Does the Student-Loan Burden Weigh into the. Decision to Start a Family?

Save this PDF as:
 WORD  PNG  TXT  JPG

Size: px
Start display at page:

Download "Does the Student-Loan Burden Weigh into the. Decision to Start a Family?"

Transcription

1 Does the Student-Loan Burden Weigh into the Decision to Start a Family? Dora Gicheva March 2011 Abstract I examine the relationship between student debt and the timing of marriage. The life-cycle consumption smoothing model implies that student loans should have a very small effect on consumption at any given point in time and should not affect the timing of family formation. I use the Survey of Consumer Finances to show that the amount of student borrowing is negatively related to the probability of marriage. I use exogenous variations in the availability of student loans since the 1970s to address the endogeneity of student debt. I supplement my results with data from a panel survey of registrants for the Graduate Management Admission Test. I find that borrowing an additional $10,000 for education decreases the long-term probability of marriage by seven percentage points or more. I use data from the GMAT Registrant Survey on reported marriage expectations to infer about the degree to which students who borrow for their education anticipate the credit constraints they face after graduation. I find evidence that among Master of Business Administration students, the size of student loans is only slightly affected by marriage expectations, so borrowers may not have perfect foresight. I thank Joseph Altonji, Fabian Lange, David Ribar, Ken Snowden and participants at the UNC Greensboro Economics Department Seminar for helpful comments. Department of Economics, University of North Carolina at Greensboro. d 1

2 The debt load keeps [Dr. Bisutti] up at night. Her damaged credit has prevented her from buying a home or a new car. She says she and her boyfriend of three years have put off marriage and having children because of the debt. The $555,000 Student-Loan Burden, The Wall Street Journal, February 13, Introduction The importance of student loans has been growing over the past four decades. In the academic year, total student loans amounted to almost $98 billion. This number, adjusted for inflation, is more than twice as large as it was ten years earlier and 15 times larger than the total amount of student loans in the academic year, when subsidized Stafford loans were introduced. This number is also considerably above the $26 billion in federal grants and $30 billion in institutional grants for the academic year. Of Bachelor s degree recipients in , 55% of those at public colleges and universities and 65% at private nonprofit institutions borrowed for their education; the average borrower accumulated over $20,000 in debt. Two-thirds of graduate degrees are financed through loans; the average graduate loan debt is currently over $40, In light of the increasing debt burdens of undergraduate and graduate students, several recent studies examine effect of student loans on the career choices that young people make. Minicozzi (2005) finds that education debt is correlated with higher earnings right after college but lower four-year wage growth and attributes the observed difference to borrowers making different career choices. Using an experiment involving financial aid 1 All data are from the College Board (2010). 2

3 assignment, Field (2009) finds that law school students career choices are also sensitive to accumulated education debt. Similarly, Rothstein and Rouse (2011) find that student borrowers are less likely to choose jobs in lower-paying sectors like government, nonprofit and education. Little academic research has been done on the role of student debt in another aspect of young people s lives: the decision to start a family. 2 However, popular media sources like The New York Times and the Wall Street Journal have turned their attention to this issue (e.g. Chaker 2009, Pilon 2010, Lieber 2010). In this paper I attempt to fill this gap and present evidence of the relationship between student loans and marriage using two very different datasets. First, I use the Survey of Consumer Finances (SCF), which surveys a large sample of households representative of the U.S. population. An advantage of the SCF data is that the timing of loan accumulation is known, which allows my empirical work to cover a long period of time and to use exogenous variations in the availability of federal and private student loans to address the endogeneity of student debt. Another useful feature of the SCF is the information it provides on respondents attitude towards different types of credit. The other dateset I use is a panel survey of men and women who registered to take the GMAT between 1990 and This survey provides a much more homogeneous sample; all student loans in these data are accrued for the same type of education (Master s of Business Administration). Removing much of the unobserved heterogeneity is another way to address potential endogeneity. I find that borrowing an additional $10,000 for education decreases the long-term 2 Two studies whose findings relate to this topic are Choy and Carroll (2000) and Chiteji (2007). Based on descriptive statistics from the Baccalaureate and Beyond longitudinal study of college graduates and the PSID, respectively, both studies find that student or other types of debt do not affect the timing of marriage. My results suggest that the apparent lack of an effect may be due to simultaneity bias. 3

4 probability of marriage by seven percentage points or more; this results is consistent across the two datasets. It is important to account for the endogeneity of student debt because otherwise the effect is attenuated considerably. I use information from the GMAT Registrant Survey on marriage expectations to show evidence that borrowers anticipate part but not all of this effect and that education expenditures and the amount of debt are correlated with anticipated marital status, although the correlation is not strong. The main results conflict with the traditional life-cycle model, which predicts that student loans should have little effect on consumption at any given point in time. Rothstein and Rouse (2011) calculate that $10,000 in student debt represents less than 1% of the present value of the typical college graduate s lifetime income. Under the permanent income hypothesis, student borrowing would not induce a dip in consumption for recent graduates. In particular, student loans should not affect the timing or probability of marriage. The assumption that people can borrow freely against their future income is critical for the predictions of the life-cycle consumption smoothing model. However, previous research suggests the presence of fairly strong liquidity constraints, for example credit card limits. Zeldes (1989) presents evidence from the PSID that borrowing constraints are an important source of observed deviations from the permanent income hypothesis. Jappelli (1990) reports that 20 percent of respondents in the SCF between the ages of 20 and 35 have been turned down by creditors recently. Gross and Souleles (2002) find evidence that credit card limits are binding as borrowing constraints because credit limit increases are associated with increased debt accumulation. The evidence on the extent to which there exist constraints on student borrowing is mixed. Keane and Wolpin (2001) 4

5 and Cameron and Taber s (2004) studies find that borrowing constraints do not affect education outcomes, but there may be an effect on consumption while in school. Carneiro and Heckman (2002) estimate that no more than 8 percent of college students in the U.S. are affected by liquidity constraints in their schooling outcomes. Lochner and Monge- Naranjo (2008) find that the borrowing constraints imposed by federal student loans are in part relaxed by the emerging private loan industry. Stinebrickner and Stinebrickner (2008) find that borrowing constraints affect the dropout decision for some low-income students. The aggregate student loan data I present in Section 2 suggest that the total amount borrowed by students in the U.S. is sensitive to changes in the federally mandated borrowing limits, which is indicative of binding constraints. Rothstein and Rouse (2011) point out that at least some level of student loans is readily available, and the terms offered by such loans are considerably better than the credit terms most young people face after graduation. In fact, student loans may exacerbate the borrowing constraints an individual faces (Minicozzi 2005). My research adds to the literature by presenting evidence that the restrictions on post-graduation borrowing are only partly anticipated at the time of student debt accumulation. Borrowing constraints can explain the delay in family formation associated with student debt when combined with a fixed cost of marriage. Mira and Ahn (2001) point out that fixed costs, such as housing and household equipment expenditures, may be part of the reason for the negative relationship they find between unemployment and age at marriage. 3 The fixed cost can be interpreted more broadly: for example, it can represent 3 From a recent Wall Street Journal article on student debt: Zack Leshetz, a 30-year-old lawyer in Fort Lauderdale, Fla., has $175,000 in student loans from his seven years in college and law school. [...] He has also been engaged since March, but has held off on marriage. There s no way I can pay for a dream wedding, or even just a regular wedding, Mr. Leshetz says (Chaker 2009). 5

6 a certain buffer amount of wealth that people seek to accumulate before starting a family. Hotz, Klerman and Willis (1993) allude to the presence of precautionary savings motives in decisions about the timing of childbearing. 4 Another interpretation of the cost comes from search models of the marriage market, in which the probability of meeting a potential spouse is increasing in the cost of search. 5 To translate this theory to a more specific example, a liquidity constrained young college graduate may need to work longer hours in order to make the required loan payments and have less time to spend on social activities that may lead to meeting a spouse. In a world in which perfect consumption smoothing is possible, it could be optimal for this worker to delay the payments until after starting a family. I focus on student loans, rather than a different type of debt, for several reasons. The timing of student loans is appropriate for the investigation of the relationship between debt and family formation because, although not always the case, it is common for the accumulation of student debt to precede marriage. The reasons for accumulating other types of debt may be related to marriage outcomes in multiple ways so it would be harder to disentangle all of the confounding factors. For example, credit card debt may be used to pay for wedding expenditures, which would result in a positive relationship between debt and the probability of being married. Medical bills are another common source of debt, and health problems may also have a direct effect on marriage. Changes in the federally mandated limits on student loans, the availability of nonfederal loans, and 4 My study focuses on the relationship between student debt and marriage; a preliminary investigation of the relationship between education loans and fertility produced similar findings. The interpretation of the fertility results is not as straightforward because the amount of accumulated debt can affect the timing and number of births directly or through changes in the probability of marriage. The results are not shown here in order to make the exposition more concise but are available from the author upon request. 5 Montgomery and Trussell (1986) offer a survey of these models. 6

