Printer Friendly

Comparative analysis of efficiency for major Northeast Asia airports.

Abstract

Growing airport capacity constraint, airports' increasing attention to profits, and regional hub formation in Northeast Asia suggest a need to study efficiency performance of the region's major airports. This article empirically evaluates the level and change of efficiency (productivity) of Tokyo (Narita), Osaka (Kansai), Seoul (Incheon and Gimpo), Beijing Capital, Shanghai Hongqiao, Guangzhou, and Hong Kong airports. We apply data envelopment analysis (DEA) to new panel data covering the 1994-2007 time period. We find that all the sample airports have recorded an increase in productivity over the period. Given the initial low productivity level of Chinese airports, this finding suggests that a series of deregulatory measures adopted by the Chinese government may have worked in improving airport productivity. On the other hand, no sample airports have experienced consistent growth in productivity. To improve productivity, the airports in Northeast Asia need not only technological revolutions at the industry level, but also technical improvements, such as better airport management, operations, and investment, at an airport level. Alternatively, the region may need to adopt more aggressive liberalization measures to stimulate further traffic growth. A number of future research areas are also discussed.

**********

Northeast Asia (NEA) has been the world's fastest growth region for the past several decades. Its three principal countries, namely, Japan, Korea, and China, all have achieved episodes of robust economic growth: Japan from the 1950s to the 1970s, Korea from the mid-1960s to the mid-1990s, and China from 1978 to the present. (1) Together they now represent about 25 percent of the world income. This rapid economic growth has significantly increased the demand for air transport services, which in turn places enormous pressure on airport infrastructure. At least eight major international airports have been built in Northeast Asia since 1991. (2) In addition, significant expansions have been, or are being, undertaken at major NEA airports, including Beijing Capital, Tokyo Narita, and Tokyo Haneda. Nevertheless, traffic growth outpaces capacity increase, creating an urgent need to improve productivity of airports to relieve demand pressure.

The need for productivity improvement also arises from the recent developments of airport corporatization and privatization in the region. Productivity certainly matters to a private airport as it will affect the airport's profit. However, with airport corporatization, even public airports have been under growing pressure from governments to be more financially self-sufficient and hence to improve productivity (e.g., Poole 2003; Zhang and Zhang 2003; Zhang and Yuen 2008).

Furthermore, while there has been substantial progress in regulatory liberalization over the last two decades, the liberalization in NEA is largely bilateral in nature. Multilateral agreements governing air services among China, Japan, and Korea have yet to be developed. As a result, the structure of airline networks remains fragmented, and at the moment there are no real hub-and-spoke operations by NEA airlines either domestically or in the NEA region (e.g., Zhang 2010). Nevertheless, the continued air transport liberalization, both within and outside of the NEA region, will facilitate airlines' network reorganization towards regional hub-and-spoke. This reorganization will not only contribute to the problem of capacity shortage at major NEA airports (Zhang 2010), but will also intensify competition among major airports in their effort to become the region' s preeminent hubs and intercontinental gateways. Part of such competition hinges on airport efficiency.

Taken together, these three observations--growing airport capacity constraints, airports' increasing attention to profits, and regional hub formation--suggest a need to study efficiency performance of airports in Northeast Asia. This article empirically evaluates efficiency (productivity) of major NEA airports in terms of both the level and change of efficiency. We construct a data set consisting of major airports, namely, Tokyo (Narita), Osaka (Kansai), Seoul (Incheon and Gimpo), Beijing Capital, Shanghai Hongqiao, Guangzhou, and Hong Kong, which, as will be discussed below, are candidate cities/airports to emerge as the mega-hubs in NEA. We apply data envelopment analysis (DEA) to this new panel data covering the 1994-2007 time period. This analysis enables us to examine the efficiency scores of each airport, as well as efficiency performance across airports and over time.

Our empirical results reveal that all the sample airports, including the Chinese airports (especially Beijing airport), have recorded an increase in productivity over the 1994-2007 period. Given the initial low productivity level of Chinese airports, this result suggests that a series of deregulatory measures adopted by the Chinese government may have worked in improving airport productivity. This improvement will be conducive to the Chinese airports' effort to become a preeminent hub in the NEA region. We further find that no sample airports have experienced consistent growth in productivity. To improve productivity, the NEA airports need not only technological revolutions at the industry level, but also technical improvements, such as better airport management, operations, and investment, at the airport level. Alternatively, the region may need to adopt more aggressive liberalization measures to open up the markets and thereby stimulate further traffic growth. As discussed below, the NEA aviation market is quite restricted, especially in comparison to the market in North America or in the EU.

Assessment of airport productivity has become the focus of a large number of studies. Different methodologies have been used to measure productivity of airports, including the widely used DEA approach. For example, Gillen and Lall (1997) applied DEA to assess terminal and airside operations of 21 top U.S. airports from 1989 to 1993. This study is followed by a large number of papers applying DEA to performance evaluation of airports in different countries (e.g., Parker 1999: Sarkis 2000: Vasigh and Hamzaee 2000; Chin and Siong 2001: Martin and Roman 2001, 2006; Pels et al. 2001, 2003; Abott and Wu 2002; Fernandes and Pacheco 2002; Bazargan and Vasigh 2003; Sarkis and Talluri 2004; Pestana Barros and Dieke 2007; see Lam et al. 2009 for a useful literature review). The majority of these papers study airports in North America and Europe. This is not surprising given that, in terms of the number of passengers carried, these two regions together accounted for more than 60 percent of the world traffic. In addition, many major liberalization events concerning aviation have involved countries in these two continents. While Northeast Asia plays an increasingly important role in world aviation, research on efficiency comparison of NEA airports is lacking, owing partly to lack of data, especially for Chinese airports. DEA studies on NEA airports have recently emerged, however. Yoshida and Fujimoto (2004) examined 67 Japanese airports using data from the year 2000; Yu et al. (2008) looked at the productivity growth of four major Taiwanese airports for the 1993-1999 time period; and Fung et al. (2008) measured productivity for 25 Chinese airports over the 1995-2004 time period. Notably, Fung et al. (2008) investigated whether airport efficiency in China was improving over time and whether productivity among the airports from different regions was converging. They found positive answers to both questions.

The present study aims to add to the body of studies on NEA airports by examining airports in Korea, Japan, and China. This multi-country approach is also in contrast to most of the above studies that examined airports within a single country or markets equivalent to domestic markets. An example of the latter case is that of Pels et al. (2003). Their study covered EU airports for the 1995-1997 period, during which the EU had started the third Package of its common market, and so it was effectively a liberalized EU "domestic" market. In addition, we work on panel data, in contrast to the dominant cross-sectional approach in the literature. Our data cover a longer period of time during which major regulatory changes had occurred in the three countries, allowing an analysis of airport efficiency trends during the period. The article is organized as follows: The next section describes recent developments in the NEA aviation market and discusses their implications for airports. Then, the empirical methodology, airports, and data employed in our analysis are discussed. Reports of the results on airport efficiency levels and changes follow, and areas for future research discussed.

