Argentine Railways: HOW INEFFICIENT ARE THEY?^(*)

In previous articles in THE REVIEW on the subject of the economics of Argentine rail services, particular attention has been called to the distinction that should be made between demand and supply prices for these services. The present article is a continuation of the series.

Readers of THE REVIEW of 30 or 40 or more years ago will recall that in those days, before State ownership, both public opinion and successive Argentine Governments pressed consistently for lower railway fares and freight tariffs. The British concessionary companies were under constant pressure to accede to these demands and, over the years, a mass of deeply ingrained prejudice against the railway companies was built up. Much of this antipathy towards the railways has persisted, although they have been State-owned now for 20 years. Incidentally, it is interesting to note that one of the purposes which this magazine was designed to serve, when it was founded in 1891, was that of exposing the fallacies on which so much public criticism of the railway services in those early days so often rested. In the interval the services seem to have failed significantly to improve their public image, although a strong case for "the defence" can be made out, as the data furnished in the accompanying tables suggests.

The fact is that no objective and competent statistical study of Argentine railway charges, in real terms, has yet been made. This is the more surprising when it is considered that, in any proper assessment of any market, whether for goods or services, the deflation of nominal prices or charges is essential.

When, a few years ago, the Senate Railway Committee made its reckoning of the accumulated increase in railway charges, it computed nominal value data only, and, from this "analysis" concluded that past increases had been excessive on the grounds that the---nominal value--- upward trend denoted had been "so steep."

The Larkin Mission was scarcely more realistic in its efforts to solve the riddle of what was wrong with the Argentine railways. It compared the rates applied in 1950 with those decreed in 1961, divided one by the other to obtain percentage increases and that was that.

The only attempt that was made to get anywhere near a deflated reckoning in this analysis of the railway deficit is seen in the computation of operating ratios (represented by expenditures divided by income or revenue). On this reckoning nominal costs are seen to have risen more steeply than did revenue. But this is all. Revenues are used as the deflator; costs or expenditure are measured in terms of revenue. The procedure is deficient in that it fails to reveal what was happening to expenditure and income in real terms, or to what specific causes the deficit was attributable.

If there is anything undesirable about public ownership of the railways it is probably that the ignorance of public opinion in matters of railway operating tends to be more pernicious in its effects when the services are government-operated than when they are in private hands. This is apparent from the data set out in Table II from which it may be seen that the privately-owned 'bus companies were allowed to increase their fares more frequently and by higher cumulative percentages than were the railways.

When railway fares and tariffs are increased it is invariably to the accompaniment of widespread public protest, yet when the 'bus companies are authorised to raise their fares---which happens more frequently--- the increase is accepted as reasonable or inevitable. Indeed the greater frequency of 'bus fare increases renders smaller percentage increases sufficient. This gives rise to the belief that, because the margin of increase in 'bus fares is less than that involved in the case of the railways, the former are relatively cheaper whereas the truth is that they are becoming dearer and are overtaking railway fares.

There can be little doubt that the refusal of successive governments to allow the railways to raise their rates to an extent commensurate with the general rise in the price level is one of the major causes of the increase in the railway deficit to its present almost unmanageable proportions. Government action in this respect has always been too slow. The recent announcement that railway fares and tariffs will not be increased again in the current year is only another pointer to the persistence in official circles of a lack of true understanding of the problem. The Government's promise should be, not that a given nominal level of rates will be maintained, but that fares and tariffs will not exceed the real level registered in some specified previous period, e.g., that of say 1948.⁽¹⁾

The irony of the present situation is seen in the fact that, notwithstanding the steep nominal increase in fares of last December, and the recent inauguration of no less than five new 'bus line services to San Carlos de Bariloche, the General Roca Railway found it necessary to control closely the tickets sold for the journey to Bariloche, to the point in fact that it resumed recourse to the sale of the so-called "pasajes personales." These tickets bear the name of the passenger to whom they are issued and cannot be used by anyone else. This is a cumbersome procedure, but its adoption was clearly necessitated by the fact that a black market had developed in railway tickets for Bariloche. If the railway problem is not tackled realistically we may soon have black markets in tickets to almost everywhere.

