Skip to main content
Log in

The evolution of U.S. rail freight pricing in the post-deregulation era: revenues versus marginal costs for five commodity types

  • Published:
Transportation Aims and scope Submit manuscript

Abstract

There have now been over three decades of experience with rate-making freedom for all modes of intercity freight transport in the United States. Most evidence suggests that regulatory change has been beneficial for the rail industry and its users. Despite evidence of positive impacts of regulatory reform of U.S. freight transport, there is limited evidence related to long-term pricing trends by commodity in the deregulated era. Moreover, U.S. shipper groups have called for increased regulation of U.S. railroads, citing increased rates and profits, and monopoly pricing to “captive shippers.” This study estimates U.S. railroad revenue-marginal cost ratios for seven different commodities between 1986 and 2008. Interestingly, we find no significant increase in revenue-cost margins for commodities thought to be “most captive” (coal and chemicals), while finding large increases for some commodities thought to be “non-captive.” These results may provide insight into the impacts of regulatory reform in other countries, where there are similar concerns of equitable pricing and financial viability. They suggest that a move toward a more market-based pricing system can enhance railroad viability without harming those with fewer transport options.

This is a preview of subscription content, log in via an institution to check access.

Access this article

Price excludes VAT (USA)
Tax calculation will be finalised during checkout.

Instant access to the full article PDF.

Institutional subscriptions

Fig. 1
Fig. 2
Fig. 3
Fig. 4
Fig. 5
Fig. 6
Fig. 7
Fig. 8

Similar content being viewed by others

Notes

  1. Interestingly, the types of regulatory constraints that have limited Australian railroads recently are similar to those that limited U.S. railroads prior to regulatory reform in the late 1970s. Docwra (1991) notes that despite deregulation of interstate road transport in 1954 in Australia, there were still significant regulations limiting intermodal competition within states in the early 1990s. Moreover, railroads were required to provide services that were uneconomic based on community service obligations (Docwra 1991).

  2. It should be noted that the impacts of U.S. regulatory reform may not be directly applicable to other countries, where there is a different mix of traffic (e.g. European rail transport is heavily focused on passengers, while U.S. rail transport is heavily focused on freight), different shipment lengths and characteristics, differences in industry structure (e.g. private vs. govt. ownership), and differences in the nature of intermodal competition. Nonetheless, there are enough similarities between U.S. railroads and railroads elsewhere to provide insight into potential impacts of regulatory reform (e.g. Australian railroads also carry a substantial amount of bulk commodities over long distances).

  3. There have been numerous contributions relating rail freight rates to deregulation. Some include Winston et al. (1990), Friedlaender (1992), and Winston et al. (2004). These studies differ from the present study in that they cover only one or two commodities (often coal), they do not disaggregate by commodity, or they are very early after the Staggers Act, so more recent evidence is useful. Ivaldi and McCullough (2007) estimate costs are based on only two estimated cost coefficients whereas our study is based on separate cost estimates for five different commodity groups. We also argue that inclusion of results for the years 2005–2008 is important for our conclusions about trends in railroad pricing. Laurits Christensen Associates (2010) make an assessment of revenues and marginal costs for individual commodities, but do not use a multiple output cost function. As we show in this study, there are large differences in estimated marginal costs for different commodities using a multiple-output cost function.

  4. We use loaded car-miles for output, as no ton-mile data by output category are available from public sources. As noted by Hensher et al. (1995), total car-miles might be considered a “supply side” output (i.e. service units available to shippers), while ton-miles would be considered a demand side measure (i.e. service units consumed by shippers). Because we use loaded car miles, our output is still a demand side measure (in line with traditional output measures used in cost studies).

  5. While it may seem that more disaggregate output categories would be desirable, available data limit the amount of disaggregation that can be done appropriately. Most commodities are carried in large volumes using more than one car type. Thus, further disaggregation of car types would count individual commodities as being separate outputs (e.g. coal carried in plain gondolas would be a separate output from coal carried in special service hoppers).

  6. Carloads are not available for tank cars. Therefore, the estimated average length of haul for the farm product/chemical/food product category is loaded car-miles in covered hoppers and refrigerator cars divided by covered hopper and refrigerator carloads.

  7. Commodities in a particular output category are assumed to have the same average lengths of haul. For example, food products, chemicals, and farm products all have the same assumed average lengths of haul.

  8. We have also included GTW in 1992 and 1995, as it had real revenues in excess of $401.4 million in every other year. Railroads in our sample account for more than 99 % of Class I revenue ton-miles in every year. Data definitions and railroads included are in an online appendix (Tables A2 and A1, respectively).

  9. Another issue of interest to some researchers is the question as to whether factor price variables should be treated as exogenous. Practically all studies of rail costs in the USA of which we are aware have noted that the fraction of total factor supply (capital, labor, fuel, and materials) accounted for by individual railroads is small. As a result, practically all studies of rail costs have treated factor prices as exogenous, and we have followed previous studies in that respect.

