Below are the first 10 and last 10 pages of uncorrected machine-read text (when available) of this chapter, followed by the top 30 algorithmically extracted key phrases from the chapter as a whole.
Intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text on the opening pages of each chapter. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.
Do not use for reproduction, copying, pasting, or reading; exclusively for search engines.
OCR for page 1
Assessing the Impact of Severe Economic Recession on the Elderly: Summary of a Workshop Assessing the Impact of Severe Economic Recession on the Elderly INTRODUCTION The economic crisis and recession that began in 2008 has had a significant impact on the well-being of certain segments of the population, and its disruptive effects can be expected to last well into the future. The National Institute on Aging (NIA), concerned with this issue as it affects the older population in the United States, asked the National Research Council to review existing and ongoing research and to delineate the nature and dimensions of potential scientific inquiry in this area. The Committee on Population thus established the Steering Committee on the Challenges of Assessing the Impact of Severe Economic ecession on the Elderly to convene a meeting of experts to discuss these issues. The primary purpose of the workshop, held at the National Academies on June 10-11, 2010, was to help NIA gain insight into the kinds of questions that it should be asking, the research that it should be supporting, and the data that it should be collecting. Workshop participants considered the impact of the recession on the well-being and behavior of older adults, as well as implications for the rest of the population; potential research questions raised by the current recession and, going forward, recessions more generally; measurement and methodological challenges that confront scientific inquiry in this area; the potential for innovative research design; and future data collection efforts. Attendees included invited experts in the fields of economics, sociology, and epidemiology; staff from NIA and the U.S. Social Security Administration (SSA); and staff from the National Academies.
OCR for page 2
Assessing the Impact of Severe Economic Recession on the Elderly: Summary of a Workshop This report highlights the major issues that were raised in the workshop presentations and discussion. The workshop agenda and roster of participants are included in the Appendix. It is important to be specific about the nature of this report, which documents the information presented in the workshop presentations and discussions. The report is confined to the material presented by the workshop speakers and participants, and its purpose is to lay out the key ideas that emerged from the workshop. Neither the workshop nor this summary is intended as a comprehensive review of what is known about the topic, although it is a general reflection of the field. The presentations and discussions were limited by the time available for the workshop; a more comprehensive review and synthesis of relevant research knowledge would have to await further development. This report was prepared by a rapporteur and does not represent findings or recommendations that can be attributed to the steering committee. Rather, the report summarizes views expressed by workshop participants, and the committee is responsible only for its overall quality and accuracy as a record of what transpired at the workshop. Furthermore, the workshop was not designed to generate consensus conclusions or recommendations but focused instead on the identification of ideas, themes, and considerations that contribute to a better understanding of the issues. THE ECONOMIC CRISIS NOW AND IN HISTORICAL PERSPECTIVE Three workshop presentations provided context for the current situation. In the first presentation, Carmen Reinhart observed that the current economic crisis has affected developed economies, including those of the United States, Europe, and Japan, in a way that has not been seen since World War II. International experience indicates that severe financial crises are protracted affairs and that not all markets recover at an equal pace. Historically, declines in housing and equity prices in the aftermath of a financial crisis have been long-lived, lasting on average for 6 years and 3.4 years, respectively. In addition, Reinhart reported that typically it takes, on average, 4 years for per capita income to recover to precrisis levels, whereas unemployment rates typically continue to increase for a period of nearly 5 years. Financial crises are also usually debt crises: they are typically preceded by surges in private debt, and, in the 3 years following a financial crisis, government debt nearly doubles on average as revenues shrink and fiscal finances deteriorate (Reinhart and Rogoff, 2009). In the next presentation, Susann Rohwedder presented findings from the RAND American Life Panel (ALP), a nationally representative longitudinal survey that is conducted over the Internet and can collect data with
OCR for page 3
Assessing the Impact of Severe Economic Recession on the Elderly: Summary of a Workshop high frequency. High-frequency interviews have the advantage of capturing certain events that are more likely to be missed in low-frequency surveys (e.g., short spells of unemployment, spells without health insurance), and they can also track variation in measures that cannot be easily or reliably elicited by means of recall (e.g., expectations, mood and affect, satisfaction in certain domains). During 2008-2009, ALP surveys were conducted every 3 months in November, February-March, and May. Since then, shorter monthly surveys have supplemented the longer ones conducted every three months. Rohwedder reported that households surveyed by the ALP were regarded as having experienced “immediate financial distress” if they experienced at least one of the following: being behind on mortgage payments, having negative home equity, having to foreclose on a property, or either the respondent or the spouse being unemployed. A total of 36 percent of households experienced this form of financial distress between November 2008 and April 2010, with younger households and those with lower incomes being more affected. Losses in retirement savings, predominantly in the stock market, disproportionately affected the older and higher income population. The younger population, however, was more