For public works projects of this magnitude, the attention tends to focus almost solely on the capital costs and thus to ignore the expenses of operating and maintaining the facilities once they are in place, not to mention those for replacing rail cars, buses, or other items. Further, as the system ages, unit operating and maintenance expenses will increase.

the

In the case of METRO, it appears that the total capital outlays will range somewhere between $5 and $6 billion, or have equivalent annual costs of about $250 to $500 million, depending on the interest rate chosen. To these capital costs must be added those for operating and maintenance, the latter of which will be at least $200 million annually by the year 1990 for both rail and bus service if the present WMATA plans are followed. Thus the estimated METRO operating and maintenance costs (not to include those for facility and equipment replacement) are potentially of the scale of annualized capital

costs.

Three points should be made at the outset: (1) By WMATA projections, about one-third of the maintenance and operating costs will be extended for bus services and the remainder for rail service on the adopted 98-mile regional rail system. With bus operations usually being more labor intensive than rail, it is reasonable to expect the bus costs to become a larger proportion of the total if transit wages for large city operations continue to rise as fast as has occurred during the last two decades. As a consequence, there may be considerable pressure in future years to trim bus service below levels anticipated by WMATA. This seems especially true since policy makers seem reluctant to trim rapid transit service once the heavy costs of fixed facilities are in place. (2) With under one-half of the rail portion of METRO under construction, the opportunity exists now for effecting operating and maintenance costs (both rail and bus services, not to speak of effecting capital outlays for way, structures and rolling stock) by a reduction in the planned 98-mile rail transit system. That is, if the METRO rail system were to be shorter than 98 miles, not only could capital requirements be decreased, but

also the costs for providing a certain amount of rail car mileage and ancillary bus mileage can be eliminated. With under one-half the rail system completed, the discussion in Chapter V indicated that about one-third of the fixed costs could be eliminated by severe curtailment, in addition to a reduction in the variable costs associated with rail and bus mileage. If alternatively the METRO rail system were extended no farther than at present, then at least $60 million annually could be saved in rail operating and maintenance costs. (3) The above maintenance and operating costs (about $125 million annually for rail and $75 million for bus) are the latest WMATA estimates for the level of service to be provided in 1990 but stated in 1976 constant dollars. However, if WMATA experience in the future follows that for other large scale public transit agencies in which increases in real wages and prices have been substantial and in excess of those in the private sector over the past twenty years, even higher real maintenance and operating cost levels can be expected. While no attempt to measure these increases has been made, some indication of the possible effects can be gauged by noting that real annual earnings for transit employees for the nation increased by about 25 percent between 1960 and 1974.2

It is not possible within this study to include a detailed analysis of the operating and maintenance cost data for the WMATA rail and bus operations or of their basis. The calculations are far too lengthy and time consuming. Also, the procedures and breakdown initialized for determining labor, materials, and in turn costs were changed (by WMATA or their consultants) from study to study, thus invalidating certain comparisons. As a consequence, only overall general statements can be made with respect to the reasonableness of WMATA's expected maintenance and operating costs for the authorized regional system in 1990 (even though the costs are stated in 1976 current dollars).

The three most recent WMATA etimates for bus and rail services indicate the following unit costs for 1990:

1The figures shown are preliminary but WMATA indicated that final values would

fall within a few percent of these.

2Jack Faucett Associates, "Inflation and the Transportation Sector: Trends, Problems, and Opportunities for Improvement," report prepared for the Office of the Secretary, U. S. Department of Transportation, October 30, 1974.

Given the fact that the rail service is to be air-conditioned and that vehicles will be over ten years old by 1990, the most recent unit cost figure for rail (about $2.58 per car mile) appears to be closer to the mark than earlier forecasts (without including any future increases in real wages and prices above the 1976 constant dollar scale). It is almost meaningless to compare this unit cost figure to that of other existing rail transit systems because of differences between operating characteristics (such as mileage between stations, speed, load factors, age of rolling stock and way, etc.) and other aspects (such as air-conditioning for METRO, local cost of power, etc.). However, simplistic comparisons do indicate that the 1975 WMATA estimate falls within the range of current unit costs for the larger rail systems and thus is not without merit. For instance, in 1973 the operating cost per car mile for rail rapid transit systems in New York was about $1.60, in Chicago about $1.20, in Boston about $3.70, and in Philadelphia about $1.40. With operating cost increases for these systems currently running about 8 to 10 percent annually, we may estimate that, measured in 1976 dollars, an additional 25 percent may be added to unit cost figures for other cities to make them comparable with the earlier cited $2.58 WMATA operating cost per car mile.

The operating costs for the alternative METRO rail systems analyzed here have been estimated using a unit cost of $2.58 per car mile corresponding to the WMATA estimate. The number of rail car miles required for each of the alternative systems has been assumed to be proportional to its track mileage. This probably tends to overstate the costs of Alternatives I and II since not

only would proportionally fewer miles be served by each train, but fewer trains would be required since there is some drop in estimated patronage (15 percent for Alternative II, 35 percent for Alternative I). However, no reductions beyond the proportion indicated above were assumed since these would partly be offset by the fact that some components of operating cost are tied to track miles as opposed to car miles (maintenance of way expense, for example); others are related to deadheading and turn-around costs and these might not drop proportionally; and others are due to minimum service frequencies at off-peak times and these costs would continue to be generated even if patronage dropped

somewhat.

Table 23 shows estimates of operating costs. The first line of the table shows the relative operating costs of the three systems. The second line shows WMATA estimated operating costs for the ARS and operating costs for the other alternatives based on this figure and the percentages in line 1.

The bus operating costs shown in Table 23 are based on the impacts on bus system operations sketched at the end of Chapter III.

Total operating costs thus estimated range from $213 million (in 1976 dollars) for Alternative I to $284 million for Alternative III.