
where taxis are located on the circle. Customers’ preferences for firm i or j are captured by distance d along the perimeter of the circle, where distance
from the taxi firm indicates to what extent the firm deviates from their most preferred taxi, multiplied by ‘disutility’ multiplier t. Thus the customer’s
utility is:
U = V − td − p, (1)
where V is the customer’s valuation for any taxicab, and p is the fare set by
the taxi they choose.
Firms maximize profits. They incur a fixed entry cost F upon entering
the market, as well as marginal cost c in the event they pick up the customer. The firms independently set their fare rates pi and pj and independently
decide whether or not to enter the market. If both firms enter, both have an equal chance of being the first taxi in the line. In the free choice environment,
the customer picks the taxi that offers the highest utility; in the FIFO environment, the customer can only pick the first taxi in line.
Assume V is large enough that all customer types are served, and that
in the FIFO environment two firms enter in equilibrium: V > 12t + 2c + 2F (this will be confirmed in equation 3). Moreover, suppose fixed costs are
relatively large: F > 4t.
Consider a welfare-maximizing regulator who considers introducing FIFO
regulation at this rank. Suppose the regulator could also potentially adopt price or entry regulation both in the FIFO and free choice environment.
This regulator would then compare producer and consumer surplus in both environments and then consider whether price regulation or entry regulation
could improve welfare. The regulator’s trade-off is considered in the next subsections.
3.2 The FIFO environment
Under the conditions specified, a taxi firm in a FIFO system – once ahead
in the line – would not be competing with the other firm. Thus, this firm would charge the monopoly price pF IF O. Since we assume the firm serves
the entire market, this monopoly price is set such that the passenger with the highest travel distance has a valuation of zero. Since the customer can pick a single firm, the largest distance on a Salop circle with perimeter one
is 12. Thus: u(d = 1 2) = V − pF IF O− 1 2t = 0 → pF IF O= V − 1 2t (2)
The expected profit that a firm makes is:
π = V −
1 2t − c
N − F (3)
Under the conditions specified, this profit is positive in equilibrium for N = 2. Thus both firms enter and make positive profits in equilibrium.
Total consumer surplus is equal to the customer’s expected utility. On a
Salop circle with perimeter 1, where the customer can only choose a single firm and where the customer’s utility decrease linearly with travel distance,
the expected utility of the customer is equal to the utility of the customer with the average travel distance d = 14. Thus the consumer surplus in a
FIFO environment, CSF IF O, is equal to:
CSF IF O= u(d = 1 4) = V − p − 1 4t = 1 4t (4)
Producer surplus is equal to the profits of both taxi firms. The profits for
the first-in-line taxi is p − c − F , while for the second-in-line taxi it it is −F . Producer surplus is:
P SF IF O= (p − c) − 2F = V −
1
2t − c − 2F (5)
Total welfare in a FIFO environment is producer surplus and consumer surplus combined:
WF IF O= P SF IF O+ CSF IF O = V −
1
4t − c − 2F (6)
3.3 Free choice
In an environment where consumers can freely choose between the two firms, both firms maximize their prices, which we shall call pi for firm i and pj for
firm j, given the price that the other firm sets. This price can be computed by finding the Nash equilibrium: we find the consumer that is indifferent
between the two firms, derive the two firms’ demand functions from this, and then maximize their profits subject to price. To derive the equilibrium,
The indifferent consumer, d∗, is at: pi+ td∗ = pj+ t( 1 2 − d ∗ ) → d∗ = pj+ t 2 − pi 2t (7)
The demand for both firms is equal to twice the distance to this consumer
(on both sides of the circle):
DF C,duo= 2d∗ =
pj+2t − pi
t (8)
From this, we can compute profits. Profits are equal to half the demand (a firm serves customers only half of the time since the other half of the time
customers are served by the other firm) multiplied by the price-cost margin minus the fixed entry costs:
