Apparently, in fields that debate such things, there’s been an attempt to apply economic principles to explain the ebbs and flows of attendance in different churches. John Lamont does some difference splitting and paints a persuasive picture (subscription required). Because “the rewards of religion are supernatural and, therefore, unseen,” the healthy religion, he explains, requires a different form of evidence, which is more visible where it is more distinctive:

Zeal and commitment are also necessary to lessen the “free rider” problem that plagues all voluntary groups — the problem of members who take the benefits of membership without contributing themselves. One can add to these considerations the fact that much of the appeal of religion comes from its providing moral principles with which to structure one’s life. Such principles are far more effective when one sees that most of the people around one are following them. A community of people who, by and large, follow the principles of a morally demanding religion is far more effective moral educator than any amount of preaching — a factor that is especially important for parents. Thus, a church has to set high standards for membership in order to be attractive, and the churches that set high standards are the churches that will grow. Those with low standards will shrink because low standards reduce the rewards for religious commitment below the required cost in time and effort. This is why, as Finke and Stark assert, “the churching of America was accomplished by aggressive churches committed to vivid otherworldliness.”

The problem arises with each incremental argument that this or that rule is arbitrary and may be discarded, often with the ultimately erroneous expectation that the church might be more attractive if its costs were lower. Lamont quotes from The Churching of America, 1776–2005: Winners and Losers in Our Religious Economy, by Roger Finke and Rodney Stark:

… other things being equal, people will always be in favor of a modest reduction in their costs. In this fashion, humans begin to bargain with their churches for lower tension and fewer sacrifices. They usually succeed, both because it is those with the most influence — the clergy and the leading laity — who most desire to lower the level of sacrifice and because each reduction seems so small and engenders widespread approval.

This perspective applies, to some degree, to cultural matters, as well. With marriage, for example, a great many people who formed their fundamental understanding of the institution long ago don’t see why an easing of divorce, here, and the erasure of gender rules, there, ought to have any effect on their own marriages. As the rules ease, though, and boundaries of the institution become less clear, those who are not already formed in their perspectives have less reason to follow the well-trodden path.
The benefits to the individual spouse are, as with religion, supernatural, but they’re also social and cultural. (Of course, the benefits to children born into stable marital homes are quite tangible.) If people don’t draw the satisfaction of feeling a part of something greater, upon which participants agree — that is, if an institution merely provides a title for something that each participant defines for him or herself — the calculated rewards for forming relationships that are insoluble even when difficult or for devoting time and energy to religious practices even when disruptive become more and more difficult to reconcile.