The Legal Bases for Government Stakes in Private Firms

President Trump’s Aug. 22 announcement that the U.S. government had taken a 9.9 percent stake in troubled chipmaker Intel is a marked shift in U.S. policy, which for decades has provided grant and tax incentives to encourage private-sector activities but has generally avoided outright ownership stakes in private firms. The idea of the government taking equity, however, did not come entirely out of the blue. In July, the Trump administration announced plans to take an equity stake in mining company MP Materials as part of a strategy to boost supply chains for critical minerals and earlier this year announced it would take a “golden share” in U.S. Steel as a condition of permitting a Japanese firm to buy the century-old steelmaker.

The concept of government stakes in companies has some bipartisan support as well. Sen. Bernie Sanders (I-Vt.) quickly expressed support for the government taking a stake in Intel. Progressive thought leaders including Vanderbilt’s Ganesh Sitaraman and Yale’s Anne Alstott have argued for a “public option,” with the government (usually a state or local government) owning companies in a range of industries to encourage competition with private competitors. New York Democratic mayoral candidate Zohran Mamdani famously supports city-owned grocery stores.

This political shift toward government equity in companies raises complex policy issues regarding the government’s role in the economy. On the one hand, when the government takes a risk on a new technology, or on sponsoring a corporate enterprise, it seems only fair that the taxpayer should have a chance to profit on the upside. On the other hand, the U.S. has typically avoided government ownership of private firms out of fear that it will reduce innovation and efficiency and encourage cronyism.

But the seeming shift toward American state capitalism raises equally important legal questions. In 1952, after President Truman issued Executive Order 10340 to seize American steel mills, the mills sued. One of them, Ohio’s Youngstown Sheet & Tube Co., went down in history as the named plaintiff in the seminal Supreme Court case on executive power, which found the seizure unlawful.

Unlike Truman, who seized the mills against the wishes of their owners, the Trump administration has so far taken equity stakes with the consent of the companies involved. In the case of Intel, Trump used money appropriated in the 2022 CHIPS Act to encourage U.S. semiconductor manufacturing to purchase the shares, and in the case of U.S. Steel, the administration made clear that it would not permit the foreign takeover without the U.S. government’s stake. Some companies may be granting the government stakes under duress; nevertheless, formally speaking they are granting them.

There is still, however, the matter of whether the government stakes are lawful. While the Trump administration has not published a detailed legal basis for its acquisitions of corporate stakes, it is likely relying on a combination of broad contracting authorities to take equity stakes as part of a number of grant programs or other contracting vehicles. Statutes do not appear to preclude the government from demanding stakes as part of certain regulatory processes, including the so-called CFIUS process that resulted in the government’s stake in U.S. Steel.

Perhaps most important from the administration’s perspective, so long as the companies formally consent to the government shares, the government is unlikely to face lawsuits in court. Intel, for example—having just agreed to sell itself to the federal government—seems unlikely to turn around and pursue its day in court. This is true even if recently announced details of the deal, which appear to relieve Intel of some of its prior commitments to the U.S. government to actually build fabrication plants in the U.S., contradict the spirit of the CHIPS Act.

That said, if Trump begins to regularly require companies to hand over shares as a condition of government contracts or receiving permits and licenses, someone will eventually sue. But a more targeted approach may let the government duck legal challenges. If Congress wants to oversee the government’s acquisition of private assets, or limit the government’s ability to acquire them, it needs to act rather than waiting on the courts.

A Brief History of U.S. Government Corporate Ownership

The concept of the U.S. government owning stakes in private companies has been debated since the early days of the Republic. In the late 18th and early 19th centuries, the federal government chartered and held a minority stake in the First Bank of the United States and its successor, the Second Bank of the United States. Both of these were federally chartered banks intended to serve public purposes as well as private interests, which were majority owned by private investors. The banks, however, were controversial from the beginning, and Congress declined to renew the Second Bank’s charter after it expired in 1832.

The demise of the Second Bank of the United States kicked off a long period in which the federal government generally avoided ownership of private firms, though in 1903 the U.S. government purchased the Panama Railroad Company from a French corporation as part of Teddy Roosevelt’s plan to build the Panama Canal.

In the 1930s, however, New Deal programs enacted in response to the Depression began to transform both the U.S. economy and the relationship between the economy and the state. By the mid-1930s, for example, the Reconstruction Finance Corporation (RFC), which Congress had established to rescue America’s failing banking sector, owned stakes in about half the nation’s banks. As the RFC’s final report in 1957 noted, “RFC had more flexibility than agencies operating under traditional Government appropriation procedures. Besides making loans of many types, the RFC subscribed for, purchased, and traded in the securities of private business enterprises.”

