Toward Productive Research Agendas in Financial Inclusion, Security, and DevelopmentWritten by Michael Sherraden, Washington University in St. Louis; Margaret S. Sherraden, Washington University in St. Louis
In this essay, we focus on key themes in research and knowledge building which have the potential to inform and improve research on financial inclusion, security, and development. In doing so, we use current examples in our own work on asset building, child accounts, and financial capability. This is the work we know best, and perhaps the examples are more meaningful and informative as a result. The implications of each theme extend broadly to other topics in financial well-being, and indeed to other topics in applied social research.
Our intent and hope is that these observations will spur interest in and engagement toward better research. Above all, we would like to see research agendas in applied social sciences — in the current case, in financial inclusion, security, and development — reach beyond simple data-based “evaluations” so that researchers can specify theory toward more coherent understanding and ask productive research questions that, when answered, may lead to positive changes in the real world.1
To be sure, these are tall expectations, but our experience over the years is that more rigorous applied social science is within reach, and that the resulting knowledge can indeed make positive contributions, with real effects on policy and practice. In that spirit, we hope this chapter may nudge both scholars and practitioners to be more thoughtful and productive in their research agendas going forward.
Thinking about Applied Social Research
We begin with several general themes in thinking about applied social research. Together, these themes express a very clear approach, with direct implications for designing research agendas. Though a bit on the idealistic side — with inevitable compromises necessary — these themes can serve as a rudder in charting a productive applied social research agenda.2
Build knowledge in actual circumstances
For this key point, we turn to the philosophy of Pragmatism, which emphasizes the value of empirical experience. Although somewhat of an oversimplification, Pragmatism defines truth as something that happens in the world. This very applied philosophy was seeded in the late nineteenth century, and flowered in the Progressive Era of the early twentieth century. William James, John Dewey, and Jane Addams are often identified as Pragmatists.3 And the social sciences, especially the applied social sciences, are sometimes interpreted as congenial applications of Pragmatist philosophy.
On this view, knowledge about the effects of social action is never divorced from experience, and instead is worked out in application. This perspective provides useful footing and an important guidepost for building and applying social knowledge. The overall implication is that assessing innovations can never depend solely on reasoning, nor can it depend on tangential but disconnected data, but must instead rest on the results of actual tests. This is a high standard for evidence, yet ultimately the correct one. One would not want to buy a new model of car that had never been driven.
Consider the individual in his or her social context
In the mid-twentieth century, both academic sociology and the field of social work evolved conceptual statements that considered the individual in his or her social context as the crux of explaining human action. One of the great contributions during that period was from Robert Merton, who, with an eye toward being useful, identified theories “of the middle range” as particularly relevant for understanding social interactions.4 Not long after, William Schwartz, William Gordon, and Harriett Bartlett in social work defined “person in environment” as a key to defining social work professional practice.5 Both of these conceptual statements consider the individual in his or her social context as the crux of explaining action.
The “middle range” lies between theories that rely solely on the biology and psychology of individuals (e.g., personality explanations of behavior), and theories that address only social structures (e.g., privilege and power explanations of social class). Both the social context and the individual contribute to action; that is, both the institutions and the individual “behave.” Indeed, sometimes institutions do most of the behaving; for example, we have often pointed out that 401(k) plans are highly institutionalized — once an individual is signed up, saving patterns are automatic and result very little from individual behaviors.
Social institutions can and should be specified and tested as specific constructs (e.g., in the above example automatic features would be an important construct), with the goal of explaining patterns of action not by individuals, but in large samples or full populations. Such tests are particularly productive for informing programs and policies — and this must be a main goal of applied social research. Scholars who plow these fertile fields of the “middle range” include social psychologists, institutional economists, institutional sociologists, some behavioral economists, and a majority of social work and public health scholars.
Ask questions well
Over several decades, we have aimed to teach our doctoral students and other younger scholars how to ask better questions. The essential steps are (1) systematically specify theory, which (2) sets the stage for clearly stated hypotheses, that (3) can be represented by a practical intervention, (i.e., the hypotheses are subject to test in the real world), with (4) such tests yielding applied knowledge, that (5) documents efficacy and efficiency (or lack of these) of the intervention being tested. In this way, theory is used not as a brilliant person’s idea, nor as a forgone conclusion, nor as a kind of decoration before data are presentation, but instead as a method for rigorous, efficient, and productive inquiry by applied scholars.
