The Two-Income Trap (35 page)

32
Johanne Boisjoly, Greg J. Duncan, and Timothy Smeeding, “The Shifting Incidence of Involuntary Job Losses from 1968 to 1992,”
Industrial Relations
37 (April 1998): 207-231, Figure 4. From 1968 to 1979, the average annual rate of job loss because of layoffs or company closing among men aged twenty-five to fifty-nine was 2.5 percent.

33
This calculation assumes that husbands and wives face roughly the same risk of getting laid off in a single year. See, e.g., Peter Gottschalk and Robert Moffitt, “Job Instability and Insecurity for Males and Females in the 1980s and 1990s,” Working Paper 408, Department of Economics, Boston College (January 1999), p. 9. The authors found that men and women had similar rates of job exits during the period 1981-1991.

34
For simplicity’s sake, this calculation assumes no correlation in the risk of job loss between husbands and wives. In fact, there is some evidence that the likelihood that a husband and a wife will both lose their jobs is weakly correlated. Husbands and wives tend to work in the same geographical area, for example, so both may face increased chances of layoff at the same time; also, they sometimes work for the same employer, who may cut back many jobs at once. As a result, the estimate in the text may slightly overstate the portion of couples in which one spouse loses a job. At the same time, however, it necessarily underestimates the proportion of couples experiencing a job loss for both spouses in a single year.

35
See Boisjoly, Duncan, and Smeeding, “The Shifting Incidence of Involuntary Job Losses.” The authors found a 28 percent increase in the incidence of involuntary
job loss between the 1968-1979 period and the 1980-1992 period. See also Daniel Polsky, “Changing Consequences of Job Separation in the United States,”
Industrial and Labor Relations Review
52 (July 1999): 565-580. Both studies found evidence that workers were more likely to lose their jobs in the 1980s and 1990s than they were in the 1970s. There is some evidence that the job loss rate improved during the expansion of the late 1990s. See Henry S. Farber, “Job Loss in the United States, 1981-1999,” Working Paper 453, Industrial Relations Section, Princeton University (June 2001). That trend, however, may have reversed during the 2001-2002 recession, when mass layoffs were 40-50 percent higher than in the 1996-1999 period. Statistics from the U.S. Department of Labor, Bureau of Labor Statistics, Archived News Releases for Extended Mass Layoffs. Available at
http://0-www.bls.gov.library.csuhayward. edu/schedule/archives/mslo_nr.htm
[3/14/2003]. There are no published studies that compare involuntary terminations during the 1970s with the early 2000s, so we have chosen, for simplicity’s sake, to document the increase between the 1970s and the early 1990s, and then to assume that the 2000s rate is comparable to that of the 1980s and early 1990s. We note, however, that sociologists and labor economists have not come to a consensus about the incidence of job loss over the past generation. Their empirical findings differ, depending on the data sets they use, the specific variables on which they focus, and the populations they examine. Several studies using the Panel Study of Income Dynamics (PSID) data detected a rise in involuntary job losses in the 1980s and 1990s compared with the 1970s. Researchers relying on Current Population Survey (CPS) data have found little increase in job instability overall, although they have noted rising instability for certain subgroups. For a review of the literature, see, e.g., John Fitzgerald, “Job Instability and Earnings and Income Consequences: Evidence from SIPP 1983-1995,” Economics Department, Bowdoin College (July 26, 1999), paper prepared for the Joint Center for Poverty Research. We also note that we have not accounted for differences in age or skill among workers. Younger parents may actually be more likely to suffer a job loss, because of a relative lack of work experience. See Boisjoly, Duncan, and Smeeding, “The Shifting Incidence of Involuntary Job Losses.”

36
Although most studies focus exclusively on men, we assume that husbands and wives, both working full-time, face an equal chance of job loss. This places the odds that one spouse would lose a job at 3.2 percent from 1980 to 1992. See Boisjoly, Duncan, and Smeeding, “The Shifting Incidence of Involuntary Job Losses.” We assume once again that there is no correlation between the chances of job loss between husbands and wives. We have not accounted for possible changes in the odds of job loss after 1992; see previous note.

37
Fully 79 percent of married women who file for bankruptcy are in the labor force, compared with 62 percent of married women in the general population. Bureau of the Census, Table F-7, Type of Family (All Races) by Median and Mean Income, 1947 to 2000, February 8, 2002.