adversely affected by the housing downturn: in April 2010, 15.1 percent of homeowners younger than age 50 had negative home equity, compared with 6.6 percent of those ages 50-64 and 4.0 percent of those ages 65 and older. Rohwedder indicated that according to data collected between May 2009 and January 2010, the most common way of coping with income loss due to unemployment was through reductions in spending (89 percent of households), followed by reductions in the amount of money put into savings (50 percent of households). The sources that most commonly made up for lost income were unemployment benefits (39 percent of households), money from savings (35 percent of households), financial help from family or friends (28 percent of households), and borrowed money or credit card debt (22 percent of households). Transitioning from employment to unemployment increased the likelihood of dissatisfaction, depressive symptoms, and difficulties in sleeping, while transitioning from unemployment to employment had opposite effects. One important research topic has to do with improving the understanding of the long-term consequences of these events. The ALP also contains information on people’s expectations about the future. Survey respondents were asked about the subjective probability of a gain in the stock market both a year from now and 10 years from now. From November 2008 through May 2010, there was a modest increase in people’s 1-year expectations of a gain but a decline in their 10-year expectations; 1-year expectations of personal job loss also did not improve over this period. Rohwedder also indicated that households ages 65 and older
OCR for page 4
Assessing the Impact of Severe Economic Recession on the Elderly: Summary of a Workshop were consistently less likely than younger households to anticipate declines in spending. A participant suggested that Hawthorne effects (i.e., effects that result from a person’s being under observation), if present in a high-frequency survey such as the ALP, might mean that persistent questioning could cause people to focus on how miserable they are or to behave differently than they would otherwise. Rohwedder responded that the ALP is careful not to directly mention the financial crisis, and that it situates mood and affect measures at the beginning of the survey rather than after questions about experiences such as job loss. With a larger sample size, it may be possible to gain insight into possible survey effects by randomly assigning some people to be interviewed less frequently. In his presentation, Matthew Shapiro laid out findings from the Cognitive Economic Study, a survey resource whose sample frame mirrors that of the Health and Retirement Study (HRS) but which has more detailed measures of cognition, wealth, and determinants of financial decision making—such as financial sophistication (e.g., knowledge about portfolio diversification), preferences (e.g., risk tolerance), and expectations. Survey rounds were conducted in 2008 and 2009, and there are plans for additional waves in 2011 and 2013. Preliminary results suggest that there was no panic or systematic dumping of stock among respondents, according to Shapiro, although it is currently unclear whether this should be attributed to farsightedness or to inertia. Using a measure of processing ability known as “fluid intelligence,” researchers found that people in the highest cognition group had significantly more wealth at baseline than those in the lower cognition groups and also experienced greater losses during the crisis. However, those with higher cognition may have been better positioned to make adjustments, which could help explain the absence of a panic response. Shapiro also noted that even those with no financial wealth were greatly affected by the economic crisis in terms of eating less food (both at and away from home) and reducing their purchases of other nondurables; they also said that they were more likely to delay the purchase of a motor vehicle and delay retirement. The mean expected delay in retirement was 5.9 years for those with less than $10,000 in financial wealth, 4.06 years for those losing less than 10 percent of their wealth, and 3.9 years for those who lost 10 percent or more of their wealth. The relatively large magnitudes of expected delay among those postponing retirement suggest both the severity of the current recession and that economic uncertainty may play an important role. In the general discussion that followed the presentations, it was noted that most of the subjective probabilities that surveys currently ask about—regarding, for example, work, retirement, and the stock market—were originally developed as part of the HRS in the early 1990s. According to
OCR for page 5
Assessing the Impact of Severe Economic Recession on the Elderly: Summary of a Workshop participants, other types of expectations that could be asked about include expectations of indebtedness, higher taxes, lower public benefits, and lower living standards and lack of prosperity; discrepancies between what people say about retirement and their actual retirement behavior; and the permanency of expectations. One participant indicated that being more leveraged may be tied to expectations of higher volatility in the stock market, since small shocks can have large effects when people are highly leveraged. Another observed that Japan, which has had a severe reduction in expectations about growth even though it has not experienced catastrophic crises, may be a pertinent comparison. It was also suggested that it might be useful to adopt a comparative approach to get a sense of whether people think that they are worse off than, for example, Europeans or the Japanese; at the very least, questions could be broadened to include other individuals as points of comparison. MARKETS, EXPECTATIONS, AND PREFERENCES In the workshop session on markets, expectations, and preferences, Robert Willis spoke about the effects of the stock market decline on the investment expectations of U.S. households. A financial crisis is a natural experiment of rare magnitude, which may affect stock market expectations and could shed light on how people’s “mental models” convert public news into personal beliefs. This is especially relevant given the link between expectations and portfolio choice (Kezdi and Willis, 2008). Individual heterogeneity in expectations about asset returns may be important in explaining inequality in wealth (especially among the