Monday, October 9, 2017

Amtrak: Avoiding the Obvious

According to The New York Times, Amtrak’s management “knew for years that they would have to replace large sections of deteriorating track in Pennsylvania Station in New York City.”[1] The management instead had engineering crews apply “short-term fixes to rows of rotted ties, crumbling concrete and eroded steel.”[2] Incredulously, the management was putting off replacing the tracks in part “to give work time to a nearby passenger hall renovation.”[3] Additionally, the management sought to minimize taking tracks out of service even on weekends so as not to disrupt service. In 2017, three accidents at the station finally got the management to commit to undertake an emergency repair program that “cut back service through the summer for thousands of passengers daily.”[4] Even by the objective of minimizing impaired service, prioritizing a hall renovation and putting off needed track repairs are problematic. The deeper problem is that of seriously misjudging utility.
The utility gained by passengers from a renovated hall is rather superficial, whereas the disutility from derailments at a station could result in passengers deciding to no longer travel by train. Put differently, limited service disruptions on weekends pale in comparison to having trains derail coming into, or leaving a station. Even if the latter are mistakenly deemed low-probability/high impact events, the business calculus that favors short-term fixes over long-term stability is problematic.
I suspect that Amtrak’s management had a hypersensitivity to passengers’ sense of utility because of basic impairments in the routine conduct of the trains. The route between Pittsburgh and Philadelphia, for instance, has been severely impacted as Amtrak trains must wait for other trains to pass because of track ownership. The prioritizing of freight over passenger trains is itself problematic, so the ownership of the tracks that Amtrak uses is as well. Barring outright ownership of the tracks, Amtrak should insist on contracts giving priority to the passenger trains.
I suspect that Amtrak’s management (including on the train level) has looked too tolerantly on delays. Traveling between California and Arizona on an overnight train, I was surprised (and dismayed) to learn that the train stopped somewhere in the desert because a man had been smoking marijuana in a bathroom. The conductor could have waited until the next station stop. A car’s designated employee even woke up all of the passengers in that car by shouting at the man in spite of the fact that he was no longer smoking. I was also surprised when an employee on the train arrogantly informed me that wifi is only for those passengers who had purchased sleepers. That customer service, and thus utility, can be so misjudged is itself a red flag on Amtrak’s management.
Besides having to put up with delays between station-stops and bad on-board service, having too many such stops on a given route can also be distressing to passengers. Why not have express trains run between San Francisco and L.A., for example, two or three days a week? Also, passengers have often had to accept traveling at slow speeds. To be sure, the local laws and bad track conditions go beyond the company’s control. Even so, that the company’s management decided to market its Acela train running between Boston and New York City as “high speed” nonetheless set up passengers to be less than satisfied. I think only twenty minutes were cut by the “high speed” train between the two cities.
In short, the utility of the product in this case may itself be so bad that the company’s management became distorted on just what constitutes real passenger utility. The management’s attention has not been on core matters, whether they be track replacements or the very functioning of the trains on the tracks. The lesson for all companies is that attention should be primarily on the basic quality of the products or services themselves.



[1] Michael LaForgia, “Delaying Repairs on Decrepit Tracks,” The New York Times, October 9, 2017.
[2] Ibid.
[3] Ibid.
[4] Ibid.