As the federal government considers spending tens of billions of dollars to counter China and boost U.S. “critical technologies” like artificial intelligence, advocates have all but ignored perhaps the easiest and most cost-effective way to do both: substantially expanding immigration. New research from economist Gordon Hanson shows just how effective the policy could be (emphasis mine):

I examine the specialization of US commuting zones in AI-related occupations over the 2000 to 2018 period. I define AI-related jobs based on keywords in Census occupational titles. Using the approach in Lin (2011) to identify new work, I measure job growth related to AI by weighting employment growth in AI-related occupations by the share of job titles in these occupations that were added after 1990. Overall, regional specialization in AI-related activities mirrors that of regional specialization in IT. However, foreign-born and native-born workers within the sector tend to cluster in different locations. Whereas specialization of the foreign-born in AI-related jobs is strongest in high-tech hubs with a preponderance of private-sector employment, native-born specialization in AI-related jobs is strongest in centers for military and space-related research. Nationally, foreign-born workers account for 55% of job growth in AI-related occupations since 2000. In regression analysis, I find that US commuting zones exposed to a larger increases in the supply of college-educated immigrants became more specialized in AI-related occupations and that this increased specialization was due entirely to the employment of the foreign born. My results suggest that access to highly skilled workers constrains AI-related job growth and that immigration of the college-educated helps relax this constraint.

Based on his findings, Hanson concludes that U.S. immigration restrictions effectively deter the development of AI in the United States: “access to high-skilled immigration relaxes the talent constraint that limits the expansion of AI. The US government, by regulating the volume and composition of high-skilled labor inflows from abroad, in effect regulates the pace of growth in AI.” This “regulation,” in turn, will greatly affect whether the United States’ market-based model for developing AI has an “advantage” over China’s state-directed approach.

Other recent research on AI and U.S. immigration policy comes to similar conclusions:

Countries all over the world are vying for limited AI talent and implementing transparent, straightforward immigration pathways to attract them. Without a similar system, the United States will fall behind. America can no longer rely on only its reputation for stellar research programs and top-ranking companies to attract the best talent. If it is too difficult to work, stay, and put down roots in the United States, AI professionals will go elsewhere. With AI so inextricably linked with both future economic and national security advances, the stakes are high.

As I’ve noted previously, immigration has also been found to be similarly essential for semiconductors — another “critical technology” that federal officials now want to subsidize — and for U.S. innovation more broadly. Unfortunately, U.S. tech and China policy today is heavy on subsidies and top-down planning and light (at best) on immigration liberalization and other market-oriented policies that have been shown to boost U.S. innovation and economic growth. 

U.S. officials thus appear more eager to copy Chinese policy than to effectively counter it.

Advocates of new U.S. restrictions on international trade often cite relatively low U.S. tariff rates while ignoring all the other ways that the federal government can – and does – discourage voluntary cross-border commerce. Perhaps the most common of these non-tariff barriers are U.S. “trade remedies” – antidumping, countervailing duty, and safeguard measures – that apply special duties to covered imports found to harm domestic companies and workers. These measures are not only common – there are more than 500 in place today – but are also subject to abuse by the agencies implementing them, resulting in prohibitive duties at ridiculously high rates (100 percent, 200 percent, or even higher). The abuse can also come directly from laws that Congress has repeatedly revised to further tilt the field toward U.S. producers/​unions.

Cato scholars have been tracking abuse of anti-dumping laws for a long time. Our latest salvo is here and focuses on an issue known as “particular market situation,” where “the cost of materials and fabrication or other processing of any kind does not accurately reflect the cost of production in the ordinary course of trade.” In these circumstances, the Department of Commerce “may use another calculation methodology under this subtitle or any other calculation methodology.” That “other calculation methodology” is almost certainly going to lead to higher anti-dumping duties than would otherwise be found to exist. Indeed, that’s the whole point of the provision, which Congress added to U.S. Antidumping Law in 2015.

