Buy America is a Scam

Streetsblog’s interview with Amalgamated Transit Union President Larry Henley hits on the normal points regarding labor issues and transit, but one bit there deserves additional followup, regarding Buy America provisions:

Tanya Snyder: Some transit advocates are also critical of things like Buy America provisions because it costs transit agencies more money.

Larry Henley: This is the Wal-Mart question. This is whether or not we have a country at all anymore.

If the goal is to race to the bottom, to get the cheapest products, which means the cheapest labor, then we ought to be mindful that while we’re preserving the fiscal integrity of the MTA, we’re ruining the lives of American kids. We’re making it impossible for them to get a job. And if you look at the unemployment rates today, as staggering as they sound, it’s 9 percent overall, but for college educated kids it’s 4 percent. Which means that people who lack a college education no longer have a future in America. They just don’t.

…So that now, we have people in China and India and all across the world competing with American kids.

…This is about a moral crisis in America. And then they have the gall to come back and make all these arguments about American people being inefficient or American people not working hard enough and why shouldn’t they all be part time. But the central issue is that we have allowed corporations like Wal-Mart to wring every ounce of hope out of young Americans’ lives.

In the comments, Stephen Smith already justly mocked Henley for complaining about China and India when the major rolling stock and bus vendors are from peer developed countries, and Buy America’s most recent derailing of a light rail order was about imports from Spain, a country with 21% unemployment. But there’s much more at stake here.

Buy America’s purported role is to create American jobs. But let’s examine the costs. Amtrak’s Sprinter locomotives, compliant with both FRA regulations and Buy America, cost 30% more than the European locomotives they’re based on, and 50% more than competitor products built only for passenger trains rather than also for freight trains. A 30% premium works out to an extra cost of about $100 million, providing 250 jobs. Since the income earned by skilled workers is normally around $100,000 or less rather than $400,000, we can conclude most of the premium doesn’t go to workers. Or, for an even more egregious example, but without job numbers specified, look to SMART’s DMUs, at twice the cost of comparable European trains.

In other words, it’s a scam. Blocking parallel imports ensures only a select number of vendors can bid, driving up prices. Usually there’s a small sop to American labor, well-publicized in the media with photo-ops of people in hard hats – e.g. the 250 jobs heralded for the Sprinter order – but the bulk of the extra money goes elsewhere. It creates makework for consultants and lobbyists. It increases vendor profits, since fewer companies, typically the largest and most global ones, can bid. (This also goes for regulations: Caltrain applied for its FRA waiver in consultation with the biggest train manufacturers, potentially locking out Stadler and other small up-and-comers.)

When the number of vendors is very small, the result can be not just high cost, but also shoddy work. The reason the US has no legacy domestic rolling stock vendors is that two of the few that remained by the 1970s, protected by Buy America but servicing an ever-shrinking market, sold New York City Transit defective trains, the R44 and R46 orders; this was one of many mishaps facing the city in the 1970s. The subsequent lawsuits bankrupted the vendors. The R44 is still a lemon, though since refurbishment the R46 has performed well. In the 1980s, NYCT switched to global vendors instead; the next order, the R62, was not federally funded due to Reagan’s cuts, so NYCT went ahead and imported trains from Kobe, which worked fine.

There is another way, but, as with most other issues facing transportation, it requires importing ideas from other developed countries. The idea in question is that parallel imports are not a bad thing, either for the economy or for workers. The US and Canada import cars from each other; neither is any worse for it. To a much smaller extent due to trade barriers and different sets of regulations, North America imports cars from Europe and Japan – and the attempts to fight it have not resulted in a union revival, but in the proliferation of non-union plants in low-wage states.

Parallel imports are not an anti-worker or anti-union tactic. The Swiss Socialist Party is for them, and, far from a neo-liberal sop, it also supports linking trade to human rights and workers’ rights and has a general roster of policy positions that most Daily Kos contributors would love to see the Democratic Party endorse.

