NAFTA Mock Talks

NAFTA Mock Talks, Round 1: Auto industry and rules of origin

Expert negotiators from each of the three countries have three sessions to hammer out a new NAFTA. Will America drive too hard a bargain on autos?

There is no greater economic issue facing Canada in 2018 than the renegotiation of the North American Free Trade Agreement, the pact that for 24 years defined this country’s economy, along with those of Mexico and the United States.

So, with the sixth round of talks under way—and a near palpable sense of pessimism in the air—Maclean’s set out to answer an essential question: Can three qualified, knowledgeable and reasonable negotiators reach an agreement to save NAFTA? Given their governments’ opening positions—but absent the racket of politics—can a trio of delegates representing their respective countries achieve what the official negotiators have so far failed to?

What would such a deal look like?

Okay, that’s three questions. But they’re vital ones. And to answer them, we’ve enlisted expert volunteers to participate in a shadow NAFTA negotiation that will unfold over the coming weeks, more or less in synch with the real ones. To state the obvious, this is an exercise in broad strokes: our delegates are tasked with tweaking the framework of NAFTA, addressing only the devilish details that stand in the way of that challenge, on five major subjects: auto and rules of origin, supply management and agriculture, conflict resolution (chapters 11 and 19), the sunset clause, and intellectual property and data. At the end of the complex three-round negotiation, they will have one choice: Deal or no deal.

To that end, we asked our representatives to focus on five high-friction areas: the auto sector, agriculture, dairy and supply management, dispute-resolution mechanisms and a proposed “sunset clause” at which point the parties must renew the deal or end it. But they face no restriction. They are free to introduce other issues if they believe it could lead to an agreement.Read curated play-by-plays of this round of negotiations on the five topics specifically here. Or read the transcript in full, here.

Now, meet our negotiators:

Christopher Sands (United States) is a Senior Research Professor and Director of the Center for Canadian Studies at Johns Hopkins University’s Paul H. Nitze School of Advanced International Studies (SAIS) in Washington, D.C. where he has previously worked for several think tanks, most notably the Center for Strategic and International Studies and the Hudson Institute. He is American, originally from Detroit, Michigan.

Dan Ciuriak (Canada) is a former Deputy Chief Economist at Global Affairs Canada, runs a consulting practice focussed on quantifying trade agreements, and holds fellowships with the C.D. Howe Institute, the Centre for International Governance Innovation, and the Asia Pacific Foundation of Canada.

Hugo Perezcano Diaz (Mexico) is the Deputy Director of International Economic Law with the International Law Research Program (ILRP), at the Centre for International Governance Innovation. Prior to joining CIGI, he was an attorney and international trade consultant in private practice. He worked for the Mexican government’s Secretariat of Economy for nearly 20 years, serving as head of the trade remedy authority, and formerly as general counsel for international trade negotiations. He was actively involved in the NAFTA and the Uruguay Round negotiations.

The negotiations began with general opening statements, which you can read here. They then progressed to discussions about the auto industry and the rules of origin, launching these negotiations with separate opening statements about their starting positions and asks on this topic in specific.


With regard to automotive, we feel strongly that we need to bring as much activity and value in North America as possible, but particularly in the United States. We would like to see a higher number than 62.5 per cent. We’re comfortable with having the same rule of origin for autos and other products, but remember that for products, not autos, we’re only at 50 per cent. The President believes that rule of origin represents a headline for the NAFTA agreement. And if it’s possible for us to come to an agreement on a higher rule of origin, and the next day The Washington Post announces that under new NAFTA, the rule of origin has gone from 50 per cent to 75 per cent, 72 per cent, then this is the clearest way to communicate to our supporters, our voters, that this is a better deal. In the same way Buy American has a tendency to communicate to people, we think that it cuts through a lot of the political nonsense that’s been spouted here over trade.

So let me unpack a little bit. Our proposal initially was for 85 per cent rule of origin, with a 50 per cent American reserve. So 50 per cent of the content had to be U.S. And we also proposed that tariff shifting, or the substantial transformation rule, be disallowed, and that tracing had to go all the way back to raw material. We feel that a lot of companies are suggesting that they meet the rule of origin, but if we scratch beneath the surface we find that they are actually not compliant, but that they’ve bundled things, either using one of the other rules and calling them local products or North American products when they don’t truly qualify. I agree with what Dan has said: that in some cases, at least in your opening statement, that in some cases, that the cheaters are not Mexico and Canada; the cheaters are our own firms who have tried to stretch the rules as far as they can. And those companies should be on notice, even if they are American headquartered, that their cheating is not acceptable either.

