Lessons Learned from Non-Marriage Experiments

Abstract

In the contemporary United States, marriage is closely related to money. Men and (perhaps to a lesser extent) women with more education, higher incomes, larger stocks of wealth, and more stable employment are more likely to marry than are people in more precarious economic positions. But is this relationship truly causal? That is, does economic insufficiency cause people to marry later and less often? Daniel Schneider reviews evidence from social experiments in areas such as early childhood education, human capital development, workforce training, and income support to assess whether programs that successfully increased the economic wellbeing of disadvantaged men and women also increased the likelihood that they would marry. These programs were not designed to affect marriage. But to the extent that they increased participants’ economic resources, they could have had such an effect. Examining these programs offers three key benefits. First, their experimental designs provide important insight into the causal role of economic resources for marriage. Second, they give us within-group comparisons of disadvantaged men and women, some of whom received economic “treatments” and some who did not. Third, they by and large assess interventions that are feasible and realistic within the constraints of U.S. policy making. Schneider describes each intervention in detail, discussing its target population, experimental treatment, evaluation design, economic effects, and, finally, any effects on marriage or cohabitation. Overall, he finds little evidence that manipulating men’s economic resources increased the likelihood that they would marry, though there are exceptions. For women, on the other hand, there is more evidence of positive effects.

Cite this work

No public copy is stored for this record, so only the metadata is cited.

Stored only when both papers are in the corpus — 27 citations exist upstream.

Papers that cite the same references — the relational fallback while embeddings are still backfilling.