High-Speed Rail

Can High-Speed Rail Save American Manufacturing?

Downtrodden Michigan seems to be pinning its hopes on a revival of train production.

With the American automobile industry in free-fall, some in Michigan are hoping that a more potent train industry could provide relief. President Obama’s commitment of $13 billion for high-speed rail is seen as the first step towards reviving U.S. manufacturing. But those ambitions are overstated, and unless GM or Ford make quick changes in their operations or acquire a train maker, Michigan seems unlikely to benefit.

Last week at a town hall in Michigan, Secretary of Commerce Gary Locke told a reporter from the Huffington Post “As you see more construction of rail cars, high-speed cars, it’s going to require new engineering, new products and services and that’s the natural fit and extension for automotive dealers and suppliers and manufacturers.” Governor of Michigan Jennifer Granholm, in Washington, agreed, saying “We have lots of capacity in Michigan and workers who know how to make things.”

As I’ve made the case before on The Infrastructurist, even a sudden increase in train orders from American rail services (even with a required Buy America pledge) would never employ nearly as many people as does the existing auto industry. The simple fact is that fewer people are needed to build trains because you don’t need nearly as many of them in a train-using society as you need cars in an automobile-dependent one; the three top existing train manufacturers today — Siemens, Bombardier, and Alstom — collectively employ fewer than 100,000 people in their transport divisions. One of the reasons mass transit is efficient is because vehicles can be shared by thousands of individuals, rather than being monopolized by one person or family.

That said, a vibrant train industry would create hundreds of thousands of service jobs, but only in places where rail travel is encouraged. France’s SNCF has more than 100,000 employees, for instance. A huge train network with hundreds of thousands of daily passengers in the United States would induce the similar creation of hundreds of thousands of jobs. Yet California, with $10 billion already committed to rail services, is far more likely to get those jobs than Michigan, which has few state funds dedicated to fast rail.

The Wolverine State shouldn’t rest its hopes on the potential of the train industry. Auto plants — whatever Ms. Granholm thinks — cannot simply be converted to train manufacturing; they’d have to be completely rebuilt and retooled. The American car industry has no affiliation at all with train manufacturing and wouldn’t know where to start, because it has no existing research done on the subject; it would take decades before Ford or GM could compete with Siemens or Alstom on the production of high-speed trains. And while many Michiganders may “know how to make things,” so do many people in Iowa or Arizona. Why would train companies choose to build their product there? For new plants, foreign auto manufacturers have shown their preference in the past for Alabama, South Carolina, Tennessee, Indiana, and Ohio, not Michigan.

The state has a long way to go before it’s healthy again, and the train industry is unlikely to be its savior.

31 replies on “Can High-Speed Rail Save American Manufacturing?”

General Motors used to have an “Electro-Motive Division” which produced electric and electrodiesel locomotives. Pretty good ones, too. They had it from the 1930s to the 1990s.

They spun it off and it was bought by another company. So they blew their inhouse expertise right there. If they’d kept it, they really *could* have retooled for train construction.

Instead, the best they can do is to try to lure Bombardier to Detroit.

lure Bombardier to Detroit..

Bombardier is headquartered in Montreal. . . they know what Detroit is like. . .

Bombardier and Alstom already have North American plants. There’d have to be lots of trains ordered over long periods of time before they would consider building new plants.

First of all, yes a conversion to rail would require a massive overhaul of tooling, factory redesign, and employee retraining. In fact, the overhaul required would be to such a degree that it would take A) a severe loss of current business combined with B) large infusions of cash and heavy-handed direction from the federal government.

Oh, wait. That’s actually happening! So I don’t think it’s an issue.

Second, when you estimate the number of jobs produced, are you sure that your focus is not too narrow? We don’t gauge the economic vitality of academia based on the number of people employed university research programs, because we know that work also supports (directly or indirectly) pharma, biotech, IT, etc. So I would encourage you to think in terms of clusters…

Bombardier is extremely unlikely to move anywhere. Montreal is a half-decent place to HQ a large company like Bombardier and being in Canada they get lots of fat subsidies and other contracts that get handed out.

