WPS6723 Policy Research Working Paper 6723 The Impact of High School Financial Education Experimental Evidence from Brazil Miriam Bruhn Luciana de Souza Leão Arianna Legovini Rogelio Marchetti Bilal Zia The World Bank Development Research Group & Latin America and Caribbean Region December 2013 Policy Research Working Paper 6723 Abstract This paper studies the impact of a comprehensive likelihood of financial planning, and greater participation financial education program spanning six states, in household financial decisions by students. “Trickle- 868 schools, and approximately 20,000 high school up” impacts on parents were also significant, with students in Brazil through a randomized control trial. improvements in parent financial knowledge, savings, The program increased student financial knowledge and spending behavior. The study also finds evidence that by a quarter of a standard deviation and led to a 1.4 the program affected students’ inter-temporal preferences percentage point increase in saving for purchases, better and attitudes. This paper is a joint product of the Finance and Private Sector Development and the Development Impact Evaluation Teams, Development Research Group; and the Finance and Private Sector, Latin America and the Caribbean Region. It is part of a larger effort by the World Bank to provide open access to its research and make a contribution to development policy discussions around the world. Policy Research Working Papers are also posted on the Web at http://econ.worldbank. org. TThe authors may be contacted at mbruhn@worldbank.org, alegovini@worldbank.org, rmarchetti@worldbank.org, and bzia@worldbank.org. The Policy Research Working Paper Series disseminates the findings of work in progress to encourage the exchange of ideas about development issues. An objective of the series is to get the findings out quickly, even if the presentations are less than fully polished. The papers carry the names of the authors and should be cited accordingly. The findings, interpretations, and conclusions expressed in this paper are entirely those of the authors. They do not necessarily represent the views of the International Bank for Reconstruction and Development/World Bank and its affiliated organizations, or those of the Executive Directors of the World Bank or the governments they represent. Produced by the Research Support Team The Impact of High School Financial Education: Experimental Evidence from Brazil Miriam Bruhn Luciana de Souza Leão Arianna Legovini Rogelio Marchetti Bilal Zia * Abstract This paper studies the impact of a comprehensive financial education program spanning six states, 868 schools, and approximately 20,000 high school students in Brazil through a randomized control trial. The program increased student financial knowledge by a quarter of a standard deviation and led to a 1.4 percentage point increase in saving for purchases, better likelihood of financial planning, and greater participation in household financial decisions by students. “Trickle-up” impacts on parents were also significant, with improvements in parent financial knowledge, savings, and spending behavior. The study also finds evidence that the program affected students’ inter-temporal preferences and attitudes. JEL Codes: D14, D91, I25, O12 Keywords: Financial Education, Inter-temporal Preferences, Impact Evaluation, Randomized Control Trial Sector Board: FSE * Miriam Bruhn and Bilal Zia are senior economists at the World Bank Development Economics Research Group. Arianna Legovini is head of the Development Impact Evaluation Unit at the World Bank. Luciana Leão was the field coordinator of this study and now a Ph.D. student at Columbia University. Rogelio Marchetti is task team leader of this project and senior financial sector specialist in the Latin America division of the World Bank. We are deeply indebted to all in-country partners without whom the project would not have been possible, including CAEd/UFJF, the Securities and Exchange Commission of Brazil (CVM), Central Bank of Brazil (BCB), the Complementary Pensions Secretary (Previc) and the Superintendent of Private Insurance (SUSEP), BM&F Bovespa, Anbima, Febraban, Instituto Unibanco, the Ministry of Education (MEC), and State Education Departments from Rio de Janeiro, São Paulo, Distrito Federal, Ceará, Tocantìns, and Minas Gerais. And finally, individual and special thanks to José Alexandre Vasco of CVM for leading the design and implementation of the financial literacy program, and Alzira Silva and Célia Bittencourt for their outstanding dedication and support. We also thank Christian Salas for research assistance. Financial assistance from CVM, BM&F Bovespa, Instituto Unibanco, the World Bank Development Impact Evaluation Unit, the World Bank Brazil Country Management Unit, and the World Bank Russia Trust Fund on Financial Literacy is gratefully acknowledged. 1 1. Introduction The fast rate of financial development around the world has made financial products and services widely available, yet such proliferation has consistently outpaced the capacity of individuals and families to make informed financial choices (Lusardi and Mitchell, 2007; Lusardi et al., 2010). Personal financial decisions are further complicated by the rapid infusion of cleverly marketed consumer products that are often coupled with expensive credit and installment plan offers. These advancements have heightened the risk of misplaced or misinformed spending decisions, especially in an environment where individuals have trouble understanding even basic financial concepts (Lusardi and Mitchell, 2011a; Xu and Zia, 2012). Indeed, the personal bankruptcy rates even in developed countries such as the US have skyrocketed, increasing by as much as 20 percent annually during the recent financial crisis. 1 The policy response to these troubling trends has been to introduce proactive measures on both the supply and demand side. On the supply side, many countries now have Consumer Protection Bureaus that are tasked with ensuring financial providers adopt and adhere to transparent consumer disclosure laws. On the demand side, there has been a significant push to focus on basic financial concepts and educate the public through various financial literacy programs. Many of these programs take the form of relatively short workshops that are geared toward adults. However, rigorous evidence on the impact of these types of courses shows only marginal improvements in financial behavior (Cole et al., 2011) and these effects tend to dissipate over time (Fernandes et al., 2013). Also, interest and participation in adult financial education workshops tends to be low, so they do not appear to be an effective way of reaching broad audiences (Bruhn et al., 2013). An alternative approach to financial education is to introduce it early, in formative schooling years. Calls for such school based programs are widespread: the U.S. Consumer Financial Protection Bureau now officially recommends introducing and building key financial education concepts throughout the K-12 school years (CFPB, 2013). 2 The UK government has mandated financial education to become part of the national secondary school curriculum starting in September 2014, and the Australian government has voted for the same. Many developing countries, including Brazil, Colombia, India, Indonesia, and Uganda, are introducing financial education in their school systems as well. The focus on youth is attractive and relevant for a number of reasons. First, good financial habits formed at an early age are likely to benefit schooling, employment, and standards of living throughout adulthood. Second, the focus on youth leverages their learning capacity as students who are primed to absorb, recall, and apply learning. Students attend school regularly and do not have to make extra time or come especially for financial education classes, which has been a 1 US Bankruptcy Abuse Prevention and Consumer Protection Act Report, 2011. 2 Likewise, the US Treasury Department recently launched a financial learning website that offers teachers lessons appropriate for instruction in math and English classes: http://www.moneyasyoulearn.org 2 problem with adult workshops. Further, well-informed students have the opportunity to modify not only their own financial choices, but also to act as agents of change in their households’ financial decisions. Yet, despite the potential benefits of financial education for youth, we simply do not know what works and the few existing studies show either conflicting results, are too narrowly focused, or suffer from important identification concerns. 3 This paper addresses these shortcomings by using a randomized control trial to study the impact of a comprehensive financial education program for public high school students in Brazil. The program spanned 17 months and was integrated in classroom curricula of Mathematics, Science, History, and Portuguese. The instruction used new textbooks with interactive classroom exercises on financial education themes, take-home exercises such as creating household budgets with parents, and role playing assignments. The curriculum was complemented by teacher training, web learning tools, and instructor handbooks. As such, the intensity of treatment of this program was much stronger than typical one-off financial education workshops. To date, our study is the largest randomized evaluation in the financial education literature, covering 868 public high schools in six Brazilian states and approximately 20,000 students. As part of our study design, schools were first stratified by state, pair-wise matched by school and community characteristics, and then randomly assigned to treatment and control. To measure the effects of the financial education program on financial knowledge and behavior, we collected data in schools through three rounds consisting of baseline (August 2010), follow-up 1 (December 2010), and follow-up 2 (December 2011). Follow-up survey results show that the program caused a quarter of a standard deviation improvement in students’ financial knowledge, as measured by an SAT-like financial proficiency test. In fact, the entire distribution of scores shifted to the right with students at all levels of capability showing marked improvements in test scores. On financial behavior, we find a statistically significant increase of 1.4 percentage points in saving up for purchases (compared to 12.9 percent in the control group), and significant improvements in the likelihood of making budgets and negotiating prices and payment methods. We also examine “trickle-up” effects of the program on parents to test whether students can act as agents of change in their households. Many of the take-home exercises involved interaction with parents such as making household budgets or researching and comparing interest rates. We survey parents and identify several findings. As proof of concept, parents in treatment schools were significantly more likely to report that their children discuss financial matters with them at home and that they volunteer to help organize household budgets. In addition, we detect improvements in parental financial knowledge on standard financial literacy questions used in the literature. And finally, we find significant improvements in parental financial behaviors, with an increase of 0.67 percentage points in the savings rate (from a control group mean of 12.2 percent) and improvements in the likelihood of keeping household budgets. 3 See section 2 of this paper for a detailed discussion of the existing literature. 3 In order to reinforce the effect of school financial education, the U.S. Consumer Financial Protection Bureau recommends educating parents at the same time as students. Our intervention complemented the student program with a standard adult workshop on financial education for parents. This involved a DVD-based intervention where parents in treated schools were randomly assigned to either a financial education screening or a health education screening. Although the attendance at these workshops was low, we detect further improvements in the percentage of disposable money saved by students from families that attended the financial education workshops. Hence, these workshops helped parents reinforce the messages taught to the students. We then study the mechanisms of impact to shed light on how and why the program influenced students’ financial behavior. The existing literature in economics and psychology highlights the importance of time preference and self-control for behavior and economic decision making (Becker and Mulligan, 1997; Frederick, et al., 2002; Mullainathan and Shafir, 2013). Among youth, the ability to control temptation and delay gratification has been shown to be an important determinant of lifetime academic, economic, and social outcomes (Mischel and Rodriguez, 1989; Duckworth and Seligman, 2005; Sutter et al., 2013). The literature also studies the influence of willpower in inter-temporal choices and finds that low willpower, lack of self-control, and impatience are related to poor economic and financial outcomes (Kocher et al., 2012). Importantly, a number of papers argue that willpower and self-control resemble a muscle that requires time and resources to replenish and becomes stronger with repeated practice (Baumeister and Heatherton, 1996; Baumeister et al. 1998; Muraven and Baumeister, 2000; Baumeister et al. 2007). The program we study provided repeated instruction and exercises, spanning three academic semesters, from August 2010 to December 2011, implying that it may well have changed students’ inter-temporal preferences and attitudes. Identifying any such changes in preferences would highlight an important intermediary channel of measured behavior change. 4 To measure students’ financial preferences, new indices of financial autonomy and intentions to save were constructed with the assistance of a local education survey firm. The financial autonomy index aggregates a series of questions designed to measure whether students feel empowered, confident, and capable of making independent financial decisions and influencing the financial decisions of their households. The intention to save index includes a series of questions that identify preferences over hypothetical savings and spending scenarios. Our analysis finds strong and statistically significant treatment effects on both these measures, with an effect size ranging between 0.08-0.12 of a standard deviation. 4 There may also be a feedback loop from financial behavior to preferences. For example, individuals may become more patient as their savings increase, but the empirical evidence on this channel is inconclusive (Carvalho, et al., 2013). 4 Finally, an important concern with adding financial education to the school curriculum is that it may crowd out learning in other areas. We analyze administrative data on student graduation rates and find an improvement in treatment schools, with a 1 percent higher passing rate (compared to a graduation rate of 84 percent in the control group). While this effect is small, the positive and significant coefficient allays concerns that the financial education curriculum diverted attention of teachers and students away from regular academic learning. Overall, our study shows that financial education can be an effective tool in improving financial outcomes of students when delivered in a comprehensive manner and over a significant period of time. Also, key complementary benefits can be derived by involving the entire household, students and parents, as indicated by the trickle-up and parents’ workshop impacts. Further our results on mechanisms show that student preferences on saving intentions and financial autonomy are important pathways for improved financial behavior. Finally, the large sample size and breadth of coverage through six states in Brazil provides strong support for the external validity of our findings. This paper proceeds as follows. Section 2 summarizes the literature on financial education for youth and details the Brazilian context. Section 3 describes the financial education curriculum, and Section 4 presents the research and sampling methodology, as well as the study timeline. Section 5 describes program take-up and implementation, and Section 6 presents summary statistics and survey participation analysis. Section 7 discusses the main results, and Section 8 analyzes mechanisms of impact. Finally, Section 9 concludes. 