πF C,duo,i =
1
2(pi− c)D − F =
pj+2t − pi
2t (pi− c) − F (9)
Firm i maximizes profits with respect to price:
F OC : 4pi+ t − 2pj− 2c
4t = 0 (10)
Since profits face this decision simultaneously and independently, we may
assume symmetry:
pi = pj → pF C,duo= c +
t
2 (11)
πF C,i=
(p − c)
2 − F =
t
4− F (12)
Note that since we have assumed that fixed costs are not small compared to
travel costs (t < 4F ), this profit is lower than zero. Thus two firms do not always enter. Instead, suppose taxis employ a mixed strategy of entering,
where they enter sometimes but not always. Suppose q is the probability of entry for a single firm. A firm maximizes profits with respect to price:
πF C,i= (1 − q)(pi− c) + q(pi− c)D(pi) − F (13)
= (1 − q)(p1− c) + q(p1− c)
pj +2t− pi
t − F, (14)
where the left side of the formula indicates the expected revenue when the second firm does not enter multiplied by that occurring, and the right side
indicates the expected revenue when the other firm enters multiplied by the probability of that event. The first order condition is:
F OC : 1 − q +q(pj+ t/2 − pi)
t −
q(pi− c)
t = 0 (15)
Since again firms set their fares independently and simultaneously we may assume symmetry: pj = pi→ pi = (1 − q 2) t q + c = t q − t 2+ c (16)
Since in equilibrium firms should be indifferent between entering and not entering, profits should be equal to zero. Thus:
(p − c)(1 − q) + (p − c)q
→ (t q − t 2)(1 − q) + (qt − t 2)q 2 − F = 0 → tq2− 4(t + F )q + 4t = 0 (17)
This equality has a unique solution for 0 < q < 1:
q = 2 +2(F −p(F + 2t)F )
t (18)
If a single firm enters, consumer surplus is calculated as in the previous subsection: CSN =1= u(d = 1 4) = V − 3 4t − t q − c (19)
If two firms enter, the average customer has travel costs of d = 18:
CSN =2= u(d = 1 8) = V − 5 8t − t q − c (20)
Total consumer surplus is the two previous equations multiplied by their
probability: CSF C = (q2)CSN =2+ (1 − q2− (1 − q)2)CSN =1 = (V −5 8t − t q − c)q 2+ (V −3 4t − t q − c)(q − q 2) = q(V +tq 8 − 3 4t − c) − t, (21) where q = 2 +2(F − √ (F +2t)F ) t .
3.4 Comparing the systems
The free choice system is preferable when:
WF C > WF IF O→ q(V + tq 8 − 3 4t − c) − t > V − 1 4t − c − 2F → q(V +tq 8 − 3 4t − c) > V + 3 4t − c − 2F (22)
The exact solution is complicated, but we can draw general conclusions. First, note that the free choice system becomes more preferable as q
rises. Consider the formula for q given at the bottom of the last page. It can be computed that q tends toward zero as t tends toward zero, and that
q tends toward 1 as t tends toward 4F . Thus, the free choice environment becomes more preferable as t rises. This is intuitive. When the customer’s
preferences for one of the cabs becomes stronger, the possibility of choosing between cabs nets greater consumer surplus. In addition, when customers
do not have strong preferences between the firms, entry becomes relatively unattractive as firms compete fiercely in price and can capture less surplus.
Our model implies that if the customer does not have strong preferences regarding taxis, the regulator should prefer FIFO regulation. If, on the
other hand, the customer has strong preferences, the regulator should not take away the customer’s ability to choose between firms.
Can the regulator improve over this situation? Since the reason that FIFO regulation may sometimes be preferable to free choice despite a higher
price is that it ensures any customer will be served, might the regulator do better by creating an environment in which a single firm serves the market,
either by introducing a medallion system or by setting a price below pF IF O
If the regulator allows only one firm to serve the market, welfare becomes: Wonef irm= CS + P S = (V − p − 1 4t) + (p − c − F ) = V − 1 4t − c − F (23)
Compared to the FIFO environment, welfare now has increased by F , i.e. the costs from idle entry have been avoided while the presence of a taxicab
is ensured. Thus entry regulation (or indirect entry regulation by setting the price such that the market is only profitable for one firm) is, under the
model’s assumptions, strictly preferable to FIFO regulation.