The majority of trade is within the developed world. To the extent Buy America is supposed to protect American workers from low-wage countries, it has failed; NYCT’s Buy America-compliant R160 trains were partially manufactured in Brazil to save money. The main function of Buy America is to protect companies that do business in the US from competition, period. At that it has done a very good job; it’s just not good for the public, which has to pay for it.


  1. ant6n

    “…but the bulk of the extra money goes elsewhere. It creates makework for consultants and lobbyists.”
    Well, those are American jobs, too. And it seems these are becoming America’s most effective and proud professions.

    • Alon Levy

      The lobbyists, sure, but I’m not so sure the consultants are American. They certainly don’t work for American companies; they work for Siemens and are involved in changing the specs, and could be based either in the US or in Germany.

      • ant6n

        In all seriousness, it would be nice if North America had better access to non-American rolling stock.
        It would even be better if a competitive manufacturing industry for modern and light equipment started here.
        – how would you propose those goals could be achieved?
        – don’t the ‘Buy America’ rules at least move some know-how back here, even if inefficiently so? (“United Streetcar” – comes to mind)

        • Alon Levy

          I’d propose encouraging railroads and transit agencies to develop in-house expertise on this matter. NYCT is a good example: it writes its own specs for equipment, which are very conservative due to the failure of the more futuristic R46, and when it needs to outsource design, it does so in an orderly fashion with a competitive bid (i.e. the R110 order) and uses the results to design future cars. An even better class of examples is the JR companies. The federal government could just tell Amtrak to do it (and earmark funding, which would be small by the standards of present Amtrak subsidies), but other agencies it would need to coach via some grant process. Private-sector rolling stock vendors aren’t necessary at the beginning; the big money, in engineering and design, would appear anyway. But eventually, employees would leave and start their own companies, creating a domestic industry that produces better products than the Colorado Railcar embarrassment.

          • JD12

            There are downsides to having each agency develop their own equipment specifications in-house. Too often, we see agencies fail to take advantage of opportunities to work together and achieve economies of scale. A recent example of this was seeing New York’s PATH and Boston’s MBTA separately designing and sourcing rapid transit cars that were essentially identical (the Boston cars replaced an earlier fleet that itself was based on a PATH design). In the case of New York’s MTA, engineers continue to forego modern lighter-weight body and truck designs and purchase overbuilt, heavy cars (the R160 car is actually heavier than the riveted steel R1 car that the Independent Subway purchased in the late 1920’s).

          • anonymouse

            The problem is that the trend has been away from in-house expertise and toward outsourcing of all kinds, both in the private sector and in the name of “privatization” in the public sector. This seems like a largely consultant- and lawyer-driven trend, because it creates more interface between companies, which means more friction and more contracts, which drives up the demand for consultants and lawyers. And of course the consultants would recommend a course of action that increases demand for their services, it’s the only rational thing for them to do.

          • Andre Lot

            In-house expertise faces an outright economy of scale problem. Maybe, only maybe, MTA has the size necessary to have a full in-house, top-notch, state-of-the-art R&D staff. Smaller agencies simple couldn’t cope with the enormous overhead it implies. Maybe BART can, Metro (LA) as well, but Marta, for instance – not so sure, let alone local authorities that manage smaller systems in Denver, Dallas, Salt Lake City…

            Even if an agency has scale to “in-source” design of railroad stock, it is not necessarily a good move. Among the pitfalls, I list:

            i) the development and design cycle varies a lot in terms of workload, but orders are not placed everyday. Therefore, there will be either excessive workforce that remains idle on “valley” months and “busy” on peak months concerning each order cycle; or insufficient workforce, meaning delayed OR low(er) quality design processes. Outsourced expertise mean specialized companies in design having clients all the time, so the manpower in design is not a constraint in the rolling stock procurement cycle.

            ii) it makes more difficult for best-practices in industry to be incorporated in the rolling stock, because it creates a tight-knight, insulated department of design very protective of their own creations and hermetic to competition (there is none) and new ideas. Related example: freight railways and their different signaling and train control systems or, if you want to look at Europe, the mess of international traffic caused by the array of “company based” (meaning usually nationally wide) solutions to communication, control etc.

            iii) transfer of the manufacturing risks, or part of it, to the company, which is usually a public authority. If a design flaw is uncovered from an outsourced project, litigation can recoup millions. What if it is the fault of an internal employee? Ok, you can fire him, but that is it. It is always good to have a private, external element that public companies can sue in case things go wrong, and the design of rolling stock is a good case of it.