Also remember that Toronto has the 2nd largest transit system in North America and is busy undergoing a massive expansion. And regardless, most of Bombardier’s train manufacturing happens in Europe, because that is where the market is and that is where the companies that it bought started and still make trains.

A bit too pessimistic in your analysis I think. Sure it’s a long-shot in terms of train manufactoring bring lots of jobs to Michigan, but not so much as you make it out to be, in my opinion.

1. I think Granholm is talking about manufacturing capacity, not specifically the companies that would use that capacity. There’s plenty of empty factories all over the industrial midwest that COULD be retooled for train manufacturing (how ’bout GE’s Diesel engine plant in Erie? ;) ). That doesn’t mean that GM or Ford have to be the ones actually using the plants does it? I could see Bombardier buying up a condemned factory in Flint and using as a new production space for North America.

2. Ok, those train manufacturers only employ 100,000 hourly workers. But remember, that after GM and Chrysler downsize, there will probably only be about 150,000 hourly auto workers in the U.S. and Canada anyway! So adding even a small fraction of that, say 15-20,000 workers for train manufacturing is nothing to sneeze at these days. Maybe in 1978 it was, but not now.

3. And I think Michigan would be a good location for new train manufacturing just like Ohio and Indiana are (your examples). More-so than Arizona. Contrary to your assertion, no, there really aren’t as many people in Arizona that know how to make things. There is no history of manufacturing, no excess capacity, and no skilled workforce for manufacturing. So, yes, Michigan (like Ohio and Indiana and even Iowa) have a comparative advantage when it comes to manufacturing. To make up for the increased labor costs (as opposed to Alabama or Tennessee, although even that may be doubtful if the plant is unionized in the South), the state government will likely add incentives to set up the plant in Michigan.

So overall, while it’s no silver bullet, and it will take much more federal investment before the demand is there to set up a manufacturing unit in Michigan, it is by no means a complete non-starter.

Adam makes good points.

Let me add that the decision of the auto transplants to locate across the Midwest and South was partly escape to non-union states — and at least as importantly it was simply smart politics.

Did you notice the line-up last winter on bailing out the big US automakers? The Senators from Michigan were all for it. The Senators from Alabama, Mississippi, South Carolina, Tennessee, Kentucky … not so much. The foreign auto makers have bought a loud voice in Congress by locating their plants where they can count on loyal local support.

We could imagine a new plant, or three or four, to supply the future passenger rail needs in this country. One for high speed trains, fewer but pricier. Probably one for trams and light rail.

And another one or two for the thousands of cars that Amtrak will need in the next decade, if not yesterday. Ray LaHood and his team are talking about expanding frequency on existing routes. That will be almost impossible without ordering more coaches. If they, or Congress, get serious about reviving routes like the Pioneer, and adding trains along the Front Range from Cheyenne through Denver to Pueblo and maybe Albuquerque and El Paso, or on dozens of other promising routes, then hundreds more coaches will be needed. And diners, sleepers, lounge cars and dome cars, baggage cars, and locomotives.

Then consider that Amtrak’s equipment is old, older, and oldest. Much of it will need replacement soon, sooner, soonest.

A couple of big orders could attract foreign manufacturers with bids tailored to catch the prevailing political wind. Locating a train car assembly plant in the Rust Belt would make good political sense.

Maybe Kawasaki will follow a strategy similar to the other Japan Inc member companies and buy political influence with an astute location for a streetcar plant. Worse things have happened to us than that.

Woody- Good point on the Amtrak cars. Not only are they talking about bringing some of the western routes back, but for all the high speed route planning I’ve seen, the planners have been talking about at least 4 round trips a day, with some corridors calling for 8 or more round trips. That right there would push demand up for new cars. Especially since as you say, Amtrak’s fleet is now mostly 30 years old or older.

