Transportation Research Part A 64 (2014) 92–109

Contents lists available at ScienceDirect

Transportation Research Part A

journal homepage: www.elsevier .com/locate / t ra

Joint impact of competition, ownership form and economicregulation on airport performance and pricing

http://dx.doi.org/10.1016/j.tra.2014.03.0080965-8564/� 2014 Published by Elsevier Ltd.

⇑ Corresponding author. Address: Jerusalem School of Business Administration, Hebrew University, Mount Scopus, Jerusalem 91905, Israel. Tel5883449.

E-mail addresses: msnic@huji.ac.il (N. Adler), v.liebert@jacobs-university.de (V. Liebert).

Nicole Adler a,b,⇑, Vanessa Liebert c

a Hebrew University of Jerusalem, Israelb University of British Columbia, Canadac School of Humanities and Social Sciences, Jacobs University Bremen, Germany

a r t i c l e i n f o

Article history:Received 3 April 2012Received in revised form 10 January 2014Accepted 18 March 2014Available online 19 April 2014

Keywords:Airport efficiencyData envelopment analysisOwnership formRegulationCompetitionPricing

a b s t r a c t

The combined impact of ownership form, economic regulation and competition on airportperformance is analyzed using data envelopment analysis to measure cost efficiency in thefirst stage and regression analysis to measure the impact of the environment in the secondstage. The empirical results of an analysis of European and Australian airports over a10 year timeframe reveal that under relatively non-competitive conditions, public airportsoperate less cost efficiently than fully private airports. Irrespective of ownership form,regulation is necessary to emulate competitive forces thus pushing airport managementtowards cost efficiency and reasonable pricing policies. Under potential regional or hubcompetition, economic regulation inhibits airports of any ownership form from operatingand pricing efficiently. Although public and fully private airports operate equally efficientlyin a competitive setting, private airports still set higher aeronautical charges. Furthermore,mixed ownership forms with a majority public holding are neither cost efficient nor lowprice, irrespective of the level of competition.

� 2014 Published by Elsevier Ltd.

1. Introduction

Historically, airports were mostly deemed state-owned entities with the objective to provide and operate infrastructurefor airlines. Frequently viewed as natural monopolies with large economies of scale, airports were subject to economicregulation in order to prevent abuse of market power. However, the nature of the airport industry has changed over the lasttwo decades. Moving away from viewing the airport as a public utility, airports have begun to operate as modern enterprisespursuing commercial objectives. A number of privatization processes have been actively promoted by governments with theproclaimed intention of reducing government involvement and increasing airport productivity and innovation. However,given the assumed profit-maximizing behavior of private companies working in a natural monopolistic environment, themajority of privatized airports in Europe remain subject to economic regulation (Gillen, 2011).

Whilst some studies have analyzed the separate impact of ownership form, regulatory regime and level of competitionfrom nearby airports on efficiency and the level of airport charges, none have examined their joint effects. In other words,the literature has yet to discuss whether the deregulation of the airline industry and changes in airport ownership andmanagement has affected the competitive situation and cost efficiency to the extent that the benefits of economic regulation

.: +972 2

N. Adler, V. Liebert / Transportation Research Part A 64 (2014) 92–109 93

are potentially unnecessary. For example, deregulation has led to increased competition between gateway hubs (e.g. Frank-furt and Amsterdam) and former military airports have opened to serve low cost carriers within the catchment area of exist-ing airports (e.g. Hahn in Germany), substantially changing the downstream airline market and potentially impacting theairport market too. Furthermore, as a result of increasing commercialization, many airports have augmented their revenuesfrom non-aeronautical sources in order to reduce aviation charges and attract additional airlines and passengers to their air-port (Zhang and Zhang, 2010). The aim of this research is therefore to analyze the impact of the structural changes in theaviation markets on airport efficiency and aeronautical charges in order to further our understanding of the most appropriateownership form and regulatory regime given the level of potential regional and hub competition.

Three well-documented quantitative methods have been applied to analyze the productivity and efficiency ofgovernment and private enterprises. A non-parametric, index number approach has been used to measure total factorproductivity (Caves et al., 1982), however this approach requires input and output prices and quantities which are notalways available. Parametric stochastic frontier analysis (SFA) assesses efficiency utilizing regression analysis anddisentangles unobservable random error from technical inefficiency (Aigner et al., 1977; Meeusen and van den Broeck,1977) based on assumptions as to the distributional forms of the production and efficiency functions and error term.Non-parametric data envelopment analysis (DEA), based on linear programming, categorizes data into efficient andinefficient groups hence produces weaker results than those of SFA, but does not require assumptions with respect to thefunctional form therefore has been chosen for the purposes of this study. Airport studies of efficiency utilizing all threeapproaches are reviewed in Liebert and Niemeier (2010).

Various environmental variables that are beyond managerial control at least in the short-term may affect the DEAefficiency estimates. Previous research argues that airport characteristics such as hub status or traffic structure, outsourcingpolicies, regulatory procedures and ownership structure all may contribute to airport efficiency (Gillen and Lall, 1997; Oumet al., 2006). Banker and Natarajan (2008) demonstrate that two-stage procedures in which DEA is applied in the first stageand regression analysis in the second stage provide consistent estimators and outperform parametric one- or two-stageapplications. Published airport studies have applied simple ordinary least squares (Yuen and Zhang, 2009), Tobit regression(e.g. Gillen and Lall, 1997) and truncated regression (Barros, 2008) for this purpose. A recent debate in the literaturediscusses the most appropriate second stage regression model to be applied when investigating DEA efficiency estimates.Simar and Wilson (2007) argue that truncated regression, combined with bootstrapping as a re-sampling technique, bestovercomes the unknown serial correlation complicating the two-stage analysis. Banker and Natarajan (2008) conclude thatsimple ordinary least squares, maximum likelihood estimation or Tobit regression dominate other alternatives. Combiningthe arguments of Simar and Wilson (2007) and Banker and Natarajan (2008), we apply robust cluster regression in order toaccount for the correlation across observations. Furthermore, in order to consider the panel structure of our sample, therandom effects model is applied which allows us to consider time-invariant environmental variables.

The second stage analysis of this research considers the impact of efficiency on ownership form, economic regulation andlevels of local and hub competition amongst other factors. Such an analysis contributes to the search for the more desirablecombinations and may indicate whether potential competition from nearby airports or gateway competition replaces theneed for economic regulation. In addition, we examine the combined impact on aeronautical revenues generated from pas-sengers, cargo handling and movements in order to further our understanding regarding the joint impacts on efficiency andpricing.

The dataset in this research consists of European and Australian airports in order to include a sufficiently heterogeneoussample with respect to ownership structure, regulatory mechanism and competitive environment. The empirical resultsreveal that under relatively non-competitive conditions, airports should be regulated to encourage cost efficiency anddual-till price-cap regulation appears to be preferable to other forms of regulation. Confirming theoretical arguments byArmstrong and Sappington (2006), privately owned and regulated airports operating under monopolistic conditions tendto be more efficient than publicly owned and regulated airports. Furthermore, unregulated airports operating under monop-olistic conditions irrespective of ownership form are likely to set higher aeronautical charges than those that are regulated.On the other hand, the existence of potential gateway or regional competition replaces the need for economic regulation,thereby supporting the argument of Vickers and Yarrow (1991) that competition rather than privatization is the key driverof efficiency. Nevertheless, the level of competition in the airport market has not proven to be sufficient to transfer theefficiency gains to consumers. Although unregulated, majority privately-owned and fully private airports tend to set loweraeronautical fees than those facing no competition, they charge higher landing based and passenger related fees than publicairports that operate equally efficiently under competitive conditions.

The paper is organized as follows: the theoretical and empirical literature discussing ownership form, economicregulation and competition is presented in Section 2, the methodology and model specifications are introduced in Section 3,the dataset for the two stage analysis is discussed in Section 4 and the results are presented in Section 5. Conclusions anddirections for future research are suggested in Section 6.

2. Competition, regulation and ownership form

Neoclassical theory states that under monopolistic conditions firms generally seek to maximize their profits by limitingoutput. The introduction of competition may lead to increased productive and allocative efficiency as a result of lower prices

94 N. Adler, V. Liebert / Transportation Research Part A 64 (2014) 92–109

and higher outputs. Consequently, social welfare may increase when market conditions exist (Leibenstein, 1966). Onthe other hand, network utilities providing substantial infrastructure may be natural monopolies in that one large firmmight produce at lower costs in which case the introduction of competition is not desired. According to Posner (1974), inorder to encourage efficiency and avoid abuse of market power, the natural monopolist ought to be subject to economicregulation.

In Europe1, airport charges have traditionally been regulated according to a rate of return or cost-plus principle as is true forthe majority of German airports (Reinhold et al., 2010). Littlechild (1983) proposes an incentive based, price-cap regulation inorder to solve the likelihood of over-investment, as described by Averch and Johnson (1962). Both cost-based and incentivebased pricing mechanisms may be applied according to a single-till or dual-till approach. Under the single-till approach, bothaeronautical (landing, passenger and aircraft parking charges, etc.) and non-aeronautical (food, retail and car parking services,etc.) revenues are constrained simultaneously. For example, Brussels airport is limited by single-till, rate-of-return regulationwhereas in the UK, the price-cap single-till approach is applied. This is similar to the North American system in which airportsare constrained to not-for-profit business models. In contrast, the dual-till approach regulates aeronautical revenues alone inorder to constrain only those activities with a monopolistic server. Whereas Vienna and Hamburg airports were moved froma cost-based to an incentive based regulation via a price-cap approach, unlike the UK model they follow a dual-till mechanism(Gillen and Niemeier, 2008). Shleifer (1985) proposes yardstick competition as an alternative approach to stimulate efficiencybased on a benchmarking process. Yardstick competition has since been applied to the British water and railway industries buthas yet to be applied to airports. To the best of our knowledge, the Dublin Airport Authority is the only European example thatattempted to implement yardstick competition. However, it was highly criticized by airport management for identifyinginappropriate peer airports and the attempt was eventually abandoned (Reinhold et al., 2010).