Moreover, note that in the model, changing the price has no influence on welfare since it is simply a transfer from consumers to producers. Thus the
case for entry regulation is also much stronger than that for price regulation. The theoretical implications of the model are that:
• FIFO regulation induces excessive entry.
• A free choice environment induces excessive entry sometimes, too little
entry at other times, and the right amount of entry most of the time (assuming q > 12).
• Prices are higher in a FIFO environment than in an environment that
allows free choice.
• Entry regulation can counter both excessive and insufficient entry.
Price regulation can too, but only to the extent that it can correctly
4
Empirical literature
This section reviews the empirical literature that relates to the theoretical results derived in the previous section. It compares countries where FIFO
regulation is applied to countries where it was abandoned, and also discusses how private taxi ranks are managed. The main question posed in this section
is: do taxi ranks in countries without FIFO regulation indeed perform better?
4.1 Regulation across countries
As previously noted, the transport economics literature has been critical of
how taxi ranks operate in practice. Economists have observed the peculiar coincidence of excessive entry and price gouging at major taxi ranks in
Australia (Abelson, 2010; Biggar, 2011), the Netherlands (Bakker, 2007; Baanders & Canoy, 2010), the United Kingdom (Toner, 1996, 2010) and the
United States (Teal & Berglund, 1987; Schaller, 2007). Very few examples of ‘perfect’ taxi rank management exist around the world (La Croix, Mak,
& Miklius, 1986; Kang, 1998). Following such observations, many support regulating taxi ranks.
Such regulation at taxi ranks can come in one of three forms. First is entry regulation, which usually comes in the form of occupational licensing.
Entry regulation is normally imposed to allow the taxi sector some positive profits, as revenue mechanism for local government, or to prevent congestion
at taxi ranks when there are capacity constraints. It may also indirectly come about as a result of quality control. The second is price regulation,
which often places a cap on the fares that taxicabs can charge, although exact price fixing is not unheard of. The last regulation is what I shall coin FIFO regulation, or ‘first in first out’ regulation, which are mandates that
the first cab in line has a right to the passenger (see, for instance, La Croix et al. (1986); Toner (1992); Baanders and Canoy (2010)), that are usually
Regulations have diverged greatly across countries and across time. For
instance, in the late 1970s, nearly all countries moved towards regulation, perhaps following the increasing push for price and entry restrictions in the
academic literature (Teal & Berglund, 1987). Yet the recent trend has been the reverse: countries increasingly move towards deregulation, for reasons
ranging from curbing rent seeking to encouraging innovation (Bakker, 2007). There have also been large differences between countries: whereas countries
such as New Zealand, Canada, and Ireland have been deregulating since the 1980s, countries such as France and Japan have been more hesitant to
leave the taxi sector up to market forces (Kang, 1998). The United States, meanwhile, is divided between states that regulate heavily (predominantly
in the Northeast of the country, such as in New York) and states with a more laissez-faire attitude (mostly in the South and the Midwest) (Schaller,
2007).
This large divergence of taxi market regulations over time and across
countries has allowed much empirical research. Hereafter, I will discuss seven of the countries which have recently moved towards deregulation where
the empirical research on the results has been most extensive. Most of these countries have done away with price and entry regulation while they
retained FIFO regulation for public order reasons: this category includes Canada, the Netherlands, Sweden, the United Kingdom and the United States. Only Ireland and New Zealand also did away with FIFO regulation.
I will discuss these nations in separate sections. Finally, a small segment of the literature has also considered the ways that the private sector regulates
private taxi ranks. Since the private sector has a direct monetary stake in optimizing its regulation, it may provide some insight on regulation for
academics and policymakers, and for completeness I include a small section on this literature as well.