Of course, this is just a description of deduction in applied social science. Deduction is a powerful tool — perhaps the single most important contribution of the Enlightenment — though many applied social scholars do not frame and conduct their research in this way. Engaging in scientific deduction is certainly not the only path to knowledge, but it has proved a very efficient and productive method of building knowledge. Therefore, applied scholars should have very good reasons before deciding not to use it.
Above all, intervention research should not be a fishing expedition, looking for anything that shows up in the net as an outcome. A disciplined focus on deduction — theory specification and testing hypotheses — is the way to avoid this.
Focus on solutions
A great deal of applied social research is designed to identify problems and causes of problems, or both. In most of this work is an implicit assumption that if we understand what causes a problem, we will know what to do about it. But this is sometimes not the case. Sometimes the causes of a problem are fixed or intractable (e.g., educational level of parents, racism), and sometimes the causes of problems do not matter in forming a solution. Overall, problem-oriented research is overdone.
More applied social research should be about identifying solutions, even, in some circumstances, when we do not know what causes the problems. Solution-oriented research directly informs interventions that can make a difference. In solution-focused research, we can identify two main types: (1) explanations of the solution (not the problem), and (2) effects of a solution that is put in place. The first type yields knowledge that informs programs and policies, and the second type provides the rationale for the intervention. At present, most applied social research does neither. If applied social science is to reach its potential, a rebalancing is necessary. A focus on empirical tests of purposeful action is consistent with the Pragmatism theme discussed above.
Guiding Concepts in Financial Well-Being
We turn next to key themes in the nature and definition of financial well-being, focusing on three areas in particular: financial capability, inclusion, and life course perspective. In fundamental ways, these three concepts are the foundation for effective financial products, programs, and policies. These are what we are all trying to achieve, so it is helpful to keep such core concepts in mind when selecting research questions. When research choices build upon core themes, knowledge accumulates more effectively.
Financial capability combines people’s ability to act and their opportunity to act in their best financial interests.6 Reflecting a person-in-environment perspective, financial capability is the interaction of individual knowledge and skills, behavior, and institutional structures that creates positive financial functioning. In other words, creating financial capability is not simply a matter of changing individual behavior, but of also changing institutions that play a role in shaping financial opportunities.7 Informed by capability theory of Amartya Sen in Development as Freedom8 and Martha Nussbaum in Women and Human Development,9 this view of financial capability refutes the assumption that financial vulnerability is a result of individual behavior alone and instead points to the interaction between individuals and social institutions that shape financial well-being. In other words, it differs from common usage that focuses on individual attributes, such as knowledge and skills, attitude, habit, motivation, confidence, self-efficacy, and behavior.10 Although some of these attributes suggest contextual variables, many current conceptualizations and measures do not specify and measure institutional variables alongside individual behaviors. This contextualized definition of financial capability follows from the “middle range” and “person in environment” discussions above.
Inclusion of the whole population
It is relatively easy to use the word “inclusion,” but much harder to achieve it in the real world. As indicated in the discussion of financial capability, access to quality financial services matters, and depending how access is structured, this could (or could not) lead to inclusion of the whole population in effective financial services. As our most successful example, by far the largest inclusive social policy in the United States is Social Security retirement, which achieves wide (though still not fully inclusive) coverage because it is mandatory and institutionalized in employment practice.
Working in a different area of social policy, the Center for Social Development (CSD) at Washington University in St. Louis is demonstrating in the SEED for Oklahoma Kids (SEED OK) experiment that full inclusion in Child Development Accounts (CDAs) — sometimes in this volume called Children’s Savings Accounts — can be designed and sustainably put into place.11 Inclusion in this case means all children in a population. Demonstrating full inclusion matters because it sets the stage for universal policy, and this research evidence matters. Results from SEED OK have directly informed CDA policy in the state of Maine that now also reaches all children with automatic accounts opened at birth.12 Full inclusion is made possible by using the state college savings (529) plan as a platform. SEED OK is an explicit and purposeful test of a fully inclusive policy. This is not the same as undertaking research on many child account projects and hoping that more of them may someday add up to something like inclusion. Instead, inclusive policy is the primary research question in SEED OK.