Over the years, the traditional perspective of airports being a natural monopoly has gradually changed as a result of thederegulation of the downstream aviation industry (Tretheway and Kincaid, 2010). Today, competition for airport servicescovers a multiplicity of markets including (1) a shared local catchment area, (2) connecting traffic through regional hubsand international gateways and (3) alternative modes of transport such as high speed rail in the medium distance markets.Although some degree of competition between airports may occur, we assume that the airport market structure can bebetter described as imperfect. Airport management often choose to serve different markets to their potential competitorsin the nearby catchment area in order to limit the degree of competition. For example, London-Heathrow is Europe’s largestgateway hub and offers different services to the airport in London-Luton which tends to serve short haul destinations withinEurope via low cost carriers. Amongst the empirical literature, only Yuen and Zhang (2009) examine the effects of regionalcompetition utilizing a Chinese airport dataset and conclude through an OLS regression that airports operating in a locallycompetitive environment tend towards efficiency. However, the outcome of a Tobit regression found competition intensityto be insignificant.

In the theoretical literature, the debate as to the necessity for and type of airport regulation has proven rathercontroversial. Oum et al. (2004) analytically assess the impact of economic regulation on the total factor productivity(TFP) of airports. They reveal that price-cap dual-till regulation provides the strongest incentives towards improving TFPdue to full exploitation of concessionary profits. In order to optimize social welfare, Czerny (2006) reveals in his analyticalmodel that for uncongested airports, single-till regulation controls profits more effectively than dual-till regulation. Yang andZhang (2011) recently extended the research by Czerny (2006) to congested airports. They argue that dual-till regulationyields higher welfare at significantly congested airports, provided aeronautical charges cover costs. Starkie (2002) obviatesthe need for economic regulation arguing that demand complementarities across aeronautical and terminal activities willprevent airports from abusing market power. In particular, airports generating additional revenues from non-aeronauticalactivities are more likely to lower their charges in order to attract airlines and greater passenger throughput, thusmaximizing their commercial revenues. If airports need to be subject to regulation, he suggests that dual-till regulation ispreferable irrespective of the level of congestion (Starkie, 2001).

The impact of regulation on efficiency and airport pricing has been empirically assessed in Barros and Marques (2008),Oum et al. (2004), Bel and Fageda (2010) and Van Dender (2007). Barros and Marques (2008) conclude that incentive basedprice-caps generally encourages cost efficiency compared to rate of return approaches. Oum et al. (2004) empirically confirmtheir analytical findings and conclude that dual-till regulation generally tends to improve economic efficiency. However, theapplication of dual-till approaches in combination with rate of return regulation leads to a complex cost allocation problem.With respect to pricing, Bel and Fageda (2010) analyze a European sample of airports and conclude that competition withnearby airports and other transport modes is likely to decrease the potential to abuse market power. Furthermore, private,unregulated airports generally charge higher prices than public, regulated airports. They also conclude that whether or notan airport is regulated is more important than the form or scope with respect to the level of aeronautical charges. Van Dender(2007) assesses the US market and finds that airports under regional competition charge lower aeronautical fees. He alsoargues that slot-constrained airports are likely to charge higher aeronautical fees, which are explained by the airportmanagement’s ability to capture scarcity rents.

Based on the theory with regard to the optimal form of economic regulation, we formulate the following hypothesis:

1 Gillen and Niemeier (2008) provide a comprehensive overview of the current economic regulation at European airports.

N. Adler, V. Liebert / Transportation Research Part A 64 (2014) 92–109 95

Hypothesis 1 (Form of regulation).

(a) When airports are subject to economic regulation, a dual-till incentive approach best aligns the management towardsproductive efficiency.

(b) The optimal scope of regulation with respect to aeronautical charges depends on the level of congestion, thus uncon-strained airports should be regulated according to the incentive, single-till approach and congested airports accordingto the incentive, dual-till regulation.

A wave of airport privatizations began in the late eighties in the UK with the stated aim of reducing government involve-ment, minimizing costs and maximizing productivity. As a result of successful initial public offerings and increasing shareprices, many European countries began to partially privatize their airports in the mid-nineties (Gillen and Niemeier,2008). The theoretical literature discussing the impact of privatization on efficiency and social welfare has led to an inter-esting debate. Sappington and Stiglitz (1987) argue that the transaction costs of government intervention are lower underpublic ownership. The government is better informed and more capable of regulating state-owned firms. With respect tocost efficiency, Armstrong and Sappington (2006) argue that regulated public enterprises facing monopolistic conditionstend to enjoy weaker budget constraints because the government may cover any losses. This was tested by Martín andSocorro (2009) who conclude that public airports not subject to budget restrictions tend to charge prices below their mar-ginal costs. Additionally, Boyko et al. (1996) argue that public ownership may lead to excessive employment. Hence, thereappears to be a trade-off between efficiency and welfare optimization: whereas public companies reduce the problem ofinformation asymmetry thereby lowering costs, they also face weaker budget constraints resulting in less cost efficient oper-ations. Under competitive conditions, Caves and Christensen (1980) reveal empirically that public and private companiescould operate equally efficiently. They argue that ownership per se does not cause cost inefficiency rather the lack of effec-tive competition. On the other hand, Shapiro and Willig (1990) argue that managers of public firms pursue personal targetsrather than maximizing social welfare. Consequently, we test the following hypothesis with regard to the combined effectsof ownership, regulation and competition:

Hypothesis 2 (Interaction of ownership, regulation and competition).

(a) Under monopolistic conditions: (i) fully private airports operate more cost efficiently than their public counterpartsand (ii) regulation does not encourage greater cost efficiency or lower airport charges at private airports due to infor-mation asymmetries.

(b) Under imperfect competitive conditions: (i) public and fully private airports operate equally efficiently when unreg-ulated and (ii) irrespective of ownership form, economic regulation restricting aeronautical charges is unnecessary toengender cost efficiency.

The emergence of partially privatized business models further complicates the debate as to the effects of ownership onproductivity. Boardman and Vining (1989) review the effects of mixed ownership structures based on theoretical argumentsand empirical studies. They conclude that large, industrial, partly privatized firms perform in a less productive and profitablemanner than their fully private counterparts, which may be caused by the public and private shareholders’ differing objec-tives. Empirical studies that attempt to assess the effects of ownership on the efficiency of airports are so far rather inconclu-sive. Parker (1999) estimates the technical efficiency of the BAA airports pre- and post-privatization. No evidence is found thatcomplete privatization leads to improved technical efficiency and he concludes that the UK government’s golden share limitsthe impact of capital market pressures. Furthermore, he argues that BAA remained subject to economic regulation henceincentives to operate more efficiently are distorted as a result of government intervention. In contrast, Yokomi (2005) findsthat the BAA airports exhibited positive changes in efficiency and technology as a result of the privatization. It should be notedthat commercial growth after privatization was substantial however this activity was not considered in Parker’s analysis.

The effects of different ownership forms on efficiency were also analyzed but again the results have not reached clearconclusions. Barros and Dieke (2007) analyze Italian airports and reveal that private airports operate more efficiently thantheir partially private counterparts. Lin and Hong (2006) find no connection between ownership form and efficiency. Oumet al. (2006, 2008) distinguish between public airports owned by public corporations and those owned by more than onepublic shareholder (multilevel). Oum et al. (2006) reach the conclusion that the productivity of a public corporation isnot statistically different from that of a major private airport. However, airports owned by a minority private shareholdingor multiple government involvement operate significantly less efficiently than other ownership forms. Oum et al. (2008)conclude that airports with major private shareholders are more efficient than public airports, particularly those with amajor public ownership structure.

Hypothesis 3 (Public–private partnerships).

(a) Under monopolistic conditions, airports owned by a public–private partnership operate less efficiently than fully pri-vate airports, where minority private holdings are less efficient than majority private holdings.

96 N. Adler, V. Liebert / Transportation Research Part A 64 (2014) 92–109

(b) Under competitive conditions, unregulated mixed ownership is the least efficient ownership form.

Whereas empirical research on airport efficiency to date has been limited to analyses of the individual effects of owner-ship, regulation and competition on efficiency, the joint impacts may be of great importance. Button and Weyman-Jones(1992, p. 440) clearly state the ‘‘degree of competitiveness in a firm’s market, the extent to which it is incorporated as partof a public-sector bureaucracy, and the nature of the regulatory regime under which a firm operates are all primary sourcesof possible X-inefficiency’’. In addition to the cost efficiency perspective, the analysis also considers the aeronautical chargingpolicy. Consequently, we assess the combined impact of ownership structure and economic regulation (or lack thereof) givenrelevant levels of potential, local and hub competition. We argue that the choice of ownership form and the regulatory pro-cedures instituted are clearly within the bounds of public policy initiatives over the medium term, whereas the competitiveenvironment remains more costly to change.

3. Methodology and model specification

The following section presents the data envelopment analysis model which we apply in the first-stage analysis in order toaccount for both the desired equi-proportional reductions in all inputs and any remaining slacks.

We then discuss the second-stage regression specifications in which the DEA efficiency estimates and the level of aero-nautical charges are regressed against environmental factors.

3.1. Data envelopment analysis

DEA is a non-parametric method of frontier estimation that measures the relative efficiency of decision-making units(DMUs) utilizing multiple inputs and outputs. DEA accounts for multiple objectives simultaneously without attaching ex-ante weights to each indicator and compares each DMU to the efficient set of observations, with similar input and outputratios, and assumes neither a specific functional form for the production function nor the inefficiency distribution. DEAwas first published in Charnes et al. (1978) under the assumption of constant returns-to-scale and was extended byBanker et al. (1984) to include variable returns-to-scale. This non-parametric approach solves a linear programming formu-lation per DMU and the weights assigned to each linear aggregation are the results of the corresponding linear program. Theweights are chosen in order to present the specific DMU in as positive a light as possible, under the restriction that no otherDMU, analyzed under the same weights, is more than 100% efficient. Consequently, a Pareto frontier is attained, marked byspecific DMUs on the boundary envelope of input–output variable space.

The weighted additive model (Charnes et al., 1985; Lovell and Pastor, 1995), chosen for its units and translation invari-ance properties, reflects all inefficiencies identified in the inputs. The input oriented model is chosen because we assume thatairport managers control operational costs and to a lesser extent airport capacities, but have relatively minor control overtraffic volume. By comparing n units with q outputs denoted by Y and r inputs denoted by X, the efficiency measure for air-port a is expressed as in model (3.1).