Another abusive approach employed by Commerce is known as “adverse facts available” — in which, loosely speaking, the Department decides that foreign respondents haven’t been “cooperative” and thus rejects their data and calculates duty rates using proxy information that’s intentionally adverse to them (often provided by the domestic companies seeking import protection). Just like PMS, AFA inflates duty rates – indeed, most of those absurdly-high rates stem from Commerce’s use of AFA. It is no doubt true that foreign companies and governments occasionally are actually non-cooperative, but their lawyers report (quietly, of course) that Commerce’s AFA decisions are increasingly abusive – for example, by invoking AFA after establishing unreasonably short and rigid deadlines for providing mountains of information (that often must be translated); by applying AFA (instead of the more benign “facts available”) for minor and unintentional technical errors; or by throwing out all evidence submitted by foreign respondents (“total AFA”) where only a small amount of data is in question.

Even without these practices, U.S. trade remedy laws would raise significant problems, but these two are probably the most egregious and harmful. And new data show that the number of AFA and PMS determinations has been on the rise (AFA cases are remaining steady as a percentage of total cases, but there are more total cases these days; PMS is being used more often in percentage terms):

Such practices not only harm U.S. consumers — the same study shows that the average duty rate applied to imports in AFA (full or partial) cases was 141.03 percent over the period examined – but also have a way of boomeranging back to American exporters too. As noted here, for example, these and other questionable trade remedies practices are spreading to other countries. Trade remedy abuse is a loss when we do it to others; the loss is compounded when others do it back to us.

Commerce’s behavior – and the laws authorizing it — also serves as a warning about the perils of U.S. protectionism and industrial policy: even in systems administered by a “neutral” bureaucratic arbiter and designed to be insulated from the political process, the administering agency can become “captured” by powerful domestic interests and the rules can be corrupted by politicians eager to reward attentive constituents with highly technical changes hidden from the vast majority of the American public.

Unfortunately, it’s that majority paying the bill.

In case you haven’t noticed, U.S. industrial policy is having (yet another) moment. Armed with the latest data and cross-country comparisons, a large and bipartisan cadre of industrial policy advocates in Washington are eager to shovel billions of taxpayer dollars into the open arms of American manufacturers of “essential goods” and “critical technologies.” The risks (China, pandemics, whatever), so the theory goes, greatly outweigh any harms that a few, scattered industrial policy failures might cause along the way, so why not just throw money at the (perceived) problems? These advocates, however, rarely acknowledge the frequency of such failures — caused by a host of well-known obstacles to effective American industrial policy (e.g., the “knowledge problem,” public choice, unintended consequences, budgetary overruns, opportunity costs, and plain ol’ inefficacy) — or fully account for the harms that industrial policies can impose on the U.S. economy and the very objectives that the policies seek to achieve. I documented some of these industrial policy failures in a recent Cato paper, but yesterday’s New York Times provides the best example that I’ve seen in a long, long time:

WASHINGTON — More than eight years ago, the federal government invested in an insurance policy against vaccine shortages during a pandemic. It paid Emergent BioSolutions, a Maryland biotech firm known for producing anthrax vaccines, to have a factory in Baltimore always at the ready.

When the coronavirus pandemic arrived, the factory became the main U.S. location for manufacturing Covid-19 vaccines developed by Johnson & Johnson and AstraZeneca, churning out about 150 million doses as of last week.

But so far not a single dose has been usable because regulators have not yet certified the factory to allow the vaccines to be distributed to the public. Last week, Emergent said it would destroy up to 15 million doses’ worth of the Johnson & Johnson vaccine after contamination with the AstraZeneca vaccine was discovered.

Emergent and government health officials have long touted their partnership as a success, but an examination by The New York Times of manufacturing practices at the Baltimore facility found serious problems, including a corporate culture that often ignored or deflected missteps and a government sponsor, the Biomedical Advanced Research and Development Authority, that acted more as a partner than a policeman.

But wait, there’s more:

Emergent is a longtime government contractor that has spent much of the last two decades cornering a lucrative market in federal spending on biodefense. The Times reported last month that sales of its anthrax vaccines to the Strategic National Stockpile accounted for nearly half of the stockpile’s half-billion-dollar annual budget throughout most of the last decade, leaving the federal government with less money to buy supplies needed in a pandemic.

And more:

In 2012, the Department of Health and Human Services awarded three contracts, including one to Emergent to retrofit and expand its Baltimore site. And between 2015 and 2019, the government placed a handful of relatively small orders for doses of treatments and vaccines being developed for use against viruses including Ebola and Zika.