older population) as well as the portfolio choices and trading behavior of households before and throughout the crisis. The HRS includes expectation questions about issues such as longevity, inheritance, labor market prospects, retirement, and stock market returns. Respondents to the 2008 wave of the HRS answered the survey during the 12 months from February 2008 to February 2009, a time period that includes the stock market crash in early October. The 2008 wave asked two probability questions about the stock market, which in principle enables researchers to derive both the mean and the uncertainty (variance) of subjective return distributions at the individual level (Hudomiet, Kezdi, and Willis, 2011); it also enables researchers to track their evolution throughout the sample period. In 2009 and 2010, the HRS asked three probability questions of interviewees that allow for more elaborate analysis of expectations. The off-year 2009 survey also contains questions about expectations over a 10-year horizon, which allows long-run effects to be analyzed. Willis said that expectation questions in the HRS are asked in a subjective probability format and thus include forms of measurement error, such
OCR for page 6
Assessing the Impact of Severe Economic Recession on the Elderly: Summary of a Workshop as excessive rounding and focal answers, and may produce inconsistent answers that contradict the law of probability. Using a structural econometric model to separate out such “noise,” researchers tracked 1-year expectations of stock market returns with data from the 2008 wave of the HRS (Hudomiet, Kezdi, and Willis, 2011). They found that the subjective mean typically moved with stock prices—with the exception of October-November 2008 (around the time of the market crash), when expectations moved in the opposite direction; this means that people expected some sort of recovery in stock prices immediately after the crash but that their expectations became more pessimistic when there was no recovery. According to Willis, people with more financial knowledge generally have more optimistic beliefs and are less uncertain about the future. Although they disagreed little about future returns before October 2008, disagreement among them after the crash increased more than it did among the less knowledgeable. These results, according to Willis, raise a number of questions. First, average expectations were more pessimistic and more uncertain than historical returns even before the crash—would people have been better off if they had known more about the historical data? Second, why did people draw such different inferences from public data? Third, what are the implications for individual and aggregate behavior? The presentation by Andrew Caplin focused on a project called the Michigan-NYU-Vanguard Panel (MINYVan), a survey resource that may eventually provide insight into how retirees perceive their risks. The sampling frame of the planned large-scale survey will be a representative sample of age-eligible (consistent with the HRS) individual Vanguard account holders and 401(k) participants; information from the survey may also ultimately be tied to various kinds of administrative data, such as Social Security. MINYVan is motivated by a number of theoretical concerns. Caplin indicated that there are gaps in knowledge about people’s “commitment aversion” and their desire to maintain control of their resources; this is why, for example, there is not yet a good explanation for low rates of annuitization. More also needs to be learned about the concept of need as it is perceived, often rudimentarily, by retirees. In contrast to other parts of the life cycle, moreover, retirement is something that may be dominated by events—and shocks—for which people cannot fully adjust. MINYVan will therefore incorporate systematic event-based factors and the final survey will include structural questions that give respondents choices that could be useful complements to behavioral data. (For example, Ameriks and colleagues (2011) asked survey respondents to divide a prize between two locked boxes—a box for bequests and a box for private long-term care.) Before it develops a structural life-cycle model to serve as the basis for a large-scale survey instrument, the MINYVan project will conduct a qualitative survey of the target population—thereby identifying new model
OCR for page 7
Assessing the Impact of Severe Economic Recession on the Elderly: Summary of a Workshop elements and concerns that may be otherwise overlooked. The questions in the qualitative survey will have overarching themes, such as the definition of retirement; anticipated spending in early retirement; perceived risks to retirement financial plans; key risks, such as medical and long-term care expenses, the education of grandchildren, reductions in Social Security benefits, and new life-enhancing technologies; awareness of and interest in retirement financial products, such as annuities, long-term care insurance, and life insurance; and the impact of a financial crisis on trust in financial institutions. During the general discussion, it was noted that retirement is a dynamic process in which expectations, assets, and health and mental capital change dramatically over time (starting from the moment that people begin to think about retirement), and that it may be useful for MINYVan to do more than just take a snapshot at a given moment in time. The measurement of retirement with such data may also be complicated by the fact that financial firms are not notified when people stop working and that people no longer appear in administrative records after they have moved their money into other accounts; however, some of the issues related to the identification of retirement could be addressed through administrative linkages to SSA data. Several participants noted that further exploration is needed into how people understand their consumption needs. Consumption patterns tend to get set in mid-adulthood, and although people may indeed be smoothing their consumption over the life cycle, they may not be doing so at the full wealth-consuming level. In the United States, the most common explanation given by people who were eligible for Supplemental Security Income (a means-tested benefit program for the elderly) but did not take up the benefits was that they did not need them. As people reduce spending in response to an economic shock, they may also settle into a new long-term equilibrium because their