Printer Friendly

The relationship between operational efficiency and customer service: a global study of thirty-eight large international airlines.


This study examined the relationship between the strategic focus of airline customer service activities and operational efficiency. The empirical investigation employed data for thirty-eight airlines for fiscal year 2000--the last full year before the events of September 11, 2001. This sample was global in nature and included large international carriers, with nine from North America, ten from Europe, six from Latin America, twelve from Asia, and one from the Middle East.

Operational efficiency was measured by means of data envelopment analysis, using the input-oriented model specified by Ali and Seiford, an approach used in related studies and much akin to that of Charnes et al. Efficiency measures were related to strategically focused expenditures on operations, passenger services, and ticketing, promotion, and sales by means of a tobit analysis.

The results of the tobit analysis suggested that focused expenditures on operations and passenger services had a negative impact on operational efficiency. At the same time, focused expenditures on ticketing, sales, and promotion had a positive impact on operational efficiency. The implications of these results are discussed within the context of a value-based approach to customer service in the new operating environment of global airlines.


The financial crisis in the airline industry, following the events of September 11, 2001, has precipitated a so-called "low-cost revolution" in this industry. National flag carriers with significant international operations have looked to "no-frills" carriers for operational models that can be emulated. This has caused large carriers to reassess the appropriate nature of customer service. A major perceived area of cost savings is passenger services, which includes meals, drinks, and other services to passengers as part of their fare, as well as meals or accommodation for transit for delayed passengers. Another area of potential savings is the cost of sales. Selling direct to customers helps reduce the commissions paid to travel agents. Low-cost carriers have realized additional savings by not placing their seats through global reservation systems such as Sabre or Galileo, which charge a fixed fee per booking.

This latter area of cost savings has increasing potential for further cost reductions because of the rapid development of information technology. Airlines need to cut distribution costs in the areas of ticketing, sales, and promotion. The pursuit of disintermediation by airlines would allow carriers to bypass travel and freight agents and link directly to their customers. The Internet may also enhance airlines' marketing power. While the Internet enables airlines to market their services worldwide in a cost-effective manner, there is another more subtle but important advantage. By allowing customers to book directly via the Internet, a database can be developed that allows airlines to offer services more attuned to customer needs as well as allowing them to proactively market to these customers. The use of the Internet may also provide an environment of more dynamic and market-focused pricing. An airline can respond more quickly to competitors' fare cuts or shortfalls in late bookings, which would normally result in unsold seats. Finally, carriers may be able to cross-sell other products and services on their Web sites (Doganis 2001).

It is interesting that consumer interests have become a heightened concern of legislative bodies. In the United States, in June 1999, the Air Transport Association (ATA), working with Congress and the Department of Transportation, developed the Airline Customer Service Commitment. In April 2000, the Aviation Investment and Reform Act provided for a number of consumer-oriented measures and an investigation requirement. The latter was carried out through an audit by the Inspector General, after which each airline incorporated the commitments into its contract for carriage in June 2001.

In June 2000, the European Commission proposed new voluntary commitments and enforcement legislation. The former, which became effective in February 2002, are the Airline Passenger Service Commitment and the Airport Voluntary Commitment on Air Passenger Service. In December 2000 the Commission also started a consultation process focusing on service quality indicator reports and in April 2002 started a similar process focusing on airline contracts. In a legislative context, the Commission issued a proposal in December 2001 for stronger measures for dealing with denied boarding, flight cancellations, and long delays.

Australia, Canada, and New Zealand apply general consumer protection laws to commercial practices of airlines. In November 2000, the Assembly of the Latin American Civil Aviation Conference adopted a recommendation on a consumer protection code for airlines and recommended its adoption for this region. The International Air Transport Association (IATA) and its member airlines have developed conditions of contract and conditions of carriage. In June 2000, this association adopted a Global Customer Services Framework, which is intended as a guide for member airlines in development of their own voluntary commitments (International Civil Aviation Association 2003).


As noted above, airlines face two major areas of potential focus with regard to customer service. The first is passenger services. The second is ticketing, sales, and promotion. However, voluntary commitments now being put into place by airlines suggest that it is the latter area that is being most aggressively targeted. Voluntary activities of member airlines of the ATA, IATA, and European Economic Community include offering the lowest fare available; allowing reservations to be held or cancelled; providing prompt ticket refunds; advising passengers regarding an airline's commercial and operational conditions; ensuring good customer service from code-sharing partners; providing notification of delays, cancellations, and diversions; assisting in case of delay including long on-aircraft delays; handling passengers denied boarding with fairness and consistency; delivering baggage on time; responding to customer complaints; and properly accommodating disabled and special needs passengers (International Civil Aviation Association 2003, Appendix A).

