Last week, the Center for Naval Analysis hosted an event titled “American Grand Strategy and Seapower.” I participated in a panel discussing “Seapower and the ‘Global System,’” along with Michael Mandelbam of John Hopkins SAIS and James Wirtz of the Naval Postgraduate School. Here is a brief synopsis of my main argument.

It has become commonplace in Washington for people to believe that the U.S. military is, and shall forever be, the world’s policeman. Polls show that a majority of Americans aren’t interested in playing this role. And this shouldn’t come as much of a surprise. As Michael Mandelbaum admits in The Case for Goliath, “to make sacrifices largely for the benefit of others counts as charity, and for Americans, as for other people, charity begins at home.”

In this view, we maintain a vast military not so much because we are threatened, per se, but rather out of the fear that others might be threatened, be denied access to global markets, or cut off from the benefits of democracy and liberal governance.

Our willingness to police the globe amounts to a form of foreign aid. Indeed our wealthy, well-governed allies have diverted their attention—and their resources—to other nice-to-have things: short work weeks, long vacations, shiny mass-transit systems, and generous health and welfare benefits.

I believe, by contrast, that many states that are currently U.S. security clients will take responsibility for securing themselves if we quit allowing them to free ride on American military promiscuity.

As a consequence of American preponderance, many countries do not see a need for power of their own. We have created dependency. America bears a larger and larger share of the burden of security in every one of its alliances. We soothe ourselves by saying that we are not paying others’ checks—we are providing “global public goods.” In fact, we are doing no such thing. The problem with “global public goods” is that they are neither global, nor public goods.

To review, public goods have two characteristics. First, once provided, their benefits cannot be denied to those for whom the original provision was not intended. Economists refer to this as nonexcludability. The other crucial feature of public goods, nonrivalrous consumption, holds that the value of the good is not diminished as additional consumers partake of it.

Consider how this works in the context America’s security alliances. The allocation of American assets to one ally necessarily diminishes the amount that can be used by another ally. Therefore, America’s provision of security via its system of alliances is rivalrous between allies.

Further, there is no global security alliance. Frequently one hears American maintenance of the freedom of the seas referred to as a global public good, but Washington is absolutely capable of excluding various states from the benefits of its actions. Alliances are collective goods for alliance members, but club goods globally, because their benefits are denied to nonmembers. Thus, none of the so-called global public goods provides benefits that are genuinely both nonrivalrous and nonexcludable, and therefore none meets the definition of a global public good.

Misapplying a theory developed in one discipline to another discipline leads to conceptual fuzziness. Clearing up these concepts is the first step toward a sensible reappraisal of American foreign policy for the twenty-first century.