AVIATION DEVELOPMENTS IN NORTHEAST ASIA

Air transport is a significant sector in China, Japan, and Korea. These three countries rank second, fourth, and seventh, respectively, in the world in terms of total tonne-kilometers transported in 2007. The future traffic growth for the region will, according to various forecasting agencies, continue to be robust. Considerable attention has, therefore, been paid to the formation of "optimal" aviation policies within each country. Inspired partly by the deregulation and liberalization experiences of the United States and other countries, significant effort has been extended to the deregulation of domestic markets by each NEA country in the last 20 years. In Korea, Asiana Airlines, a trunk carrier, was allowed to enter the industry in 1988 to compete against the incumbent monopoly Korean Air. The increased airline capacity and competition, together with the country's rapid economic growth, resulted in a large increase in air traffic volume. Ten new domestic routes were introduced between 1988 and 1993, as compared to just two routes between 1980 and 1987. On the international routes, the market share of Korean carriers was 48 percent in 1990 when Korean Air was a monopoly, but it rose to 67 percent in 1998 with the duopoly (Park 2000). Major recent developments include the signing of an open-skies agreement with the United States in 1998, and the liberalization of airfare setting for domestic routes in 1999.

In Japan, Skymark Airlines and Air Do, the two low-cost carriers, entered, respectively, the Tokyo-Fukuoka and Tokyo-Sapporo routes in 1998. Although limited in their scope, these had been the first independent entries in the Japanese airline industry since the 1960s. The passage of the new Civil Aeronautics Law in 1999 represents a significant deregulatory step, as it substantially liberalized the operating license system, airfare approval system, and other regulatory provisions. The liberalization also allowed airlines to set fares freely beginning in 2000. In April 2007, the Japanese government unveiled its "Asia Gateway Plan," aimed at removing restrictions on foreign airlines' access to its regional airports, boosting trade and tourism, and addressing the issue of increasing regional economic disparity. The government has opened up 23 regional airports to strengthen its gateway position for international traffic.

That airline liberalization, both domestically and internationally, has come relatively slowly to Japan may have to do with the capacity constraint at its major airports, particularly landing slots at Tokyo Haneda, Tokyo Narita, and Osaka (Itami). In the past Japan's Council for Transport Policy argued that because of airport capacity constraints, an American style of deregulation did not suit the circumstances of Japan (Takahashi 2003). While the overall national capacity has increased in parallel with the deregulation, slot shortage at congested airports has not been resolved. Yamauchi (2000) points out that slot shortage has been the single most important barrier to expanding air services to and from Japan for a long time. (One consequence of the capacity shortage is the high airport charge at Narita. In fact, Narita has been among the most expensive airports in the world.) He attributes two factors to this capacity shortage. First, the Japanese government had rendered too much support to airport construction in local areas as a result of strong political pressure. As a result, Japan is able to build a very dense nationwide airport network, which has nevertheless resulted in an airport capacity shortage in the metropolitan areas. (3) Second, these physical barriers are caused to a large extent by financial barriers, since capacity expansion in the Tokyo area is extremely costly. On the other hand, the budget for airport construction has been substantially cut by other expenditures such as the expenditure for noise pollution abatement.

In addition, there appeared to be problems with airport profitability. For instance, 14 of the 22 airports managed by the (then) Ministry of Land, Infrastructure and Transport lost money amounting to 33.3 billion yen in the 1999 fiscal year (Takahashi 2003). To reduce government subsidies and improve airport management efficiency, Japan announced plans to privatize, via public share offerings, three major international airports: Narita (New Tokyo International), Osaka Kansai, and Nagoya. The original plan would have privatized the three airports as a package and used the proceeds to pay off much of the debt of money-losing Kansai. But after considerable airline protests, the government backed down and announced that the New Tokyo International Airport Authority would be privatized on its own, after first being corporatized in 2004. Shares would be sold in tranches over the subsequent five years (Poole 2003).

The Chinese market shifted from a monopoly to a more competitive market structure in the late 1980s. Since then, China's international aviation policy appears to have shifted away somewhat from the previous restrictive approach to a proactive regime that views aviation as a facilitator of national trade, foreign direct investment, tourism, and economic development (Zhang and Chert 2003). As argued further by Zhang and Chen (2003), the liberalization efforts have contributed not only to a more competitive marketplace, but also to airline productivity growth and the industry's dramatic traffic growth. As for airports, to improve their productivity, China embarked upon an airport localization program, in which airports are turned over to local governments. As a test run, operation of the Xiamen International Airport and Shanghai Hongqiao International Airport (including all fixed and working capital and all personnel) was transferred to their municipal governments in 1988 and 1993 respectively (Zhang 1998). The Civil Aviation Administration of China (CAAC)-the industry's regulator-was, however, still heavily involved in airport business during the late 1980s and 1990s. The localization program regained momentum in the early 2000s and was completed in 2003, when the CAAC transferred ownership and control of all its remaining airports, except the Beijing and Tibet Airports, to their respective local governments. While the localization program increased the initiatives for local and private investment into airport capacity expansion, airport productivity was expected to improve after the implementation of the localization program. As pointed out by Zhang and Yuen (2008), as opposed to the "soft budget" approach previously taken by the CAAC, the localization program made the airports more financially accountable. As part of the localization program, the central government began to phase out its subsidization of airports in 2006. Furthermore, as the efficiency of airports has significant implications for local economies, local governments may have greater incentives to improve airport efficiency than would the CAAC.

The second major policy change that aimed to improve productivity of Chinese airports is to allow their initial public offering (IPO). Although attracting private funds was one rationale for listing, the principal objective was to improve airport efficiency (Zhang and Yuen 2008). Since the IPO of Xiamen airport, six Chinese airport companies have been listed on stock exchanges. Zhang and Yuen (2008) investigated the effect of public listing on Chinese airport productivity. They found that the listed airports had higher efficiency scores than did unlisted airports, while the correlation between productivity growth and listing was statistically insignificant.

Despite the important progress made within each NEA country, the air transport market for the region as a whole is largely restricted. For the past several years there have been some important regional developments in market liberalization. In June 2006, China and Korea signed an "open skies" agreement on the routes between Korea and China's Shandong Province and Hainan Province. This agreement removes capacity restrictions and pricing control, and allows multiple airline designation. As discussed in Lee (2008), the agreement has dramatically reduced airfares on the Shandong routes from about $500 to $100, while stimulating new demand. Furthermore, somewhat surprisingly, Chinese carriers, which were perceived as weaker carriers and hence likely "victims" of liberalization prior to the signing of the agreement, actually gained market shares on the routes at the expense of their Korean counterparts. In August 2007, Korea and Japan signed an "open skies" agreement which liberalized the Third and Fourth Freedom traffic rights with the exception of the routes involving Tokyo area airports. Moreover, the triangular air shuttle (charter) services among Shanghai Hongqiao, Seoul Gimpo, and Tokyo Haneda started in late 2006. These three airports were previously regarded as the "domestic" airports (and hence only handling domestic traffic) in their respective countries.

While bilateral liberalization will continue, efforts to negotiate an NEA "open-skies" bloc will get its own momentum, especially in light of the fact that China, Japan, and Korea have all participated in the discussion with the ASEAN "common air agreement." (4) It is highly likely that an open-skies bloc will be created in the medium term (e.g., 10 years) in the region. Such an open-skies bloc would induce major NEA airlines to set up hub-and-spoke networks in an effort to fully realize economies of traffic density and network as has occurred in the U.S. and EU airline businesses (e.g., Zhang et al. 2009b). When the dust is settled, there would be a very limited number of super-hubs emerging in NEA. More specifically, consider the three global airline alliances, namely, Star Alliance, OneWorld, and SkyTeam, which together account for more than two-thirds of the world's passenger traffic. Each alliance has airlines in the three main markets-North America, Europe, and Asia-as its senior members. These senior members set up mega-hubs in each continent as their major traffic collection and distribution centers. In North America, these hubs are Chicago (for Star's United Airlines), Dallas/Fort Worth (for OneWorld's American Airlines), and Atlanta (for SkyTeam's Delta Airlines), as well as, to a lesser extent, New York and Los Angeles. In Europe, the mega-hubs are Frankfurt (for Star's Lufthansa Airlines), London Heathrow (for OneWorld's British Airways), Paris (for SkyTeam's Air France-KLM), and Amsterdam (for SkyTeam's Air France-KLM).