That said, since the 1950s the U.S. government has typically owned stakes in private companies only as part of temporary bailout programs designed to resolve problems at specific companies or in industries seen as important to the U.S. economy. In 1980, for example, while the U.S. government did not take direct ownership of Chrysler as part of a congressionally authorized bailout of the automaker, the government did receive warrants to purchase stock as part of its security interest in bailout loans—warrants the government ultimately sold at a profit in 1983. Federal banking regulators have generally avoided taking direct ownership stakes in failed banks, instead putting them into receivership and seeking sales to other financial institutions. But in 1984, the Federal Deposit Insurance Corporation (FDIC) took a majority stake in Continental Illinois National Bank, then one of the nation’s largest banks, as part of a bailout. In 2008 and 2009, the government took ownership in a range of companies from banks to carmaker GM as part of the congressionally authorized Troubled Asset Relief Program (TARP), passed to address the global financial crisis. In 2020 and 2021, the government took warrants to purchase airline stock as part of the CARES Act’s support for the U.S. airline industry, which was then reeling from the coronavirus pandemic. Across these cases, however, the U.S. government generally planned to own the stakes for only a limited period of time, until the crisis had passed: The TARP program had a statutory sunset, and the government subsequently sold off the stakes in companies it acquired as part of TARP and later sold the airline warrants it received pursuant to the CARES Act.

The U.S. also has a history of government-sponsored private or nonprofit enterprises that serve specific public purposes. In 1933, Congress passed the Tennessee Valley Act (TVA), chartering the act to help develop an area spread across eight southeastern states. In 1970, Congress passed the Rail Passenger Service Act, establishing Amtrak to preserve U.S. passenger rail in the face of the likely bankruptcy of most passenger rail companies, then reeling from competition from automotive and airline travel. The government set up the Federal National Mortgage Association, better known as Fannie Mae, as a federal agency in 1938 to provide liquidity to the mortgage market. It began taking lenders as shareholders and became a publicly traded company in 1954 before going public in the late 1960s—until it was taken over by the federal government again during the 2008-2009 financial crisis. Congress established the Federal Home Loan Mortgage Corporation (Freddie Mac) in 1970 and 1989, but it, too, was taken over again in 2008. Recently, President Trump said he wants to re-privatize both companies.

The U.S. has also supported government-backed venture capital firms, the best known being the CIA’s In-Q-Tel, which was established in 1999 and today has $1 billion in assets. NASA and the Army also have government-backed venture capital firms. These funds, however, do not represent direct government ownership in private firms. From a legal perspective, In-Q-Tel is structured as an independently operated 501(c)(3) charitable entity. It receives grant money from the government and may have contractual obligations to its governmental partners, but the government is not a shareholder in In-Q-Tel or the companies In-Q-Tel invests in.

The Legal Authority to Take Stakes in Private Companies

Against this backdrop, what legal theories support government stakes in private companies such as Intel, MP Materials, and U.S. Steel?

This question should be broken into two parts: First, can a relevant statute, such as a grant program, contracting vehicle, or regulatory process, be read to authorize the government to take a stake as part of a deal? Second, is there a crosscutting prohibition on the government taking an equity stake that would prohibit the government from taking equity, even if a particular grant program, contracting statute, or regulatory authority by its own terms could be read to authorize the government doing so?

With respect to the first question, the government appears to be relying on the fact that while almost no statutes expressly authorize the government to take equity stakes in companies as part of a grant, contract, or regulatory process, the government can interpret broad language in a number of statutes as permitting the government to do so.

Take the U.S. government’s golden share in U.S. Steel, which reportedly gives the U.S. government a perpetual say in certain governance decisions, such as any attempt to redomicile outside the U.S., waive or reduce planned investments, or relocate jobs. The U.S. government, Nippon Steel, and U.S. Steel negotiated this golden share as part of the so-called Committee on Foreign Investment in the United States (CFIUS) review of Nippon Steel, a Japanese firm, acquiring U.S. Steel. Under Section 721 of the Defense Production Act (DPA), the government has a broad authority to review foreign acquisitions of U.S. companies for national security risks and to negotiate with companies to mitigate identified national security risks associated with a transaction. Agreements between the government and parties to mitigate these risks are commonly called “national security agreements” or “mitigation agreements.” The president has broad authority to block or suspend a transaction if the risks are not mitigated.