At CSD, we are aiming for a CDA policy model that can one day bring in all children in the nation — meaning every newborn. A nationwide 529 plan, or something like it, will be able to support such an inclusive policy. This is possible because government(s) can require that a 529 financial provider automatically open accounts for all children. In other words, there can be a positive public role in CDAs, even while assets are managed in the private sector. At the end of the day, a public role will be necessary for full inclusion in CDAs.
This example of applied research on inclusive policy is relevant for other types of financial services as well. Other potential candidates are tests of inclusive transaction accounts, tests of inclusive emergency savings accounts, and as a likely delivery system, tests of universal financial access via cellphone. We have little doubt that more such tests of inclusive financial delivery will occur in the not-too-distant future — internet technology now makes them very possible, perhaps inevitable. Indeed, we can imagine a world where inclusive finance becomes as efficient, easy, and safe as plumbing and clean water running to all homes. But for this to occur, purposeful research on financial inclusion must be in the forefront.
Developing financial capability and assets is a lifelong process.13 The life course offers a framework for studying financial well-being “at the nexus of social pathways, developmental trajectories, and social change.”14 The life course perspective underscores the reality that although individual choices and decisions matter a great deal, these decisions are shaped by and interact with other factors, such as the timing of events in people’s lifetimes, the circumstances and actions of the people around them, and the historical times and the place when and where people live.15
Timing, especially when people experience advantage or disadvantage, may shape the next stage in life in important ways, often with persistent effects over the life span. For example, the age when a person gets a job with retirement savings benefits has significant implications for total savings the person has in retirement. Linked lives draws attention to how others affect people’s well-being. For instance, a woman may withdraw from paid work to care for a sick spouse or an older relative. Historical time and place affects access to opportunities and constraints in socioeconomic, cultural, historical, and geographical conditions. As Elder and colleagues write, “when times change lives change.” Periods of war, economic recession, technological shifts, and cultural transformations affect entire age cohorts in different ways.16 For example, in financial services we see cohort effects in lower homeownership rates for the “millennial” generation following the Great Recession. Moreover, these effects are compounded for populations of color. Understanding these historical and sociological differences in financial circumstances is a very important research agenda.
Testing Financial Innovations in the Field
We now turn to some “nuts and bolts” issues in applied social research. These topics arise only in actual research in the field (recall the Pragmatism theme above). Unfortunately, these field research topics are seldom addressed as research methods in scholarly articles, but they are fundamental.17
Assessing delivery in the field
Well-designed intervention studies can rigorously document effects on accounts, savings, and asset building, as well as the well-being of families and children. However, effective intervention research rests on several types of knowledge that research methods texts rarely address. In effect, putting the intervention in place in the field — sometimes called the “fidelity” of the treatment — can never be taken for granted. In this discussion, we sometimes use the example of child account research, though similar lessons apply broadly to other innovations in financial services and policies.
The financial product. As indicated above, the purposes and characteristics of the financial instrument are important to document.
- Primary purpose. What is the primary purpose of the financial product? To take the child account example, is the main purpose financial experience, education, and savings behavior, or is the primary purpose long-term asset building for development goals? Different financial instruments by design serve different purposes. Indeed, a standard recommendation of financial planners is to have both a liquid savings account and a long-term asset-building account. To avoid confusing apples with oranges, it is important to state the primary purpose upfront.
- Does the primary purpose include full inclusion? If so, how is inclusion defined? In the case of child accounts, does this mean all children? Or as in some “universal” child accounts, does this mean only a public school population? If so, we might ask what happens with children who are in parochial or other private school, those who are home schooled, and those who are in institutional care?
