Income, Work and Marital Dissolution: Canadian Experimental Evidence

Abstract

Marital dissolutions occur for a variety of reasons. Among low income families, the added stress of inadequate earnings, intermittent job spells, and high unemployment may contribute to the decision to end a marriage. One approach to end poverty and marital instability is to give income assistance to the poor. At the same time, cash transfers might foster family breakups, intentionally or otherwise. For example, if the welfare system is designed such that individuals are better off living separately than together, there will be a financial incentive to split. The effect of income transfers on family structure is theoretically ambiguous. On the one hand, low income families may become more stable if stress factors associated with low earnings, intermittent job spells and the like are lessened by the receipt of assistance. On the other hand, unstable marriages held together strictly for economic convenience may have their bonds weakened if individuals, when separated, are eligible for support. Consequently, whether income assistance engenders, on balance, effects which are stabilising or destabilising is an empirical question which depends upon the specific population group and the assistance program under consideration. This paper employs microdata from MINCOME (Canada’s experimental test of guaranteed income) to examine the effect of various income transfers on family dissolution. Employing a path model, we find that family income level is principally an intervening rather than a direct factor in determining whether or not a family will stay intact. Our results suggest that whether or not the male head has a stable work pattern, and whether or not the female head works at all, or earns more than her spouse, are more direct contributors to marital instability than the level of family income itself.

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 — 19 citations exist upstream.

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