THE EFFECTS OF PARENTAL ADVICE AND FINANCIAL LITERACY ON ASSET ACCUMULATION AMONG AMERICAN YOUTH A Thesis submitted to the Faculty of the Graduate School of Arts and Sciences of Georgetown University in partial fulfillment of the requirements for the Degree of Master of Public Policy in Public Policy By Alexander Joel Rosenberg, B.A. Washington, D.C. April 12th, 2017 Copyright 2017 by Alexander Joel Rosenberg All Rights Reserved ii THE EFFECTS OF PARENTAL ADVICE AND FINANCIAL LITERACY ON ASSET ACCUMULATION AMONG AMERICAN YOUTH Alexander Joel Rosenberg, B.A. Thesis Advisor: Jeffrey P. Thompson, Ph.D. ABSTRACT Financial literacy is an important body of knowledge and set of skills that consumers need to successfully navigate the 21st century economy. Prior research shows financial literacy bears a significant relationship, along with other factors, to the wealth outcomes of adults. While some of this research has examined how specific behaviors related to self-control affect wealth, few include the effects of parental socialization as measured through advice given from parents to children. This paper estimates an empirical relationship amongst wealth, literacy, and parental advice using data from the National Longitudinal Study of Youth’s 1997 Cohort (NLSY 97). I find financial literacy and parental advice are strongly related to wealth. I also find that women on average have lower wealth than men, even after controlling for literacy, advice, and other demographics. The source of the parental advice also proves statistically important. iii ACKNOWLEDGEMENTS The author thanks his advisor Jeffrey P. Thompson for leading him through the stages of preparation of the thesis. He also thanks his wife, Jessica, who has supported him while he worked towards this degree. iv TABLE OF CONTENTS INTRODUCTION AND BACKGROUND ................................................................................... 1 REVIEW OF PRIOR LITERATURE ............................................................................................ 2 CONCEPTUAL MODEL AND HYPOTHESES ........................................................................... 9 DATA DESCRPTION .................................................................................................................. 13 EMPIRICAL STRATEGY ........................................................................................................... 15 DESCRIPTIVE STATISTICS ...................................................................................................... 24 ANALYSIS AND REGRESSION RESULTS ............................................................................. 30 DISCUSSION AND CONCLUSION .......................................................................................... 44 APPENDIX ................................................................................................................................... 47 REFERENCES ............................................................................................................................. 58 v LIST OF FIGURES AND TABLES Figure 1. Determinants of Wealth Outcomes Through Parental Advice and Literacy……12 Figure 2. Determinants of Wealth Outcomes Differentially Affected Through Literacy by Gender………………………………………………………………………………….13 Figure 3. Determinants of Wealth Outcomes Differentially Affected Through Advice by Gender………………………………………………………………………………….13 Figure 4. Distribution of Wealth of NLSY 1997 Respondents as of Age 25……………. 18 Table 1. Summary Statistics of Categorical Variables in NLSY 97 Data Set ……………26 Figure 5. Men Have Higher Financial Literacy Scores Than Women On Average………27 Table 2. Summary Statistics of Numerical Variables in NLSY 97 Data Set ……………28 Table 3. Variable Construction Details and Definitions …………………………………29 Table 4. Quantile Regression Results with No Controls …………………………………31 Table 5. Quantile Regression Results with Controls …………………………………… 33 Table 6. Quantile Regressions with Interactions and No Controls ……………………… 36 Table 7. Quantile Regressions with Interactions and Controls……………………………39 Table 8. Quantile Regressions by Biological Parent and No Controls……………………41 Table 9. Quantile Regressions by Biological Parent with Controls……………………… 43 vi INTRODUCTION AND BACKGROUND Policymakers have devoted a great deal of attention to improving the state of financial literacy in the United States since the Great Recession. Post hoc analyses, including the Financial Crisis Inquiry Commission’s report, concluded that the build-up and eventual collapse of home prices tied to complex subprime and adjustable rate mortgages were a substantial contributor to the Recession (Angelides, 2011). Following large macroeconomic policy responses, policymakers promoted financial literacy and continued education as a bulwark against future unsustainable build-ups in household and consumer debt (Bernanke, 2012). Although many groups have endeavored to increase financial literacy and education in the United States in the public, non-profit, and private sectors, there is little uniformity in the design and implementation of these programs. Prior literature has examined the effects of demographics on financial literacy itself (Lusardi et al., 2009), including its effects on the financial outcomes of adults and young adults. This body of literature shows that financial literacy has important implications for economic outcomes. I follow in the tradition of Tang et al., 2015 and Letkiewicz & Fox, 2014, who have examined the relationship of financial literacy and additional wealth-sustaining behaviors to wealth. This paper explores financial literacy in the context of prior work on financial capacity, meaning that demonstrating knowledge of financial concepts alone is presumed insufficient for accruing savings and practicing responsible financial behavior. In particular, I empirically quantify the effect on wealth of parents providing advice to youth using the National Longitudinal Survey of Youth’s 1997 Cohort. My findings show that parents influence their 1 children’s financial behavior and outcomes, that women face particular obstacles to achieving sound financial futures holding