- Characteristics. Describe the financial product and its characteristics. In what ways do these characteristics reflect the primary purpose, or not? Characteristics may include universal or targeted, automatic versus sign up (or “opt out” vs. “opt in”), initial deposit or not, types of subsidies if any, and how subsidies are distributed.
Delivery and administration. As the proverb notes, “There’s many a slip ‘twixt the cup and the lip.” We know this to be true — alas, inevitable in the human condition — yet for convenience we often do not assess the slippage. We typically assume that the “intervention” is being “delivered” as designed. Some things to pay attention to include the following:
- Vehicle of delivery. What is the delivery system for the financial product, (i.e., how is the customer engaging with the product and who is in charge of that experience?)? Describe the purpose, auspice, applications, footprint of the delivery system, and details on traffic and use.
- Business fit. Is financial services the primary “business” of the organization? Is the organization involved in financial services, but this account is not a main focus? Especially, to what extent is offering the account “sustainable” based on revenue patterns for the business? For example, is a provider undertaking a boutique project for Community Reinvestment Act compliance or other purposes, or can this be a product that goes to scale and is sustained indefinitely?
- Tasks and responsibilities. What account administration tasks are performed by the sponsoring organization? What tasks are outsourced? For example, is there a third-party record keeper? How is all this working in practice?
- Efficacy of delivery. To what extent is the product delivered consistent with the original design and purpose?
- Data systems. Is there a reliable centralized data system for tracking performance?
- Potential for expansion. Is the provider likely to expand accounts to other settings or groups? If so, describe the likely expansion. If not, explain the situation.
Policy platform. As indicated above, many different local applications of a financial innovation will not necessarily add up to a comprehensive policy. Policy structure matters if full inclusion is to be achieved.
- Policy. To what extent is the financial product and delivery built on an underlying policy structure (i.e., defined and supported by local, state, or federal legislation)?
- Replication. Does a policy platform facilitate implementation of accounts to other geographical locations? What is the evidence that this is occurring, or not?
Fees, investment options, and expected growth of assets. Particularly in long-term asset building, net return (total earnings and gains, minus fees) matters. Topics to assess include the following:
- Fees. What are the separate and total fees associated with the accounts? Are any of the operational costs subsidized by a public funder or other party? Explain in detail.
- Investment options. What investment options are available in the accounts?
- Expected growth of assets over time. What is the range of anticipated growth of assets in the accounts over time?
Assessing actual financial results
Of course, real experience can (and often does) differ from intent and design. To make actual assessments, it is fundamental to have trusted and accurate data on what is occurring in the accounts.
Quality of data sources. Data quality is fundamental, yet this topic is undervalued in social science reporting.
- Data sources on account holding and amounts. What are the data sources on account holding and savings? Is there a systematic database? Are program sponsors entering data? How complete and accurate are the data?
- Reporting. What is the basis of prior reporting of program results? Did objective researchers clean, analyze, and present the data? Are systematic published reports available? And if not, why?
Data on Accounts, Savings, and Asset Accumulation. If long-term financial capability and asset building is the goal, important financial results include account holding, saving patterns, and asset accumulation. It is also important to ask how results differ by subgroup.18
- Data on account holding. What is the number of accounts? How does this compare with the target population for this initiative? How does it compare with the full population in this age group?
- Data on savings patterns and performance. What are the patterns of depositing in and withdrawing from the accounts?
- Data on asset accumulation over time. What is the asset accumulation (deposits minus withdrawals, plus earnings, minus fees) in the accounts over time? How do these results change over time?
Assessing financial security and development
The effect on individuals, families, and communities is the ultimate test of a financial innovation. For example, depending on purpose and design, child accounts may have effects beyond saving and asset accumulation. Indeed, some are specifically designed to achieve other important goals. Types of data sources and research studies, and their quality, are always important considerations.
Types of well-being outcomes. It is possible to assess a wide range of potential outcomes. Below we offer key categories of hypothesized effects of CDAs, based on theory and prior research. Certainly these are not the only categories of potential effects in financial innovation, but they are among the most important.
- Attitudes and outlooks of parents. Do child accounts cause parents to think differently about their child’s future, especially their expectations for the child’s educational attainment? Do parents in any way show attitudinal or mental health effects related to the accounts?