2. Literature and Context 2.1. Financial Education for Youth There is a growing literature on financial education and its determinants for youth. A consistent finding in this literature is that financial literacy tends to peak among adults in the middle of the life cycle and is significantly lower among youth. In the U.S., for example, those in the prime age group (25-65) tend to perform about 5 percent better on financial literacy questions than those under 25 (Lusardi and Mitchell, 2011b). Strikingly, Lusardi et al. (2009b) find that less than a third of American teenagers (ages 12-17) possess basic knowledge of interest rates, inflation, and risk diversification. Mandell (2006) notes that there is even evidence that youth financial literacy has been declining in the U.S. since the late 1990s. Similar evidence comes from Australia where Beal and Delpachitra (2003) identify low levels of financial literacy among youth. Perhaps in response to such trends and combined with the aftermath of recent global financial crises, policy makers around the world have made financial education for youth a priority, with many school based initiatives now part of education reform. Yet, the impacts of such school programs on financial knowledge and behavior are still not well understood. Existing studies 5 show contradictory results and many of them suffer from identification issues. For example, Bernheim et al. (2001) employ a difference-in-difference approach to analyze the impact of state high school financial education mandates on savings behavior in the U.S. and find that mandates appear to effectively increase exposure to financial education, and have a significant subsequent effect on future savings. However, Cole and Shastry (2009) replicate and extend the analysis using a much larger sample from U.S. census data and find, in contrast, no significant impact of high school financial education on future savings. Other impact evaluations of financial literacy programs additionally rely on small samples, which raises concerns about the external validity of their findings. Carlin and Robinson (2010) examine a financial literacy course for high school students in the U.S. The course provided 19 hours of financial training that included credit card management, taxes, budgeting, and simple investments. Students’ financial skills are measured in role-playing games of fictitious budget situations at visits to a “Finance Park”. The authors find that training raised completion rates (successfully crafting a balanced budget) from 5 percent to over 50 percent and savings increased four-fold. However, the identification strategy relies on comparing outcomes of 125 students before and after the financial literacy course, so that it is not clear whether improvements are due to the course, repeated visits to the park, or other time-varying factors. Similarly, Varcoe et al. (2005) conduct a pre-post analysis to evaluate a program (Money Talks) in which 13-18 year-olds in various settings, including public high schools, received four newsletters which covered different topics, such as savings habits, shopping tips, car costs, and money values. Using a sample of 114 students, the authors find that students show both greater financial knowledge and improved financial behavior after the course compared to before. Walstad et al. (2010) study the effects of a DVD-based curriculum for high school students – Financing Your Future. The five video segments cover such topics as saving, money management, banking, credit and debt, and investing, and add up to six hours of instruction. They find that 673 students who participated in the education program showed a significant gain in financial knowledge (as measured by pre-test and post-test scores) compared to 127 students in a matched control group, but they do not study subsequent behavior change. Luhrmann et al. (2012) evaluate 90 minute financial education sessions delivered to 14-16 year old students in lower stream German high schools. They compare 558 treatment students to 158 control students and find significant improvements in financial knowledge and a hypothetical savings scenario. However, schools are chosen to receive financial education sessions based on how busy teachers feel they are with students prior to the end of the academic year, which raises serious selection concerns since teachers in control schools may also be dealing with relatively poorly performing students or other unobserved underlying student, class, or school characteristics. 6 Randomized evaluations of school based financial education programs are scarce; in fact we are aware of only two other such studies. Berry et al. (2012) conduct an evaluation of a program offering voluntary after-school clubs in Ghana for primary and junior high students in 135 schools over a 10 month period. The study randomly assigns a group of 5th and 7th graders to a social and financial treatment and another to basic training. It is not clear how exposed students are to financial decision making at such a young age, and eliciting reliable and consistent responses in this age group is generally difficult; hence there are some measurement concerns. The treatment is also short and participation voluntary. The findings are unsurprisingly muted – while the study identifies some effects on savings, there are no improvements in financial knowledge, test scores, or social and psychological measures. Another experiment is a study among 17 to 19 year old high school students in Italy by Becchetti et al. (2011), covering 944 students in 36 classes. The authors offered a 16 hour long course on finance over three months. Difference-in-difference estimates show no statistically significant effect on financial knowledge. Instead, both the treatment and control groups show improved knowledge over time, suggesting that students adapt to repeated financial literacy tests. Becchetti and Pisani (2012) extend the sample to 3,820 students in 118 classes. They now detect positive effects of the course on financial knowledge in a difference-and-difference analysis. Surprisingly, however, treatment group students already performed better on the financial knowledge test before the course, raising questions about the implementation of the random assignment. The paper does not examine effects on financial behavior or attitudes. Against the state of the existing literature on high school financial education, our study fills an important gap by bringing together: (1) a randomized evaluation methodology, (2) a comprehensive financial education intervention that was mandatory for students and lasted 3 semesters over 17 months, (3) a very large sample size (868 schools and nearly 20,000 students), (4) widespread coverage over six states in Brazil, (5) a measure of financial knowledge that changes from survey to survey to avoid remembered answers, while still being comparable over time, (6) studying impacts on financial behavior, (7) a unique set of measures of financial attitudes and preferences developed specifically for the youth segment of the population, (8) analysis of “trickle-up” effects for parents, and (9) estimating the additional effects of a complementary workshop for parents. As such, our study offers new insights into the impact of financial education for youth and their parents. 2.2. The Brazilian Context Brazil has a relatively low national saving rate of around 16 percent of GDP and levels of financial awareness are also low. For example, a survey conducted by the Instituto Data Popular in 2008 found that 82 percent of Brazilian consumers were unaware of the interest rate when borrowing money. To address these issues, policy makers set up a working group consisting of 7 representatives from the financial sector including public, civil society, and private institutions.5 In 2010, the working group launched a National Strategy for Financial Education (ENEF) with the goal of fostering a culture of financial education in the country and enabling citizens to make sound financial decisions. The strategy’s scope is national, targeting children, youth and adults, and its goals ambitious, involving a large set of actors and a multiplicity of delivery mechanisms for financial education. 6 One of the first pilot projects of the ENEF was a financial education program for public high schools. This program was developed in collaboration with the Federal Ministry of Education and State and Municipal Secretariats of Education. We joined the project team to help assess whether the high school pilot program had the desired effects on financial knowledge, attitudes and behavior. The results of our study would be used to inform improvements to, and the eventual national scale up of, the program. 7 3. Financial Education Curriculum The financial education curriculum was developed by a team of education experts, psychologists, and sociologists in Brazil who were commissioned by ENEF. The content encompasses didactically innovative material designed to capture the interest of young adults and be relevant to their lives. It consists of 72 case studies that can be integrated into regular school subjects such as Mathematics, Portuguese, Science, Geography, and History. The decision to incorporate financial education into other subjects instead of creating a separate subject was taken after reviewing impacts on teacher and student workloads. In contrast to typical workshop based financial literacy programs that are delivered in one shot and vary in length from 90 minutes to a few hours, the case study based program provides textbooks with material for between 72 and 144 hours of teaching (1-2 hours per case study), spread out over one and a half school years (three semesters). The material is interactive and includes exercises that students complete with their parents such as household budgeting. The student textbook is divided into three blocks and covers nine different themes: family life, social life, personal property, work, entrepreneurship, large expenditures, public goods, country economy, and world economy. 8 These themes were chosen to provide a solid foundation of 5 Participating public sector institutions included the Brazilian Securities and Exchange Commission (Comissão de Valores Mobiliários, CVM), the Central Bank, and pensions and insurance regulatory agencies (Superintendência Nacional de Previdência Complementar, PREVIC, and Superintendência de Seguros Privados, SUSEP). The group also included the association of financial institutions (Associação Brasileira das Entidades dos Mercados Financeiro e de Capitais, ANBIMA), BM&F BOVESPA (the stock exchange), and the federation of banks (Federação Brasileira de Bancos, FEBRABAN). 6 After the working group fulfilled its purpose, a new committee, the National Committee on Financial Education (CONFEF – including Minister of Education, Minister of Justice, Ministério da Fazenda e Ministério da Previdência Social) was created to spearhead the financial education agenda. 7 CVM coordinated the pilot project, with collaboration and support from BM&F BOVESPA, ANBIMA, FEBRABAN, Instituto Unibanco and Citi Foundation. 8 A detailed outline of the curriculum is provided in Appendix 1. 8 knowledge on financial issues and also to address the specific financial education needs of high school students. More specifically, the program aimed at providing students with tools to overcome financial challenges they were facing or about to face, such as family finances, entry into the labor market or further studies, and preparing for financial independence. The themes are taught through 72 case studies called didactic situations (SDs) that include theoretical and applied content, exercises and activities, as well as self-evaluative questions. SDs make use of texts, stories, images, and tables to convey the material in an accessible way. SDs also contain “experiment” sections that are designed to make the material relevant to students’ daily life. For example, when discussing the international economy, students are asked to identify the imported products they use in everyday life. Each SD concludes with a short outline of what students are expected to learn. Teacher guidelines explain how to integrate the financial education case studies into the regular curriculum, and teachers have discretion over the order in which the cases are taught. The program trained teachers in advance through an introductory seminar that lasted 1 to 2 days, a reference DVD, and made available a training website with distance education material which remained active throughout the program. Teacher training was also important for generating interest and support for the program, which ultimately led to a large fraction of teachers applying the financial education material (see section 6.1 for discussion of take-up rates). 4. Research Design, Sampling, and Timeline 4.1. Research Design We use a randomized control trial to measure the causal effects of the high school financial education program. Treatment was assigned at the school level. Treated schools received financial education material and teacher training. Control schools did not receive any material or training, but participated in surveys and testing in the same manner as the treated schools. One eleventh grade class in each school participated in our study. Treated classes received financial education during the second semester of eleventh grade (fall 2010) and the second and third semesters of financial education throughout twelfth grade (spring 2011 and fall 2011), the last year of high school. Students in the sample were between 15 and 17 years of age at the start of the intervention. In addition to the school financial education program, we also study the impact of a complementary parent intervention in treatment schools. 9 During the second semester of financial education (spring 2011), parents of students in treatment schools were invited to participate in a school workshop. At the workshop, they were directed to watch either a financial education video (workshop treatment group) or a health education video (workshop control group) through random assignment. Parents learned which video they had been assigned to 9 Due to political constraints, we were unable to extend the parent intervention to control schools. 9 watch only after they arrived at the workshop. The purpose of the parents’ workshop was to raise awareness among parents about the importance of financial education and to encourage them to interact more with their children on financial matters, thereby leveraging and reinforcing the material students were taught through the school program. The reason for showing a health education video to the workshop control group was that our counterparts did not think it was logistically and politically feasible to invite only certain parents within a school to a workshop, but not others. We do not expect the health video to have an impact on financial outcomes. Figure 1 summarizes the study design. 