In order to get a higher number—70, 72, 75 per cent, we’ll see what we can get from you—we’re willing to give ground on substantial transformation and tracing. We’re interested in the headline number. The other pieces are counted in that. But I think that you can change how you calculate things in a way that lets you fairly and transparently meet that higher number.

And let me make a proposal with regard to autos, because it is such a unique sector. What we’ve seen in the United States, and I think Canada and Mexico share this, is that if you look at not just origin but value, an increasing amount of the value of what our car companies produce is in the R&D, in the engineering, in the marketing, in the technology. And our NAFTA rule of origin really counts the bits and bobs that go into making the car, with indirect cost receiving only sort of a cursory allowance. We’d like to see that allowance increased so that it reflects the value that we’re contributing. And yeah, that means white collar jobs in engineering—technology centres that are helping to push this new tech forward. Right now we don’t feel that’s adequately calculated in, and we want to reflect where we are really adding value.

We remember, and I’m sure you do too, the debate over semiconductors and things like VCRs. And the U.S. would take a similar position now: making the gear is nice, but if you own the intellectual content, the movie that the VCR plays or the software that the semiconductor runs, you have opened up on a lot of value, and it’s better to be on that idea side, which is more renewing and more profitable. We want to recognize the intangibles—indirect costs—more in our rule of origin. And we think that that will go a long way to achieving a higher number that we hope we can come to terms with you on.


The U.S. demands of 85/50 for rules of origin—that’s over the top. They’ve not been validated by industry. I don’t think that’s a viable starting point. I think NAFTA rules—the 62.5 per cent regional value content with the existing structure of the content rules, including tracing and so forth, is the starting point, now that the U.S. has pulled out of TPP. I think we throw the TPP stuff under the bus because that was essentially the U.S. and Japan’s position. So we have to go back to NAFTA rules of origin as the starting point, and we have to recognize that we’re very much in the hands of what are basically US multinational corporations in terms of what is viable for North America, and, in particular, in view of the fact that we need to keep the industry viable in competition with global firms.

The second thing on autos is that we have to recognize that the sector is undergoing fundamental transformation. The solution of a Fortress North America based upon 20th-century rules of origin—which put a primacy on things like steel and plastic—is inappropriate for the future. Cars are becoming technological instruments, with artificial intelligence and other kinds of technologies being the dominant form of value added, that is where we want to go.

Now, this actually provides a potential solution to the political optics of rebalancing and addressing specific concerns about how the auto industry has worked in North America, because Canada is in the same position as the United States in terms of having seen its share of the auto sector value-added flowing down to Mexico because of low wages. As I mentioned in my opening statement, that’s part of NAFTA’s problem: it did not support the wage and income gains in Mexico that we hoped for.

What I would suggest is potentially possible are to set up the rules of origin for Canada-U.S. and for U.S.-Mexico. We keep NAFTA as one deal, but we devise separate rules of origin which address the differences in the issues facing U.S.-Mexico and facing Canada and Mexico. But Canada’s particular position will be to ensure that our contribution on the technological front is safeguarded, and meanwhile the U.S. can put forth at least with a straight face – and without a red face – that it’s ensuring a very high U.S. content in the traditional portions of Mexican automotive assembly that then flow basically to the U.S. market.

That would be the position I take: that we address the autos issue as a rules of origin issue, and that we consider bifurcating the rules of origin.


To begin with, the U.S. position on the origin content is quite helpful, in the sense that it’s easier to recognize that 85 per cent domestic content may actually do more harm to the three parties’ auto industries than good. And I think it’s a creative proposal to consider other elements that have, I would agree with Chris, that other elements have been left out that do provide value but have not been considered simply because this has been the traditional way of looking at things in the rules of origin context.

Mexico would like to understand the U.S. position better, and I’m sure we’ll come to that. We’ll have time to discuss it in more detail. But I think that that provides a good starting point for our discussion. We do want to see more production in North America. We do want to see more business here in the auto sector, and other sectors as well. But again, we need to be careful not to put ourselves in a position that will make us less competitive. Where there were probably originally the big three U.S. companies, I think that we can safely now say that they are the three big NAFTA North American companies as they all do substantial business in North America, seeking other places from where to produce and not achieving any of the goals that each of us is seeking. That is a good starting position.