Recently Air China, the largest carrier in China, and Shanghai Airlines joined Star Alliance (the other two NEA member carriers are All Nippon Airways and Asiana). China Southern joined SkyTeam (the other NEA member carrier is Korean Air, a founding and senior member of SkyTeam). China Eastern and Hainan Airlines, the third and fourth largest carriers in China, have applied to join OneWorld (the existing NEA member carriers are Cathay Pacific, plus its subsidiary Dragonair, and Japan Air Lines).

Based on experiences of the U.S. and Europe, four or five mega-airport hubs will likely emerge in an integrated NEA. Which cities or airports will become such hubs will depend on government policies, as well as positioning by airports and airlines, over the next 5-10 years. Some of the policy instruments are those related to airport developments. Here, policies are designed and infrastructure investment is made to ensure that sufficient hard and "soft" capacities are available for national hub airports, and that these airports are run efficiently. An integral part of this policy formation and positioning is a comparative study of airport efficiency, which is discussed next.

EFFICIENCY STUDY OF MAJOR AIRPORTS IN NORTHEAST ASIA

Methodology

As indicated in the introduction, data envelopment analysis (DEA) has been widely used in measuring airport performance. As compared to other efficiency measurement techniques, the DEA approach does not require any assumption concerning either the technology or behavior of the actors (e.g., cost minimization) (Pels et al. 2001), and it can be done without some detailed operating information (such as input costs).

We use DEA to evaluate airport efficiency. In general, DEA is a method that can be used to assess the technical (productive) efficiency of a firm, which is reflected by the relationship between the outputs the firm produces and the inputs it uses in a given period of time. One way to measure a firm's efficiency is to compare it with other firms' in the same industry. In the simple case where firms in an industry produce a single output with a single input, Farrell (1957) measured efficiency in terms of potential input per unit of output. Here the most efficient firm is used to define the potential input and output. The efficiency measure for any firm in the industry is then defined as the ratio of the potential input to the actual input the firm is using to produce one unit of output. Alternatively, efficiency may be measured in terms of potential output per unit of input.

Empirical applications of such efficiency measurements later became feasible by a nonparametric technique known as DEA, developed by Charnes, Cooper, and Rhodes (1978, 1981) and based on Farrell's (1957) efficiency measurement. While Farrell's original concept is concerned with singular input and output, the DEA method of Charnes, Cooper, and Rhodes (CCR) can deal with the case where firms use multiple inputs to produce multiple outputs. More specifically, similar efficiency measurements can still be obtained in this more complicated case, with the most efficient firms forming an efficient frontier. A DEA model gives an efficiency score for each firm in a given industry. For the output-oriented model, the efficiency score has a value between zero and one. Firms with an efficiency score of unity are located on the frontier in the sense that their outputs cannot be further expanded without a corresponding increase in input. Firms with an efficiency score below one are deemed inefficient. The DEA model defines the efficiency score of any firm as the fraction of the firm's outputs that can be produced for a firm on the efficient frontier with the same level of inputs. So it is an extreme point method and compares each producer with only the "best" producers.

We use the CCR method to assess the productive efficiency of major airports in NEA. More specifically, technical efficiency (TE) of decision-making unit (DMU) k is obtained by solving the following linear programming problem:

[MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII] (1)

where [Y.sub.kr] = the [r.sup.th] output, [x.sub.ki] = the [it.sup.th] input, [u.sub.r] = weight of the [r.sup.th] output, [v.sub.i] = weight of the [i.sup.th] input, s = the number of outputs, m = the number of inputs, and n is the number of DMUs (airports). We use (1) to compute the efficiency score, [E.sub.k], where subscript k indicates an airport whose performance relative to the other airports is under investigation. With a total of n airports being evaluated, [v.sub.i] and [u.sub.r] are the weights of the inputs and outputs, respectively. Defining the efficiency frontier and solving the linear programming (1) gives the efficiency score for one airport. To estimate efficiency scores for all the airports, the linear programming must be solved n times (adjusting index k each time).

While the CCR method is widely regarded as a fundamental model of DEA, various refinements have since been made. (5) In particular, the CCR method assumes a constant return to scale technology, which occurs when a twofold increase in input factors is offset by a similar increase in output factors. To relax this assumption, Banker, Charnes, and Cooper (1984) decompose the CCR TE into "pure technical efficiency" (PTE) and "scale efficiency" (SE), where PTE measures the extent of efficient utilization of inputs, while allowing for variable return to scale. Thus, the model of Banker, Charnes, and Cooper (BCC) allows for potential productivity gains from achieving the optimal size of airport. It may also allow one to see sources of productivity differences among airports. Specifically, the PTE of airport k is obtained by solving the following BCC function:

[MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII] (2)

We shall apply both the CCR and BCC models to our airport data. Since the CCR model assumes constant returns to scale and the BCC model allows variable returns to scale, the efficiencies evaluated from these two models define the scale efficiency.

Both the CCR and BCC models are used to measure the static efficiency of an airport in a certain year and so it may be difficult to infer from these models airport efficiency trends, which is one of the present study's objectives. In order to capture the variations of efficiency over time, Charnes et al. (1985) proposed a technique called "window analysis" in DEA, which assesses the performance of a DMU over time by treating it as a different entity in each period. The DEA-window analysis works on the principle of moving averages, and measures the level of relative efficiency during a certain period (called a "window"). Use of windows helps overcome the problem of potentially unstable efficiency indices produced by the standard, one-year DEA methods (CCR or BCC).

Major Hub Airports

The foregoing discussion on aviation developments in NEA indicates that four or five mega airport hubs will likely emerge in an integrated NEA market. It is less clear, however, where those hubs are located. Based on the North American and European experiences, it appears that Hong Kong likely competes with Guangzhou for one such mega-hub, whereas Shanghai competes with Taipei for another. Notice here that Taipei is in the picture only after Mainland China and Taiwan fully implement the "direct flights" policy, which has been implemented only partially. Finally, it appears that Tokyo, Seoul, Beijing, and Osaka are likely to compete against one another for the remaining two or three mega-hubs. Nevertheless, as indicated earlier, which cities or airports will become such hubs will depend on, among other factors, government policies. With the large stake involved for national winners and losers, "strategic trade policy" will take place by governments to ensure that their own airports eventually emerge as the mega-hubs of the region. Governments of the three NEA countries may, for instance, engage policies with the effect of channeling traffic originating in the other countries' hinterland regions into their own airports for onward carriage to North America or Europe. So all of these candidate cities and airports must be considered in the analysis.

We construct a data set consisting of the major cities and airports mentioned above, namely, Tokyo Narita, Osaka Kansai, Seoul (including both Incheon and Gimpo airports), Beijing, Shanghai Hongqiao, Guangzhou, and Hong Kong. (Unless specified otherwise, these airports will, for convenience, be hereafter referred to as Narita, Kansai, Incheon, Shanghai, Guangzhou, and Hong Kong.) We started with a sample period of 1991-2007, which was later reduced to the present 1994-2007 period due to missing data for Chinese airports. We also note that the Seoul Gimpo data were added into the Incheon data. Further, due to the opening of the new Guangzhou airport in August 2004 and the closing of the old airport in the same month, we adjusted two inputs of the two airports-namely, the terminal size and runway length--on a pro-rata basis for that year.

