FREE ELECTRONIC LIBRARY - Abstract, dissertation, book

Pages:   || 2 | 3 | 4 |

«FS IV 97 - 25 Long Run Properties of Technical Efficiency in the U.S. Airline Industry Ila M. Semenick Alam* Robin C. Sickles** * Tulane University ...»

-- [ Page 1 ] --




discussion papers CENTER BERLIN

FS IV 97 - 25

Long Run Properties of Technical

Efficiency in the U.S. Airline Industry

Ila M. Semenick Alam* Robin C. Sickles** * Tulane University ** Rice University October 1997 ISSN Nr. 0722 - 6748 Forschungsschwerpunkt Marktprozeß und Unternehmensentwicklung Research Area Market Processes and Corporate Development


Ila M. Semenick Alam, Robin C. Sickles, Long Run Properties of Technical Efficiency in the U.S. Airline Industry, Discussion Paper FS IV 97 - 25, Wissenschaftszentrum Berlin, 1997.

Wissenschaftszentrum Berlin für Sozialforschung gGmbH, Reichpietschufer 50, 10785 Berlin, Tel. (030) 2 54 91 - 0


Long Run Properties of Technical Efficiency in the U.S. Airline Industry by Ila M. Semenick Alam and Robin C. Sickles* This paper takes an innovative approach to test the relationship between technical efficiency and the market structure hypothesis which states that competitive pressure enhances relative efficiency. DEA and FDH time series of technical efficiency scores, for a panel of 11 US airlines observed quarterly during 1970-1990, are examined for cointegration and convergence. For almost all firm pairs (90% for both efficiency measures), the null hypothesis of cointegration cannot be rejected; meanwhile, the null of no cointegration is rejected for approximately one-third of the firm pairs. Furthermore, convergence tests document less dispersion in firm performance over time. These results indicate that the scores move together and, in fact, the firms are becoming more alike one another in terms of efficiency.


Langfristige Eigenschaften der technischen Effizienz in der US-Luftfahrtindustrie In diesem Beitrag wird ein innovativer Ansatz entwickelt, um die Beziehung zwischen technischer Effizienz und der Marktstrukturhypothese zu überprüfen, die besagt, daß Wettbewerbsdruck die relative Effizienz erhöht. Die DEA- und FDH-Zeitreihen der technischen Effizienz für ein Panel-Datensatz von 11 US-Fluggesellschaften, die während der Periode 1970 bis 1990 beobachtet wurden, werden im Hinblick auf Kointegration und Konvergenz analysiert. Für nahezu alle Unternehmenspaare (mehr als 90% bei beiden Effizienzmaßen) kann die Nullhypothese der Kointegration nicht zurückgewiesen werden, wo hingegen die Nullhypothese, daß keine Kointegration vorliegt, für etwa ein Drittel der Unternehmenspaare zurückgewiesen wird. Außerdem zeigt der Konvergenztest, daß die Streuung des Unternehmenserfolgs im Zeitablauf abnimmt. Die Ergebnisse deuten darauf hin, daß die Effizienzwerte sich auf einander zu bewegen und daß in der Tat, die Unternehmen im Hinblick auf Effizienz sich immer ähnlicher werden.

* Previous versions of this paper were presented at the Fifth Conference on Panel Data in Paris, France, June, 1994; the Productivity Seminar in Athens, Georgia, October, 1994; and the American Economic Association Meeting in Washington, DC, January, 1995. The authors thank Severin Borenstein, William W. Cooper, Peter Schmidt, Larry Jenney, David Good, Gerald McDougall and Bruce McClelland for their helpful comments; we also thank Leola Ross for her assistance (in obtaining the concentration measures).


An industry’s production technology is the set of all feasible pairs of input-output vectors. The production frontier consists of those combinations which, under existing production processes, maximize output production for given input levels or, conversely, minimize input usage for given output levels. Firms not operating on this frontier are identified as technically inefficient.

The frontier literature has largely been concerned with documenting inefficiency of firms in various sectors. Technical efficiency techniques have a wide-spread appeal because both government policy makers and industry managers are concerned about productive performance. More importantly, upon determination of efficiency differentials, these techniques can be used as decision making tools since they indicate areas of deficiency and direction for change.

Our objective, beyond simply presenting evidence of technical inefficiency, is to explore some of the long-run properties of these scores. Specifically we are interested in the link between market structure and performance first made explicit in Leibenstein's (1966) article on X-efficiency which states that, given “proper motivations”, firms can achieve increased efficiency. A prime motivational factor is the degree of competitive pressure.

