US President Donald Trump and his Mexican counterpart, Enrique Peña Nieto, recently announced resolution of major sticking points that have held up the overall renegotiation of the NAFTA treaty (or whatever new name Trump confers on the expected trilateral agreement).


At first glance, there are some marginal improvements on the existing treaty, especially in terms of higher local content sourcing, and the theoretic redirection of more “high wage” jobs back to the US. We present you a report in this regard titled: “Trump’s NAFTA deal can’t solve America’s manufacturing problems” which appeared in Asia Times.

These benefits are more apparent than real. The new and improved North American Free Trade Agreement won’t mean much, even if Canada ultimately signs on. The deal represents reshuffling a few deck chairs on the Titanic, which constitutes American manufacturing in the 21st century: a sector that has been decimated by policies of globalization and offshoring.

Additionally, what has remained onshore is now affected adversely to an increasing degree by the Pentagon. The experience of companies that have become largely reliant on military-based demand is that they gradually lose the ability to compete in global markets.

As early as the 1980s, this insight was presciently confirmed by the late scholar Seymour Melman. He was one of the first to state the perhaps not-so-obvious fact that the huge amount of Department of Defense (DoD) research and development pumped into the US economy has actually stifled American civilian industry innovation and competitiveness, most notably in the very manufacturing sector that Trump is seeking to revitalize with these “reformed” trade deals.

The three biggest reasons are:

  1. The huge diversion of national R&D investment into grossly overpriced and mostly unjustifiable DoD programs has tremendously misallocated a large proportion America’s finest engineering talent toward unproductive pursuits (for example the tactical-fighter fiascos, such as the F-35 Joint Strike Fighter, which among myriad other deficiencies cannot fly within 40 kilometers of a thunderstorm; and producing legacy systems that reflect outdated Cold War defense programs to deal with a massive national power, as opposed to combatting 21st-century terrorist counterinsurgencies). Indicative of this waste, former congressional aide Mike Lofgren quotes a University of Massachusetts study illustrating that comparable civilian expenditures “would produce anywhere from 35% to 138% more jobs than spending the same amount on DoD [projects].” The NAFTA reforms won’t change any of that.
  2. By extension, the wasteful, cost-is-irrelevant habits of mind inculcated into otherwise competent engineers by lavish DoD cost-plus contracting have ruined these engineers for innovation in competitive, cost-is-crucial civilian industries.
  3. The ludicrously bureaucratized management systems (systems analysis, systems engineering, five-year planning and on and on through a forest of acronyms) that the DoD has so heavily propagandized and forced on contractors has, in symbiosis with the Harvard Business School/Wall Street mega-corporate managerial mindset, thoroughly wrecked efficient management of most sectors of American industry.

Let’s drill down to the details of the pact, notably automobiles, which have made up a big part of NAFTA. Under the new deal, 25% of auto content can be produced elsewhere than North America, a reduction from 37.5% that could be produced outside before, because of the multinational nature of every major automobile manufacturer. Twenty-five percent is still a very large proportion of the high-end auto content, much of which is already manufactured in Europe – especially expensive parts like engines and transmissions, especially for non-US manufacturers, which won’t be much affected by this deal.

Additionally, much of the non-North American auto content that can be or is being manufactured in Europe is the high end of the value-added chain. Certainly the workers producing the engines and transmissions have wage rates higher than US$16 per hour, which are trumpeted in the new agreement as proof that more “good jobs for working people” are being re-established by virtue of this deal. Since when is $16 per hour a Trumpian boon for US auto workers?

Objectively, $16 is only 27% above the 2018 Federal Poverty Threshold for a family with two kids; even worse, $16 is only 54% of today’s actual average hourly pay ($29.60) in US automobile manufacturing, according to 2018 Bureau of Labor Statistics numbers.

But beyond cars, here’s the real problem: Although the ostensible goal of all of Trump’s trade negotiations is to revitalize American manufacturing, the truth is that US manufacturing basically suffered a catastrophic setback when China entered the World Trade Organization (WTO) back in 2001. Along with liberalized capital flows, the extensive resort to “offshoring” of manufacturing to China has sapped US manufacturing capabilities, as well as engendering a skills shortage. This includes (to quote a recent Harvard Business Review study co-authored by Professors Gary Pisano and Willy Shih):

“[the] tool and die makers, maintenance technicians, operators capable of working with highly sophisticated computer-controlled equipment, skilled welders, and even production engineers [all of whom] are in short supply.

“The reasons for such shortages are easy to understand. As manufacturing plants closed or scaled back, many people in those occupations moved on to other things or retired. Seeing fewer job prospects down the road, young people opted for other careers. And many community and vocational schools, starved of students, scaled back their technical programs.”

The one ready source of demand for US-manufactured goods is the military. High-tech enthusiasts like to claim that the US Defense

Department had a crucial role in creating Silicon Valley. The truth is far more nuanced. Silicon Valley grew like a weed with in essence no DoD investment; in fact, until quite recently, most successful Silicon Valley enterprises avoided DoD contracts like the plague.