Variable Construction and Data Sources

To measure productive efficiency "using DEA, one must first identify outputs that an airport produces and the inputs it uses in producing those outputs. The characteristics of the DEA input/output variables are summarized in Table 1. On the input side, we consider two physical capital input measures--runway length and terminal size-along with the number of employees. On the output side, we consider passenger volume, air cargo volume, and the number of aircraft movements as outputs of airports. The number of passengers served is the most commonly used output for measuring airport productivity. A look at the data shows that the outputs of Kansai and Shanghai Hongqiao have been stable over time, while the outputs of the other airports have been increasing steadily.

The data are compiled from various sources. The Japanese data are drawn from the Japan Aeronautic Association (Nippon Koku Kyoukai), 1995-2008; the Ministry of Land, Infrastructure, Transport and Tourism, 2008; and the National Airport Building Association (Zenkoku Kuukou Buil Kyoukai), 1982-2007. The Chinese data are obtained from the Statistical Data on Civil Aviation of China, 19952008, for runway length, terminal size, passenger, cargo, and aircraft movement; and company annual reports of listed airports for the number of employees. Finally, the Korean data are from the Statistical Data on Incheon International Airport Corporation and the Statistical Data on Korea Airports Corporation for runway length, terminal size, passenger, cargo, and aircraft movement; and from the authors' direct contact for the number of employees.

EMPIRICAL RESULTS

Efficiency Scores

The results of the DEA technical efficiency (TE) scores of the sample airports in the period between 1994 and 2007 are shown in Table 2. Guangzhou had a sharp decline of TE in 2005, owing to a huge increase of inputs (especially the runway length) in late 2004, when the airport moved to a new location. Similarly, Incheon had a big drop in TE in 2001, when the Incheon Airport opened in early 2001. Beijing saw a sharp TE decline in 1999 but has since enjoyed a steady increase in TE. Hong Kong has gradually increased TE since 1998, when its new airport opened. The average TE scores have shown an upward trend over time, with the exception of 2003. The sharp decline in 2003 is caused mainly by the SARS breakout in NEA that had a severe negative impact on air travel demand in the region.

The technical efficiency is then decomposed into pure technical efficiency (PTE) and scale efficiency (SE), as shown in Tables 3 and 4, respectively. As in the case of TE, PTE of Guangzhou sharply declined in 2005 due to the airport opening, whereas that of Beijing showed a sudden decrease in 1999, but has since seen a steady rise (especially since 2004). On the other hand, Shanghai and Kansai have kept a very high level of PTE during the entire sample period. In terms of SE, Incheon had a sharp decline in 2001 due to a sudden increase in inputs, whereas the scores for Kansai and Shanghai fell in 2003 because of declines in passenger traffic. On the other hand, Hong Kong and Narita have more or less kept a high level of SE over the period. Other highlights include a sharp increase of SE for Guangzhou since 2004 owing to a large reduction of employees, and Shanghai's significant increase of SE since 2004 owing to output expansion.

We have also conducted a DEA-Window analysis on the data, with Table 5 providing average efficiency scores for both DEA (discussed above) and DEA-Window. Table 5 shows that DEA-Window, while stabilizing the efficiency scores, shows a very similar trend of efficiency scores, as seen from the DEA analysis. Furthermore, the time-series patterns of TE, PTE, and SE under the DEA-Window analysis are drawn in Figure 1. As can be seen from the figure, when the sample NEA airports are taken as a whole, both TE and SE have shown an upward trend. On the other hand, PTE has exhibited a fair amount of fluctuation over time with no noticeable upward trend observed, suggesting that more efforts are needed to improve PTE.

[FIGURE 1 OMITTED]

Productivity Growth Analysis Having examined levels of productivity for the major hub airports, we will now turn to the changes in levels of productivity. This examination is useful in that, if the low level of productivity at some airports is due to their low starting point, then a faster growth rate in productivity could reduce and eliminate the gap. For instance, Fung et al. (2008) found that productivities among Chinese airports from different regions were converging.

The Malmquist productivity index (MPI) is used to measure changes in the overall productivity of each airport over time:

[MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII] (3)

where [D.sub.k] is an output distance function of airport k. Thus, MPI is defined with use of the distance function, referring to the distance between the observed point and point on the production frontier. DEA methods are generally used for the computation of distance functions. Here, the distance function is the inverse of the output-oriented efficiency score calculated above. The superscripts on [D.sub.k] indicate the time periods within which the efficiency scores are calculated. The superscripts on x and y indicate the time periods of the data used in the calculation of the efficiency scores. As a measure of the "total factor productivity change" (TFPC), an MPI [M.sup. t + 1.sub.k]] greater (less) than unity indicates that the overall efficiency of airport k has increased (declined) from period t to period t + 1.

The Malmquist productivity changes can be decomposed into two useful components. Note that equation (3) also represents a decomposition of efficiency change from period t to period t + 1. The ratio outside the bracket on the right-hand side of (3) measures the "technical efficiency change" (TEC) of airport k from period t to period t + 1. Greater (smaller) than unity implies that the technical efficiency has improved (declined) in reference to the production frontier from period t to period t + 1. The bracketed term, on the other hand, represents the geometric mean of the shift in production frontier. When the value of this term is greater (less) than unity, it implies that the technology of the industry has progressed (regressed) from period t to period t + 1. This value measures the extent of airport business innovations and is termed as TC for "technology change."

Tables 6--8 present the MPI results of TFPC, TEC, and TC, respectively. (6) Kansai experienced the highest growth rate of TFPC (of all TFPC growth rates in Table 6) in 1994--95. On the other hand, the overall productivity of Hong Kong dropped by 78 percent in 1997-98, whereas Beijing saw a sharp decline of TFPC in 1998-99. All these TFPC changes are driven by changes in its TEC component. For example, the sharp rise of Kansai's TFPC in 1994 95 was affected predominantly by TEC, as the outputs increased greatly in 1995 (recall Kansai International Airport opened in September 1994). In contrast, Beijing experienced the highest growth of TFPC in 2005-06, but this change is driven by the TC component rather than the TEC component. Furthermore, Table 7 shows that in 1997-98--except for Hong Kong, which saw its TEC fall by 72 percent--most of the sample airports recorded high growth rates in TEC. In comparison, in 199798 all the airports recorded a large fall in TC (see Table 8). The average TC increased by 46 percent in 2003-04, with every airport experiencing an increase in TC (Beijing recorded the highest growth of 72 percent in 2004).

Other noticeable points include the following:

--In 2005-06, Beijing experienced the highest growth of overall efficiency (TFPC), with TEC having a stronger impact on this growth than TC. The increase of TEC was influenced by the significant drop in the number of employees.

--In 1997-98, Hong Kong's TFPC dropped by 78 percent, with TEC having a greater impact on this decline than TC. The drop in TEC (by 72 percent) is due to the expansion of inputs as the new Chek Lap Kok Airport opened and replaced the old Hong Kong Airport in July 1998.

--The productivity rate of Guangzhou Airport fell suddenly in 2004--05, with TEC having a greater impact on the decrease than TC. This is because Guangzhou's new airport opened and replaced its old airport in August 2004.

--Except for 2001, the productivity of Incheon was trending upwards. In 2001, both TEC and TC recorded a sharp drop. The inputs increased dramatically due to the opening of the Incheon International Airport in March 2001.

