“Friendfunding” Industrial Policy

After President Trump announced so-called “Liberation Day” reciprocal tariffs under the International Emergency Economic Powers Act (IEEPA) in April 2025, many US trading partners hurried to Washington to negotiate. Much of the bargaining focused on tariffs and trade, but the administration also pressured major trading partners, particularly those running significant trade surpluses with the United States, to pledge large-scale investments in the US. Many responded with promises to pour capital into American industries in exchange for lower tariff rates. Japan was the first to sign, pledging $550 billion in investment commitments and receiving a baseline 15 percent tariff rate in return. This became the template for subsequent deals.

The European Union (EU) and South Korea followed. The EU committed $600 billion, and South Korea agreed to $350 billion in investments, including a separate $150 billion shipbuilding tranche. Both secured the same 15 percent tariff rate in return. Switzerland and Liechtenstein closed their deal in November 2025, pledging $200 billion for a tariff rate of “no higher than 15 percent.” Most recently, Taiwan finalized a semiconductor-focused deal with $250 billion in direct investment backed by an equal amount of state credit guarantees, also in exchange for a tariff rate of “no more than 15 percent” (Figure 1).

Recall that the negotiating process began with the IEEPA tariffs. After the US Supreme Court struck the tariffs down in February 2026, President Trump invoked Section 122 of the Trade Act of 1974 and imposed a temporary 10 percent import surcharge, which expired in July. On July 23, the Office of the US Trade Representative (USTR) imposed Section 301 tariffs on 60 economies over their supposed failure to impose and enforce forced labor import rules. For covered products from the EU, Taiwan, Japan, South Korea, and Switzerland, the USTR set rates of 10 or 12.5 percent, net of normal most-favored-nation duties and subject to exemptions. These forced labor tariffs are now being challenged in a lawsuit at the US Court of International Trade. Against this shifting backdrop, Japan and South Korea, security-dependent US allies already familiar with USTR pressure, have gone the furthest by formalizing their investment commitments in government-to-government memoranda of understanding (MOUs).

Both the US–Japan and US–South Korea MOUs state that they do “not create legally binding rights and obligations.” They are not traditional trade agreements with formal dispute settlement. In theory, either Tokyo or Seoul could decline to fund a project. But in practice, Washington could threaten to impose higher tariffs on goods from whichever country balks. Because Japanese and South Korean exporters compete head-to-head in the US market, a penalty on one can benefit the other. Enforceability therefore rests on Washington’s economic leverage rather than legal obligation.

Regarding the substance of the MOUs, Washington retains much of the authority over where the investments are directed. Under “Investment Selection,” both MOUs designate the US president as the final decisionmaker to approve investments. Recommendations would be channeled through an Investment Committee, established by the president and chaired by the secretary of commerce.

Japan and South Korea, however, do have some say in the investments. Before the Investment Committee recommends any investment to the president, both MOUs require consultation involving representatives from both sides. The US–Japan MOU states that the Investment Committee “should consult with a Consultation Committee comprising designees from both nations,” without specifying which side would lead. The US–South Korea MOU states that the Investment Committee “will consult with [South] Korea directly or through a Consultation Committee, chaired by the [South] Korean Minister of Trade, Industry and Resources and comprising designees from [South] Korea and the US.”

In both MOUs, the US side presents possible investments to Japan and South Korea for review. Japan is generally expected to fund a selected investment no sooner than 45 business days after notice. South Korea looks to fund its $200 billion strategic investment portion in milestone-based tranches subject to a $20 billion annual cap. If the capital calls threaten foreign exchange market stability, South Korea may request changes in amount and timing, which the United States will consider in good faith, according to the deal’s bilateral fact sheet.

While neither MOU explicitly states who determines the investment amount, the structure speaks for itself: Washington selects and presents each project. If either country declines to provide funding, both MOUs state that the president may raise tariffs on the other country’s imports. Japan’s investments are to be made through January 19, 2029, while South Korea’s projects are to be selected by then and can be funded later. Neither agreement requires the underlying factories, terminals, or power plants to be completed before Trump leaves office.

How would investment work in practice? The key mechanism in both MOUs is a special-purpose vehicle (SPV) managed and governed by the United States. If President Trump approves a semiconductor plant after consultation and the Investment Committee’s recommendation, for example, Japan or South Korea would be bound to fund the deal and the United States governs the financing through an SPV. Available cash flow is split evenly until the financing country recoups its investment, after which 90 percent goes to the United States and 10 percent to Japan or South Korea.