The contract also required Emergent to stand ready to produce up to 50 million doses of vaccine within four months in the event of an influenza pandemic. To show that it was up to the task, Emergent was supposed to work with a company developing a flu vaccine candidate and seek F.D.A. approval to manufacture it by June 2020.…

When the June 2020 deadline arrived, Emergent hadn’t met the flu requirement, but the government nevertheless awarded it a $628 million contract to manufacture Covid-19 vaccines.

Overall, the Times reports that Emergent situation — bearing almost all of the hallmarks of American industrial policy in action — has imperiled tens of millions of Covid-19 vaccine doses. Fortunately, the United States has other vaccine alternatives, including J&J doses imported from the Netherlands, and, of course, the hundreds of millions of jabs made here by Moderna and Pfizer.

Pfizer’s case, in particular, provides a clear alternative to the top-down planning and industrial policy that’s so hot right now. As I explained a few weeks ago:

The vast majority of the hysteria we heard about empty store shelves and the failures of globalization proved meritless—instead reflecting a single, shocking moment in time that markets, companies, and governments mostly addressed in a matter of months (“seemingly out of nowhere”). Today, some of the biggest impediments to our economic recovery are U.S. trade restrictions, not “free market fundamentalism” or whatever.

Meanwhile, Pfizer and [Germany’s] BioNTech were quietly working on their miraculous COVID-19 vaccine, leveraging Pfizer’s substantial pre‐​existing U.S. manufacturing capacity, multinational research teams stocked with immigrants, global capital markets and supply chains, and a logistics and transportation infrastructure that had developed over decades to deliver those “cheap consumer goods” that the populist punditocracy loves to hate. The companies went from concept to final delivery of millions of vaccine doses in about nine months—just as their management boldly predicted last April. Surely, some state support (e.g., grants for previous mRNA research and vaccine purchase commitments) was involved, but the Trump administration’s contract with Pfizer expressly excluded from government reach R&D, clinical trials, and manufacturing supply chain issues (i.e., “activities that Pfizer and BioNTech have been performing and will continue to perform without use of Government funding”). Thus, even the uber‐​nationalist Trump White House realized that government attempts to “nationalize” and micromanage the vaccine’s development and delivery would have delayed—if not thwarted—those processes, costing numerous American lives along the way.

When time was of the essence and success really, really mattered, they just got out of the way. And it worked.

The Pfizer vaccine thus provides a valuable lesson regarding the dynamism and capabilities of the U.S. manufacturing sector and the free(ish) market more broadly — a lesson now greatly amplified by the Emergent fiasco and its contrasting industrial policy approach.

It’s an open question, however, as to whether anyone in Washington will heed it.

A new federal court ruling has invalidated some of President Trump’s “national security” tariffs, and in the process addressed one problematic procedural aspect of one of the worst U.S. trade laws on the books. Congressional action is still needed, however, to fix the statute’s many other problems.

This week, the U.S. Court of International Trade ruled against the Trump administration’s January 2020 action (Proclamation 9980) to expand tariffs of 25% on steel and 10% on aluminum, implemented pursuant to Section 232 of the of the Trade Expansion Act of 1962, to “derivative” (downstream) products like nails and car fenders. As we discussed in a recent policy analysis, Section 232 allows the president to “adjust” imports that “threaten to impair the national security,” but the statute suffers from numerous procedural and substantive flaws – errors that allow the executive branch to declare imports of almost anything a threat and cut procedural corners in restricting those imports.

The new USCIT case, brought by PrimeSource Building Products, Inc., addresses one of those corners, which we discussed in our paper:

After the tariffs went into effect, an importer of Turkish steel sued the Trump administration in the USCIT, arguing (among other claims) that the president cut corners in the procedures required by law. In November 2019, a unanimous three‐ judge panel agreed that “the President’s expansive view of his power under Section 232 is mistaken, and at odds with the language of the statute, its legislative history, and its purpose.”

Despite the court’s order that the Trump administration’s “expansive view” of presidential power is “mistaken,” however, President Trump soon advanced an even more expansive view of his power. On January 24, 2020, Trump announced that he was broadening the original “national security” tariffs to include “derivative” steel and aluminum products such as nails, pins, and staples. Again, he pointed to the original proclamations as his source of legal authority—the same justification that the USCIT had denied weeks earlier. Furthermore, the tariffs were scheduled to go into effect on February 8, 2020, just 16 days after the announcement, leaving little room for public consultation and thorough consideration of the impact of such actions. Several companies filed lawsuits against these tariffs at the USCIT, challenging the actions being taken outside Section 232’s prescribed 90‐ day window, the lack of public consultations, and disrespect to their due process. Since February 13, 2020, the USCIT has granted injunctions to these companies, preventing U.S. Customs and Border Protection from collecting tariffs on their “derivative” steel and aluminum imports, until it hears their cases.

