President Donald Trump has repeatedly complained that the United States carries too much of the economic and military burden in NATO. He has even gone so far as to call the European alliance “obsolete” and to suggest that his administration might not fulfill the treaty’s Article 5 obligation that commits NATO countries to come to the defense of any member that is attacked (Note: administration officials have repeatedly sought to reassure NATO allies that we remain committed to the collective defense of Europe, and Trump has contradicted himself on this score).

Many think this provocative rhetoric is just a ploy to get our NATO allies, who habitually underspend on defense and free-ride on America’s security guarantees, to pay more of their fair share of the burden. At the Washington Post’s Monkey Cage blog, Andrea Gilli argues this approach is unlikely to jolt NATO allies into spending more on defense, though. Among other reasons, most NATO allies “face financial and political constraints to increasing military expenditure” in part because U.S. security assurances “have freed up state funds in Europe for other priorities, including a robust system of social services.” And since cutting welfare benefits is typically a political non-starter, we shouldn’t necessarily expect NATO countries to boost defense spending due to Trump’s abrasive rhetoric.

But the historical record seems to contradict Gilli’s argument. According to the RAND Corporation, Europe has historically spent between 43 percent and 78 percent of U.S. spending on defense. The ratio reached its peak in 1980, and then again in 2000 — years that were at the tail end of periods of defense budget cuts. And according to the RAND report, one of the the most successful techniques in getting NATO allies to share more of the burden was “threats by Congress to withdraw its troops from Europe.”

The only period of signficant real growth in European defense spending was during the 1970s; otherwise European defense expenditure has been remarkably flat in real terms…

Historically, efforts to create incentives or to manage the burden-sharing problem have taken four different approaches. The first approach (1966 to the mid-1980s) was based on the threat of U.S. troop withdrawals. With a series of resolutions and amendments from 1966 to 1975, Senator Mike Mansfield sought to use the threat of U.S. troop withdrawals to force Europe to contribute more and to lessen U.S. costs. As noted, that effort—plus other factors relating to economic growth and the Soviet threat—may have had a positive effect: European defense spending grew by 44 percent between 1970 and 1984.

Certainly other factors contributed to this period of growth in NATO burden sharing — higher rates of economic growth, increased perceptions of the Soviet threat, defense budget cuts as we withdrew from Vietnam, etc. But U.S. threats to pare back its commitment to the region seem to have had a significant impact.

That said, European defense spending may never reach the levels that the Trump administration, or for that matter the Washington foreign policy community generally, would prefer. And while U.S. security guarantees are surely one reason for this, it also may be the case that European countries aren’t boosting defense spending levels because they don’t face any major threats. Increasing defense spending to 2 percent of GDP or higher won’t do much about the terrorism problem European countries face. And the supposed geopolitical threat from Russia, meddling in Georgia and Ukraine aside, is consistently exaggerated.