38
Among two-income families filing for bankruptcy, about 83.3 percent identified a job problem as leading to their bankruptcy. Among married couples with only one income, the percentage identifying a job problem was 74.6 percent. Among single filers, that is, men and women who filed for bankruptcy without a spouse, the proportion identifying a job problem is even lower: 63.6 percent.

39
Among all respondents, 68.0 percent identify a job problem in the two-year period before they filed for bankruptcy. In 2001, this amounted to an estimated 1,047,000 households in bankruptcy in the aftermath of a job problem.

40
Rosalind C. Barnett and Caryl Rivers,
She Works/He Works: How Two-Income Families Are Happier, Healthier, and Better-Off
(San Francisco: HarperSanFrancisco, 1996), pp. 2, 5. See discussion in chapter 3.

41
John M. Broder, “Problem of Lost Health Benefits Is Reaching into the Middle Class,”
New York Times,
November 25, 2002.

42
D. U. Himmelstein, S. Woolhandler, and O. Carrasquillo, unpublished analysis of data from the Current Population Survey and the National Health Interview Survey.

43
Comparisons between bankruptcy filers in the early 1980s and today are difficult, because no one collected precisely the same data then as now. The best comparative estimate can be made from a 1981 survey that the San Antonio courts required of all families who filed for bankruptcy, which shows that about 8 percent of the filers cited a medical reason for filing. Sullivan, Warren, and Westbrook,
As We Forgive Our Debtors,
p. 175, n. 1. Extrapolating that sample to all filers in 1980 would suggest that about 23,000 families filed for bankruptcy in the wake of a medical problem. Other reports from about the same time estimate a lower number of medical-related bankruptcies, but they rely exclusively on court records to identify medical debt still outstanding at the time of filing and do not ask the debtors directly what happened. For example, a 1978 study in Albany, New York, found that medical bills constituted less than 2 percent of scheduled debts, but the study was based entirely on identifying medical bills in court records. B. A. Gold and E. A. Donahue, “Health Care and Personal Bankruptcy,”
Journal of Health Politics and Law
7 (1982): 734-739. For comparability to the 1981 data, the 2001 calculation includes only families who specifically identified a medical reason, thus excluding those who identified lost time at work because of medical problems. That subset would suggest that about 424,500 families in 2002 filed because of an identified medical reason—more than a twentyfold increase. For more information on current filings for medical reasons, see David Himmelstein, Deborah Thorne, Elizabeth Warren, and Steffie Woolhandler, “Illness and Injury as a Cause of Bankruptcy in the United States” (forthcoming, 2003).

44
Bureau of the Census, Current Population Reports, Special Studies, P23-190,
65+ in the United States
, 1996, Table 2-1, Elderly Population by Age: 1900-2050.

45
Twenty-two percent of working parents report receiving at least one hour of unpaid assistance from their own parents each month, compared with 38 percent of
working parents who report providing at least one hour of unpaid assistance to their own parents each month. Jody Heymann,
The Widening Gap: Why America’s Working Families Are in Jeopardy and What Can Be Done About It
(New York: Basic Books, 2000), pp. 103-104.

46
In 2000, there were 36 million hospital discharges in the United States. Centers for Medicare and Medicaid Services, Health Care Indicators, Table 1, Selected Community Hospital Statistics, 1998-2002.

47
In 2001, approximately 27,500 single-income couples, or 0.13 percent of all single-income couples in the United States, filed for bankruptcy after missing two or more weeks of work without pay because of the illness of the worker or of another family member. By comparison, 82,800 two-income couples, or 0.25 percent of the total, filed for bankruptcy after missing work because of illness.

48
Approximately 43 percent of marriages end in divorce. Data from the National Center for Health Statistics, cited in Margaret F. Brinig and Steven L. Nock, “Marry Me, Bill: Should Cohabitation Be the Default Option?” paper prepared for the Emory Series on Marriage and Religion, March 2003.

49
Barnett and Rivers,
She Works/He Works,
p. 22.

50
Scott J. South, “Time-Dependent Effects of Wives’ Employment on Marital Dissolution,”
American Sociological Review
66 (April 2001): 226-245. See also Steven L. Nock, “When Married Spouses Are Equal,”
Virginia Journal of Social Policy and the Law
9 (Fall 2001): 48-70. Nock found that marriages between equally dependent spouses (defined as marriages in which each spouse earns at least 40 percent of the total family income) are 57 percent more likely than other marriages to end in divorce.