--The average (over the years) growth rates for individual airports are given in the last columns of Tables 6--8. The indices of TFPC, TEC, and TC are larger than unity, suggesting that all the sample airports recorded an increase in productivity. The productivity increase is especially high for Beijing, suggesting that a series of deregulatory measures adopted by the CAAC may have worked in improving airport productivity. TFPC was affected by TEC and TE at the same time, but the graphs of average productivity changes showed that TC has had a greater effect on TFPC than TEC.

--The average (over the sample airports) growth rates for individual years are given in the last rows of Tables 6--8. Except for 2000-01 and 2002--03, TFPC increased gradually. High growth rates of TEC were recorded in 1994--95 and 1997--98, while TC decreased suddenly in 1997--98 and also experienced a large degree of up-and-down fluctuations.

A number of implications can be drawn from our MPI analysis. First, all the sample airports, including the Chinese airports (especially Beijing airport), have recorded an increase in productivity over the 1994--2007 period. Given the initial low productivity level of Chinese airports, this finding suggests that a series of deregulatory measures adopted by the Chinese government may have worked in improving airport productivity. This improvement will be conducive to these airports' efforts to become preeminent hubs in the NEA region.

Second, TFPC of most sample airports grew at a similar rate to TEC, except for that of Narita. In view of the results, while TFPC is affected by both TEC and TE, the technical efficiency of individual airports appears to have a stronger effect on the overall productive efficiency than the level of technological innovations in the industry.

Third, no sample airports have experienced a consistent growth in productivity, and the rise in the overall average efficiency level over the sample period is not impressively high. To improve productivity, the major airports in NEA need not only technological revolutions and effective policies at the industry level, but also improvement in airport management, operations, and investment so that technical efficiency is improved at an individual airport. Alternatively, the region may need to adopt more aggressive liberalization measures to open up the markets, thereby increasing traffic output and hence TEC. As our earlier discussions indicate, the NEA aviation market is still quite restricted, especially in comparison to the markets in North America or in the EU. These results will be useful not only for governments' policy formulation, but also for airports, both the ones examined in this study and those not examined here.

FUTURE WORK

In addition to yielding useful implications, our study has raised a number of other issues and avenues for future research. First, unlike the present study, in which only a few major hub airports in Northeast Asia are analyzed, it will also be important to study a large number of non-major airports in the region. A follow-up study may evaluate airport performance with, for instance, 8--10 airports from Korea, 20 airports from Japan, and 25--40 airports from China. (7) The associated city pairs are the markets where full-service airlines (FSAs) tend not to operate, but low-cost carriers (LCCs) can thrive with their business model (regional jets, one fare-class, "no-frill" services, less congestion at secondary airports and hence fast aircraft turnaround, etc.). As discussed in Zhang et al. (2009a), the LCC model, while being a much more recent phenomenon and being less successful in NEA than in North America and Europe, will grow and prosper in NEA in the near future. Unlike FSAs, LCCs in North America and Europe de-emphasize hub-and-spoke networks. Instead, they tend to provide point-to-point services linking non-major cities, or using secondary airports in a metro area (Tretheway 2004; O'Connell and Williams 2005; Zhang et al. 2009b). This strategy will likely be followed by NEA LCCs, especially in a more liberal NEA airline market, thereby suggesting the importance of studying efficiency performance of non-major airports.

Second, due to data limitations, we considered only three inputs and three outputs when estimating efficiency scores using DEA methods. In practice, different airports are very different in their operating characteristics and services provided; as a consequence, several other measures have been considered in the literature. Exclusion of certain inputs and outputs may yield biases in measuring airport productivity. For instance, if the so-called "soft cost input," which is measured by all expenses not directly related to capital and personnel, is not included, the efficiency measurement may favor the airports that outsourced most of their services. Similarly, non-aeronautical businesses, i.e., concessions, retailing, car rental, car parking, advertising, and other commercial services at an airport, are becoming more important for airports around the world, including major NEA airports (e.g., Zhang and Zhang 1997, 2003; Yu et al. 2008). The proportion of non-aeronautical revenue to the total revenue may vary significantly among airports. In general, the portion of non-aeronautical revenue for airports in China is still relatively low compared to airports in Japan and Korea. For example, in 2006 the non-aeronautical revenue contributed only 27.3 percent to the total revenue of Beijing Airport, and this figure rose to 47.3 percent of total revenue at Guangzhou Airport. Exclusion of non-aeronautical revenue may bias productivity against the airports that generate more revenues from non-aeronautical businesses. Consequently, further research may need to consider extending our analysis by using a larger set of inputs and outputs in measuring the airports' efficiency, although data collection will remain a formidable challenge.

Third, analysis of this article does not account for intrinsic differences among airports from different countries. It would be interesting, and important, to investigate factors (including country-specific factors) that might influence the efficiency performance of airports in NEA. Airports in different countries serve domestic networks and compete in international networks, and hence efficiency of an airport is expected to be significantly affected by the country's characteristics. For instance, how have country--specific factors affected airport productive efficiency? How have various aviation liberalization policies implemented by a country-e.g., airport localization and privatization programs as well as domestic and international airline deregulation measures-affected airport productivity? Our new data set might provide a basis for investigating these questions. To achieve this goal, one might run regressions to examine the effects of such factors as country-specific and policy liberalization factors on the efficiency scores, while controlling for airport characteristics and other shifting variables. While such regressions have been used extensively, being either OLS, the Tobit model (Tobin 1958), or other models (see, e.g., the airport studies mentioned in the introduction, Ali and Flinn 1989, and Kalirajan 1990 for applications in other industries), recent econometric work by Simar and Wilson (2007) suggests that strict application of regressions on DEA efficiency scores might not be apt. Instead, one may want to follow Simar and Wilson's procedure and conduct necessary econometric adjustments prior to the regression analysis.

Finally, previous work looking at airport efficiency and productivity has commonly used either linear programming approaches, notably DEA, or parametric methods, especially the stochastic production frontier model (see, e.g., Oum et al. 2009 for a review on methodology). Here, we use DEA (a non-parametric model) to estimate airport efficiency. There are a number of advantages of applying the DEA method in production efficiency measurement. Since DEA does not require financial or price data, DEA has relatively low data requirements compared to other approaches such as cost-function-oriented econometric approach or an index-number approach. Unlike the cost-function-oriented method, DEA can be applicable to multi-output production technology. It is also free from the a priori assumptions on functional forms. While there are a number of advantages to using DEA, it also has some disadvantages. Since the production possibility set is defined by using only a subset of observations, its estimation efficiency is rather low, and hence the efficiency score is extremely sensitive to outliers. Also, unlike the econometric approach, it generates multiple best performers. It would, therefore, be important to conduct, if relevant data were available, both DEA and a parametric approach and compare results of the two approaches. We see both this and the analyses indicated above as natural extensions of the analysis presented here.

REFERENCES

Abbott, M. and S. Wu (2002), "Total Factor Productivity and Efficiency of Australian Airports," Australian Economic Review, 35, pp. 244-260.

Ali, M. and J.C. Flinn (1989), "Profit Efficiency among Basmati Rice Producers in Pakistan Punjab," American Journal of Agricultural Economics, 71(2), pp. 303-310.

Banker, R.D., A. Charnes, and W.W. Cooper (1984) "Some Models for Estimating Technical and Scale Inefficiencies in Data Envelopment Analysis," Management Science, 30, pp. 1078-1092.

