What are the Economic Benefits of Marriage?

By Yhu Kuni

---

In the United States, it has been noted for some time that currently married men earn substantially more than unmarried men. An important measure used to determine the differences is called the wage premium. Robert Schoeni, using the data collected by the Luxenbourg Income Study (LIS) argues that married males earn between 10 and 30 percent more than men who have never been married. Jeffrey Gray, using data from the National Longitudinal Surveys, observes that the wage premium decreased by more than 40 percent during the 1980s. He argues that this was due largely to a decline in specialization associated with marriage.

Theory

There are several hypotheses to explain the wage premium paid to married men. They are 1. productivity differences 2. selection into marriage, 3. discrimination by employers, and 4. omitted unobservable.

The first hypothesis suggests that marriage per se makes workers more productive. This may occur through specialization by spouses, or perhaps that a married man feels a stronger sense of responsibility because he is contributing not only to his own well-being but also to that of his wife and family. The second hypothesis proposes that individuals who are financially successful are more attractive marriage partners and, therefore, are more likely to be selected into marriage. The third hypothesis suggests that employers have some preference for married men and thus discriminate in their favor. The fourth hypothesis is that marriage is simply correlated with unobservable characteristics that are rewarded in the labor market.

Results/Analysis

Schoeni used the Luxembourg Income Study (LIS) to base his research. The LIS data studied fourteen countries: Australia, Canada, France, West Germany, Israel, Italy, Luxembourg, Netherlands, Norway, Poland, Sweden, Switzerland, United Kingdom, and the United States. The earliest data are from 1979 and the latest are from 1986. The empirical analysis consists of estimating log earnings equations for each country and year. The sample is restricted to those men who are heads of households and 25-55 years old.

This study finds that there are large and statistically significant different earnings between men of different marital states. Married men earn on average 0-30 percent more than those who are not married. In addition, both separated and divorced men earn more than men who are never married but less than those who are currently married.

Grey observes a recent fall in the marriage wage premium and explores why this is so.

Grey borrows data from both the National Longitudinal Survey (NLS) of Young Men and the National Longitudinal Survey of Youth (NLSY). In 1966, the NLS surveyed a cohort of men aged 14 to 24 and reinterviewed them periodically through 1981. Data from the 1976, 1978, and the 1980 surveys are used in the analysis. To be consistent with the NLSY, the sample was restricted to men aged 24-31 as of January 1, 1976. The NLS sample of white men in this cohort attributed from 2836 in 1966 to 204 in 1980. Due to sample restrictions and incomplete data, the sample consists of 1248 observations.

Results combining the NLS and NLSY suggest that over the 1976-80 period a marriage wage premium did exist, as married earned 11 percent more than never-married men in the same industry and occupation and with the same socioeconomic background. Divorced or separated men earned about 10 percent more than comparable never-married men.

Survey results reveal a decrease in real wages irrespective of marital status. In 1976, the average wage of never-married men was 12 percent lower than the average wage of divorced or separated men and 13 percent lower than the average wage of married men. By 1989, the differential between married and never-married men had dropped to 9 percent, while the mean wage of divorced or separated men was 7 percent lower than the mean wage of never-married men.

Gray also tests whether the productivity effect of marriage has changed over time. Both cross-sectional and longitudinal estimates reveal a negative relationship between wages and years divorced or separated over both the 1976-80 and 1989-93 periods. The relative fall in wages as men remain divorced is consistent with both the depreciation of human capital acquired while married and the selection of higher-ability divorced men back into marriage. Much of the wage premium experienced by divorced or separated men during the 1976-80 period appears consistent with productivity gains occurred while married, gains that were slowly lost as men remained divorced.

Grey then tests whether the recent fall in the marital status premium is attributable to a fall in the productivity effects of marriage. This fall in the productivity of marriage could result either from less specialization taking place in marriages or from a decrease in the return to specialization. If the marriage wage premium is due to specialization within marriage, then the wage premium should be larger when wives work less in the labor force. This is indeed the case where the marriage wage premium was nearly 17 percent for husbands whose wives didn’t work, and 9 percent for husbands whose wives worked 40 hours per week in the labor force.

Results show that gains from specialization decreased by the 1989-93 period. The premium paid to husbands whose wives specialized in home production decreased by nearly 7 percentage points. This fall in the marriage wage premium coincides with the apparent decline in specialization, as the mean labor market hours worked by wives nearly doubled from 15 hours per week in 1976 to 28 hours per week in 1989.

The results of Gray’s report suggest that the marriage premium paid to men decreased dramatically between the late 1970s and the early 1990s. The marriage wage premium is consistent with productivity-enhancing effects of marriage. The data also suggest that the marriage premium is very sensitive to the degree of household specialization taking place.

References:

Robert F. Schoeni, "Marital Status and Earnings in Developed Countries", Journal of Population Economics, Vol. 8, 1995, pp. 351-359.

Jeffrey S. Gray, "The Fall in Men's Return to Marriage: Declining Productivity Effects or Changing Selection?", The Journal of Human Resources, Vol. 32, No. 3,

1997, pp. 481-504.