4.2 Deregulation without FIFO laws
In the Republic of Ireland, entry restrictions were up until the 1990s some
of the strictest in the world, demonstrated by the unprecedented market value of taxi licenses (cf. Barrett, 2004). Price regulation was strict and
was intended to protect consumers, although consumer satisfaction was not high. The regulation changed in 2000, when the Irish High Court judged
entry restrictions illegitimate (Barrett, 2003). Shortly after, the Commission for Taxi Regulation in Ireland declared FIFO regulation to be a violation of
antitrust laws, and moved to abandon it. All measures intended to protect producers were abandoned. Only some price regulation (in the form of
maximum prices) remained.
The literature has noted that deregulation in Ireland was very successful.
Researchers have noted that consumer satisfaction rose substantively in the wake of deregulation (Barrett, 2003; Seibert, 2006). Prices were no longer
fixed at the maximum price and now fluctuated freely under that threshold. The average fare dropped substantially, as did entry to the market (Barrett,
2010); and the enhanced flexibility of prices helped combat congestion and reduced the volatility of taxi producer profits.
Deregulation had earlier taken place in New Zealand, where regulatory changes were similar. Initially, the New Zealand Ministry of Transport had
strictly regulated entry, fares, and quality nation-wide (Kang, 1998); the high cost of licenses demonstrate the strictness of regulation in this period.2
After the 1983 land transport reforms in New Zealand, which protected the taxi sector more than any other, increasing public pressure forced the legislature to review the ‘modernity’ of its taxi policies. In 1989, parliament
passed the Transport Services Licensing Act, which abolished regulation of entry and fares, as well as FIFO laws.
2For a detailed treatment of the relation between taxi medallion prices and regulatory
In New Zealand, too, the effects of deregulation are judged positively.
Deregulation substantially reduced fares in larger cities, particularly in the capital city Wellington, and there was evidence of a substantial decrease in
passenger waiting times. Deregulation also sparked a wave of innovation in the taxi industry. Taxicab firms introduced mail delivery and taxi-van
services, and became more effective in advertising these services (Gaunt, 1996).
These findings are in line with the FIFO hypothesis proposed in the present paper. From a welfare perspective, markets functioned significantly
better after New Zealand and the Republic of Ireland abandoned FIFO regulation, and the dismissal of entry regulation caused a significant decline
in prices. Deregulation in the Republic of Ireland and in New Zealand was very effective and made taxi markets function better.
4.3 Deregulation while retaining FIFO laws
In other countries, deregulation of the taxi sector did not coincide with a
dismissal of ‘first in, first out’ laws. Deregulation efforts in countries such as the Netherlands, Sweden, the United Kingdom and the United States
focused on eliminating entry and/or price controls, but FIFO regulation remains applied at these countries’ taxi ranks.
In the United States, taxi market regulation is applied at the city level, and deregulation took place in various cities. Empirical studies focused
on deregulation in cities such as Charlotte (NC), Kansas City, San Diego, San Jose, and Seattle (Teal & Berglund, 1987; Kang, 1998; Schaller, 2007).
These cities initially employed combinations of price and entry regulation, and deregulation focused mostly on removing entry restrictions. However,
when cities deregulated entry to their taxi industries, FIFO regulation was rarely abandoned, and to this day FIFO laws are nearly universally applied
The literature studying the effects of deregulation in the United States is
distinctly less positive about the results on fares and quality, although there remain large differences between states and between cities. Major studies
by Teal and Berglund (1987) and Schaller (2007) note that deregulation of entry caused little to no reduction in price gouging and sparked excessive
entry at taxi ranks. Few judge the empirical evidence of deregulation in the United States as positive.
Schaller (2007) draws the same conclusion about Canada. In Canada, also, taxi market regulation is universally enforced at the county and province
level, rather than on a national level. Yet FIFO rules have been retained across the board, and as Schaller (2007) notes in his study, many Canadian
jurisdictions now suffer from prices that are too high and entry that is too excessive. It should be noted, however, that in Canada there had been much
less a ‘wave’ of deregulation than in the United States, and accordingly many entry restrictions in the form of medallions or “plate systems” (ibid., p. 4)
remained in place even after deregulation.