The two MOUs differ in several respects. The US–South Korea structure is two-tiered, with an umbrella investment SPV and separate project SPVs. Its “Korean Vendors, Suppliers, and Project Manager” section says that the United States should, “where feasible and available,” select a project manager suggested by South Korea, a condition that is absent from the US–Japan MOU. It also limits recommendations to investments considered “commercially reasonable,” defined as those expected in good faith to generate sufficient cash flow to cover distributions. Japan’s MOU has no comparable standard. Despite these differences, both countries provide the capital, while the United States manages and controls the financing structure.

On the operational side, the Commerce Department’s Investment Accelerator executes, documents, manages, and administers the investments and serves as the interface for each country’s financing institutions. For Japan, this means the Japan Bank for International Cooperation, which announced a dedicated investment facility in September 2025. In South Korea, the state-backed Korea–US Strategic Investment Corporation was launched in June under the Korea–US Strategic Investment Special Act. It manages the $200 billion strategic investment portion and supports the separate $150 billion shipbuilding program.

Following the signing of the US–Japan MOU last September, President Trump visited Tokyo in October. The White House subsequently released a fact sheet outlining, for the first time, companies’ commitments to specific investments. It identified companies across energy infrastructure, AI, electronics, supply chain, and critical minerals, along with approximate investment figures. Yet all of these projects were simply commitments to pursue rather than contracts to deliver.

In February 2026, implementation of the first batch of projects was officially announced. It comprised a $33 billion natural gas power facility in Ohio operated by SB Energy, a SoftBank subsidiary; a $2.1 billion deepwater crude oil export terminal off the Texas coast operated by Sentinel Midstream; and an approximately $600 million synthetic diamond grit facility in Georgia operated by Element Six. According to the Commerce Department’s press release, the Ohio facility would be “the largest natural gas generation facility in history”; the export terminal was “expected to generate $20–30 billion annually in US crude exports”; and the Georgia facility would “onshore production 100 percent of US demand.”

During Japanese Prime Minister Sanae Takaichi’s March 2026 visit to Washington, a second batch added three more projects: up to $40 billion in small modular reactor construction in Tennessee and Alabama by GE Vernova Hitachi, a $17 billion natural gas generation facility in Pennsylvania, and a $16 billion natural gas generation facility in Texas. NextEra Energy Resources would operate the latter two.

In South Korea’s case, the two sides signed another MOU in May 2026 on the Korea–US Shipbuilding Partnership Initiative. South Korea has highlighted devoting a separate $150 billion tranche of its $350 billion package to the Make American Shipbuilding Great Again (MASGA) project. The initiative established a Korea–US Shipbuilding Partnership Center in Washington, which opened on July 23, with another center proposed for Texas. The centers are intended to translate MASGA into concrete projects.

On the $200 billion tranche, President Trump announced South Korea’s long-awaited investment commitments on September 30, including a $22.3 billion natural gas-fired power plant in Texas, $120 billion for eight large-scale nuclear reactors across the country, and a liquefied natural gas (LNG) project in Alaska. The South Korean government, however, quickly pushed back on the suggestion that all three projects had been finalized. Seoul said the Texas project had been confirmed and the two countries had agreed on a broader nuclear framework, including its overall funding, but no specific reactor projects had yet been agreed to. It also said that on Alaska LNG, the two sides had agreed only to review the project’s commercial viability before deciding whether to proceed. 

The headline numbers should not be treated as completed foreign direct investment (FDI). The packages combine private investment intentions, government equity, loans, loan guarantees, purchases, and gross project values, with different legal force and timelines. The EU statement says only that European companies are “expected” to invest an additional $600 billion through 2028, even though EU institutions cannot order private companies to do so. The still larger announcements with the Gulf states combine US-bound investment with arms, aircraft, and other purchases and are nonbinding. A Peterson Institute for International Economics assessment concluded that several pledges “strain credibility” and that Gulf investment could fall well below the headline figures.

Japan’s and South Korea’s agreements are more concrete, but their deadlines still do not guarantee that projects will be completed by 2029. The six Japanese projects announced through March 2026 carry headline values of about $109 billion. Yet the first signed financing was a $2.2 billion loan for the initial three projects, with total stated values of about $36 billion. South Korea did not launch its administering corporation until June, and while it has finally unveiled its first projects, the two governments are already at odds over the status of some of them, with key details still subject to review and negotiation. Fully funding South Korea’s $200 billion strategic investment portion at the annual cap would take at least 10 years, and the foreign exchange safeguard could further delay the schedule. The useful measures of delivery are therefore signed financing, incremental capital expenditures, and completed projects—not recycled announcements or top-line project values.