A bit of history: The transistor was invented by entirely privately funded research at Bell Labs in 1947. Next, the first integrated circuit patent was filed by Werner Jacobi, a Siemens commercial engineer, in 1949 for application to hearing aids.

The next advance, the idea of a silicon substrate, was patented in 1952 as a cheaper way of using transistors by Geoffrey Dummer, a reliability engineer working at a British government radar lab; for all of the talk of the defense establishment’s role in developing high tech, ironically the British military showed no interest and Dummer couldn’t secure the funding support to produce a working prototype.

In 1958, Jack Kilby, a newbie at Texas Instruments (which had developed the first transistor radio as a commercial product in 1954), came up with the idea of multiple transistors on a germanium substrate. Almost simultaneously, in 1960, Robert Noyce at Fairchild Electronics patented a cheaper solution, a new approach to the silicon substrate, and implemented working prototypes. Both men envisaged mainly civilian applications (Kilby soon designed the first integrated-circuit pocket calculator, a commercial success).

It is true that the first customers for both the Noyce and Kilby chips were the US Air Force’s B-70 and Minuteman I projects, which gave rise to the idea that the Pentagon played the key role in developing US high-tech manufacturing, although it is worth noting that the electronics on both projects proved to be failures. Both companies soon developed major commercial applications for their integrated-circuit innovations, Texas Instruments with considerably more success than Fairchild.

The Defense Advanced Research Projects Agency (DARPA) is generally credited with inventing the Internet. That’s overblown. In fact, civilian computer-science labs in the US, the UK and France developed the idea of wide area networks in the 1950s.

In the early 1960s, DARPA started funding ARPANET (Advanced Research Projects Agency Network) design concepts to connect DoD laboratories and research facilities, initially without the idea of packet switching. Donald Davies at the UK’s National Physics Lab first demonstrated practical packet switching in 1967 and had a full inter-lab network going by 1969.

The first two nodes of the ARPANET were demonstrated in 1969 using a primitive centralized architecture and the Davies packet switching approach. In 1973, DARPA’s Vint Cerf and Bob Kahn borrowed the idea of decentralized nodes from the French CYCLADES networking system, but this wasn’t fully implemented as the TCP/IP protocol on the ARPANET until 1983.

In 1985, the civilian US National Science Foundation started funding the NSFNET, based on the ARPANET’s TCP/IP protocol, for a much larger network of universities, supercomputer labs and research facilities. NSFNET started operations with a much larger backbone than ARPANET in 1986 (ARPANET itself was decommissioned in 1990) and started accepting limited commercial service providers in 1988, and, with further expansion and much-needed protocol upgrades, NSFNET morphed into the Internet in 1995, at which time the NSFNET backbone was decommissioned.

Today, of course, the DoD is showering the largest Silicon Valley companies with multibillions of dollars and begging them to help the US military out of the hopeless mess it has made of its metastasizing computer, communications and software systems. Needless to say, if this DoD money becomes a significant portion of the income stream of Google, Microsoft, Apple etc, it is safe to predict their decay and destruction at the hands of new innovators unencumbered by DoD funding, much as occurred with aviation companies such as Lockheed. NAFTA’s reforms won’t change that reality.

As a result of the militarization of what’s left of US manufacturing, along with the enlargement of the trans-Pacific supply chains with China (brought about through decades of offshoring), a mere tweak of the “new NAFTA” is unlikely to achieve Trump’s objective of revitalizing America’s industrial commons. With China’s entry to the WTO, it is possible that the US manufacturing has hit a “point of no return,” which mitigates the utility of regional trade deals, as a means of reorienting the multinational production networks in a way that produces high-quality, high-paying jobs for American workers.

Offshoring and the increasing militarization of the US economy, then, have rendered reforms of the kind introduced by NAFTA almost moot. When it becomes more profitable to move factories overseas, or whenit becomes more profitable, more quickly, to focus on finance instead of manufacturing, then your average red-blooded capitalist is going to engage happily in the deconstruction of the manufacturing sector (most, if not all, outsourced factories are profitable, just not as profitable as they might be in China or, for that matter, Mexico).

The spoils in a market go to the fastest and strongest, and profits from finance, measured in nanoseconds, are gained more quickly than profits from factories, whose time frame is measured in years. Add to that the desire to weaken unions and the championing of an overvalued dollar by the financial industry, and you have a perfect storm of national decline and growing economic inequality.

Seen in this broader context, the new NAFTA-that’s-not-called-NAFTA is a nothing-burger. The much-trumpeted reforms give crumbs to US labor, in contrast to the clear-cut bonanza granted to corporate America under the Republican Party’s new tax “reform” legislation.

In fact, by the time the big-bucks corporate lawyer-lobbyists for other mega corporations have slipped in their little wording refinements exempting hundreds of billions of special-interest dollars, the new NAFTA-that’s-not-called-NAFTA will most likely include an equal screwing for textile, steel, energy sector, chemical, and other US industry workers, and anything else that is left of the civilian economy.



Sep 13, 2018 09:30 UTC