Airlines must also carefully consider their choices in the passenger services provided to their customers. In a recent white paper, IBM outlined a strategy of Customer Relationship Management (CRM) that suggests an approach for selecting and prioritizing the customer services that will be afforded to their passengers (IBM Institute for Business Value 2001). The white paper makes three recommendations:

1. Airlines need to segment customers using value-based and needs-based approaches. Traditional approaches such as mileage-based segmentation are no longer adequate.

2. In developing systems of customer relationship management, airlines can no longer pursue the "fast follower" approach. They must be proactive in pursuing initiatives that respond to the needs of their customers and also provide a high rate of return.

3. Airlines need to pursue an organizational design that imbues employees with a service mentality and provides them with a clear understanding of their role and a complete view of the customer.

Value-based segmentation allows an airline to understand the profitability of each customer. This allows for the identification of high-valued customers to be retained and lower-valued customers who might be migrated upward to higher-value segments. The white paper suggests that instead of grouping customers by demographic characteristics or frequent flyer program attributes alone, an airline might want to characterize customers along the two dimensions of monetary value and travel frequency.

As argued above, airlines must proactively address the needs of their high-valued customers. For example, there may be considerable value in in-flight Internet and e-mail access for customers concerned about personal work productivity. Using direct and indirect customer inputs, such as complaints and requests received, customer satisfaction surveys, government ratings, focus-group research, competitor analysis, anecdotal reporting by employees, and customer behavior and technology usage, airlines can gain insights into the habits and needs of these high-valued customers.

Such proactive strategic behavior is also important in addressing the needs of lower-valued customer segments and in lowering overall costs. For these lower-valued segments, self-service initiatives can be quite successful with relatively low implementation costs. Drawing from the example of the banking industry, airlines may want to drive customers in these segments to self-service options by charging for "high-touch" services. At the same time, use of these self-services can be rewarded to reinforce such behavior. Some airlines already offer frequent flyer rewards for purchasing tickets online or for using self check-in kiosks.


The major international airlines have looked at the low-cost airlines in an attempt to achieve better operational efficiency, thus improving profit margins. However, the discussion above suggests that improved margins will not be achieved solely by means of simple models of emulation. This study empirically explored the identification of the proper strategic foci, with regard to customer services, in the pursuit of the desired improvement in operational efficiency. It utilized operational data from fiscal year 2000, the last full year before the events of September 11, 2001, in examining the operational impacts of the two categories of customer services: passenger services and ticketing, sales, and promotion.

Among the items included in passenger services are the premiums for passenger liability insurance and passenger accident insurance; meals and accommodation, including the cost of supplies and personal services furnished to passengers; the expense of handling passengers incurred because of interrupted flights, including hotels, meals, taxi fares, and other expense items; the cost of other services provided to passengers such as pay, allowances, and expenses of passenger service personnel; and all other services provided for the comfort of passengers in transit. It also includes pay and allowances, pensions, insurance, traveling, and other similar expenses (uniforms, etc.) of cabin crew as well as the training costs of cabin crew.

Ticketing, sales, and promotion includes pay, allowances, and related expenses of all staff engaged in reservations, ticketing, sales, and promotion activities; accommodation costs; agency fees for outside services; and advertising and publicity through various media, and expenses related thereto. It also includes the net commission payable to others for the sale of transportation on the reporting carrier's service less the commission receivable from the reporting carrier's sale of transportation on other air carriers' services.


The operational efficiency model used in this study is derived from that used by Schefczyk (1993). He defines an input-output model characterized by three inputs and two outputs. Fethi et al. (2002) note that both physical units and financial values can be used to measure the relevant variables. In this study, financial values were used when no corresponding data were available for physical units. This was the issue faced by Schefczyk and it is his combination of financial values and physical units that is used in this study. The inputs are available ton-kilometers, operating cost, and non-flight assets. The outputs are revenue passenger-kilometers and non-passenger revenue ton-kilometers. Available ton-kilometers reflect available aircraft capacity. The value for operating cost reflects operating cost excluding capital and aircraft cost captured by available ton-kilometers. Non-flight assets reflect all assets not included in available ton-kilometers such as facilities, reservation systems, and current assets.