7 the cost of undergraduate and graduate education offer interesting exogenous variations in the amount borrowed by a typical student. Additionally, in most cases these loans cannot be discharged in personal bankruptcy. 6 The rest of the paper proceeds as follows. In the next section I describe in more detail the two datasets that I use. Sections 3.1 and 3.2 present the empirical results for the SCF and GMAT Survey samples, respectively. I offer an interpretation of the results in Section 4, and Section 5 concludes. 2 Data 2.1 Survey of Consumer Finances The Survey of Consumer Finances, sponsored by the Federal Reserve Board in cooperation with the Department of the Treasury, draws a series of representative samples of the U.S. population, and each installment surveys a different group of about 4,500 families. The survey, conducted triennially, collects detailed information about respondents six most recent student loans, including their amounts and the years these loans were taken out. This allows me to use student debt data that cover a long time period. 7 I use the five surveys conducted between 1995 and 2007 because the marital status variable is consistent for these five instalments. I use survey respondents, rather than the household heads, in the estimation; the respondent is the household head 73 percent of the time. The SCF does not provide 6 Under Title 11, Section 523 of the United States Code. 7 The most informative empirical strategy would be to conduct survival analysis using the exact timing of debt accumulation. However, in order to ensure the privacy or respondents, the public use data do not contain the timing of first marriage for respondents who have multiple marriages. Thus, I am unable to estimate a hazard model without encountering a serious sample selection problem. 7

8 information on the household head s race or attitude towards credit. I use the first imputation provided in the public use data. The estimation sample consists of respondents who were born after 1952; this ensures that people in the sample turned 18 in 1971 or later, so they made their education borrowing decisions when guaranteed student loans were available. The final sample consists of 5,139 men (772 of whom have students loans) and 5,804 women (1,062 with loans). The main independent variable in the analysis is the amount borrowed for education. I use the available information about the six most recent student loans. I use the CPI for all urban consumers to convert the amounts to 2007 dollars and sum them for each individual to construct a measure of the total amount borrowed. My results may understate the effects of student loans on the probability of marriage because respondents are asked to report combined household loans, which would include any debt accumulated by one s spouse after marriage. The dependent variable I use is constructed from the survey question asking about legal marital status; this variable equals 0 if a respondent has never been married and 1 otherwise. The SCF also asks respondents about their attitude towards different types of credit. In the IV specifications in Section 3.1 I use the self-reported attitude towards consumption smoothing credit as an instrument. 8 The other instrument for student loans that I use exploits the exogenous variations in the availability and popularity of student loans. Using aggregate loan data from the College Board (2010), I construct a variable that equals the average amount (in real dollars) borrowed per full-time student in the year when the respondent was 18 years old. This instrument should be exogenous in the marriage regressions, since the estimation also 8 The wording of the question is People have many different reasons for borrowing money which they pay back over a period of time. [...] Please tell me whether you feel it is all right for someone like yourself to borrow money to cover living expenses when income is cut 8

9 includes a cubic in age that would absorb any cohort effects, but the variable is strongly related to the propensity to borrow for one s education. Figure 1 illustrates the annual fluctuations in the average amount borrowed by fulltime students between the and academic years. The amounts are adjusted for inflation, and the calculations include students who did not borrow. The first notable jump in student debt occurred in the late 1970s and early 1980s with the passage of the Middle Income Student Assistance Act of 1978, which made subsidized loans available to more students by removing the family income cap. Additionally, in 1979 Congress passed an amendment that provided more capital for the guaranteed loan program from private lenders, and the 1980 reauthorization of the Higher Education Act (HEA) of 1965 provided more borrowing sources for the parents of dependent college students and for financially independent undergraduates. The Reagan administration cut back on some of these provisions, and Figure 1 shows that the amount borrowed per student leveled off in the 1980s. There was a small spike after the 1986 reauthorization of the HEA, which provided a slight increase in the limits on borrowing. The most pronounced increase in student loans occurred after the 1992 reauthorization of the HEA. Under the Stafford loan program, the upper limit on the cumulative amount borrowed increased from $17,250 to $23,000 for dependent undergraduates. The limits on annual borrowing also went up. As a result, the total volume of federal loans increased from $16,222 million to $22,551 million between the and academic years: a 39% increase. The amount borrowed per full-time student went up by 32 percent (College Board 2010). The other important step in the development of the student loan program was the emergence of private loans starting with the school year. The and academic years marked the highest point for nonfederal 9

10 loans, when they accounted for a quarter of all debt. Figure 1 also shows the average amount borrowed from the SCF data divided by the number of SCF respondents who were at least 18 years old in a given year. The trends in the SCF data are similar to the aggregate student borrowing trends, including, for instance, a jump in borrowing after the 1992 reauthorization of the HEA. Table 1 shows summary statistics for the SCF variables I use in the analysis. The sample is divided by gender and based on whether a respondent has accumulated student loans. It is consistent with the data in Figure 1 that borrowers tend to be younger. This can account in part for why fewer of them have been married, but as the more formal analysis in Section 3.1 shows, age differences do not account for all of the gap. Student borrowers are also more likely to consider using credit for consumption smoothing. It is not surprising that respondents who have student loans tend to be higher educated: they are more than 10 percentage points more likely to have a Bachelor s degree or higher. 9 The average amount borrowed, conditional on having any student debt, is $30,000 for men and $23,000 for women. Table 1 also shows the distribution of responses by survey wave. Later survey years have higher representation because of the age restriction I impose on the sample, but the distribution does not differ much for borrowers and non-borrowers. 9 The proportion of SCF respondents with a postsecondary degree is somewhat higher than the average for the U.S. population, which is consistent with the fact that the SCF slightly oversamples high-asset households. 10

11 2.2 GMAT Registrant Survey The second dataset used in this study is a four-wave panel survey of registrants for the Graduate Management Admission Test. 10 It is typical for MBA students to graduate with large amounts of student debt. 11 Other than the wide use of loans observed in the data, a big advantage of the survey of GMAT registrants is that students in the sample borrowed for the same type of education and tend to hold similar occupations, which eliminates much of the heterogeneity present in other datasets. Hence, the empirical results are likely to be subject to less endogeneity bias. An additional benefit of the survey is that the age at which most MBA students graduate is close to the median age at first marriage for the highly educated. 12 The universe for the survey consists of everyone who registered to take the GMAT between June 1990 and March 1991 and was living in the U.S. at the time of registration. The GMAT Registrant Survey was conducted in four waves. The first one was sent out shortly after test registration and had a response rate of 84 percent (5,853 responses out of 7,006 randomly selected test registrants). The final interviews took place between January 1997 and November 1998 and 3,771 of the 5,853 initial respondents returned completed questionnaires. The marriage variable I use equals 1 if a respondent is married at the time of the last interview and 0 otherwise and is only defined for the sample of 10 The survey was conducted by the Batelle Memorial Institute on behalf of the Graduate Management Admission Council. For examples of other studies using this survey, see Arcidiacono, Cooley and Hussey (2008) and Montgomery and Powell (2003). 11 According to data from the National Center for Education Statistics Baccalaureate and Beyond Longitudinal Study, 47.3 percent of the students in the study who completed and MBA degree between 1997 and 2003 borrowed money for their graduate education. Among those with loans, the average amount borrowed was around $33, Goldin and Katz (2008) find that for men and women who graduated from Harvard in 1990, the median age at first marriage is 30 years. The median MBA graduation age in the GMAT Registrant Survey is 29 years. 11

12 respondents who are not married when first interviewed (1,392 men and 1,266 women). Eliminating observations with missing values on key variables leaves an estimation sample consisting of 1,316 men and 1,154 women. There are 799 males and 618 females in the sample who completed an MBA degree by the last installment of the survey. 13 As a measure of the main regressor of interest, the size of school loans, I use the reported total amount borrowed for business school by the time of the last interview. The variable equals zero if the reported amount is zero or if a respondent did not attend a graduate management program. The amount borrowed is censored at $99,999, but this affects only one observation. I use the nominal amounts reported in the survey. I do not have information on the exact timing of the loan, but the time period over which all debt was accumulated covers about five years, so deflating the amounts should not make a difference. The GMAT Registrant Survey also asks respondents who attended business school to report their expenditures on two main categories: the first is tuition and fees, and the second is books and supplies. I use these as instruments for the amount of accumulated debt in Section 3.2. The empirical model that I estimate includes two variables designed measure respondents attitudes towards their career and family. I include them because they are likely to be related to the decision to borrow and the decision to start a family. To construct these variables, I use a question from the first wave of the GMAT Registrant Survey that asks about the importance attributed to One s own family and children and Career 13 It can be argued that those who did not complete an MBA degree either never intended to, so they are inherently different from completers and do not belong in the sample, or self-selected into the zero-loans group, in which case I should include them in the estimation. The summary statistics in Table 2 show that MBA graduates are similar in most respects to respondents without a degree, and all estimation results are almost identical regardless of whether I use the unrestricted or restricted sample. In the marriage regressions I only show results for the unrestricted sample. 12

13 and work. 14 The Values family variable is set to equal one for respondents who indicate that family and children are very important and zero for those who select somewhat important, not very important or not at all important. Similarly, Values career equals 1 if the respondent selected very important and 0 otherwise. The GMAT Survey also provides information on respondents expected marital status. This information is important because it can provide an insight as to whether men and women in the sample can predict well the relationship between graduate school loans and the probability of marriage. I estimate two sets of specifications: one excluding expected marital status, and another one in which marriage expectations are included. If the first specification produces estimates that are significantly different from zero, and in the second specification I find no relationship between loans and marital status, this would be a good indication that borrowers are aware of the medium to long term effects of student debt. The specific question that I use comes from the first wave of the survey, where respondents are asked whether they expect to marry within the next two years. The possible answers are Yes, No and Don t know, so I use this question to construct two binary variables: Expect to be married and Expect not to be married. Table 2 shows summary statistics for the GMAT Survey sample. The survey slightly oversamples women and oversamples minorities to a larger extent. 15 Male and female respondents are similar in their ages (on average 25 years old at the time of the first interview), but men are slightly more likely to marry by the end of the survey period. 14 The exact wording used in the survey is Here is a list of various aspects of life. We would like to know how important each of these aspects of life is for you. Responses can vary from 1 (Very) to 4 (Not at all). 15 In the testing cycle, women constituted 36.8 percent of test takers, while the proportion reporting their race as something other than white (non-hispanic) was 16.3 percent (Profile of Graduate Management Admission Test Candidates, to n.d.). 13