Trump’s derivatives tariffs clearly violated the law’s time limits for implementing an “action” against imports (the president has 90 days to decide whether to act, and 15 more days to take any such action), and they were a prime example of cascading protectionism. On the latter point, the Trump administration admitted as such in the proclamation expanding the Section 232 tariffs to steel and aluminum “derivates” where it asserted that the original tariffs (25% on steel and 10% on aluminum) had not achieved the administration’s initial capacity utilization goals, in part because the tariffs lead to price increases in steel and aluminum imports (as one would expect when you tax something), and this prompted industrial consumers to purchase “derivative” products from abroad, which ended up hurting domestic derivative products producers and depressing domestic demand for the tariffed metals.

The USCIT ruled narrowly on the former point (the deadlines for action). The court’s decision will likely be extended to other companies that filed lawsuits similar to Primesource’s and should be welcomed by free traders and importers – particularly U.S. construction companies that are now reeling from sky high materials prices.

It does not, however, solve the problem of Section 232.

As we explain in our paper, Trump did far more than simply ignore this procedural requirement in implementing the steel and aluminum tariffs and investigating other products under Section 232. His administration’s actions revealed several major substantive and procedural flaws in Section 232 – flaws that are in desperate need of reform but that the USCIT, by its own admission, won’t touch. Absent legislative action, the abuse of Section 232 will thus remain a serious risk – checked only by the president’s self-restraint.

The ruling will also provide a test for the Biden administration because it can appeal the USCIT’s decision, thus keeping Trump’s derivatives tariffs alive, and defending Trump’s expansive (abusive) interpretation of executive power. Doing so, however, would not only hurt many U.S. companies (and the economy more broadly) but also provide a clear and unfortunate indication that the Biden administration – for all its talk of countering Trump’s trade abuses and restoring U.S. alliances – condones Trump’s approach to Section 232 and “national security” and might even seek to open pandora’s box even further.

Earlier this week Wall Street Journal columnist Mary O’Grady penned an excellent piece detailing the myriad ways in which U.S. maritime protectionism interferes and distorts trade. Some of the examples presented may be familiar, such as the imperilment of Alaska’s summer cruise season. But the column also delves into effects of the law that are often overlooked, such as its impact on ports:

The Jones Act is particularly costly to Puerto Rico because it keeps the island from capitalizing on its comparative advantage as a transit point for cargo. Colin Grabow, who heads the Cato Institute’s Project on Jones Act Reform, puts it this way: “Absent the Jones Act we would see large ships drop their cargo in San Juan or Ponce for placement on smaller ships to various ports in the region, including the U.S.—basically a hub-and-spoke model. But no one will ever use Puerto Rico as such a transshipment hub so long as shipping between PR and the world’s largest economy is subject to the Jones Act.”

This merits further explanation. For those unfamiliar, transshipment is the transfer of cargo from one vessel to another before it is brought to its final destination. The massive containership Ever Given that recently blocked the Suez Canal, for example, was bound for the port of Rotterdam where much of the ship’s containers would be unloaded for placement onto smaller vessels for destinations around Europe.

But such transshipment of containers for destinations elsewhere in this country does not happen because any goods moved between U.S. ports—including those originating from abroad—must use expensive and uncompetitive Jones Act shipping. As the Congressional Research Service notes, “Transshipment of international containerized cargo by feeder ships is prevalent abroad, but the practice does not exist in the United States.” Instead, containers are not moved by water to their final U.S. destination but rather by rail and trucks (with their attendant environmental costs).

This lack of transshipment extends to Puerto Rico. While neighboring Jamaica and the Dominican Republic have emerged as major transshipment hubs where large ships arriving from Europe and Asia (via the expanded Panama Canal) can have their cargo loaded onto to smaller vessels for transport to destinations throughout the Caribbean, Puerto Rico isn’t even a part of the transshipment conversation. 

The Jones Act figures prominently here. Why use the island as a shipping hub when other ports in the region not beholden to the law offer far less expensive options for transport to the world’s largest consumer market? The end result is a missed economic opportunity for Puerto Rico. 

