Railway Planning: Confused Economists^(*)

In a previous reference to Argentine railway economics (THE REVIEW, September 12, p. 407) it was pointed out that rail transport services in Argentina are rationed, indicating that users of the services are willing to pay more for them than they are asked to. In the same article a way of estimating the excess of the demand price over the supply price was indicated. This excess also serves to illustrate two other related issues: the position of the road transport industry and the quality of transport planning in Argentina. On the subject of the first of these (road transport) the previous argument may be briefly recapitulated in order to show yet another way of proving that the demand for rail transport is greater than is usually thought to be the case.

It is frequently argued that railway traffic has been steadily diminishing, that it will continue to do so and that shippers of goods are tending to transfer their preference to other carriers. This line of reasoning derives its seeming support from the trend of tonnage figures of freight carried by the railways. But such figures only show the quantities or facilities actually supplied. To assess changes in shippers' preference one has to use data relating to the quantities demanded. The users of rail transport services determine the number of wagons ordered; the number supplied is determined by the other side.

The first of the three accompanying tables shows the excess of the quantities demanded over those supplied. The freight tonnage for which transport by rail is ordered invariably exceeds the tonnage that the railways are able to carry. From this it follows that shippers of goods are always willing to pay more than the railway administration asks them to pay. (It is a case of the law of demand operating in reverse: if at the going price they want more than is available, they will take exactly what is available only at a higher price).

Since the goods that the railways cannot carry must be shipped anyway, the volume of road transport activity is largely determined by the state of the railways. The orders for wagons taken back may be assumed to represent the traffic that the railways lose to the roads. It may be estimated as amounting to at least 10 per cent of the road transport total.

According to the report of the World Bank Transport Planning Group (relating to the reorganisation of Argentine transport services) "an unnecessarily large amount of traffic is being diverted from the railways to the roads because of the inability of the railways to carry traffic that would otherwise be given to them."(⁽¹⁾) This statement can be made much more meaningful by reference to other information from the same source. Tables II and III show actual quantities of different commodities shipped by rail or otherwise. To the original information we have added the data showing averages and standard deviations, the latter expressed both in tons and as a percentage of the corresponding averages. The standard deviations measure fluctuations in the different traffic divisions around the average in each case. Thus, while total maize shipments (Table II) undergo variations of approximately 65 per cent, shipments of maize by road suffer much larger variations (82 per cent), while rail transport remains relatively unaffected, varying by only 44 per cent. This means that when the total demand for transport increases by 65 per cent, rail transport movement increases by a smaller margin since there is an effective limit to the railways' capacity. Road transport has to increase too and, to make up for the railways' deficiency, it does so by a larger margin than that of the total demand. On the other hand, should the demand fall by 65 per cent, railway traffic will drop at most by 44 per cent while road transport movement will fall by at least 82 per cent. The significance of the reckoning "at most by 44 per cent" is that it assumes that the demand for railway transport will still exceed the supply. The figures tabulated show what the railways were willing to supply, they are minimum estimates of the demand for rail transport. By the same token, road transport figures are maximum estimates of the demand for such facilities, since part of the freight carried by road hauliers is a spill-over from the railways.

The greater variability, or volatility, of road transport, compared with rail transport, shows that rail transport is preferred to road haulage. Since the railways' capacity is limited, the crucial test comes when total demand for transport falls. When there is less maize to be carried, road transport suffers most of the reduction because shippers tend to "stick to" the railways more than they do to the road transport services.

The preference for rail transport and the volatility of that for road haulage services is, of course, more marked the greater the average length of haul is. Maize is carried over an average distance of 316 kilometres, but dried fruit from San Juan and Mendoza involves transport over distances exceeding 1,000 kilometres. Thus the variability of road transport is, in the case of dried fruit not twice, but six times that of the railways. Road transport in Argentina, in fact bears evidence of serious instability and seems to be founded on highly unsound economic premises.

The fact that the road hauliers' share of total inland transport should be increasing is, however, quite meaningless. For example, I consume more margarine than butter. But there is no question that I prefer butter to margarine. The same applies to the transport of goods in Argentina. More goods are carried by road than by rail. But this in itself does not prove that the shippers' preference is for road rather than rail transport. The total volume of goods to be carried is increasing year by year. But there is an effective limit to the absolute quantity that the railways can carry. Thus, their relative share declines automatically. This does not prove that shippers are transferring their preference from rail to road transport. It does show that there is a great need for increased rail transport facilities.

There is, obviously, ample room for an improvement in railway services, but, unfortunately or otherwise, any such improvement will mean a loss of traffic for road haulage services.

