Trouble in River City
In Meredith Willson’s classic play and later movie “The Music Man,” Robert Preston perhaps said — or sang — it best:
Trouble, oh we got trouble,
Right here in River City!
With a capital “T”
That rhymes with “P”
And that stands for Pool…
Yes, we got trouble, trouble with a thing called Pool. Today, though, we are talking not billiard tables but the national pool of Amtrak services. Fifty years after “The Music Man” won an Academy Award, five Tonys, one Grammy, and twelve weeks as number one on Billboard’s charts (plus an additional four-and-a-half years as a best-selling record), passenger railroad history may again also recapitulate the dismal events documented by Trains magazine in the mid-1960s.
The last decade has seen a resurgence of corridor passenger rail service in these United States. From California to Maine, and from Seattle to Miami, states that have seen ridership and tourism blossom from better rail service are looking to add trains. But no state seems eager to pay much more for today’s service.
The Passenger Rail Investment and Improvement Act of 2008, better known as the PRIIA, is set to levy a burdensome tax on the states, in the form of funding 100% of what Amtrak says is its cost of providing corridor services. Frank Wilner, in his column, “Competitive bidding for corridor passenger rail service in play“, Railway Age, 2 February 2013, counts 27 such services in 19 states.
The Wall Street Journal, just last month, considered what might happen (“States Weigh Picking Up Train Tab” by Mark Peters, 22 January 2013). Some states might cut services, while others might eliminate some routes altogether.
Meanwhile, down in Maine — and we say “down” since Portland and Rockland are down current from Boston, and clearly Boston is the Hub of the Universe, or at least it was to Oliver Wendell Holmes — down in Maine, we read, “Rockland passenger train service likely to see cutback,” and to quote, “…according to Maine Eastern’s Vice President and Director of Passenger Service Gordon Page, the Downeaster schedule is not one that allows for adequate connections [at Brunswick] to Rockland. As a result, Maine Eastern is likely to reduce its two round-trip seasonal passenger train trips from Rockland to Brunswick to one.” (Bangor Daily News, 4 February 2013) This on the heels of Amtrak actually adding three daily trains north of Portland: “Freeport, Brunswick welcome long-awaited passenger rail service to Boston” (Bangor Daily News, 1 November 2012)
Here we have a prime example of what might happen if PRIIA-induced cuts sweep the nation.
A Review of the Network effect
What happens as we add cities and connections to a rail line? The phenomenon is known as the Network effect. Wikipedia explains, “The classic example is the telephone. The more people own telephones, the more valuable the telephone is to each owner.” Furthermore, “Over time, positive network effects can create a bandwagon effect as the network becomes more valuable and more people join, in a positive feedback loop.”
The Maine Eastern, for example, serves four stations: Brunswick, Bath, Wiscasset, and Rockland. Their pricing and schedule page conveniently lists the six city-pair travel combinations: Brunswick to Rockland, Bath to Rockland, Wiscasset to Rockland, Brunswick to Bath, Brunswick to Wiscasset, and Bath to Wiscasset.
In general, the utility factor of a train in relation to the number of stations, N, is:
( N x ( N – 1 ) ) / 2
Where N=4, as for the Maine Eastern, this is ( 4 x 3 ) / 2 = 12 / 2 = 6… as we see in the Maine Eastern timetable. The equation above is known as Metcalfe’s_law. In the real world, of course, the value of each node (station) varies with its population or tourism level; see also Reed’s Law and Dunbar’s Number for some examples of how the Network effect may be less, or greater, than this simple equation might suggest.
We cannot consider the Rockland train without understanding that most of its riders arrive on Amtrak’s Downeaster. The first train from Boston arrives Brunswick at half past noon; Maine Eastern’s morning departure left two hours previously, leaving your only option for connection as the 6:20 p.m. departure which arrives Rockland at 8:15 p.m. (consulting Amtrak’s and M.E.’s most recent timetables). Returning, your one choice is the southbound Downeaster leaving Brunswick at 5:55 p.m. — and the 3:55 from Rockland is scheduled to arrive at 5:50, leaving only five minutes leeway… which would be a perfectly ordinary connection in the United Kingdom or Germany, but perhaps a little close for comfort here.
Shift one of those timetables half an hour, and the connections fall apart, and ridership evaporates. But keep even one good connection and let’s see what happens when we consider the two trains as one service: N changes from 4 (M.E. stations only) to 14 with the southernmost being Boston’s North Station. Now instead of 6, we get (14 x 13)/2 = 91… fifteen times the utility factor. We are even ignoring the multitudinous connections at North Station (the MBTA commuter trains, the Green Line streetcars and the Orange Line subway).
But the reverse of the Network effect is also true: As people, or cities, or more generally “nodes” leave the network, the value of the network plummets ever faster.
Cut that connection at Brunswick, and the Maine Eastern trains will quit. Cut half the Hiawathas in Wisconsin, and you ratchet up the cost of Illinois running its trains to St. Louis.
And suddenly, it’s 1963 all over again. One train-off begets three others, and the entire network unravels in ever-increasing disarray.
The River Cities, of course, was Amtrak’s Kansas City – St. Louis – Centralia – New Orleans train, discontinued in 1993. (Wikipedia entry) Today, to travel from Kansas City to New Orleans, you must first take train 316, departing at 4:30 p.m., to St. Louis where you arrive at 10:10 p.m. Then you board a bus at 11 p.m. for Carbondale, arriving at the eye-opening hour of 1 a.m. You then await the City of New Orleans for transfer back to a comfortable railway seat at something like 1:20 a.m. One easily imagines a single bus should suffice to handle the load of a couple dozen bleary-eyed folks ready for this encounter in the wee hours.
Train 316 is one of a pair, the other being 314, travelling daily between Kansas City and St. Louis. Start cutting that already minimal corridor service, and even that hardy few dozen connecting passengers go away. Repeat that a few places down the line and the City of New Orleans itself might be back to only fifteen restless riders.
What must be done?
The harsh medicine of the PRIIA is, in the long run, probably a good thing; if done right, any hidden acrimony over who-subsidizes-whose-train ought to be eradicated. If done right, we may be able to bury the debates about whether long distance trains pay for the corridor trains, or vice versa. (Momentarily omitted from this discussion is the Northeast Corridor, a whole other kettle of fish.)
The question on the floor is — How will the several State Departments of Transportation work with Amtrak, with each other, with the host railroads, and possibly with companies like Herzog and Veolia who already operate commuter trains, to resolve this situation? Because without a coherent plan, each state will find its costs increased not only by what Amtrak has already quoted them, but by the ever-increasing impact of the network effect of every other state’s decisions.