14 About a quarter of all respondents expect to marry within two years of their first interview, while half of all men and slightly fewer than half of women in the sample expect to remain single within the two-year period. Part-time and full-time MBA education is about equally popular for men, while women are significantly more likely to complete a part-time degree during the survey period. Only 4 percent of female graduates and 5 percent of male graduates attend an executive program. Slightly over a quarter of MBA completers in the sample borrow for their graduate management education, and the average loan is $25,000 for men and $20,000 for women (conditional on this amount being positive). Men are likely to attend MBA programs with higher tuition costs, but the amount spent on books and supplies does not differ much by gender. Only about 16 percent of men and 14 percent of women assign less than high importance to their family, but career is not a main priority for 40 percent of men and 30 percent of women in the sample. 3 Empirical Results The dependent variable in the main specifications is a binary indicator of marital status. For this reason, I estimate a probit model, but the results are likely to be affected by simultaneity bias if marital status affects borrowing. To correct for this, I also estimate an instrumental variable probit model by maximum likelihood. Marital status m i is determined by the unobserved continuous variable m i = βloan i + X i δ + u i, 14

15 where loans are endogenous, and X i is a vector of other explanatory variables that are exogenous. The outcome observed in the data is m i = 1 if m i 0 and m i = 0 otherwise. The reduced form relationship between the loan size and the exogenous regressors, including the vector of instruments Z i, is Loan i = Z i Π 1 + X i Π 2 + ν i. The error terms u i and ν i are jointly normally distributed, and the usual normalization is made: σ u = 1. Let σ denote the standard deviation of ν i and ρ - the correlation between u i and ν i. Then the log likelihood contribution of the i th observation is [ βloani + X i δ + ρ σ ln L i = m i ln Φ (Loan ] i Z i Π 1 X i Π 2 ) (1 ρ 2 ) ( [ 0.5 βloani + X i δ + ρ σ + (1 m i ) ln 1 Φ (Loan ]) i Z i Π 1 X i Π 2 ) (1 ρ 2 ) [ ] 0.5 Loani Z i Π 1 X i Π 2 + ln φ ln σ, σ where Φ and φ are the standard normal cdf and pdf, respectively. All estimation is performed separately for men and women to account for potential gender differences in the correlates of the probability of marriage. The results I report are marginal effects estimated at the sample means. 3.1 SCF Results Table 3 shows the estimation results for the SCF samples of men (first two columns) and women (last two columns). Columns 1 and 3 show results for the probit model without instrumenting for the amount borrowed. The estimated marginal effect is close 15

16 to zero; this result is similar to the findings in Choy and Carroll (2000) and Chiteji (2007). However, Columns 2 and 4 show that when I instrument for the size of student debt, loans have a fairly strong impact on the probability of marriage, especially for women. A $10,000 increase in student debt decreases the probability of marriage by 9 percentage points for men and 21 percentage points for women whose characteristics fall in the middle of the distribution. Estimating the IV probit model separately for various age groups does not change the size of the estimated coefficient much but makes the results noisier because this eliminates much of the time variation in average loans per student, which the identification relies heavily upon (these results are not shown but are available upon request). This provides some evidence that the effect I find is persistent over the long run. The IV results in Table 3 suggest that, holding student loans constant, acquiring more education increases the probability of marriage for both genders. Table 3 also shows the reduced form, or first stage, regression results. The maximum likelihood approach estimates both equations jointly, so this is not a true first stage. The two instruments are the average amount of debt per full-time student in the year when respondents were 18 years old and the reported attitude towards consumption smoothing credit. Both are highly significant and have the expected signs. A $1,000 increase in average loans per full-time student increases the amount borrowed by $1,900 for men and $1,400 for women in the sample. An increase in the number of full-time students in periods of more generous student loan provision can explain why the estimated elasticity is higher than 1, but we also cannot reject the null hypothesis that the true coefficient is equal to 1 for women. People who indicate approval of borrowing in periods when income is uncharacteristically low tend to accumulate between $1,300 and 16

17 $1,400 more in student debt GMAT Survey Results The probit and IV probit estimation results for the sample of GMAT registrants are shown in Table 4. Panel A contains the results for men, and Panel B - for females. I estimate two sets of results: excluding and including the marriage expectation variables discussed in Section 2.2. I do this in order to examine the degree to which expected marital status is correlated with the amount borrowed. The probit results in Column 1 indicate that the effect of student loans on the probability of marriage is negative and statistically significant, but the absolute value of the estimated marginal effect is considerably smaller than the SCF instrumental variable estimates in Section 3.1: for men and for women. Since the GMAT Survey sample is less heterogeneous, it is reasonable to expect that the probit results will be affected less by endogeneity bias and the estimated coefficient is more likely to be statistically different from zero. Adding the marriage expectation variables reduces the absolute value of these marginal effects: they become for men (significant at the 5 percent level) and for women (not significant at the 10 percent level). This is consistent with the hypothesis that people anticipate some of the the borrowing constraints they face after graduation and adjust their marriage expectations or size of debt accordingly. 16 Two-step estimation of the IV probit model allows me to perform an overidentification test of the exclusion restriction. The minimum chi-square statistic has a p-value of for men and for women. This should raise some concern that the one (or both) of the instruments may not be exogenous in the estimation model for women. Excluding either instrument does not change the sign or significance of the coefficient estimate on the amount borrowed for women, so the overall conclusions would be similar if only one of the instruments is truly exogenous. 17

18 The IV probit results in Columns 3 and 4 suggest that not instrumenting for the amount borrowed still biases the estimates up, although the bias is smaller than it is in the more widely representative SCF data. The marginal effects are comparable to the estimated effect for men in the SCF: for male GMAT registrants and for females when marriage expectation are excluded, and and respectively when expected marital status is included in the estimation. All of these effects are statistically significant. The marriage expectation variables are highly significant in all specifications and have the expected signs: positive for the Expect to be married variable and negative for the Expect not to be married variable. The family valuation indicator is significant in all specifications for men, and the estimated marginal effect is positive: 0.16 without expected marital status and 0.11 when expected marital status is included. The estimates effects are about 6 percentage points smaller for women and are not significant when the marriage expectation indicators are included. The career valuation variable is not significant in any of the specifications; all of the estimated marginal effects are negative for men and positive but smaller than 0.02 for women. Both instruments are good predictors of the amount borrowed in the reduced form first stage regressions. MBA students use loans to finance about half of their tuition expenditures. For women, loans increase by almost a dollar with each $1 spent on books and supplies, while men borrow 72 cents for each dollar spent on books and supplies. These results are highly significant. 17 To provide a closer look at the relationship between graduate school expenditures, 17 The overidentification test based on two-stage estimation of the IV probit model fails to reject the null hypothesis of exogeneity of the instruments; the p-value of the test statistic is above 0.5 for men and 0.9 for women. 18

19 student loans and expected marital status, Table 5 shows Tobit estimates of the determinants of the amount borrowed and amounts spent on tuition and books by MBA graduates. The results in Panel A are for men in the sample of MBA graduates, and Panel B shows results for female MBA graduates. The dependent variable in Column 1 is the amount borrowed, and in Columns 2 and 3: the amounts spent on tuition and books or other supplies, respectively. The results suggest that there is some correlation between marriage expectation on the one hand and business school expenditures and debt on the other, but this correlation is not very strong. None of the coefficients are significant at the 10 percent level for women, but all three coefficient estimates for Expect to be married are negative, while the corresponding estimates for Expect not to be married are positive. Three of the six coefficients are significant for men; male MBA students who expect to marry within two years attend cheaper programs on average. Men who anticipate not to marry are expected to borrow an additional $6,000; this is the largest point estimate and it is significant at the 10 but not 5 percent level. The Values career variable is positively correlated with business school expenditures: men who indicate high career valuation are expected to spend $3,400 more on MBA tuition and $300 more on books and supplies. Women who value their career spend on average $2,800 more on tuition. The coefficients are not significant when the dependent variable is the amount borrowed. The Values family variable has negative coefficients for men and positive coefficients for women, but none of the estimates is statistically significant. The sample sizes for these regressions are small: 799 men and 618 women. This may be part of the reason why the results are noisy. 19

