Analyzing Amtrak

In his article in TRAINS’ January 2019 issue (“Amtrak’s Money Mystery”), Bob Johnston points out, correctly, in my opinion, that “The expense burden has become Amtrak’s justification for systemwide cost cutting.”  This, as he discusses near the end of the article, also creates the problem that if a train is eliminated for cost reasons, system overhead costs will be spread over the remaining runs, thereby increasing the costs of the trains that were not eliminated.

While this is true, in a well-run organization, true overhead costs (i.e. those that can’t be ascribed to a particular service, such as Human Resources, Legal, Corporate Finance etc.), should be much smaller than direct operating costs, and, in any case, if the organization shrinks overall, good management does its best to shrink overhead, as well. 

The really startling news, in my mind, however, was the statement that Amtrak engaged the DOT’s Volpe Center “to construct clear assignments of revenues and costs to replace Amtrak’s previous system.”  Why? Because that system failed to provide “reliable cost accounting essential to making prudent business decisions”. 

In reality, another item further upstream of accounting policies and economic analysis is also at fault, namely that Amtrak has no formal raison d’etre, beyond what Bob states as the need to “sustain and grow surface transportation mobility options nationwide”.  This is similar to language in Public Law 91-518, which created Amtrak; it indicates that this legislation was to “provide financial assistance for and establishment of a national rail passenger system, to provide for the modernization of railroad passenger equipment, to authorize the prescribing of minimum standards for railroad passenger service”, and in Section 305, it refers to “the operation of intercity trains operated for the purpose of providing modern, efficient, intercity transportation of passengers”, as well as mail and express. 

Pleasant platitudes, but how can these be measured and assessed?  If Amtrak had been an ordinary corporation, albeit with initial funding provided by the Federal Government and payments from those railroads that were exiting the privately-operated intercity passenger train business, what would it have done?

In that case, Amtrak’s management would have been under the gun to generate sufficient cash flow to continue to operate after the initial investment was used up, and probably would have been very interested in having a system that let them know which trains generated positive cash flows, and how much of a drain on its finances the other ones were.

This did not happen, of course.  In reality, Amtrak was and is dependent on government funding to make up the difference between the revenues it brings in, and the costs it incurs (including overhead), whether via state-supported service payments, federal grants, or other sources (although I doubt that crowd funding is the answer).   

Airlines, for example typically have systems to measure the profitability (or lack) of individual flights and markets.  In the early 1970s, when I worked for TWA, we had a very sophisticated system that used the company’s operating budget as the basis for developing standard cost rates, including the use of appropriate cost drivers, as well as assigning revenues to flight segments both for passengers traveling only on that flight segment and those with connecting itineraries.  Forecast costs were re-calibrated based on actual results as the year progressed, and the system adjusted as needed to closely mirror the airline’s operating profit and loss figures.  (I’d like to say that this induced robust profitability, but at least we knew where we were losing money). 

Since the cost rates used to measure profitability were built from “bottom-up” budget information, there was less need to find ways to allocate costs, an example being station costs.  Bob alludes to Amtrak allocating snow removal cost to Miami; this wouldn’t have happened in the TWA system, since the Miami (airport) station didn’t budget for this activity. 

Effort had to be made, and money spent for the airline to accomplish this.  In 1972, a group comprising a manager and four analytical personnel were assigned full-time to the Flight Profitability Analysis system at the company’s main accounting office in Kansas City; other entities, including data processing and financial analysis at company headquarters in New York also participated.

I’d suggest that Amtrak take a hard look at current airline profitability-measurement systems (or utilize a large accounting firm with knowledge of them), to see how this type of tool could be adapted for its use.  While these are designed with private-enterprise company profitability in mind, the principles are similar; whether revenue comes from customers or a government entity, it’s still revenue.  Even if new equipment is provided on a grant basis, the value of the asset is known, and depreciation, allocated properly to individual routes and segments, is an appropriate method for doing this.  Maintenance cost, whether for new or fully-depreciated items, or whether expensed or capitalized, is a separate item. 

Knowing whether trains 1) have revenues below their cash operating costs, or 2) revenues exceed cash operating costs, but not fully allocated costs, or 3) by some miracle, revenues exceed fully allocated costs, is valuable information that management needs to have in order to make decisions about what to keep, what to get rid of, and, in the case on entity like Amtrak, know how much supplemental revenue will have to be obtained to keep the operation going.  What I’m advocating is not the “glide path to profitability”; I hope that it prevents Amtrak, via sound analysis, from being on the glide path to oblivion.  This is long overdue, and, while there is significant complexity inherent in the process, it’s not “Rocket Science”.  And finally, “overhead” is not, or at least should not be, the real problem.

(Photo by George W. Hamlin; Amtrak train 19 at Bristow, Virginia, September 7, 2016)