Data envelopment analysis (DEA) was used to compute relative efficiency scores for each of the airlines in this study. This methodology has been used in the transportation literature to assess the relative efficiency of firm operations. Mejza and Corsi (1999) utilized DEA to assess motor carrier safety processes. Peck, Scheraga, and Boisjoly (1997) employed this framework to examine the relative efficiency of aircraft maintenance technologies. In a similar manner, Poli and Scheraga (2002) assessed motor carrier maintenance strategies with this methodology. Schefczyk (1993) used this technique in the study described above. A study done by Scheraga (2004) examining the relationship between airline operational efficiency and financial mobility linked data envelopment efficiency scores to a tobit analysis as described below. Other studies that employ this two-stage approach are own and Yu (1994), Kerstens (1996), and Gillen and Lall (1997).

The particular model employed was the input-oriented model specified by Ali and Seiford (1993). This model is much akin to that of Charnes et al. (1978). Like traditional microeconomic analysis, DEA uses the concept of the frontier of the production possibility set where the production possibility set consists of the feasible input and output combinations that are available from a particular production technique or plan. A particular production plan is efficient if there is no way to produce more output with the same inputs or to produce the same output with fewer inputs. DEA is a nonparametric technique that estimates a best practice production frontier from the observed inputs and outputs of individual airlines. DEA derives, using a projected point on the efficient frontier, how an airline can move from its point of actual operation to a point of relative operational efficiency. In the case of the input-oriented model, the methodology seeks a projected point on the efficient frontier such that the proportional reduction in inputs is maximized. The implicit underlying premise in such a model is that the primary objective of an airline under evaluation is to gain efficiency by reducing excess input utilization while continuing to operate with its current technology mix.

For the input-oriented model, IOTA measures the proportionate level of all inputs a relatively efficient airline should be using to produce its current level of output. Thus a relatively efficient airline would have a value of IOTA equal to one, which implies that it should be using 100 percent of its current input levels. Alternatively, the variable 1-IOTA measures the possible proportionate reduction in inputs, which for a relatively efficient airline would be zero. The efficiency measure IOTA does convey information about possible reorientations of an airline's production strategy. Consider the following case: Suppose Airline A has a peer group of airlines that have comparatively efficient production techniques allowing them to achieve the levels of output of Airline A more efficiently. If IOTA is very small, then the production technique of Airline A is really off the mark. Attention should be focused on a shift of the input/production technique. If, on the other hand, IOTA is close to 1, then the airline could remain with its current production technique and achieve the same levels of output with a small scaling down of inputs.


Once the efficiency scores were derived from the data envelopment analysis, they were regressed against a set of operational and environmental variables, which previous studies have shown to impact operational efficiency (Caves et al. 1984, Banker and Johnston 1993, Schefczyk 1993, Siau and Van Lindt 1997, and Fethi et al. 2002). These include average flight length, passenger revenues as a percentage of total revenues, scheduled service revenues as a percentage of total revenues, international passenger revenue kilometers as a percentage of total passenger revenue-kilometers, average load factor, and the percentage of state ownership in the airline. These variables, in effect, describe the operating environment of an airline.

Average flight length captures economies of distance. It has been suggested that there is a correlation between average flight length and unit cost. This occurs because for a given aircraft size, increasing the distance of a flight results in larger output volume as measured either in passenger-revenue-kilometers or ton-kilometers. However, it must be noted that empirically this posited effect has been shown to be ambiguous (Caves et al. 1981 and Trethaway 1984).

Passenger revenues as a percentage of total revenues represent the passenger focus for an airline. As Oum and Yu (1999) note, air cargo accounts for a large portion of total output for many Asian and European carriers based in export-oriented countries. U.S. carriers have traditionally been primarily passenger-focused in their operations. Cargo service is seen as requiring less input than passenger services, but it generates less revenue. O'Connor (2001) does suggest specific operational advantages in carrying cargo as opposed to passengers. Cargo is usually carried one-way, while passengers usually travel roundtrip. Passengers have a preference for day travel, while cargo generally moves at night. Unlike passengers, shippers and recipients of cargo are not concerned about indirect routes or plane changes as long as the cargo arrives when expected. Additionally, the aesthetics of the aircraft environment are not a concern in the transport of cargo. Thus the anticipated impact of this variable is unclear.

Scheduled service revenues as a percentage of total revenues is anticipated to have a positive impact on operational efficiency. Scheduled flights require different product and marketing facilities than unscheduled charter flights. An increase in the percentage of regularly scheduled services allows for a rationalization of operational routines leading to greater overall efficiency performance.