20 4 Discussion of the Results The main question posed in this study is whether there is a relationship between the amount of accumulated student debt and the timing of family formation. The instrumental variable probit results in Sections 3.1 and 3.2 offer strong evidence that student loans have a negative and significant, both statistically and economically, impact on the probability of marriage. Controlling for age and education, both men and women are less likely to marry if they hold student loans. This result is interesting because it conflicts with the predictions of the Permanent Income Hypothesis. My findings are in line with other recent studies that present evidence that young graduates choices of career paths are more sensitive to the amount of accumulated student debt than lifecycle consumption smoothing would imply (Minicozzi 2005, Field 2009, Rothstein and Rouse 2011). I argue that incorporating a fixed cost of marriage and binding borrowing constraints faced after the completion of schooling can reconcile the observed trends with the lifetime consumption theory. Another important observation is that marriage expectations are weakly correlated with the amount of accumulated student debt. It is a convenient feature of the GMAT Registrant Survey that respondents are asked about their anticipated marital status, and the panel aspect of the data allows me to compare respondents predictions with their actual outcomes. Including marriage expectations in the regressions decreases the absolute value of the estimated effect of loans. This is consistent with the hypothesis that, first of all, marriage expectations are somewhat accurate so they are correlated with marriage outcomes, and second, that people anticipate to some degree the borrowing constraints they are faced with after graduation. The latter would result in a correlation 20

21 between student debt and expected marital status. Interestingly, the effect of student debt does not disappear completely in the IV specifications when marriage expectations are included. One plausible explanation is that borrowers do not plan for all postgraduation liquidity constraints. If this is indeed the case, then the findings in this paper would carry policy implications beyond concerns about the decrease in marriage rates. Potential student borrowers would need to be educated better about education and other types of credit in order to make efficient life-cycle consumption decisions. Both datasets used for the estimation reveal that the absolute value of the coefficient on student debt is considerably lower when borrowing is treated as exogenous. The estimated marginal effects are essentially zero in the SCF sample. This result is related to the finding that MBA students adjust their schooling expenditures and loans with regards to expected marital status. If anticipated or realized marriage outcomes affect students borrowing decisions, this would lead to simultaneity bias in the non-iv specifications. Hence, it is important to treat debt as endogenous in the estimation. Not doing so could explain why other studies, such as Choy and Carroll (2000) and Chiteji (2007), do not find a relationship between student debt and marriage. The trends depicted in Figure 1 reveal the presence of substantial exogenous variations in the availability of student loans, which drive observed borrowing behavior in the SCF. Using these variations as an instrument for student debt is another contribution of this paper. 5 Conclusion The relationship between student borrowing and the timing of family formation has not been studied at length before, but the question has been raised in the popular press 21

22 (Chaker 2009, Pilon 2010, Lieber 2010). Data on student debt are publicly available from surveys like the Survey of Consumer Finances, but the main empirical problem is the likely endogeneity of school loans. The empirical estimates would be biased if, for example, marital status affects schooling decisions. To address the endogeneity of education-related debt, I resort to an instrumental variables approach. Data from the SCF allow me to study different cohorts of students, who made their borrowing decisions at different points in time: between 1971 and As Figure 1 shows, the average amount borrowed by full-time students in the U.S. has been exhibiting a well-pronounced upward trend since the early 1970s. This can be explained in part by the increasing cost of education, but changes in the federally mandated borrowing limits are also responsible for the rising student debt. I use the fact that changes in the Higher Education Act of 1968 are unlikely to have a direct effect on marriage decisions, but are strongly correlated with the size of student debt observed among SCF respondents. The other instrument I use is based on respondents attitudes towards using credit in order to achieve consumption smoothing. The other dataset I use is a survey of GMAT registrants. Since this sample is much more homogeneous, the endogeneity bias is not as strong in these data. The borrowing decisions of MBA students I observe were made over a relatively short time period, so it is not possible to use variations in the availability of federal and private student loans. Instead, I use variations in the cost of MBA education as an instrument. Both datasets reveal similar patterns: instrumenting for the loan amounts increases the absolute value of the coefficient on the amount borrowed, compared to the non-iv estimates. An additional $10,000 in loans decreases the probability of marriage by 7 percentage points or more. A useful question that the GMAT Registrant Survey asks 22

23 is about expected marital status. I use this information to examine to what extent borrowers anticipate the decrease in the probability of marriage associated with student debt. I find that including marriage expectations in the regressions decreases the absolute value of the marginal effect of loans, but the change is small compared to the size of the estimated effect. The results provide some evidence of a relationship between marriage expectations prior to enrollment in an MBA program and the cost of the program attended. I also see some evidence in the SCF that the relationship between loans and marriage outcomes does not vary by age, which would suggest that the observed effect does not represent a short-term delay in marriage. Although more work needs to be done to tie the results to a more formal theoretical model, my findings indicate that the consumption smoothing model is insufficient in this context. One possible explanation is that young people may not be fully aware of the liquidity constraints they face after completion of schooling, and that in addition there exist fixed costs to starting a family. This could be a source of inefficiency in the long-term schooling and consumption decisions that students make. As school loans are becoming more popular, there may be the need for better education about student debt and its consequences. 23

24 Figure 1: Average Loans per Full-Time Student Federal/Private Loans per FT Student $8,000 $7,000 $6,000 $5,000 $4,000 $3,000 $2,000 $1,000 $0 Federal Loans per FT Student Total (Federal+Private) Loans per FT Student Average SCF Loans $1,000 $900 $800 $700 $600 $500 $400 $300 $200 $100 $0 SCF Loans Year Source: The College Board (2010) and author s calculations using Survey of Consumer Finances data. The calculations include nonrecipients. All amounts are in 2009 dollars. The SCF averages are calculated as the total amount borrowed reported for each year divided by the number of respondents who were at least 18 years old in that year. 24

25 Table 1: SCF Summary Statistics No School Loans Have School Loans Male Female Male Female N = 4367 N = 4742 N = 772 N = 1062 White Black Hispanic Age (8.427) (8.401) (8.579) (8.482) Has been married OK to borrow to smooth consumption College degree Master s degree Doctorate (includes PhD, MD, JD) Amount borrowed/ (38.01) (31.69) Survey year (% of sample) Standard errors in parentheses. 25

26 Table 2: GMAT Registrant Survey Summary Statistics All MBA Graduates Male Female Male Female N = 1316 N = 1154 N = 799 N = 618 Asian Black Hispanic Age at t = (4.146) (4.821) (4.219) (5.199) Married at t = Expect to be married in 2 years Expect not to be married in 2 years Obtained FT MBA Obtained PT MBA Obtained Executive MBA Percent who borrowed Amount borrowed / (18.39) (15.46) (18.39) (15.46) Total amount spent on tuition (13.51) (11.51) (15.61) (14.23) Total amount spent on books and supplies (2.013) (1.778) (2.329) (2.187) Values family (t = 1) Values career (t = 1) Standard errors in parentheses. The calculations for the average amount borrowed include only MBA graduates with positive amount borrowed. 26

27 Table 3: SCF Estimation Results Male Female Probit IV Probit Probit IV Probit Amount borrowed/ ** ** (0.0004) (0.0043) (0.0003) (0.0019) College degree * ** ** ** (0.0155) (0.0249) (0.0151) (0.0209) Master s degree ** ** ** ** (0.0230) (0.0295) (0.0232) (0.0341) Doctorate ** ** ** (0.0283) (0.0628) (0.0406) (0.0450) First stage results: Average loans per FT student ** ** (0.0005) (0.0004) OK to borrow to smooth consumption ** ** (0.4640) (0.3100) * p<0.10, ** p<0.05 Dependent variable: whether respondent has ever been married. The reported coefficients are marginal effects estimated at the sample means. The IV coefficients are estimated by maximum likelihood, hence there is no actual first stage. All specifications include controls for race and a cubic in age. N = 5139 for men and N = 5804 for women. 27

28 Table 4: Marriage Rates and the Size of Student Loans: MBA Students A. Males Probit IV Probit Amount borrowed/ ** ** ** ** (0.0013) (0.0013) (0.0020) (0.0021) Values family (t=1) ** ** ** ** (0.0391) (0.0401) (0.0390) (0.0400) Values career (t=1) (0.0289) (0.0295) (0.0289) (0.0296) Expect to be married in 2 years ** ** (0.0424) (0.0424) Expect not to be married in 2 years ** ** (0.0346) (0.0346) First stage results: Total amount spent on tuition/ ** ** (0.0248) (0.0248) Total amt on books & supplies/ ** ** (0.1610) (0.1611) B. Females Probit IV Probit Amount borrowed/ ** ** ** (0.0018) (0.0019) (0.0028) (0.0029) Values family (t=1) ** ** (0.0441) (0.0453) (0.0440) (0.0452) Values career (t=1) (0.0332) (0.0341) (0.0333) (0.0341) Expect to be married in 2 years ** ** (0.0418) (0.0418) Expect not to be married in 2 years ** ** (0.0364) (0.0364) First stage results: Total amount spent on tuition/ ** ** (0.0224) (0.0224) Total amt on books & supplies/ ** ** (0.1419) (0.1419) * p<0.10, ** p<0.05. The reported coefficients are marginal effects estimated at the sample means. The IV coefficients are estimated by maximum likelihood, hence there is no actual first stage. The dependent variable is whether married by t = 4. All regressions include controls for the type of MBA degree, age, age squared and race. There are 1316 observations for men and 1154 observations for women. 28

29 Table 5: Do Marriage Expectations Affect the Cost of MBA Degree and Amount Borrowed? A. Males Loans Tuition Books Expect to be married in 2 years ** ** (3.9513) (1.9442) (0.2523) Expect not to be married in 2 years * (3.5189) (1.6821) (0.2351) Values family (t=1) (3.7362) (1.8533) (0.3070) Values career (t=1) ** * (2.7294) (1.3551) (0.1737) B. Females Loans Tuition Books Expect to be married in 2 years (3.3990) (1.7887) (0.2441) Expect not to be married in 2 years (3.0107) (1.7848) (0.2548) Values family (t=1) (4.2057) (1.9844) (0.2393) Values career (t=1) * (2.9201) (1.5934) (0.2198) * p<0.10, ** p<0.05. Tobit estimation results for the subsample of MBA graduates. The reported standard errors are heteroskedasticity robust. All regressions include controls for race and a quadratic in age. There are 799 observations for men and 618 observations for women. All dependent variables are measured in thousands of nominal dollars. 29

Does the Student Loan Burden Weigh into the Decision to Start a Family?