We’re also happy to hear that the U.S. may rethink its position in terms of substantial transformation, and especially tracing. Tracing is complicated. Autos are very complex goods, and it was not easy to come up to negotiate and to agree on the rules for tracing in the original NAFTA negotiations. And Mexico was quite concerned that, by making them even more complex and going all the way back to raw materials, it will probably defeat many of the purposes that Chris has already placed on the table. We do not want to see more complexity in terms of the rules of origin that will encourage firms to circumvent the rules, that will encourage more cheating as opposed to finding ways to prevent that cheating. But by adding more complexity, I think we will be encouraging just that.

So I think that the tracing provisions as they are, they are already complex, they are far different from how origin verification works in pretty much every other sector. And again, if Mexico would like to see anything, it would be simplification of that, rather than adding complication. I’ll leave it at that for the moment, but I do look forward to a more detailed explanation by Chris and his team on how they envisage reaching a higher content by bringing other elements into the equation.



The United States began by acknowledging that the demand that at least 85 per cent of a car must be made in North America (or regional value content, or RVC) under a new NAFTA was a bit of a high number, suggesting a number closer to 75. Canada agreed that 85 was too high, saying that it “risks undermining the competitiveness of North America,” as well as causing firms that gear their supply chain so heavily to North America costing themselves when it comes to participating in the key U.S. market.

Ciuriak, representing Canada, made note of this:

“Countries don’t trade; it’s firms that trade. And the firms will be in a position where they will have to actually move their production offshore entirely to be able to compete wit the firms that supply the U.S. market from abroad. And here, I would put on the table the Ford Focus issue. Ford had a dispute with the U.S. Department of Commerce over a treatment of its supply chain, and it basically resolved the deal by moving the production of the Ford Focus to China.

“So I think going for higher RVC, while interesting on a communications perspective, would have to be dealt with in a way which ultimately actually doesn’t undermine the competitiveness of firms. And I’m talking about U.S. firms, but certainly about Canadian firms exporting to the U.S. We don’t want to see our market share dribbled away to third countries as a result of this. It’s one thing to rebalance in favour of the U.S., but it’s simply not on the table for anyone—I think this goes for the U.S. as well—to have this dribble away to third countries.”

Perezcano Diaz, representing Mexico, jumped in to say that before the parties could talk about a compromise RVC number, they should first figure out whether they needed better mechanics to punish companies—including U.S. companies—who were circumventing these rules-of-origin requirements, or moving production offshore. He used the opportunity to push for less complexity and bureaucracy on tracing, one of Mexico’s biggest asks in these discussions. (Ciuriak would later offer his support to Mexico’s position on tracing, saying that “every time that we add costs to complying with NAFTA, we use up the margin that is provided by our existing external tariffs to keep things in North America.”)

That’s when Sands made a bold proposal: taking the auto industry out of NAFTA’s purview and operating it through a separate automotive customs union. In his own words:

I know this is out of the box, but I’m trying to be creative—we’ve always treated autos differently, whether it was during the Mexican Auto Decree System or the Canada-U.S. Auto Pact era, and maybe we should take autos out [of NAFTA] and, for the purpose of the auto sector, agree to a North American customs union for autos in which we all increase our MFN (most-favoured nation) tariff rate for autos to 25, maybe even 30 per cent. And this, I think, would provide the kind of backstop to a higher rule of origin that would force our companies to be creative—and I am open to counting things into the formula to make it easier for them. But we want that powerful symbol that these companies are North American and are willing to stay.

It wasn’t that long ago that we had to bail these American-headquartered companies out. And through the policy NAFTA put in place with 62.5 per cent rule of origin, we got companies like Toyota, Honda, and Nissan to make much bigger commitments into North America. So this has been a strategy that’s worked. I have limited sympathy for the Detroit guys. If they’re not willing to make a good effort, I’m happy to talk to them and ask them to bring it around. But I also think we need to set the parameters that bring them to the table and close off some of these avenues for offshoring that they’ve used in the past.

This produced a long pause. Perezcano Diaz was the first to throw cold water on the idea, warning that it would raise the price on cars for consumers. He also reminded the parties that Mexico had deployed a customs union around the sugar trade between the U.S. and Mexico which, in his view, “has maintained a very high price on sugar in the North American region that has benefited very few groups and group of companies.”

Ciuriak didn’t rule out taking autos out of NAFTA completely, but felt that a customs union idea for autos that raises tariffs would collide with Canadian and American agreements with the World Trade Organization, which would threaten its survival: “I can safely say that that would be a non-starter for Canada.”