International passenger revenue kilometers, as a percentage of total passenger revenue-kilometers, captures the international focus of an airline. A priori, it is not unambiguous what the impact of this measure should be, although there are arguments to suggest a potential negative influence on operational efficiency. Fethi et al. (2002) suggest that an increase in the international focus of an airline exposes it to spatial disparities in its operating environment. In structuring bilateral agreements, the international air transport system has tended to focus on individual or small sets of routes between countries. This has impeded the achievement of high levels of efficiency over global networks of air services. There are unresolved issues with regard to ownership and control, cabotage, and the right of establishment. There is still divergence across geographic regions with regard to competition law and policy in air transport. There are differences in fiscal policies, with air transport being subjected to many taxes of varying levels whose function is to finance general governmental expenditure. Further, customs clearance can impede both speed and reliability. Finally, airport infrastructure constraints can significantly affect the level of competition in particular markets.

There is evidence of a positive correlation between average load factors and operational performance (Caves et al. 1981, 1983). Oum and Yu (1999) suggest that average load factor reflects an airline's control on the choice of aircraft and flight frequencies. A higher load factor indicates better utilization of aircraft and thus it should positively impact operational efficiency.

Doganis (2001) suggests a variety of idiosyncratic factors that impact the ability of state-owned airlines to operate efficiently. These factors either impede or encourage state-owned airlines to operate with a less than optimal production technology. Consequently, the impact of the operational variables delineated above may be differentiated in the case of these air carriers.

State airlines have received indirect subsidies from their governments, which explicitly reduce their operating costs. Such support makes these airlines vulnerable to influences from taxpayers and governments who impose obligations on management. National cultural constraints may impose further constraints on management and employment practices. Artificially reduced domestic fares may generate high demand and load factors. However, such numbers may obscure the fact that routes facing such excess demand may have become unprofitable. Some governments expect their state-owned national carriers to fly scheduled services to certain domestic and international destinations in order to promote domestic, social, or economic objectives or simply to "show the flag." Poor financial performance means that these airlines cannot optimally re-equip their fleets, which tend to exhibit a non-rationalized array of aircraft types. Finally, state-owned airlines are typically known for poor air and ground service quality.

To capture these effects, several variables were added to the regression analysis. A variable measuring the percentage of state ownership in a given airline was added to capture operating sub-optimality caused by the degree of non-privatization. Additionally, interaction terms between the percentage of state ownership and average flight length, passenger revenues as a percentage of total revenues, international passenger revenue kilometers as a percentage of total passenger revenue-kilometers, and average load factor were added to capture any modifying impacts on these operational variables.

Three focus variables were added to the set of explanatory variables in the regression. These are operational expenses (not including passenger services and ticketing, sales, and promotion) per available ton-kilometer; passenger services expenses per revenue passenger-kilometer performed; and ticketing, sales, and promotion expenses per revenue passenger-kilometer performed. The denominators of these focus variables are those suggested by Siau and Van Lindt (1997).

A rationale for utilizing these three focus variables is provided by Shank and Govindarajan (1993). They suggest that an airline's value chain consists of three stages:

1. Providing reservation information and ticketing services.

2. Operating the aircraft from point A to point B.

3. Providing service to the passengers before flight, during flight, and after arrival.

Thus the three focus variables are an attempt to capture an airline's strategic emphasis on each of its value chain stages and its allocation of operating resources to each of these stages. The construction of the focus variables is that suggested by Shank and Govindarajan.

IOTA is a censured variable, i.e., 0 [less than or equal to] IOTA [less than or equal to] 1. Utilization of ordinary least squares would yield biased and inconsistent parameter estimates. The more appropriate technique (tobit analysis) is that developed by Tobin (1958) for a left-censured variable. To transform IOTA into a left-censured variable, a transformation suggested by Fethi et al. (2002) was used. The new variable is defined as (1/IOTA) -1, which is greater than or equal to zero in a continuous fashion. The tobit analysis was performed using the LIFEREG procedure in the SAS statistical package.


The current study employed data for thirtyeight airlines for fiscal year 2000. This sample represents large international carriers, with nine from North America, ten from Europe, six from Latin America, twelve from Asia, and one from the Middle East. The sample membership is listed in Table 1, along with the percentage of state ownership in each airline.

The financial and operating data used in this study came from the International Civil Aviation Organization's two publications on financial data: Commercial Air Carriers, Series F and Traffic: Commercial Air Carriers, Series T for 2000 and 2001. The 2001 data were necessary so that the traffic data could be carefully matched to the financial data to allow for the fact that different airlines had different fiscal year ending dates. All financial data are converted to U.S. dollars at the rate of exchange, which is the average of the twelve month IATA rates for the year reported. In those cases where the rate changed considerably only in the last month of the financial year, the rate prevailing prior to this change was adopted, and if the rate for a currency changed frequently during the year reported, a twelve-month average was taken. Because exchange rates capture the effects of changes in relative inflation and interest rates as well as trade deficits or surpluses, the conversion of all financial values to a common currency helps capture the impacts of idiosyncratic forces on country-specific economies. Information on the percentage of country ownership in an airline was obtained from Doganis (2001). The software used for the data envelopment analysis was an updated version of that employed by Schefczyk (1993), Integrated Data Envelopment Analysis System, Version 6.1.7, obtained from 1 Consulting.


