Relative burdens: family ties and the safety net.

• Author: Fennell, Lee Anne

TABLE OF CONTENTS INTRODUCTION I. THE NATURE OF THE PROBLEM A. Defining Dependency B. Distribution and Efficiency C. Shared Concerns, Divergent Interests D. Implications of a Constrained Policy Spectrum 1. Bracketing the Feasible Spectrum 2. Counterintuitive Results II. FAMILY STRATEGIES A. Exiting (or Not Entering) the Circle of Support B. Influencing Dependence of Existing Family Members C. Spending Less on Dependent Care D. Childbearing Decisions E. Shifting Costs onto Other Family Members F. Using Financial Products To Address Illiquidity and Risk III. TOWARDS A THEORETICAL APPROACH TO DISTRIBUTING DEPENDENCE COSTS A. The Valence and Use of Familial Strategies B. Categorizing Dependence 1. Controllability 2. Predictability 3. Verifiability 4. Temporal Lumps and Variances in Outcomes 5. Availability of Familial Exit and Nonentry C. Policy Directions 1. Harnessing Family Advantages 2. Altering the Relative Attractiveness of "Good" and "Bad" Strategies 3. Disaggregating Liquidity from Public Assistance 4. Recognizing the Potential of Public Risk Pooling D. Some Complications 1. Individuals and Families 2. Society Plays Games, Too CONCLUSION INTRODUCTION

Vast numbers of people require unreciprocated assistance from others in order to enjoy a minimally decent life. (1) Dependency is endemic to the human condition, (2) touching everyone for at least part of the life cycle. (3) Its causes are heterogeneous, (4) but its manifestations are always expensive. (5) How should the costs be distributed?

No domestic policy question of our time carries higher stakes or triggers more divisive public discourse.

One significant axis of controversy concerns the appropriate respective roles of the family and the state--society acting collectively--in bearing dependency burdens. (6) The modern welfare state serves some of the purposes that family units and larger voluntary social networks have served historically: insuring individuals against risk, smoothing fortunes over the life cycle, and providing for those who are not capable of self support. (7) Yet, families continue to bear much of the burden associated with dependency. (8) Whether the role of families should expand or continue to shrink is a subject of much debate. The large and diverse body of scholarship on this topic resists simple generalizations, (9) but both those who favor a larger cost-bearing role for the family and those who favor a larger cost-bearing role for the state tend to give insufficient attention to three crucial facets of society's cost-allocation problem. By engaging these shortcomings, this Article illuminates underappreciated features of the dependence allocation task, and takes some tentative steps towards outlining a policy agenda that accounts for them.

First, society's feasible choice set does not comprise an unlimited spectrum that runs from the extreme of no dependence assistance at all to the other extreme of completely socialized assistance. In fact, because unaddressed dependence is extremely costly to society as a whole, society has committed itself to provide at least a minimal level of public support for dependents who lack personal or familial resources. (10) The conventional tack of evaluating policy interventions against an implicit baseline of complete nonintervention misapprehends the nature of society's choice. Moreover, recognizing that society will serve at least a gap-filling function in addressing dependence leads to radically different policy prescriptions than those that might follow from an inquiry that assumes all options lie open.

Second, and closely related, dependence costs left to fall on families will not necessarily come to rest upon families. (11) Current and potential family members have at their disposal a number of strategies designed to reduce their exposure to dependence costs. These strategies produce results ranging from desirable cost reductions to costly deadweight losses. A high-profile example of costly family strategizing is the "Medicaid divorce" undertaken to qualify a dependent spouse for publicly funded, means-tested benefits. (12) Such cost-shifting techniques carry the potential to alter not only the final distribution of the costs, but also the total costs involved, and hence the efficiency of the arrangement. Yet, cost-shifting strategies are usually treated as isolated epiphenomena to be attacked in situ, rather than as symptoms of an overarching set of interactions between families and society that should globally influence our thinking about arrangements for dependence care. As a result, they have received relatively little sustained or systematic theoretical attention.

