There is an aphorism in the United States that “you can’t be kinda pregnant.” The idea, of course, is that there is no halfway — either a woman is pregnant or she isn’t. America’s China policy, and its recent decision to upgrade Taiwanese F‑16s, reflect widespread “kinda pregnant” thinking in Washington about China policy. Nothing good is likely to result from this over the long term.

For decades, America’s China policy has been an incoherent stew of what a RAND Corporation report [.pdf] termed “congagement” — that is, part containment, part engagement. On the one hand, Washington has posed itself as the hub of a “hub and spokes” series of bilateral alliances in Asia and liberally sprinkled U.S. military assets throughout the region. On the other hand, it has traded and engaged economically with China, helping fuel its remarkable economic growth and by extension, its military power.

The “containment” aspect of Washington’s China policy has involved building alliances and quasi-alliances throughout Asia, including those with South Korea, Japan, India, Taiwan, and a number of others. The dilemma here is that when Washington commits itself to the defense of these clients, the rational thing for these governments to do is shirk their responsibility for their own defense. As Mancur Olson and Richard Zeckhauser pointed out in their article “An Economic Theory of Alliances,” when a number of units band together to provide a collective good like security within an alliance, it is rational for the smaller, weaker members of the alliance to shirk any responsibility to pay their fair share of the good as long as the large, wealthy member is willing to pay it for them.

This is the crucial flaw of this approach. When Washington continually reassures its clients of its commitment to their security any time a problem arises, such as in the context of the recent disputes in the South China Sea, that has the effect of discouraging them from paying for their own defense. The Economist magazine recently fretted that were Washington to distance itself from Taipei, the result may be that “the region’s democracies [would] worry that America might be willing to let them swing too.” But these worries would in turn cause governments to take on more responsible military postures in the region. This phenomenon should be seen as a feature, not a bug.

For its part, Taiwan’s contribution to its own defense has shriveled as a percentage of both its GDP and its share of US-Taiwan military spending. America’s tight embrace seems to have led Taiwan’s elite to conclude that they need not waste resources on security matters, since they have an implicit commitment from Washington. This is one example of how Washington’s infantilization of its allies in Asia has led to atrophy among partners who could be carrying a much larger share of the cost of constraining any Chinese proclivity for adventurism in the region.

The recent arms deal with Taiwan demonstrates that the current approach leaves Washington with the worst of both worlds. In tuning the dials on its provision of “defensive” military materiel to Taiwan, the Obama administration decided to upgrade Taiwan’s existing F‑16s rather than providing C/D versions. This decision has already reaped a diplomatic whirlwind, with China’s foreign minister Yang Jiechi complaining that Washington had “grossly interfered in China’s internal affairs and seriously undermined China’s security,” and reportedly promising to sever some military-to-military engagements.

Moreover, there is little reason to believe that A/B upgrades or even C/D new planes would substantially improve Taiwan’s defense capabilities against China should a military conflict erupt. One recent study indicated that, should it want to, China could disable all of Taiwan’s airbases with missile strikes, keeping airplanes — of whatever type — on the ground and therefore useless. In order to militarily deter any Chinese attack, what would be needed in Taipei is a wholesale transformation of spending priorities and mass opinion. To put it mildly, this does not appear to be in the cards.

A far better, more prudent policy would be to offload the front-line responsibility for regional defense to America’s Asian allies (and quasi-allies like Taiwan) by severing any implicit or explicit security commitment. Washington should sell arms to its friends in the region without committing itself to any individual power’s defense. China would no doubt complain about arms sales, but Washington would no longer be on the hook as the balancer of first resort, which it is today.

Our present policy has three effects. First, by trading with China we make it more difficult to militarily contain her. Second, by constantly reassuring our allies of a firm U.S. commitment to their security, we make them less likely to spend their own resources to ensure their own security. These two outcomes interact to produce a third effect, which is a balance of power in Asia that will inexorably shift toward China unless its economic growth slows dramatically or else Washington devotes ever more of its own resources to militarily constrain China. Surely the world’s only superpower can come up with something better.