Instead, he thought about how the rules of origin number could go higher if the United States was willing to give up some room on intellectual property, the intangible content that drives economic growth:

I think the far more promising avenue to pursue is that of reflecting intellectual property—R&D, and those headquarters services basically in the rules of origin. Now, here I would need to do some homework because this sounds an awful lot like what we do with the trade remedies, where we take into account the cost structure of (inaudible) and we assign fair market value to them, part of our rules of origin sort of valuation. But I think we do have the intellectual framework for that already in the trade remedy area, and deduct it in duties. So it’s something to think about.

Has anyone done any homework on this that would allow us to think about how we might actually realistically include the headquarters services) which includes R&D and the IP into the North American regional value content, that would also include, by the way, U.S. corporations parking all their IP abroad in Ireland or elsewhere, and us bringing that stuff back home, where it’s taxed and the income is taxed on that in the U.S., so that would be salutary from an American perspective directly, without actually affecting Canada or Mexico very much. So you would achieve your higher regional value content without actually undermining Canada or Mexico.

So I think this is a promising area to pursue. I’d like to see the United States actually put something on the table in terms of even a working paper setting out how this might work. Because I’m not aware of anything that’s been done so far on this score. I did think that the idea of (inaudible) with much higher external tariffs is a non-starter.

Sands was intrigued, especially given his desire to make it clear that it looks like America wins any deal: “I think this would be both important in terms of capturing important economic activity for North America and in dealing with the public relations issues of the headline numbers on value.” But Perezcano Diaz was wary about the trickiness of measuring the value of intangibles. It would also have an impact on enforcement of rules of origin, Perezcano Diaz noted:

We would not want companies to overvalue or overestimate the value of their research and development or IP in order to satisfy a particular percentage of RVC and actually lower the goods’ regional content value in the equation.

Sands said that while it would be very complicated, this challenge actually represented an opportunity:

…there’s room for us to include in some sort of formula the cost of labour input into production. I don’t want the indirect labour cost, the executives who are getting paid six-figure salaries. I’m much more interested in the workers. Because if we in fact recognize the value of the workers’ contribution in a monetary way as a contribution towards origin, you provide an incentive or, in some ways, a fairer response to those companies that continue to employ highly skilled labour in the production of vehicles. And if it was the case that there was a Mexican supplier or firm that was underpaying its employees, this would provide an incentive for them, in order to meet origin calculations more fully, to pay their workers better.

I feel a little bit more comfortable with those sorts of incentives rather than “social responsibility,” at least as far as I interpret that as being somewhat voluntary. I’d like something that was a bit more reliable and serious and provided a stronger incentive for the behaviour.

Indeed, Ciuriak flagged that Mexico’s productivity in the auto sector has been slow, and that having some reflection of corporate social responsibility or productivity wages in the value content of production might be a way to help.

Part of the reason why companies have been shifting production, possibly to an excessive degree, down to Mexico, while Mexico itself has not been benefiting because in fact, while it’s getting the production, it’s not getting the wages, and therefore not getting the demand, and we’re not getting this kind of the knock-on macroeconomic impact on our own economies of rising Mexican demand. It’s being pocketed by American corporations and multinational corporations.

…So if you worked out a formula that said value content is … you subtract the gap between productivity and wages from your value calculation as to what qualifies, and in point of fact a company producing in Mexico that is underpaying relative to the productivity of its workers would then be facing a rules of origin penalty for that. And so that would work to drive up the wages; that would work to reduce the tensions, wage-driven tensions within North America in the auto sector.

While he was obviously eager to improve conditions in Mexico, Perezcano Diaz said he did feel that rules of origin were not the way to incentivize better labour practices within a country.

While I fully recognize that wages in general are much lower in Mexico, in the auto industry especially, we have literally the highest wages, and the most qualified workers pretty much in all the countries. So that is the auto industry, and mostly export-related companies are there, but probably the auto industry in terms of wages is probably the highest-paying. … But we would like to look at the numbers, the implications, and certainly other ways where we can productively improve standards of living and improve wages generally.

After putting forward their policy ideas and some general structure about what they wanted to achieve, the three agreed that innovative ideas were being thrown around, and that more homework needed to be done to see how and where they could land. Encouraged by the “dents” they made in this unit, they agreed to move to a negotiation on supply management.

Our next negotiation will take place in mid-to-late February. Keep your eyes out for our negotiators to take a second crack at this and other high-friction issues to see if they can find common ground.

Read the transcript in full, here. Or click below to go to the section topic they negotiated:

Edited by Charlie Gillis, Adrian Lee, Nick Taylor-Vaisey and David Thomas

Looking for more?

Get the Best of Maclean's sent straight to your inbox. Sign up for news, commentary and analysis.