Maxs;r

wts

s:t: Yk� s ¼ Ya

� Xk� r ¼ �Xa

ek ¼ 1k; s; r P 0

ð3:1Þ

where k represents a vector of DMU weights chosen by the linear program, wt a transposed vector of the reciprocals of thesample standard deviations, e a vector of ones, r and s vectors of input and output slacks respectively and Xa and Ya the inputand output column vectors for DMUa respectively. Hence DMUa, the airport under investigation, is therefore efficient if andonly if all input slacks equal zero. Variable returns-to-scale is assumed (ek = 1) because the sample dataset consists of air-ports of substantially different sizes, ranging from 0.5 million passengers to more than 50 million per annum. It should benoted that the objective function value of Eq. (3.1) lies between zero and infinity with a DMU deemed efficient when thesum of slacks equal zero. In order to interpret the coefficients obtained from the second-stage regression as percentages,the efficiency scores were normalized to a range from zero to one according to a slack-based measure (Tone, 2001).

3.2. Second-stage regression

The raw inefficiency scores estimated in the first stage may be explained by factors beyond managerial control in additionto managerial inefficiency. In order to conduct hypothesis testing, regression analyses is often applied in a second stage inwhich the DEA efficiency estimate is regressed against a set of potential environmental variables. As a result, the net inef-ficiency estimators (ui) are computed after taking the exogenous variables into account. Banker and Natarajan (2008) andSimar and Wilson (2007) independently review appropriate forms to conduct second-stage regressions of DEA estimateswhich lead them to different conclusions. Based on Monte Carlo simulations, Banker and Natarajan (2008) argue that

N. Adler, V. Liebert / Transportation Research Part A 64 (2014) 92–109 97

ordinary least squares, Tobit (censored) regression and maximum likelihood estimation in the second-stage outperform one-stage and two-stage parametric methods. Simar and Wilson (2007) argue that the majority of empirical two-stage studies donot properly define the data generating process. The efficiency estimates are likely to be serially correlated via the efficiencyfrontier hence the error term, eit, will also be serially correlated thereby violating the common assumption that the errors areidentically and independently distributed. They also state that any bias in the efficiency estimate is ignored and will be auto-matically included in the error term. Consequently, Simar and Wilson advocate truncated regression in the second stage,which removes the efficient units from the sample. The problem with this approach is that we would then ignore all airportsdeemed to be lying on the efficient frontier, yet we are searching for the most appropriate form of ownership and regulationgiven the competitive environment.

Drawing from both approaches, we apply a linear regression model to explain the DEA efficiency estimates thereby fol-lowing the Banker and Natarajan (2008) results. In order to account for the issues identified in Simar and Wilson (2007), weapply a robust between-cluster approach which is clearly described in Williams (2000). The sample will be clustered in orderto overcome the limitation that the error terms are not independently distributed at the airport level. The airport observationis repeated for multiple periods hence the correlation within the cluster may be different to that across clusters, which theapproach estimates ensuring robust variance estimators. Accounting for heteroscedastic robust standard errors as proposedby White (1980), we construct unbiased t-tests and confidence intervals hence attempt to solve the limitation that the errorterm is not necessarily identically distributed. Consequently, in order to account for the panel structure2 of our sample, arandom effects model has been applied as shown in Eq. (3.2).

2 Forordinarcomparsame. A

3 Accprofits rsome fo

4 It ssuch as

5 The

ln yit ¼ aþ bkxitk þ ui þ eit ð3:2Þ

where lnyit represents the efficiency estimate for airport i at time t in the first set of regressions and the aeronautical rev-enues respectively for the second set. In order to apply linear regression according to Banker and Natarajan (2008), thedependent variable is transformed to logarithmic form. The k regressors, xitk, represent the exogenous environmental vari-ables in our analysis, ui indicates the random effect at the airport level and eit represents the error term. This model accountsfor time-invariant covariates because the form of ownership, regulation and degree of competition remained constant overtime for the majority of airports. The robust cluster approach accounts for correlation within a cluster thus we consider theimpact of analyzing the same airport over time. An alternative approach would be to include an instrumental variable whichmay ameliorate any limitations related to endogeneity, however the collection of appropriate data is problematic and maycause additional bias in the case of weak instruments, namely were the instrumental variable to be only loosely correlatedwith the independent variables.

4. Dataset

In this section we discuss the variables collected for the first-stage relative efficiency analysis and then the environmentalvariables included in the second-stage regression. Appendix A lists the complete set of airports in the sample, which includes48 European airports of which half are located in Germany and the United Kingdom. To ensure a sufficiently heterogeneousdataset with respect to the form of ownership, economic regulation and level of local and hub competition, we also includethree fully privatized Australian airports, Melbourne, Perth and Sydney, which are neither ex-ante regulated3 nor located in acompetitive environment due to the great distances across the continent. The pooled data consists of an unbalanced set of 398observations covering the time period between 1998 and 2007.4 Given the rather short time period of on average 7 years perairport and limited technological changes for such an industry, we construct a single intertemporal frontier from the entire dataset according to Tulkens and van den Eeckaut (1995).

4.1. Variables in the first-stage efficiency analysis

For the first-stage efficiency analysis, three inputs and four outputs were collected as summarized in Table 1. Similar datahas been collected for multiple airport analyses as documented in Liebert and Niemeier (2010).5 The operating inputs consistof staff costs and other operating costs, including materials and outsourcing. It is also necessary to consider capital however thisvariable is extremely problematic since it is often unreported. Furthermore, when the dataset covers more than one country, themonetary measurement of physical capital creates difficulties due to different national accounting standards and depreciationmethods or periods across countries. For example, the airports of the British Airports Authority depreciate their runways over

purposes of sensitivity analysis, we also conducted pooled truncated regressions, ordinary least squares and (censored) Tobit regressions. Undery least squares and Tobit regression we obtain very similar results. Under truncated regression, slightly lower statistical significance was founded to the other functions due to the removal of all relatively efficient observations from the sample however the qualitative conclusions remain thell the results are presented in Appendix C.ording to the Trade Practices Act 1974, the Australian Competition & Consumer Commission (ACCC) is responsible for the monitoring of prices, costs andelated to aeronautical and airport car parking services and facilities in Adelaide, Brisbane, Melbourne, Perth and Sydney. Hence the airports experiencerm of ex-post regulation (Forsyth, 2004).

hould be noted that we have attempted to maximize the number of observations available but could not include airports belonging to airport groupsAdP and Aena, due to the lack of disaggregated information.data is available to any interested reader by contacting Prof. Hans-Martin Niemeier, the leader of the German Airport Performance Project.

Table 1Variables in analysis (DEA).

Variable Description Average Standard deviation Maximum Minimum Source

Staff costs Wages and salaries, other staffcosts (2000 = 1 and US$ = 1)

63,654,765 120,554,070 1,080,756,267 3,655,825 Annual Reports

Other operating costs Costs of materials, outsourcing andother (2000 = 1 and US$ = 1)

84,811,284 117,603,464 725,987,196 3,631,353 Annual Reports

Declared runwaycapacity

Number of movements per hour 46 20 110 15 IATA (2003), AirportCoordinator

Passengers Annual passenger volume (onlyterminal passengers)

11,091,246 12,761,170 67,673,000 480,011 Annual Reports

Cargo Metric tons per year (truckingexcluded)

172,922 366,881 2,190,461 0 Annual Reports

Air transportmovements

Annual number of commercialmovements

132,482 109,190 492,569 19,397 Annual Reports

Non-aeronauticalrevenues

Revenues from concessions ownretail and restaurants, rents,utilities and ground handlingactivities (2000 = 1 and US$ = 1)

124,647,578 186,748,031 1,167,377,411 6,194,408 Annual Reports

98 N. Adler, V. Liebert / Transportation Research Part A 64 (2014) 92–109

100 years whereas the airports operated by the Aéroports de Paris apply a 10–20 year depreciation rule (Graham, 2005).Consequently, physical data such as the number of runways, gates or check-in-counters and terminal size are often collectedfor cross-border studies as a proxy for capital (Gillen and Lall, 1997; Pels et al., 2003). In this study, we include declared runwaycapacity as a proxy for capital, which is defined as the capacity constraint on the number of departure and arrival movementsper hour. Declared runway capacity is negotiated twice a year in agreement with airport stakeholders and is primarily used toavoid congestion at schedule facilitated airports and to allocate slots at coordinated airports (IATA, 2010). Compared to purephysical information, the capacity measure considers the runway configuration, weather impacts and/or environmental restric-tions thus accounting for bottlenecks in the system that may be amenable to shorter term solutions. Furthermore, this param-eter restricts the airport’s aeronautical revenues hence reflects a real measure of capacity. Consistent terminal data proved verydifficult to collect hence has been excluded in this study, however runway capacity is highly correlated with terminal capacitytherefore this omission should not greatly impact the results.

On the output side, the annual traffic volume is represented by the number of passengers, commercial air transport move-ments and tons of cargo (trucking was excluded). Each of these throughputs generates their own costs and is priced sepa-rately, therefore need to be included individually. The fourth output variable captures revenues from the non-aeronauticalactivities, including concessions, car parking and rent. In addition to the traditional income sources, non-aeronautical reve-nues also include revenues from labor intensive ground handling activities. Ignoring this operation on the output side wouldotherwise bias the results since the input data could not be adjusted to exclude this service. At least theoretically we shouldbe able to compare all three airport models, namely airports that produce ground handling services in-house and have rel-atively higher employee costs and requisite revenues, those who outsource which appear in the other costs category andtheir respective revenues and the third case in which airports do not provide the service nor earn revenue beyond perhapsa nominal fee from third party contractors. All financial data is deflated to the year 2000 and adjusted by the purchasingpower parity according to the United States dollar in order to ensure comparability across countries. We are aware that mul-tiple airport systems, such as BAA, may pay higher salaries than individual airports however, due to lack of data we are notable to capture this effect and note that our efficiency measure will not separate technical inefficiency from labor inefficien-cies. In addition, the data has been normalized by the standard deviation which ensures that all inputs are considered equallywithin the additive model.

4.2. Variables in the second-stage regression

Variables describing ownership structure, economic regulation and the level of potential hub and local competition havebeen collected for this research in addition to specific airport characteristics and information on managerial strategies. Allfactors are at least in the short-term beyond managerial control yet may impact the inefficiency measurement process. Fur-thermore, these variables enable us to capture the likelihood that publicly owned airports, as opposed to their privatelyowned counterparts, may be faced with different objectives to that of profit maximization. All data is expressed in the formof categorical variables. To further assess efficiency changes over the review period and remove time-related effects, cate-gorical variables on the financial years (TIME)6 have been included.

We note that revenue source diversification exploits demand complementarities across aeronautical and non-aeronauti-cal services and may impact airport efficiency, therefore we consider the share of non-aeronautical revenues (NA) generated atan airport. Airports are split between those that earn less than 50% of their revenues from non-aeronautical activities and

6 Financial data has been adjusted by the different reporting periods.

Table 2Combination of environmental variables analyzed.