4.2. Sample Selection for Student Intervention Brazil is divided into 26 states and a Federal District. The Federal District and five states were part of the study, including three of the most populous and developed states (São Paulo, Minas Gerais, and Rio de Janeiro), and two states (Ceará and Tocantins) that represent the less developed areas of Brazil. In 2009, the Federal District had the highest GDP per capita in Brazil (US$ 28,951). Minas Gerais, Rio de Janeiro and São Paulo are located in the Southeast, with GDP per capita of US$ 8,289, US$ 12,524, and US$ 14,872, respectively. Ceará and Tocantins are located in the Northeast and North of Brazil with GDP per capita of US$ 4,399 and US$ 5,960, respectively. In April and May 2010, the Secretariat of Education in each state assembled a list of public high schools that volunteered to participate in the financial education pilot program, totaling 815 schools. We divided these schools into a treatment and control group through stratified and matched randomization as soon as we received each list, so that the teacher training for treatment schools could be organized and conducted before the mid-year school break in July. São Paulo being the state with the largest number of schools sent four separate lists on different dates. In addition, one of the project’s partner institutions (Instituto Unibanco) provided us with an additional list of 101 public schools that they partnered with in Rio de Janeiro, São Paulo, Minas Gerais, and the Federal District. For each list, we first stratified schools by whether they were located in a municipality with above or below median number of financial institutions per capita. Within strata, we formed matched pairs of schools and randomly assigned one school in each pair to be in the treatment group and the other school to be in the control group. We matched on the following school and municipal variables to improve balance on these characteristics across the treatment and control groups: GDP per capita of the municipality, savings volume per capita of the municipality, school location, number of students in the school, number of teachers in the school, school drop- out rate, and school continuation rate (percent of students moving on to the next grade). 10 We 10 For Ceará, we did not have number of teachers, drop-out rates, and continuation rates, so we matched on the remaining variables only. For Tocantins, we did not have drop-out rates and continuation rates, so we matched schools on the remaining variables only. Also, since all schools in the Federal District and in Minas Gerais (Juiz de Fora) were located in the same municipality, we only matched on school level variables in these states. For the Rio de Janeiro and São Paulo schools from the 10 chose these variables since they may be correlated with the impact of the material on financial knowledge, attitudes, and behavior. We were not able to match on variables collected through our surveys because the randomization had to take place before the baseline survey to enable the program to train the teachers on time. Municipal level variables used in the randomization come from the Brazilian Statistical Institute (IBGE) and refer to 2009. School level variables used in the randomization are for 2008 and were provided by the Federal Ministry of Education (2009 data were not yet available at the time of randomization). Randomization was done by us and through a computer, implying that any differences across the treatment and control group are due to pure chance. After the randomization was completed, we had to move three control schools to the treatment group manually since some states requested that at least one school in each school district participate in the program. We chose these schools at random among the schools in the school district and drop them and their pairs from the analysis. Also, after the randomization was completed but before the program was implemented, we discovered that twelve schools indicated by the Instituto Unibanco decided not to participate in the pilot. We drop these schools and their pairs from the analysis also. In addition, five schools had accidentally been listed twice (three were on both the state ministries’ and Instituto Unibanco lists and two were duplicated on the state ministries’ lists). We randomly chose which entry to drop (along with their pair). Another six treatment group schools from São Paulo did not participate in any of our surveys for unknown reasons. We drop these schools from our sample along with their pairs. In the end, we are left with 868 schools for the analysis – 432 in the treatment group and 436 in the control group. The majority of the schools are located in the most populous states, São Paulo, Rio de Janeiro, and Ceará. 4.3. Sample Selection for Parent Intervention For the implementation of the parent workshops, treatment schools sent in a list of current students enrolled in the class receiving financial education during the spring of 2011. Only schools that sent the list were included in the study as this was taken as a signal that they were willing to implement the workshop. We used these lists instead of relying on information from earlier surveys because of potential student turnover from one school year to the next. A total of 264 treatment schools provided a list, covering 8,534 students. We matched the lists with data from the baseline parent survey on the basis of the student’s name, and stratified and randomly assigned parents in each school into treatment and control groups by computer. We used the following strata: (1) no baseline information on parents; (2) parent had low baseline financial literacy; and (3) parent had high baseline financial literacy. We defined the level of baseline financial literacy based on the number of correct answers to the two Instituto Unibanco list, we stratified by financial institutions per capita, but we only matched on school-level variables, not on municipality level variables, since many of these schools were located in the same municipalities, restricting the possible matches for the remaining schools. 11 standard financial literacy questions on interest rates and inflation asked on the survey. 11 About 41 percent of parents did not answer either question correctly. We classify these parents as having low baseline financial literacy. Parents who answered one or two questions correctly are classified as having high financial literacy. Each school that provided a list of their students received two separate lists in return: one with the names of the students whose parents would watch the financial education video during the school workshop (treatment group) and one with the names of students whose parents would watch the health video (control group). Schools were provided the financial education and health videos and parent exit questionnaires. Each school was responsible for organizing and implementing the workshop at a time of their choosing and administering the questionnaire at the end of the workshop. 4.4. Empirical Specifications Based on the random assignment, the impact of student financial education is measured as the difference in average outcomes in the treatment and control groups using the following intention- to-treat OLS regression: yi,s,f = α + βTreatment i,s + ∑ γs ds + δyi,s,b + εi,s (1) where yi,s,f is a measure of the financial knowledge, attitude, or behavior, of student or parent i R in school pair s at follow-up f. The variable Treatment i,s indicates whether an individual is in a school that was randomized into treatment or not and is thus equal to one for the treatment group and equal to zero for the control group. We include school pair dummies, d s in all specifications and control for baseline values of available dependent variables, y i,s,b as per Bruhn and McKenzie (2009) and McKenzie (2012). When baseline values have missing observations, we replace these with zero and include a dummy variable indicating that the observation was missing. Standard errors are clustered at the school level. Finally, note that results from the first and second follow-up surveys are not fully comparable due to changes in class composition from one year to the next, as discussed in Section 6.3. The analysis of the parent workshops is based on workshop attendance rather than invitation, hence the treatment variable in specification (1) takes on the value of 1 if the parent watched the 11 These two questions are: Question 1. Assume that you would like to get a loan for R$ 50,000. Bank A offers you a loan of R$ 50,000 that needs to be repaid after one year with an interest rate of 15% payable at the end of the loan. Bank B offers you a loan for the same amount, R$ 50,000, for which you will have to pay R$ 60,000 also after a period of one year. Which option do you prefer? A) Bank A, B) Bank B, C) Don’t know. Question 2. Assume that the interest rate on your savings account is 1% per year and the inflation rates 2% per year. After one year without making any transactions on our saving account, would you be able to purchase more, the same amount, or less than what you could purchase today with the money in the account? A) More than today, B) The same as today, C) Less than today, D) Don’t know. 12 financial education video, and 0 if the parent watched the health education video. As mentioned before, these assignments were unknown to the parents prior to arriving at school, and both videos were shown simultaneously and in identical settings. 4.5. Study Timeline The sample selection and randomization for the student intervention occurred in April-May 2010. A baseline survey was conducted in early August 2010 among students and parents in both treatment and control schools. The financial education program was rolled out immediately after. By mid-August 2010, teachers began using the financial education materials in the classroom. The program continued until November 2011 for a total of three school semesters. Over the course of this study, two rounds of follow-up surveys were conducted. The first follow- up survey was implemented in early December 2010, four months after the program started. The results of this survey measure the short-term effects of the program. A second follow-up survey was implemented in December 2011 to assess the longer-term impacts. The parent intervention was introduced in May 2011, and parent outcomes recorded through an exit survey and the December 2011 follow-up survey. 5. Program Take-up and Implementation 5.1. Student Program Following school assignment to treatment and control, the program distributed textbooks to treatment schools and organized training sessions for the teachers. The vast majority of treatment schools received the financial education textbooks and distributed them to students. In both rounds of the follow-up survey, we asked school principals as well as teachers and students a series of questions regarding implementation and usage of the financial education program. Follow-up 1 spans the first semester of the program, while follow-up 2 covers the second and third semesters. Over 95 percent of treatment school principals report that they received the textbooks for the first semester, and 93 percent report receiving them for semesters 2 and/or 3. The large majority of teachers also say that students received the textbooks (i.e. the books were actually distributed to students), 94 percent and 92 percent in follow-up 1 and 2, respectively. On training, 78 percent of teachers report that they received training on how to use the financial education material in the first semester, and 65 percent report receiving it in the second/third semesters. In terms of usage, 87 percent of students report that teachers actively used the financial education textbooks in classrooms in the first semester, though a drop-off occurs in semesters 2 and 3, with 74 percent of students reporting usage. However, the percentage of principals reporting that 13 financial education was taught in school remained high throughout the study period (93 percent for all semesters). Schools in the control group did not receive textbooks or teacher training through the financial education program studied in this paper, but they may have implemented other types of financial education. The principal and teacher questionnaires were only applied in treatment schools during the first follow-up survey, but during the second follow-up survey control schools also answered these questionnaires. 12 In this survey, 16.6 percent of control group principals stated that the school had a financial education program and 11 percent of control group teachers reported receiving some training related to financial education. We do not have detailed information on the financial education program implemented in control schools. However, only 5 percent of control group principals said that the school received a textbook with financial education material, suggesting that these programs may be less intensive than the one studied in this paper. We supplemented this quantitative analysis with qualitative work in the form of teacher focus groups, which were organized with our Brazilian counterparts. The purpose of these meetings was to learn how teachers implemented the program, how material was integrated into the regular curriculum, and how the contents and/or delivery could be improved for future implementation. Six such meetings were held in September and October 2011, with teachers and educators from all six states. All meetings were conducted in facilities provided by the State Education Departments and meals were provided. The meetings were generally well attended (more than 200 educators attended in São Paulo), and lasted an average of four hours each. In all states, the teachers and educators greatly approved of the textbooks and the financial education program. They said that the material allowed students to learn by themselves, that students liked the case studies and felt that they connected well with their daily life situations. The teachers also complemented the clarity with which concepts were conveyed in the books. One teacher commented, “What motivates me the most about this project are the books. The content is directly related to students’ lives and helps to insert them in a highly-competitive market society. I learned a lot from the material, and similarly to what happened to my students, it has helped me to plan better for the future.” In terms of implementation, most teachers employed work group strategies to teach the material, where students were divided into small groups and asked to work on different tasks. Examples of such tasks varied from identifying steps for opening a small firm to creating a school market. Several teachers reported assigning projects to students related to fundraising and organizing their own graduation party. Other teachers reported undertaking field trips to local markets, 12 The student questionnaire did not include questions on program implementation in control group schools in either follow-up survey. 14 universities, and companies to learn how they operate. Others used various forms of media to explore the topics in the textbook, such as computer simulations and videos. Overall, teachers and educators felt the financial education program was valuable, that the textbooks were extremely relevant, and that they as well as their students enjoyed the new learning opportunities afforded to them. 5.2. Parent Workshops For the parent workshops, we asked schools to mail the filled-in parent exit questionnaires back to us after the workshop, and we determine which schools implemented the workshops based on our receipt of these questionnaires. We received 1,553 filled-in parent questionnaires from 109 treatment schools, implying that parent workshops did not take place in the other 153 treatment schools that had originally provided a current list of their students for the parent intervention (these schools either did not organize the workshop or parents did not attend the workshop). When planning the parent intervention, our counterparts had cautioned that attendance rates may be low since parents of public high school students in Brazil tend to not participate in school events. Comparing the lists of current students we received from schools before the workshop to the filled-in exit questionnaires gives an average attendance rate of 46 percent across the 109 schools that returned questionnaires. The attendance rate does not differ across parents who were randomized into the treatment and control groups. Parents did not know whether they had been assigned to watch a financial education or health video until the video was screened at the workshop. That is, the decision to attend was independent from treatment status. Hence, in our impact analysis of the parents’ workshop, we only keep students and parents if the parent attended a workshop. The interpretation of the results is applicable to families who are more interested in school events and respond to school invitations. 