The importance of distinguishing between the supply of, and demand for transport services has been stressed. It has been necessary to insist on the distinction although the issue may seem elementary. It is, but the railway problem must be recognised as a problem of supply and demand. Behind the supply side of the issue are to be found all the questions relating to cost, efficiency, personnel redundancy, etc. They have all been exhaustively discussed and examined by the experts---even if they have not yet been solved. But there has been surprisingly little discussion about the demand side of the railway problem. Here it is relevant to recall a well-known saying of Alfred Marshall to the effect that supply and demand are like the two blades of a pair of scissors. One alone is useless, the two have to be used simultaneously.

Although we have had many economic plans and projects in Argentina, no adequate analysis of the demand for rail transport has been made. The existing transport plans are, in fact, only half-plans and, as such, are utterly useless.

The authors of the Long-Range Transportation Plan for Argentina (the so-called "Larkin Plan") were well aware of the fact that the railway services were rationed, yet the economists working with them apparently overlooked the significance of the fact. They based all their recommendations on actual traffic, i.e., on supply alone. They made no attempt to estimate the demand for rail transport. In Appendix V of their report, which contains the chapter on railway rates and fares, and which is where one would expect to find an analysis of demand, nothing at all is said on the subject. There are descriptions of the rate-making system, and tables showing road and rail transport rates. There is no enquiry into the possible margin by which rail transport rates might be increased without losing traffic to the roads, and thus decreasing the deficit at zero cost, without new capital expenditure. Instead, the reader is afforded sundry strange recommendations such, for example, as the following: "A policy should be established of operating the railways on a self-supporting basis, with the necessary changes in the levels of passenger fares and freight rates to be made from time to time without undue delay. Two qualifications may be made to this statement however. First, users of the railway services should not be required to support gross inefficiency in the operating of the railways . . . In so far as existing deficits result from inefficient operations . . . the shipping and travelling public should not be required to bear the burden." (⁽²⁾) But who, it may be asked, should bear the burden? The man in the moon? Why should the taxpayer rather than the user of the service, or the inflation-burdened public, have to bear the burden? Perhaps the economists of the Transport Planning Group did not really mean what they appear to be saying, and that they felt they had to pay this degree of lip service to popular---if "phony"---beliefs held in Argentina at that time.

It remained for the National Development Council (CONADE) to repair the deficiency and produce its own draft of a railway plan which began with an analysis of demand prices for railway transport. The CONADE report begins by stating that "there is no substitution [between rail and road transport] since there seems to be no correlation at all between changes in the general level of railway rates and the short-run changes in the railways' share of total traffic" (⁽³⁾) and that "the low levels of freight (and passenger traffic) carried by the railways cannot be accounted for by any problem of [insufficiency of] the demand for them but rather of the unavailability and inadequacy of the supply" (⁽⁴⁾) . In plain English a low elasticity of substitution means that, whatever the rates may be, users do not tend to substitute road transport for rail and vice-versa; there is no road-rail competition. On the other hand, CONADE too knows that the railways ration their services.

However, the CONADE economists have fallen into a statistical trap. The reader may have noticed that the correlation they speak of is one between railway rates (i.e., the supply price) and changes in the railways' share of total traffic. The railways' share of total traffic measures, if it measures anything at all, quantities supplied. Thus they have found a correlation between supply prices and quantities supplied. This is the supply of railway services, but they call it demand. They are confusing supply with demand. They should have correlated the quantities demanded (wagons ordered) with the prices users have to pay. But they go on with their planning and leave the reader wondering about the reasons for the "demand analysis," since they never apply it. As in the "Larkin Plan," their recommendations, of which the most controversial are those relating to the closure of branch lines, are made only on the basis of the actual passenger and freight traffics and the actual revenue from them. Again no attempt has been made to assess the traffic and receipts that might be obtained in the absence of rationing. This is very bad economic analysis indeed. If it were representative of all economic planning in Argentina, one would have to mistrust the lot.

There was a time in the early 50's when potatoes were scarce. Housewives had to queue for hours to obtain a few pounds---or none at all. Imagine what their reaction would have been if---with total disregard of their unsatisfied demand for potatoes---they had been told that what was wrong was that too few potatoes were being bought; that the low level of potato consumption proved that the housewives had shifted their preference to other foodstuffs and that the solution to the potato problem was not only to produce less of them, but even to make them wholly unobtainable in large parts of the country. This is obviously nonsense, yet it is by an analogous process of reasoning that the planners hope to solve the railway problem.

The crux of the matter is that we want to have a free enterprise economy, which means an economy regulated by supply and demand alone. But our economists are seemingly unable to distinguish supply from demand.

Sylvester Damus

*. Published in The Review of the River Plate, October 31, 1966, pp. 149-152.

1. Ministry of Public Works and Services, A Long Range Transportation Plan for Argentina, Buenos Aires, 1962, Appendix V, page 70.

2. Ibid., Appendix V, page 3, italics are mine.

3. CONADE, Informe sobre el Plan de Transporte Ferroviario, 1965, (mimeo.) Chapter I, page 4.

4. Ibid., Chapter II, page 1.