- Parent–child interactions. Do child accounts affect how parents interact with their children?
- Development of children. Do child accounts affect social–emotional and cognitive development of children?
- Primary schooling of children. Do child accounts affect school attendance, behaviors, and academic performance?
- Attitudes and outlooks of children. Do child accounts affect how children see their future, especially their expectations for educational attainment?
- Secondary schooling of children. Do child accounts affect high school attendance, behaviors, and performance? Especially, do child accounts affect high school graduation?
- Postsecondary schooling. Do child accounts affect college (or other postsecondary) enrollment, progress, and especially attainment of an educational credential?
- Financial capability of children and families. Do child accounts affect financial knowledge and skills of children and families? Are child accounts related to types and quality of other financial relationships?
- Long-term financial well-being. Do child accounts affect long-term asset building, avoidance of debt problems, and other measures of financial well-being?
This is a long list of potential outcomes, all based on our work with CDAs — and this list may also have broad applicability for research in financial innovation. What can we say about such a list? Foremost, it is important in policy and practice that an intervention, inasmuch as possible, has multiple positive effects. The notion of an intervention as solving a single problem is needlessly restrictive, and is another weakness of the problem-oriented approach. A far better idea is an intervention that creates multiple positive outcomes — a concept we have previously called a “strong intervention” or “strong policy.” At the same time, however, a single intervention cannot do all things, and happy talk must be avoided. The standard for testing effects should be reasonable hypotheses given theory and empirical data. For any financial innovation, careful thought should be given to this range of reasonable outcomes, and research methods should aim to document whether or not they occur.
Using Field Demonstration and Applied Research to Inform Financial Products, Programs, and Policies
Effective demonstration coupled with intervention research matters. The work in CDAs exemplifies this. Following the first proposal for lifelong asset building starting as early as birth,19 CFED, CSD, New America Foundation, and other partners implemented a major child account demonstration in SEED.20 Continuing this work, CSD is now testing fully inclusive CDAs with rigorous experimental methods in SEED OK. As a result of all this, and other demonstration and research activity, today the United States has reached a period of increased understanding and heightened interest in child accounts.
Little of this has occurred by chance. The strategy of purposeful demonstration, as Bob Friedman points out in his essay, in itself informs policy and practice by making the concept of child accounts real in the world. These concrete examples, with real people, generate media attention and serve to educate ordinary people, opinion leaders, policymakers, and others. At the same time, the knowledge gained in research has informed policy development in several states that have recently enacted and implemented CDAs.21 On this promising foundation, step by step, the United States can now move toward an inclusive and progressive CDA policy.
Further demonstration and research — models and learning — will be required. Researchers must always ask: What is most effective? What strategies can become fully inclusive, progressive, lifelong, and sustainable? As applied researchers continue to ask these important questions, knowledge will continue to spur policy and practice.
Every topic taken up in this volume has its particular history regarding development of concept and research. Content may be very different, yet the important themes and research considerations are largely the same. We hope this chapter has illuminated some of these themes.
As a closing thought, if productive research questions are asked, then research results will find an interested audience — a “social market.” This is a very different strategy than asking research questions and then later asking who will be interested in the results. As a simple guideline to avoid potential irrelevance, applied social researchers should ask themselves the following question before finalizing a research topic and plan: If the proposed research supports my hypotheses, what will happen in application? As applied researchers, we should have a clear and reasonably confident answer to this question to justify the time and expense of conducting high-quality applied social science. This is once again a high bar, yet when we achieve it, research can contribute to positive change.
- Here we focus on applied social science research, particularly focused on interventions. We acknowledge the importance of other data collection and analysis efforts to help us better understand the current state of household financial management in context — efforts such as the U.S. Financial Diaries project, the J.P. Morgan Chase & Co. Institute’s report on their own customer data, the Center for Financial Services Innovation financial health study, and the Assets & Opportunity Scorecard. However, we focus specifically on applied social research in this essay.
- This section is summarized from a more extended presentation by Michael Sherraden entitled Social Innovation: Vision, Knowledge, and Action (Oxford University Press, in progress).