That the law greatly reduces Puerto Rico’s attractiveness for transshipment isn’t controversial. To wit:

• A 2012 report from the Federal Reserve Bank of New York states that most experts agree the Jones Act “diminishes the viability of the Island as a major regional trans-shipment port.”
• The Inter-American Development Bank flatly states that the port of San Juan is not part of a transshipment cluster “since it is a Jones Act port.”
• A 2000 report authored for Puerto Rico’s government about the island’s feasibility as a transshipment point conceded that “transshipment between foreign and U.S. destinations via a Puerto Rican transshipment port would usually not be attractive as cabotage [i.e. Jones Act] shipping rates are on average higher.”

A 2014 article in American Shipper also cited the Jones Act, along with high labor costs, as key reasons why the Puerto de Las Américas in Ponce is not terribly attractive for transshipment purposes. As a visit by the Cato Institute’s multimedia team shows, to call the port underutilized is putting it mildly:

A common perception of the Jones Act and other forms of maritime protectionism is that their costs manifest themselves in higher transportation costs that are then passed along to consumers. But that’s just a starting point. These laws also mean the forced diversion of tourist dollars to Canada. Higher taxpayer expenditures for dredging. The inability to transport liquefied natural gas by ship to Puerto Rico and New England. Reduced opportunities for Puerto Rico’s port. In ways large and small the Jones Act and similar laws exact their tolls. Often overlooked, but always there and chipping away at the country’s economic vitality. 

During the Trump administration, national security rose in prominence as a reason for restricting trade, investment and other economic activity. Among other examples, purported national security concerns led to the imposition of steel and aluminum tariffs under Section 232 of the Trade Expansion Act of 1962 (as our colleagues Scott Lincicome and Inu Manak describe here); there were bans on Chinese social media apps WeChat and TikTok; and most recently, the shares of certain Chinese companies, including electronics maker Xiaomi, were prohibited from being traded by Americans.

There are, of course, some legitimate security concerns relating to China specifically and to the world more broadly. However, when U.S. courts have been asked to review the actions of the Trump administration that were supposedly based on national security considerations, they have often rejected the justifications being offered. WeChat and TikTok were granted preliminary injunctions against the Trump administration’s bans (see here and here). And just recently, Xiaomi received a preliminary injunction of its own. While these decisions are not the final word, they do offer some indications of how U.S. courts will approach the issue.

In this post, we first explain the restrictions imposed on Xiaomi, and then turn to the lawsuit and the court rulings. The flaws in the Trump administration’s national security justification in this case are a reminder that the Biden administration should review carefully the Trump administration’s decisions that invoked national security, and should carry out policy in this area in a way that addresses these complex issues with more balance and nuance.

Designation of Xiaomi as a “Communist Chinese Military Company,” and the Resulting Punishment

In the Xiaomi case, the U.S. government actions at issue started with President Trump exercising power under Section 1237 of the National Defense Authorization Act for Fiscal Year 1999 (as amended), which specifies that the Secretary of Defense “shall make a determination of those persons operating directly or indirectly in the United States or any of its territories and possessions that are Communist Chinese military companies and shall publish a list of those persons in the Federal Register.” The President is authorized by this statute to take certain actions against these companies pursuant to the International Emergency Economic Powers Act (IEEPA). Possible actions under the provisions include: (1) prohibiting foreign exchange transactions or (2) prohibiting the import or export of currency or securities. In terms of creating a list, the statute covers any “person” who is “engaged in providing commercial services, manufacturing, producing, or exporting” and is “affiliated with” the Chinese military or a ministry of the government, or “owned or controlled by any entity affiliated with the defense industrial base” of China.

After many years of inaction, in June of 2020 the Department of Defense (DoD) began publishing lists of Communist Chinese Military Companies (CCMC). All together, the DoD has designated 44 Chinese companies as CCMCs, including companies in the aerospace, energy, shipbuilding, construction, technology and communication sectors. Xiaomi was added to the list on January 14, 2021, just before Trump left office. (Other well-known companies on the list include Huawei, China Mobile Communications, China Telecommunications Corp and China Unicom.)