Weak local competition Potential local or hub competition

No. of airports No. of obs. No. of airports No. of obs.

Public No ex-ante regulation 6 37 5 36Ex-ante regulation 9 59 7 60

Minor private No ex-ante regulation 0 0 2 8Ex-ante regulation 2 13 3 24

Major private No ex-ante regulation 1 2 2 19Ex-ante regulation 2 9 2 15

Fully private No ex-ante regulation 5 33 6 53Ex-ante regulation 2 6 3 24

N. Adler, V. Liebert / Transportation Research Part A 64 (2014) 92–109 99

those that exceed this share.7 The threshold of 50% was chosen such that a rich set of airports exists in the two categories andsensitivity analyses show no change in the results when reducing the threshold to 40% or increasing it to 60%.

High levels of delay (D) may cause an efficiency overestimation as suggested in Pathomsiri et al. (2008). In the Europeanmarket, airport related delays per movement is not publicly available data, hence we include a categorical variable drawnfrom the ranking of the most delayed airports (departure and arrival) as reported by the European air traffic control pro-vider.8 An airport appearing on the list of the Top 50 most delayed airports in Europe was categorized accordingly.9 Since delaysoccur for many reasons, we also consider runway utilization (RWY) in order to directly assess the effects of congestion, which areexpected to positively impact efficiency.10 The variable was calculated based on annual air transport movements divided byestimated annual declared capacity.11 We define three categories including (1) airports with less than 50% runway utilizationindicating under-utilization, (2) airports between 51% and 90% runway utilization and (3) airports achieving more than 90% run-way utilization indicating the likelihood of a congested facility.

For the aeronautical charge based regression, in which we explain the annual aeronautical revenues of an airport, weinclude the log forms of air transport movements (ATM)12 and average aircraft size (AC) in order to capture the airport sizeand substantial differences in fleet mix that exist within the sample. Hence, after accounting for these variables we reduceany size effects that may influence our results.

Ownership (O) form is defined according to (1) fully public airports, (2) public–private airports with minority private hold-ings (less than 50%), (3) public–private airports with majority private holdings (above 50%) and (4) fully private airports.

The form of economic regulation (R) has been categorized according to (1) no ex-ante regulation,13 (2) single-till cost-plusregulated, (3) dual-till cost-plus regulated, (4) single-till price-cap regulated and (5) dual-till price-cap regulated airports.Furthermore, in order to test the analytical findings of Czerny (2006) and Yang and Zhang (2011), we include a dummy (I)to represent the combinations of single or dual till price-cap regulation at congested and uncongested airports respectively.Unfortunately, this level of refinement is not possible in the combined model due to an insufficient level of data hence weaggregated the classification to ex-ante unregulated and regulated airports in the combined environmental modeling approach.

Regional competition (C) has been defined as the number of commercial airports with at least 150,000 passengers perannum within a catchment area of 100 km around the airport. The radius of the catchment area has been defined in line withBel and Fageda (2010). In addition, competition between gateway airports is considered. Consequently, weak competition isdefined at the regional level as no more than one additional airport within the catchment area and potential competition as alocation with at least two airports within the catchment area. However, an airport possessing hub status that serves as aregional or international gateway will also be categorized as working in a competitive market, irrespective of their localcatchment area. Due to lack of information for the whole sample, we were not able to capture different product diversifica-tion strategies, such as low cost carrier traffic, which may limit the level of competition between closely located airports.Hence, our conservative measure indicates an upper level of likely competition across airports.

The random effects model regresses the logged form of airport cost efficiency (lneff) in Eq. (4.1) and aeronautical revenues(lnarev) in Eq. (4.2) against the following set of exogenous variables:

7 Rev8 The

system9 Aus

sufferin10 It is11 The12 The

multi-c13 For

ln effit ¼ aþ b1NAit þ b2Dit þ b3RWYit þ b4Oit þ b5Rit þ b6Cit þ b7TIMEt þ ui þ eit ð4:1Þ

enues generated from ground handling activities are not included in this computation which may reduce any potential endogeneity issues.measure reported by Eurocontrol does not capture airport-related delays, rather delays caused by many factors including airlines, the air traffic controland weather, which are beyond the control of the airport manager.tralian airports are included in the group of non-delayed airports for lack of further information. Reallocating these airports to be designated as thoseg from delay does not impact the final results of the analysis.also possible for airports not operating near their capacity to cause delays due, for example, to a lack of manpower.annual capacity has been estimated based on the declared hourly capacity obtained from the airport coordinator.number of passengers is highly correlated with the number of air transport movements (>94%) thus were not included in the regression due to issues of

ollinearity.simplicity we refer to airports subject to ex-post, standard, anti-trust regulation as unregulated airports.

100 N. Adler, V. Liebert / Transportation Research Part A 64 (2014) 92–109

ln arev it ¼ aþ b1 ln ACit þ b2 ln ATMit þ b3NAit þ b4Dit þ b5Iit � RWYit þ b6Oit þ b7Rit þ b8Cit þ b9TIMEt þ ui þ eit

ð4:2Þ

Table 2 presents the combinations and the number of airports and observations that belong to the different groups in thecombined model.

Consequently, the models assessing the combined effects of ownership, regulation and competition against cost efficiency(Eq. (4.3)) and aeronautical revenues (Eq. (4.4)) are as follows:

ln effit ¼ aþ b1NAit þ b2Dit þ b3RWYit þ b4OitRitCit þ b5TIMEt þ ui þ eit ð4:3Þ

ln arev it ¼ aþ b1 ln ACit þ b2 ln ATMit þ b3NAit þ b4Dit þ b5RWYit þ b6OitRitCit þ b7TIMEt þ ui þ eit ð4:4Þ

5. Empirical results

In the following section we first discuss the DEA efficiency results. The complete set of normalized DEA efficiency scores islisted in Appendix B. The second part of this section discusses the results of the regression analyses explaining the DEA effi-ciency estimates and aeronautical revenues, initially discussing the individual impacts and followed by the combined envi-ronmental regression approach. A complete set of regression results can be found in Table 3 and Appendix C.

5.1. Estimates from the first-stage efficiency analysis

The average efficiency score of the dataset obtained from the input-oriented, slack-based measure is 0.62 with 14% of allairports categorized as relatively efficient. The majority of airports exhibit an efficiency decrease over time, which provedsignificant according to the paired sample Wilcoxon sign rank test (p-value = 0.000). As a result of the general economicdownturn and the attacks on the World Trade Center in New York in 2001, the majority of airports experienced stable ordeclining traffic rates with disproportionate increases in staff and other operating costs. Increased security measures for bag-gage screening require the recruitment and training of specialized workers, expenses which have been covered at least par-tially by the airports (Vienna International Airport, 2004). Hence one could argue that the additional costs provide anincreased quality with respect to security.

Consistently efficient airports include Ljubljana and Malta that represent the smaller airports in the dataset, many of theAustralian airports, as well as the largest operators, Frankfurt and London-Heathrow. In Australia, domestic terminals areoften operated by incumbent airlines under long-term leases, thereby lowering maintenance and staff costs (Hooperet al., 2000). Frankfurt and London-Heathrow obtain reasonably high cost efficiency estimates over time. Both airports placegreat emphasis on cost efficiency with Heathrow attempting to minimize staff costs and Frankfurt tending to reduce otheroperating costs. The diverse strategies are not surprising given the different levels of outsourcing, including ground handlingprovision. Whereas Frankfurt provides ground handling services in-house, this operation has long been operated by airlinesand third-party providers at Heathrow. It should be noted that both airports are severely congested and require airsidecapacity expansions.

Except for Sydney, no airport consistently improved their relative efficiency scores over time. Between 2003 and 2007,Sydney increased its score from 0.56 to 1.00 which is mainly attributable to a large increase in non-aeronautical revenueswith fairly constant cost inputs. In contrast to Melbourne and Perth, a cost increase from the sale of the Ansett terminal backto Sydney’s airport management is not reported as the review period begins in 2003. However, the ACCC responsible for theprice monitoring of aeronautical charges and car parking fees at the top five Australian airports accused Sydney airport ofabusing market power. In March 2010, the ACCC reported that the airport had almost doubled car parking fees from 2008to 2009 (ACCC, 2010).

The airports of Athens, Budapest, Cologne-Bonn, Hanover, Leipzig, Lyon, Manchester and Munich appear to be the leastrelatively cost efficient airports in the sample. Athens underwent substantial capacity expansions and a new green-field loca-tion hence capacity utilization is low in comparison to reference airports such as Gatwick and Nice. Furthermore, the Germanairports of Cologne-Bonn and Leipzig suffer from excess airside capacities despite the extensive cargo operations resultingfrom their positions as the European hubs for UPS and DHL respectively. Hanover also suffers from excessively low capacityutilization and exhibits relatively high operating costs compared to its benchmarks. However, service quality indicators suchas congestion and delay are not included in the first stage analysis, hence we refer to these estimators as gross relative mea-sures to be adapted after considering these additional variables in the second-stage of the analysis.

5.2. Second-stage regression results

In this section we analyze the impact of additional variables on the DEA cost efficiency scores and aeronautical revenuesand also include time indicators as identified in Section 5.1. Table 3 presents the results obtained from the regression modelsintroduced in Section 3 with respect to the individual effect models (Eqs. (4.1) and (4.2)) and the combined models (Eqs. (4.3)and (4.4)). With respect to model (4.3), the monopolistic, minority private shareholding, regulated airport is defined as thebase case because it represents the least efficient group of airports in the dataset. The monopolistic, fully private, regulated

Table 3Second-stage regression results.