Studies have become more formalized with the introduction of technical efficiency measurement techniques which are useful tools to measure and partition X-inefficiency (Leibenstein and Maital, 1992). Caves and Barton (1990), for example, consider the relationship between technical efficiency levels and competitive conditions for 285 US industries. Overall they find support for public policies designed to maintain competition among producers since these policies promote efficiency. Other studies focus on sectors such as utilities (e.g., Reifschneider and Stevenson, 1991), since the dependency between efficiency and competitive pressure has significant regulatory relevance. Button and Weyman-Jones (1992) find, in their literature survey, that those industries subject to bureaucratic control generally exhibit lower efficiency levels than those which are competitive or weakly regulated.

We take a unique approach to empirically examine this relationship between competitive forces and the time pattern of technical efficiency. Our procedure is made possible by bringing together recent advances in various areas of the economics literature. In the technical efficiency arena, studies focused exclusively on cross-sectional results until developments in the parametric estimation of technical efficiency (Cornwell, Schmidt and Sickles, 1990; hereafter CSS) and generalizations of the linear programming approaches (Tulkens and Vanden Eeckaut, 1995) allowed the recent exploitation of panel data sets.

The ability to capture the dynamic nature of a firm's performance relative to its competitors has stimulated even greater interest in the topic. However, little is known about the time series nature of technical efficiency. To explore the dynamics, we introduce the concepts of cointegration and convergence to this literature in order to determine if, under increasing competitive forces, the technical efficiency scores of firms move together in the long run (cointegration) or, in fact, move closer together over time (convergence). Evidence of these time series characteristics would be indicative of a greater concern among firms to maintain high relative technical efficiency. To this end, we use developments in the cointegration (including those by Kwiatkowski, Phillips, Schmidt and Shin, 1992; hereafter KPSS), and convergence literatures (including those by Färe, Grosskopf, Norris and Zhang, 1994; hereafter FGNZ).

The US airline industry from 1970 to 1990 is an ideal candidate for this analysis since, over this period, competition among the airlines increased due to deregulation1. The importance of studying technical efficiency in this industry is highlighted by previous papers which have found technical inefficiency to be a much greater source of distortion than that which is due to allocative inefficiency2. Furthermore, US airlines continue to face substantial upheavals in the form of mergers, failures, bankruptcy filings, reorganizations and operating loss reports. This apparent inability of the industry to reach equilibrium, has raised concern that the future is bleak in terms of the number of carriers which will survive and prosper. One other empirically attractive feature of this industry, a consequence of the strict filing requirements imposed by the federal government, is the wealth of accessible data not available in most other industries.

The paper is organized as follows: section 2 briefly discusses the methodologies; section 3 describes the production data set; section 4 contains the empirical findings; section 5 concludes.


2.1. Efficiency Measurement. Assume a panel with n=1,...,N firms, t=1,...,T periods, j=1,...,J inputs and k=1,...,K outputs. Thus, xjnt is the level of input j used by firm n in period t and yknt is the level of output k produced by firm n in period t. Further, assume The airline industry moved from service-based to price-based competition under deregulation.

See for example, Sickles, Good and Johnson (1986), Sickles (1987), Good, Nadiri and Sickles (1992), and Good, Roller and Sickles (1993, 1994). These results are somewhat at variance with the finding of Atkinson and Cornwell (1994) who use airline data covering the subset period 1970However, for industries in general, the empirical evidence overwhelmingly finds that allocative inefficiency is trivial compared to technical inefficiency (Leibenstein and Maital, 1992).

Furthermore, Berger and Hannan (1994) have preliminary empirical evidence from the banking industry showing that the potential social loss over time from firms operating inefficiently is many times larger than the loss from mispricing that typically accompanies a concentrated market structure and the exercise of market power. This is because the inefficiency loss is present for each unit produced, while the “welfare triangle” loss applies only to customers who do not receive the product because of the higher prices.

an intertemporal3 production set where input and output observations from all time periods are used. The production technology, S, is

–  –  –

Holding the input vector constant, this expression expands the output vector as much as possible without exceeding the boundaries of S. An output efficient firm has a score of 1 while an output inefficient firm has OD(x,y) 1. We utilize the methods of Data Envelopment Analysis (DEA) and Free Disposal Hull (FDH) to define the boundary of S5.

2.2 Data Envelopment Analysis. DEA creates an “envelope” of observed production points (Charnes, Cooper and Rhodes, 1978). It provides for flexible piecewise linear approximations to model the “best-practice” reference technology. Programming methods require neither the assumption of cost minimization or profit maximization nor the specification of a production function. The intertemporal output-based efficiency

score is obtained from the following linear programming model:

–  –  –

Reference sets can also be contemporaneous (data from only one time period are used) or sequential (data from the first time period up to the time period s, t=1,…s where s≤T, are used) (Tulkens and Vanden Eeckaut, 1995). We adopt the intertemporal approach since it allows us to compare various measures of technical efficiency. The Malmquist Index discussed later uses the contemporaneous definition but is able to capture the relationship between time periods.