  10. Note that this problem can be closely related to the more familiar issue of simultaneous determination of price and output. If demand were lower than anticipated, as described above, then price would be too high to clear the market; similarly, if demand is higher than anticipated, price would be too low to clear the market. It can thus be noted that correcting for the bias pointed out by Borts should contribute to the correction of this more familiar problem. However, we do not statistically estimate price-output relationships (demand) in this study.

  11. To the extent that our procedure has failed to correct for the problem pointed out by Borts, our estimates of the cost coefficients (and hence the marginal costs) will be biased downwards. The implications of that will be discussed later.

  12. Our cost function is increasing in factor prices, and continuous in factor prices by assumption. It is concave in factor prices at the means of all variables. However, in testing for concavity in factor prices for individual observations, the condition is only met for 18.5 % of observations (41 out of 222). As noted in Pels and Rietveld (2008), failure to find global concavity is common in empirical studies. To test for concavity in factor prices, the characteristic roots of the Hessian matrix are taken for every observation in the sample. If all characteristic roots are non-positive, this suggests the Hessian matrix is negative semidefinite, and therefore, the cost function is concave in factor prices.

  13. There are a variety of reasons why these elasticities may vary, including, but not limited to differences in equipment maintenance and investment costs among car types, differences in operating characteristics of different types of traffic (e.g. lengths of haul, percent empty, percent loaded on-line, speed of operation), and differences in the routes traveled by different types of traffic, with some traveling on high traffic routes where more density economies are exhausted and others traveling on lower density routes.

  14. Estimated revenues per car-mile, marginal costs per car-mile, revenue-to-marginal cost ratios, and revenue-to-stand-alone cost ratios, along with their standard errors are presented in Table A3 of the appendix.

  15. There are several ways one might assess the overall marginal costs of handling various commodities. In addition to the method used here (simulate MC at the overall means of all variables), a weighted average of simulated individual railroad marginal costs might be used. However, whichever method is used, the trends in marginal costs are similar. Figures 9, 10, 11, 12 in the online Appendix show a comparison of MCs estimated with overall sample means to those estimated via a weighted average of individual railroad MCs.

  16. Costs could be higher on high-density routes either because of increasing long-run costs, which stem from long-run diseconomies of high-density operation, or from increasing short-run costs, stemming from inadequate investment in plant capacity on high-density routes. The latter is possible in the case of mountain passes and tunnels, where expansion of capacity is expensive, or in situations of underinvestment, which some claim has occurred on some transcontinental lines in the West. Our results of increasing returns to traffic density, over the range of densities in our sample, suggest that neither of the above-mentioned cases is likely to be typical or representative, in terms of our present analysis and results. Further, a 95 % confidence interval on our output measure shows .3962 ≤ βOutput ≤ .7846, suggesting strong economies of density at the means of all variables.

References

  • Association of American Railroads: Class I Railroad Statistics (2010)

  • Australian Government, Department of Transport and Regional Services, Bureau of Transport and Regional Economics: rail infrastructure pricing: principles and practice (2003)

  • Baumol, W.J., Panzar, J.C., Willig, R.D.: Contestable markets and the theory of industry structure. Harcourt Brace Jovanovich, New York (1982)

    Google Scholar 

  • Bereskin, C.G.: Econometric estimation of the effects of deregulation on railway productivity growth. Transp. J. 35(Summer), 34–43 (1996)

    Google Scholar 

  • Bereskin, C.G.: Sequential estimation of railroad costs for specific traffic. Transp. J. 40(Spring), 33–45 (2001)

    Google Scholar 

  • Berndt, E.R., Friedlaender, A.F., Chiang, S.W., Vellturo, C.: Cost effects of mergers and deregulation in the U.S. railroad industry. J. Prod. Anal. 4(1–2), 127–174 (1993)

    Google Scholar 

  • Bitzan, J.D., Keeler, T.E.: Economies of density and regulatory change in the U. S. railroad freight industry. J. Law Econ. 50(1), 156–179 (2007)

    Article  Google Scholar 

  • Bitzan, J.D., Keeler, T.E.: Intermodal traffic, regulatory change and carbon energy conservation in U.S. freight transport. Appl. Econ. 43(25–27), 3945–3963 (2011)

    Article  Google Scholar 

  • Bitzan, J.D., Keeler, T.E.: Productivity growth and some of its determinants in the deregulated U.S. railroad industry. South. Econ. J. 70(2), 232–253 (2003)

    Article  Google Scholar 

  • Borts, G.H.: The estimation of rail cost functions. Econometrica 28(1), 108–131 (1960)