Dependent variable Individual effects Combined effects

DEA efficiency scores(log) (Model 4.1)

Aeronautical revenues(log) (Model 4.2)

DEA efficiency scores(log) (Model 4.3)

Aeronautical revenues(log) (Model 4.4)

(a) Airport characteristics and management strategiesAverage aircraft size (log) – 0.418 (0.094)*** – 0.438 (0.103)***

Number of air transport movements (log) – 0.841 (0.077)*** – 0.849 (0.059)***

Share of non-aviation revenues > 50% 0.101 (0.052)** �0.146 (0.034)*** 0.101 (0.050)*** �0.140 (0.023)***

Heavy delays �0.115 (0.046)*** �0.019 (0.020) �0.139 (0.048)*** �0.009 (0.015)Runway capacity utilization between 50%

and 90%0.127 (0.034)*** – 0.123 (0.037)*** 0.009 (0.015)

Runway capacity utilization > 90% 0.538 (0.124)*** – 0.615 (0.164)*** 0.133 (0.074)*

(b) Ownership, regulation and competition – individual effectsMinor private airport �0.180 (0.090)*** 0.067 (0.059) – –Major private airport 0.150 (0.129) 0.085 (0.046)* – –Fully private airport 0.095 (0.084) �0.046 (0.065) – –Heavy competition 0.003 (0.070) �0.052 (0.047) – –Cost-plus regulation, single-till �0.253 (0.082)*** 0.045 (0.061) – –Cost-plus regulation, dual-till �0.192 (0.078)*** �0.117 (0.062) * – –Price-cap regulation, single-till �0.183 (0.092)*** – – –Price-cap regulation, dual-till 0.030 (0.097) – – –Price-cap regulation, single-till

(uncongested)– �0.080 (0.073) – –

Price-cap regulation, single-till (congested) – 0.224 (0.095)*** – –Price-cap regulation, dual-till (uncongested) – �0.220 (0.050)*** – –Price-cap regulation, dual-till (congested) – 0.584 (0.089)*** – –

(c) Ownership, regulation and competition – combined effectsLow competition Public No regulation – – 0.349 (0.154)*** 0.408 (0.103)***

Regulation – – 0.316 (0.143)*** 0.335 (0.081)***

Minor private No regulation – – n/a n/aRegulation – – Base case 0.498 (0.108)***

Major private No regulation – – 0.250 (0.143)* 0.466 (0.099)***

Regulation – – 0.767 (0.194)*** 0.416 (0.091)***

Fully private No regulation – – 0.465 (0.166)*** 0.279 (0.028)***

Regulation – – 0.480 (0.181)*** Base case

Potential competition Public No regulation – – 0.589 (0.131)*** 0.218 (0.094)***

Regulation – – 0.081 (0.146) 0.389 (0.092)***

Minor private No regulation – – 0.322 (0.181)* 0.281 (0.104)***

Regulation – – 0.128 (0.154) 0.387 (0.087)***

Major private No regulation – – 0.646 (0.215)*** 0.351 (0.125)***

Regulation – – 0.433 (0.130)*** 0.373 (0.096)***

Fully private No regulation – – 0.564 (0.143)*** 0.329 (0.094)***

Regulation – – 0.358 (0.156)*** 0.484 (0.090)***

(d) Time trends1999 �0.012 (0.024) 0.025 (0.011)*** �0.157 (0.025) 0.034 (0.009)***

2000 �0.046 (0.035) 0.043 (0.013)*** �0.054 (0.037) 0.047 (0.011)***

2001 �0.126 (0.045)*** 0.083 (0.016)*** �0.133 (0.048)*** 0.093 (0.014)***

2002 �0.179 (0.056)*** 0.089 (0.016)*** �0.182 (0.058)*** 0.086 (0.018)***

2003 �0.234 (0.062)*** 0.114 (0.017)*** �0.239 (0.064)*** 0.112 (0.018)***

2004 �0.210 (0.061)*** 0.113 (0.021)*** �0.213 (0.063)*** 0.132 (0.023)***

2005 �0.261 (0.061)*** 0.174 (0.021)*** �0.258 (0.061)*** 0.173 (0.023)***

2006 �0.285 (0.065)*** 0.170 (0.029)*** �0.279 (0.065)*** 0.170 (0.031)***

2007 �0.294 (0.071)*** 0.205 (0.031)*** �0.279 (0.069)*** 0.204 (0.031)***

Intercept 0.307 (0.086)*** 2.846 (0.380)*** 0.730 (0.149)*** 2.370 (0.321)***

R2 0.5096 0.8767 0.6399 0.9001Sigma u 0.1852 0.1328 0.1371 0.1257Sigma e 0.1495 0.0632 0.1498 0.6336Rho 0.6055 0.8152 0.4561 0.7961Observations (n) 398 398 398 398

Note: Robust standard errors in parentheses; efficiency model clustered at airport level. Uncongested airports with a runway utilization < 90%, congestedairports otherwise.* Significance at 10% level.** Significance at 5% level.*** Significance at 1% level.

N. Adler, V. Liebert / Transportation Research Part A 64 (2014) 92–109 101

102 N. Adler, V. Liebert / Transportation Research Part A 64 (2014) 92–109

airport has been chosen as the base case in model (4.4) as an example of the least expensive airport in the sample set. Thetime trend dummies prove to be statistically significant across all regressions from 2001. Hence, after accounting for the effi-ciency decreases and revenue increases over time, we conclude that ownership form and regulation play an important role inexplaining efficiency and pricing differences across airports.

The results with respect to airport characteristics and management strategies reveal that airports earning more than 50%of their revenues from non-aeronautical sources achieve higher levels of efficiency and set lower aeronautical charges, sup-porting the results of Oum et al. (2006) and indicating a marginal contribution to cost efficiency of approximately 10%. Themodel of airport pricing reveals that the same airports charge on average 15% less for aeronautical activities than otherwise.This confirms the analytical findings of Zhang and Zhang (2010) in which it was argued that airports may reduce their aero-nautical costs in order to attract additional airlines and passengers. Delay and congestion have a statistically significantimpact on airport cost efficiency, as discussed in Pathomsiri et al. (2008). Delay impacts cost efficiency negatively in theregion of 12%. In contrast, efficiency significantly increases with runway capacity utilization. Congestion, proxied by capacityutilization above 90%, has a large positive impact of approximately 50% on airport efficiency compared to underutilized air-ports. Given that the positive impact of congestion is higher than the negative effect of delays, this would indicate the needfor service quality indicators written into contracts between airlines and airports or internalization through compensation toairlines and passengers for airport related delays. Furthermore, whereas delay was found to have no impact on aeronauticalrevenues, charges are significantly higher at congested airports thereby supporting the empirical outcome of van Dender(2007) that congested airports are in a position to exploit scarcity rents.

Reviewing the individual effects of ownership, regulation and competition in model (4.1), we see that mixed ownershipwith less than 50% of shares traded privately is significantly less efficient than other ownership forms. Furthermore, majorand fully private airports appear not to be statistically significantly different in terms of cost efficiency than their fully publiccounterparts, in line with Oum et al. (2006). Consequently, ignoring interactions between potential competition and regula-tion, the results indicate that fully public or private ownership operate equally efficient and behave similarly in setting theirprices. The importance of local and gateway competition is not statistically significant in either model. With respect to regu-lation, it would appear that unregulated airports generally operate more efficiently than their regulated counterparts, with thesingle exception being price-cap dual-till regulation. Cost-plus regulation appears to be the least appropriate form of eco-nomic regulation whether single-till or dual-till, reducing efficiency by 25% and 19% respectively. In addition, single-tillprice-caps also appear to be dominated by dual-till price-caps and standard ex-post anti-trust monitoring. Hence we accepthypothesis (1a) in which it is argued that a dual-till, incentive approach such as a price-cap would be the most preferable formof economic regulation, were it to prove necessary to implement. From the airport pricing model (4.2) we find that congested,single-till, incentive regulated airports charge lower aeronautical prices than its dual-till regulated counterparts. Hence weconfirm the argument of Beesley (1999) that single-till regulation of large, congested airports such as London Heathrow leadsto lower aeronautical charges due to large concession revenues, possibly further increasing congestion. On the other hand, wenote that at uncongested airports the incentive dual-till mechanism leads to lower charges than the single-till incentiveoption. These results support the argument by Starkie (2001) that regulated, uncongested airports set lower aeronauticalcharges in order to earn additional revenues from non-aeronautical sources. Hence we do not accept hypothesis (1b) forthe uncongested case and do not find support for the analytical findings of Czerny (2006) and Yang and Zhang (2011). Further-more, in contrast to the empirical outcome of Bel and Fageda (2010), we find that the form and scope of regulation may sub-stantially affect the level of aeronautical charges of an airport. In summation, ignoring the interaction between ownership,regulation and competition, the results reveal that airports should either be unregulated or, if subject to economic regulation,dual-till price-capped in order to encourage cost efficiency. Airports should either be fully privatized or remain entirely in pub-lic hands because the mixed models result in greater inefficiency and higher aeronautical prices. Finally, the level of local orhub competition does not affect cost efficiency or the pricing behavior of an airport in the individual effects model.

Under the combined models (4.3) and (4.4), the impact of potential competition becomes clearer. Under weak compet-itive conditions, defined as at most one airport within the catchment area, we find that privatized airports with at least 50%of the shares in private hands are the most cost efficient ownership form. In comparison to airports with minority privateownership (the base case), the major or fully privatized airports are on average 60% more efficient when regulated and15% more efficient when unregulated also suggesting that economic regulation is desirable. Referring back to the individualregression model results, dual-till price-cap regulation would appear to be the most preferable instrument. ConfirmingArmstrong and Sappington (2006), fully private airports operate more cost efficiently than their public counterparts in amonopolistic setting as argued in hypothesis (2ai). Consequently, monopolistic public airports may exhibit excessiveemployment and face lower budget constraints than major and fully private airports. Furthermore, unregulated airports irre-spective of the ownership form charge higher aeronautical prices than their regulated counterparts, suggesting that regula-tion is also essential in order to prevent exploitation of market power. Hence economic regulation may improve both welfareand cost efficiency of public and private airports, therefore we reject hypothesis (2aii). Publicly owned and regulated airportsperform slightly less efficiently than their unregulated counterparts. Furthermore, charges at regulated public airportsexceed those of regulated fully private airports. This may be explained by the fact that regulated, monopolistic, public air-ports in this sample are all subject to cost-plus regulation whereas regulated, monopolistic, fully private airports follow aprice-cap regulation approach. Interestingly, the price difference between regulated and unregulated fully private airportsis fairly high. Considering that the efficiency level between unregulated and regulated fully private airports is fairly similar,this result indicates the exploitation of market power under this ownership form.

