The most surprising aspect of the saga of American investment in Chinese companies is an unwillingness among many policymakers to even find out what’s going on. Though such investment had surged in 2017, a spike in 2019 prompted the first concerns that the departure to China of large amounts of American money was inconsistent with President Donald Trump’s goals. In 2021, the Biden administration promised action. That action was finally announced in 2023, but implementation was kicked down the road.

This year, bills intended to restrict American investment in China are again being circulated in Congress, their prospects unknown because of uncertainty about President Trump’s second-term goals. Restrictions on American support of Chinese companies are long overdue, as is much greater transparency on where exactly this money is going. They were already overdue in 2021, and they might still be overdue in 2029. The U.S. blocks inbound investment to protect advanced technology, and it restricts exports of advanced technology. Yet money to support such technology development in our biggest rival is not even monitored.

There are two types of investment: direct investments, made by companies — for example, in a new factory — and securities (or “portfolio”) investments made by companies but also by other stock and bond investors. Direct investment is more valuable, dollar for dollar, but recent changes in U.S. portfolio investment in the People’s Republic of China have been far larger.

At the end of 2023, according to the latest figures available, the stock of U.S. direct investment in mainland China and Hong Kong was $218 billion. From 2017 to 2023, the largest annual change in U.S. direct investment in China was an $18 billion increase. At the end of 2023, the stock of U.S. portfolio investment in China and Hong Kong was $798 billion. More important, the largest change was a $386 billion increase. (There was also, during the 2017–23 period, a $258 billion one-year decline.) A great deal of money can move between the U.S. and the PRC when securities investment conditions change.

The Department of the Treasury tells us almost nothing useful about it. Treasury claims that the Cayman Islands, which has no securities markets, is the dominant recipient of American securities investment, at more than $2.6 trillion. This is absurd. For the past few years, an annual paper published by the Federal Reserve documents investments by the true national origin, not legal residence, of the companies involved. The Federal Reserve’s number for investment in the Caymans is essentially zero, as it should be. Moreover, the direct investment data include subtotals of U.S. investment in a few industries, such as chemicals. The portfolio data, however, offer no similar breakdown by industry.

Thanks to the Fed, therefore, we have a reasonable total estimate of U.S. investment in the PRC, but we have no idea what exactly was supported by the nearly $800 billion in additional American portfolio investment in Trump’s first term. Year after year, Treasury makes clear that it does not want the public to know too much about American money overseas. The implementation regulations for the Biden administration’s executive order on outbound investment were published a week before the 2024 election. The Biden Treasury cited PitchBook, a private market-data service, regarding U.S. investment in advanced technology overseas and admitted that further detail “cannot be determined with precision.” If only there was a department with the authority to fix this. At that point, Treasury could not have dismissed this as an unimportant political matter; it’s much more likely the data were not gathered because they would have been embarrassing.

This is a bipartisan failure — Janet Yellen’s most recently and Steven Mnuchin’s before that. It’s also an extended bipartisan failure for Congress. When a review of inbound investment reform passed Congress overwhelmingly in summer 2018, outbound investment was considered and set aside. Congress has since been unable to pass even basic disclosure requirements for, much less restrictions on, outbound investment. It all may sound mysterious, but the same phenomenon has long been typical of American policy on China broadly. On one side are all the concerns about the PRC that politicians raise publicly; on the other is the potential for their donors to make money through foreign investment, trade, or other activity. The need to win the competition against China continues to control political speech; making money by helping China continues to control political (in)action.

This failure of oversight on outbound investment is long-standing, but the risks have increased since President Trump’s second election victory. This may have less to do with the president himself than with his advisers and with how the financial community sees his presidency. The average annual change in the stock of American portfolio investment in the PRC during Trump’s first term was a $213 billion increase, including a $348 billion rise in 2020, during Covid. No actions were taken by the first Trump administration to check or even monitor these flows. Wall Street’s grotesque conclusion that causing a global pandemic made China more attractive as an investment destination was complemented by Secretary Mnuchin’s refusal to act against China with respect to outbound U.S. investment or anything else.

Mnuchin’s successor, Secretary Yellen, has long been known as a friend of China’s. It took her Treasury two years to prepare an executive order on outbound investment in China, something that could have taken two months, and a remarkable 14 more months to implement it. With all this time, Treasury was still unable or unwilling to provide a figure for how much American investment would actually be restricted. Perhaps executive orders should be permitted only when the executive can decently explain their anticipated impact. But Yellen had an excuse Mnuchin did not: the average annual change in the stock of American portfolio investment in the PRC from 2021 to 2023 was a $192 billion decline. Since the Biden administration did not act until late 2024, this is certainly not a result of its (almost nonexistent) policies. Instead, the decline was due to private investment decisions, most likely influenced by China’s intensified Covid-related repression.

In 2025 or 2026, with Covid farther in the rearview mirror, will BlackRock and other financial houses change their minds and again help the PRC develop technologies that may threaten the U.S. economically or militarily? American brokerages boast that they are better at identifying high-value Chinese companies than are local Chinese state-owned banks focused on fulfilling PRC government mandates. These boasts are very likely justified. But it’s not in our national interest that American money be allowed to find and fund the next DeepSeek.

What can prevent it? Transparency and restrictions. Good policy is impossible without good information, and the U.S. is nowhere near having good information on outbound portfolio investment. The failure has persisted through so much debate that it appears deliberate. Markets work better with more information; it’s a legitimate government function to provide it. If defenders of American investment in China are indeed defending markets, they should have no objection to far greater transparency.

First, the embarrassing pretense that the Caymans receives trillions in American securities investment should be completely and permanently scrapped. Treasury should stop publishing nonsense. Next, reporting of the true national recipients of outbound U.S. investment should be done quarterly instead of annually, to more quickly identify large movements such as those seen since 2017 and to evaluate responses to any new policy.

More important, the broad industry-level breakdowns by which direct investment can be reported should be used to report portfolio investment too. This would entail figures for “computers and electronic products” as well as information and services. If it’s discovered that portfolio flows in potentially sensitive areas, such as advanced computing techniques, are large or rising, more detail about them could be mandated. If Congress and the administration won’t do even this, it will be a clear sign that China policy remains unserious.

With regard to restricting investment in the PRC, the House Select Committee on the Chinese Communist Party did good work in the last Congress by identifying specific technologies that should not receive U.S. funding. That work was not made public, because of clashes with a few well-placed congressional Republicans, but this Congress has a foundation to build on, as long as it’s willing. The idea is to define technologies, from artificial intelligence to biology, that could lead to China’s developing capabilities beyond what’s commercially available. The U.S. should not be supporting this in any way — be it with products, knowledge, or capital — and the list of prohibited technologies should be updated periodically.

In this case, as in all others, legislation is superior to executive action. Restrictions on outbound U.S. investment to China would be better enshrined in law. The legislative process would also allow the standards instituted to be bipartisan and therefore more durable. The last-minute, half-hearted Biden order should be replaced by (belated) congressional action to greatly enhance transparency and prevent the American financing of advanced technology in China. And the Department of the Treasury should implement such legislation quickly and rigorously. Yes, the Chinese will make some advances anyway. We should, however, finally stop helping them.