Really, all these tie together with the fact that we need a dedicated funding source for rail like the gas tax is for the highway trust fund. This means that if want to seriously expand passenger rail and high speed rail we need to work with our congressional representatives to make sure the funding mechanism is put in place in the next transportation reauthorization bill.

FYI Kawasaki already has a railcar plant in Lincoln, Nebraska, which has built subway cars for New York City.

Toyota has more than 300,000 employees; that’s how big a large healthy automaker is.

Yonah is right: the economics of mass transit are such that it costs more and provides more jobs locally than cars, but less in one central manufacturing facility. To replace each of its 6,400 subway cars with a new model, New York would have to spend about $10 billion; these new cars would last for 40 years. If instead the subway were shut down and New York’s car ownership rate rose to the same level as this of the rest of the US, it would gain 4.5 million more cars, costing about $70 billion, and lasting 10-20 years. It’s this factor-of-20 difference that makes it impossible to run an urban economy on rolling stock the way Detroit (or Nagoya) runs on cars.

it would gain 4.5 million more cars, costing about $70 billion, and lasting 10-20 years.

Not to mention you’d have to tear down half of Manhattan to drive them and park them. Half the Bronx, Brooklyn, big chunks of Queens, the pricier parts of Jersey City…. Do that and Manhattan isn’t Manhattan any more. One of the reasons they are building ARC and East Side Access is that they are the cheaper options.

I don’t think we should focus on manufacturing jobs, but on real wealth.

Alon Levy writes a good example: NY’s subway costs $10billion, the equivalent car ownership would cost a multiple of $70billion over the same decades. That means New Yorkers have a multiple of $60billion to spend over the next decades that they wouldn’t otherwise, which not only makes them de facto wealthier but also creates American jobs in healthcare, internet retailing, electronic gadgets, etc. (for those interested in the current-account deficit, that means more American exportable products and services, such as pharma R&D, Amazon services, Kindles and iPhones, etc.).

In the late 1970s, as Carter was cutting the defense budget, Boeing scrambled for new markets and decided to start making rail cars. San Francisco Muni bought some. They were a disaster. Ask anyone who rode Muni Metro in the 80s and 90s. But by then, Reagan was in, defense spending was up, Boeing got out of the business and stopped returning Muni’s calls.

I look forward to seeing new rail building capacity in the US, but I hope it’s provided by a company that knows what it’s doing, and in the beginning that may really have to be a “foreign” company like Bombardier. I’m nervous that “Buy America” provisions, designed mainly to protect American jobs, have a way of preferring US-based corporations, which is another matter entirely.

Be vigilant about this.

It’s this factor-of-20 difference that makes it impossible to run an urban economy on rolling stock the way Detroit (or Nagoya) runs on cars.

If Gov. Granholm is hoping that she can run an urban economy on rolling stock, then yes she is dreaming and it will never happen. But is that actually what she is hoping or rather something more modest where some fraction of the excess manufacturing capacity in the state right now could be put to productive use building rail cars and engines?

Diego –
I think it’s odd you say let’s not focus on manufacturing but then one of your examples is “electronic gadgets”. Aren’t those manufactured somewhere? And unfortunately I fairly sure they’re manufactured in a low-wage country like China.


I obviously was referring to software and design, not manufacturing, which is the lion’s share in electronic gadgets’ cost.

Diego: the actual difference is probably a fraction of $60 billion, because subways incur most operating costs and maintenance than freeways. My point is not that mass transit is cheaper than cars, but that the costs are spent locally, on train operators and track workers, rather than on manufactured goods.

Adam: Michael Moore has called for revamping Detroit for manufacturing rolling stock. He’s not the only one – I’ve seen some lesser known people suggest that as a way of wedding clean technology with semi-skilled heavy manufacturing.

Adirondacker: they’d probably just build huge parking buildings, and get people to take the bus. There are dense cities with high car ownership, like San Francisco and Sao Paulo. In Sao Paulo the main consequence of high car use is not endless parking lots, but severe air pollution. I’d venture a guess that the extra health care costs that would come from extra driving would trump all other costs or benefits.