definition of need has undergone a fundamental shift. More could also be learned about how people create expectations about long-term care needs and whether those expectations are affected by their own experiences with family members. It was suggested during the discussion that although it may be interesting to ask people about their expectations, there is also reason to believe that the link between what people say they expect to do and what they ultimately do can be tenuous. For example, people tend to retire earlier than they say they will and are more likely to take Social Security benefits at age 62 rather than at age 65 or older. Also, people tend to overestimate their competence and capabilities and may not even know or understand what it is that they are being asked about. Even so, the battery of financial sophistication items in the Cognitive Economic Study, which provides a quasi-objective measure of what people actually know, suggests that they have a fairly good idea of what they do and do not know; most people felt
OCR for page 8
Assessing the Impact of Severe Economic Recession on the Elderly: Summary of a Workshop that they did not know much about the financial measures in the survey and proved themselves correct. The problems faced by survey respondents in articulating subjective probabilities were also raised during the discussion, in response to which one participant referred to an innovative method of visual representation to elicit subjective probabilities from survey respondents (Delavande and Rohwedder, 2008). Participants discussed the possibility that people may not know things (e.g., about stock market returns) because they do not need to know them (e.g., because they do not have any money in the stock market). It was pointed out, however, that knowledge can be measured using the HRS by comparing what people say about their pension plans and Social Security benefits with the actual plans and programs. One observer noted the puzzling fact that although numeracy is a very important determinant of wealth, knowledge of pensions and Social Security—which could be taken as a sign of financial sophistication—has almost no relation to total wealth (see Gustman, Steinmeir, and Tabatai, 2010a). Indeed, although people may not know a lot about the specifics of their pension plans, they are still very capable of retiring when they would lose financially by not doing so. A participant also commented that even though many people have suggested that individual retirement accounts (IRAs) should be annuitized because people will otherwise spend their money recklessly, a very large fraction of people do not make withdrawals from their IRAs until they are required to do so by minimum distribution requirements. One participant said that behavioral economics may also need to do a better job of looking at age differences and connecting with the literature on the psychology of aging. It should not be assumed that different groups of people think in the same way. For example, according to the participant, experiments suggest that older people tend to filter out negative information—not because they are cognitively incapable, but for reasons of salience and affect. Another participant noted that the HRS asks respondents about the probability that they will be able to live independently and the probability that they will be able to think and reason well enough to manage their own affairs, which may be relevant to the potential vulnerability of older people to financial fraud. WORK, LABOR MARKETS, AND RETIREMENT Presenters Courtney Coile and Alan Gustman discussed aspects of work, labor markets, and retirement. According to Coile, since the beginning of the current economic crisis, the media and the public have focused on retirement delays that may result from plunging equity and housing markets. What has been missing, however, is a recognition that weak labor markets may lead to earlier retirement. Recent research (Coile and Levine, 2010) using
OCR for page 9
Assessing the Impact of Severe Economic Recession on the Elderly: Summary of a Workshop 30 years of Current Population Survey (CPS) data has found that long-term declines in stock prices (5-10 year returns) lead more educated workers, who are more likely to own stocks, to delay retirement, whereas increases in the unemployment rate lead less educated workers, who are more likely to be affected by labor market shocks, to retire earlier. In general, people really start responding to stock and labor market fluctuations at age 62, suggesting an important role for the Social Security program in influencing how people respond to economic shocks. According to simulations, the net effect of the current crisis will be an increase in retirement, with approximately 250,000 people delaying retirement because of the stock market decline and 375,000 retiring early because of problems finding employment. The authors also found that fluctuations in housing prices do not affect retirement; Coile noted that this is consistent with work showing that households tend not to consume their housing wealth in retirement until they experience a shock, such as the death of a spouse or the entry of a family member into a nursing home (Venti and Wise, 2001). Recent research using the decennial census and the American Community Survey has also explored the effect of stock and labor market conditions around the time of retirement on retiree income (Coile and Levine, 2010). Poor stock returns reduce investment income for retirees in the top third of the income distribution, whereas a high unemployment rate reduces Social Security income (by an amount roughly corresponding to the earlier claiming of benefits) for retirees in the bottom two-thirds of the income distribution. The latter effects are much larger relative to total income than the former, suggesting that the problems faced by older workers when the labor market is weak merit greater attention from researchers, policy makers, and the public. In his presentation, Gustman indicated that, according to research using 2006 HRS data, the wealth held by those approaching retirement age is cushioned by Social Security and defined-benefit pensions (which have been in existence longer than defined-contribution plans) and is therefore not very vulnerable to stock market or housing price declines. Projections based on behavior during the dot-com boom and bust also suggest that retirement changes due to the stock market decline are likely to be small (Gustman, Steinmeir, and Tabatai, 2010b). Once layoffs and their effects are allowed for, the result may be earlier rather than later retirements as a result of the recession. For other descriptive studies of outcomes during the recession, see Munnell, Muldoon, and Sass, 2009; Sass, Monk, and Haverstick, 2010. According to Coile, other work on the effect of economic conditions on labor supply has found that changes in state or industry employment affect the employment rates of older workers (von Wachter, 2007); cities that were more affected by industry shocks had larger drops in labor force participa-