In other countries that retained FIFO regulation at ranks, the effects
of deregulation were similarly unimpressive. In the United Kingdom, FIFO regulation is nationally applied (Toner, 1992) to ensure public order at taxi
ranks; the same policy is upheld in the Netherlands (Baanders & Canoy, 2010) and Sweden (Marell & Westin, 2002) (although in the Netherlands an exception has been made for Schiphol Airport Amsterdam, where due to
political bargaining the airport authority, rather than the national transport authority, is in charge of taxi regulation). Empirical studies in all three of
these countries have noted little effect of the liberalization of entry on taxi fares and quality, both positive and negative. However, these countries did
experience significantly increased entry and shrinking profits for the taxi sector, leaving some to argue that deregulation had caused a net negative
4.4 Public versus private sector management
At private taxi ranks, FIFO regulation is rarely employed, even though
quite typically private firms employ some for of entry regulation and, very occasionally, price regulation. Typically, a set amount of parking spots in
front of hotels is reserved for taxis, which are either selected by the hotel or are freely accessible by non-affiliated taxicabs. Hotels sometimes make
arrangements with taxi firms about prices or offer their own services, which can either be more affordable than calling a taxicab (as a means of competing
with other hotels) or more expensive (to put a premium on the service of reducing search costs for those frequenting the hotel). When multiple taxis
are waiting in front of the hotel, no law or rule is regularly in place for picking the first cab in line, though as Baanders and Canoy (2010) note
taxis often collusively enforce such rules themselves when there are capacity constraints, taking advantage of the fact that the lack of capacity prevents
other taxis from entering the rank as a response to increased fare rates. An interesting case of private sector management of taxi ranks that
Baanders and Canoy (2010) discuss is that of Schiphol Airport Amsterdam in The Netherlands. Although the airport taxi rank is publicly owned, it is
managed by the airport authority, rather than by the transport authority (which administers all other ranks in the country), and thus the airport falls
under different regulation than the taxi industry in the remainder of The Netherlands. As opposed to ranks outside the airport, the airport authority
avoids enforcing FIFO rules by assigning each taxi firm a zone where it can advertise its services. Taxi drivers are prohibited from entering the terminal and the sidewalk leading up to it. In this way, it has preserved competition
while avoiding public order nuisances. Baanders and Canoy (2010) note that passengers evaluate Schiphol taxis considerably more positively than their
5
Discussion and conclusion
The model presented in this paper provides tentative evidence that, at a small taxi rank with a single customer and up to two taxi firms, ‘first in
first out’ laws lead to excessive entry and higher prices than a free choice environment. The model, however, similarly illustrated that a free choice
environment may at times bring about insufficient entry at the taxi rank. Under certain circumstances (particularly, when the customer has no strong
preferences regarding what taxi to travel with) the FIFO environment can be preferable over the free choice environment. Additionally, the model
shows that entry regulation, when set precisely, can prevent excessive entry under FIFO regulation, but not prevent entry problems in a free choice
environment. Price regulation can accomplish the same to the extent that it induces the correct number of entrants.
The empirical literature, however, is less positive about FIFO regulation. When countries deregulated their taxi industries, countries that abandoned
FIFO regulation as well, such as New Zealand and the Republic of Ireland, performed substantially better than those that retained it, such as Canada,
the Netherlands, Sweden, the United States and the United Kingdom. In places where FIFO regulation was retained, furthermore, there were large
differences in the extent to which regulators were successful in employing entry and price regulation to combat excessive entry and pricing at taxi
ranks, demonstrating that setting the correct number of entrants and fare rates is difficult in practice. Moreover, there is little empirical evidence
of customers having to wait for taxicabs (signaling insufficient entry) in those places where FIFO regulation was abandoned, as the model predicts.
Finally, we have found that regulation is rarely employed at private taxi ranks, where managers should have a strong economic interest in setting the
There are a few reasons why the empirical reality differs from the model
presented in this paper. First, we have made various assumptions in the model that helped illustrate when FIFO regulation would be preferable to
a free choice environment, but might not hold in the real world. In reality, there is often more than one customer requiring service in a given time
period, and there are often more than two taxi firms that could enter the market, reducing the chance that a customer arrives at an empty taxi rank.