          • EngineerScotty

            TriMet (along with the city of Portland, who runs the Streetcar and are actively involved in promoting United Streetcar) actually has a fair bit of in-house expertise on rail. And said expertise, along with the in-house planning staff, is a frequent target of critics who think it ought to be jettisoned, with the funds being redeployed (depending on who is doing the criticizing) to increasing bus service, dealing with pension obligations, or reducing taxes.

        • Wad

          America could regain its manufacturing expertise through what is called import replacement. It’s a slow process, but at least it is organic economic growth.

          First would be the import of a durable good (a bus or train). One of the first steps in the substitution process is to train a local workforce in maintenance. The next step would be to find domestic sources for parts, first from commodity sources and then to more specialized components. The third would be to establish a secondary market for used vehicles, while the original buyer would want new vehicles from the manufacturer. Once a robust secondary market develops, they would then consider buying vehicles new once they gain capabilities and money.

          After all this, it becomes feasible for an import to be made domestically.

          However, this process may take a generation or more. Politicians think in 2-, 4- or 6-year time frames. They’ll hurry the process through Buy America and thereby give up loaves for a slice of bread.

        • Wad

          @ant6n, to answer your question of whether Buy America rules produce some know-how domestically, the short and simple answer is NO.

          I describe Buy America as a policy in which the government trades away several loaves of bread to get back a slice.

          One of the big trade-offs is that in exchange for a “policy factory” — one that is created solely to comply with Buy America laws but otherwise serves no market advantage — Buy America gives up a broader organic economy.

          For one thing, the manufacturer retains the right for trade secrets and closed-loop sourcing. Just as Buy America locks out upstart manufacturers of the final product, it also locks out smaller manufacturers of sourced materials.

          The manufacturer will want to go with what it knows. In most Western industrial countries, suppliers are so robust that a Buy Domestic law would have no bearing on their overall bottom line since they have so many other markets to serve.

          A foreign manufacturer can just turn to GE, Detroit Diesel, Cummins, PPG, Dow, 3M, United Technologies, Goodyear et al. for relatively common components. The foreign manufacturer will be sure to retain rights to high-value, high-expense and specialized components — the very things Americans need to get their hands on and minds around.

          Buy America laws trade away these economic factors for a final assembly plant, which is essentially solving a math problem backwards by writing the answer and then figuring out how to show the work.

          • ant6n

            Interesting. You make it seem that even buying second hand equipment from Europe would be better at moving expertise to the US than ‘Buy America’.

          • Wad

            @ant6n, it would. Refurbished used train sets would save a lot of money, but the U.S. could afford to buy stock units from Europe built new.

            There would have to be a large enough order for a domestic factory. It would make sense if the U.S. order overwhelms existing capacity, but even with a few trainsets the U.S. would still gain maintenance and sourcing expertise.

  2. EngineerScotty

    The biggest problem with Buy America–or one of the biggest IMHO–is that it only applies to transit. Road projects, as well as transportation projects in other modes, aren’t subject to its requirements, just transit. While I doubt that hobbling transit was the intent, I’m certain that the road, auto, and petro industries don’t mind at all that the “competition” is hindered by such protectionist legislation.

    At any rate, we’re not going to have meaningful pro-labor legislation, and arrest the race to the bottom, until public-sector labor and private-sector labor are on the same page. As long as labor unions engage in the tactic of seeking carve-outs for their own members, rather than advancing the broader cause of workers’ rights, including (and especially) those who aren’t represented (and some who may never be), the broader working class will continue to view them as part of the problem, not part of the solution.