6. Survey Data and Summary Statistics 6.1. Survey Data For the data collection, we partnered with the Centro de Políticas Públicas e Avaliação da Educação (CAEd/UFJF), a local education research and survey institute with extensive experience in designing and implementing knowledge tests in Brazilian schools. CAEd has a wide network of supervisors and surveyors across Brazil and was able to implement simultaneous surveys in all schools in our sample. Survey implementation took three days per school during each survey round (baseline and both follow-ups). On the first day, CAEd staff administered a financial knowledge test and distributed parent questionnaires to students. The students were instructed to take the parent questionnaire home, ask one of their parents to fill out the questionnaire and return the questionnaire on one of 15 the following days. On the second day, students filled out a self-administered questionnaire measuring financial attitudes and behavior. The third day provided an opportunity for any student who had missed one or both of the previous days to fill out the test and/or questionnaire. The student tests and questionnaires were administered in the classroom in the same way as a regular school exam, i.e. distributed to students, supervised by the surveyor and collected by the surveyor at the end of the allocated time. School teachers were not privy to the tests or questionnaires beforehand, and were not involved in proctoring them. 6.2. Baseline Summary Statistics Table 1 shows pre-program summary statistics of school and student characteristics for treatment and control schools. The school level variables are for 2008 and were provided by the Federal Ministry of Education. As a result of the school matching and randomization procedures described in Section 4, the pre-program school characteristics were the same on average in the treatment and control groups. For example, the drop-out rate was about 10 percent and the class passing rate was 74 percent in both treatment and control schools. The remaining variables in Table 1 were collected through the baseline survey that was conducted in August 2010. Student background characteristics show that 56 percent of students participating in the study were female, 67 percent had some form of income (from work or from parents) and about 35 percent were working at baseline. Additionally, 33 percent were beneficiaries of the Bolsa Família government cash transfer program, indicating that they belonged to low-income households. About 60 percent of the students’ parents had less than high school education. In terms of financial characteristics, only 11 percent made a list of their expenses every month, but 75 percent negotiated the price or payment method when making purchases. There were no differences in baseline levels of financial proficiency, financial autonomy, or intention to save. Overall, the data in Table 1 indicate that students’ background and financial characteristics were the same across the treatment and control groups at baseline, as expected since treatment status was randomly assigned. 6.3. Survey Participation and Attrition Table 2 shows the number of schools, students and parents that participated in the surveys at baseline and at follow-up. At baseline, 866 of 868 schools in our study sample participated in the survey, although two of these schools did not implement the financial literacy test and the parent questionnaire. In each follow-up survey, about 40 schools did not implement the survey. Reasons for non-implementation varied and were mostly related to scheduling difficulties. The schools that did not participate in the follow-up surveys were different in each round, so we have follow- up data for most schools from either follow-up 1 or follow-up 2. After realizing that survey participation had dropped between baseline and follow-up 1, we provided incentives for survey completion during follow-up 2. Both treatment and control 16 schools where more than 75 percent students completed at least 80 percent of the survey questions were entered into a lottery for one of twenty-five computers. Despite this incentive, we did not have a higher number of schools participate in follow-up 2 than in follow-up 1. However, a greater number of students within the participating schools answered the surveys in follow-up 2 than in follow-up 1. The number of students surveyed per school in follow-up 2 is still lower than in the baseline (about 22 vs. 28 students per school). One reason for this decline in the number of students per school is that the drop-out rate is quite high in our sample (about 10 percent per year at baseline). Note that student rotation is also common in our study sample. As shown in Table 1, 30 percent of students reported that they had repeated at least one school year at baseline. In part due to high drop-out and repetition rates, some schools reshuffle classrooms from one year to the next. In our sample, the majority of students in follow-up 1 were present in baseline (follow-up 1 was implemented in the same semester as the baseline, about four months apart). However, only about 60 percent of students in follow-up 2 were present at baseline (follow-up 2 took place about 16 months after baseline). For the financial education program, this high rotation implies that more than a third of the sample was not exposed to the material for a full three semesters, but rather for only one or two semesters. With respect to the parent questionnaire, Table 2 shows that participation was quite high, considering that this questionnaire was self-administered at home. About 88 percent of students returned parent questionnaires at baseline. In the follow-up surveys, this number dropped to about 76 percent. To test whether survey attrition differed across treatment and control schools, we perform the following analysis for follow-up 2. We use enrolment counts at the beginning of the fall 2011 semester to calculate response rates to the survey conducted at the end of the semester. These enrolment counts were available for 631 schools in our sample. We then run a regression of response rates on a treatment dummy and randomization school pair dummies. The results are reported in Table 3 and show that survey attrition is uncorrelated with treatment status for both the student and parent surveys. The response rate for the student (parent) survey was about 80 (55) percent in both treatment and control schools. 7. Analysis and Discussion of Main Results 7.1. Student Financial Proficiency Table 4 shows the impact of the school financial education program on students’ financial proficiency score. The proficiency tests were tailored to the program’s material and objectives, and the same tests were administered in both treatment and control schools. Tests were constructed using item response theory (which is also used to construct the GRE and SAT in the United States), and were comparable across students and across time. This implies that a series 17 of equivalent questions were used to test the same concept, leading to different combinations of questions on each test. The questions that each student received were thus likely to be different in each survey round and different students within the same class received different questions. This minimized the risk that students simply remembered correct answers from previous rounds without truly understanding the question and also reduced the scope for cheating. Teachers were not privy to the tests at any point prior, after, or even the day of the tests and CAEd proctors were present in classrooms for the entire duration. No awards were given for performance on tests. These steps protect the analysis from concerns of teachers teaching to the test. Student financial proficiency was calculated and scored on a scale of 0 to 100. Table 4 shows the average level of financial proficiency was significantly higher in the treatment group than in the control group in both follow-up 1 and in follow-up 2. The difference is of 3.5 points and 3 points, respectively, corresponding to a 5 to 7 percent increase in financial knowledge. These increases are equivalent to a quarter of a standard deviation improvement by follow-up 1 and a fifth of a standard deviation improvement by follow-up 2. Further, as shown in Figure 2, test scores improved across their distribution benefiting low and high achieving students. Specifically, the proportion of students that performed exceptionally well increased by 28 percent and the proportion of students that performed exceptionally poorly decreased by 26 percent. This effect represents a rightward shift in the distribution of test scores for treated schools compared to control schools. Hence, the financial education program helped poorly performing students to improve significantly, and well performing students to do even better. These distributional effects are important and show that the program benefited students along a broad performance spectrum rather than being driven by any one category, and that the curriculum speaks to the learning needs and interests of all types of students. Comparing these results with impacts in the previous literature is difficult because the small number of existing studies on financial education in schools is fraught with identification concerns, as discussed in Section 2.1. But there are school based studies outside the realm of financial education that are important for comparison purposes. The types of education interventions in secondary schools that have been tested through random control trials include the provision of monetary incentives to students and student tutoring. In general, these studies identify improvements in student learning during the periods studied, although the effects are not always statistically significant. Perhaps the most well-known studies of monetary incentives for secondary school students are Angrist and Lavy (2009) and Angrist et al. (2002). In the first study, Israeli students were provided cash incentives to pass their graduation exams, and while the mean estimates are positive, they are not statistically significant. In the second study, lotteries were used in Colombia to distribute vouchers to partially cover the cost of private secondary schooling for 18 1,600 students who maintained satisfactory academic progress. Three years after the lotteries, winners were about 10 percentage points more likely to have finished 8th grade, and scored 0.2 standard deviations higher on achievement tests, although the latter result is only marginally significant. Barrera-Osorio and Linden (2009) study the introduction of computers to supplement learning in Colombian schools and find negligible improvements in language and math test scores and limited use of the computers for academic learning in higher grades despite the program’s focus to do so. Compared to these studies, our finding of an increase of 0.2-0.24 standard deviations in financial proficiency lies in the top end of statistically significant improvements in test scores. 13 7.2. Purchasing and Spending Behavior Next we turn to analyzing changes in financial behavior. The student survey asked questions on outcomes related to topics emphasized in the financial education curriculum. Impulsive spending was an important behavior that the financial education program targeted. Consumers nowadays are exposed to a multitude of cleverly marketed consumer products, and falling prey to such temptation is particularly easy for young adults who are likely influenced by additional factors such as social norms, trends, and fashion. Furthermore, consumer items in Brazil such as cell phones are readily available to purchase on installment plans that come with large implicit (and often hidden) interest charges. A sizeable proportion of our student sample had part-time employment at baseline (35 percent) and an even greater proportion had disposable money available from other sources (67 percent), hence these temptations were real and meaningful. Against this setting, the financial education program highlighted the importance of responsible inter-temporal financial choices. Our expected channel and direction of impact has precedent in the literature. Existing evidence from psychology supports a mechanism by which individuals can be taught to self-regulate better. The literature identifies self-control as being important in overriding temptation (Barkley, 1997; Baumeister et al., 1994), and how abiding and following learned rules can help individuals exhibit self-control and delay gratification (Hayes, 1989; Hayes et al., 1996). Repeated practice of such rules can further strengthen self-control mechanisms (Baumeister and Heatherton, 1996; Baumeister et al. 1998; Muraven and Baumeister, 2000; Baumeister et al. 2007). Some of the rules that were taught and repeated over a 17-month period in our course and measured in the follow-up surveys included saving up for purchases instead of buying items on installments, comparison shopping, negotiating prices, and keeping track of expenses by making budgets. Table 5 shows treatment effects on these margins. We study saving for purchases on the extensive margin (whether you save for purchases?) in columns (1) and (2), and the intensive 13 There are also a number of papers that study classroom interventions in primary schools. These papers tend to find effects on test score between 0.15 and 0.47 standard deviations (for example, Banerjee et al., 2007; Cabezas et al., 2011; Muralidharan and Sundararaman, 2011; He et al. 2008; Linden, 2008). 19 margin (percentage of disposable money saved?) in columns (3) and (4). The results show statistically significant improvements in variables among students in treatment schools. Compared to 40 percent of students in the control group, the financial education program prompted an additional 5 percent of students in the treatment schools to save up for their purchases. This result is similar and highly significant (at the 1 percent level) in both follow-up rounds. Table 5 also shows that students’ average percentage of money saved increased significantly, by 1.4 percentage points in follow-up 2 (14.3 percent of disposable money saved in the treatment group as compared to 12.9 percent in the control group). 14 This 1.4 percentage point improvement represents a 0.07 standard deviation increase. We find significant treatment effects on spending behavior as well. In line with the concepts taught in the curriculum, the results show a 2.1-2.5 percentage point greater likelihood of comparison shopping before making purchases, and a 3.1-4.1 percentage point greater likelihood of negotiating price or payment method prior to purchases by students in treated schools. Further, the results show that 16 percent of students in treated schools made a list of monthly expenses as part of a budgeting exercise compared to 13 percent in the control schools in follow-up 1. These numbers are 17 percent and 14 percent respectively, in follow-up 2. All treatment effects are statistically significant at the 1 percent level. Overall, these results show a clear improvement in student behavior as a result of the financial education course, in line with concepts taught in the curriculum. 