Does the Student Loan Burden Weigh into the Decision to Start a Family? Does the Student Loan Burden Weigh into the Decision to Start a Family? Department of Economics Working Paper Series Dora Gicheva University of North Carolina at Greensboro September 2011 Working Paper

More information

In Debt and Alone? Examining the Causal Link. between Student Loans and Marriage

In Debt and Alone? Examining the Causal Link. between Student Loans and Marriage In Debt and Alone? Examining the Causal Link between Student Loans and Marriage Dora Gicheva June 2013 Abstract I examine the relationship between student debt and marital status. I use exogenous variations

More information

In Debt and Alone? Examining the Causal Link. between Student Loans and Marriage

In Debt and Alone? Examining the Causal Link. between Student Loans and Marriage In Debt and Alone? Examining the Causal Link between Student Loans and Marriage Dora Gicheva June 2012 Abstract I examine the relationship between student debt and marital status. I use exogenous variations

More information

Postponing the Family? The Relationship between Student Debt and. Lifecycle Transitions. Philip Anderson, University of Notre Dame!

Postponing the Family? The Relationship between Student Debt and. Lifecycle Transitions. Philip Anderson, University of Notre Dame! Postponing the Family? The Relationship between Student Debt and Lifecycle Transitions Philip Anderson, University of Notre Dame! April 2013 Abstract This paper uses the Survey of Consumer Finances and

More information

Worker Mobility, Employer-Provided General. Training, and the Choice of Graduate Education

Worker Mobility, Employer-Provided General. Training, and the Choice of Graduate Education Worker Mobility, Employer-Provided General Training, and the Choice of Graduate Education Dora Gicheva March 2011 Abstract This paper links inherent mobility to observed schooling choices. A job search

More information

Tax Benefits for Graduate Education: Incentives for. Whom?

Tax Benefits for Graduate Education: Incentives for. Whom? Tax Benefits for Graduate Education: Incentives for Whom? Steven Bednar and Dora Gicheva March 2012 Abstract This paper examines how changes in the tax treatment of graduate tuition affect enrollment in

More information

Optimal Consumption with Stochastic Income: Deviations from Certainty Equivalence

Optimal Consumption with Stochastic Income: Deviations from Certainty Equivalence Optimal Consumption with Stochastic Income: Deviations from Certainty Equivalence Zeldes, QJE 1989 Background (Not in Paper) Income Uncertainty dates back to even earlier years, with the seminal work of

More information

The Effects of Credit Status on College Attainment and College Completion

The Effects of Credit Status on College Attainment and College Completion DISCUSSION PAPER SERIES IZA DP No. 6719 The Effects of Credit Status on College Attainment and College Completion Dora Gicheva Felicia Ionescu Nicole Simpson July 2012 Forschungsinstitut zur Zukunft der

More information

Tax Benefits for Graduate Education: Incentives for Whom?

Tax Benefits for Graduate Education: Incentives for Whom? Tax Benefits for Graduate Education: Incentives for Whom? Steven Bednar and Dora Gicheva October 2012 Abstract Numerous studies have examined the enrollment responses of traditional undergraduate students

More information

The question of whether student debt levels are excessive

The question of whether student debt levels are excessive College on Credit: How Borrowers Perceive Their Education Debt By Sandy Baum and Marie O Malley Sandy Baum is professor of economics at Skidmore College in Skidmore, PA. Marie O Malley is vice president

More information

Most discussions of student loans focus on the

Most discussions of student loans focus on the January 2003 The Role of Student Loans in College Access Sandy Baum, Ph.D., Professor of Economics, Skidmore College Most discussions of student loans focus on the difficulties they generate for students.

More information

The Effects of Focused Area of Study on Self-Reported Skill Gains and Satisfaction of MBA Graduates

The Effects of Focused Area of Study on Self-Reported Skill Gains and Satisfaction of MBA Graduates The Effects of Focused Area of Study on Self-Reported Skill Gains and Satisfaction of MBA Graduates Andrew J. Hussey, University of Memphis, USA ABSTRACT Using a longitudinal survey of registrants for

More information

Tax Benefits for Graduate Education

Tax Benefits for Graduate Education Tax Benefits for Graduate Education Steven Bednar and Dora Gicheva November 2011 Abstract This paper examines how changes in the tax treatment of graduate tuition affect enrollment in and financing of

More information

Characteristics of College Students Who Graduate with No Debt

Characteristics of College Students Who Graduate with No Debt Student Aid Policy Analysis Characteristics of College Students Who Graduate EXECUTIVE SUMMARY Mark Kantrowitz Publisher of Fastweb.com and FinAid.org August 24, 2011 About two fifths of undergraduate

More information

Social Security Eligibility and the Labor Supply of Elderly Immigrants. George J. Borjas Harvard University and National Bureau of Economic Research

Social Security Eligibility and the Labor Supply of Elderly Immigrants. George J. Borjas Harvard University and National Bureau of Economic Research Social Security Eligibility and the Labor Supply of Elderly Immigrants George J. Borjas Harvard University and National Bureau of Economic Research Updated for the 9th Annual Joint Conference of the Retirement

More information

Tax Benefits for Graduate Education: Incentives for Whom?

Tax Benefits for Graduate Education: Incentives for Whom? Tax Benefits for Graduate Education: Incentives for Whom? Steven Bednar and Dora Gicheva June 2013 Abstract Numerous studies have examined the enrollment responses of traditional undergraduate students

More information

Education and Wage Differential by Race: Convergence or Divergence? *

Education and Wage Differential by Race: Convergence or Divergence? * Education and Wage Differential by Race: Convergence or Divergence? * Tian Luo Thesis Advisor: Professor Andrea Weber University of California, Berkeley Department of Economics April 2009 Abstract This

More information

Online Survey of Employees Without Workplace Retirement Plans

Online Survey of Employees Without Workplace Retirement Plans Online Survey of Employees Without Workplace Retirement Plans Report of Findings Conducted for: State of California October 2015 Prepared by Greenwald & Associates 2015 1 Table of Contents Methodology

More information

An Analysis of the Health Insurance Coverage of Young Adults

An Analysis of the Health Insurance Coverage of Young Adults Gius, International Journal of Applied Economics, 7(1), March 2010, 1-17 1 An Analysis of the Health Insurance Coverage of Young Adults Mark P. Gius Quinnipiac University Abstract The purpose of the present

More information

Federal Student Loan Debt: 1993 to 2004

Federal Student Loan Debt: 1993 to 2004 Issue Brief June 2005 Federal Student Loan Debt: 1993 to 2004 University officials, lenders, and policy makers typically monitor annual federal student loan volume (the number of loans made and the total

More information

At What Cost? h o w s t u d e n t d e b t. r e d u c e s. l i f e t i m e w e a l t h. r o b e r t. h i l t o n s m i t h A U G U S T 2 0 1 3

At What Cost? h o w s t u d e n t d e b t. r e d u c e s. l i f e t i m e w e a l t h. r o b e r t. h i l t o n s m i t h A U G U S T 2 0 1 3 Debt-For-Diploma Series At What Cost? h o w s t u d e n t d e b t r e d u c e s l i f e t i m e w e a l t h by r o b e r t h i l t o n s m i t h A U G U S T 2 0 1 3 a b o u t d ē m o s Dēmos is a public

More information

AGENDA ITEM 12. Action requested:

AGENDA ITEM 12. Action requested: AGENDA ITEM 12. SUMMARY OF: INCREASING COLLEGE ACCESS OR JUST INCREASING DEBT? A DISCUSSION ABOUT RAISING STUDENT LOAN LIMITS AND THE IMPACT ON ILLINOIS STUDENTS Submitted for: Summary: Information The

More information

Profile of Pell Grant Recipients Quick Reference Guide

Profile of Pell Grant Recipients Quick Reference Guide Profile of Recipients Family Income The average family adjusted gross income (AGI) for recipients is 123% of the poverty line (median 112%), compared with 395% of the poverty line for non-recipients. Adjusted

More information

Medical Bills and Bankruptcy Filings. Aparna Mathur 1. Abstract. Using PSID data, we estimate the extent to which consumer bankruptcy filings are

Medical Bills and Bankruptcy Filings. Aparna Mathur 1. Abstract. Using PSID data, we estimate the extent to which consumer bankruptcy filings are Medical Bills and Bankruptcy Filings Aparna Mathur 1 Abstract Using PSID data, we estimate the extent to which consumer bankruptcy filings are induced by high levels of medical debt. Our results suggest

More information

Rethinking the Cultural Context of Schooling Decisions in Disadvantaged Neighborhoods: From Deviant Subculture to Cultural Heterogeneity