Third, the true impact of dependence burdens on families turns, in part, on the ability of families to spread costs temporally, or to pool the risks that give rise to them. Risk exposure and periods of illiquidity can lead to suboptimal investments in human capital, yielding results that are both distributionally problematic and inefficient. The question of access to financial products capable of spreading costs temporally and pooling risks, however, is often conflated in policy discussions with the question of who should pay for those financial products. Although linked in often complex ways, these issues can be separated, as the existence of private, market-based financial products designed to address liquidity shortfalls and risk attests. Where market products are unavailable, as is generally the case where the asset in need of protection and timely investment is human capital, policy interventions could focus on making appropriate products available rather than on directly altering the allocation of the costs of dependency. (13) The desirability of such alternatives depends on their relative performance in harnessing the useful elements of familial strategies while limiting the more destructive manifestations of family strategizing.

The analysis proceeds in three parts. Part I frames society's cost allocation problem. Of central importance is the convergence of societal and familial interests on the goal of appropriately addressing dependence. This convergence has the effect of limiting the feasible spectrum of policy choices. When coupled with often divergent interests about cost bearing, it sets the stage for strategic interactions between families and society as each attempts to bluff the other into bearing a larger share of dependency care costs. Part II details the strategies that families might employ in their attempts to reduce dependency exposure. Part III shows how this analysis generates a more useful way of approaching the question of distributing dependence burdens.

1. THE NATURE OF THE PROBLEM

   This Part examines the nature of society's cost allocation task and the constraints it faces in making allocation choices. Part I.A discusses some possible meanings of "dependence" and arrives at a working definition capable of meaningfully guiding our inquiry. Part I.B discusses two criteria that typically inform a society's choice about allocating dependency burdens--distributive justice and allocative efficiency. Part I.C sets out the ingredients of the strategic dynamic that society faces in assigning dependent care costs and explores how this dynamic restricts the feasible spectrum of policy choices. Part I.D illustrates how this restricted spectrum suggests very different policy approaches than those generated under the assumption that any and all cost assignments are feasible.

   1. Defining Dependency

      The first task is to specify with some precision what dependence means in the context of this Article's analysis. I have already suggested that a dependent person is someone who requires unreciprocated assistance in order to lead a minimally decent life, but I have not yet fleshed out what I mean by "requires" and "unreciprocated" and "assistance." By doing so here, I distinguish the meaning of dependence I employ from broader and narrower meanings that the term might plausibly be given. Defining dependence is often an ideologically charged exercise. Moreover, a society's legal, cultural, and institutional features set the background conditions against which the notion of dependence is constructed. (14) My objective here is not to take sides in ideological debates or to state what "is" and "is not" dependence for all times and purposes. Instead, I seek to arrive at a workable definition that fits with broad societal intuitions and that represents a meaningful category for which cost allocations are open to ongoing, broad-based debate. Only by using such a definition can the analysis developed here have meaningful traction for law and policy.

      To start, consider how we might best construe the idea of assistance. None of us is truly self-sufficient. (15) A healthy, working-age person with plenty of marketable skills and no disabilities is still incapable of single-handedly constructing, concocting, and cultivating all of the things necessary for a minimally decent life, while simultaneously protecting those things from theft and destruction. (16) Thus, what is generally regarded as "dependence" in America relates only to those needs for outside assistance that arise within a societal framework featuring a well-developed market economy and governmental institutions capable of reliably protecting resources. Those arrangements, themselves quite costly, (17) are essential in constructing the condition we think of as "independence."

      This line of reasoning presents a plausible case for viewing everyone as dependent in some sense, but such an expansive view of dependence is not very useful for present purposes. Society has already chosen to spread the costs associated with maintaining our particular form of social order throughout society, rather than to concentrate those costs on individuals, families, or other voluntary social networks. That fundamental decision is so firmly rooted in our system of government...