7.3. Student Participation in Household Finance Apart from students’ own financial behavior, we investigate whether students become more proactive in their households’ financial decisions. Indeed, the curriculum included several take- home exercises that were meant to be completed with parents such as making a household budget and exploring savings account options. Data on student participation in household financial decisions come from the parent questionnaires. The parent surveys included questions on socio-demographic characteristics, measured financial literacy through standard questions used in the literature, and elicited parents’ financial behavior regarding budgeting and savings. The parent questionnaires also asked whether parents discuss financial matters with students and whether students help to organize the household budget. We find that a significantly larger percentage of students in the treatment group talked to their parents about finances and participated in organizing the household budget. Table 6 shows that 71 percent of students in treatment schools participated in household financial decisions compared to 67 percent in control schools in follow-up 1, and 74 percent compared to 70 percent 14 This question was not asked in follow-up 1. 20 in follow-up 2. Due to the program, students were also significantly more likely to help organize the household budgets with significant improvements of 3.6-4.8 percentage points over the control group. 7.4. Parent Outcomes Next, we examine impacts of student financial education on parents’ financial knowledge and behavior. Financial knowledge is measured through two standard questions on interest rates and inflation (listed in a footnote in Section 4.1). The results in Table 7 show no impact on parents’ financial knowledge in follow-up 1. However, in follow-up 2, parents of treatment school students were significantly more likely to correctly answer both financial knowledge questions than parents of students in control schools. In follow-up 2 we also detect a positive and statistically significant difference in knowledge of budgeting between the two groups: Compared to 68 percent of parents in the control group, parents of students in the treated schools were 6.3 percentage points more likely to understand the composition of a budget. This is a promising result as budgeting was keenly taught in the student curriculum, including the take-home exercises discussed earlier. When examining the impact of the student financial education on parents’ financial behavior in Table 8, we detect no effects in follow-up 1 but again see several positive impacts in follow-up 2. The percentage of parents who save more than zero increased from 76 percent in control schools to 78 percent in treatment schools. The average percentage of income saved increased from 12 percent in control schools to close to 13 percent in treatment schools. Parents in student treatment schools were also more likely to list monthly expenses in a budget, with an increase from 37 percent of parents in the control schools to 39 percent of parents in the treatment schools. These results indicate that the student financial education program had a “trickle-up” impact or spillover effect on parents. The next subsection discusses whether the parents’ workshop reinforced this effect. 7.5. Parent Workshops To measure the impact of the parent workshops, we compare outcomes of parents who attended the financial literacy and the health literacy sessions. We only include parents who actually attended the workshops in this analysis, excluding parents in the same schools who were invited but did not come. As emphasized earlier, parents were not informed of their video assignment (health vs. financial education) until they arrived at school, hence restricting the sample to attendees is a statistically valid exercise. Note that parents who went to the workshop have different baseline characteristics from the ones who did not attend. They are more likely to be recipients of the Bolsa Familia cash transfer 21 program and fathers are less likely to have completed at least some secondary education, suggesting that they are from relatively more disadvantaged households. Drop-out rates are also higher in schools that held a parent workshop. These differences imply that the parent workshop results are not representative of the full sample, although they are valid for the group of parents who attended a workshop (and the corresponding students). As shown in Table 9, we find no significant improvements in parents’ financial behavior as a result of watching the financial literacy video – parents who watched this video were no more likely to make a budget or improve savings behavior compared to those who watched the health video. The lack of significant impact of the parents’ workshop may be due to the relatively low intensity of the treatment or other constraints parents face in responding to the information. The exposure to financial education material through the DVD was relatively short and the workshops included no supplementary discussions. This finding is consistent with the existing literature on financial education (see for example, Bruhn et al., 2013, Cole et al., 2011, and Fernandes et al., 2013). The remarkable result from this intervention, however, is the impact of the parents’ workshop on student behavior. The parent workshops significantly improved the percentage of disposable money saved among students by 2.5 percentage points. Specifically, students whose parents participated in the health literacy workshops saved on average 13.5 percent of their disposable money. In comparison, students whose parent participated in the financial literacy workshops saved 16 percent of their money. This result suggests parents were able to use their improved awareness of financial issues to reinforce the school messages with their children. 7.6. Student Graduation An important corollary effect of the financial education program was on student graduation rates. One hypothesis on student graduation suggests a positive impact especially if the financial education material covered was engaging and interesting, with the new textbooks providing both teachers and students more motivation and interest in teaching and learning. At the same time, however, a greater emphasis on financial education topics and exercises may have come at the expense of neglecting aspects of the regular curriculum. Hence, it is important to empirically investigate impacts on graduation rates. The data on graduation comes from administrative records of the Brazilian Ministry of Education for 2011 – the last year of the financial education program when the students in our sample where in the third and final year of high school. While we were not allowed access to individual student-level or class-level data, we were provided information at the aggregate level for final grade classrooms. Table 10 shows the impact of the financial education program on 2011 grade- level graduation/passing, failure, and dropout rates. Column (1) presents the treatment coefficient on passing rates and finds a 1 percent improvement over the control group, a statistically 22 significant effect at the 10 percent level. Columns (2) and (3) identify treatment effects on failure and dropout rates. While we do not detect statistically significant effects on the likelihood of dropping out of school in 2011, the effect on failing is negative and statistically significant at the 5 percent level. An important point to note is that the analysis on graduation likely underestimates the true effect since the administrative data is at the grade level, whereas our financial education program was provided to a single classroom in each school. Moreover, what is encouraging is that the effect on passing rates is positive and significant, hence allaying concerns about the financial education program diverting attention and jeopardizing grades for the regular curriculum. The evidence, in fact, suggests a complementary role of financial education. Part of the explanation is that financial education was not offered as a separate course but rather integrated into the regular curriculum, and concerns about teaching and student workloads were at the forefront when designing the materials. 8. Mechanisms of Impact This section explores potential mechanisms or channels of the impacts reported in our analysis. Specifically, we measure effects on financial attitudes and preferences of students in our sample. As outlined in the introduction, the existing literatures in economics and psychology argue that inter-temporal preferences can be influenced by practices which may be learned, and that the adoption of certain practices such as delaying gratification can be important drivers of future financial behavior. The outstanding question we address in this section is whether financial education can directly influence students’ preferences regarding current and future financial decisions. We do so by studying two new indicators of financial autonomy and intention to save which were developed for this project by CAEd, and which use insights from sociology and child psychology for age- appropriate measurement (Micarello et al., 2012). As such, these measures aim to signal preferences on current as well as future financial decision-making potential of students. The development phase of these measures involved first synthesizing the main messages of the financial education curriculum, which were identified as the capacity to make responsible financial choices, the capacity to plan finances, and the capacity to act to improve one’s reality. The first concept on responsible financial choices refers to the ability to critically review financial aspects of purchases such as prices, competition, risks, and payment obligations. The second concept on financial planning involves long-term budgeting, cost-benefit analyses, financial prudence, and spending discipline. Finally, the third concept on individual capacity refers to optimism, self-confidence, and perseverance in financial decision-making. 23 The measures on financial autonomy and intention to save were then constructed to highlight and embody these main concepts. Autonomy has previously been studied in the psychology and sociology literatures to understand and encompass attributes of independence, confidence, optimism, self-control, and conformity to parents and peers (Berndt, 1976; Steinberg and Silverberg, 1986; Reichert and Wagner, 2007). Noom et al. (2001) synthesize autonomy into attitudinal, emotional, and functional parts. Attitudinal autonomy refers to the ability to set goals, thinking before acting, and encompasses the notions of knowledge, consciousness, and responsibility. Emotional autonomy relates to the perception of emotional independence in the face of parents, relatives, and peer groups, and the feeling of confidence in one’s own choices. Lastly, functional autonomy refers to perceptions of competence, control, and responsibility in making decisions. The measure of financial autonomy used in this paper and developed in Micarello et al. (2012) most closely follows these definitions. Specifically, the autonomy measure was designed to capture students’ confidence, independence, and willingness to participate and influence household financial decisions. For example, the survey asked students the extent to which they agree or disagree with statements on (i) reflexive/attitudinal autonomy, such as “I like to think carefully before deciding to buy something;” (ii) emotional autonomy, such as “o I feel prepared to talk to my parents about money matters;” and (iii) functional autonomy, such as “I always try to save some money to do things I really like.” Five questions were asked in each category, totaling 15 questions. 15 Student responses to these questions were then aggregated into a summative scale, the financial autonomy index, which ranges from 0 to 100. The intention to save measure follows a similar development methodology and aggregates responses to questions on (i) attitudes towards financial behavior, such as “In my opinion, saving some money every month is extremely beneficial;” (ii) subjective norms and expectations, such as “My family has the habit of saving some money every month;” and (iii) perceptions of controlling one’s behavior, such as “I believe I can save some money every month.” The treatment effects on these preference measures are presented in Table 11. Columns (1) and (2) show that the average financial autonomy score increased from 49 in the control group to 51 in the treatment group by follow-up 1; and from 51 in the control group to 52 in the treatment group by follow-up 2. These results are statistically significant at the 1 percent level and correspond to a 0.09 standard deviation improvement over the control group. In addition, we find that students in the treatment group had a higher measured intention to save (51) than those in control group (49) at follow-up 1, and 53 compared to 51 in follow-up 2. Normalizing by the standard deviations in the control group, these effect sizes represent 9-12 percent of a standard deviation. 15 The full list of questions is available in Appendix 2. 24 Hence, we find statistically significant evidence on improvements in financial preferences of students. Combined with measured behavior change identified in the earlier analysis, these results on preferences distinguish an intermediary channel of the behavioral impacts. In addition, these results are suggestive of the longevity of behavior change since improved underlying preferences have the potential to continue impacting future financial decisions. 9. Conclusion Financial education in schools is an important policy focus in both developed and developing countries, yet its impacts are not well understood. This paper contributes to the literature by demonstrating that a financial education program targeted to youth in secondary school can improve both knowledge and behavior, as well as influence financial attitudes and preferences. Our results offer some important policy lessons. First, financial education in schools may help reduce anomalies in inter-temporal choices, and in fact affect preferences directly. Indeed, self- control strategies among youth have been shown to benefit lifetime economic and social outcomes independent of cognitive ability (Duckworth et al., 2012), hence financial education can be a valuable complement to regular academic learning in school. Second, financial education can be understood more widely as a life skill needed for making better inter-temporal decisions, being aware of opportunities ahead and planning to take advantage of them. As such, financial education can strengthen and complement the effectiveness of other public policies. Third, the partnership between the highly educated and resourceful facets of financial and private sectors with the less resourced education systems can play a transformative role in making public schools an exciting place for learning and personal growth. Fourth, educating parents can strengthen their involvement in their children’s education and generate valuable learning dynamics within the household. Moreover, adult education can provide an important feedback element in children’s education. In terms of realized policy impact, the findings of this study are already being used to guide policy discussions on the impact of financial education in schools. The audience includes financial institutions and education sectors in Brazil and several other interested countries in South America and elsewhere. In fact, the Ministry of Education in Brazil has recently approved a continuation of the financial education program. The program will be scaled-up to 3,000 public high schools as part of two federal programs: Ensino Médio Inovador and Mais Educação. Brazilian stakeholders also decided to develop financial education textbooks and a pilot project for primary schools, as well as a financial education program targeted towards beneficiates of Bolsa Familia and retirees. Furthermore, several other countries in the region have expressed interest in the Brazilian experience to learn and adapt the program to their respective environments and school systems. 