Alon- Well to the degree you’re debunking some oversized dreams by the likes of Michael Moore, then yes I agree. But I also don’t think it’s unreasonable to try to bring some jobs into the industrial midwest through rolling stock as a means to retool and reuse some of the excess capacity there rather than watch it further whither into derelict landscapes.

And yes, trains serve more people so there is less volume demand to manufacture individual units. But of course it’s not a one to one comparison since it takes more man-hours, capital, etc. to manufacture a train car or engine than an individual automobile (think of how many workers it takes to manufacture on commercial aircraft at Boeing). So, no there won’t be 100,000 train manufacturing jobs coming to Detroit. But I think the state would be very happy if 10% could happen. That would equate to half the planned total GM hourly layoffs across the country.

It’s possibly, but unlikely. The existing rolling stock plants in the US are mostly in the Northeast, to take advantage of proximity to the largest markets. For the Midwest to lure Kawasaki away from Yonkers, Alstom away from Upstate New York, or Bombardier away from Vermont, it will need to spend large amounts of funds in subsidies, or start an ambitious high-speed rail program eclipsing the existing market in New York.

The post could be better titled “Should high-speed rail save American manufacturing?”

High-speed rail would have profound economic benefits as game-changing as the interstates had more than a half century ago.

But high-speed rail should be decoupled (no pun intended) from any efforts to revive the American manufacturing sector.

There are deeper problems with manufacturing that high-speed rail policy will not solve, or worse, end up masking the problems that are there.

If we forgo manufacturing the trains, we would instead reap the benefits of a far broader economic gain.

How? In the same path Japanese carmakers took.

Let’s get over the fact that the rolling stock will be foreign-built. Let’s even throw Buy America rules out the window. Right now, foreign builders have the labor, resources and know-how to give us the rolling stock we need. Neither they nor passengers need to be hobbled by creating redundant capacity which raises costs and/or diminishes quality.

Will be be paying billions to foreign governments? Yes.

It will all come back, but not in ways as visible as a mammoth factory churning out trains.

First, the foreign builders will need to source parts. We’re not building trains, but we do produce parts that can either be bought off the shelf or on lines much easier to retool than a whole rolling stock-making plant.

So builders will more than likely turn to American suppliers to source parts. It will start with smaller components and cascade up to more expensive, heavy-duty parts.

Second, the rolling stock will need to be maintained here. We won’t have the time or money to ship a train set back to Europe for repairs. The manufacturers will set up a repair base and train American workers. This is an even better economic multiplier since maintenance jobs pay very high wages, are apprenticeable (skills need to be passed on to others), are liquid (maintenance skills can be applied to other machines) and are portable (demand for workers is everywhere).

Third, we will hire American workers to run and clean the trains, as well as provide customer service. The benefit of high-speed rail is that paying a good wage is offset by high productivity (a ratio of passengers to train crews).

Fourth, once the trains are actually running, this would enable long distances to become day trips and what is now a day trip by car to become a commuting distance. This sets up a new dynamic of economic transactions.

For one thing, lower-value economic activities from high-cost cities can filter down to other cities along the line. Example: a major Chicago real estate developer is looking to expand, and it has two very important units: sales and data research. It needs to expand in both, but the high costs in Chicago put a limit to its expansion. Ah, but Chicago is Midwest hub for high-speed rail.

The sales office would need to be kept in Chicago, since the agents are the developer’s front line. They need to be close to where the market is. Data research is also important, but it is a job that can be done remotely yet still requires face-to-face interaction with the home base. So within Chicago’s sphere is Milwaukee, Minneapolis-St. Paul, St. Louis, Indianapolis, all of Ohio’s big cities and Detroit. These are all cities that are smaller metropolitan areas yet still have metropolitan qualities of life (recreation, universities, etc.). Any of these cities could vie for the data research office.

Also, once far-flung metropolitan areas would begin to understand what network effects are — how high-speed rail produces them and how they can benefit from them. In the Midwest, a declining metropolitan area like Detroit, Milwaukee or St. Louis gains more by being along the high-speed rail network then if they would have to play the diminishing returns game of economic development.