    • betamagellan

      While it’s not in the transportation sector, there’s also been a sustained push for forcing Buy American regulations onto any renewable energy as well (I don’t think this ultimately went through–the last references I found were from March 2010).

      • betamagellan

        Make that “any renewable energy receiving tax credits or government subsidies.” Sigh.

    • Andre Lot

      Road projects differ from most transit project in regard of that they don’t include the purchase of the vehicles that are going to run over the roads. However, many road projects are bounded to provisions requiring them to buy American-manufactured steel and other components.

      However, the bulk of road project costs are earth movements, concrete, pavement and bulk construction, none of which is outsource-able. It is just insanely expensive to, for instance, manufacture bridge columns in Bangladesh and ship them over to US.

      • Wad

        @Andre, California has ordered sections of the Bay Bridge replacement that are being fabricated in the manner you describe, but from China. Supposedly, there was not a single U.S. company that had the capability of doing this!

        However, it’s more than just insanely expensive in the case of a bridge. You can get tilt-ups or pre-fabs for commonly built structures, but for rare projects like bridges, there are a lot more site-specific concerns to be factored in, and you don’t get the cost savings from uniform construction and knowledge trees.

    • Danny

      Excellently put. I was in a union plant a while ago, in a right to work state. The management was surprisingly friendly to the union…it was the younger workers that were most vehemently opposed to it. They were the ones that had to experience layoffs every year, work in the worst conditions and the worst hours…regardless of how much better they were at their job than the old union elite.

      When unions aren’t busy alienating their own future generations, they do a pretty good job of destroying the competitiveness of the company they work for…essentially killing their own best interests.

      Unions don’t have to be this way, but so many of them are managed in the US makes them their own worst enemy. I can think of a few exceptions (like a local welders unions that enforces quality as a criteria for membership, thereby adding value to the companies they are employed by), but they are just exceptions.

      • Nathanael

        Seniority-based systems turn out to be simply nasty, and this is at the root of most of the problems-with-unions which you describe. Seniority is no basis for anything and should not be used as one.

        • Paulus Magnus

          Eh, seniority makes a good deal of sense I think when it comes to large organizations (especially as favoritism can easily be the alternative), but it may perhaps be overused. Certainly it shouldn’t override measurable and metrics based competence and performance.

  3. Stephen Smith

    The biggest problem with Buy America–or one of the biggest IMHO–is that it only applies to transit.

    Bingo. There would be rioting in the streets if the US government tried to stop Americans from being foreign cars.

    …the broader working class will continue to view them as part of the problem, not part of the solution.

    I think you put way too much faith in “the broader working class” to see their interests clearly here.

    • EngineerScotty

      I think you put way too much faith in “the broader working class” to see their interests clearly here.

      And here you have a point–much of the “broader working class” seems more outraged about the black guy in the White House than about the real reasons for their economic misery.

      But still–currently, the plutocracy can (and does) deflect blame for the present state of things to public employees, as though allegedly outsized compensation for bus drivers and teachers (and work rules in some agencies which amount to featherbedding) is the biggest problem which bedevils the private-sector working class.

      At any rate, today’s recall races in Wisconsin will be telling, on many fronts.

    • Andre Lot

      However, there are no federal funds to match private expenses in the purchase of foreign cars. As Alon pointed out, transit projects not required federal funding are not subject to Buy America provisions.

  4. Stephen Smith

    What exactly do you mean by “parallel imports”? It looks like you’re using it to mean imports from similarly-developed countries, but I always thought it was a term related to gray market goods/arbitrage trade…

    • Alon Levy

      It generally means imports bypassing price discrimination (e.g. in Australia, DVDs are legally mandated to be region-free). But it also means imports bypassing protectionist regulations, for example the special crash safety laws that the US government requires for cars that make it impossible to import off-the-shelf European cars. I think it’s a valid term in light of the way FRA regulations are used to shore up protectionism.