Rethinking the Cultural Context of Schooling Decisions in Disadvantaged Neighborhoods: From Deviant Subculture to Cultural Heterogeneity Rethinking the Cultural Context of Schooling Decisions in Disadvantaged Neighborhoods: From Deviant Subculture to Cultural Heterogeneity Sociology of Education David J. Harding, University of Michigan

More information

TAX POLICY, LUMP-SUM PENSION DISTRIBUTIONS, AND HOUSEHOLD SAVING ANGELA E. CHANG *

TAX POLICY, LUMP-SUM PENSION DISTRIBUTIONS, AND HOUSEHOLD SAVING ANGELA E. CHANG * TAX POLICY, PENSION DISTRIBUTIONS, AND SAVING TAX POLICY, LUMP-SUM PENSION DISTRIBUTIONS, AND HOUSEHOLD SAVING ANGELA E. CHANG * Abstract - The Tax Reform Act of 986 (TRA) imposed a ten percent penalty

More information

Saving and Investing Among Higher Income African-American and White Americans

Saving and Investing Among Higher Income African-American and White Americans The Ariel Investments 2010 Black Investor Survey: Saving and Investing Among Higher Income African-American and Americans July 21, 2010 1 For further information, contact Merrillyn J. Kosier at Ariel Investments

More information

The Effects of Student Loans on Long-Term Household Financial Stability. Dora Gicheva, UNC Greensboro. And. Jeffrey Thompson, Federal Reserve Board**

The Effects of Student Loans on Long-Term Household Financial Stability. Dora Gicheva, UNC Greensboro. And. Jeffrey Thompson, Federal Reserve Board** The Effects of Student Loans on Long-Term Household Financial Stability Dora Gicheva, UNC Greensboro And Jeffrey Thompson, Federal Reserve Board** DRAFT: October 18, 2013 ** The analysis and conclusions

More information

Comparing the Labor Market Return to An Associate Degree and to A Bachelor s Degree

Comparing the Labor Market Return to An Associate Degree and to A Bachelor s Degree Comparing the Labor Market Return to An Associate Degree and to A Bachelor s Degree Carol Xiang Pomona College April 25, 2015 Abstract College education has been shown to improve the labor market success

More information

SPECIAL SECTION: Indebtedness

SPECIAL SECTION: Indebtedness SPECIAL SECTION: Indebtedness Introduction Student indebtedness is a topic of great import for universities, public policy officials, and of course, doctoral students and their families. This special section

More information

FACULTY RETIREMENT PLANS: THE ROLE OF RETIREE HEALTH INSURANCE

FACULTY RETIREMENT PLANS: THE ROLE OF RETIREE HEALTH INSURANCE TRENDS AND ISSUES SEPTEMBER 2015 FACULTY RETIREMENT PLANS: THE ROLE OF RETIREE HEALTH INSURANCE Robert L. Clark Zelnak Professor Poole College of Management North Carolina State University Retiree health

More information

Can Annuity Purchase Intentions Be Influenced?

Can Annuity Purchase Intentions Be Influenced? Can Annuity Purchase Intentions Be Influenced? Jodi DiCenzo, CFA, CPA Behavioral Research Associates, LLC Suzanne Shu, Ph.D. UCLA Anderson School of Management Liat Hadar, Ph.D. The Arison School of Business,

More information

Expansion of Higher Education, Employment and Wages: Evidence from the Russian Transition

Expansion of Higher Education, Employment and Wages: Evidence from the Russian Transition Expansion of Higher Education, Employment and Wages: Evidence from the Russian Transition Natalia Kyui December 2012 Abstract This paper analyzes the effects of an educational system expansion on labor

More information

Student Debt: Bigger and Bigger. By Heather Boushey. September 2005

Student Debt: Bigger and Bigger. By Heather Boushey. September 2005 cepr Briefing Paper CENTER FOR ECONOMIC AND POLICY RESEARCH Student Debt: Bigger and Bigger By Heather Boushey September 2005 CENTER FOR ECONOMIC AND POLICY RESEARCH 1611 CONNECTICUT AVE., NW, SUITE 400

More information

Gender Differences in Employed Job Search Lindsey Bowen and Jennifer Doyle, Furman University

Gender Differences in Employed Job Search Lindsey Bowen and Jennifer Doyle, Furman University Gender Differences in Employed Job Search Lindsey Bowen and Jennifer Doyle, Furman University Issues in Political Economy, Vol. 13, August 2004 Early labor force participation patterns can have a significant

More information

Who Graduates with Excessive Student Loan Debt?

Who Graduates with Excessive Student Loan Debt? Student Aid Policy Analysis Papers Who Graduates with Excessive Student Loan Debt? www.studentaidpolicy.com/excessive debt/ Mark Kantrowitz President MK Consulting, Inc. December 14, 2015 Executive Summary

More information

Household debt and consumption in the UK. Evidence from UK microdata. 10 March 2015

Household debt and consumption in the UK. Evidence from UK microdata. 10 March 2015 Household debt and consumption in the UK Evidence from UK microdata 10 March 2015 Disclaimer This presentation does not necessarily represent the views of the Bank of England or members of the Monetary

More information

Saving and Life Insurance Holdings at Boston University A Unique Case Study

Saving and Life Insurance Holdings at Boston University A Unique Case Study Saving and Life Insurance Holdings at Boston University A Unique Case Study by B. Douglas Bernheim Stanford University and National Bureau of Economic Research bernheim@stanford.edu Solange Berstein Boston

More information

THE ECONOMIC RETURNS TO AN MBA. Duke University, U.S.A.; University of Wisconsin Madison, U.S.A.; University of Memphis, U.S.A. 1.

THE ECONOMIC RETURNS TO AN MBA. Duke University, U.S.A.; University of Wisconsin Madison, U.S.A.; University of Memphis, U.S.A. 1. INTERNATIONAL ECONOMIC REVIEW Vol. 49, No. 3, August 2008 THE ECONOMIC RETURNS TO AN MBA BY PETER ARCIDIACONO, JANE COOLEY, AND ANDREW HUSSEY 1 Duke University, U.S.A.; University of Wisconsin Madison,

More information

Economic Brief. The Role of Option Value in College Decisions

Economic Brief. The Role of Option Value in College Decisions Economic Brief April 2016, EB16-04 The Role of Option Value in College Decisions By Jessie Romero and Nicholas Trachter Despite the large and persistent wage premium earned by college graduates, college

More information

Student Loan Information Provision and Academic Choices

Student Loan Information Provision and Academic Choices Student Loan Information Provision and Academic Choices By Maximilian Schmeiser, Christiana Stoddard, and Carly Urban As the cost of pursuing post-secondary education in the United States has continued

More information

Using the National Longitudinal Survey

Using the National Longitudinal Survey Who goes to college? Evidence from the NLSY97 s from the National Longitudinal Survey of Youth 997 show that sex, race, and ethnicity are unrelated to the student s decision to complete the first year

More information

In contrast to the large and rapidly growing literature on information technology (IT) investments and firm

In contrast to the large and rapidly growing literature on information technology (IT) investments and firm MANAGEMENT SCIENCE Vol. 52, No. 2, February 2006, pp. 187 203 issn 0025-1909 eissn 1526-5501 06 5202 0187 informs doi 10.1287/mnsc.1050.0479 2006 INFORMS The Personal Computer and Entrepreneurship Robert

More information

The Price of a Science PhD: Variations in Student Debt Levels Across Disciplines and Race/Ethnicity

The Price of a Science PhD: Variations in Student Debt Levels Across Disciplines and Race/Ethnicity Center for Education & Innovation at American Institutes for Research Broadening Participation in Graduate Education Issue Brief May 2013 Kristina L. Zeiser Rita J. Kirshstein Courtney Tanenbaum The Price

More information

Women s Participation in Education and the Workforce. Council of Economic Advisers

Women s Participation in Education and the Workforce. Council of Economic Advisers Women s Participation in Education and the Workforce Council of Economic Advisers Updated October 14, 214 Executive Summary Over the past forty years, women have made substantial gains in the workforce

More information

Student Loan Debt: Causes and Consequences DISSERTATION

Student Loan Debt: Causes and Consequences DISSERTATION Student Loan Debt: Causes and Consequences DISSERTATION Presented in Partial Fulfillment of the Requirements for the Degree Doctor of Philosophy in the Graduate School of The Ohio State University By Xue

More information

The Risk of Losing Health Insurance Over a Decade: New Findings from Longitudinal Data. Executive Summary

The Risk of Losing Health Insurance Over a Decade: New Findings from Longitudinal Data. Executive Summary The Risk of Losing Health Insurance Over a Decade: New Findings from Longitudinal Data Executive Summary It is often assumed that policies to make health insurance more affordable to the uninsured would

More information

The Big Payoff: Educational Attainment and Synthetic Estimates of Work-Life Earnings

The Big Payoff: Educational Attainment and Synthetic Estimates of Work-Life Earnings The Big Payoff: Educational Attainment and Synthetic Estimates of Work-Life Earnings Special Studies Issued July 2002 P23-210 Does going to school pay off? Most people think so. Currently, almost 90 percent

More information

Default and Repayment among Baccalaureate Degree Earners

Default and Repayment among Baccalaureate Degree Earners Default and Repayment among Baccalaureate Degree Earners Lance J Lochner Alexander Monge-Naranjo October 4, 2013 Abstract Lenders are interested in the expected return on their loans. In this paper, we