OCR for page 10
Assessing the Impact of Severe Economic Recession on the Elderly: Summary of a Workshop tion (Black and Liang, 2005); job transitions and labor force participation are related to metropolitan statistical area and state unemployment rates (Friedberg, Owyang, and Webb, 2008; Munnell et al., 2008); and early retirement is related to trends in industry-level employment (Hallberg, 2008). Research on the difficulties facing displaced workers has found that older workers are significantly less likely to be in the labor force several years after the displacement than are comparable nondisplaced workers, and that those who do find new jobs have substantially lower earnings than at their previous job (Chan and Stevens, 1999, 2001, 2004; Johnson and Kawachi, 2007; von Wachter, 2007). There is also evidence of age discrimination in hiring for women (Lahey, 2008). Coile noted that older workers who have been laid off tend to rely on Social Security rather than unemployment benefits for assistance (Coile and Levine, 2007), and record numbers of people have been collecting Social Security during the current downturn (see, e.g., Johnson and Mommaerts, 2010). A participant noted that Disability Insurance (DI) claims also go up when the unemployment rate increases, although it is difficult to take a month or even a year of new awards to estimate the precise effect because of the sizeable number of denied applications that end up in the appeals process (in which many denials are eventually overturned). Gustman said that although case studies of firms hiring older workers on a part-time basis suggest that they will have greater demand for older workers as the workforce ages, a full demand-side analysis is missing from the literature. At the market level, one would like to know about the vacancies that are suitable for older workers by industry and occupation, but there are no models of demand for continuing workers (including long-term job attachment and retirement) by industry and occupation. Demand-side structural analyses in this area have not been possible because production function analyses have not been able to adequately account for such factors as implicit contracts and specific training. Demand-side analysis could be facilitated by matching firm-based data to the HRS using Employer Identification Numbers (as pioneered by John Abowd and colleagues), although this would be a time-intensive undertaking. Coile listed a number of questions for the future study of labor supply. Do shocks affect labor force reentry, the use of bridge jobs, or retirement expectations? In the household context, are the effects of shocks muted (e.g., because of an added worker effect) or magnified (since both spouses may be affected)? Do shocks have an effect on the labor supply of workers in their 50s (perhaps related to retirement expectations)? What are the long-term labor market effects of the recession? Do shocks have a differential impact by socioeconomic status? Several participants also raised the issue of labor market expectations and noted that the HRS asks people how difficult it would be, if they were to lose their job, to get a similar one.
OCR for page 11
Assessing the Impact of Severe Economic Recession on the Elderly: Summary of a Workshop Gustman noted that one important challenge is to isolate the effects of the recession from other changes that have occurred over time, such as the rise in labor force participation of women and requirements that defined-benefit pensions credit work after the usual retirement age. Other changes include trends in the incentives from defined-benefit plans, first toward earlier retirement, then the weakening of those plans as a share of pension wealth. Coile also suggested several questions for the future study of well-being. If market conditions cause workers to retire earlier or later, what are the consequences for retiree well-being—which could be measured in a number of ways (e.g., income, consumption, health/mortality, happiness, living arrangements)? To what extent can government programs mitigate the effects of economic crisis on well-being? Gustman listed a number of analytical approaches for thinking about recession and retirement: descriptive analysis of dependent variables pertaining to retirement and the labor market; descriptive analysis of retirement intentions; reduced-form analysis with business cycle measures as independent variables and labor market outcomes as dependent variables; supply-side structural analysis that includes reoptimization in the face of recession; demand-side structural analysis, including demand-side responses to recession (which is complicated by long-term job attachment); and market-level analysis that joins supply and demand, determining the work of the older population in the long run and in the face of an economic downturn. Dependent variables of interest in these analyses include labor market outcomes (at each age and into the future) such as full-time work; job tenure; part-time work; full retirement; job search; flows among these states, including leaving the labor market and then coming back; and flows for husbands and wives. (Gustman observed that although structural models can handle the decisions of both husbands and wives, the interdependence of these decisions is also important.) Coile suggested that the ideal data set would have extensive labor market data (to identify retirements and layoffs), extensive data on income and wealth holdings (e.g., stocks, housing, pensions, Social Security), and data on multiple measures of well-being (e.g., consumption, health, happiness). It would also include a large sample of older workers; follow the same workers over time to examine reentry, the long-term effects of shocks, and so on; and include many cohorts of workers to cover multiple periods of expansion and recession. Although no single data set does all of this in reality, many do meet some of these criteria—for example, the Census/American Community Survey, the Consumer Expenditure Survey, the Current Population Survey, the Displaced Worker Supplement, the General Social Survey, the Health and Retirement Study, the Panel Study of Income Dynamics (PSID),
