National Security Regulation and the Decline of Cost-Benefit Analysis

A regular series on the choices faced by international economic policymakers

Under both the Donald Trump and Joe Biden administrations, the U.S. government has had a seemingly unquenchable thirst for national security regulations limiting investment, trade, and information flows. Rare bipartisan policy alignment has yielded a growing array of restrictions premised on concerns about China, Russia, terrorism, and other international sources of angst.

The restrictions have been accompanied by a worrisome trend: the government has given short shrift to cost-benefit analysis. While available data is sparse, preliminary indications suggest some restrictions could, on balance, harm rather than advance U.S. security.

The U.S. government should facilitate data gathering about the costs and benefits of those proliferating restrictions, as it does in connection with regulation on the environment, labor, and health. Decisions about whether to implement or maintain national security restrictions should be informed by that data.

Cost-Benefit Analysis, Except For National Security

For decades, the U.S. government generally has required analysis that benefits outweigh costs when issuing new regulations. For example, raising vehicle emission standards can make cars more expensive, but those costs should be outweighed by the benefits of reduced emissions, such as fewer costly doctor visits because of pollution-related health problems.

Recently, however, when the phrase “national security” is uttered, the U.S. government has not appeared to engage in a serious weighing of the costs and benefits. Regardless of whether the benefits appear marginal or the costs dramatic (or both), the government generally concludes—often without examination—that the benefits override any cost concerns. 

This summer, for example, the U.S. government proposed outbound investment rules to restrict contributions to companies working on semiconductors, quantum computing, and artificial intelligence if those companies have certain links to China. Many of these soon-to-be-restricted companies operate in the United States: an artificial intelligence company in San Francisco could be restricted if it has significant Chinese ownership. Further, because artificial intelligence is prominent among technology companies of many types, and links to China are not self-evident, it will be difficult to discern which companies are restricted under the new rules.

So, potential U.S. investors, lenders, and other commercial enterprises engaged in ostensibly domestic transactions will need to hire lawyers to confirm that the restrictions do not apply to their potential business partners. They will repeat this exercise for thousands of transactions annually, resulting in higher costs of capital for virtually all U.S. technology companies.

Yet it does not appear that the intended impediments to Chinese advancements will outweigh aggregate costs. The restrictions apply to U.S. businesses but not businesses anywhere else in the world. Because U.S. dollars can be replaced by euros, yen, or money flowing to and from other countries, it seems doubtful that U.S. restrictions will impede Chinese advancements. Rather, U.S. parties will likely be replaced by foreign investors, lenders, and other partners.

In a prefatory statement to explain the benefits of the new rule, the government argues “the national security benefits, while qualitative, will outweigh the compliance costs of the proposed rule.” However, an analysis of the potential reduction in money flow to the Chinese sectors targeted by the new rules is entirely missing. Any serious attempt to compare costs and benefits would address the degree to which that goal (i.e., reduction in money flows to China-linked companies) could be achieved.

This example is merely illustrative. The flood of national security regulations is unabated. A multitude of regulations are pending to ratchet up government scrutiny of any foreign investment into U.S. companies, regardless of the investor’s nationality; those rules already implicate thousands of transactions annually, most involving investors from Europe and other allies. Restrictions on the export of U.S. products and technology have expanded significantly, and there has been a dramatic increase in U.S. government sanctions: the lists of sanctioned parties now collectively include more than fifteen thousand names. But cost-benefit analysis has been absent from virtually all those and other new national security regulations.

Data Is Sparse but Ominous

Business lobbyists have been surprisingly quiet amidst this wave of regulation. The government’s application of the national security restrictions—say, a prohibition on an investment—often occurs under the cloak of national security secrecy. Many actions taken by the government are not even subject to court review and leave no public trace.

With each business knowing only its own hardships, it is difficult to form a coalition to push back against those national security regulations, particularly when advocacy runs the risk of being deemed unpatriotic.

Preliminary data, though, suggests that the U.S. government is often not making sensible decisions. Earlier this year, several economists, including staff of the U.S. Federal Reserve Board, estimated that “export controls cost the average affected U.S. supplier $857 million in lost market capitalization, with total losses across all the suppliers of $130 billion.”

These controls are significantly harmful to U.S. businesses developing the technologies that the U.S. government deems essential to national security. Moreover, as the authors suggest, China has been able to boost domestic innovation and reliance on non-U.S. firms so that the U.S. export controls may not materially impede Chinese progress. In other words, the export controls seem to produce high U.S. costs and potentially illusory benefits.

Other restrictions could similarly lack reasonable cost-benefit justifications. Take the Committee on Foreign Investment in the United States (CFIUS), which reviews foreign investments into domestic companies. Other countries, generally following CFIUS’s lead, maintain similar systems (e.g., Dutch regulators review investments into Dutch companies). A new study estimates that those investment screening restrictions reduce dealmaking by more than 10 percent. The authors conclude by exhorting policymakers to “weigh the economic costs of screening against the security benefits.”

Another recent study from the French central bank has found that U.S. sanctions on Russia have reduced global reliance on the U.S. dollar as the currency for international transactions. Because large international transactions generally have necessitated access to U.S. financial institutions and dollars, U.S. government sanctions that remove this access can isolate sanctioned parties. Overuse of sanctions, however, threatens to undermine those levers: a global shift away from dollars and U.S. financial institutions could reduce the power of U.S. sanctions. Yet the government continues ratcheting up the number of sanctioned parties, from the Cuban nation to Russian oligarchs to drug traffickers. On the benefits side of the ledger, with a few notable exceptions (generally involving broad multilateral sanctions, such as those against apartheid South Africa), sanctions generally have not produced changes in behavior sought by the U.S. government. Cuba, for example, has not jettisoned its communist leadership after more than sixty-two years of U.S. sanctions.

Willful Blindness

A central problem with ignoring the relative costs and benefits of national security regulation is that U.S. commerce and entrepreneurship suffer, potentially undermining the goal of staying ahead of competitor nations in a technology and security race.

The U.S. government should focus on collecting data to assess whether, for example, there are some export controls that harm U.S. industry more than they impede foreign competitors. Early evidence suggests at least some export controls could be overkill.

Similarly, the U.S. government should ask whether there are investment screening rules that generate significant costs—for example, legal fees or opportunity costs from deals that are deterred or reduced in size—without commensurate benefits. As I have argued elsewhere, the CFIUS mandatory filing rules, which necessitate lawyerly reviews of thousands of transactions each year (mostly involving investors from Europe and other U.S. allies), have no cost-benefit justification.

The U.S. government has not facilitated much cost-benefit data gathering in connection with national security regulation. Without available data, it is difficult to determine whether proliferating national security restrictions are sensible or counterproductive. Regardless, one point seems clear: the government should not be wittingly blind to the relative costs and benefits of these rules.

As for sanctions, and the possibility that overuse of this tool could undermine the centrality of U.S. financial institutions, President Barack Obama’s deputy national security advisor, Ben Rhodes, recently commented that “we don’t think about the collateral damage of sanctions the same way we think about the collateral damage of war. . . . But we should.”

*Stephen Heifetz is a former U.S. government national security official and currently a partner at the law firm Wilson Sonsini Goodrich & Rosati. The views expressed do not necessarily represent the views of the firm or any of its clients.