Those control and distribution rights give the federal government a contractual financial interest in the SPVs, even when it does not own shares of the operating companies. Public fact sheets for the first six Japanese projects do not disclose whether their financing is equity, debt, or guarantees. Japan says the overall package may use equity, loans, and loan guarantees. It would therefore be premature to say that Washington owns any part of the six operating companies. Rather, the government controls the project vehicles and claims a share of their returns.

The Westinghouse plan goes further. A binding term sheet with the Commerce Department calls for the federal government to arrange financing for at least $80 billion in new reactors, a program the administration linked to the Japanese investment framework. If the government makes the final investment decision and enters into definitive construction agreements, it receives a participation interest entitled to 20 percent of Westinghouse’s cash distributions above $17.5 billion. If that interest has vested and Westinghouse undertakes an initial public offering valued at $30 billion or more by January 2029, it converts into a five-year warrant to buy equity securities based on 20 percent of the public value after deducting $17.5 billion. This is not a current 20 percent federal stake, but it is an explicit path to one.

The publicly released documents do not identify a statute authorizing Commerce to acquire and hold that interest or set rules for valuation, voting, conflicts, disposition, or losses. Nor did Congress create the investment SPVs or prescribe how the foreign funding and government proceeds should be handled. The MOUs say only that the arrangements are subject to applicable law, disclaim fiduciary duties, and create no binding rights. Federal control, private financial returns, and possible equity without a congressionally enacted framework heighten the legal and oversight concerns discussed below.

South Korea’s nuclear investment plan adds another layer, potentially giving Korean companies a direct stake in Westinghouse. Under the bilateral framework, Korean companies, including state-run Korea Electric Power Corp., are seeking to acquire a stake in the nuclear company and share in its profits. According to South Korea’s industry minister, Seoul is discussing a stake of “less than 10 percent” with Washington, without board representation for the Korean companies, although the precise size and other terms remain to be worked out among the companies involved.

The United States did not need these tariff bargains to attract foreign investment. The United Nations Conference on Trade and Development reports that the United States remained the world’s largest host economy in 2025, with $277 billion in inflows. The Bureau of Economic Analysis (BEA) places the inward FDI position at $5.86 trillion at the end of 2025. Europe accounted for $182.4 billion of that year’s increase; Japan was the largest source when measured by ultimate beneficial owner, at $827.1 billion. In a separate release, the BEA reported $232.2 billion in new FDI expenditures in 2025, including $116.6 billion from Europe and $50.5 billion from Japan; South Korea supplied $2.2 billion in greenfield spending. Acquisitions accounted for $218.4 billion of the total, while new establishments and expansions accounted for only $13.8 billion.

Those baselines make attribution difficult. Announcements can include projects already in the works; expansions of established affiliates, acquisitions, loans, and guarantees; or gross facility values rather than new foreign financing. South Korea’s separate $150 billion corporate announcement expressly included previously announced projects. The available numbers show that the United States remains highly attractive to foreign investors. They do not yet show how much additional investment can actually be attributed to the tariff threats. Any net assessment must also account for projects that the administration’s trade-policy uncertainty may have deterred, delayed, or canceled.

The Trump administration’s friendfunding approach is a new variation on an old policy with a long history. Traditional concerns about industrial policy still apply, and this structure creates additional ones.

Along with the traditional concerns of industrial policy comes a set of related issues concerning industry influence and transparency. For instance, where government-provided benefits exist, companies will seek to take advantage of them. Specific sectors will push to have their industry deemed crucial for economic or security reasons, and that definition can be stretched quite broadly. Are shoes crucial to the military, and thus must shoe production be done domestically? If shoes qualify, it becomes difficult to identify any industry unrelated to security.

Within a sector, companies will maneuver to maximize their benefits at the expense of competitors. Rather than focusing on production efficiency, companies will hire more lobbyists to take advantage of government benefits more effectively. This jockeying for government largesse inherently leads to problems with both transparency and corruption. Lobbying done behind the scenes renders the decisionmaking process opaque. Decisions begin to favor those with connections rather than those who are most qualified. Will those with connections to key administration players reach out to shape or join investment deals? If the investors have close ties to top political officials, does that make bailouts more likely should the investments turn sour?

Problems in choosing projects can lead to issues in how they are carried out. Without market discipline, unprofitable projects might be selected. Projects could go bankrupt or require perpetual subsidies to stay afloat. It is worth considering why market actors did not pursue a supposedly valuable project and government help is required.

