6.1 Conclusion
In order to investigate the effects of various degrees of accidents on the technical efficiency of a bus transit, the current research incorporates both desirable and undesirable outputs into a stochastic frontier model. An aggregate score of accidents is chosen as the undesirable output and a log-linear stochastic output distance function is specified to estimate the technical efficiency for Taipei bus transit systems. The aggregate score of accident converts fatality, major injury, minor injury, and property loss only into proper weighted score, and thus can distinguish the severity of accidents. For comparison, the research also estimates the technical efficiency without consideration of accidents by specifying a standard production function frontier.
The findings indicate that the inefficiency term in the stochastic output distance function model is significant taking Taipei bus transit industry as a whole, which means that the industry needs to curtail inputs and expand outputs so as to improve their productive efficiency. In addition, the results also reveal that the productive efficiency with adjustment of either accident rate or weighted accident severity is somehow different from that measured without adjustment of accident effects.
Furthermore, the empirical results also show that both desirable output (vehicle-kms) and undesirable output (aggregate accident score) have affected the technical efficiency of Taipei bus transit. The elasticity of vehicle-kms (=1.0055) is greater than that of accident rate (=-0.0055).
The managerial implication is that the bus firms can improve technical efficiency by increasing their desirable outputs and/or reducing the inputs and accidents, thus a bus company can promote its efficiency via safer operation. This thesis contributes to identify the effects of undesirable outputs in bus transit efficiency measurement, through which one can propose more practical strategies for improving the bus operating efficiency.
The case study shows that the elasticity of stochastic output distance function associated with fuel consumption (0.7088) is much greater than that associated with the other two inputs (fleet size=0.1751 and number of employees=0.0555), implying that energy consumption can be a dominant factor affecting the efficiency of Taipei bus transit. Thus, one promising strategy for improving the efficiency is to provide more bus exclusive lanes with preemption signals.
Another strategy is perhaps to train the drivers to operate the vehicles more smoothly. As such, the energy consumption can be saved. Introduction of innovative fuel economy technologies, such as diesel- or gas-electricity hybrid, is of course also promising for the enhancement of bus transit technical efficiency.
Moreover, the case study of the current research shows that the privatized MBC (2004 to 2006) has higher efficiency than the public operator TMB (2001 to 2003), indicating that privatization has indeed improved the technical efficiency. In addition, the empirical results indicate that the scale economy for Taipei bus transit exhibits decreasing returns to scale, suggesting downsizing its scale should be able to enhance the technical efficiency. It is also found that the ratio of aged vehicles can deteriorate the technical efficiency, while the length of bus exclusive lane has positive contribution on productive efficiency. These findings suggest that new vehicles should be substituted for the old ones and that more bus exclusive lanes should be introduced so as to promote the technical efficiency of bus transit systems.
Comparisons between SFA and other methods are provided. It is found that the results are generally in common with Taipei bus transit appraisal, indicating that the results in this thesis do not violate the real situation. In addition, from the comparison between DEA and SFA methods, one can see that SFA method is better-behaved than DEA in Taipei bus transit industry. This result indicates that random error (such as traffic jam, caused by accidents or malfunction of traffic lights) has significant influence on the efficiency of bus transit, thus SFA may be more suitable than DEA when measuring bus efficiency. Also, the result support that aggregated
accident score is more appropriate to be the undesirable output than accident rate, as we expected.
6.2 Recommendations
The remaining issue to be resolved in the future study is of four aspects. First, this thesis uses accident rate and aggregate score of accidents as the undesirable outputs for bus transit. In fact, there are other undesirable outputs, such as complaints from passengers, noise and pollutants, which may also influence the technical efficiency of bus operation. It deserves further exploration in the future study.
Second, the case study shows that translog functional form is not suitable for measuring the efficiency of Taipei bus transit, thus a log-linear stochastic output distance function is specified in the current research. However, a log-linear functional form may not be general enough for the production surface. Other functional form may be needed to further investigate, in order to have a more approximate frontier for bus transit.
Third, this thesis concludes that the ratio of aged vehicles and the length of exclusive lanes are two external factors which influence the efficiency. In fact, this may not be enough to explain the determinants of efficiency. In other words, the technical efficiency may be influenced by some other external factors, thus it deserves to further investigate the efficiency by taking into account more determinants.
Finally, this research measures the efficiency by considering only one kind of vehicle, in fact, apart from the city bus; there are still mini buses in Taipei bus transit systems. Thus, the one of the potential avenue for future study is that it deserves further investigation by dividing the vehicle into two categories: city bus and mini bus when measuring the efficiency of Taipei bus transit industry if the data is available.
References
1. Aigner, D. J., C. A. K. Lovell and P. Schmidt (1977), “Formulation and estimation of stochastic frontier production function models,” Journal of Econometrics, 6, 21-37.
2. Battese, G. E. and T. J. Coelli (1988), “Prediction of firm-level technical efficiencies with a generalized frontier production function and panel data,”
Journal of Econometrics, 38, 387-399.
3. Battese, G. E. and T. J. Coelli (1995), “A model for technical inefficiency effects in a stochastic frontier production function for panel data,” Empirical Economics, 20, 325-332.
4. Bhattacharyya, A., S. C. Kumbhakar and A. Bhattacharyya (1995),
“Ownership structure and cost efficiency: a study of publicly owned passenger-bus transportation companies in India,” Journal of Productivity Analysis, 6, 47-61.