Designation of an entity as a CCMC does not automatically lead to a specific punishment, but last November President Trump gave the CCMC list some teeth. He issued Executive Order 13959 entitled “Addressing the Threat from Securities Investments that Finance Communist Chinese Military Companies” (the Order was amended in January of 2021). The Order declared a national emergency pursuant to IEEPA, on the basis that China is “increasingly exploiting United States capital to resource and to enable the development and modernization of its military, intelligence, and other security apparatuses, which … directly threaten[s] the United States homeland and United States forces overseas.” The Order prohibited any U.S. person from investing in or trading any publicly traded securities, or derivatives thereof, of designated companies and their subsidiaries as of March 15, 2021, and requested U.S. persons to divest any such holdings by November 11, 2021. The Office of Foreign Assets Control under the Treasury Department later published guidance clarifying the scope of these prohibitions.

As a result of these actions, the New York Stock Exchange (NYSE) decided to delist China Mobile Communications, China Telecommunications Corp and China Unicom. With regard to Xiaomi, which is not listed on the NYSE, multiple U.S. banks, including Morgan Stanley, JP Morgan Chase and Goldman Sachs, announced that they would suspend trading of Xiaomi stock, and drop their Xiaomi Hong Kong Stock Exchange listings. And Global index publisher FTSE Russell said it would drop Xiaomi from its global and Chinese indexes. These actions almost certainly contributed to Xiaomi’s 9.5 percent stock price drop.

Xiaomi and certain shareholders filed a lawsuit in U.S. district court in late January 2021 (later filing an amended complaint) claiming that the U.S. government actions violate the Administrative Procedures Act, exceed the scope of authority granted by Section 1237 as well as the authority of Executive Order 13959, and violate the Fifth Amendment’s due process clause.

At the time Xiaomi was included on the list, no explanation for the Trump administration’s decision on Xiaomi was provided, but during the course of the litigation the U.S. government supplied the decision document relied on by the DoD. As explained in the court ruling, this document noted that Lei Jun, Xiaomi’s CEO, had been recognized as an “Outstanding Builder[] of Socialism with Chinese Characteristics” by the PRC’s Ministry of Industry and Information Technology; and the document highlighted Xiaomi’s plan to invest in 5G and AI, which are both “[c]ritical [t]echnologies essential to modern military operations,” according to the 2019 DoD Industrial Capabilities Report. The ruling explains that “[t]he decision memo concludes by stating perfunctorily that ‘Xiaomi meets the criteria’ for CCMC classification.”

The U.S. District Court Ruling

In a ruling issued on March 12, 2021, U.S. District Judge Rudolph Contreras focused on the claims under the Administrative Procedures Act. He agreed with Xiaomi’s claims and issued a preliminary injunction against the prohibitions imposed by the Trump administration on Xiaomi. The key parts of his reasoning were as follows.

In evaluating the likelihood of success on the merits (the first prong in the standard for granting a preliminary injunction), Contreras first found that the DoD’s explanation for designating Xiaomi as a CCMC is “inadequate.” In this regard, he said that the DoD memo that provided the basis for the decision to include Xiaomi “skips the most ‘critical step’ of an agency ‘connecting the facts to the conclusion.’”

Furthermore, he noted that “Xiaomi is a publicly traded company that produces commercial products for civilian use, is controlled by its independent board and controlling shareholders, and is not effectively controlled or associated with others under the ownership or control of the PRC or its security services.” He concluded that “Xiaomi is not an affiliate of any of the proscribed entities identified by Congress in Section 1237.”

In addition, he found that the DoD failed to provide substantial evidence to support its determination. DoD had cited two facts related to Xiaomi’s designation: The award received by Xiaomi’s CEO, and Xiaomi’s plan to invest in 5G and AI. Contreras was not persuaded by these arguments.

In the end, he concluded that “[t]he Department of Defense’s CCMC designation process as to Xiaomi was deeply flawed and failed to adhere to several different [APA] requirements.” After finding that the other factors related to preliminary injunctions were also met, Contreras granted the injunction.

As a result of this injunction, other Chinese companies on the list are pursuing similar lawsuits. In the meantime, the Biden administration said that it is still in the process of deciding the “appropriate path forward” in the Xiaomi case.