25 References Angrist, Joshua, and Victor Lavy. 2009. “The Effects of High Stakes High School Achievement Awards: Evidence from a Randomized Trial.” American Economic Review, 99 (4): 301–331. Angrist, Joshua, Eric Bettinger, Erik Bloom, Elizabeth King, and Michael Kremer. 2002. "Vouchers for Private Schooling in Colombia: Evidence from a Randomized Natural Experiment." American Economic Review, 92(5): 1535–1558. Banerjee, A., Cole, S., Duflo, E. and Linden, L. 2007. “Remedying Education: Evidence from Two Randomized Experiments in India,” Quarterly Journal of Economics, 122(3): 1235–1264. Barrera-Osorio, Felipe, and Leigh L. Linden. 2009. "The use and misuse of computers in education: evidence from a randomized experiment in Colombia." World Bank Policy Research Working Paper No. 4836. Baumeister, Roy, and Todd Heatherton. 1996. “Self-Regulation Failure: An Overview,” Psychological Inquiry, 7(1): 1–15. Baumeister, Roy F., Ellen Bratslavsky, Mark Muraven, and Dianne M. Tice. 1998. “Ego Depletion: Is the Active Self a Limited Resource,” Journal of Personality and Social Psychology, 74(5), 1252–1265. Baumeister, Roy F., Kathleen D. Vohs, and Dianne M. Tice. 2007. “The Strength Model of Self- Control,” Current Directions in Psychological Science 16(6), 351–355. Beal, Diana, and Sarath Delpachitra. 2003. “Financial literacy among Australian university students.” Economic Papers, 22 (1): 65–78. Becchetti, L., S. Caiazza and D. Coviello, 2011. “Financial education and investment attitudes in high schools: Evidence from a randomized experiment.” CEIS Research Paper No. 210, Tor Vergata University. Becchetti, L. and F. Pisani, 2012. “Financial education on secondary school students: The randomized experiment revisited.” Aiccon Working Paper No. 98. Becker, G., and C. Mulligan. 1997. “The Endogenous Determination of Time Preference.” The Quarterly Journal of Economics, 112(3): 729–758. Berndt, Thomas J. 1979. “Developmental Changes in conformity to peers and parents,” US: American Psychological Association, November 1979. Bernheim, B.Douglas, Daniel M. Garrett, and Dean M. Maki. 2001. “Education and saving: The long-term effects of high school financial curriculum mandates.” Journal of Public Economics 80 (3): 435–465. 26 Berry, James, Dean Karlan, and Menno Pradhan. 2012. "Evaluating the Efficacy of School Based Financial Education Programs with Children in Ghana.” The Abdul Latif Jameel Poverty Action Lab, Poverty Action Lab. 17 Oct. 2012. Bruhn, Miriam, Gabriel Lara Ibarra, and David McKenzie. 2013. “Why Is Voluntary Financial Education So Unpopular? Experimental Evidence from Mexico.” World Bank Policy Research Working Paper No. 6439. Bruhn, Miriam, and David McKenzie. 2009. "In Pursuit of Balance: Randomization in Practice in Development Field Experiments." American Economic Journal: Applied Economics, 1(4): 200–232. Cabezas, Victor, Ricardo Paredes, and C. Quezada. 2011. “Peer effects, segregation and school performance in Chile.” Working Paper, Department of Industrial Engineering, PUC. Carlin, Bruce Ian, and David T. Robinson. 2010. “What Does Financial Literacy Training Teach Us?” National Bureau of Economic Research Working Paper Series No. 16271 (August). Carvalho, Leandro, Silvia Prina, and Justin Sydnor, “The Effects of Savings on Risk-Taking and Intertemporal Choice Behavior: Evidence From a Randomized Experiment,” Working Paper, Case Western University. Cole, Shawn. A, and Gauri K Shastry. 2009. “Smart Money: The Effect of Education, Cognitive Ability, and Financial Literacy on Financial Market Participation." Harvard Business School Finance Working Paper Series. Harvard Business School, February 17. Cole, Shawn A., Thomas Sampson, and Bilal Zia. 2011. “Prices or Knowledge? What Drives Demand for Financial Services in Emerging Markets?” Journal of Finance, 66(6): 1933–1967. Consumer Financial Protection Bureau. 2013. “Transforming the Financial Lives of a Generation of Young Americans: Policy Recommendations for Advancing K-12 Financial Education,” CFPB Office of Financial Education Policy White Paper, April 2013. Duckworth, AL, and MEP Seligman. 2005. “Self-Discipline Outdoes IQ in Predicting Academic Performance of Adolescents,” Psychological Science, Vol. 16, 939–44. Duckworth, Angela, Patrick Quinn, and Eli Tsukayama. 2012. “What No Child Left Behind Leaves Behind: The Roles of IQ and Self-Control in Predicting Standardized Achievement Test Scores and Report Card Grades,” Journal of Educational Psychology, Vol. 104(2), May 2012, 439-451. Fernandes, Daniel, John G. Lynch Jr., and Richard G. Netemeyer. 2013. “Financial Literacy, Financial Education and Downstream Financial Behaviors.” Forthcoming in Management Science. 27 Frederick, Shane, George Loewenstein, and Ted O'Donoghue. 2002. “Time Discounting and Time Preference: A Critical Review,” Journal of Economic Literature, 40(2), 351–401. He, F., Linden, L. and MacLeod, M. 2008. “How to Teach English in India: Testing the Relative Productivity of Instruction Methods within the Pratham English Language Education Program,” mimeo Columbia University. Instituto Data Popular. 2008. Annual Report. Brazil, 2008. Kocher, Martin G., Daniela Rützler, Matthias Sutter, and Stefan T. Trautmann. 2012. “Cognitive Skills, Self-Control, and Life Outcomes: The Early Detection of At-Risk Youth,” Vox (April 16), www.voxeu.org Linden, L. 2008. “Complement or Substitute? The Effect of Technology on Student Achievement in India,” mimeo Columbia University. Lührmann, Melanie, Marta Serra-Garcia, and Joachim Winter. 2012. "The effects of financial literacy training: Evidence from a field experiment with German high-school children." Discussion Papers in Economics 14101, University of Munich, Department of Economics. Lusardi, Annamaria, and Olivia S. Mitchell. 2007. “Financial Literacy and Retirement Preparedness: Evidence and Implications for Financial Education,” Business Economics, 42(1): 35–44. Lusardi, Annamaria, and Olivia S. Mitchell. 2008. “Planning and Financial Literacy: How Do Women Fare?” Center for Research on Pensions and Welfare Policies, Turin (Italy), January. Lusardi, Annamaria, Olivia S. Mitchell, and Vilsa Curto. 2010. "Financial Literacy among the Young: Evidence and Implications for Consumer Policy." Lusardi, Annamaria, and Olivia S. Mitchell. 2011a. “Financial Literacy Around the World: An Overview.” Journal of Pension Economics and Finance, October 2011. Lusardi, Annamaria, and Olivia S. Mitchell. 2011b. “Financial Literacy and Retirement Planning in the United States.” CeRP Working Papers. Center for Research on Pensions and Welfare Policies, Turin (Italy), February. Lusardi, Annamaria, Olivia S. Mitchell, and Vilsa Curto. 2009. “Financial Literacy Among the Young: Evidence and Implications for Consumer Policy.” National Bureau of Economic Research Working Paper Series No. 15352 (September). Lusardi, Annamaria, Peter Tufano, and S. Field. 2009. “Debt Literacy, Financial Experiences, and Overindebtedness.” National Bureau of Economic Research Working Paper Series No. 14808 (March). 28 Mandell, Lewis. 2006. “Financial Literacy: If It’s so Important, Why Isn’t It Improving?” Networks Financial Institute Policy Brief 2006-PB-08 (April). McKenzie, David. 2012. “Beyond baseline and follow-up: the case for more T in experiments,” Journal of Development Economics, 99(2): 210–221. McKenzie, David, and Bilal Zia. 2012. “Who you Train Matters: Identifying Complementary Effects of Financial Education on Migrant Households,” World Bank Policy Research Working Paper No. 6157. Micarelli, Hilda, Manuel Palacios and Marcelo Burgos. 2012. “Application of the CAEd autonomy scale to assess the impact of financial education”, CAEd academic publications online. Mischel, W. 1974. “Process in Delay of Gratification,” in L. Berkowitz (Ed.), “Advances in Experimental Social Psychology,” Vol 7, 249-292, San Diego: Academic. Mischel, W., Y. Shoda, and P. K. Peake. 1988. “The Nature of Adolescent Competencies Predicted by Preschool Delay of Gratification,” Journal of Personality and Social Psychology, 54: 687–696. Mischel, W, S Yuichi, and ML Rodriguez. 1989. “Delay of gratification in children”, Science, 244:933–38. Mullianathan, Sendhil, and Eldar Shafir. 2013. “Scarcity: Why Having too Little Means so Much,” Times Books, New York. Muralidharan, K. and Sundararaman, V. 2011. “Teacher Performance Pay: Experimental Evidence from India,” Journal of Political Economy, 119(1): 39–77. Muraven, Mark and Roy Baumeister. 2000. “Self-Regulation and Depletion of Limited Resources,” Psychological Bulletin, 126(2): 247–259. Noom, M. J., Dekovic, M., and Meeus, W. 2001. “Conceptual analysis and Measurement of Adolescent Autonomy,” Journal of Youth and Adolescence, 30(5): 577–595. Reichert, C. B., and A. Wagner. 2007. “Considerations Regarding Autonomy in the Contemporaneity,” Estudos e Pesquisas em Psicologia (Online), 7(3): 46–59. Steiberg, Laurence, and Susan Silverberg. 1986. “The Vicissitudes of Autonomy in Early Adolescence,” Child Development, 57: 841–851. Sutter, Matthias, Martin G. Kocher, Daniela Glätzle-Rüetzler, and Stefan T. Trautmann. 2013. "Impatience and Uncertainty: Experimental Decisions Predict Adolescents' Field Behavior." American Economic Review, 103(1): 510-31. U.S. Bankruptcy Abuse Prevention and Consumer Protection Act Report, 2011 29 Varcoe, Karen P., Allen Martin, Zana Devitto, and Charles Go. 2005. “Using A Financial Education Curriculum For Teens.” Journal of Financial Counseling and Planning, 16(1): 63–71. Walstad, William B., Ken Rebeck, and Richard A. MacDonald. 2010. “The Effects of Financial Education on the Financial Knowledge of High School Students.” Journal of Consumer Affairs, 44(2): 336–357. Xu, Lisa, and Bilal Zia. 2012. “Financial Literacy Around the World: Overview of the Evidence and Practical Suggestions for the Way Forward,” World Bank Policy Research Working Paper #6107, July 2012. 30 Figure 1: Study Design Group 1 Schools Group 2 Schools Student Intervention Control Treatment Parent Intervention -- ½ Treatment, ½ Control Figure 2: Distribution Shift in Financial Proficiency Scores FOLLOW-UP 1 .025 .02 .015 .01 .005 0 20 40 60 80 100 120 Financial Proficiency Score Control Treatment FOLLOW-UP 2 .025 .02 .015 .01 .005 0 20 40 60 80 100 120 Financial Proficiency Score Control Treatment This figure compares density plots for student financial proficiency scores in treatment and control schools. Follow-up 1 and follow-up 2 plots are presented separately. 31 Table 1: Baseline Summary Statistics Control Treatment Difference in Number of Number of Standard Standard Means Test Mean Mean (p-value) Schools Students Deviation Deviation School Level Variables Used to Form Matched Pairs: Number of students in school 867 640.08 460.20 669.86 508.08 0.366 Number of teachers in school 748 37.41 24.28 38.11 25.77 0.702 School dropout rate 698 9.83 10.36 9.90 10.01 0.925 Grade passing rate 698 74.24 13.20 74.26 13.51 0.981 Student Background Characteristics: Student is female 864 26131 0.55 0.56 0.069* Mother attended secondary school 863 22474 0.45 0.43 0.392 Father attended secondary school 863 22357 0.41 0.40 0.449 Student has failed at least one school year 863 22435 0.30 0.32 0.100 Student's family receives Bolsa Familia cash transfer 863 22662 0.32 0.33 0.281 Student has computer with internet at home 863 22502 0.52 0.53 0.542 Student has some form of income 866 24319 0.66 0.67 0.111 Student is employed 866 24303 0.35 0.35 0.902 Student Financial Characteristics: Financial proficiency score 864 23255 49.80 15.18 50.15 14.85 0.461 Makes a list of expenses every month 866 24080 0.10 0.11 0.680 Negotiates prices or payment methods 866 23867 0.75 0.76 0.462 Financial autonomy index 866 22789 49.11 19.83 49.04 19.58 0.844 Intention to save index 866 22797 48.29 18.95 48.19 18.57 0.772 This tables presents baseline summary statistics as well as p-values for difference in means tests between students in treatment and control schools. The first four rows show school level variables from administrative data for 2008 obtained from the Brazilian Ministry of Education. These four variables were used to form matched school pairs prior to randomization. One school in each pair was then randomly assigned to treatment and the other to control. The subsequent rows in the table summarize survey data from the baseline survey. To calculate p-values, standard errors are clustered at the school level and randomization pair dummies are added as controls. * denotes statistical significance at the 10% level. Table 2: Survey Participation Number of Schools Number of Students Baseline Follow-up 1 Follow-up 2 Baseline Follow-up 1 Follow-up 2 Financial Knowledge Test (Day 1) 864 829 824 23290 17831 18420 Student Questionnaire (Day 2) 866 826 824 24473 16512 17680 Parent Questionnaire 864 810 818 21603 13532 14080 This table maps the number of schools and the number of students in our study sample across the three waves of surveys: baseline, follow-up 1, and follow-up 2. The survey team visited each school for three days for each round. On day 1, the financial knowledge test was administered and parent questionnaires were handed out. On day 2, the student questionnaire on behaviors and attitudes was administered. Day 3 was reserved for any make-up surveys for students who missed any of the earlier surveys, and the completed parent questionnaires were also collected on day 3. Table 3: Survey Attrition Student Survey Parent Survey Response Rate Response Rate (1) (2) Treatment School 0.02 0.03 (0.019) (0.020) R-squared 0.656 0.663 Sample Size (Number of Schools) 631 631 Dependent Variable Mean in Control Group 0.77 0.55 Dependent Variable Standard Deviation in Control Group 0.19 0.24 This table presents attrition regression results for the follow-up 2 sample. The dependent variable is the survey response rate at the end of the semester, which is the ratio of survey response to enrolment at the beginning of the semester. Column (1) present the response rate among students and column (2) presents the response rate among parents. Enrolment counts for 631 schools were obtained. The regressions include school pair dummies and shows robust standard errors in parentheses. Table 4: Student Financial Proficiency Financial Proficiency Score Follow-up 1 Follow-up 2 (1) (2) Treatment School 3.548*** 3.020*** (0.296) (0.355) R-squared 0.455 0.328 Sample Size (Number of Students) 17831 18415 Number of Schools 829 824 Dependent Variable Mean in Control Group 56.135 59.003 Dependent Variable Standard Deviation in Control Group 14.804 14.908 This table shows OLS regression results for the impact of the financial education program on student financial proficiency. The sample in columns (1) and (2) includes students present in follow-up waves 1 and 2, respectively. The number of students and schools included in the sample fluctuate within a wave because not all students answered every question; and they fluctuate across waves because of student turnover. The outcome variable in this table is a student financial proficiency score, which aggregates financial knowledge questions included in the survey on a 0-100 scale. All regressions control for baseline outcomes and include school pair dummies. When baseline outcomes have missing values, they are replaced by zero and a dummy variable indicating such missing values is included. Robust standard errors, clustered at the school level, are in parentheses. *** denotes statistical significance at the 1% level. Table 5: Student Purchasing and Spending Behavior Percentage of Disposable Comparison Shops Before Negotiates Prices or Payment Lists Monthly Expenses in a Saves for Purchases? Money Saved Making Purchase? Methods? Budget? Follow-up 1 Follow-up 2 Follow-up 1 Follow-up 2 Follow-up 1 Follow-up 2 Follow-up 1 Follow-up 2 Follow-up 1 Follow-up 2 (1) (2) (3) (4) (5) (6) (7) (8) (9) (10) Treatment School 0.047*** 0.052*** N/A 1.389*** 