Banker, R.D., A. Charnes, W.W. Cooper, J. Swarts, and D.A. Thomas (1989) "An Introduction to Data Envelopment Analysis with some of Its Models and Their Uses," Research in Government and Nonprofit Accounting, edited by J.L. Chan, 5, pp. 125-164, Greenwich, CT: JAI Press.

Bazargan, M. and B. Vasigh (2003) "'Size versus Efficiency: A Case Study of U.S. Commercial Airports," Journal of Air Transport Management, 9, pp. 187-193.

Charnes, A., W.W. Cooper, B. Golany, L. Seiford, and J. Stutz (1985), "Foundation of Data Envelopment Analysis for Pareto-Koopmans Efficient Empirical Production Functions," Journal of Econometrics, 30, pp. 91-107.

Charnes, A., W.W. Cooper, and E.L. Rhodes (1978), "'Measuring the Efficiency of Decision-making Units," European Journal of Operation Research, 2, pp. 429-444.

Charnes, A., W.W. Cooper, and E.L. Rhodes (1981), "Evaluating Program and Managerial Efficiency: An Application of Data Envelopment Analysis to Program Follow Through," Management Science, 27, pp. 668-697.

Chin, A.T.H. and L.E. Siong (2001), "Airport Performance: A Comparative Study between Changi Airport and Airports in the New York-New Jersey Metropolitan Area," paper presented at the World Conference on Transport Research (WCTR), Seoul, Korea.

Farrell, M.J. (1957), "'The Measurement of Productive Efficiency," Journal of the Royal Statistics Society, Series A: General, 120, pp. 253-281.

Fernandes, E. and R.R. Pacheco (2002), "'Efficient Use of Airport Capacity," Transportation Research Part A, 36(3), pp. 225-238.

Fung, M., K. Wan, Y.V. Hui, and J. Law (2008), "Productivity Changes of Chinese Airports 1995-2004," Transportation Research Part E, 44(3), pp. 521-542.

Gillen, D. and A. Lall (1997), "'Developing Measures of Airport Productivity and Performance: An Application of Data Envelopment Analysis," Transportation Research Part E, 33(4), pp. 261-273.

Kalirajan, K.P. (1990), "On Measuring Economic Efficiency," Journal of Applied Econometrics, 5 (1), pp. 75-85.

Lam, S.W., J.M.W. Low, and T.C. Tang (2009), "Operational Efficiencies across Asia Pacific Airports," Transportation Research Part E, 45, pp. 654-665.

Lee, Y.H. (2008), "Open Sky and Current Issues in Korean Air Transport Policy," paper presented at the International Forum on Shipping, Ports, and Airports (IFSPA), held at The Hong Kong Polytechnic University, May 2008, Hong Kong.

Lovell, C.A.K. (1993), "Production Frontiers and Productive Efficiency," in The Measurement of Productive Efficiency: Techniques and Applications, edited by H.O. Fried, C.A.K. Lovell and S.S. Schmidt, 3-67, New York: Oxford University Press.

Martin, J.C. and C. Roman (2001) "An Application of DEA to Measure the Efficiency of Spanish Airports prior to Privatization," Journal of Air Transport Management, 7, pp. 149-157.

Martin, J.C. and C. Roman (2006), "A Benchmarking Analysis of Spanish Commercial Airports: A Comparison between SMOP and DEA Ranking Methods," Networks and Spatial Economics, 6, pp. 111-134.

O'Connell, J.F. and G. Williams (2005), "Passengers' Perceptions of Low Cost Airlines and Full Service Carriers: A Case Study Involving Ryanair, Aer Lingus, Air Asia and Malaysia Airlines," Journal of Air Transport Management, II, pp. 259-272.

Oum, T.H., H. Yamaguchi, and Y. Yoshida (2009), "Efficient Measurement Theory and Its Application to Airport Benchmarking," in Handbook of Transport Economics, edited by A. de Palma, R. Lindsey, E. Quinet, and R. Vickman, Edward Edgar, forthcoming.

Park, Y.H. (2000), "Reducing Legal and Institutional Barriers to Liberalize the Air Transportation Market: Korea's Perspective," paper presented at the EWC/KOTI Conference on Northeast Asia Transportation, August 2000, Hawaii.

Parker, D. (1999), "The Performance of BAA before and after Privatization," Journal of Transport Economics and Policy, 33, pp. 133-145.

Pels, E., P. Nijkamp, and P. Rietveld (2001), "Relative Efficiency of European Airports," Transport Policy, 8(3), pp. 183-192.

Pels, E., P. Nijkamp, and P. Rietveld (2003), "Inefficiencies and Scale Economies of European Airport Operations," Transportation Research Part E, 39, pp. 341-361.

Pestana Barros, C. and P.U.C. Dieke (2007), "Performance Evaluation of Italian Airports: Data Envelopment Analysis," Journal of Air Transport Management, 13, pp. 184-191.

Poole, R.W. (2003), "Global Airport Privatization Regains Altitude," report, Reason Public Policy Institute, Los Angeles.

Sarkis, J. (2000), "'An Analysis of the Operational Efficiency of Major Airports in the United States," Journal of Operations Management, 18, pp. 335-351.

Sarkis, J. and S. Talluri (2004), "Performance-based Clustering for Benchmarking of U.S. Airports," Transportation Research Part A, 38, pp. 329-346.

Seiford, L.M. and R.M. Thrall (1990), "Recent Developments in DEA: The Mathematical Programming Approach to Frontier Analysis," Journal of Econometrics, 46, pp. 7-38.

Seiford, L.M. (1996), "Data Envelopment Analysis: The Evolution of the State of the Art (1978-1995)," Journal of Productivity Analysis, 7, pp. 99-137.

Simar, L. and P.W. Wilson (2007), "Estimation and Inference in Two-Stage, Semi-parametric Models of Production Processes," Journal of Econometrics, 136, pp. 31-64.

Tobin, J. (1958), "Estimating the Relationship for Limited Dependent Variable," Econometrica, 26, pp. 24-36.

Takahashi, N. (2003), "Airline Deregulation in Japan: Its Economic Effects and Issues," Kansai University Review of Business and Commerce, 5, pp. 47-63.

Tretheway, M.W. (2004), "Distortions of Airline Revenues: Why the Network Airline Business Model is Broken," Journal of Air Transport Management, 10, pp. 3-14.

Vasigh, B. and R. G. Hamzaee (2000), "Airport Efficiency: An Empirical Analysis of the US Commercial Airports," paper presented at the 4th Air Transport Research Society (ATRS) Conference, Amsterdam.

Yamaguchi, K. (2005), "Inter-regional Air Transport Accessibility and Macro Economic Performance in Japan," paper presented at the 9th Air Transport Research Society (ATRS) Conference, Rio de Janeiro.

Yamauchi, H. (2000), "Air Transport Policy in Japan: Policy Change and Market Competition," paper presented at the EWC/KOTI Conference on Northeast Asia Transportation, August 2000, Hawaii.

Yoshida, Y. and H. Fujimoto (2004), "Japanese-airport Benchmarking with the DEA and Endogenous-weight TFP Methods: Testing the Criticism of Overinvestment in Japanese Regional Airports," Transportation Research Part E, 40, pp. 533-546.

Yu, M.-M., S.-H. Hsu, C.-C. Chang, and D.-H. Lee (2008), "Productivity Growth of Taiwan's Major Domestic Airports in the Presence of Aircraft Noise," Transportation Research Part E, 44, pp. 543-554.