51
For a discussion of contributors to changing divorce rates among working and nonworking women, see South, “Time-Dependent Effects.”

52
Bureau of the Census, Table CH-5, Children Under 18 Years Living With Mother Only, by Marital Status of Mother: 1960 to Present (June 29, 2001).

53
The number of unmarried, opposite-sex-couple households with children under age fifteen increased from 204,000 in 1977 to 1.7 million in 2000. Bureau of the Census, Table UC-1, Unmarried-Couple Households, by Presence of Children: 1960 to Present (2001). In 2000 there were 27.1 million married couples with children under eighteen. Bureau of the Census, Table F-10, Presence of Children Under 18 Years Old by Type of Family and Median and Mean Income, 1974-2001.

54
Larry Bumpass and Hsien-Hen Lu, “Trends in Cohabitation and Implications for Children’s Family Contexts in the United States,”
Population Studies
54 (2000): 29-41.

55
Wendy D. Manning and Pamela J. Smock, “The Relative Stability of Cohabiting and Marital Unions for Children,” unpublished paper (2002). See also Larry L. Bumpass and James A. Sweet, “National Estimates of Cohabitation,”
Demography
26 (November 1989): 615-625. Cited in Renata Forste, “Prelude to Marriage or Alternative
to Marriage? A Social Demographic Look at Cohabitation in the U.S.,” Symposium on the ALI’s Family Dissolution Principles: Blueprint to Strengthen or to Deconstruct Families? Brigham Young University, February 3, 2001. The authors found that 29 percent of cohabiting unions end within the first two years, compared with only 9 percent of marriages.

56
See, for example, W. Jean Yeung and Sandra L. Hofferth, “Family Adaptations to Income and Job Loss in the U.S.,”
Journal of Family and Economic Issues
19 (Fall 1998): 255-283.

57
Among families with children, 38.7 percent gave only one of these three reasons (job loss, medical problems, or a family breakup), 40.4 percent gave two reasons, and 7.8 percent identified all three reasons.

58
Among those in bankruptcy with out-of-pocket medical expenses of more than $1,000, 62.0 percent, or about 243,000 families, had continuous medical insurance coverage. Calculated from Himmelstein et al., “Illness and Injury as a Cause of Bankruptcy.”

59
Calculated from Himmelstein et al., “Illness and Injury as a Cause of Bankruptcy.”

60
Only 43 percent of workers have long-term disability coverage, and only 19 percent have at least six months of short-term coverage. Helen Levy, “Disability Insurance: Where Are the Gaps in Coverage?” unpublished paper (July 2002).

61
In Texas, for example, an individual “must be physically and mentally able to perform full time work” in order to qualify for unemployment benefits. Texas Workforce Commission, “Unemployment Insurance Benefits, Frequently Asked Questions.” Available at
http://m06hostp.twc.state.tx.us/CLAIMS/common/help.html#faqs
[3/23/03].

62
The proportion of families reporting that the debtor or spouse lost two or more weeks without pay because of illness or injury was 21.3 percent.

63
U.S. Office of Personnel Management, “Long Term Care Insurance Background.” Available at
http://www.opm.gov/insure/ltc/ltcbackground.pdf
[3/22/03].

64
Statement of Senator Orrin Hatch Before the United States Senate Committee on the Judiciary, S.1301—The Consumer Bankruptcy Reform Act of 1998 (May 21, 1998). Available at
http://judiciary.senate.gov/oldsite/ogh52198.htm
[3/22/03].

1
Lenore J. Weitzman,
The Divorce Revolution: The Unexpected Social and Economic Consequences for Women and Children in America
(New York: Free Press, 1985): 323.

2
A subsequent analysis of Weitzman’s data showed the decline for women to be 27 percent and an increase for men of 10 percent. Richard R. Peterson, “A Re-evaluation of the Economic Consequences of Divorce,”
American Sociological Review
61
(June 1996): 528, 532; see also Lenore J. Weitzman, “The Economic Consequences of Divorce Are Still Unequal: Comment on Peterson,”
American Sociological Review
61 (June 1996): 537; Richard R. Peterson, “Statistical Errors, Faulty Conclusions, Misguided Policy: Reply to Weitzman,”
American Sociological Review
61 (June 1996): 539. While Weitzman’s study is controversial and her reported numbers are disputed, the overall effect of divorce on women and the greater economic suffering that divorce inflicts on women are generally well established in the scholarly literature.