Historically, the government has structured mitigation agreements as enforceable contracts between the government and the parties. Parties might, for example, agree that the U.S. company, now owned by a foreign company, will not transfer sensitive technology outside of the U.S., or that it will keep U.S. citizen data inside the U.S. The government may also insist on vetting the board members of the U.S. firm. Yet while the government has historically implemented mitigation agreements via contract, the language of the statute is broad and does not appear to preclude the government asking for or receiving a golden share as part of a mitigation agreement. Specifically, the relevant statute provides that the CFIUS Committee may “negotiate, enter into or impose, and enforce any agreement or condition with any party to the covered transaction in order to mitigate any risk to the national security of the United States that arises as a result of the covered transaction.” While this language could not be read to require a golden share—a company could always choose to abandon the transaction rather than give a golden share to the government, much as a company could always abandon a transaction rather than agreeing to CFIUS mitigation terms—by its terms it does not preclude the government from asking for one or a company from offering one.

The government is likely taking a similarly broad read of the CHIPS Act—which Congress passed as part of an annual defense bill in early 2021 and funded in 2022 to provide incentives for manufacturers to make computer chips in the U.S.—as authorizing the government to take its 9.9 percent stake in Intel. Here, the specifics of the government’s actions in some ways appear to violate the spirit of the law, albeit they likely technically comply.

Under the CHIPS Act, the Commerce Department is specifically authorized to provide companies with funding to expand semiconductor manufacturing in the U.S. The statute makes clear that the federal support could include outright grants as well as loans, and also that the government can negotiate clawback provisions and other contractual provisions to protect the taxpayer if companies failed to perform what they agreed to under the grants. Under the Biden administration, the Commerce Department awarded Intel nearly $8 billion to build factories in the U.S. (The Defense Department awarded another $3 billion.)

During congressional debate over funding the CHIPS Act in 2022, Sen. Sanders and Sen. Elizabeth Warren (D-Mass.) proposed an amendment that would have required the Commerce Department to obtain a warrant or equity interest in CHIPS grant recipients, alongside other restrictions. While the Senate never adopted the amendment, the language of the CHIPS Act also does not explicitly prohibit the department from asking for or receiving equity as part of a grant agreement. Instead, it simply provides that the Commerce Department shall establish a program that “provides Federal financial assistance to covered entities [relevant semiconductor firms] to incentivize investment in facilities and equipment in the United States for semiconductor fabrication, assembly, testing, advanced packaging, or research and development” and then stipulates the kinds of companies that are eligible and criteria for selecting grantees (50 U.S.C. § 4652). The law appropriating funding for the grants sets limits on grant and loan amounts, but it, too, neither explicitly authorizes nor precludes taking equity.

Moreover, the law that funded the CHIPS Act expressly and broadly provided the Commerce Department with “other transaction authority” governing the way in which it can contract with grant recipients. This language provided that the department can “enter into and perform such contracts, including cooperative research and development arrangements and grants and cooperative agreements or other transactions, as may be necessary in the conduct of its work and on such terms as it may determine appropriate, in furtherance of the purposes of this Act” (15 U.S.C. § 272(b)(4)). Because the Commerce Department is authorized to provide “financial assistance,” using the “other transaction authority” would likely require interpreting an equity investment as “other financial assistance” under 2 C.F.R. 200.1, the regulation that generally defines “financial federal assistance.” It would be an expansive interpretation, but not an implausible one.

The Biden administration chose not to pursue warrants or equity stakes in the chipmakers in part because Congress appeared to intend the program for grants and loans—after all, the Senate did not adopt the Sanders-Warren amendment requiring it to take equity. There were also legal considerations related to the Government Corporation Control Act of 1945, which are discussed below. Most important, from a practical perspective, when the Commerce Department was establishing the program in 2022 and 2022, most chipmakers would have been unwilling to accept the money to manufacture semiconductors in the U.S. if they had to give the U.S. government equity. Instead, the chipmakers would likely have simply continued to manufacture chips in Asia, and the CHIPS Act would not have achieved its objectives.

That said, the Biden administration did try to negotiate provisions into CHIPS Act grant agreements stipulating that if U.S. semiconductor factories supported by the CHIPS Act proved more profitable than expected, the government would benefit. Specifically, the Commerce Department announced that it required companies receiving more than $150 million in CHIPS grants to commit to “upside sharing” in which the government would get a share of returns that exceeded projections.

