The Cato Institute has urged the Supreme Court for several years now to find that the exclusion of gay couples from legal marriage violates the equal protection clause of the Fourteenth Amendment to the Constitution. At the end of June, it did, in the case of Obergefell v. Hodges. Throughout the 20‐​year debate over gay marriage, some libertarians have insisted that legal marriage should not be extended to include gay people; rather, government should get entirely out of the business of licensing marriage. Let people write contracts, and marry in churches if they choose, but leave government out of it.

That’s an appealing libertarian position. In fact, I wrote one of the first articles proposing it (“Privatize Marriage,” Slate, 1997). I said that social conflicts, such as the then‐​growing conflict over same‐​sex marriage, can “be depoliticized and somewhat defused if we keep them out of the realm of government.”

But there’s a problem: The country was not and is not ready to privatize marriage. So then should libertarians advocate only a radical libertarian policy solution that won’t be implemented any time soon, leaving some people excluded from a legal institution open to others? Or do they advocate a second‐​best solution, equality under the law for whatever services government supplies? As Steven Horwitz, author of the forthcoming book Hayek’s Modern Family: Classical Liberalism and the Evolution of Social Institutions, puts it:

Cato chairman Robert A. Levy has written that “marriage today should be a private arrangement, requiring minimal or no state intervention … [but] whenever government imposes obligations or dispenses benefits, it may not ‘deny to any person within its jurisdiction the equal protection of the laws.’ ”

Cato Unbound editor Jason Kuznicki argues, however, that privatizing marriage would mean “much greater government interference in family life, higher taxes for married couples, invasions of privacy, difficulties related to child custody, and other negative consequences.” So that’s another perspective. And of course my colleagues and I intend to fight for religious liberty, as we have always done, including the liberty of bakers, florists, and others not to participate in weddings or other activities that offend their conscience.

Trade agreements present similar challenges. Scholars at Cato’s Herbert A. Stiefel Center for Trade Policy Studies have generally supported trade agreements such as NAFTA and the new Trans‐​Pacific Partnership. In the past few months they have advocated giving President Obama “trade promotion authority” (TPA), also known as “fast‐​track,” to let the administration negotiate trade agreements that Congress can reject or ratify, but not amend. Some libertarian critics say that 2,000-page agreements such as NAFTA are by definition not free trade: a free‐​trade agreement would take one page. Unilateral free trade would be even better. They have a point.

But again, it isn’t within Cato’s power to wave a magic wand and make free trade or marriage privatization happen. So our scholars usually opt for trying to move policy in a better direction. Center director Dan Ikenson writes, “Despite their flaws, free trade agreements have helped reduce domestic impediments to trade, expand our economic freedoms, and lock in positive reforms, even if only as the residual byproduct of an ill‐​premised mercantilist process. Ultimately, free trade agreements have delivered freer trade.” Not free trade, alas. But freer trade.

As I put it in a Facebook debate in June, “Best is best, but better is better than worse.” And that’s the standard that has mostly guided us at Cato for 38 years. We want to push public debate and public policy in a direction consistent with liberty and limited government. Sometimes, as in my book The Libertarian Mind and much of the material on Lib​er​tar​i​an​ism​.org, that entails laying out the case for libertarianism and strictly limited government. And sometimes, as in many of our policy studies, it involves offering politically realistic reform plans or second‐​best solutions.