On the heels of the biannual summit of NATO leaders, there has been a lot of talk about the return on investment the United States gets from the transatlantic alliance. Prior to the summit, President Donald Trump once again questioned its value—returning to themes from his campaign about NATO members failing to “pay their bills,” among other complaints that more demonstrate his ignorance of how the alliance actually works. As the summit got underway, his complaints about European allies not pulling their weight reached absurd levels. And following the meeting with leaders of the alliance, and his one-on-one meeting with Russian President Vladimir Putin, the president expressed skepticism about the very premise of NATO in an interview with Fox News host Tucker Carlson.

The president, however, is not necessarily alone in believing America’s defense of Europe is not worth the cost. While many Beltway foreign policy analysts are quick to defend the alliance as representing a vital interest of the United Statesand to sometimes lob accusations of isolationism at those who disagreeothers make legitimate claims that the United States does not need NATO to defend itself and therefore American taxpayers are getting a raw deal as rich European countries free ride on American security guarantees. Common ground remains elusive in these debates because the problem is not a matter of objective analysis, but rather a value judgment.

Critiques of NATO are often—though not exclusively—the province of academics in the realist school of international relations theory, as well as a number of allied analysts in the think tank community. Realists see a state’s security as its primary foreign policy goal. From this perspective, allies are supposed to be additive in terms of increasing the state’s military capabilities to balance against threats to their security.

But the United States is very secure already. It is not in any danger of invasion, nor is a hegemon likely to dominate Europe any time soon. Therefore, the United States gains little from NATO in the absence of the possibility that the Red Army might pour through the Fulda Gap. Instead, it spends large sums of money defending allies that do not directly contribute to its security. Since it is safe, Washington can reduce the size of its military forces and its defense spending. If threats reemerge, former allies such as France, Germany, the United Kingdom, or Italy will either spend their own money or ally with one another to defend against it. If they fail to do so, the United States will then build up its own military forces to tip the balance against a would-be hegemon.

Realists acknowledge other foreign policy aims, but many who seek to end America’s involvement in NATO either assume that these aims can be achieved autonomously or through non-military instruments. They believe military instruments should be used solely for security. As Barry Posen, a realist political scientist and critic of America’s role in NATO, wrote as part of a roundtable on his book outlining a realist grand strategy he calls “restraint,”

The explicit purpose of the book is to develop a grand strategy based on a strictly construed assessment of U.S. security interests, or what once might have been termed ‘vital interests.’  These interests are the minimization of threats to U.S. sovereignty, safety, territorial integrity, and the power position necessary to ensure against these threats. I wished to demonstrate that on the basis of such an approach, one could build a relatively coherent, effective, and efficient grand strategy and accompanying military strategy and force structure that would involve the U.S. in fewer costly and indecisive wars.  The present grand strategy of the U.S. includes many other elements–supporters argue that trade and the spread of democracy are so important that they merit a significant military effort… My purpose was to show that U.S. national security can be protected without adding these objectives. Those who wish to employ military power to pursue them should demonstrate their importance to U.S. security, or they should admit forthrightly that they believe them worthy of vast expenditures of treasure and sometimes blood because of their inherent value, not because of their contribution to national security.

But many people, including many realists, do believe things like trade and democracy matter. Nor can security be analyzed in isolation from them. Alliances, especially NATO, can produce benefits for non-security reasons—some of which, in turn, produce greater security benefits. It does so both directly and indirectly.

Indirectly, NATO allows European countries to focus on commerce and non-security goals rather than security. The continent has largely “desecuritized” at this point. In this sense, that America’s European allies are “free riders” is a feature, not a bug of the alliance. As NATO’s first secretary general, Lord Hastings Ismay, supposedly said, the purpose of the alliance was “to keep the Soviet Union out, the Americans in, and the Germans down.” That last part of the equation is key since the lack of a German threat—which was still very much on the minds of Europeans when NATO was founded in 1949—means other European countries who might otherwise arm themselves to defend against it are now more likely to focus on non-security priorities, such as economic interdependence.

However, NATO also provides non-security benefits directly. For example, political scientist Marina Henke demonstrates that American alliances are constructed in a way to achieve benefits beyond increased security. She writes,

The breadth of institutions matters because it allows officials to gather information on the potential coalition partner’s deployment preferences beyond straightforward security considerations—such as what kind of economic and political considerations affect its willingness to join the coalition. Diplomatic embeddedness also helps American officials identify linkages between military and non-military interests. This facilitates the construction of side-payments.

NATO accession also requires political and economic reforms by potential member states. That a greater number of European countries are more liberal, democratic, and economically open can be seen both as an end unto itself and even a means to increased security if economic openness decreases the likelihood of war. But the institutionalized cooperation embedded in alliance membership can also facilitate multilateral efforts to address transnational problems, such as climate change.

There are legitimate questions about the alliance’s success in some of these areas. Nationalist populist movements across the continent have gained steam in recent years. Brexit is producing a messy divorce between Great Britain and the European Union. Democratic backsliding is evident in Poland and Victor Orbán’s Hungary, while Recep Tayyip Erdoğan consolidates authoritarian rule in Turkey. The shared values of democracy and the rule of law that were enshrined in the alliance’s founding treaty do seem to be waning to a certain degree.

Risk tolerance also plays a role in how one views the value of the alliance. Many scholars believe NATO is not worthwhile because it encourages what Posen calls “reckless driving” by allies that might draw the United States into a war. If you believe the chances of such entanglement are high, then the non-security benefits of NATO might not mean as much. Some scholars have shown that fears of entanglement are exaggerated or can be mitigated by structuring alliance agreements to reduce the risk.

Other risks are more difficult to mitigate. For example, NATO’s expansion in the 1990s and 2000s to include former Warsaw Pact members and Soviet possessions was done at a time of low military threats and thus the requirements for defending new members were not thoroughly considered. American leaders did so without any sense that they might have to honor those commitments through the expenditure of American blood and treasure. Resurgent Russian aggression should put that illusion to rest and caution Western leaders when they seek to expand the alliance.

Yet not every potential new member of NATO is created equally. Some create more cost and risk than others. The admission of the Baltic states was misguided not only—or even primarily—because it antagonized Russia, but rather it did so while creating new security liabilities for the United States. Such commitments will be particularly difficult for the U.S. military to defend should it be needed. Similarly, potential offers of NATO membership to Ukraine and Georgia would surely increase the chances of war with Russia—or, at the very least, would drastically increase the requirements of deterring Moscow without providing commensurate benefits.

Yet the admission of Montenegro—about which much ado was made last year—and, more recently, the possibility of including Macedonia in the alliance, have no similar dynamics. Neither is at risk of a Russian invasion, they do not necessarily encroach on what Moscow perceives as its “near abroad,” and therefore their inclusion in the alliance does not add to America’s defense burden.

There is no “right” answer here. Throughout this post I have highlighted a number of reasons why I think NATO is worth the price. As persuasive as I think this case is, it is entirely reasonable to come to the opposite conclusion if you believe that the purpose of a military alliance is—or should be—entirely about additive military capabilities and that NATO is simply not worth the price given America’s inherent security. Others have made that case, and even if I am no longer persuaded by it (as, admittedly, I once was), they often make it well. But common ground on issues like further NATO expansion or how to address imbalances in burden sharing among alliance members will remain elusive until reasonable people on both sides accept the legitimacy of the other’s position.

The value of the Atlantic alliance is ultimately a matter of judgment about the purpose of alliances, the ability to achieve the non-security benefits NATO provides, and the risks alliance membership or withdrawal might create. There are legitimate differences of opinion on these issues. Accepting that is necessary for any debate over its future.