Click here for search results

Sources of Data


Several sources have been used to develop and update the database: 
Annual Reports
When available, the Annual Report or specific government statistics for the country and the railway are used even when other data exist.  For example, a number of railways, including Indian Railways, Bangladesh Railways and Turkish Railways, publish detailed Annual Reports. These are publications of official record and, even though they may contain some inaccuracies or approximations in derivation, are usually consistent from year to year.  In any event they represent the best estimates available.
In a number of cases, such as the three major Japanese railways (
JR EastJR West and JR Central), the Annual Reports are published in English on the company website. 
International Union of Railways
A major source is the International Union of Railways (UIC) in Paris.  The UIC publishes “
International Railway Statistics” annually that contains useful data about UIC members and associate members.  “International Railway Statistics” tends to be complete for European and several major international railways but is often incomplete for smaller railways.  In addition, UIC publishes a “flash” report for the most recent year that includes a limited set of more recent data for a number of countries.  The flash report is often based on preliminary data, and the numbers are often revised, sometimes significantly, when the “International Railway Statistics” data for the same year are eventually reported.  The latest issue of the flash report is for the year 2005, and much of the 2005 data provided are derived from the 2005 flash report.  Finally, the UIC publishes “Railway Time-Series Data” that contains a limited number of measures over a number of years.  The latest issue of this report covers the period 1970-2000, and is the primary source for data on railways (e.g. Russia) that did not exist before 1990, but that have chosen to recreate earlier data.  The UIC is currently regionalizing its organization which is likely to result in a much more comprehensive international membership and hopefully more comprehensive statistical returns from railways in many developing countries.
World Bank Data
World Bank analyses will often contain railway sector or transport sector data that have been developed directly from official sources.  In addition, World Bank staff often collect data from specific railways.

Regulatory Authorities Websites
The PCD was originally developed with the use of a PPIAF grant, and much of the data prior to 2002 were developed by consultants financed by that grant.  Data entered after that point is mostly based on regulator websites or direct collection by consultants.  In particular, both 
Argentina and Brazil have regulatory websites that contain very useful information on the concessions in those countries. 

The U.S. railway regulator, the Surface Transportation Board (STB) assembles the annual “Statistics of Class I Railroads” that contains very detailed statistics on the larger railways in the U.S.. 

  • By contrast, unfortunately, regulators in Mexico and Peru either do not collect, or do not report, any information.  The Association of American Railroads publishes a “Handbook of Railroad Facts” that is based on the STB “Statistics of Class I Railroads” that (because they are members of the Association of American Railroads (AAR)) publishes limited but useful data on two of the Mexican concessions and on the two Canadian railroads (CN and CP). The Canadian Government also publishes data pertaining to rail transport in Canada.
  • In the course of the Bank’s work in railway concessioning, the Bank was able to encourage some of the countries involved to create regulatory authorities that in turn required the concessions to file performance reports.  The Sitarail concession in Cote d’Ivoire/Burkina Faso was a particularly good example of a highly useful report that was, unfortunately, discontinued during the civil war.

International Finance Corporation (IFC)
In a number of cases, and with full permission, reports to IFC on the performance of their concession investments have furnished data that can be used in public reports.

Railway Consultants
There are a number of consultants and consulting companies who work in many of the countries advising on rail issues and who have been willing to provide data they collected and to comment on data collected by the Bank.  In some cases, these consultants developed specific analyses of concession performance in Africa and Latin America that included data that were used in the CDB.

Journals and newspapers
Finally, when nothing else is available, some data have been taken from newspapers or journal articles.  This is often the source of recent data (2006) that has not yet been officially reported.  These data are replaced when more official data become available.

As discussed in the introduction, however, the existence of multiple sources has inevitably meant that there are conflicts in the data between or among sources and/or that some information has been taken from one source and some from other sources.  These conflicts arise for a number of reasons.  Definitions differ, partly because of railway practice and partly because of differences in accounting requirements.  Sources differ because reporting time periods differ (the UIC’s “International Railway Statistics” usually is issued after a two-year time lag, and may have later data than the data reported in the initial annual or Flash statistics reported by a railway) or because some statistics refer only to the main national railway company and exclude smaller railways or niche operators.  Labor force (“staff”) levels are often reported to one source as the yearly average, and to another as the year-end level, which can be significantly different if the labor force is changing rapidly.  The two major Canadian Railways own large subsidiaries in the U.S. and it is not always clear whether the “Canadian” statistics report solely on Canadian operations or also include the performance of the U.S. subsidiaries (if, indeed, they can actually be separated).
Along side the conflicting or unclear definition issue is the issue of gaps in the data.  In some significant cases, railways withhold or aggregate data for reasons of commercial confidentiality.  A good example is Spoornet in South Africa where the latest available reports include significantly less data than used to be regularly made public because Spoornet is being more and more deeply integrated within its parent Transnet, and the parent apparently does not want to reveal details of its constituent parts.  In other cases (Nigeria), the railway simply does not report (or calculate) normally available data such as tonne-km (tonnes are reported, but not tonne-km).  Even in the UIC’s “International Railway Statistics,” members often simply do not report data in disaggregate form, such as separate reports of freight and passenger revenues.


Permanent URL for this page: