Wives' Income and Marital Quality: Are There Reciprocal Effects?

Abstract

The nature and direction of the relationship between wives' income and marital quality are investigated in this research. Competing hypotheses are tested using panel data for a nationally representative sample of 771 married women and men (not couples) and structural equation modeling. The results indicate that increases in wives' income do not significantly affect either husband,s' or wives' perceptions of marital discord. Instead, increases in marital discord contribute significantly to increases in wives' income. Further, increases in marital discord contribute to increases in wives' income by increasing the likelihood that nonemployed wives will enter the labor force. Key Words: marital quality, wives' income, work. The complex association between women's growing economic independence and marital quality and stability is an important part of the debate over whether marriage is declining as an institution (Cherlin, 1992; Coontz, 1997; Hood, 1983; Oppenheimer, 1997; Ruggles, 1997; Scanzoni, 1972, 1978). Some observers have pointed to the parallel increase in rates of women's participation in the labor force and rates of divorce during this century (Cherlin, 1992; Ruggles, 1997). There is substantial research on the relationship between married women's changing economic opportunities and the quality and stability of marriage, but conclusions remain both controversial and equivocal. In particular, the direction of the association-whether increases in wives' participation in the labor force and increases in wives' income primarily affect marital quality or whether declines in marital quality primarily affect wives' laborforce behavior and income-remains unclear. There is support in previous research for both directional relationships. Some observers claim that wives' income affects the quality of marriage and the likelihood of divorce. Previous thinking on this topic suggests both positive effects on marriage (Conger et al., 1990; Hood, 1983; Scanzoni, 1972, 1978), as well as detrimental effects on marriage (Becker, 1981; Booth, Johnson, White, & Edwards, 1984; Hiedemann, Suhomlinova, & O'Rand, 1998; Treas, 1993). An opposing perspective suggests that declines in marital quality affect wives' behavior in the labor force and their income. This possibility has been addressed less frequently. Some observers emphasize that economic resources lessen women's dependence on marriage at both the aggregate and individual levels (Cherlin, 1992; Johnson & Skinner, 1986; Ruggles, 1997), and others emphasize the contributions that employment and income make to the well-being of both women and men, regardless of marital status (Hochschild, 1997; Hood, 1983; Scanzoni, 1978). The goal of the research presented here is to extend previous work by clarifying the nature and direction of the association between wives' income and marital quality. Past research has been unable to address this question due to the limitations of cross-sectional data and the use of statistical techniques that permit the investigation of only one path at a time. The present research overcomes these limitations by investigating reciprocal paths between wives' income and marital quality using longitudinal data and structural equation modeling. This approach makes it possible to investigate both of the hypothesized directional relationships and to test the relative strength of each. ECONOMIC CHANGES AND WIVES INCOME Data come from 1980-1988, the years when most of the dramatic transformations in married women's employment and earnings became particularly visible. Substantial gender differences persist between married women and married men in their employment outcomes and responsibility for paid work and family work (Blair & Lichter, 1991; Brayfield, 1995; Thompson & Walker, 1989). However, there have been important changes in married women's attachment to the labor force and in their earnings. …

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 — 95 citations exist upstream.

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