N. Adler, V. Liebert / Transportation Research Part A 64 (2014) 92–109 103

In a potentially competitive environment, unregulated, purely public airports and major or fully privatized airports operatein an equally cost efficient manner. It is clear that such airports do not require economic regulation to maintain cost efficiencyas compared to airports operating in a weakly competitive environment. Major and fully private unregulated airports performto the order of 20% more cost efficiently than their regulated counterparts. Apart from private airports with a minority of sharesin private hands, airports located in a potentially competitive environment generally operate more efficiently than those oper-ating in weakly competitive markets. Similarly we find that regulation generally increases aeronautical prices. Consequently,we accept hypothesis (2b). Government intervention would appear to incur high transaction costs and is required to emulatethe competitive environment when missing but is very expensive when such conditions already exist in the market. Althoughunregulated public and major or fully private airports may operate equally efficient, the level of charges between these airportsvaries considerably. Private airports tend to charge higher prices compared to their public counterparts. Finally, we find minorprivate airports to be the least efficient ownership form, irrespective of the level of competition, hence we reject hypotheses(3a) and (3b). In summary, the results obtained from the combined approach in models (4.3) and (4.4) reveal that the level ofcompetition has a substantial impact on the appropriate form of ownership and the need for economic regulation. As shown inTable 3, airports should be fully privatized in order to encourage cost efficiency and dual-till incentive regulation should beapplied in order to avoid excessive pricing under weakly competitive conditions. Under conditions of potentially (imperfect)competitive markets, we conclude that public and major or fully private airports operate equally efficiently. Nevertheless, air-port charges are higher at unregulated, private airports as compared to their public counterparts.

6. Conclusions and future research

The inefficiency of airports may be explained not only by input excess and output shortfalls but also by exogenous factorsover which management have little to no control. A number of empirical studies have assessed the impact of ownershipstructure, economic regulation and levels of competition on efficiency and aeronautical charges however the effects werealways considered separately hence significance was sometimes an issue. The aim of this research has been to extend thecurrent literature on airport ownership, regulation and competition by assessing the combined impact of the environmentalvariables in order to gain insight into the most efficient ownership form and regulatory framework whilst accounting forlevels of regional and hub competition as well as other managerial strategies.

The two-stage analysis combined data envelopment analysis in the first stage and regression analysis in the second stage.The non-radial, weighted additive, input-oriented DEA model has been chosen to identify all relative cost inefficiencies. Fol-lowing the recent debate on the most appropriate second-stage regression model, Banker and Natarajan (2008) and Simarand Wilson (2007) propose ordinary least squares and truncated regression respectively. Due to the fact that we have a paneldataset, we apply robust clustering to each of these forms as well as the random effects model in order to consider unob-served heterogeneity and allow for time-invariant covariates. The regression results were consistent across the pooledOLS, truncated and tobit, robust cluster, random effects models.

Data availability remains a serious issue and the attempt to include all combinations of institutional settings proved dif-ficult. However, the results of the joint analysis provide additional information beyond that of the individual regression mod-els. The results suggest that private, regulated airports are more efficient under monopolistic conditions whereas pure publicand private airports operate equally efficiently given potential local, regional and gateway competition. Furthermore, ex-ante regulation at all airports located in a competitive environment is unnecessary and generates X-inefficiency of the orderof 15%, which rises substantially at purely public airports. However, unregulated, fully private and majority privately ownedairports located in a competitive setting, charge higher aeronautical fees than unregulated public airports. Combining theresults from the efficiency and revenue models reveal that imperfect competition is sufficient to encourage airport cost effi-ciency and reduce the likelihood of abuse of market power, although private airports still transfer fewer of these gains toairlines. On the other hand, non-hub airports with weak local competition generally require economic regulation in orderto prevent an exploitation of market power and to encourage cost efficiency. Dual-till price-cap regulation is shown to bethe most efficient form. Having defined competition at the regional and hub level in a relatively simple manner, we may haveassumed excessive levels of competition having ignored product diversification strategies such as market destination sepa-ration. In other words, we categorized airports as competitive that may be avoiding direct competition, however despite thisshortcoming, the impact of competition remains clear.

The policy implications from this analysis suggest that if an airport is to be privatized, for example to apply to the capitalmarkets to fund further investment, it is better to completely privatize the airport. Partial privatization appears to lead toconflicting objectives. Irrespective of ownership form, under weakly competitive markets, regulation is worthwhile anddual-till price caps appear to be the most appropriate form. Under relatively competitive markets, regulation appears tobe unnecessary to encourage cost efficiency. However, unregulated, privatized airports are likely to set higher charges thantheir public counterparts. Consequently it is up to the government to determine which is less problematic; the additionalcost of regulation or the higher prices that the airlines are likely to be charged.

Future research would require substantially more data to permit an improved analysis of all the categories describedhere. Finer distinctions with respect to ownership form and regulation might better highlight the most efficient institutionalsetting given alternative levels of competition. The consideration of product differentiation may render the actual degree ofcompetition more precisely. Additional environmental variables, including airport-related delays, noise and air pollution,

104 N. Adler, V. Liebert / Transportation Research Part A 64 (2014) 92–109

would enable the development of a social welfare analysis of airports and the trade-off across the different stakeholders. Wewere in a position to compare delay and congestion, but only in a limited manner. We find that the positive impact of con-gestion is higher than the negative effect of delays, which indicates the need for service quality indicators written into con-tracts between airlines and airports or internalization through compensation of airport related delays to airlines andpassengers. Malmquist measure of changes in technical and managerial efficiency over time would also strengthen suchan analysis, however the dataset must be balanced and substantial for this to be a possibility. Finally, in order to improvethe benchmarking process, a more accurate measure of the capital required to build, maintain and expand or reduce airportswould only be possible if the ACI, ICAO or equivalent organization were to develop standardized data collection proceduresthat airports globally reported annually. This is extremely important if we are to be able to analyze this industry and providesensible recommendations for future development.

Acknowledgments

We would like to thank the German Airport Performance research project funded by the German Ministry of Educationand Research for providing the data. Further, we thank Prof. David Gillen, Prof. Hans-Martin Niemeier, and the participants ofthe World Bank roundtable on Transport Productivity in February 2010, the German Aviation Research Society Conference inNovember 2010 and the Aviation seminar at the University of British Columbia in February 2011 for fruitful discussions.Nicole Adler thanks the Recanati Fund for partial funding of this research.

Appendix A. List of airports

Code

Airport Country Time period Ownership Regulation Competition

ABZ

Aberdeen UK 99-05 Fully private No ex-ante regulation Weak AMS Amsterdam NL 98-06 Public Cost-plus, dual till Heavy

2007

Incentive, dual till ATH Athens GR 05-07 Minor private Cost-plus, dual till Weak BFS Belfast UK 99-07 Fully private No ex-ante regulation Weak BHX Birmingham UK 98-07 Major private No ex-ante regulation Heavy BLQ Bologna IT 00-05 Public No ex-ante regulation Heavy BRE Bremen DE 98-07 Public Cost-plus, dual till Heavy BRU Brussels BE 99-04 Major private Cost-plus, single till Heavy BTS Bratislava SK 03-05 Public No ex-ante regulation Weak

06-07

Major private BUD Budapest HU 00-01 Public No ex-ante regulation Weak CGN Cologne Bonn DE 98-07 Public Cost-plus, dual till Heavy CPH Copenhagen DK 01-04 Major private Incentive, dual till Weak DRS Dresden DE 98-06 Public Cost-plus, dual till Weak DTM Dortmund DE 98-07 Public Cost-plus, dual till Heavy DUB Dublin IE 06-07 Public Incentive, single till Weak DUS Dusseldorf DE 99-07 Major private Cost-plus, dual till Heavy EDI Edinburgh UK 98-07 Fully private No ex-ante regulation Heavy EMA East Midlands UK 98-06 Fully private No ex-ante regulation Heavy FRA Frankfurt DE 02-07 Minor private Incentive, dual till Heavy GLA Glasgow UK 98-06 Fully private No ex-ante regulation Heavy GVA Geneva CH 98-07 Public No ex-ante regulation Weak HAJ Hanover DE 98-07 Minor private Cost-plus, dual till Weak HAM Hamburg DE 98-99 Public Cost-plus, dual till Heavy

00-07

Minor private Incentive, dual till LBA Leeds Bradford UK 98-02, 06-07 Public No ex-ante regulation Heavy LCY London-City UK 99-07 Fully private No ex-ante regulation Heavy LEJ Leipzig DE 98-06 Public Cost-plus, dual till Weak LGW London-Gatwick UK 98-05 Fully private Incentive, single till Heavy LHR London-Heathrow UK 98-05 Fully private Incentive, single till Heavy LJU Ljubljana SI 98-06 Major private No ex-ante regulation Heavy LTN London-Luton UK 00-07 Fully private No ex-ante regulation Heavy LYS Lyon FR 98-06 Public No ex-ante regulation Weak MAN Manchester UK 98-07 Public Incentive, single till Heavy MEL Melbourne AU 99-01 Fully private Incentive, dual till Weak

14

N. Adler, V. Liebert / Transportation Research Part A 64 (2014) 92–109 105

Appendix A (continued)

Code

A score of

Airport

1.000 represents a relatively

Country

cost efficient ai

Time period

rport and a value belo

Ownership

w that indicates cost in

Regulation

efficiency.

Competition

02-07

No ex-ante regulation MLA Malta MT 02-06 Major private Incentive, dual till Weak MLH Basel Mulhouse FR 98-07 Public No ex-ante regulation Weak MRS Marseille FR 98-06 Public No ex-ante regulation Heavy MUC Munich DE 98-05 Public Cost-plus, dual till Heavy NCE Nice FR 98-06 Public No ex-ante regulation Heavy NUE Nuremberg DE 98-07 Public Cost-plus, dual till Weak OSL Oslo NO 99-03 Public Cost-plus, single till Weak

04-07

Incentive, single till PER Perth AU 99-01 Fully private Incentive, dual till Weak

02-07

No ex-ante regulation RIX Riga LV 04-06 Public No ex-ante regulation Weak SOU Southampton UK 99-05 Fully private No ex-ante regulation Heavy STN London-Stansted UK 98-06 Fully private Incentive, single till Heavy STR Stuttgart DE 98-07 Public Cost-plus, dual till Weak SYD Sydney AU 03-07 Fully private No ex-ante regulation Weak SZG Salzburg AT 2004 Public Cost-plus, single till Weak

05-07

Incentive, single till TLL Tallinn EE 02-07 Public Cost-plus, dual till Weak VCE Venice IT 00-04 Public No ex-ante regulation Heavy

2005

Minor private VIE Vienna AT 98-07 Minor private Incentive, single till Heavy ZRH Zurich CH 01-07 Minor private No ex-ante regulation Heavy