The results of the data envelopment analysis are also presented in Table 1. IOTA ranged from a low of 0.73 to a high of 1.00. Twenty of the thirty-eight airlines in the sample were operating below the efficient frontier. So, for example, Avianca, which had an IOTA of 0.73, could have reduced all inputs proportionately by 27 percent and still produced the same level of outputs. Other notable examples where an airline could have reduced all inputs proportionately by more than 10 percent include Austrian Airlines, Air Canada, Canadian Airlines, All Nippon Airways, Kuwait Airways, Pakistan International Airlines, LOT, Scandinavian, and US Airways.

Table 2 presents the mean values of the explanatory variables used in the tobit analysis. The results of the tobit analysis are presented in Table 3. (Recall that for the transformed value of IOTA, an efficient airline will have a value of zero, while an inefficient airline will have a value greater than zero. Thus variables positively correlated with IOTA will be negatively correlated with the transformed value of IOTA.) Nine variables were statistically significant. The international percentage of operations had a negative impact on operational efficiency, suggesting that the spatial disparities discussed above had a significant impact. Average load factor had the expected positive impact on operational efficiency. The negative impact of the percentage of total operating revenues from passenger services is interesting. This may very well reflect the differing characteristics of passenger service and air cargo, as suggested above. However, the percentage of total operating revenues from scheduled services had a positive impact, indicating that opportunities for rationalization of service schedules had the anticipated impact.

Of particular interest, however, is the positive impact on efficiency of the interaction term between the percentage of country ownership and the percentage of total operating revenues from passenger services. Again, as Doganis notes, it is not uncommon for state-owned airlines to operate in environments of very low cargo tariffs so structured to stimulate the export of particular commodities. The result may be stimulated cargo traffic but with unprofitably low yields over sub-optimally scheduled networks. In these cases, the ability to reduce the degree of dependency on revenues from cargo may, in fact, increase operational efficiency. Similarly, the need for state-owned airlines to operate over networks of less than optimal schedule configuration and the desire of governments for their national airlines to operate international routes for their own economic objectives or to "show the flag" may account for the negative impact of the interaction term between the percentage of country ownership and average flight length. Also of interest is the negative impact on operational efficiency of the interaction term between the percentage of country ownership and the percentage of total operating revenues from scheduled services. This suggests that state ownership may mitigate against the positive rationalization benefits of scheduled as opposed to non-scheduled services.

The results for the three focus variables are of particular interest. A more intensive operations focus had a statistically significant negative impact on operational efficiency. This suggests that airlines may have been investing in operational technologies that were less than optimally efficient so that inputs were being used in inefficient combinations. Focusing on passenger services had a statistically negative impact on operational efficiency. Thus, a "no-frills" approach to customer service that eliminates such nonvalue-based activities does indeed enhance operational efficiency. Ticketing, sales, and promotion did exhibit a statistically significant positive impact on operational efficiency. This is consistent with the proliferating voluntary commitments being made by ATA, IATA, and European Airlines, as discussed above. It may also be consistent with attempts to reduce cost of sales and to exploit information system technologies.


In an attempt to develop better customer-focused products, Scandinavian Airlines System utilized a model developed by Kano (1993) (Ekdahl et al. 1999). The value-based perspective argued above to be critical in a customer service analysis was not an overarching objective in this particular exercise. Kano defined three categories of product requirements that influence customer satisfaction: must-be requirements, one-dimensional requirements, and attractive requirements. Must-be requirements are the basic criteria of the product or service. While the fulfillment of these requirements will not increase the customer's satisfaction, if they are not fulfilled the customer will be extremely dissatisfied. One-dimensional requirements are usually demanded by the customer in an explicit manner. Here, customer satisfaction is proportional to the level of fulfillment. Attractive requirements are neither specifically expressed by the customer nor are they expected. The fulfillment of these requirements leads to a more than proportional level of satisfaction. However, if these requirements are not met, there is no feeling of dissatisfaction on the part of the customer.