These activities have happened in Japan and Germany. This is also how American metropolitan areas grew, although rail has become an afterthought here.

By establishing just five rail hubs with networks extending out 500 miles, you will see these economic effects in more than three-quarters of where Americans live right now.

We wouldn’t even need those train factories now. But that may not stop the train manufacturers, who may see U.S. demand rise to the point where they would need to produce here.

This is not a dreamland scenario. The Japanese carmakers have increasingly become American carmakers that happen to have Japanese names. All have U.S. headquarters (Toyota, Honda and formerly Nissan were all but next-door neighbors in Southern California) and all are planning for more production in the U.S.

There are the more obvious ways how the fortunes of U.S. and foreign carmakers reversed, but not enough attention is paid to the silent process of the foreign carmakers exporting a few jobs at a time and building business relationships with local suppliers. This is the wealth production that we don’t see, but we tend to think as the factories that produced the wealth.

On the other hand, we can always demand that we make American high-speed rail as American as possible. In the near term, some lucky town gets a factory and a boost of self-esteem from its workers.

Then what?

If things go badly, this reinforces the stereotype that America can’t make anything well and the Rust Belt’s decline was self-inflicted and deserved.

If things go well, that is also bad news. Why? They solved the make-work problem. They now must figure out the keep-work problem. If workers build a train of solid workmanship and durability, they are not needed because they did a good job. Factories can’t hum along for a few years, producing a machine that last decades and sit dormant until a new round of equipment is produced.

America didn’t lose its manufacturing base because we did a poor job. We lost our manufacturing base because we did our jobs too well. The Upper Midwest is littered with abandoned factories because when those factories became successful, they had enough knowledge, technology and capital to replicate their operations anywhere. They were no longer constrained by location. In “Cities and the Wealth of Nations,” Jane Jacobs calls these industrial activites transplants. Relying on transplants is an unhealthy way to build an economy, because if a corporation can choose to place its businesses anywhere … they will.

It’s the locomotives that will be the locomotives of economic progress, not the places in which they’re made.

Wad: most of what you say is true, but with the US auto industry specifically, it really did do a bad job. Detroit wasn’t a healthy city region to begin with, but in the 1970s, the inability of GM and Ford to compete with Toyota and Honda struck a blow to the region. Detroit crashed more than any other city of similar size, and would’ve crashed even more if GM hadn’t gotten the federal government to impose import quotas on Japanese cars.

Alon, if I may be so bold, I would trace the beginning of the end for Detroit back to the end of World War II.

Even if the Big Three had always been building bad cars, they still would not have been knocked off their perch. After all, bad cars make for good mechanics and tinkerers.

Anyway, Detroit didn’t need good cars because people were scooping them up like mad anyway. It’s positive feedback. Detroit thought it was doing well, so it continued doing things the way it had always done.

After World War II, the U.S., the Allies and the defeated nations embarked on an idealistic policy to avoid future wars and allow all societies to build themselves up. One key plank was to form a beneficial market system with each country encouraged to rebuild through industry.

The U.S. flourished during the next two decades while Europe and Japan had been rebuilding. Then comes along the turbulent period of the mid-1960s onward, with the first shocks of deindustrialization.

This happened at a time when the industries of Europe and Japan had finally matured and taken stride. War had the effect of modernizing once bombed-out factories, and trade policies helped Europe and Japan get back on their feet.

Europe and Japan also had industrial policy that tempered corporate profits with corporate responsibilities. Germany and Japan, for instance, required a percentage of revenues be set aside for R&D. Here in the U.S., we’d never go for such nonsense, since the free market knows best. :>

As for Detroit — the city region — it died for the same reason it became rich. It specialized in building cars. The whole economy revolved around one industry, which produced a preponderance of both jobs and wealth. The factories in effect became a supply region.