OCR for page 24
Assessing the Impact of Severe Economic Recession on the Elderly: Summary of a Workshop are still employed. Such people may be subject to higher levels of stress because of greater demands from their employers, longer working hours, and heightened economic insecurity. Fourth, what are the long-term implications of the negative mental health effects that may be caused by economic downturns? Ruhm (2000) found that a variety of physical health measures improved with unemployment, but that increases in state unemployment were associated with an increase in suicide and mental health disorders. Similarly, pilot data from an Internet consumer panel in Arizona, California, Florida, and Nevada, in summer 2008 suggest that serious psychological distress is more prevalent among those who experience greater strain in their housing situation (Pollack et al., 2010). Contradictory findings for physical and mental health need to be further investigated. Fifth, not enough is known about the unique effects of different exposures that increase in prevalence during recessions, such as unemployment, reductions in income, reductions in wealth, food insecurity, reduced spending on health care, and unaffordable housing, including mortgage default. Which types of disadvantage are particularly harmful to health? (See Alley et al., 2009, for a discussion of the later life health consequences of material disadvantage in health care, food, and housing.) One promising avenue of research is to find ways of using variation in such factors to determine their effect (distinct from income) on health. For example, state housing laws may be related to foreclosure rates in ways that are not related to income and could serve as good instruments for looking at the potential relationship between foreclosures and health. Other examples may exist for such factors as food insecurity and health care access. Sixth, how do the health effects of individual-level disadvantage interact with community-level economic factors? Does the health effect of becoming unemployed, for example, depend on whether there is a high level of unemployment in the community? On one hand, if an experience is normative, there may be less stigma and easier access to supportive resources. On the other hand, a recession might have negative effects on the community environment (e.g., with respect to crime) and reduce the local resource base for services. Looking more specifically at the current recession, David Weir reported that the Gallup Well-Being Index (WBI) registered significant declines in fall 2008 but recovered substantially by May 2009. Changes in the overall WBI during this period were driven by changes in the “life satisfaction” component—that is, “Cantril’s ladder.”1 The health behavior component 1 Cantril’s self-anchoring striving scale consists of two ladders from 0 to 10, in which people put themselves between the worst possible life (0) and best possible life (10) today and in 5 years; the two are then averaged. People who end up at a 3 or below are said to be in “misery.”
OCR for page 25
Assessing the Impact of Severe Economic Recession on the Elderly: Summary of a Workshop (but not the physical or emotional health components) also tracked the overall WBI relatively closely. The movement in the overall WBI during this period, though noticeable, was small relative to the gradient between the top and bottom income quartiles. Weir went on to explain that core HRS data from 2008, 2010, and beyond will help identify the magnitude of the effects of the downturn on health; international sister studies will permit cross-national comparisons. The HRS also conducted a postcrash Internet survey in May and June 2009 and a mail survey on well-being in November and December 2009, providing further data on the effects of the crisis. The 2009 HRS Internet survey included information on consumer sentiment; positive and negative affect; satisfaction and control across a variety of domains; assets, ownership, losses, and rebalancing of investment portfolios; mortgage issues for self and family; consumption; employment and retirement expectations; and health behaviors. Comparing the same people across 2008 and 2009, Weir reported several findings: (1) the average reported chances of working full-time past both age 62 and age 65 increased; (2) the percentage of people not satisfied with their financial situation increased, and the percentage very satisfied with their financial situation decreased; (3) the percentage with no depressive symptoms decreased, and the percentage with four or more depressive symptoms increased; and (4) the percentages experiencing both frequent pain and severe pain increased. Although the smoking rate went down, drinking (including binge drinking) and church attendance did not change much. The 2009 HRS well-being mail survey primarily had to do with hedonic well-being,2 although it also included several global well-being measures, such as Gallup’s Cantril’s Ladder, general life satisfaction, and domain satisfaction. All the measures of well-being are characterized by income and wealth gradients. Well-being also tends to increase until people are in their late 60s, after which it plateaus and then, around age 75, begins to decline. Weir suggested that the fact that these different measures of well-being track each other so well raises interesting questions about how people answer the survey questions and what the various measures reflect. A number of questions are common to the HRS core (2008), Internet (May and June 2009), and mail (November and December 2009) surveys. According to Weir, the mean difficulty of paying monthly bills increased steadily through all three surveys and over the course of the recession. Similarly, the percentage of people having difficulty paying bills increased between the core and Internet surveys and, to a lesser extent, between the 2 Hedonic well-being was based on a one-day recall of time spent on eight main activities and the seven affects associated with each of those activities; it was designed to be comparable to the Disability and Use of Time survey being done by PSID as a time diary.