Appendix B. DEA efficiency scores14

1998

1999 2000 2001 2002 2003 2004 2005 2006 2007

ABZ

– 0.579 0.599 0.546 0.522 0.505 0.515 0.536 – – AMS 1.000 1.000 1.000 0.496 0.309 0.249 0.253 0.297 0.259 0.280 ATH – – – – – – – 0.421 0.432 0.450 BFS – 0.512 0.489 0.487 0.480 0.485 0.575 0.502 0.500 0.502 BHX 0.549 0.577 0.598 0.582 0.568 0.583 0.535 0.530 0.514 0.481 BLQ – – 0.755 0.715 0.710 0.733 0.698 0.740 – – BRE 0.587 0.648 0.602 0.589 0.585 0.568 0.575 0.568 0.562 0.562 BRU – 1.000 1.000 0.653 0.455 0.417 0.431 – – – BTS – – – – – 0.854 0.732 0.575 0.579 0.563 BUD – – 0.357 0.336 – – – – – – CGN 0.390 0.410 0.390 0.417 0.409 0.388 0.361 0.350 0.337 0.323 CPH – – – 1.000 0.617 0.562 0.635 – – – DRS 0.639 0.588 0.575 0.547 0.559 0.574 0.548 0.538 0.532 – DTM 1.000 0.843 0.732 0.472 0.433 0.418 0.408 0.410 0.410 0.408 DUB – – – – – – – – 0.504 0.475 DUS – 1.000 1.000 0.605 0.682 0.472 0.588 0.531 0.508 0.574 EDI 0.664 0.712 0.714 0.702 0.713 0.718 0.738 0.711 0.620 0.661 EMA 0.637 0.628 0.629 0.606 0.629 0.640 0.637 0.584 0.558 – FRA – – – – 1.000 1.000 1.000 1.000 0.732 1.000 GLA 0.702 0.712 0.714 0.705 0.693 0.685 0.697 0.703 0.692 – GVA 0.635 0.660 0.791 0.720 0.709 0.700 0.676 0.638 0.645 0.687 HAJ 0.313 0.319 0.331 0.316 0.314 0.310 0.308 0.301 0.302 0.297 HAM 0.535 0.524 0.426 0.394 0.399 0.381 0.420 0.404 0.406 0.414 LBA 1.000 1.000 0.901 0.830 0.842 0.881 0.888 – – – LCY – 1.000 1.000 0.933 0.887 0.867 0.799 0.744 0.888 1.000

(continued on next page)

106 N. Adler, V. Liebert / Transportation Research Part A 64 (2014) 92–109

Appendix A (continued)Appendix B (continued)

1998

1999 2000 2001 2002 2003 2004 2005 2006 2007

LEJ

0.571 0.570 0.392 0.382 0.374 0.369 0.367 0.337 0.359 – LGW 1.000 0.861 1.000 0.744 0.682 0.619 0.658 0.537 – – LHR 1.000 1.000 1.000 0.907 1.000 0.852 1.000 1.000 – – LJU 1.000 1.000 1.000 1.000 1.000 1.000 1.000 1.000 1.000 – LTN – 0.584 0.518 0.525 0.513 0.520 0.564 0.511 0.529 0.473 LYS 0.487 0.504 0.525 0.509 0.459 0.443 0.448 0.414 0.417 – MAN 0.477 0.425 0.407 0.334 0.352 0.434 0.504 0.442 0.434 0.397 MEL – 0.862 1.000 1.000 0.815 0.760 0.903 1.000 1.000 1.000 MLA – – – – 1.000 1.000 1.000 1.000 1.000 – MLH 0.887 0.821 0.670 0.542 0.502 0.439 0.432 0.438 0.414 0.413 MRS 0.832 0.844 0.932 0.808 0.621 0.586 0.597 0.600 0.588 – MUC 0.335 0.338 0.355 0.353 0.332 0.263 0.264 0.234 – – NCE 1.000 1.000 1.000 0.857 0.684 0.612 0.543 0.526 0.538 – NUE 0.628 0.613 0.636 0.619 0.596 0.574 0.578 0.570 0.568 0.564 OSL – 0.555 0.517 0.515 0.440 0.479 0.441 0.458 0.494 0.543 PER – 1.000 1.000 0.972 0.790 0.689 0.678 0.673 0.703 1.000 RIX – – – – – – 0.547 0.558 0.563 – SOU 0.847 0.919 0.943 0.720 0.617 1.000 0.751 – – – STN 0.670 0.770 0.651 0.595 0.723 0.707 0.661 0.635 – – STR 0.561 0.564 0.559 0.534 0.513 0.505 0.462 0.516 0.513 0.519 SYD – – – – – 0.563 0.748 1.000 1.000 1.000 SZG – – – – – – 0.733 0.716 0.722 0.723 TLL – – – – 0.904 0.690 0.628 0.522 0.516 0.513 VCE – – 0.625 0.599 0.627 0.671 0.682 0.545 – – VIE 0.428 0.456 0.471 0.396 0.365 0.338 0.338 0.325 0.320 0.310 ZRH – – – 0.656 0.532 0.516 0.481 0.342 0.315 0.306

Appendix C. Results from second-stage OLS, truncated and censored regressions over DEA efficiency scores

Individual effects

Combined effects

RobustClusterOLS

RobustClusterTobit

RobustClusterTruncated

RobustClusterOLS

RobustClusterTobit

RobustClusterTruncated

(a) Airport characteristics and management strategies

Share of non-aviation

revenues > 50%

0.109(0.060)**

0.137(0.038)***

0.074(0.067)

0.095(0.055)**

0.112(0.038)***

0.049(0.055)

Heavy delays

�0.253(0.062)***

�0.292(0.034)***

�0.196(0.071)***

�0.278(0.066)***

�0.311(0.034)***

�0.220(0.069)***

Runway capacity utilizationbetween 50% and 90%

0.136(0.055)***

0.154(0.033)***

0.081(0.065)

0.141(0.054)***

0.154(0.031)***

0.084(0.053)

Runway capacity utilization > 90%

0.708(0.106)***

0.972(0.105)***

0.774(0.064)***

0.733(0.164)***

1.019(0.100)***

0.718(0.109)***

(b) Ownership, regulation and competition – individual effects

Minor private airport �0.252

(0.070)***

�0.256(0.050)***

�0.255(0.076)***

–

– –

Major private airport

0.210(0.089)***

0.279(0.049)***

0.086(0.097)

–

– –

Fully private airport

0.028(0.072)

0.028(0.039)

0.057(0.086)

–

– –

Heavy competition

0.027(0.060)

0.028(0.030)

0.004(0.068)

–

– –

Cost-plus regulation, single-till

�0.272(0.076)***

�0.303(0.083)***

�0.267(0.076)***

–

– –

Cost-plus regulation, dual-till

�0.275 �0.286 �0.274 – – –

N. Adler, V. Liebert / Transportation Research Part A 64 (2014) 92–109 107

Appendix A (continued)Appendix C (continued)

Individual effects

Combined effects

RobustClusterOLS

RobustClusterTobit

RobustClusterTruncated

RobustClusterOLS

RobustClusterTobit

RobustClusterTruncated

(0.071)***

(0.036)*** (0.081)***

Price-cap regulation, single-till

�0.206(0.084)***

�0.211(0.048)***

�0.148(0.096)

–

– –

Price-cap regulation, dual-till

0.026(0.089)

0.046(0.063)

�0.091(0.080)

–

– –

(c) Ownership, regulation and competition – combined effects

Low competition Public No regulation – – – 0.436 (0.112)*** 0.427 (0.077)*** 0.453 (0.123)***

Regulation

– – – 0.296 (0.120)*** 0.277 (0.074)*** 0.333 (0.123)***

Minor

No regulation – – – n/a n/a n/a Private Regulation – – – Base case Base case Base case Major No regulation – – – 0.379 (0.103)*** 0.357 (0.180)** 0.419 (0.114)***

Private

Regulation – – – 0.785 (0.134)*** 0.900 (0.113)*** 0.538 (0.104)***

Fully

No regulation – – – 0.378 (0.147)*** 0.364 (0.087)*** 0.345 (0.137)***

Private

Regulation – – – 0.580 (0.151)*** 0.685 (0.148)*** 0.958 (0.331)***

*** *** ***

Potential competition Public No regulation – – – 0.662 (0.087) 0.672 (0.078) 0.687 (0.099) Regulation – – – 0.112 (0.104) 0.100 (0.073) 0.095 (0.119)

Minor

No regulation – – – 0.268 (0.107)*** 0.272 (0.106)*** 0.282 (0.113)***

Private

Regulation – – – 0.154 (0.119) 0.117 (0.085) 0.096 (0.105) Major No regulation – – – 0.661 (0.151)*** 0.724 (0.087)*** 0.495 (0.090)***

Private

Regulation – – – 0.518 (0.097)*** 0.541 (0.093)*** 0.418 (0.122)***

Fully

No regulation – – – 0.581 (0.106)*** 0.574 (0.075)*** 0.603 (0.114)***

Private

Regulation – – – 0.406 (0.131)*** 0.425 (0.099)*** 0.573 (0.114)***

(d) Time trends

1999 �0.022 (0.028) � 0.033 (0.068) 0 .013 (0.029) �0.028

(0.028)

�0.041 (0.062) 0.007 (0.031)

2000

�0.057 (0.042) � 0.063 (0.067) � 0.029 (0.041) �0.068(0.041)

�0.074 (0.062)

�0.042 (0.041)

2001

�0.134(0.049)***

�

0.176 (0.066)*** � 0.038 (0.047) �0.146(0.049)***

�0.193 (0.060)***

�0.070 (0.043)

2002

�0.160(0.057)***

�

0.198 (0.065)*** � 0.067 (0.047) �0.165(0.058)***

�0.204 (0.060)***

�0.087(0.042)***

2003

�0.221(0.061)***

�

0.270 (0.065)*** � 0.132(0.047)***

0.220(0.062)***

�0.271 (0.059)***

�0.143(0.044)***

2004

�0.192(0.061)***

�

0.230 (0.065)*** � 0.113(0.047)***

�0.192(0.061)***

�0.230 (0.059)***

�0.123(0.043)***

2005

�0.234(0.060)***

�

0.271 (0.065)*** � 0.172(0.044)***

�0.226(0.059)***

�0.262 (0.059)***

�0.178(0.042)***

2006

�0.264(0.065)***

�

0.312 (0.067)*** � 0.190(0.050)***

�0.251(0.064)***

�0.294 (0.061)***

�0.191(0.047)***

2007

�0.280(0.080)***

�

0.324 (0.072)*** � 0.221(0.066)***

�0.236(0.115)***

�0.272 (0.066)***

�0.195(0.057)***

Intercept

�0.227(0.084)***

�

0.169 (0.063)*** � 0.313(0.088)***

�0.695(0.115)***

�0.635 (0.085)***

�0.786(0.122)***

R2

0.5626 0 .5622 0 .4995 0.6333 0.6187 0.5444 Observations

(n)

398 3 98 (56 right-

censored)3t

42 (56runcated)

398

398 (56 right-censored)

342 (56truncated)

108 N. Adler, V. Liebert / Transportation Research Part A 64 (2014) 92–109

References

Australian Competition and Consumer Commission, 2010. Airport Monitoring Report 2008–09: Price, Financial Performance and Quality of ServiceMonitoring. ACCC, Canberra.