- For example, William James, “Philosophical Conceptions and Practical Results,” paper delivered at Philosophical Union of the University of California at Berkeley, printed in the University Chronicle 1 (4) (1898).
- Robert Merton, Social Theory and Social Structure (New York: Free Press, 1968).
- For example, William E. Gordon, “A Critique of the Working Definition,” Social Work 7 (4) (1962): 3–13.
- This definition of financial capability was first presented by Elizabeth Johnson and Margaret S. Sherraden and is expanded upon in Margaret S. Sherraden, “Introduction.” In Financial Capability and Asset Building: Research, Education, Policy, and Practice, edited by Julie Birkenmaier, Margaret S. Sherraden, and Jami Curley (New York & Oxford: Oxford University Press, 2013).
- For example, Michael Sherraden and Michael S. Barr, “Institutions and Inclusion in Saving Policy. In Building Assets, Building Credit: Bridges and Barriers to Financial Services in Low-Income Communities, edited by Nicolas Retsinas and Eric Belsky (Washington, DC: Brookings Institution Press, 2005).
- Amartya Sen, Development as Freedom (New York: Anchor Books, 1999).
- Martha Nussbaum, Women and Human Development: The Capabilities Approach (Cambridge: Cambridge University Press, 2000).
- For example, Mike Dixon, “Rethinking Financial Capability: Lessons from Economic Psychology and Behavioural Finance.” (London: IPPR, June 2006); Annamaria Lusardi, “Americans’ Financial Capability.” (Financial Crisis Inquiry Commission, 2010).
- Yunju Nam et al., “Do Child Development Accounts Promote Account Holding, Saving, and Asset Accumulation for Children’s Future? Evidence from a Statewide Randomized Experiment,” Journal of Policy Analysis and Management, 32 (1) (2013): 6–33.
- Margaret Clancy and Michael Sherraden, “Automatic Deposits for All at Birth: Maine’s Harold Alfond College Challenge.” CSD Policy Report 14–05. (St. Louis: Washington University, Center for Social Development, 2014).
- Nancy Morrow-Howell and Margaret Sherraden, editors, Financial Capability and Asset Holding in Later Life (New York & Oxford: Oxford University Press, 2015). Content of this section on life course is drawn principally from this book.
- Glen Elder, Jr., Monica Johnson, and Robert Crosnoe, “The Emergence and Development of Life Course Theory.” In Handbook of the Life Course, edited by Jeylan Mortimer and Michael Shanahan (New York: Kluwer Academic/ Plenum, 2003).
- Glen Elder, Jr. and Janet Giele, “Life Course Studies: An Evolving Field.” In The Craft of Life Course Research, edited by Glen Elder, Jr. and Janet Giele (New York: Guilford, 2009).
- Karl Mayer, “The Sociology of the Life Course and Lifespan Psychology: Diverging or Converging Pathways?” In Understanding Human Development: Dialogues with Lifespan Psychology, edited by Ursula Staudinger and Ulman Lindenberger, (Dordrecht, The Netherlands: Kluwer Academic, 2003).
- This section is based on CSD research briefs in development by Michael Sherraden and Margaret Clancy. We have realized in discussions of child accounts that a number of nuts and bolts issues in design, delivery, data availability and quality, and measures of outcomes may benefit from clarification. This section is not at all “academic,” but rather is very practical, illustrating on-going issues.
- For example, Sondra Beverly et al., “Can Child Development Accounts be Inclusive? Early Evidence from a Statewide Experiment,” Children and Youth Services Review, 53 (C) (2015): 92–104.
- Michael Sherraden, Assets and the Poor: A New American Welfare Policy (Armonk, NY: M. E. Sharpe, 1991).
- See Lessons from SEED: A National Demonstration of Child Development Accounts, edited by Michael Sherraden and Julia Stevens (St. Louis: Washington University, Center for Social Development, 2010).
- For recent research and policy impacts see Sondra Beverly, Margaret Clancy, and Michael Sherraden, “The Early Positive Impacts of Child Development Accounts.” CSD Research Brief 15–08. (St. Louis: Washington University, Center for Social Development, 2015).