5. Caves, D. W., L. R. Christensen and W. E. Diewert (1982), ”Multilateral comparisons of output, input, and productivity using superlative index numbers,” Economic Journal, 92, 79-86.
6. Chung, Y. H., R. Färe and S. Grosskopf (1997), “Productivity and undesirable outputs: a directional distance function approach,” Journal of Environmental Management, 51, 229-240.
7. Coelli, T. (1995), “Recent developments in frontier modeling and efficiency measurement,” Australian Journal of Agricultural Economics, 39, 219-245.
8. Coelli, T. (1996), “A Guide to FRONTIER Version 4.1: A Computer Program for Frontier Production Function Estimation,” CEPA Working Paper 96/07, Department of Econometrics, University of New England, Armidale.
9. Coelli T., L. Lauwers and G. V. Huylenbroec (2005), “Formulation of technical,
economic and environmental efficiency measures that are consistent with the materials balance condition,” CEPA Working paper series, No. 06/2005, University of Queensland, Australia.
10. Coelli, T., D. S. P. Rao, C. J. O’Donnell, and G. E. Battese (2005), An Introduction to Efficiency and Productivity Analysis, second edition, Kluwer Academic Publishers, Boston.
11. Dalen, D. M. and A. Gómez-Lobo (2003), “Yardsticks on the road: regulatory contracts and cost efficiency in the Norwegian bus industry,” Transportation, 30, 371-386.
12. Färe, R., S. Grosskopf, C. A. K. Lovell and C. Pasurka (1989), “Multilateral productivity comparisons when some outputs are undesirable: a nonparametric approach,” Review of Economics and Statistics, 71, 90-98.
13. Färe, R., S. Grosskopf, C. A. K. Lovell and S. Yaisawarng (1993), “Derivation of shadow prices of undesirable outputs: a distance function approach.”
Review of Economics and Statistics, 75, 374-380.
14. Farrell, M. J. (1957), “The measurement of productivity efficiency.” Journal of the Royal Statistical Society, Series A, 120, part 3, 253-290.
15. Fernández, C., G. Koop and M. F. J. Steel (2002), “Multiple-output production with undesirable outputs: an application to nitrogen surplus in agriculture,”
Journal of the American Statistical Association, 97, 432-442.
16. Filippini, M., R. Maggi, and P. Prioni (1992), “Inefficiency in a regulated industry: the case of Swiss regional bus companies,” Annuals of Public and Cooperative Economics, 63, 437-455.
17. Gathon, H. J. (1989), “Indicators of partial productivity and technical efficiency in the European urban transit sector,” Annuals of Public and Cooperative Economics, 60, 43-59.
18. Jørgensen, F., P. A. Pedersen and R. Volden (1997), “Estimating the inefficiency in the Norwegian bus industry from stochastic cost frontier models,” Transportation, 24, 421-433.
19. Kumbhakar, S. C. and C. A. K. Lovell (2000), Stochastic Frontier Analysis, Cambridge University Press.
20. Lan, L. W. and E. T. J. Lin (2003), “Measurement of railways productive efficiency with data envelopment analysis and stochastic frontier analysis,”
Journal of the Chinese Institute of Transportation, 15, 49-78 (in Chinese).
21. Lan, L. W. and E. T. J. Lin (2006), “Performance measurement for railway transport: stochastic distance functions with inefficiency and ineffectiveness effects,” Journal of Transport Economics and Policy, 40, 383-408.
22. Lin, E. T. J. and L. W. Lan (2006), “Measuring technical efficiency with consideration of undesirable outputs: the case of Taipei bus transits,” Asia and Pacific Productivity Conference, Seoul, Korea, August 17-19.
23. Lin, E. T. J., L. W. Lan and A. K. Y. Chiu (2007), “Evaluation of Productive Efficiency with Adjustment of Accidents: Case of Taipei Bus Transit,” paper submitted to 12th HKSTS International Conference.
24. Lovell, C. A. K., S. Richardson, P. Travers, and L. L. Wood (1994),
“Resources and functionings: a new view of inequality in Australia,” in W.
Eichhorn (ed.), Models and Measurement of Welfare and Inequality, Berlin, Springer-Verlag.
25. Meeusen, W. and J. van den Broeck (1977), “Efficiency estimation from Cobb-Douglas production functions with composed error,” International Economic Review, 18, 435-444.
26. Pittman, R. W. (1983), “Multilateral productivity comparisons with undesirable outputs,” The Economic Journal, 93, 883-891.
27. Reinhard, S., C. A. K. Lovell and G. J. Thijssen (2000), “Environmental efficiency with multiple environmentally detrimental variables estimated with SFA and DEA,” European Journal of Operational Research, 121, 287-303.
28. Sakano, R. and K. Obeng (1995), “Re-examination of inefficiencies in urban transit systems: a stochastic frontier approach,” Logistics and Transportation Review, 31, 377-392.
29. Sakano, R., K. Obeng and G. Azam (1997), “Subsidies and inefficiency:
stochastic frontier approach,” Contemporary Economic Policy, 15, 113-127.
30. Shephard, R. W. (1970), Theory of Cost and Production Functions, Princeton University Press, Princeton.
31. Thiry, B. and H. Tulkens (1992), “Allowing for inefficiency in parametric estimation of production functions for urban transit forms,” Journal of Productivity Analysis, 3, 45-65.
32. Tyteca, D. (1996), “On the measurement of the environmental performance of firms—a literature review and a productive efficiency perspective,” Journal of Environmental Management, 46, 281-308..