Detroit was blindsided, and it chose to cope by waging a low-intensity racial civil war. Whites not only left most of Detroit, they took their wealth with them and waged a suburb vs. city zero-sum game of economic attrition. The whole city region still has not able to escape this toxic state of affairs. It’s a shame, too, since Detroit occupies a unique geographic location that it could leverage to become a powerhouse again.

Corey Berger is quite wrong about Toronto having the second biggest transit system in North America. Perhaps he is thinking about Canada alone. He doesn’t get around or read much, it would appear. Has he ever heard of Chicago? That’s what I love about the theoretical democracy about the internet and its blogs: so much
wrong data given as “fact” from “experts.”


Has he ever heard of Chicago?

yes. I’d add Washington and Chicago are almost tied up.

Anyway, North America has nothing to be proud of when compared to much smaller European cities: Berlin or Madrid’s rapid transit networks put any North American network to shame.

(NYC Subway would be a bit larger, but for a much bigger population and with crime and rat problems, unheard of in Europe).

Jonathan, your article is a good overview of the problems of the US rolling stock industry, but it also contains the main argument for why Detroit can’t retool itself to make trains. The global subway industry, according to your article, produces $9.3 billion’s worth of subway trains per year. By the standards of the auto industry, it’s nothing. Toyota alone has annual revenues of $263 billion. The problem is that with subways, most of the cost is in infrastructure, construction, and operations, which are done locally, rather than in vehicle production.

By the way, in terms of metro area transit mode shares, Toronto is the second largest city in North America. Washington, which is second in the US, ranks seventh, behind New York, Toronto, Montreal, Ottawa, Vancouver, and Calgary. I’m not sure whether Chicago has a larger overall ridership than Toronto or not – Chicago has nearly twice the population of Toronto whereas Toronto has nearly twice the mode share.

I say yes in that when you look at athe Pennsyvinia Railroad’s steel catenary mast system all that metal was made and refined in the US. If we were to start extending this catenary system but have a law that only US Steel can be used in it then that would most likely need hunderds of people making new catenary masts in US steel mills.

I don’t think these rail car projects are going to make that many jobs do to right now. Such as why should they build them here when they can import them from France or say Japan along with China. If we are copping the high speed rail plans from Europe and China then it is very easy to simply import the parts we need from those places to built a high speed rail copy of the high speed rail systems in China or Europe. It would be easer for them to import these rail cars from existing factories outside of the county then to come in and open up new factories which you have to build.

Thanks for the comments. It is not just the subway market. There are multiple markets that include: light rail, commuter rail, high speed rail (domestic and foreign). The global market for rail is multiple billions. Then, there is the question of leveraging into other related markets, beyond transit but related to them.

Those other markets you mention are smaller than subways, and much smaller in the case of HSR. For example: the entire Shinkansen fleet would be worth about $10 billion, new – and trains are only replaced once every 20 years. The TGV fleet would be worth $15 billion, replaced once every 40 years. As a result, the top trainmaker, Bombardier Transportation, has about an order of magnitude fewer workers and less revenue than any of the top automakers.

I appreciate Mr. Levy’s comments, but I am really addressing another problem than the one posed. I weighed in on this because it is somewhat related to what I see as the real problem. The real issue here is not the absolute size of mass transit markets for high speed rail. I think we agree there. The idea is mass transit as a launch market. Mass transit production in and of itself is not going to work. Key producers are in mutliple markets (light rail, subways, high speed rail), so discussing high speed rail in a vacuum is kind of a strawman argument. That is not how the firms (nor I) see it. Furthermore, some auto manufacturers are in high speed and subway production or aerospace and rail production. So, even focusing just on mass transit markets is wrong. The South Koreans and Spanish have enough to sustain themselves and gain market share in the U.S. Why can’t the U.S.? This increased share would have saved some now dead auto suppliers. All this does not mitigate the larger auto market which might have to be replaced gradually by mass transit to meet climate change needs, to promote energy efficiencies, to overcome energy supply shortages, etc. As for the political will, that is another story that I will address elsewhere.

Leave a Reply