Why have fewer Americans formed partnerships as marriage gave way to cohabitation?
Core argument: Four decades of monthly U.S. demographic data show marriage declining while cohabitation rose, with both union types becoming more stable over time and fertility falling substantially after the 2008 financial crisis.
Using four decades of monthly U.S. demographic data, I show that: (i) marriage has declined while cohabitation has risen, (ii) both unions have become more stable over time, and (iii) there are two inflection points: the mid-1990s, when substitution from marriage to cohabitation accelerated, and after the 2008 financial crisis, when remaining single became more common. Figure 3 summarizes these changes in one-year hazard rates. The first phase of the transition corresponds to substitution from marriage toward cohabitation; the second marks a regime in which remaining single became increasingly common. I refer to these as the first and second inflection points of the second demographic transition. Figure 3d also shows that fertility declined substantially after the 2008 financial crisis. Estimates of a nonstationary matching model show that rising match-formation costs outpaced surplus gains, reducing partnership formation, while marriage stability is driven by rising surplus and cohabitation stability by higher dissolution costs. Policy reforms in the 1990s coincide with the marriage-cohabitation substitution, and counterfactuals indicate that, coming out of the 2008 crisis, had house prices not risen so sharply, marriage formation would have been higher while cohabitation would have decreased.