More information

The Life-Cycle Motive and Money Demand: Further Evidence. Abstract

The Life-Cycle Motive and Money Demand: Further Evidence. Abstract The Life-Cycle Motive and Money Demand: Further Evidence Jan Tin Commerce Department Abstract This study takes a closer look at the relationship between money demand and the life-cycle motive using panel

More information

This PDF is a selection from a published volume from the National Bureau of Economic Research

This PDF is a selection from a published volume from the National Bureau of Economic Research This PDF is a selection from a published volume from the National Bureau of Economic Research Volume Title: Fiscal Policy and Management in East Asia, NBER-EASE, Volume 16 Volume Author/Editor: Takatoshi

More information

Analysis of the CSLP Student Loan Defaulter Survey and Client Satisfaction Surveys

Analysis of the CSLP Student Loan Defaulter Survey and Client Satisfaction Surveys Western University Scholarship@Western Centre for Human Capital and Productivity. CHCP Working Papers Economics Working Papers Archive 2013 Analysis of the CSLP Student Loan Defaulter Survey and Client

More information

4. Work and retirement

4. Work and retirement 4. Work and retirement James Banks Institute for Fiscal Studies and University College London María Casanova Institute for Fiscal Studies and University College London Amongst other things, the analysis

More information

The Effect of Credit Constraints on the College Drop-Out Decision: A Direct Approach Using a New Panel Study

The Effect of Credit Constraints on the College Drop-Out Decision: A Direct Approach Using a New Panel Study The Effect of Credit Constraints on the College Drop-Out Decision: A Direct Approach Using a New Panel Study Todd Stinebrickner The University of Western Ontario Ralph Stinebrickner Berea College Abstract:

More information

The Economic Returns to an MBA

The Economic Returns to an MBA The Economic Returns to an MBA Peter Arcidiacono, Jane Cooley, Andrew Hussey January 3, 2007 Abstract Estimating the returns to education is difficult in part because we rarely observe the counterfactual

More information

The Ariel/Schwab Black Investor Survey:

The Ariel/Schwab Black Investor Survey: The Ariel/Schwab Black Investor Survey: Saving and Investing Among Higher Income and Americans Ariel Investments, LLC and The Charles Schwab Corporation are non-affiliated entities, but co-sponsor the

More information

a topic for a future issue brief, please Minnesota is heading into challenging territory. In the decades

a topic for a future issue brief, please Minnesota is heading into challenging territory. In the decades RESEARCH BRIEF September 2011 Minnesota Educational Needs and Higher Education Finance Policy This is one in a series of research-based briefs prepared by the Minnesota Private College Council for members

More information

Student Loan Market Trends Is College Worth It. Presenter: Kelly Savoie, Director Business Development April 2016

Student Loan Market Trends Is College Worth It. Presenter: Kelly Savoie, Director Business Development April 2016 Student Loan Market Trends Is College Worth It Presenter: Kelly Savoie, Director Business Development April 2016 Agenda This presentation is an overview of trends in the industry and the value of a college

More information

Education and Default Incentives with Government Student Loan Programs

Education and Default Incentives with Government Student Loan Programs Education and Default Incentives with Government Student Loan Programs Lance Lochner University of Western Ontario NBER Alexander Monge-Naranjo Northwestern University December, 2003 Preliminary and Incomplete

More information

Why Laura isn t CEO. Marianne Bertrand March 2009

Why Laura isn t CEO. Marianne Bertrand March 2009 Why Laura isn t CEO M i B t d Marianne Bertrand March 2009 Based on: Dynamics of the Gender Gap for Young Professionals in the Corporate and Financial Sectors Joint with: Claudia Goldin (Harvard University)

More information

The Effects of Student Loans on Long-Term Household Financial Stability

The Effects of Student Loans on Long-Term Household Financial Stability The Effects of Student Loans on Long-Term Household Financial Stability Department of Economics Working Paper Series Dora Gicheva University of North Carolina at Greensboro Jeffrey Thompson Federal Reserve

More information

CHAPTER 11. AN OVEVIEW OF THE BANK OF ENGLAND QUARTERLY MODEL OF THE (BEQM)

CHAPTER 11. AN OVEVIEW OF THE BANK OF ENGLAND QUARTERLY MODEL OF THE (BEQM) 1 CHAPTER 11. AN OVEVIEW OF THE BANK OF ENGLAND QUARTERLY MODEL OF THE (BEQM) This model is the main tool in the suite of models employed by the staff and the Monetary Policy Committee (MPC) in the construction

More information

The Student Debt Review

The Student Debt Review New America Education Policy Program February 2014 Policy Brief The Student Debt Review Analyzing the State of Undergraduate Student Borrowing Ben Miller 2 The Student Debt Review About the Author Ben

More information

What It s Worth: Field of Training and Economic Status in 2009

What It s Worth: Field of Training and Economic Status in 2009 What It s Worth: Field of Training and Economic Status in 2009 Household Economic Studies Issued February 2012 P70-129 INTRODUCTION The relationship between educational attainment and economic outcomes

More information

Americans Current Views on Smoking 2013: An AARP Bulletin Survey

Americans Current Views on Smoking 2013: An AARP Bulletin Survey Americans Current Views on Smoking 2013: An AARP Bulletin Survey November 2013 Americans Current Views on Smoking 2013: An AARP Bulletin Survey Report Prepared by Al Hollenbeck, Ph.D. Copyright 2013 AARP

More information

HELP Interest Rate Options: Equity and Costs

HELP Interest Rate Options: Equity and Costs HELP Interest Rate Options: Equity and Costs Bruce Chapman and Timothy Higgins July 2014 Abstract This document presents analysis and discussion of the implications of bond indexation on HELP debt. This

More information

Trends in Household Wealth Dynamics, 2005-2007

Trends in Household Wealth Dynamics, 2005-2007 Technical Series Paper #09-03 Trends in Household Wealth Dynamics, 2005-2007 Elena Gouskova and Frank Stafford Survey Research Center - Institute for Social Research University of Michigan September, 2009

More information

DO THE POOREST AMONG THE POOR BENEFIT LESS FROM ACTIVE LABOR MARKET PROGRAMS? EVIDENCE FROM PERU S PROJOVEN

DO THE POOREST AMONG THE POOR BENEFIT LESS FROM ACTIVE LABOR MARKET PROGRAMS? EVIDENCE FROM PERU S PROJOVEN DO THE POOREST AMONG THE POOR BENEFIT LESS FROM ACTIVE LABOR MARKET PROGRAMS? EVIDENCE FROM PERU S PROJOVEN Miguel Jaramillo José Galdo Verónica Montalva Motivation There is increasing interest in program

More information

AMERICA'S YOUNG ADULTS AT 27: LABOR MARKET ACTIVITY, EDUCATION, AND HOUSEHOLD COMPOSITION: RESULTS FROM A LONGITUDINAL SURVEY

AMERICA'S YOUNG ADULTS AT 27: LABOR MARKET ACTIVITY, EDUCATION, AND HOUSEHOLD COMPOSITION: RESULTS FROM A LONGITUDINAL SURVEY For release 10:00 a.m. (EDT) Wednesday, March 26, 2014 USDL-14-0491 Technical information: (202) 691-7410 nls_info@bls.gov www.bls.gov/nls Media contact: (202) 691-5902 PressOffice@bls.gov AMERICA'S YOUNG

More information

DOES HOUSEHOLD DEBT HELP FORECAST CONSUMER SPENDING? Robert G. Murphy Department of Economics Boston College Chestnut Hill, MA 02467

DOES HOUSEHOLD DEBT HELP FORECAST CONSUMER SPENDING? Robert G. Murphy Department of Economics Boston College Chestnut Hill, MA 02467 DOES HOUSEHOLD DEBT HELP FORECAST CONSUMER SPENDING? By Robert G. Murphy Department of Economics Boston College Chestnut Hill, MA 02467 E-mail: murphyro@bc.edu Revised: December 2000 Keywords : Household

More information

WOMEN S PERSPECTIVES ON SAVING, INVESTING, AND RETIREMENT PLANNING

WOMEN S PERSPECTIVES ON SAVING, INVESTING, AND RETIREMENT PLANNING WOMEN S PERSPECTIVES ON SAVING, INVESTING, AND RETIREMENT PLANNING 1 WOMEN S PERSPECTIVES ON SAVING, INVESTING, AND RETIREMENT PLANNING November 2015 Insured Retirement Institute 2 WOMEN S PERSPECTIVES

More information

A Comparative Analysis of Income Statistics for the District of Columbia

A Comparative Analysis of Income Statistics for the District of Columbia Occasional Studies A Comparative Analysis of Income Statistics for the District of Columbia ACS Income Estimates vs. DC Individual Income Tax Data Jayron Lashgari Office of Revenue Analysis Office of the

More information

Graduating to a Pay Gap

Graduating to a Pay Gap Graduating to a Pay Gap The Earnings of Women and Men One Year after College Graduation Executive Summary and Recommendations Executive Summary Fifty years after the passage of the Equal Pay Act of 1963,

More information

Double your major, double your return?