SAS saw the specific challenge it faced as being able to provide attractive and exciting attributes to its customers. This implied a need for more product variations and services with few limitations. SAS sought to allow customers to design their own travel experience in order to identify attractive requirements above and beyond one-dimensional and must-be requirements. To understand the interaction between itself and its customers, SAS amassed thousands of hours of video and photographic data of passengers on the ground and in-flight. From this information, three categories of customer activities were identified: procedural activities, personal activities, and planning and preparing activities. In identifying which areas were suffering most from a lack of customer focus, SAS developed three design principles: give passengers control, make the process transparent, and empower the staff. These steps then led SAS to prioritize in-flight, ground, and information services.

However, it is interesting to note that in terms of the current study, SAS achieved the third lowest efficiency score in the sample of thirty-eight airlines. Its IOTA value of 0.78 indicates that it could have continued to produce the same level of output but with a 22 percent proportional reduction in all inputs. Not surprisingly, in the sample of this study, SAS had the second highest level of expenditures of passenger services, consistent with its perceptions formed from its customer service study. It should be noted that All Nippon Airways, which had the highest level of these expenditures, also displayed an IOTA value of 0.74. Given the empirical results of this study, these two airlines were prioritizing an activity, passenger services, which had a demonstrated statistically significant negative impact on operational efficiency.

This study suggests that airlines need to focus on those core competencies that allow them to design the operation of their networks of air services from a value-based perspective. Airlines clearly need to be customer-focused. However, further liberalization and overcapacity in some markets, coupled with the impact of low-cost carriers and the commoditization of air travel, will lead to a long-term decline in yields (Doganis 2001). Perhaps airlines should outsource as many non-core functions as possible. This may mean abandoning peripheral services such as catering or ground handling services. Should they not want to do this, they may be required to perform these activities as external specialist companies defined to be independent profit centers. In any case, it is not clear that the traditional paradigm of the self-sufficient airline, which provides all of its ancillary support services in-house, can survive.

On a positive note, many of the current voluntary commitments by member airlines of the ATA, IATA, and European Economic Community, with regard to the provision of services to enhance the quality of the customer experience, include a large number of the factors captured in the ticketing, sales, and promotion variable. It is precisely this variable that is shown, in this study, to enhance operational efficiency. Thus, such commitments reflect a prioritization of services to customers that reflect a valued-based perspective.
Table 1. Airline Sample Membership, Percentage State Ownership, and
Efficiency Parameter IOTA

 % State
Airline Country Ownership IOTA

Aerolineas Argentinas Argentina 5 1.00
Austrian Airlines Austria 52 0.84
Biman Bangladesh Bangladesh 100 0.90
Lloyd Aereo Boliviano Bolivia 48 1.00
Air Canada Canada 0 0.84
Canadian Airlines Canada 0 0.86
Lan Chile Chile 0 1.00
Cathay Pacific China, Hong Kong 0 1.00
Avianca Colombia Colombia 0 0.73
Czech Airlines Czech Republic 84 1.00
Air France France 64 1.00
Lufthansa Germany 0 0.99
Air India India 100 0.93
All Nippon Airways Japan 0 0.74
Japan Airlines Japan 0 1.00
Kuwait Airways Kuwait 100 0.89
Malaysia Airlines Malaysia 25 0.92
Aeromexico Mexico 0 0.92
Mexicana Mexico 0 0.90
KLM Royal Dutch Airlines Netherlands 25 1.00
Pakistan International Airlines Pakistan 56 0.86
LOT Poland 52 0.84
Asiana Republic of Korea 0 1.00
Korean Air Republic of Korea 0 1.00
Scandinavian Scandinavia 50 0.78
Singapore Airlines Singapore 54 1.00
Iberia Spain 54 0.93
Srilankian Sri Lanka 74 0.99
Thai Airways International Thailand 93 1.00
British Airways United Kingdom 0 0.99
Virgin Atlantic United Kingdom 0 1.00
American Airlines United States 0 1.00
Continental Airlines United States 0 1.00
Delta Air Lines United States 0 1.00
Northwest Airlines United States 0 1.00
Trans World Airline United States 0 0.92
United Airlines United States 0 1.00
US Airways United States 0 0.84

Table 2. Mean Values of Tobit Regression Explanatory Variables (N=38)

Variable Mean Minimum

Percentage state ownership (%) 27.261 0
International percentage of operations (%) 74.453 13.430
Average flight length (km) 2,059.368 770.000
Load factor: ton-kilometers (%) 62.389 44.430
Percentage of total operating revenues
 from passenger services (%) 82.421 56
Percentage of total operating revenues
 from scheduled services (%) 91.737 79
Operations focus($/ATKm) 0.352 0.187
Passenger services focus($/TRPKm) 0.009 0.002
Ticketing, sales, promotion focus($/reve-
 nue passengers) 0.029 0.002