0.025*** 0.021*** 0.041*** 0.031*** 0.026*** 0.030*** (0.006) (0.007) (0.316) (0.006) (0.006) (0.005) (0.006) (0.005) (0.005) R-squared 0.210 0.106 0.045 0.099 0.067 0.219 0.131 0.149 0.093 Sample Size (Number of Students) 16288 17320 16695 16119 17295 16175 17343 16358 17475 Number of Schools 825 822 822 825 822 825 822 825 822 Dependent Variable Mean in Control Group 0.440 0.404 12.897 0.656 0.674 0.740 0.740 0.129 0.139 Dependent Variable Standard Deviation in Control Group 18.958 This table presents OLS regression results for the impact of the financial education program on student purchasing and spending behavior. The sample in odd numbered columns is follow-up 1 and in even numbered columns is follow-up 2. The number of students and schools included in the sample fluctuate within a wave because not all students answered every question; and they fluctuate across waves because of student turnover. The outcome variables in this table are: an indicator variable equal to 1 if the student saves for purchases (columns 1 and 2); the percentage of monthly disposable money that is saved (columns 3 and 4); an indicator variable equal to 1 if the student comparison shops before making a purchase (columns 5 and 6); an indicator variable equal to 1 if the student negotiates the price or the payment method when making a purchase (column 7 and 8); and an indicator variable equal to 1 if the student makes a list of monthly expenses in a budget (columns 9 and 10). All regressions control for baseline outcomes and include school pair dummies. When baseline outcomes have missing values, they are replaced by zero and a dummy variable indicating such missing values is included. Robust standard errors, clustered at the school level, are in parentheses. *** denotes statistical significance at the 1% level. Table 6: Student Participation in Household Finance Student Discusses Financial Student Helps Organize Matters with Parents? Household Budget? Follow-up 1 Follow-up 2 Follow-up 1 Follow-up 2 (1) (2) (3) (4) Treatment School 0.036*** 0.039*** 0.048*** 0.036*** (0.007) (0.006) (0.007) (0.008) R-squared 0.171 0.104 0.194 0.115 Sample Size (Number of Parents) 13357 13844 13345 13867 Number of Schools 810 816 809 816 Dependent Variable Mean in Control Group 0.673 0.701 0.487 0.522 This table presents OLS regression results for the impact of the financial education program on student participation in household finance. The sample in odd numbered columns is follow-up 1, and in even numbered columns is follow-up 2. The number of parents and schools included in the sample fluctuate within a wave because not all parents answered every question; and they fluctuate across waves because of student turnover. The outcome variables in this table are: an indicator variable equal to 1 if a student discusses financial matters at home (columns 1 and 2); and an indicator variable equal to 1 if a student helps organize the household budget (columns 3 and 4). Both questions are based on responses in the parent questionnaires. All regressions control for baseline outcomes and include school pair dummies. When baseline outcomes have missing values, they are replaced by zero and a dummy variable indicating such missing values is included. Robust standard errors, clustered at the school level, are in parentheses. *** denotes statistical significance at the 1% level. Table 7: Parent Financial Knowledge Interest Rate Question Inflation Question Budgeting Question Follow-up 1 Follow-up 2 Follow-up 1 Follow-up 2 Follow-up 1 Follow-up 2 (1) (2) (3) (4) (5) (6) Treatment School -0.001 0.017* 0.007 0.014* N/A 0.063*** (0.008) (0.009) (0.008) (0.008) (0.009) R-squared 0.091 0.071 0.132 0.104 0.080 Sample Size (Number of Parents) 13368 12691 13327 12609 12710 Number of Schools 809 816 809 816 816 Dependent Variable Mean in Control Group 0.444 0.443 0.329 0.320 0.678 This table presents OLS regressions results for the impact of the financial education program on parent financial knowledge. The sample in odd numbered columns is follow-up 1 and in even numbered columns is follow-up 2. The number of parents and schools included in the sample fluctuate within a wave because not all parents answered every question; and they fluctuate across waves because of student turnover. The outcome variables in this table are three financial literacy questions in the parent surveys: the interest rate question (columns 1 and 2) tests the ability to calculate an interest rate using percentages; the inflation question (column 3 and 4) tests the understanding of how inflation affects future purchasing power; and the budgeting question (columns 4 and 6) tests the knowledge of what goes into a budget. All regressions control for baseline outcomes and include school pair dummies. When baseline outcomes have a missing values, they are replaced by zero and a dummy variable indicating such missing values is included. Robust standard errors, clustered at the school level, are in parentheses. * denotes statistical significance at the 10% level; and *** denotes statistical significance at the 1% level. Table 8: Parent Savings and Spending Behavior Lists Monthly Expenses in a Has Formal Savings? Percentage of Income Saved Budget? Follow-up 1 Follow-up 2 Follow-up 1 Follow-up 2 Follow-up 1 Follow-up 2 (1) (2) (3) (4) (5) (6) Treatment School 0.003 0.014** N/A 0.667** -0.012 0.021*** (0.006) (0.006) (0.263) (0.007) (0.007) R-squared 0.334 0.206 0.050 0.190 0.116 Sample Size (Number of Parents) 13079 13533 12953 13187 13566 Number of Schools 810 816 816 810 816 Dependent Variable Mean in Control Group 0.734 0.762 12.171 0.366 0.369 Dependent Variable Standard Deviation in Control Group 16.521 This table presents OLS regressions results for the impact of the financial education program on parent savings and spending behavior. The sample in odd numbered columns is follow-up 1 and in even numbered columns is follow-up 2. The number of parents and schools included in the sample fluctuate within a wave because not all parents answered every question; and they fluctuate across waves because of student turnover. The outcome variables in this table are based on responses in the parent survey and include: an indicator variable equal to 1 if the parent has formal savings such as a current account, savings account, debit card or checks (columns 1 and 2); the percentage of monthly income that is saved (column 4); and an indicator variable equal to 1 if the parent makes a list of monthly expenses in a budget (columns 5 and 6). The percentage of income saved question was not asked in follow-up 1. All regressions control for baseline outcomes and include school pair dummies. When baseline outcomes have missing values, they are replaced by zero and a dummy variable indicating such missing values is included. Robust standard errors, clustered at the school level, are in parentheses. ** denotes statistical significance at the 5% level; and *** denotes statistical significance at the 1% level. Table 9: Parent Financial Education Workshop Parent Outcomes Student Outcomes Lists Monthly Percentage of Budgeting Has Formal Percentage of Saves for Expenses in a Disposable Question Savings? Income Saved Purchases? Budget? Money Saved (1) (2) (3) (4) (5) (6) Attended Financial Education Workshop -0.042 0.018 1.350 0.008 0.024 2.445** (0.026) (0.026) (1.057) (0.030) (0.027) (1.135) R-squared 0.197 0.375 0.202 0.254 0.253 0.170 Sample Size (Number of Parents/Students) 1022 1059 1016 1063 1273 1239 Dependent Variable Mean in Control Group 0.824 0.705 12.550 0.376 0.474 13.484 Dependent Variable SD in Control Group 14.838 17.567 This table presents OLS regression results for the impact of the parent financial education workshop. The sample in columns 1-4 includes parents in treatment schools who attended either a financial education or health education workshop. Since workshop assignment was not revealed in advance, the analysis only includes parents who attended. All data in this table is from follow-up 2 since the parent workshops occurred between the two follow-up survey rounds. Four parent outcomes are presented in this table: an indicator variable equal to 1 for correct answers to a question that tests the knowledge of what goes into a budget (column 1); an indicator variable equal to 1 if the parent has formal savings such as a current account, savings account, debit card or checks (column 2); the percentage of monthly income that is saved (column 3); and an indicator variable equal to 1 if the parent makes a list of monthly expenses in a budget (column 4). In addition, two student outcomes are presented in columns (5 )and (6): an indicator variable equal to 1 if the student saves for purchases (column 5); and the percentage of monthly disposable money that is saved (column 6). All regressions control for baseline outcomes and include parent workshop stratification dummies. When baseline outcomes have missing values, they are replaced by zero and a dummy variable indicating such missing values is included. Robust standard errors are in parentheses. ** denotes statistical significance at the 5% level. Table 10: Student Graduation Grade-Level Passing Grade-Level Failing Grade-Level Rate Rate Dropout Rate (1) (2) (3) Treatment School 0.013* -0.010** -0.003 (0.006) (0.005) (0.004) R-squared 0.582 0.592 0.632 Sample Size (Number of Schools) 858 858 858 Dependent Variable Mean in Control Group 0.841 0.093 0.066 Dependent Variable Standard Deviation in Control Group 0.103 0.081 0.069 This table presents OLS regression results for the impact of the financial education program on student graduation and retention rates. The analysis uses administrative data from the Brazilian Ministry of Education from the 2011 school year for grade-level outcomes at the end of the school year. Passing rate, failing rate, and dropout rate sum to 1 for each school. Column (1) show treatment effects on the graduation rate, while columns (2) and (3) identify treatment effects for students who did not pass, by whether they failed the grade (column 2) or dropped out of the grade during the school year (column 3). Robust standard errors, clustered at the school level, are in parentheses. * denotes statistical significance at the 10% level; and ** denotes statistical significance at the 5% level. Table 11: Student Financial Preferences Financial Autonomy Index Intention to Save Index Follow-up 1 Follow-up 2 Follow-up 1 Follow-up 2 (1) (2) (3) (4) Treatment School 1.703*** 1.774*** 2.471*** 1.914*** (0.234) (0.305) (0.297) (0.324) R-squared 0.449 0.250 0.278 0.168 Sample Size (Number of Students) 14283 16019 14919 16286 Number of Schools 824 822 823 821 Dependent Variable Mean in Control Group 49.035 50.544 49.020 51.342 Dependent Variable Standard Deviation in Control Group 19.785 20.724 21.157 21.232 This table presents OLS regression results for the impact of the financial education program on student financial preferences. The sample in odd numbered columns is follow-up 1 and in even numbered columns is follow-up 2. The number of students and schools included in the sample fluctuate within a wave because not all students answered every question; and they fluctuate across waves because of student turnover. The outcome variables in this table are a student financial autonomy index (columns 1 and 2) and an intention to save index (columns 3 and 4). The financial autonomy index aggregates responses to questions that elicit future financial preferences, confidence, and decision making independence. The intention to save index aggregates responses to questions on hypothetical and savings and spending scenarios. All regressions control for baseline outcomes and include school pair dummies. When baseline outcomes have missing values, they are replaced by zero and a dummy variable indicating such missing values is included. Robust standard errors, clustered at the school level, are in parentheses. *** denotes statistical significance at the 1% level. For Online Publication: Appendices 1 and 2 APPENDIX 1: Details of the Financial Education Curriculum A1.1 General Description The material used in the school financial education program in Brazil includes: (i) a student textbook, (ii) a student exercise book, (iii) a teacher guidebook, and (iv) a teacher training DVD. All materials were developed by the Pedagogical Support Group (GAP: Grupo de Apoio Pedagógico). This Appendix describes the content of the student textbook in detail.1 The student textbook is divided into three blocks and covers nine different themes. Each theme is taught through case studies/didactic situations (SDs), consisting of theoretical and applied content, activities and self-evaluative questions. SDs make use of texts, stories, images, and tables to convey the material in an accessible way. SDs also contain “experiment” sections that are designed to make the material relevant to students’ daily life. SDs conclude with a short outline of what is expected from the student in terms of learning. The rest of this Appendix describes the themes included in each of the three blocks. A1.2 Block 1 The first block discusses three themes: everyday family life, social life, and personal property. THEME 1: EVERYDAY FAMILY LIFE In everyday family life, students are exposed to common situations where they have to make decisions that impact their family’s financial wellbeing. This theme is covered in 7 sections: 1. Agenda and Planning  Record expenses regularly  Know where you spend money  Estimate the value of items purchased 2. Calendar  Make a list of personal and family expenses  Classify expenses as “fixed” or “variable”  Prepare a monthly budget separating “fixed” and “variable” expenses 3. House Repairs  Compare different repair cost estimates  Compare interest rates for personal loans  Decide whether to take a loan or use money previously saved 4. Supermarket  How to avoid overspending on “temptation” goods  To distinguish good and bad behaviors when going shopping  Advance decision making on items to purchase 1 The student exercise books and the teacher guidebooks and DVDs were meant to support the material covered in the student textbooks and did not offer new material of their own. The student exercise book provided assignments to students based on material taught through the textbook. Similarly, the teacher guidebook and DVD provided instructions on teaching and assessment methods for the course material, as well as examples of how to integrate the financial education curriculum into regular school learning. 5. Balancing  Distinguish and categorize personal and family expenses  Assess the importance of these different expenses  Identify spending categories where cuts can be made  Prepare a 5% spending-cut plan 6. Unforeseen Circumstances  Understand the value of insurance  Understand the specific vocabulary of insurance products  Identify alternative methods of prevention 7. Matching spending to earnings  Classify income sources as “fixed” and “variable”  Prepare a table with family incomes  Analyze how family spends and saves money THEME 2: SOCIAL LIFE In social life, students are exposed to situations where they have to make financial decisions about their personal and social lives. This theme is covered in 7 sections: 1. What a waste  Analyze personal expenses and identify waste  Avoid waste  Identify actions that can lead you to spend more than necessary 2. Let’s get this party started  Make estimates of the quantity of food and drinks necessary for a party  Make a budget for a party  Plan an environmentally friendly party  Identify pitfalls when making estimates 3. To give in or not to give in to peer pressure… that is the question  Organize financial information in a way that can be easily explained to others  Learn and apply concepts such as interest rates, risks and returns to everyday situations 4. Buying on credit  Identify elements of a credit card bill  Identify financial behaviors that lead to credit card debt  How to use a credit card in a responsible manner 5. Camping  Identify expenses involved in going camping  Always keep funds for unforeseen events  Prepare a financial plan to go camping 6. “Viva São João!”  