    Article  Google Scholar 

  • Caves, D., Christensen, L.R., Tretheway, M.W., Windle, R.: Analytical studies in transport economics. In: Daughety, F. (ed.) Network effects and the measurement of returns to scale and density for US railroads. Cambridge University press, Cambridge (1985)

    Google Scholar 

  • Deaux, J.: Regulations could derail railroad profits. The Street, August 21 (2011)

  • Docwra, G.: Transport regulation, deregulation and regulatory reform in Australia. In: Button, K., Pitfield, D. (eds.) Transport deregulation. Palgrave Macmillan, New York (1991)

    Google Scholar 

  • Ellig, J.: Railroad deregulation and consumer welfare. J. Reg. Econ. 21(2), 143–167 (2002)

    Google Scholar 

  • Faulhaber, G.R.: Cross-subsidization: pricing in public enterprises. Am. Econ. Rev. 65(5), 966–977 (1975)

    Google Scholar 

  • Friedlaender, A.F.: Coal rates and revenue adequacy in a quasi-regulated rail industry. Rand. J. Econ. 23(3), 376–394 (1992)

    Article  Google Scholar 

  • Friedlaender, A.F., Spady, R.: Freight transport regulation. MIT Press, Cambridge (1981)

    Google Scholar 

  • Gaskins, D.W.: Regulation of freight railroads in the modern era: 1970–2010. Rev. Net. Econ. 7(4), 561–571 (2008)

    Google Scholar 

  • Hensher, D.A., Daniels, R., Demellow, I.: A comparative assessment of the productivity of Australia’s public rail systems 1971/72–1991/92. J. Prod. Anal. 6(3), 201–223 (1995)

    Article  Google Scholar 

  • Interstate Commerce Commission: Coal Rate Guidelines: Nationwide, Ex Parte No. 347 (sub-no. 1) (1985)

  • Ivaldi, M., McCullough, G.: Railroad pricing and revenue-to-cost margins in the post-staggers era. Res. Tran. Econ. 20, 153–178 (2007)

    Article  Google Scholar 

  • Keeler, T.E.: Railroads, freight, and public policy. The Brookings Institution, Washington, DC (1983)

    Google Scholar 

  • Keeler, T.: The economics of passenger trains. J. Bus. 44(2), 148–174 (1971)

    Article  Google Scholar 

  • Laurits Christensen Associates: An update to the study of competition in the U.S. freight railroad industry: final report, prepared for the U.S. Surface Transportation Board, January (2010)

  • Meyer, J., Peck, M., Stenason, J., Zwick, C.: The economics of competition in the transportation industries. Harvard University Press, Cambridge (1959)

    Google Scholar 

  • Nash, C., Rivera-Trujillo, C.: Rail reform in Europe: issues and research needs. In: Rietveld, P., Stough, R. (eds.) Institutions and sustainable transport: regulatory reform in advanced economies. Edward Elgar, Cheltenham (2007)

    Google Scholar 

  • Pels, E., Rietveld, P.: Cost functions in transport. In: Hensher, D., Button, K. (eds.) Handbook of transport modelling, 2nd edn. Bingley, Emerald (2008)

    Google Scholar 

  • Pels, E., Rietveld, P.: Rail cost functions and scale elasticities: a meta-analysis. Free University Department of Spatial Economics, Amsterdam (2003)

    Google Scholar 

  • Pittman, R.: The economics of railroad ‘captive shipper’ legislation. economic analysis group discussion paper 10-1, Antitrust Division, U.S. Department of Justice, January (2010)

  • Surface Transportation Board, U.S. Department of Transportation: Class I Annual Reports (R1), (1986–2008)

  • Wilson, W.W.: Cost savings and productivity in the railroad industry. J. Reg. Econ. 11(1), 21-40 (1997)

    Google Scholar 

  • Winston, C., Corsi, T.M., Grimm, C.M., Evans, C.A.: The economic effects of surface freight deregulation. The Brookings Institution, Washington, DC (1990)

    Google Scholar 

  • Winston, C., Scott, M.D., and Vikram M.: Duopoly in the railroad industry: bertrand, cournot, or collusive? Brookings institution working paper (2004)

Download references

Author information

Authors and Affiliations

Authors

Corresponding author

Correspondence to John D. Bitzan.

Electronic supplementary material

Below is the link to the electronic supplementary material.

Supplementary material 1 (DOCX 246 kb)

Rights and permissions

Reprints and permissions

About this article

Cite this article

Bitzan, J.D., Keeler, T.E. The evolution of U.S. rail freight pricing in the post-deregulation era: revenues versus marginal costs for five commodity types. Transportation 41, 305–324 (2014). https://doi.org/10.1007/s11116-013-9463-8

Download citation

  • Published:

  • Issue Date:

  • DOI: https://doi.org/10.1007/s11116-013-9463-8

Keywords

Navigation