OCR for page 26
Assessing the Impact of Severe Economic Recession on the Elderly: Summary of a Workshop Internet and mail surveys. Conversely, life satisfaction fell considerably between the core and Internet surveys and, to a lesser extent, between the Internet and mail surveys. LIVING ARRANGEMENTS AND TRANSFERS Two presenters spoke about the role of the family in times of recession. Kathleen McGarry suggested that it is useful to think of the family as a potential provider of insurance against financial shocks. The family may be able to insure against some risks more efficiently than the market (e.g., they may have lower monitoring costs and more complete information) and help correct for market failures. Insurance may have important behavioral effects even when a risky event does not occur and a “benefit” is not paid. For example, someone could go to school or take a new job in a higher risk occupation or industry because they know that the family will be there to assist if necessary. It is therefore important to remember that looking at actual transfers in the data misses the (implicit) transfers that would exist if someone were to need them; the insurance role of the family will be underestimated accordingly. One participant commented that family insurance may also be important for mental health, as the “stress buffering hypothesis” in the sociology literature suggests that social support can be effective in buffering the effects of stress on mental health (see, e.g., Wethington and Kessler, 1986). What matters is whether there is someone that people could call on for help, rather than if they actually received help. The importance of family insurance for physical health is less clear. McGarry went on to note that inter vivos transfers (i.e., transfers between living persons) are relatively common, with approximately 30 percent of HRS respondents making a transfer to at least one child in any wave and 15 percent of children receiving a transfer in any wave. Transfers are also relatively large, at approximately $1,500 per child and $3,000 per family; these amounts are even larger when summed over many years. Not surprisingly, these transfers are positively related to the income and wealth of the donor and negatively related to the income of the recipient—although parents do not entirely make up for the lost income of their children with these compensatory transfers. According to data from the HRS, financial transfers for college are also common, with approximately 60 percent of parents making such transfers and 40 percent of children receiving them. These transfers are large: the average contribution covered approximately half of total tuition and room and board, with total transfers per child averaging $11,445. When making transfers to older parents, according to McGarry, donors in the highest income quartile were more likely to give cash than time (which may be regarded as an in-kind transfer): 6.8 percent in the highest
OCR for page 27
Assessing the Impact of Severe Economic Recession on the Elderly: Summary of a Workshop income quartile gave “only time” to elderly parents, 11.6 percent “only cash,” and 1.3 percent “both.” The opposite was true for donors in the lowest income quartile: 5.9 percent in the lowest income quartile gave “only time” to elderly parents, 4.5 percent “only cash,” and 0.6 percent “both.” The increase in coresidence after welfare benefits were reduced by welfare reform is an example of the role played by in-kind transfers in the lower part of the income distribution. McGarry indicated that according to data from the PSID, older people who live alone had the highest incomes, followed by those living with children, those living with children who never left home (who may be disabled), and those living with others; concerns about the potential endogeneity of income (in which the level of income is affected by the choice of living arrangement3) are alleviated by the fact that similar patterns existed when these same individuals were younger (age 58). Comparing older people’s own incomes with the incomes of the households in which they reside (relative to the poverty line) also suggests that coresidence may reduce poverty among the older population. According to McGarry, older people who were less affected by the economic crisis might be expected to provide assistance to their adult children, some of whom may have returned home to live with them—that is, “boomerang” children. For other parents, the increased need for cash transfers to their adult children may be a drain on their resources. Those nearing retirement who lost jobs and/or assets may reduce cash transfers, invest less in the schooling of children (which raises the question of whether grandparents make up any of the difference), reduce bequests, or experience an increase in coresidence at later ages. The recession may also have delayed effects: parental support of adult children may affect the availability of resources late in life, bequests may decline, and schooling investments for the next generation may decrease. There could also be delayed positive effects, if boomerang children practice reciprocity or families grow closer. For a comparative perspective on economic shocks on family transfers, one could look at such events as the reunification of Germany (e.g., many experienced huge windfall gains as pension plans in East Germany were made identical to the pension plans in West Germany, whereas younger workers were likely to experience a positive shock in terms of lifetime labor income) and the financial crisis in Indonesia, which resulted in many families moving in together (see Thomas and Frankenberg, 2007). McGarry observed that the current recession makes it necessary to ana- 3 For example, Supplemental Security Income benefits are reduced for those who move in with others, people may stop working early because they know that they can live with their children, and children may give up their jobs and go back to school because they know that they can move in with their parents.