Variable Maximum

Percentage state ownership (%) 100
International percentage of operations (%) 100
Average flight length (km) 8,978.700
Load factor: ton-kilometers (%) 80.600
Percentage of total operating revenues
 from passenger services (%) 97
Percentage of total operating revenues
 from scheduled services (%) 100
Operations focus($/ATKm) 0.689
Passenger services focus($/TRPKm) 0.026
Ticketing, sales, promotion focus($/reve-
 nue passengers) 0.062

Table 3. Regression Results--Tobit Model (Dependent Variable:
Transformed IOTA 2000 = (1/IOTA 2000)-1)

Variable Estimate Std. Error

Intercept 133179 0.4954
Percentage state ownership
 (%/102) 1.3914 1.5921
International percentage of
 operations (%/102) 0.4192 0.0869
Average flight length (km/103) -0.0580 0.0418
Load factor: ton-kilometers
 (%/102) -1.5291 0.2855
Percentage of total operating
 revenues from passenger
 services (%/102) 0.9530 0.3592
Percentage of total operating
 revenues from scheduled
 services (%/102) -1.6860 0.7172
Operations focus($/ATKm) 0.5345 0.2993
Passenger services focus
 ($/TRPKm) 13.9647 3.5321
Ticketing, sales, promotion focus
 ($/TRPKm) -7.7019 4.4215
Percentage state ownership x
 international percentage of
 operations -0.0426 0.3690
Percentage state ownership x
 average flight length 0.1514 0.0534
Percentage state ownership x
 load factor -1.4478 1.1154
Percentage state ownership x
 percentage of total operating
 revenues from passenger
 services -2.7653 0.8085
Percentage state ownership x
 percentage of total operating
 revenues from scheduled
 services 1.6011 1.1095

Variable Chi-square PR > Chi Sq.

Intercept 7.0766 0.0078
Percentage state ownership
 (%/102) 0.7638 0.3821
International percentage of
 operations (%/102) 23.2494 <0.0001
Average flight length (km/103) 1.9261 0.1652
Load factor: ton-kilometers
 (%/102) 28.6842 <0.0001
Percentage of total operating
 revenues from passenger
 services (%/102) 7.0412 0.0080
Percentage of total operating
 revenues from scheduled
 services (%/102) 5.5262 0.0187
Operations focus($/ATKm) 3.1895 0.0741
Passenger services focus
 ($/TRPKm) 15.6312 <0.0001
Ticketing, sales, promotion focus
 ($/TRPKm) 3.0342 0.0815
Percentage state ownership x
 international percentage of
 operations 0.0134 0.9080
Percentage state ownership x
 average flight length 8.0286 0.0046
Percentage state ownership x
 load factor 1.6847 0.1943
Percentage state ownership x
 percentage of total operating
 revenues from passenger
 services 11.6997 0.0006
Percentage state ownership x
 percentage of total operating
 revenues from scheduled
 services 2.0826 0.1490


Ali, A. and L. Seiford (1993), "The Mathematical Programming Approach to Efficiency Analysis," in Fried, H. Lovell, C. and S. Schmidt., eds,. The Measurement of Productive Efficiency, New York, New York; Oxford University Press.

Banker, R.D. and H.H. Johnston (1993), "An Empirical Study of Cost Drivers in the U.S. Airline Industry," The Accounting Review, July, pp. 576-601.

Caves, D.W., L.R. Christensen, and M.W. Trethaway (1981), "U.S. Trunk Air Carriers 1972-1977: A Multilateral Comparison of Total Factor Productivity," in Cowing, T. G. and R. E. Stevenson, eds., Productivity Measurement in Regulated Industries, New York, New York; Academic Press.

Caves, D.W., L.R. Christensen, and M.W. Trethaway (1983), "Productivity Performance of US Trunk and Local Service Airlines in the Era of Deregulation," Economic Inquiry, Vol. 21, pp. 312-324.

Caves, D. W., Christensen, L. R, and M. W. Trethaway (1984), "Economies of Density versus Economies of Scale: Why Trunk and Local Service Airlines Costs Differ," Rand Journal of Economics, Winter, pp. 471-489.

Charnes, A., W. Cooper, and E. Rhodes (1978), "Measuring the Efficiency of Decision Making Units," European Journal of Operational Research, Vol. 2, No, 6, pp. 429-444.

Doganis, R., The Airline Business in the 21st Century, London, England and New York, New York; Routledge, 2001.

Ekdahl, F., A. Gustafsson, and B. Edvardsson (1999), "Customer-Oriented Service Development at SAS," Managing Service Quality, Vol. 9, No. 6, pp. 403-410.