      • EngineerScotty

        As an aside–do you consider US safety and emissions standards to be protectionist, assuming they apply equally to domestic and foreign-produced automobiles? Obviously, when and where different standards apply depending on the source, that is protectionism; and I can see cases where supposedly-neutral regulations are drafted which only exist to cause friction and discourage foreign suppliers–but I’m not sure that is the case with US automobile crashworthiness standards. (Whether they represent best practice is another matter). Putting the shoe on the other foot, do the various hurdles (both technical and regulatory) that might discourage someone from importing a stock Ford F350 into Europe constitute protectionism?

        I’m playing devil’s advocate here… but it it your position that US auto regs are so flawed that they cannot be justified on any rational basis, and therefore constitute a protectionist measure?

        • Alon Levy

          The regulations’ basic problem is the same as with the FRA’s: they assume buff strength equals safety. We know that American cars aren’t especially safe; the Smeed curve of the US has been flatter than that of the UK, the other country studied by John Adams, and the UK’s per-passenger-km accident rate (which is admittedly low by any standard) is lower than the US’s. They’re also not friendly to pedestrians: the regulations prioritize safety for car occupants but neglect safety for pedestrians who are run over.

          Europe and Japan are every bit as protectionist as the US, but at least there isn’t some unmet demand for imports from the US. Americans have demanded and gotten somewhat modified versions of VWs and Toyotas; in contrast, Europeans have no use for American cars, except for a few models geared specifically for the European market, such as the Ford Focus. At any rate, best practice would be for US, EU, and Japanese regulatory agencies to actually talk to one another and come up with unified standards; if they invited agencies from developing countries, then all the better.

          • Danny

            That would go for much more than automotive and rail safety standards. Medical Drug regulation could also benefit…being more robust and less costly…if they could pick one unified standard.

      • Andre Lot

        Alon, some agencies in some sectors might play that way, but I don’t think FRA developed its policy towards crash worthiness with an specific objective of protecting American manufacturers. It represents a whole different philosophy which could be argued for and against on many level, but that it makes more difficult to import European train cars is a mere unintended consequence. To prove it by contradiction: the safety culture surrounding this approach was born when US had a large rail car manufacturing industry.

        • Alon Levy

          Andre, to clarify, I do not think there’s a deliberate intention to lock out foreign competition. I agree with you that it’s an unintended consequence. The reason I’m ascribing some degree of malice to the process is that once the different regulatory culture is in place, the vendors that are used to it lobby against changes that expose them to more competition. To take the example of the Caltrain waiver, Alstom, Bombardier, and Siemens’ interests are served by a narrow FRA waiver that locks out Stadler and CAF; more broadly, if FRA reform becomes a serious possibility, you can bet US Railcar will lobby against it and employ all the usual FUD used by defenders of the status quo.

  5. Nathaniel M Hood

    Great post! This is another example of how transit in America starts with a deficit versus its competition (the automobile). Add in this, and the much longer and arduous planning, alternative analysis and approval process; and that is usually before it needs to fight for limited transit dollars – whereas it feels like they just blindly hand-out road construction dollars.

  6. Pingback: Free Trade In Trains | 1FENS
  7. Pingback: The costs of a ‘Buy Canada’ policy on the LRT | Ottawa Citizen Blogs
  8. Paulus Magnus

    The particularly stupid bit about it is that companies will gladly put manufacturing in America or even individual states if the order is large enough, it’s just the small orders that get nailed. Take the occasional SCAG desire to electrify Southern California’s railroads. That’s 720+ locomotives that they plan on purchasing in such a project. No matter who won the contract, most of the locomotives would be built in America and likely CA regardless of any regulations requiring it.

    Actually, doing one or more large purchases like that might be exactly what is needed to essentially obsolete Buy America and bring much of the know how and capability to America.

  9. The Overhead Wire

    I don’t want to defend Buy America because that Houston deal was stupid. But the extra $100 million, it can’t all be going to the producer and those 250 jobs. Isn’t that just the added up value of all the components that have to be created in America, each one a bit more expensive than the other versions one could possibly get? Yes Siemens added 250 jobs, but there were probably a lot of other jobs added up throughout the whole economy because the Buy America provision means that you have to have all of your components made in America as well, all of them being a little more expensive. Am I missing something here or is this some fuzzy math on the $100M divided by only 250 jobs that Siemens discussed?