Conclusions

A review of the Xiaomi designation as a CCMC was clearly warranted, but the issues here go beyond Xiaomi. What the Biden administration needs to do is carry out a broader review of the Trump administration’s invocations of “national security.” (The Biden administration already started such a review on the TikTok and WeChat bans.) National security is undoubtedly a real concern, but the Trump administration applied it arbitrarily and, in some instances, as a disguised means of protectionism. An overbroad approach to the issue could undermine legitimate security concerns by calling into question the credibility of the concept as it is being practiced.

In a bid to leave his mark on Mexico’s energy sector, Mexican President Andres Manuel López Obrador (AMLO) is rolling back reforms by his predecessor Enrique Peña Nieto that were designed to increase competition in the energy market. For years, Mexico’s energy market was in decline due to a lack of foreign investment and competitive pricing, governance challenges, capital constraints, and a drop in crude oil production. Peña Nieto sought to reverse this trend with ambitious reforms that required amending Mexico’s Constitution to allow for much needed foreign direct investment (FDI). But in his populist push, AMLO has railed against this market opening, and sought to increase the role of state intervention in Mexico’s energy markets.

Through a recent bill, which passed comfortably through Mexico’s Congress on March 2, 2021, AMLO is fulfilling his promise to make changes to Mexico’s Law of the Electric Power Industry (LIE) by putting more power in the hands of state regulators. The changes are significant enough to have generated broad opposition from energy experts, environmental activists, investors, and business associations. While the law was supposed to go into effect within six months, a temporary reprieve has been handed down by Mexican district courts, suspending implementation until they rule on the law’s constitutionality. AMLO is confident that, if the dispute escalates, the Supreme Court of Mexico will rule in his favor, but if it does, it is likely to lead to further discord between Mexico and its foreign investors, and add to an already strained U.S.-Mexico relationship. With U.S. Trade Representative Katherine Tai putting emphasis on enforcing existing trade rules, and pressure from Congress likely to build, the electricity bill could trigger international litigation if it is not resolved in Mexico’s courts.

Why doesn’t the electricity bill add up to “reform”?

AMLO’s “reform” of the Law of the Electric Power Industry represents the first instance of his administration pushing for major legislative changes in energy. But what does AMLO’s bill do, exactly? The bill undercuts competition and the protections to private investors established through Mexico’s energy reforms (passed in 2014), as well as efforts to combat climate change and transition to clean sources of energy. It’s not so much “reform” as an energy policy “regression” for Mexico. There are a number of concerns with the law, but it’s the potential impact on foreign investors that could undo much of the gains in the energy sector since the energy reforms went into place. We address some of these issues below.

First, the law mandates that Mexico’s independent system operator for the electricity market, Centro Nacional de Control de Energía (CENACE), which manages the national electricity system, change the order in which electricity is dispatched to the grid to serve large consumers. Currently, these consumers acquire energy through a wholesale market made up of private firms and the state-owned power utility, the Federal Electricity Commission (CFE). CFE is involved in electricity generation, transmission, distribution and retail, though generation “is by far its least profitable activity,” according to a report by the Wilson Center. CENACE then authorizes the dispatch of the cheapest energy available from these producers. Excluding hydroelectric, geothermal, and nuclear plants – which do not have the combined capacity to satisfy demand – private solar and wind generators and natural gas-powered plants represent the cheapest available sources.

Challenging the independence of the country’s electricity management, AMLO’s new law requires electricity to be dispatched by CENACE according to a fixed order: 1) Hydroelectric plants; 2) CFE-owned nuclear and geothermal plants; 3) CFE-owned natural gas-powered plants; 4) CFE-owned carbon- and fuel-powered plants; 5) Privately-operated natural gas-powered plants under contract with CFE; 6) Wind and solar electric generation systems; 7) Private natural gas-powered plants not contracted by CFE.

This order appears to discriminate against most private electricity generators in favor of the state-owned CFE. If implemented, this aspect of the legislation could be challenged in international tribunals for giving preference to a state-owned enterprise (it is already being challenged in domestic courts).

While AMLO has defended Mexico’s sovereignty on energy issues on the basis of the United States-Canada-Mexico Agreement (USMCA), which recognizes Mexico’s “sovereign right to reform its Constitution and its domestic legislation,” the reality may be a little more complicated. Depending on how the law is implemented, it is possible that it could violate a number of provisions of the USMCA, and both state-to-state and investor-state dispute procedures are available. Notably, legacy investments are still covered under the scaled back investor-state dispute settlement rules for three years after entry into force; and for new investments, the oil and gas and power generation sectors are still covered for investor-state disputes related to covered government contracts. While Canada does not have recourse to ISDS against Mexico under USMCA, it could pursue a claim under the Comprehensive and Progressive Trans Pacific Partnership (CP-TPP), to which Mexico is also a party.