Fethi, M., P. Jackson, and T. Weyman-Jones (2002), "Measuring the Efficiency of European Airlines: An Application of Tobit Analysis," Working Paper, Leicester, United Kingdom; University of Leicester, Management Center.

IBM Institute for Business Value, The Future of CRM in the Airline Industry: A New Paradigm for Customer Management, Somers, New York; IBM Global Services, 2001.

Gillen, D. and A. Lall (1997), "Developing Measures of Airport Productivity and Performance: An Application of Data Envelopment Analysis," Transportation Research Part E, Vol. 33, no. 4, pp. 261-273.

International Civil Aviation Organization, Financial Data: Commercial Air Carriers, Series F Montreal, Quebec, Canada; 2000.

International Civil Aviation Organization, Traffic: Commercial Air Carriers, Series T, Montreal, Quebec, Canada; 2000 and 2001.

International Civil Aviation Association, Worldwide Air Transport Conference: Challenges and Opportunities of Liberalization, Montreal, Canada; March 24-29, 2003.

Kano, N., N. Seraku, F. Takahashi, and S. Tsuji, (1993), "Attractive Quality and Must-be Quality," The Journal of the Japanese Society for Quality Control, April, pp. 39-48.

Kerstens, K. (1996), "Technical Efficiency Measurement and Explanation of French Urban Transit Companies," Transportation Research Part A, Vol. 30, No. 6, pp. 431-452.

Mejza, M. and T. Corsi (1999), "Assessing Motor Carrier Potential for Improving Safety Processes," Transportation Journal, Vol. 38, No. 4, pp. 36-50.

O'Connor, W. E., An Introduction to Airline Economics, Westport, Connecticut; Praeger, 2001.

1 Consulting, Integrated Data Envelopment Analysis System, Version 6.1.7, Amherst, Massachusetts, 2000.

Oum, T. and C. Yu (1994), "Economic Efficiency of Railways and Implications for Public Policy: A Comparison of the OECD Countries' Railways," Journal of Transport Economics and Policy, Vol. 28, No. 2, pp. 121-138.

Oum, T. and C. Yu, Winning Airlines: Productivity and Cost Competitiveness of the World's Major Airlines, Norwell, Massachusetts; Kluwer Academic Publishers, 1999.

Peck, M., C. Scheraga, and R. Boisjoly (1997), "Assessing the Relative Efficiency of Aircraft Maintenance Technologies: An Application of Data Envelopment Analysis," Transportation Research Part A, Vol. 32, No. 2, pp. 261-269.

Poli, P. and C. Scheraga (2002), "A Quality Assessment of Motor Carrier Maintenance Strategies: An Application of Data Envelopment Analysis," Quarterly Journal of Business and Economics, Vol. 40, No. 1, pp. 24-43.

SAS Institute, Inc., SAS/STAT User's Guide, Version 8, Cary, North Carolina, 2002.

Schefczyk, M. (1993), "Operational Performance of Airlines: An Extension of Traditional Performance Paradigms," Strategic Management Journal Vol.14, pp. 301-317.

Scheraga, C. (2004), "Operational Efficiency versus Financial Mobility in the Global Airline Industry: A Data Envelopment and Tobit Analysis," Transportation Research Part A, Vol. 38, No. 5, pp. 383-404.

Shank, J., and V. Govindarajan, Strategic Cost Management, New York, New York; The Free Press, 1993.

Siau, C. and D. Van Lindt (1997), "Cost Driver Analysis in the Airline Industry," Journal of Cost Management, Vol. 11, pp. 37-47.

Tobin, J. (1958), "Estimation of Relationships for Limited Dependent Variables," Econometrica, Vol. 26, pp. 24-36.

Trethaway, M. (1984), "An International Comparison of Airlines," Proceedings of the Canadian Transportation Research Forum, pp. 34-43.

Mr. Scheraga, EM-AST&L, is associate professor of business strategy and technology management, Charles F. Dolan School of Business, Fairfield University, Fairfield, Connecticut 06824; e-mail
COPYRIGHT 2004 American Society of Transportation and Logistics, Inc.
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 2004 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Author:Scheraga, Carl A.
Publication:Transportation Journal
Geographic Code:1U2NY
Date:Jun 22, 2004
Previous Article:Importing and exporting: a longitudinal study of process component importance and expertise.
Next Article:U.S.-Canada transportation and logistics: border impacts and costs, causes, and possible solutions.

Terms of use | Copyright © 2017 Farlex, Inc. | Feedback | For webmasters