Prepare a business plan for organizing a party to celebrate the São João holiday 7. Don’t fall victim to advertising  Identify financial pitfalls of credit card advertisements  Analyze various options available for credit THEME 3: PERSONAL PROPERTY In personal property, students learn from situations where they have to make personal shopping decisions. This theme is covered in 7 sections: 1. In search of the perfect shoes  Calculate the difference between the price paid in cash and the one paid with credit  Decide if it is better to pay in cash or with credit  Find specific information in the Consumer Defense Code 2. Computer  Identify the opportunity cost of owning a computer  Balance wants and needs when choosing a computer  Compare prices  Calculate the necessary savings in order to buy a computer 3. Digital camera  Follow similar steps as with purchasing a computer 4. If by magic…  Identify the elements of advertising aimed at generating consuming desire  Identify the conflict between desires and needs  Be vary of temptation traps and impulsive spending 5. Cell phone  Choose a cell phone that best fit your needs  Choose a plan that best fits your needs  Understand your cell phone bill 6. Consumer protection measures?  Identify cases of abusive practices and consumer rights violations  When to reach out to the Foundation for the Protection and Consumer Advocacy (PROCON) 7. Changing money  Convert the value of products priced in foreign currency to local currency  Know how the value of a credit card purchase in foreign currency appears in local currency on the bill A1.3 Block 2 The second block discusses three themes: work, entrepreneurship, and large projects. THEME 4: WORK In work, students discuss several aspects of their current and future professional lives. This theme is covered in 7 sections: 1. What line of work?  Identify the type of work that attracts you the most  Decide on the most appealing type of job according to your life ambitions 2. First job  Prepare a CV  Identify the skills that are compatible with advertised positions  Appropriately highlight your professional qualities in a simulated job interview  Combine your desired job with the type of life you want to have 3. Gross vs net income  Differentiate gross and net income  How to explain this concept to others 4. Good times and bad times  Understand the concept of structural unemployment in a made-up story  Identify measures to overcome unemployment 5. The incredible story of the 13th salary that disappeared  Make a budget based on data and estimates  Make a financial plan in order to achieve a positive balance at the end of the month  Consider future situations in current monthly planning 6. Lifelines  Prepare an outline of a retirement plan, harmonizing long-term goals and the means to achieve them 7. Antenor, the wary employee  Develop a product and the message for an information campaign about insurance  Utilize the vocabulary of insurance as it applies to an information campaign THEME 5: ENTREPRENEURSHIP In entrepreneurship, students learn about practical issues of creating and running a business. This theme is covered in 7 sections: 1. A great idea  Differentiate entrepreneurs driven by necessity and by opportunity  Relate own characteristics with business opportunities  Identify needs in own community that may generate a business opportunity  Brainstorm to generate good business ideas 2. What are your talents?  Distinguish between “knowledge”, “skill”, “attitude”, and “competencies” in the context of entrepreneurship  Evaluate if you possess the necessary knowledge to open a particular business  Evaluate if you possess the necessary skills to open a particular business  Evaluate if you possess the necessary attitudes to open a particular business 3. Profession: entrepreneur  Identify the characteristics of an entrepreneur  Differentiate entrepreneurship from intrapreneurship  Test if you have the profile of an entrepreneur 4. The soul of a business  Identify the target audience of a fictitious business  Create a brand and slogan for a fictitious product or service  Put together a fictitious marketing plan  Carry out market research for a fictitious product or service 5. Hands to work  Identify resources necessary to open and run a business  Budget for opening and running a fictitious business  Determine the knowledge, skills, attitudes, and competencies of the personnel necessary to work in a fictitious business 6. Victory  Make sales and profit projections for a fictitious business  Measure the profit of a fictitious business  Cut costs and expenses related to products or services of a fictitious business 7. Beyond profit  Distinguish between philanthropy and socio-environmental responsibility  Make a plan of socio-environmental responsibility for a fictitious business  Put together in a business plan all the information on entrepreneurship learned in this theme THEME 6: LARGE EXPENDITURES In large projects, students are exposed to situations that involve significant financial outlays. This theme is covered in 7 sections: 1. Brick by brick  Balance the desires and needs of your family when choosing a house to purchase  Search for information on prices and financing for a house  Decide how much your family is willing to spend as a function of the household budget  Plan financially for the down payment and installments of a home mortgage 2. Surprise  Create a budget for a party  Plan a party that suits your financial situation  Make provisions for unforeseen expenses  Cut expenses according to your priorities 3. In your corner  Make investment decisions in a simulated market situation  Make an initial investment decision, taking into consideration family and personal preferences 4. She talks about the same thing all day long  Identify rights and duties that are not being met in a certain situation  Generate arguments to debate rights and duties of investors 5. Consumption and savings  Make consumption and savings decisions in a simulated situation 6. Now it’s my turn to help my parents  Decide between two debt application options, taking into account interest rates  Explain how to avoid indebtedness  Come up with options to pay off a debt of R$ 1000 (US$ 583) 7. How much distance separates you from your future  Estimate fixed and variable expenses in order to study in another city  Calculate the monthly income necessary to study in another city  Make a financial plan to study in another city A1.4 Block 3 The third block discusses three themes: public goods, the country’s economy, and the world economy. THEME 7: PUBLIC GOODS In public goods, students address several issues concerning the use and financing of public goods and services. This theme is covered in 7 sections: 1. Everything has a price  It is always the case that someone pays for the public goods you consume for free  Calculate how much the government spends to sustain a high school class in a public school 2. School budget  Think about the school and its budget  Suggest improvements to the school that are feasible 3. School books  Identify the reasons for high environmental cost of school books  Calculate the consumption of paper in school  Identify actions that can save paper  Develop and engage in a campaign to save paper 4. Public spaces  Everyone has the right to access free public spaces  Maintenance of public spaces is costly and is paid for through taxes  The individual tax burden can be reduced if all citizens pay their taxes  Consult the community in order to know which public spaces need to be improved 5. Public services  A public budget is very similar to a family budget  The legislature – senators and congressmen – decides the public budget  Link the public duty to pay taxes with the government’s duty to provide public services 6. Corruption  Corruption affects the lives of everyone because it reduces the money that the government can invest in public services  Check public accounts through public records 7. Taxation  Link the public duty to pay taxes with the government’s duty to provide public goods and services  Understand the purpose of different taxes paid by citizens  Develop and engage in a campaign to provide incentives for citizens to pay their taxes THEME 8: THE COUNTRY ECONOMY In the country economy, students are exposed to several aspects of their country’s economy that are relevant for their personal lives, including the concept of inflation, the law of supply and demand, the concept of minimum wage, and the basics of the national financial system. This theme is covered in 7 sections: 1. Culture and sports  Prepare an outline of a project for a cultural or sporting activity  Align the objectives of a project to the Rouanet Law (Law that provides tax incentives to private firms for supporting cultural activities)  Understand the sections of the laws concerning education that contain financial vocabulary 2. Inflation  Recognize the problems that inflation can generate when changes in income do not follow the increase in prices  Make adjustments in the family budget taking inflation into consideration  Explain the concept of inflation to someone else 3. Supervisors of the national financial system  Explain the national financial system to someone else 4. Speaking in economic terms  There exists a direct relationship between the nation’s economic growth and the growth of a family’s personal income  Families with low income can also organize themselves financially 5. Markets  Demand and supply simulations 6. Foresight  Estimate the income and expenses of a retired person  Prepare a simulated financial plan for a retired person 7. Minimum wage  Research prices to estimate the total value of the basic needs of a person  Link the value of the basic needs of a person with the value of the minimum wage THEME 9: THE WORLD ECONOMY In the world economy, students are exposed to several aspects of the world economy that are relevant for their personal lives, such as the concept of imports and exports, international economic blocks, and measures of a country’s wellbeing. This theme is covered in 7 sections: 1. Special issue on money  Contextualize the role of money in society  The importance of saving money 2. International cooperation  Identify the complications involved in international negotiations  International economic blocks organize themselves through arrangements that are negotiated 3. The game of economic blocks  Experience, in a game, some of the issues concerning international economic blocks  Think about simulated strategies of global conflict resolution 4. The business of China  Identify the imported products that you use in everyday life  Locate the countries where the imported products you use my everyday life come from  Search for data on national and international exports 5. International Organizations  Reflect upon the profile and the performance of representatives of a country in an international community  Develop a funding proposal for an international financial institution 6. The well-being of your country  Compare the Human Development Index and the GDP per capita for different countries  Link the economic performance of a country with its environmental impact 7. Moment of crisis: do I care?  Establish the relationship between an economic crisis and situations of your personal life  Identify ways to overcome the impact of an economic crises for individuals APPENDIX 2: Questions Used to Construct Preference Measures Financial Autonomy Index  Reflexive Autonomy: o I like to think thoroughly before deciding to buy something o I like to research prices whenever I buy something o I make sure to get information on warranty periods o I always try to obtain more information on product quality o I pay attention to news about the economy as it may affect my family  Emotional Autonomy: o I like to participate in family decision making when we buy something expensive for home o I usually have a critical view of the way my friends deal with money o I take part in domestic expense planning o I try to advise my parents on money matters o I feel prepared to talk to my parents about money matters  Functional Autonomy: o I always try to save some money to do things I really like o I always like to negotiate prices when I buy o I suggest at home that we keep money aside for emergencies o I keep an eye on promotions and discounts o I am willing to make sacrifices now to buy something important Intention to Save Index A total of 21 questions go into the intention to save index. The possible answers to each question are on a seven point scale, with answers ranging from extremely negative to extremely positive.  Attitudes towards behavior o Complete the following statement “In my opinion, saving some money every month is...” – Answers ranging from 1. Extremely harmful to extremely beneficial 2. Extremely unpleasant to extremely pleasant 3. Extremely bad to extremely good 4. Extremely useless to extremely useful  Subjective norms o To which extent do you agree with the following statement – Answers ranging from completely false to completely true 5. “My family has the habit of saving some money every month.” 6. “I have friends who save some money every month.”  Perceived control over one’s own behavior 7. To which extent do you agree with the following statement: “I believe that I can save some money every month.” – Answers ranging from completely false to completely true. 8. Complete the following statement “For me, saving some money every month is…” – Answers ranging from extremely difficult to extremely easy. 9. To which extent do you agree with the following statement: “Whether or not I save some money every month is in my own hands.” – Answers ranging from completely disagree to completely agree.  Attitudes about the possible effects of behavior 10. To which extent do you agree with the following statement: “Saving some money every month would allow me to buy things I need.” – Answers ranging from completely agree to completely disagree. 11. Complete the following statement “For me, saving to allow me to buy the things I need is…” – Answers ranging from extremely bad to extremely good. 12. To which extent do you agree with the following statement: “Saving some money every month is a good way of making sure I never accumulate debt.” - Answers ranging from completely agree to completely disagree. 13. Complete the following statement “In my opinion, borrowing money and accumulating debt is…” – Answers ranging from extremely bad to extremely good.  Motivation for following subjective norms o All Answers ranging extremely unlikely to extremely likely. 14. To which extent do you agree with the following statement: “My family hopes that I am able to save some money every month.” 15. “When it comes to spending and saving, how likely are you to follow your family’s opinion?” 16. To which extent do you agree with the following statement: “My friends would support me if I decided to save some money every month.” 17. “How likely are you to follow your friends’ opinion about savings habits?”  Perceived benefits of one’s own behavior 18. To which extent do you agree with the following statement: “Managing my expenses and not wasting money is essential for saving some money every month.” – Answers ranging from completely agree to completely disagree. 19. Complete the following statement “For me, managing my expenses and not wasting money is…” – Answers ranging from extremely difficult to extremely easy. 20. To which extent do you agree with the following statement: “In order for me to save, it’s essential that I avoid borrowing money and getting into debt.” – Answers ranging from completely agree to completely disagree. 21. Complete the following statement: “For me, being able to avoid borrowing money and getting into debt is…” – Answers ranging from extremely difficult to extremely easy.