The CP-TPP adds an additional layer of complexity because Mexico “locked in” some aspects of its energy reforms in that agreement. This is important because the USMCA grants Mexico the right to adopt additional non-conforming measures on investment and trade in services, as long as these do not accord USMCA parties treatment less favorable than that accorded to parties from any other trade agreement that Mexico has ratified – a clear reference to CP-TPP. Thus, insofar as Mexico cannot restrict investments in the areas covered under CP-TPP, it also cannot do so under USMCA.

The order in which electricity is dispatched also has potential environmental implications, as carbon- and fuel-powered plants would be relied upon more heavily for electric power generation than cleaner wind and solar electric generation systems, which will now sit near the bottom of the queue. Not only could this undermine Mexico’s transition to cleaner energy sources, it could also violate Mexico’s commitments under the eight multilateral environmental agreements covered by the USMCA’s environmental chapter.

In addition, the legislation undercuts protections for private firms that entered the industry prior to the enactment of the energy reforms. Beginning in 1992, these firms were able to generate electricity for their associates, including large industrial firms such as Ford, Continental, and Hershey’s, and as contractors for CFE on a limited basis. When the LIE overhauled the legal framework governing the industry in 2014, the permits and contracts obtained by these firms were grandfathered up until their original expiration date. But AMLO’s reform now provides for their potential cancellation within a span of 6 months, should a revision by the Energy Regulatory Commission (CRE) and CFE find these permits and contracts to be “fraudulent” or not “cost-effective.”

What’s at stake?

Mexico’s electric power sector received more than $11 billion USD in net foreign direct investment between 2014 and 2020. Most of this FDI comes from EU countries, Canada, and the United States. Since the energy reforms, opportunities for U.S. investments in the energy sector have grown, and a vibrant cross-border electricity trade has developed, as electricity prices in Mexico became more competitive. As can be seen in Figure 1, U.S.-Mexico electricity trade has witnessed pronounced growth in recent years. AMLO’s actions could therefore threaten to undo some of the gains of integration in our energy markets.

In fact, since AMLO took office, there have been a number of alarms raised over his actions in the energy sector that have concerned foreign investors. Former Secretary of State Mike Pompeo sent a letter to Mexico’s cabinet officials warning that “recent regulatory actions by the Mexican government have created significant uncertainty about Mexico’s regulatory process, especially regarding the energy sector, and have damaged Mexico’s overall investment climate.” He went on to say that these actions also “raise concerns regarding Mexico’s commitments under the USMCA.” Other trading partners have put forward similar concerns. For example, an attempt last year by CENACE to effectively restrict access to the electrical grid for renewable energy, under the guise of safeguarding its “reliability” and “security,” prompted Canada and the EU to send formal letters to the Secretary of Energy, stating that this and other actions from the administration undermined Canadian and European investments in the sector.

But Mexican consumers may pay the heaviest price. In 2020, CFE’s average generation costs were 252 percent higher than those of private firms employing renewable sources. Experts further estimate that the fixed order for electricity dispatch mandated by AMLO’s bill would lead to the average cost of electricity increasing by more than 180 percent – an impact that would not only be reflected directly on utility bills, but also in higher prices for other goods and services.

In cementing his legacy, AMLO may turn off the lights

With his presidential term coming to an end in 2024, López Obrador has a few years left to leave his mark on Mexican politics and cement his legacy against what he claims to be Mexico’s “long neoliberal nightmare.” A key part of this is his crusade against the energy reforms. The fact that AMLO’s party, Morena, remains highly popular and may clinch another majority in Mexico’s Congress in June’s mid-term elections, gives AMLO little reason to back away from his controversial policies.

AMLO is not likely to easily give in to international pressure to change his approach, so the best outcome we may be able to hope for is a domestic resolution whereby Mexican courts declare the law unconstitutional. But in case this does not happen, Canada and the United States, as Mexico’s neighbors and closest trade partners, will have to raise the issue with AMLO directly to ensure that his actions do not jeopardize the hard gains of the energy reforms and the future of North American trade and cooperation on energy-related issues.