Assume that in 1943 the railways were so efficient that they could provide transport facilities---at the then current ruling tariff---up to the limit required of them. There would have been, in that event, no back-log of waggon orders and no rationing of the supply of waggons.⁽²⁾ This is only another way of saying that, in these circumstances, the demand price would have been equal to the supply price. The real demand price for the transport of goods was thus a unitary symbol. A service sold for one peso had a nominal and real demand value of one peso. In 1957/58 (November to October), however, the supply price of this goods-transport service (which, in 1943, was one peso), had risen to 6.37 pesos. But this was the nominal reckoning; in real terms the supply price was between 50 and 67 centavos (see Table I).

Since the demand price is a measure of the value which the user assigns to the service the difference noted in the interval (i.e., between 1943 and 1957/58) indicates--other things being equal---the change that has occurred in the quality of the service.

It will be recalled from a previous article in this series (THE REVIEW, January 31, 1967) that the demand price for rail transport exceeded the supply price by just over 50 per cent. Thus, in 1957/58, the demand price for rail transport exceeded the supply price by just over 50 per cent, which put it between 75 and 100 per cent of the 1943 level. It may be reckoned, therefore, that, between 1943 and 1958 the quality of the service declined by at most 25 per cent. In the second quarter of 1965, the demand price exceeded the supply price by 63 per cent. Thus, a service which in 1943 was worth one peso could in 1965 have been sold for 83 centavos, the difference of 17 centavos being a measure of the deterioration of the service since 1943. In the interval between 1958 and 1965, this factor underwent virtually no change. In view of the widely publicised references to the "appalling deficiencies" of the services today, it is difficult to say whether this statistical reckoning indicates a larger or lesser change in quality than that usually alleged.

Finally, let us rehabilitate the operating ratio, but not without due attention to its correct use. It has been reported that in 1965 this ratio amounted to 2.18 (THE REVIEW, December 30, 1966). Since, in 1965, the real value of passenger fares was between 0.21 and 0.37 pesos for suburban services and goods tariffs were between 0.36 and 0.49 pesos, the weighted average of the real revenues in 1965 is approximately 0.364. Now, for every peso which the railways took in, they spent 2.18, so that their expenditure per unit of service was 2.18 x 0.364 = 0.79352. This is extremely interesting and what follows is no mere coincidence.

In 1965 the real value of expenditure per unit of service was 0.79 pesos. If the railways had charged a real price of one peso per unit---and suffered no loss of traffic---they would have been able to show an operating ratio of 0.79. This, it may be noted, is identical with that obtained in 1943. In that year the operating ratio was 0.71 for the State Railways, 0.82 for the privately owned railways and 0.79 as the overall average (M.O.P., Dirección Nacional de Ferrocarriles: Estadística de los Ferrocarriles en Explotación, Vol. L, p. 351, col. 8 ).

Of course, the railways could not have charged one peso per unit and still sold the same quantity as they did in 1965, since the demand price was a little more than 0.80 pesos for goods transport and probably slightly lower for passenger transport. Even if the railways had been allowed to charge as much as the demand price---in which case there would have been no reduction of output---they would have found it extremely difficult to earn even a microscopic surplus.

But the more interesting conclusion is this: railway operating costs did not rise by as much as is usually thought to be the case. Apparently, they did not rise at all. This is far from being surprising, although many readers will not have expected it. The reasons for this result are not far to seek. But they have to be explained. This is a matter of considerable importance, having serious implications for past and present railway policies. It will be dealt with in a subsequent article.⁽³⁾

S.D.

*. Published in The Review of the River Plate, April 21, 1967, pp. 87-89.

1. This was a suggestion of what decades later became advocated as "price-cap regulation."

2. The assumption is made for the purpose of the argument only. In fact, there was rationing already in 1942/43. Railway rates were artificially depressed and the companies in financial difficulties even then.

3. The next instalment was not written. It suffices to say that the operating ratio of Argentine railways, equal to 0.79 when properly evaluated at the prices that the public was willing to pay rather than the prices actually charged, compared very well with the operating ratios of US railways that fluctuated between 1960 and 1967 from 77.95 to 79.52 (Association of American Railroads, Statistics of Railroads of Class I in the United States, Years 1957 to 1967, Washington D.C., August 1968, page 1.)