The input-based distance function is ID(x,y)=max{λ|(x,y/λ)∈S}. Under constant returns to scale, OD(x,y)=[ID(x,y)]-1.

A third method, Stochastic Frontiers (SF), is an econometric approach which, unlike DEA and SFA, requires a priori specification of technology to describe the boundary. For completeness we also calculated efficiencies using the translog functional form and the CSS approach. However, because the SF estimates are log-linear in time and, by definition, are automatically cointegrated they are not very illuminating with respect to our cointegration hypothesis.

where the condition on the weights, wnt, gives constant returns to scale (CRS)6.

2.3. Free Disposal Hull. Compared to DEA, FDH imposes one less restriction on the data: it does not require that convex combinations of every observed production plan be included in the production set (Deprins, Simar and Tulkens, 1984). Thus, an FDH score is relative to an observed point on the frontier and, since managers can look at an actual rather than a theoretically possible alternative in order to modify current practices and improve performance, it is argued that FDH is more valuable for managerial decision making. The FDH frontier is obtained by replacing the last line in Equation (3) with ∑n∑twnt=1, wnt∈{0,1}, n=1,...,N, t=1,...,T.

2.4. Second Stage Regression. The efficiency results obtained from DEA and FDH are regressed7 on a matrix of firm characteristics as well as on firm and time dummies8. The residuals from this regression provide the appropriate measure of performance since they capture the efficiency score minus the effects of these other characteristics9.

2.5. Cointegration. Cointegration analysis examines the existence of stationary relationships between nonstationary variables. For a stationary time series xt, a shock has only a temporary effect on xt, xt has a finite variance and its autocorrelation function declines quickly to zero as the distance between neighboring data points in the xt series increases.

If the series has a unit root, however, the series is nonstationary. It is said to be integrated of order one, xt~I(1), if first differencing results in a stationary series. In general a series is integrated of order d, I(d), if it becomes stationary after differencing d times. An I(1) series is permanently affected by a shock, has a variance approaching infinity as T goes to infinity and had an autocorrelation function which does not decline quickly.

Pages:   || 2 | 3 | 4 |

Similar works:

«A Guidebook of Industrial Traffic Management & Forklift Safety Accident Research Centre Monash University Victoria 3800 Australia (Authors: TJ Larsson, T Horberry, T Brennan J Lambert & I Johnston) An initiative funded by WorkSafe, Victoria. April 2003 Table of Contents SECTION 1. OVERVIEW: THE SYSTEM OF WORK SECTION 2. LOGISTICS, DESIGN AND SAFETY THE HIERARCHY OF HAZARD CONTROL PUBLICATIONS AND TOOLS FOR SAFER FORKLIFTS SECTION 3. PEOPLE AND TASKS PEOPLE AT WORK: THE GHOST IN THE MACHINE...»

«DISS. ETH Nr. 19674 Eects of Litter Diversity and Identity on Leaf Decomposition in Streams Patterns and Mechanisms ABHANDLUNG zur Erlangung des Titels DOKTOR DER WISSENSCHAFTEN der ETH ZÜRICH vorgelegt von Andreas Bruder Dipl. UWIS ETH geboren am 10. Juli 1979 aus Mörschwil SG Angenommen auf Antrag von Prof. Dr. Mark O. Gessner Dr. Eric Chauvet Dr. Brendan G. McKie Prof. Dr. Jukka Jokela April 2011 Summary Growing concern about the consequences of current global biodiversity loss has led to...»

«DER LANDKREIS OSTERHOLZ Die NBauO-Novelle 2012 Wesentliche Änderungen im Überblick Entwurfsverfasserveranstaltung 6. Juli 2012 DER LANDKREIS OSTERHOLZ Begrüßung Richard Eckermann Dezernent für Ordnung, Bauen und Umwelt Dominik Vinbruck Leiter des Bauordnungsamts Arend Bewernitz Leiter des Technischen Sachgebiets DER LANDKREIS OSTERHOLZ Inhalte NBauO 2012 im Überblick Bauanträge mit Servicegarantien Abschließende Hinweise und Fragen DER LANDKREIS OSTERHOLZ NBauO 2012 im Überblick DER...»