Double your major, double your return? ARTICLE IN PRESS Economics of Education Review 27 (2008) 375 386 www.elsevier.com/locate/econedurev Double your major, double your return? Alison F. Del Rossi a,, Joni Hersch b a Department of Economics,

More information

WORKING THEIR WAY THROUGH COLLEGE: STUDENT EMPLOYMENT AND ITS IMPACT ON THE COLLEGE EXPERIENCE

WORKING THEIR WAY THROUGH COLLEGE: STUDENT EMPLOYMENT AND ITS IMPACT ON THE COLLEGE EXPERIENCE Issue Brief May 2006 WORKING THEIR WAY THROUGH COLLEGE: STUDENT EMPLOYMENT AND ITS IMPACT ON THE COLLEGE EXPERIENCE During the 2003 04 academic year, 78 percent of undergraduates worked while they were

More information

Society of Actuaries Middle Market Life Insurance Segmentation Program (Phase 1: Young Families)

Society of Actuaries Middle Market Life Insurance Segmentation Program (Phase 1: Young Families) Society of Actuaries Middle Market Life Insurance Segmentation Program (Phase 1: Young Families) September 2012 Sponsored By: SOA Marketing and Distribution Section SOA Product Development Section SOA

More information

Borrowing During Unemployment: Unsecured Debt as a Safety Net

Borrowing During Unemployment: Unsecured Debt as a Safety Net Borrowing During Unemployment: Unsecured Debt as a Safety Net James X. Sullivan Northwestern University May 2002 Abstract This paper explores two related questions: Does unsecured debt help households

More information

Measure. Georgia State University. Funding

Measure. Georgia State University. Funding Technical Series Paper #15-02 Comparing Estimates of Student Loans Measure ed in the PSID-TAS with the National Postsecondary Studen nt Aid Survey Patricia Andreski University of Michigan Daniel Kreisman

More information

When Will the Gender Gap in. Retirement Income Narrow?

When Will the Gender Gap in. Retirement Income Narrow? When Will the Gender Gap in Retirement Income Narrow? August 2003 Abstract Among recent retirees, women receive substantially less retirement income from Social Security and private pensions than men.

More information

En h a n c ed Ea r n i n g s a n d Ta x Revenues from a

En h a n c ed Ea r n i n g s a n d Ta x Revenues from a En h a n c ed Ea r n i n g s a n d Ta x Revenues from a University of Wisconsin-Whitewater Degree: A Synthetic An a ly s i s by Russ Kashian Fiscal and Economic Research Center University of Wisconsin

More information

The Labor-Market Returns for Community College Degrees, Diplomas, and Certificates. September 2008

The Labor-Market Returns for Community College Degrees, Diplomas, and Certificates. September 2008 The Labor-Market Returns for Community College Degrees, Diplomas, and Certificates September 2008 by Christopher Jepsen * Kenneth Troske Paul Coomes University of Kentucky University of Kentucky University

More information

Working Long Hours and Career Wage Growth

Working Long Hours and Career Wage Growth Working Long Hours and Career Wage Growth Dora Gicheva September 2009 Abstract This study establishes empirically a positive relationship between hours worked per week and growth in hourly wages. A four-wave

More information

Liquidity Constraints in the U.S. Housing Market

Liquidity Constraints in the U.S. Housing Market Liquidity Constraints in the U.S. Housing Market Denis Gorea Virgiliu Midrigan May 215 Contents A Income Process 2 B Moments 4 C Mortgage duration 5 D Cash-out refinancing 5 E Housing turnover 6 F House

More information

Student Loans: Do College Students Borrow Too Much (or Not Enough)? Appendix: Data and Methods

Student Loans: Do College Students Borrow Too Much (or Not Enough)? Appendix: Data and Methods Student Loans: Do College Students Borrow Too Much (or Not Enough)? A. Processing of March CPS Data Appendix: Data and Methods We use the March CPS from the 1964 through 2009 (1963 through 2008 earnings

More information

Institutional Grants and Baccalaureate Degree Attainment

Institutional Grants and Baccalaureate Degree Attainment Institutional Grants and Baccalaureate Degree Attainment Derek V. Price, Ph.D., DVP-PRAXIS, LTD. Ryan J. Davis, NASFAA June 2006 ACKNOWLEDGMENTS Thanks to Professor Scott L. Thomas of the University of

More information

SUMMARY OF BORROWER INTAKE FORM DATA COLLECTED for the STUDENT LOAN BORROWER COUNSELING PROGRAM

SUMMARY OF BORROWER INTAKE FORM DATA COLLECTED for the STUDENT LOAN BORROWER COUNSELING PROGRAM SUMMARY OF BORROWER INTAKE FORM DATA COLLECTED for the STUDENT LOAN BORROWER COUNSELING PROGRAM Prepared by the For the Center for Excellence in Financial Counseling LEADERSHIP COUNCIL SEMI-ANNUAL MEETING

More information

In the past two decades, the federal government has dramatically

In the past two decades, the federal government has dramatically Degree Attainment of Undergraduate Student Borrowers in Four-Year Institutions: A Multilevel Analysis By Dai Li Dai Li is a doctoral candidate in the Center for the Study of Higher Education at Pennsylvania

More information

Affirmative Action in MBA Programs: Investigating the Mismatch Hypothesis for Race and Gender

Affirmative Action in MBA Programs: Investigating the Mismatch Hypothesis for Race and Gender Affirmative Action in MBA Programs: Investigating the Mismatch Hypothesis for Race and Gender Wayne A. Grove 1 Professor, Economics Department, Le Moyne College grovewa@lemoyne.edu Andrew J. Hussey Assistant

More information

STUDENT DEBT MYTHS AND FACTS Second Edition

STUDENT DEBT MYTHS AND FACTS Second Edition STUDENT DEBT MYTHS AND FACTS Second Edition April 2014 The Issue Student loan debt is now approximately $1 trillion, and the delinquency rate for student loans has increased to 10 percent. Critics are

More information

Volume URL: http://www.nber.org/books/feld87-2. Chapter Title: Individual Retirement Accounts and Saving

Volume URL: http://www.nber.org/books/feld87-2. Chapter Title: Individual Retirement Accounts and Saving This PDF is a selection from an out-of-print volume from the National Bureau of Economic Research Volume Title: Taxes and Capital Formation Volume Author/Editor: Martin Feldstein, ed. Volume Publisher:

More information

What Do Stafford Loans Actually Buy You? The Effect of Stafford Loan Access on Community College Students

What Do Stafford Loans Actually Buy You? The Effect of Stafford Loan Access on Community College Students NATIONAL CENTER for ANALYSIS of LONGITUDINAL DATA in EDUCATION RESEARCH TRACKING EVERY STUDENT S LEARNING EVERY YEAR A program of research by the American Institutes for Research with Duke University,

More information

Inflation. Chapter 8. 8.1 Money Supply and Demand

Inflation. Chapter 8. 8.1 Money Supply and Demand Chapter 8 Inflation This chapter examines the causes and consequences of inflation. Sections 8.1 and 8.2 relate inflation to money supply and demand. Although the presentation differs somewhat from that

More information

Selection of Student Loans and College Performance

Selection of Student Loans and College Performance Selection of Student Loans and College Performance Maximilian D. Schmeiser Federal Reserve Board* Christiana Stoddard Montana State University Carly Urban Montana State University Abstract The rising cost

More information

Marriage and divorce: patterns by gender, race, and educational attainment

Marriage and divorce: patterns by gender, race, and educational attainment ARTICLE OCTOBER 2013 Marriage and divorce: patterns by gender, race, and educational attainment Using data from the National Longitudinal Survey of Youth 1979 (NLSY79), this article examines s and divorces

More information

Today, more households than ever before are paying off student loan debt. Fry (2012)

Today, more households than ever before are paying off student loan debt. Fry (2012) Is Student Debt Jeopardizing the Short-Term Financial Health of U.S. Households? William Elliott and IlSung Nam In this study, the authors use the Survey of Consumer Finances to determine whether student

More information

RETIREMENT SAVINGS OF AMERICAN HOUSEHOLDS: ASSET LEVELS AND ADEQUACY

RETIREMENT SAVINGS OF AMERICAN HOUSEHOLDS: ASSET LEVELS AND ADEQUACY RETIREMENT SAVINGS OF AMERICAN HOUSEHOLDS: ASSET LEVELS AND ADEQUACY Report to the Consumer Federation of America and DirectAdvice.com Catherine P. Montalto, Ph.D. Consumer and Textile Sciences Department

More information

Canadian Labour Market and Skills Researcher Network

Canadian Labour Market and Skills Researcher Network Canadian Labour Market and Skills Researcher Network Working Paper No. 4 The Effect of Student Loan Limits on University Enrolments Christine Neill Wilfrid Laurier University December 2008 CLSRN is supported

More information

Determining Future Success of College Students

Determining Future Success of College Students Determining Future Success of College Students PAUL OEHRLEIN I. Introduction The years that students spend in college are perhaps the most influential years on the rest of their lives. College students

More information

Life Cycle Asset Allocation A Suitable Approach for Defined Contribution Pension Plans

Life Cycle Asset Allocation A Suitable Approach for Defined Contribution Pension Plans Life Cycle Asset Allocation A Suitable Approach for Defined Contribution Pension Plans Challenges for defined contribution plans While Eastern Europe is a prominent example of the importance of defined

More information

Measures of College Affordability and Student Aid in North Carolina. Analysis by Laura Greene Knapp Education Research Consultant.

Measures of College Affordability and Student Aid in North Carolina. Analysis by Laura Greene Knapp Education Research Consultant. Benchmarks: Measures of College Affordability and Student Aid in North Carolina Analysis by Laura Greene Knapp Education Research Consultant for the North Carolina State Education Assistance Authority

More information