Another problem with US industrial policy is its “seesaw” quality. Favored sectors change from administration to administration. The shift from pro–electric vehicle and pro-wind policies under Biden to anti–electric vehicle and anti-wind policies under Trump is a recent example. Little progress can be made in any direction when successive administrations reverse one another. Former Biden administration official Peter Harrell has written positively about the Japanese and South Korean arrangements, suggesting that their funds could serve as “patient capital” for “lower-return but strategically important projects.” A future Democratic administration, however, may decide to stop allocating money to a crude oil export terminal in favor of a more “environmentally friendly” project.

Turning to the specific circumstances of the Trump administration’s friendfunding, the announced framework lacks an oversight mechanism prescribed by Congress. The Commerce Department’s Investment Accelerator manages the transactions, but Congress did not create the program or set rules for project selection, valuation, conflicts of interest, government ownership, losses, disposition, or reporting. Nor has Congress designated an independent body to review these decisions. Harrell himself argues that Congress should establish oversight mechanisms while acknowledging that this would require more legislative initiative than Congress has shown. Without such guardrails, the usual problems of industrial policy, including opacity and industry capture, are likely to be worse.

Concerns abound about what impact this will have on US alliances. Japanese and South Korean officials have publicly expressed support for these investment plans. But it is unclear how they and other trading partners truly feel about being pressured in this way. If the projects underperform, or if people in those countries come to feel they were bullied, the resulting resentment could exact a cost that no trade deal can easily repair.

Part of the unease stems from the allocation of control and risk. The United States selects and oversees the projects, while Japan and South Korea supply much of the financing and bear the corresponding exposure. They have consultation rights and can decline a proposal, but their control is limited relative to their financial stake. If a Washington-selected investment goes sour, that imbalance can breed grievances, even among willing allies.

Signs of strain are already visible. Funding for the Japan-backed $40 billion nuclear project in Tennessee and Alabama has hit a roadblock because Japanese lenders want binding assurances that they will not be held liable in the event of a nuclear accident. The greatest risk may be political. For longstanding allies, including South Korea, a free-trade agreement partner for more than a decade, the deals reinforce the sense that established commitments have given way to transactional bargaining.

That concern is particularly visible in Alaska. President Trump’s announcement of South Korean investment in the Alaska LNG project—which Seoul later said it had not committed to—was made at the White House alongside Republican Sen. Dan Sullivan (R‑AK), who is facing a competitive reelection race in November. It created the impression that billions of dollars in South Korean taxpayer money could become intertwined with the administration’s domestic political priorities.

The arrangements may also raise legal concerns. Can the executive branch direct foreign money into government-controlled vehicles, claim their proceeds, and spend or dispose of them without congressional authorization? As the Trump administration learned from the IEEPA tariffs, courts may intervene when the executive branch pushes statutory boundaries too far. Legal experts have raised concerns under the Constitution’s appropriations clause and the statutes that enforce it, including the Miscellaneous Receipts Act, the Anti-Deficiency Act, and the Government Corporation Control Act. As they note: “Directing and spending money are among Congress’s most fundamental checks, not exceptions to them.”

Finally, President Trump and Commerce Secretary Howard Lutnick’s personal direction of investments creates another problem. This “personalist” approach frequently brings the risk of corruption and continues to substitute political officials’ judgment for the knowledge of market participants. It also makes for a messy transfer to a future administration with different priorities.

On its face, a US government push for more foreign investment looks sensible. Foreign investment is valuable, so why not encourage more of it? But the Trump administration has gone far beyond mere encouragement. It has tied foreign investment to tariff pressure, government-selected projects, and presidential discretion. Those are poor substitutes for a coherent investment strategy.

American allies such as Japan and South Korea may well see value in having their companies undertake more US investment, including in semiconductors, shipbuilding, energy, and other sectors. But that does not mean the arrangements are politically and diplomatically costless. Stronger economic ties with the United States through these sorts of investments may make strategic sense for them. Yet when those investments are secured through tariff pressure, even mutually beneficial cooperation can appear coercive.

That concern is heightened when commercial investment overlaps with security priorities. In South Korea’s case, the $150 billion shipbuilding investment is formally part of the investment package, but it also sits alongside broader maritime, security, and defense efforts, including cooperation on US commercial and naval shipbuilding capacity. Taken together, friendfunding blurs the lines among trade policy, industrial policy, alliance management, and national security.

The better approach is to ensure that these investments take into account market forces and their implications for relations with American allies. If these countries are pushed to undertake projects that markets otherwise would not choose, exposing their companies to unnecessary financial risks and raising doubts about whether the United States is treating friendfunded investment as partnership or leverage, the broader project could end up as a cautionary tale.