    • Alon Levy

      I doubt that the components are sourced in the US. Buy America has loopholes; the R160 complied with it even though the car shells were made in Brazil.

        • Stephen Smith

          It’s easy to say “sixty percent” on paper, but actually determining what counts and what doesn’t is a bit more difficult. I’ve heard that manufacturers can play a lot of tricks to play with the domestically manufactured content requirements, though I don’t remember where I heard that.

      • david vartanoff

        the Buy America component counts both parts and labor. The stainless steel shells from Brazil (likely the old Mafersa plant, a Budd licensee) are fairly common including the California Cars for Amtrak operated corridor service. This is not very different from GM practice in the early 6os when body panel stampings for Opels were manufactured in the US and shipped to Germany for assembly. Apparently the long lead times and ocean shipping costs were lower thanmaking those parts in Europe.

      • david vartanoff

        Not loopholes, merely the un required content. OTOH NABI buses(Ikarus of Hungary in Alabama) w/the diesel engines from Cummins of Indiana, Van Hool buses built in Belgium also Cummins engines. Not very different from Daimler producing components in Europe for the Chrysler PT Cruiser or GE diesels for MN and Amtrak w/ europeam sourced components assembled in Erie PA.

  10. david vartanoff

    I am not clear that insisting 52 % of the contract price be spent stateside is inherently wrong. The misguided FRA buff strength requirements can be done anywhere–the first few Hyundai-Rotem Silverliner Vs for SEPTA were made in Korea with the balance of the cars to be finished in Philadelphia. As to in house v consultants v vendor designs, the agency specific designs are a huge mistake. As pointed out by others, a PATH design could be used on both Orange and Blue lines in Boston. CDoT and SEPYA should have co designed the new cars insisting on a better price per car.

  11. anonymouse

    One thing about in-house designs versus vendor designs is that if in-house expertise is insufficient, there’s a possibility of making stupid specs, and there’s always the possibility of getting something special and incompatible with the rest of the world. On thevother hand, if you let the vendors do the design, you may end up with something incompatible with the rest of your operation, or something that the vendor happens to have that doesn’t really match your needs. As an example of the latter, consider the epic disaster of the SelTrac deployments on the Jubilee Line and especially the Muni Metro. The problem is that there’s no one right answer and choosing the right answer for any given situation is hard.

  12. John Alexander Thacker

    Politically speaking, there’s simply no coalition for getting rid of Buy American. Most Republicans voted against Buy American– but they’d vote against the Amtrak funding too. The Democrats who favor Amtrak funding almost all vote for Buy American.

    The number of people who are against Amtrak funding or subsidies if it has Buy American but favor it without are very small; the number of people who would oppose the funding if it didn’t have the subsidies are larger. Even most pro-transit folks (like our bloghost) who realize that the rules are stupid will (however reluctantly) favor funding even if it has these subsidies, generally justifying to themselves that it somehow “balances out” other subsidies to other transportation modes. Anti-transit subsidy folks will oppose it, Buy American or no.

    Therefore, expect these rules to persist. Any time Amtrak gets a lot of money, it will have these rules.

    • Nikko P

      Well keep in mind that a transit agency doesn’t have to abide by Buy America if it doesn’t receive federal funding. Theoretically, I guess it would be possible to seperate a new project into two smaller projects: one for right of way and construction and one for acquiring new vehicles. The “vehicle project” could use exclusively local and state funding to avoid the Buy America requirement.

      • Nikko P

        But obviously nothing could help Amtrak (except state-funded corridors) since it’s a federal company.

  13. Pingback: Free Trade In Trains | ThinkProgress « mawillits
  14. Pingback: Betting the Farm on Oregon Iron Works

Leave a Reply to betamagellan Cancel reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

This site uses Akismet to reduce spam. Learn how your comment data is processed.