Moreover, the model assumes that all customer types are served, which in reality might not necessarily be true and ignores substitution effects that
occur when fares go up, strengthening the case for free choice (where the price is lower). Finally, the model was set up such that in equilibrium, entry
was not profitable for two firms in a free choice environment. If it were to be, then free choice should be strictly preferable over a FIFO environment.
Future research could expand on these findings by including more than one customer and more than two taxis at the ranks.
Altogether, this dissertation offers a tentative case that at small taxi ranks where customers do not have strong preferences between firms, there
is a strong case for enforcing FIFO rules. However, when customers do have strong preferences between firms, or are not all served in equilibrium, the
model and empirical research both illustrate that the FIFO system at taxi ranks should create a welfare loss. In addition, this dissertation has shown that FIFO regulation provides a salient explanation as to why deregulation
of taxi industries has failed in some countries but succeeded in others. While further studies are imperative, policymakers should take note of this lesson.
References
Abelson, P. (2010). The high cost of taxi regulation, with special reference to Sydney. Agenda: A Journal of Policy Analysis and Reform, 17 (2),
41-70.
Baanders, A., & Canoy, M. (2010). Ten years of taxi deregulation in the
Netherlands: The case for re-regulation and decentralization. Associ-ation for European Transport Working Paper.
Bakker, P. (2007). Deregulation of the taxi industry: Experiences in the Netherlands. In: (De)regulation of the taxi industry, Roundtable 133,
ECMT Paris.
Barrett, S. D. (2003). Regulatory capture, property rights and taxi
deregu-lation: A case study. Oxford, UK: Blackwell Publishing.
Barrett, S. D. (2004). Privatization in Ireland. CESifo Working Paper, No.
1170.
Barrett, S. D. (2010). The sustained impacts of taxi deregulation. Economic
Affairs, 30 (1), 61-5.
Biggar, D. (2011). Why and how should we regulate taxis? Melbourne, AU:
The Victorian Taxi Inquiry.
Cairns, R. D., & Liston-Heyes, C. (1996). Competition and regulation in
the taxi industry. Journal of Public Economics, 59 (1), 1-15.
Gaunt, C. (1996). The impact of taxi deregulation on small urban areas:
Some New Zealand evidence. Transport Policy, 2 (4), 257-62.
Kang, C. (1998). Taxi deregulation: International comparison (Unpublished
master’s thesis). Institute of Transport Studies, University of Leeds, Leeds, UK.
La Croix, S. J., Mak, J., & Miklius, W. (1986). Airport taxi service reg-ulation: An analysis of an exclusive contract. Transportation, 13 ,
145-61.
areas of Sweden. Journal of Transport Geography, 10 (2), 135-44.
Schaller, B. (2007). Entry controls in taxi regulation: Implications of US and Canadian experience for taxi regulation and deregulation. Transport
Policy, 14 (6), 490-506.
Seibert, C. (2006). Taxi deregulation and transaction costs. Economic
Affairs, 26 (2), 71-3.
Shreiber, C. (1975). The economic reasons for price and entry regulation of
taxicabs. Journal of Transport Economics and Policy, 3 , 268-79. Small, K. A., & Verhoef, E. T. (2007). The economics of urban
transporta-tion. London, UK: Routledge.
Teal, R. F., & Berglund, M. (1987). The impacts of taxicab deregulation in
the USA. Journal of Transport Economics and Policy, 21 (1), 37-52. Toner, J. P. (1992). Regulation in the taxi industry. Institute for Transport
Studies Working Paper. Leeds, UK.
Toner, J. P. (1996). English experience of deregulation of the taxi industry.
Transport Reviews, 16 (1), 79-94.
Toner, J. P. (2010). The welfare effects of taxicab regulation in English
towns. Economic Analysis and Policy, 40 (3), 299-312.
Tullock, G. (1975). The transitional gains trap. Bell Journal of Economics,
6 (2), 67-78.
Williams, D. J. (1980). The economic reasons for price and entry regulation of taxicabs: A comment. Journal of Transport Economics and Policy,
