ProRailHelping people travel by train
-- and have more trains to travel

(Note: The following was published in the December 31, 2004, issue of Keeping Track, the periodic newsletter published by ProRail.)

According to recent reports, the full cost of the Midwest Regional High Speed Rail Initiative is up to $7.7 billion now. In 1998 it was estimated that capital costs (track work and trains) to create a 3,000-mile system in eight states would come to $3.5 billion, more or less.

The increase is not unexpected. All large-scale capital projects cost more than the original estimates, which in this case were were somewhat seat-of-the-pants affairs. Besides, as more detailed field engineering has been performed, the estimated costs per mile of infrastructure work have increased. (And prices of almost everything everything connected with railroading at least are up, in raw dollars if not necessarily in inflation-adjusted dollars.)

The current breakout calls for $6.6 billion for infrastructure, $1.1 billion for rolling stock (60-some trainsets). But why such a large increase (a little over 100% in six years)?

Midwest Regional recently underwent a thorough restudy. Interestingly, the initial study was done carefully enough that very few system changes were proposed in the restudy.

The biggest change (by far) is the rerouting of the nine round trips between Cleveland and Chicago. The original plan used the same route (Norfolk Southern's ex-New York Central main line) as Amtrak's Lake Shore Limited -- Chicago, South Bend, Waterloo, Bryan, Toledo, Cleveland.

How to get NS, famously opposed to even one more Amtrak operation on that route, to accommodate eight more round trips was a needle left to be threaded at a future date.

In the revised plan a totally new routing was nominated as far as Toledo, namely Chicago-Fort Wayne-(Fort) Defiance-Toledo. The Chicago -- Fort Wayne segment will be over the ex-Pennsylvania main line (later Penn Central, Conrail, and [briefly] CSX, now a short line, the Chicago, Fort Wayne, and Eastern); Fort Wayne-Toledo over ex-Norfolk & Western ex-Wabash track, now understood to be an independent short line (Indiana & Ohio?).

This rerouting added greatly to the infrastructure costs: the ex-Pennsylvania main line, once the route of a whole fleet of PRR passenger trains (including the Broadway Limited), was impressively straight and flat and double-tracked, generally following US 30 between Fort Wayne and Chicago. It's still straight and flat, of course, but it's been single-tracked and sketchily maintained for twenty years (with a generation of pretty sketchy maintenance under the PRR and PennCentral before that).

The ex-Wabash line never saw much passenger traffic over a route that likewise is both straight and flat from a little east of Fort Wayne to Toledo. At a guess, the upgrading required of both segments, something like 250 miles, will be not unlike what is needed for the thirty miles between Madison and Watertown, namely just about everything except the grade.

Another large new cost is the planned access to the Fox Valley and Green Bay. The original plan assumed that trains to Green Bay would take the Canadian Pacific (ex-Milwaukee Road) main line to Duplainville and there switch to the Canadian National (ex-Wisconsin Central ex-Soo) line to Fond Du Lac.

However, even before the CN took over the operation, the prospective landlords objected that the freight operation on that single-tracked line was already at capacity and could not possibly accommodate five round-trip passenger trains without constructing a second track.

Since that would involve, among other things, new tracks through wetlands and other environmentally-sensitive zones, attention shifted back to the former Chicago and North Western route from Milwaukee via West Bend.

Most of the track on that line has been removed, but the grade is still in good shape and there's lots more population en route. (And a good chance for the state to make some pin money by charging CN rack rents if they ever want to use the line for detours or over-capacity operations.)

But whichever choice is made, it is obvious that the cost of getting to Green Bay and intermediate points is going to have to be a lot higher than the estimates in the original TEMS study; $7.7 billion may look like a lot of money, but for a 3,000-mile system serving eight states over 10 or so routes and with a phased construction plan extending over some thirteen years, it is actually quite reasonable for a transportation project.

The $6.6 billion for infrastructure comes to $2.2 million per mile. Who knows how much it would cost to add two lanes (one in each direction) to 3,000 miles of Interstate highway in the same regions, but it's unlikely to be any cheaper than $2.2 million per mile on average.

In any case, the plan still calls for 80% of the capital costs to come from the federal government, which has yet to come up with a penny. And the feds are broke and getting broker, and on top of that there seems to be no stomach for real national transportation planning in this country or anywhere else.

Recent articles in leading newspapers, The Economist magazine, and elsewhere, have bemoaned the fact that hardly any government in the world seems to be really thinking about looming transportation crises, the most outstanding exception being (of all things) Communist China.

Of course, Chinese planning isn't always distinguished: for a time the government was seduced by the Transrapid consortium into planning a 775-mile Maglev line from Shanghai to Beijing for the low, low cost of $23 million a mile, or $22 billion total. At least, that was according to the enthusiastic burblings in the coverage of the opening of the little Shanghai-Pudong Airport Maglev line in January 2004.

Well, here was a pair of eyebrows that went up at such low-ball figures. (Third-world construction worker wages?) Less than a month later, though, there was a new official announcement from Beijing: notwithstanding nine years of planning, the grand Shanghai -- Beijing Maglev project was being cancelled.

Among the reasons was the cost of "$66 to $90 million per mile"; clocking at something like 300% in a month, that must be a record for cost overruns. (Possibly the orignal estimate didn't include the powerful electric substations that have to be constructed at intervals along a line that long.)

Another stated reservation had to do with the problem that all existing infrastructure is useless for a Maglev line bridges, tunnels, existing rail lines. As Prof. Vukan Vuchic, a transportation expert, has drily remarked, Maglev does have advantages over steel wheels, but they are few and slight.