Zhang, A. (1998), "'Industrial Reform and Air Transport Development in China," Journal of Air Transport Management, 4, pp. 155-164.

Zhang, A. and H. Chen (2003), "Evolution of China's Air Transport Development and Policy towards International Liberalization," Transportation Journal, 42, pp. 31-49.

Zhang, A. and A. Yuen (2008), "Airport Policy and Performance in Mainland China and Hong Kong," Aviation Infrastructure Performance: A Study in Comparative Political Economy, edited by Clifford Winston and Gines de Rus, pp. 159-192, Washington, D.C.: Brookings Institution Press.

Zhang, A. and Y. Zhang (1997), "Concession Revenue and Optimal Airport Pricing," Transportation Research Part E, 33, pp. 287-296.

Zhang, A. and Y. Zhang (2003), "Airport Charges and Capacity Expansion: Effects of Concessions and Privatization," Journal of Urban Economics, 53, pp. 5475.

Zhang, A., S. Hanaoka, H. Inamura, and T. Ishikura (2009a), "'Low Cost Carriers in Asia: Deregulation, Regional Liberalization and Secondary Airports," Research in Transport Economics, 24, pp. 36-50.

Zhang, A., Y. Zhang, and J.A. Clougherty (2009b), "Competition and Regulation in Air Transport," Handbook of Transport Economics, edited by A. de Palma, R. Lindsey, E. Quinet, and R. Vickman, Edward Edgar, forthcoming.

Zhang, Y. (2010), "Network Structure and Capacity Requirement," Transportation Researeh Part E, 46, pp. 189-197.

ENDNOTES

(1) In this article the word China refers to Mainland China, and the word Korea refers to South Korea.

(2) These include Shenzhen, China (opened in 1991), Osaka Kansai (1994), Macau (1995), Hong Kong (1998), Shanghai Pudong (1999), Incbeon (2001), Guangzbou (2004), and Nagoya, Japan (2005).

(3) Nevertheless, analysis by Yamaguchi (2005) suggested that there had been a significant productivity gain to the national economy from the improvement in air transport accessibility between 1995 and 2000, particularly for the agglomerated areas such as the Tokyo metropolitan region. See also Yoshida and Fujimoto (2004) for related discussions.

(4) ASEAN stands for "'Association of Southeast Asian Nations." China, Japan, and Korea have been participating in the annual ASEAN summit, "ASEAN+3," while holding tripartite summits on the sidelines as a means to promote economic integration in Northeast Asia.

(5) Useful general references on DEA include Banker et al. (1989), Seiford and Thrall (1990), Lovell (1993), and Seiford (1996).

(6) The MPI analysis here is based on a one-year change only. To obtain a "smoothed surface" frontier over a multiyear period, one may also introduce the DEA windows into the MPI analysis. For the present case, the results are similar with or without such windows.

(7) Because of a large number of airports involved, it is unlikely that one will have data for the number of employees, although the other input and output categories may still be available. Due to such data deficiency, it is difficult to assess airport "productivity." Rather, one may focus only on the capital (or capacity) efficiency of these airports.

Mr. Ha is associate professor, Asia-Pacific School of Logistics, Inha University, Incheon, South Korea. Mr. Yoshida is associate professor, National Graduate Institute for Policy Studies, Minato-ku, Tokyo, Japan; email yoshida@grips.ac.jp. Mr. Zhang is Vancouver International Airport Authority, Professor in Air Transportation, Sauder School of Business, University of British Columbia, Vancouver, British Columbia, Canada; email anming.zhang@sauder.ubc.ca.

The authors would like to thank two anonymous reviewers for their constructive comments, which have significantly improved the focus and readability of this article. We thank seminar participants at the Workshop on Northeast Asia Aviation Market, February 2009, Inha University, and at the 3rd International Forum on Shipping, Ports, and Airports, May 2009, Hong Kong Polytechnic University, where an early version of this article was presented. This work was supported by the Korea Research Foundation Grant funded by the Korean Government (MOEHRD) (KRF-2008-005-J01601).
Table 1. Characteristics of DEA Input and Output Variables

 Standard
 Definition Mean deviation
Input

Runway length Sum of runway 5,739 2,954
(meter) lengths

Terminal size Total area of 397,413 292,705
(square meter) passenger and cargo
 terminals

Employee Sum of full-time and 2,361 1,953
(persons) part-time employees

Output

Passenger volume Passenger throughput 24,967,344 10,516,235
(persons) of the airport

Cargo volume Cargo throughput of 1,412,000 936,698
(tons) the airport

Aircraft Number of flights at 159,782 82,442
movement the airport

Table 2. DEA Technical Efficiency (TE) of Airports, 1994-2007

 1994 1995 1996 1997 1998 1999 2000 2001

Beijing 0.47 0.62 0.68 0.72 0.85 0.46 0.52 0.61
Guangzhou 0.75 0.64 0.66 0.67 0.71 0.76 0.81 0.84
Shanghai 0.56 0.67 0.79 0.67 0.78 0.87 0.73 0.74
Hong Kong 0.87 0.91 0.96 1.00 0.67 0.75 0.84 0.81
Incheon 0.57 0.64 0.70 0.76 0.87 0.91 1.00 0.55
Kansai 0.23 0.77 0.87 0.92 0.90 0.90 0.93 0.89
Narita 0.79 0.82 0.80 0.85 0.83 0.91 0.96 0.85

 2002 2003 2004 2005 2006 2007

Beijing 0.66 0.63 0.81 0.92 1.00 1.00
Guangzhou 0.89 0.84 1.00 0.58 0.58 0.65
Shanghai 0.74 0.62 0.84 0.95 0.98 1.00
Hong Kong 0.87 0.88 0.96 1.00 0.98 1.00
Incheon 0.57 0.58 0.59 0.59 0.62 0.67
Kansai 0.83 0.69 0.75 0.83 0.85 0.92
Narita 0.99 0.77 0.86 0.87 0.93 1.00

Table 3. DEA Pure Technical Efficiency (PTE) of Airports, 1994-2007

 1994 1995 1996 1997 1998 1999 2000 2001

Beijing 0.69 0.71 0.75 0.77 0.83 0.48 0.50 0.58
Guangzhou 1.00 0.96 0.96 0.96 0.97 0.97 0.97 0.98
Shanghai 1.00 1.00 1.00 0.96 0.96 0.97 0.95 0.96
Hong Kong 1.00 1.00 1.00 1.00 0.71 0.77 0.82 0.80
Incheon 0.56 0.62 0.66 0.70 0.88 0.97 1.00 0.55
Kansai 1.00 1.00 1.00 1.00 1.00 0.99 0.99 0.99
Narita 0.85 0.85 0.85 0.87 0.85 0.90 0.93 0.86

 2002 2003 2004 2005 2006 2007

Beijing 0.63 0.61 0.80 0.90 1.00 1.00
Guangzhou 0.98 0.98 1.00 0.55 0.60 0.67
Shanghai 0.96 0.95 0.98 0.99 0.99 1.00
Hong Kong 0.86 0.91 0.97 1.00 0.96 1.00
Incheon 0.59 0.60 0.59 0.59 0.64 0.73
Kansai 0.99 1.00 1.00 1.00 1.00 1.00
Narita 0.95 0.83 0.89 0.89 0.96 1.00