Today, at least with respect to Intel, the Trump administration is confronted with a different practical reality. Intel faces profound technological issues and has reportedly considered a major strategic overhaul to break up its business. The company this year announced repeated delays of its semiconductor fabrication plant in Ohio and plans to lay off factory workers in Oregon and Arizona. While neither Intel nor the government has ever published the full details of Intel’s CHIPS award, it seems plausible that Intel’s troubles mean that the company could not meet commitments under the award. Pursuant to the CHIPS statute, noncompliance with a grant gives the government an opportunity to claw back the money.

Here is where the Trump administration’s deal with Intel appears to violate the spirit, if not the letter, of the law. A document Intel recently filed with the Securities and Exchange Commission indicates that, instead of trying to claw back CHIPS Act money for the delayed fabrication plants, the Commerce Department actually accelerated planned CHIPS Act payments to Intel. In addition, the Commerce Department agreed that “that to the maximum extent permissible under applicable law, the Company’s obligations pursuant to the [CHIPS grant] will be considered discharged” (except with respect to some of Intel’s obligations to the Defense Department). In short, under the specific terms of the deal, Intel got CHIPS Act money quicker and without actually having to meet the full extent of its prior U.S. manufacturing obligations. In exchange for the accelerated payment and being relieved of these obligations, Intel’s board awarded the government its 9.9 percent stake.

Relieving Intel of prior obligations while accelerating payments appears contrary to the spirit of the CHIPS Act’s requirement that funding be used “to incentivize investment in facilities and equipment in the United States for semiconductor fabrication.” In essence, the government is getting equity in the company rather than the U.S. being guaranteed additional new fabrication plants. But the Trump administration can likely argue that language in the CHIPS Act is broad enough, and Intel has done just enough in terms of fabrication plant construction (between breaking ground in Ohio and completing the planned expansion in New Mexico), that the deal complies with the letter of the law. It is also of course possible that the government has secured private commitments from Intel to continue pursuing the fabs, but if so, those commitments are not public.

Finally, there is the Defense Department’s deal with mining company MP Materials, in which the government has taken stock and warrants that will give the government 15 percent of the company’s shares as part of a deal to support MP’s plans to mine rare earth elements in the U.S. and to process them into industrial magnets. In addition to the U.S. government taking an equity stake in the company, the Defense Department agreed to set a “price floor” for MP’s production of certain rare earths, as well as to acquire magnets that the company plans to manufacture.

The MP Materials deal appears to be executed pursuant to Title III of the Defense Production Act of 1950, which is intended to allow the government to take provisions to expand the country’s productive capacity. Title III authorizes the president to “make provision … for purchases of or commitments to purchase an industrial resource or a critical technology item, for Government use or resale; [and/or] … for the development of production capabilities” (50 U.S.C. § 4533(a)(1)(A&C). Moreover, the president may do so “without regard to the limitations of existing law,” except that the government cannot spend unappropriated money (50 U.S.C. §  4533(b)). While this language, much like the CHIPS Act language, does not expressly authorize the government to take equity stakes, by its own terms it is very broad and does not preclude it if a company agrees. The “without regard” language can also be read to waive statutory restrictions that might exist in other parts of the U.S. Code, such as limitations on some specific procurement authorities.

Finally, there is one clear standing authority for the U.S. government to take equity stakes in private companies. In 2018, when Congress established the U.S. Development Finance Corporate (DFC) to promote global economic development, it specifically authorized the DFC to make equity investment in companies, so long as the DFC was a minority investor. The BUILD Act specifically provided that the DFC “may, as a minority investor, support projects with funds or use other mechanisms for the purpose of purchasing, and may make and fund commitments to purchase, invest in, make pledges in respect of, or otherwise acquire, equity or quasi-equity securities or shares or financial interests of any entity, including as a limited partner or other investor in investment funds, upon such terms and conditions as the Corporation may determine.” This is extremely broad authority for the DFC, which is up for renewal this year, to purchase minority stakes in companies—or, arguably, pursuant to the “otherwise acquire” language, to accept stakes in companies provided as part of actions a company might agree to with another government agency.

Are There Any Limitations on the Government’s Authority to Take Equity?

This is not to say that the government’s ability to take equity in U.S. companies is unbounded. If the government plans to spend money for the equity, money must be appropriated, and there must be a contract vehicle for the government to do so.

There is also the second question of the analysis: Is there a crosscutting statute that prohibits the government from taking equity stakes in companies?