«APM-1-W Aquatic Plant Management Barley Straw for Algae Control Carole A. Lembi, Professor of Botany Botany and Plant Pathology, Purdue University E-mail: lembi@purdue.edu The use of barley straw for algae control has received a lot of publicity in recent years. It is now common to find small barley bales being sold in nurseries and garden shops for use in water gardens and small pools to control algae. The word-ofmouth reports of success with this method have led many people to suspect that...»

«Pr´cis of Mechanisms for Cross-Situational Learning of e Word-Referent Mappings: Empirical and Modeling Evidence George Kachergis Introduction Language acquisition is a ubiquitous, challenging problem involving fundamental cognitive abilities of attention, learning, and memory. From infants to adult travelers, language learners are faced with figuring out which words refer to which referents from situations that contain many words and referents. By remembering the words and referents (e.g.,...»

«Q2014 European Conference on quality in official statistics Vienna, 2-5 June 2014 Web scraping techniques to collect data on consumer electronics and airfares for Italian HICP compilation Riccardo Giannini (rigianni@istat.it), Rosanna Lo Conte (rolocont@istat.it), Stefano Mosca (stmosca@istat.it), Federico Polidoro (polidoro@istat.it), Francesca Rossetti (frrosset@istat.it)1 Table of contents Foreword 1. Centralised data collection for Italian HICP estimation and identification of the products...»

«USGS Award No. G13AP00070 SITE RESPONSE MAPPING WITH ONE LESS PROXY: COLLABORATIVE RESEARCH WITH SDSU AND THE USGS FINAL TECHNICAL REPORT SUBMITTED: November 2014 Eric M. Thompson Geological Sciences San Diego State University 5500 Campanile Dr. San Diego, CA 92182-1020 ethompson@mail.sdsu.edu Tel: 858-848-0095 David J. Wald U.S. Geological Survey Golden, CO AWARD PERIOD: May 2013 to April 2014 Abstract The goal of this project is to process and analyze site-specific empirical amplification...»

«EOS Departmental Awards The following awards are given annually at the EOS Fall Barbecue to recognize excellence in teaching and administrative andtechnical performance within the Department: NEW in 2011 EOS Leadership and Service Award This new award recognizes long-term leadership and service, and is open for nomination on a two-year cycle. 2011: Mary Lou Bevier Excellence in Administrative and Technical Services: To formally recognize outstanding performance, dedication and contribution to...»

«Technische Fakultät Albert-Ludwigs-Universität, Freiburg Lehrstuhl für Kommunikationsysteme Prof. Dr. Gerhard Schneider Master thesis Abstract Unattended Workflow Interactions January 18, 2012 Supervisors Dirk von Suchodoletz Klaus Rechert First Reviewer Prof. Dr. Gerhard Schneider Alibek Kulzhabayev Second Reviewer Matr.-Nr.: 2950774 Prof. Dr. Christian Schindelhauer Erklärung Hiermit erkläre ich, dass ich diese Abschlussarbeit selbständig verfasst habe, keine anderen als die...»

«THL D Schutzjacke und Schutzhose Protective jacket and Protective trousers Verwenderinformation User information Verwenderinformation D User information GB Rosenbauer THL D, Schutzjacke / Schutzhose THL D Schutzjacke / Schutzhose zur technischen Hilfeleistung und Brandbekämpfung im freien Gelände Die Schutzjacke und Schutzhose sind eine Feuerwehrschutzbekleidung ▪ gemäß EN 15614 Die Feuerwehtschutzjacke THL D muss immer gemeinsam mit der ▪ Feuerwehrschutzhose THL D getragen werden. Wir...»

«Sex and the Situs Sam Cooper I want to talk about not talking about sex; or, more precisely, I want to talk about why someone who professes total social liberation should choose to not talk about sex. The Situationist International (SI) famously held that all social interaction is mediated by the spectacle, which is the culmination and controlling mechanism of consumer capitalism. Guy Debord argued that within the society of the spectacle “everything that was directly lived has receded into a...»

«OPTIMAL PRODUCTION OF POTABLE WATER CASE STUDY: AGONA SWEDRU GHANA WATER COMPANY LIMITED (GWCL) BY NYAME, MOSES (B.Ed. Mathematics) A thesis submitted to the Institute of Distance Learning to the Department of Mathematics, Kwame Nkrumah University of Science and Technology In partial fulfillment of the requirement for the degree of Master of Science Industrial Mathematics September, 2011 DECLARATION I hereby declare that this submission is my own work towards the MSc. And that, to the best of...»

<<  HOME   |    CONTACTS
2016 www.abstract.xlibx.info - Free e-library - Abstract, dissertation, book

Materials of this site are available for review, all rights belong to their respective owners.
If you do not agree with the fact that your material is placed on this site, please, email us, we will within 1-2 business days delete him.