Table 4. DEA Scale Efficiency (SE) of Airports, 1994-2007

 1994 1995 1996 1997 1998 1999 2000 2001

Beijing 0.68 0.87 0.91 0.93 0.99 0.90 0.99 1.00
Guangzhou 0.75 0.66 0.68 0.64 0.68 0.72 0.75 0.77
Shanghai 0.56 0.67 0.79 0.65 0.74 0.77 0.60 0.67
Hong Kong 0.87 0.91 0.96 1.00 0.95 0.93 0.97 0.97
Incheon 0.93 0.95 0.99 0.99 0.94 0.95 1.00 0.99
Kansai 0.23 0.76 0.86 0.90 0.87 0.87 0.89 0.86
Narita 0.88 0.91 0.89 0.93 0.90 0.96 0.97 0.92

 2002 2003 2004 2005 2006 2007

Beijing 1.00 1.00 0.99 0.99 1.00 0.98
Guangzhou 0.82 0.80 1.00 0.95 0.96 0.98
Shanghai 0.67 0.62 0.84 0.93 0.96 1.00
Hong Kong 0.95 0.92 0.98 1.00 1.00 1.00
Incheon 0.97 0.96 0.98 0.98 0.97 0.92
Kansai 0.80 0.67 0.73 0.81 0.84 0.92
Narita 0.99 0.84 0.91 0.93 0.96 1.00

Table 5. Average Efficiency Scores of DEA vs. DEA-Window, 1994-2007

 1994 1995 1996 1997 1998 1999 2000

TE: DEA 0.59 0.71 0.76 0.77 0.76 0.75 0.77
TE: DEA-Window 0.60 0.72 0.78 0.80 0.80 0.80 0.83
PTE: DEA 0.87 0.88 0.89 0.89 0.89 0.86 0.88
PTE: DEA-Window 0.88 0.89 0.92 0.95 0.91 0.88 0.91
SE: DEA 0.70 0.82 0.87 0.86 0.87 0.87 0.88
SE: DEA-Window 0.71 0.82 0.85 0.85 0.89 0.91 0.91

 2001 2002 2003 2004 2005 2006 2007

TE: DEA 0.71 0.74 0.68 0.81 0.79 0.84 0.89
TE: DEA-Window 0.76 0.80 0.72 0.83 0.82 0.85 0.89
PTE: DEA 0.82 0.85 0.84 0.89 0.85 0.88 0.91
PTE: DEA-Window 0.87 0.93 0.89 0.92 0.88 0.90 0.92
SE: DEA 0.88 0.89 0.83 0.92 0.94 0.96 0.97
SE: DEA-Window 0.87 0.86 0.81 0.90 0.93 0.94 0.96

Notes: TE = technical efficiency; PTE = pure technical efficiency;
SE = scale efficiency

Table 6. MPI Total Factor Productivity Changes (TFPC) by Airports

 95 96 97 98 99 00 01 02

Beijing 1.29 1.10 1.06 1.15 0.51 1.13 1.15 0.94
Guangzhou 0.98 1.04 0.95 1.05 1.06 1.16 1.01 1.12
Hong Kong 1.15 1.06 1.10 0.22 1.24 1.15 1.20 1.11
Incheon 1.13 1.09 1.08 1.58 1.13 1.12 0.36 1.05
Kansai 5.69 1.41 1.23 0.92 0.97 1.01 0.95 0.91
Narita 1.17 0.83 1.30 1.04 1.03 1.03 0.87 1.16
Shanghai 1.19 1.15 0.92 1.15 1.16 0.81 1.00 0.94
Average 1.80 1.10 1.09 1.02 1.01 1.06 0.93 1.03

 03 04 05 06 07 Average

Beijing 0.83 1.41 1.25 2.08 0.98 1.15
Guangzhou 1.06 1.69 0.53 1.08 1.08 1.06
Hong Kong 0.93 1.28 1.15 0.92 1.06 1.04
Incheon 1.07 1.04 1.00 1.02 1.09 1.06
Kansai 0.79 1.07 1.10 1.08 1.09 1.40
Narita 0.76 1.13 1.12 1.31 1.21 1.07
Shanghai 0.84 1.37 1.27 1.01 1.06 1.07
Average 0.90 1.28 1.06 1.21 1.08 1.12

Table 7. MPI Technical Efficiency Changes (TEC) by Airports

 95 96 97 98 99 00 01 02

Beijing 1.23 1.04 1.02 2.63 0.50 1.03 1.54 0.85
Guangzhou 0.93 0.98 0.91 2.00 1.00 1.34 0.92 1.06
Hong Kong 0.99 0.96 1.01 0.28 1.00 1.01 2.00 1.02
Incheon 1.08 1.03 1.03 2.72 1.02 1.01 0.43 0.99
Kansai 5.59 1.01 1.00 1.09 0.92 0.83 1.26 0.94
Narita 0.88 0.63 0.94 2.03 0.99 0.97 1.00 1.03
Shanghai 1.14 1.09 0.88 2.42 1.07 0.86 1.02 0.85
Average 1.69 0.96 0.97 1.88 0.93 1.01 1.17 0.96

 03 04 05 06 07 Average

Beijing 0.89 0.82 1.84 1.73 0.95 1.24
Guangzhou 1.26 1.13 0.60 0.53 1.11 1.06
Hong Kong 0.97 1.07 1.11 0.91 1.00 1.03
Incheon 1.19 0.80 0.90 1.04 1.05 1.10
Kansai 0.99 0.69 1.03 1.11 0.81 1.33
Narita 0.88 0.78 0.95 1.32 1.02 1.03
Shanghai 0.89 0.89 1.91 0.97 1.01 1.15
Average 1.01 0.88 1.19 1.09 0.99 1.13

Table 8. MPI Technological Changes (TC) by Airports

 95 96 97 98 99 00 01 02

Beijing 1.05 1.06 1.04 0.44 1.03 1.10 0.74 1.10
Guangzhou 1.05 1.06 1.04 0.52 1.06 0.87 1.09 1.05
Hong Kong 1.16 1.09 1.09 0.76 1.24 1.14 0.60 1.08
Incheon 1.05 1.06 1.04 0.58 1.11 1.11 0.84 1.06
Kansai 1.02 1.39 1.23 0.84 1.05 1.22 0.75 0.97
Narita 1.34 1.32 1.39 0.51 1.04 1.07 0.87 1.13
Shanghai 1.05 1.06 1.04 0.48 1.08 0.94 0.98 1.11
Average 1.10 1.15 1.12 0.59 1.09 1.06 0.84 1.07

 03 04 05 06 07 Average

Beijing 0.94 1.72 0.68 1.20 1.03 1.01
Guangzhou 0.84 0.50 0.88 2.03 0.98 1.07
Hong Kong 0.96 1.20 1.04 1.01 1.05 1.03
Incheon 0.90 1.29 1.11 0.99 1.04 1.01
Kansai 0.80 1.54 1.07 0.96 1.35 1.09
Narita 0.87 1.46 1.18 0.99 1.18 1.10
Shanghai 0.95 1.54 0.67 1.04 1.05 1.00
Average 0.89 1.46 0.95 1.17 1.10 1.05
COPYRIGHT 2010 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 2010 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Author:Ha, Hun-Koo; Yoshida, Yuichiro; Zhang, Anming
Publication:Transportation Journal
Geographic Code:9SOUT
Date:Sep 22, 2010
Words:10709
Previous Article:Managing Risk and Security: The Safeguard of Long-Term Success for Logistics Service Providers.
Next Article:Air transport liberalization and its impacts on airline competition and air passenger traffic.
Topics:

Terms of use | Privacy policy | Copyright © 2020 Farlex, Inc. | Feedback | For webmasters