Here, the only significant potential issue is the Government Corporation Control Act of 1945 (GCCA)—which Congress enacted and has updated to regulate the governance of government companies such as Fannie Mae, Freddie Mac, Amtrak, the Export-Import Bank, and the TVA, among others—and might be read to impose limits on acquiring majority interests in U.S. companies. The GCCA stipulates that “[a]n agency may establish or acquire a corporation to act as an agency only by or under a law of the United States specifically authorizing the action” (30 U.S.C. § 9102). Several Justice Department memos from the 1990s—one of which has been released—as well as a General Accounting Office (GAO) letter regarding a company the Federal Communications Commission established in 1997 offer an executive branch interpretation of the act. The executive branch appears to have viewed the GCCA as potentially prohibiting government agencies from acquiring equity stakes in companies—and certainly controlling equity stakes in companies—in violation of the GCCA. But the one case that addressed the scope of the GCCA appeared to take a more lenient approach, suggesting that the multifactor test the Supreme Court used in 1995 to determine whether Amtrak is a government instrumentality for the purposes of the First Amendment would apply to the GCCA. Against that backdrop, the government’s stake in Intel will be a minority stake and the government will not be a controlling shareholder. Moreover, with respect to the GCCA, the government would almost certainly argue that it is not acquiring Intel to “act as an agency.”

Furthermore, the government could try to stack its legal tools in a bid to skirt the GCCA. Defense Production Act Title III, for example, authorizes the government to enter into purchases and purchase commitments “without regard” to the limitations of existing law. Should the Defense Department attempt to take equity stakes in U.S. defense contractors—something Commerce Secretary Howard Lutnick recently publicly suggested—the department could potentially use its DPA Title III authority to dodge the GCCA. (In order to do this, the president would have to follow the procedural requirements of the DPA.)

There could also be arguments that the Supreme Court’s emerging “major questions doctrine” forbids the government from taking equity stakes without clear authorization from Congress. In recent years, the Supreme Court has held on several occasions that Congress must “speak clearly” to authorize the executive branch to take actions of major economic or policy significance, holding, for example, that “[w]e expect Congress to speak clearly if it wishes to assign to an agency decisions of vast ‘economic and political significance’” (Utility Air Regulatory Group v. EPA). A company seeking to challenge a government stake could argue that, against a backdrop of the government generally not taking equity stakes in the private sector, the government should not be allowed to read broad appropriations and regulatory authorities as today authorizing the government to do so.

But this raises the government’s potential ace in the hole: Who will sue? Companies that agree to the government taking stakes will presumably not challenge in court the legality of the government taking those stakes. Potentially, a company denied a grant or a license because it refused to give the government a stake might bring suit down the road, a competitor of a new American state-owned enterprise might sue, or a shareholder might be able to bring a challenge against a company if his or her stake was diluted by the government. But it may be some time before any such suits are brought, and they would almost certainly face procedural hurdles, as well as having to overcome substantive arguments in favor of the legality of the government’s actions. 

Implications for the Future

To suggest that the government has colorable legal authority to take stakes in private companies—far more frequently and on a far greater scale than it has in the past—is not to suggest that the government should regularly do so. Greater government stakes in the private sector raise profound policy questions: What are the limits on the government’s ability to demand equity in exchange for funding or regulatory approvals? Will the government favor firms in which it owns stakes over other competitors that might have better technology or processes? Will the government impose new political mandates on American state-owned enterprises? The Chinese model of state capitalism for sectors such as electric vehicles typically encourages multiple firms to fight for competitive advantage to ensure that more innovative ones weed out weaker competitors. Will the U.S. do the same, or will it simply back well-connected firms in which it owns interests, regardless of how noncompetitive they would be in a free market? There are important governance questions about how the government will treat other shareholders in mixed publicly owned and privately owned corporations.

But from a legal perspective, the executive branch may well be able to act unless Congress decides to intervene. There are a couple of steps Congress could take.

First, Congress could include language in appropriations bills forbidding the government from using any funds appropriated by the bill from being used to purchase stakes in companies, or from making the distribution of any funding contingent on a company granting the government equity, unless equity is specifically authorized by law. This would ensure that the government could not use funding to take equity unless Congress specifically gave its consent to do so.

Second, with the Defense Production Act up for reauthorization this year, Congress could express views on whether and under what circumstances it views equity stakes as appropriate. More broadly, Congress could enact legislation clarifying if and under what circumstances the government could request or accept equity as part of regulatory approvals.

If Congress wants the government to expand its role in the private sector, it could also adopt more general legislation—a modern-day Government Corporation Control Act—stipulating how and under what circumstances the government could take equity, and setting out how such equity should be governed.

The bottom line is that if Congress wants greater oversight of this new age of American state capitalism, it will need to act to assert its authority—rather than waiting for the courts.