all else equal, and that the source of the advice also matters. REVIEW OF PRIOR LITERATURE State of Financial Literacy in the United States Financial literacy is relatively low in the United States. Researchers have assessed such knowledge through surveying respondents’ understanding of compound interest, inflation, and basic diversification theory of financial portfolios (Lusardi & Mitchell, 2014). In a 2004 survey, only 34% of American adults were able to answer all three questions correctly, 10% of adults could not answer any of these three questions correctly.1 These general results have been confirmed across a variety of demographic groups in different surveys, including when other financial topics have been tested (Lusardi & Mitchell, 2007b; Duca & Kumar, 2014; Breitbach & Walstad, 2013; Huston, Finke, & Smith, 2011). Additionally, American young adults perform quite poorly on financial literacy tests when compared to their international peers. American young adults fell below the Organization for Economic Co-operation and Development (OECD) average of a five-part financial literacy question battery measured in 2012 through the Programme for International Student Assessment (PISA) (Lusardi & Mitchell, 2014). In the United States and internationally, these results differ substantially by demographic group. Lusardi, Mitchell, and Curto analyzed the number of correct responses to the three- question battery included in 2007 wave of the National Longitudinal Survey of Youth’s 1997 Cohort (NLSY 97) (Lusardi et al., 2009). The authors found that only 27% of respondents could 1 The compound interest question asks, “Suppose you had $100 in a savings account and the interest rate was 2- percent per year. After 5 years, how much do you think you would have if you left the money to grow? More than $102; exactly $102; less than $102; do not know; refuse to answer.” The correct answer is “More than $102,” as 2% interest would leave the account with a little more than $110 after 5 years. Thirty-two percent of American adults near retirement were unable to correctly answer this question (Lusardi & Mitchell, 2014). 2 correctly answer all three questions. The authors also found that women, racial minorities, less educated, and lower ability individuals scored worse than their male, white, well educated, or high ability peers. Additionally, parents’ education and financial sophistication were associated with higher probability of a correct answer to the financial literacy questions. Low financial literacy spans various age categories. Research shows that even Americans close to retirement had trouble answering the compound interest question correctly on the test (Lusardi & Mitchell, 2007a). Comparisons of cohorts in the Health and Retirement Study show that Americans lack of financial knowledge was evident across all age groups (Lusardi & Mitchell, 2007b). Financial literacy is also positively correlated with parental income, survey respondents’ income, one’s neighborhood growing up, and strongly influenced by gender (Tang et al., 2015; Lachance, 2014). Tang et al. show that there are strong interactive effects between financial knowledge and gender. Financial Literacy’s Effect on Economic Outcomes Prior literature also establishes a link between financial literacy and economic outcomes early on and later in life. Lusardi, Michaud, & Mitchell show that between 30% and 40% of retiree wealth inequality can be explained by differences in financial knowledge (Lusardi et al., 2014). Investors with greater financial knowledge earn more in their retirement portfolios than those with less knowledge (Clark, Lusardi, & Mitchell, 2015). More financially knowledgeable consumers are more likely to develop a plan for retirement, save, and accumulate more wealth in both liquid and illiquid asset classes (van Rooij, Lusardi, Alessie, 2012; Behrman, Mitchell, Soo, Bravo, 2012; Letkiewicz & Fox, 2014). The less financial literate are more likely to misunderstand the expected value of certain retirement products, such as a private annuity, and accumulate less wealth (Brown, Kapteyn, Luttmer, & Mitchell, 2013; Stango & Zinman, 2007). 3 The less financially literate are also more likely to mismanage their own housing wealth through home-equity withdrawals (Duca & Kumar, 2014). Those with less financial literacy spend more for credit and banking services, are more likely to miss a mortgage payment, hold higher amounts of credit card debt, and use high-cost cash management services rather than traditional bank accounts compared to more literate individuals (Lusardi & Tufano, 2009; Tang et al., 2015; Breitbach & Walstad, 2013). Researchers have encountered problems identifying causality in the relationship between literacy and wealth due to financial knowledge’s endogeneity in modelling. While it may be that those who are more financially literate make smarter decisions and thereby accrue more assets, it may also be true that those with a desire to accumulate wealth acquire more financial knowledge in order to attain it. Therefore, estimating the relationship between literacy and wealth will not identify a pure effect of literacy on wealth, but may be biased by motivation to obtain wealth. Those with low literacy and wealth may simply be content not to acquire financial knowledge. Different instrumental variable approaches have been proposed to correct this issue, including instrumenting for consumers’ level of economics education or information learned prior to acquiring assets (Lusardi & Mitchell, 2007; Behrman et al., 2012). Policy Responses to Low Financial Literacy In response to low levels of financial literacy schools, firms, and governments have attempted to improve the public’s understanding of finance. However, the results of such efforts are mixed. Some evidence comes from prior estimations of the effects of changes in state education policy mandating financial education in several U.S. states. In 2004, Georgia, Texas, and Idaho changed their public high school curricula to mandate financial education. Researchers found that those graduating from high schools in these states after implementation of the policy 4