Aigner, D., Lovell, C.A.K., Schmidt, P., 1977. Formulation and estimation of stochastic frontier production function models. J. Econom. 6, 21–37.Armstrong, M., Sappington, D.E.M., 2006. Regulation, competition and liberalization. J. Econ. Lit. 44 (2), 325–366.Averch, H., Johnson, L.L., 1962. Behavior of the firm under regulatory constraint. Am. Econ. Rev. 52 (5), 1052–1069.Banker, R.D., Natarajan, R., 2008. Evaluating contextual variables affecting productivity using data envelopment analysis. Oper. Res. 56 (1), 48–58.Banker, R.D., Charnes, A., Cooper, W.W., 1984. Some models for estimating technical and scale inefficiencies in data envelopment analysis. Manage. Sci. 30

(9), 1078–1092.Barros, C.P., 2008. Airports in Argentina: technical efficiency in the context of an economic crisis. J. Air Transp. Manage. 14 (6), 315–319.Barros, C.P., Dieke, P.U., 2007. Performance evaluation of Italian airports: a data envelopment analysis. J. Air Transp. Manage. 13 (4), 184–191.Barros, C.P., Marques, R.C., 2008. Performance of European airports: regulation, ownership and managerial efficiency. Working Paper 25/2008/DE/UECE,

School of Economics and Management, Technical University of Lisbon, Portugal.Beesley, M.M., 1999. Airport regulation. In: Beesley, M.E. (Ed.), Regulating Utilities: A New Era? Institute of Economic Affairs, London.Bel, G., Fageda, X., 2010. Privatization, regulation and airport pricing: an empirical analysis for Europe. J. Regul. Econ. 37 (2), 142–161.Boardman, A.E., Vining, A.R., 1989. Ownership and performance in competitive environments: a comparison of the performance of private, mixed, and state-

owned enterprises. J. Law Econ. 32 (1), 1–33.Boyko, M., Shleifer, A., Vishny, R.W., 1996. A theory of privatization. Econ. J. 106, 309–319.Button, K.J., Weyman-Jones, T.G., 1992. Ownership structure, institutional organization and measured X-efficiency. Am. Econ. Rev. 82 (2), 439–445.Caves, D.W., Christensen, L.R., 1980. The relative efficiency of public and private firms in a competitive environment: the case of Canadian railroads. J. Polit.

Econ. 88, 958–976.Caves, D.W., Christensen, L.R., Diewert, W.E., 1982. Multilateral comparisons of output, input, and productivity using superlative index numbers. Econ. J. 92,

73–86.Charnes, A., Cooper, W.W., Rhodes, E., 1978. Measuring the efficiency of decision making units. Eur. J. Oper. Res. 2 (6), 429–444.Charnes, A., Cooper, W.W., Golany, B., Seiford, J., 1985. Foundations of data envelopment analysis for Pareto-Koopmans efficient empirical production

functions. J. Econom. 30 (1–2), 91–107.Czerny, A.I., 2006. Price-cap regulation of airports: single till versus dual till. J. Regul. Econ. 30 (1), 85–97.Forsyth, P., 2004. Replacing regulation: airport price monitoring in Australia. In: Forsyth, P., Gillen, D., Knorr, A., Mayer, O., Niemeier, H.M. (Eds.), The

Economic Regulation of Airports: Recent Developments in Australasia, North America and Europe. Ashgate, Aldershot.Gillen, D., 2011. The evolution of airport ownership and governance. J. Air Transp. Manage. 17 (1), 3–13.Gillen, D., Lall, A., 1997. Developing measures of airport productivity and performance: an application of data envelope analysis. Transport. Res. Part E 33 (4),

261–273.Gillen, D., Niemeier, H.M., 2008. The European Union: evolution of privatization, regulation, and slot reform. In: Winston, C., de Rus, G. (Eds.), Aviation

Infrastructure Performance: A Study in Comparative Political Economy. Brookings Institute, Washington.Graham, A., 2005. Airport benchmarking: a review of the current situation. Benchmarking: Int. J. 12 (2), 99–111.Hooper, P., Cain, R., White, S., 2000. The privatisation of Australia’s airports. Transport. Res. Part E 36 (3), 181–204.International Air Transport Authority, 2003. Airport Capacity/Demand Profiles, 2003 ed. Air Transport Consultancy Services, Montreal, Geneva.International Air Transport Association, 2010. World Scheduling Guidelines. 19th ed. <http://www.iata.org/NR/Content Connector/CS2000/SiteInterface/

sites/whatwedo/scheduling/file/fdc/WSG-12thEd.pdf> (accessed 17.03.10).Leibenstein, H., 1966. Allocative efficiency versus X-efficiency. Am. Econ. Rev. 56 (3), 392–415.Liebert, V., Niemeier, H.M., 2010. A review of empirical studies on the productivity and efficiency of airports. In: World Conference on Transport Research

Proceedings.Lin, L.C., Hong, C.H., 2006. Operational performance evaluation of international major airports: an application of data envelopment analysis. J. Air Transp.

Manage. 12 (6), 342–351.Littlechild, S.C., 1983. Regulation of British Telecommunications’ Profitability: Report to the Secretary of State. Department of Industry, London.Lovell, C.A.K., Pastor, J.T., 1995. Units invariant and translation invariant DEA models. Oper. Res. Lett. 18, 147–151.Martín, J.C., Socorro, M.P., 2009. A new era for airport regulators through capacity investments. Transport. Res. Part A 43 (6), 618–625.Meeusen, W., van den Broeck, J., 1977. Efficiency estimation from Cobb–Douglas production functions with composed error. Int. Econ. Rev. 18 (2), 435–444.Oum, T.H., Zhang, A., Zhang, Y., 2004. Alternative forms of economic regulation and their efficiency implications for airports. J. Transp. Econ. Policy 38 (2),

217–246.Oum, T.H., Adler, N., Yu, C., 2006. Privatization, corporatization, ownership forms and their effects on the performance of the world’s major airports. J. Air

Transp. Manage. 12 (3), 109–121.Oum, T.H., Yan, J., Yu, C., 2008. Ownership forms matter for airport efficiency: a stochastic frontier investigation of worldwide airports. J. Urban Econ. 64 (2),

422–435.Parker, D., 1999. The performance of BAA before and after privatisation: a DEA study. J. Transp. Econ. Policy 33 (2), 133–145.Pathomsiri, S., Haghani, A., Dresner, M., Windle, R.J., 2008. Impact of undesirable outputs on the productivity of US airports. Transport. Res. Part E 44 (2),

235–259.Pels, E., Nijkamp, P., Rietveld, P., 2003. Inefficiencies and scale economies of European airport operations. Transport. Res. Part E 39 (5), 341–361.Posner, R.A., 1974. Theories of economic regulation. Bell J. Econ. Manage. Sci. 5 (2), 335–358.Reinhold, A., Niemeier, H.M., Kamp, V., Müller, J., 2010. An evaluation of yardstick regulation for European airports. J. Air Transp. Manage. 16 (2), 74–80.Sappington, D.E.M., Stiglitz, J.E., 1987. Privatization, information and incentives. J. Policy Anal. Manage. 6 (4), 567–582.Shapiro, C., Willig, R.D., 1990. Economic rationales for the scope of privatization. In: Suleiman, E.N., Waterbury, J. (Eds.), The Political Economy of Public

Sector Reform and Privatization. Westview Press, Boulder.Shleifer, A., 1985. A theory of yardstick competition. Rand J. Econ. 16 (3), 319–327.Simar, L., Wilson, P.W., 2007. Estimation and inference in two-stage, semi-parametric models of production processes. J. Econom. 136 (1), 31–64.Starkie, D., 2001. Reforming UK airport regulation. J. Transp. Econ. Policy 35, 119–135.Starkie, D., 2002. Airport regulation and competition. J. Air Transp. Manage. 8 (1), 63–72.Tone, K., 2001. A slacks-based measure of efficiency in data envelopment analysis. Eur. J. Oper. Res. 130, 498–509.Tretheway, M.W., Kincaid, I., 2010. Competition between airports: occurrence and strategy. In: Forsyth, P., Gillen, D., Müller, J., Niemeier, H.M. (Eds.), Airport

Competition: The European Experience. Ashgate, Aldershot.Tulkens, H., van den Eeckaut, P.V., 1995. Non-parametric efficiency, progress and regress measures for panel data: methodological aspects. Eur. J. Oper. Res.

80 (3), 474–499.Van Dender, K., 2007. Determinants of fares and operating revenues at US airports. J. Urban Econ. 62 (2), 317–336.Vickers, J., Yarrow, G., 1991. Economic perspectives on privatization. J. Econ. Perspect. 5 (2), 111–132.Vienna International Airport, 2004. Geschäftsbericht. Flughafen Wien AG.White, H., 1980. A heteroskedasticity-consistent covariance matrix estimator and a direct test for heteroskedasticity. Econom.: J. Econom. Soc. 48 (4), 817–

838.Williams, R.L., 2000. A note on robust variance estimation for cluster-correlated data. Biometrics 56, 645–646.

N. Adler, V. Liebert / Transportation Research Part A 64 (2014) 92–109 109

Yang, H., Zhang, A., 2011. Price-cap regulation of congested airports. J. Regul. Econ. 39, 293–312.Yokomi, M., 2005. Measurement of Malmquist index of privatized BAA plc. In: 9th ATRS Conference 2005, Rio de Janeiro Brazil.Yuen, A.C.-L., Zhang, A., 2009. Effects of competition and policy changes on Chinese airport productivity: an empirical investigation. J. Air Transp. Manage.

15 (4), 166–174.Zhang, A., Zhang, Y., 2010. Airport capacity and congestion pricing with both aeronautical and commercial operations. Transport. Res. Part B 44 (3), 404–

413.