OCR for page 28
Assessing the Impact of Severe Economic Recession on the Elderly: Summary of a Workshop lyze all types of transfers (cash, time, coresidence) and in all directions. One can expect to see differences across the income distribution with, for example, high-income families being more likely to give cash and low-income families being more likely to coreside and make other in-kind transfers. The HRS is a good source of information for cash transfers to family members, although it does not contain information on transfers between siblings. There is also a measurement issue with regard to reciprocity, as more people report making transfers than report receiving them; it would therefore be useful to interview both parents and children. With regard to coresidence, McGarry thought it desirable to know not only the current situation but also potential alternatives. Who else could people have lived with, and why were those options not chosen? Another question is how resources are shared within the household—who pays for what? As with cash transfers, there are issues related to reciprocity—when people are asked whom the relationship benefits, they tend to report that it benefits the other person. It would therefore be useful to interview both parents and children to get information from both perspectives. It is also important, McGarry said, to measure the potential for assistance. The HRS has questions about whether someone is available who could be counted on for help. It would also be useful to find out whether children and parents share expectations, whose beliefs “win out,” and whether expectations change over time in response to financial and personal changes. More could also be learned about the preferences of both older people and their children for formal care and whether the care provided by children was a last resort because formal care was not affordable. The presentation by Linda Waite focused on the resources, demands, and environment associated with the living arrangements of older adults. Homeownership is high among older adults, and one of the obvious responses of younger adults affected by the economic downturn is to move in with them. At the same time, according to Waite, older adults who have been laid off may have a harder time finding work and may move in with their middle-aged children. Adding more people to living arrangements affects crowding, disorder, and the resources available and demands made (in terms of time and money). The National Social Life, Health, and Aging Project (NSLHAP), a nationally representative sample of adults ages 57-85, contains useful information on the living arrangements of older adults; interviews were carried out in 2005 and 2006, and the same respondents reinterviewed in 2010 and 2011. In the first wave of the NSLHAP, the most common living arrangements among older adults were living with a spouse (47.2 percent) and living alone (28.6 percent). Other living arrangements included living with a spouse and children (8.7 percent); children (4.3 percent); others (3.6 percent); a spouse and others (2.6 percent); a spouse, children, and others (2.6
OCR for page 29
Assessing the Impact of Severe Economic Recession on the Elderly: Summary of a Workshop percent); and children and others (2.5 percent). Perceived social support and demands varied greatly across living arrangements. The worst off (in terms of low perceived support and high perceived demands) were single people living with children and single people living with children and others (most of whom were presumably grandchildren); spouses, in contrast, were a big source of support and did not make many demands. Interviewers also rated the rooms in which the interviews were conducted according to noise, disorder, dirt, smell, clutter, and the repairs that needed to be made. People who lived only with their spouse had the most ordered households, while people who were single and lived with their children and others had the least ordered ones. Household disorder was highly correlated with C-reactive protein levels, so that people in the most disordered households had higher levels of systemic inflammation. Waite listed a number of questions raised by the economic downturn. How many older adults moved in with others? How many older adults took in others? Did home environments change and how? How many older adults lost family resources? Did relationships deteriorate in families under financial stress? (The NSLHAP also contains information on elder abuse.) Did social networks became smaller or change composition?4 How many older adults lost lobs, lost homes, or declared bankruptcy? How many older adults had family members who lost jobs, lost homes, or declared bankruptcy? Did financial transfers to and from family members change? What were the characteristics of family members most likely to be affected by these changes? During the general discussion, it was suggested that kin availability has been changing and is about to change even more rapidly, raising the question of whether people in the future are going to have fewer daughters and daughters-in-law (who are currently the primary caregivers) and whether stepchildren would provide equivalent care. African American men are particularly disadvantaged in kin availability because they are less likely to be married and less likely to have lived with their children. It was also noted that the PSID allows one to infer the amount of time stepchildren spend with their stepparents when growing up. HOUSING It was the increase in the foreclosure rate in the 2006-2007 period that ultimately led to the collapse of the subprime mortgage market and the current economic downturn, and two presentations at the meeting therefore 4 One participant noted that the HRS asks respondents whether they have access to the Internet and, if so, what they use it for. Many people say that they use the Internet to communicate with their children and grandchildren.
OCR for page 30
Assessing the Impact of Severe Economic Recession on the Elderly: Summary of a Workshop focused on aspects of housing. John Weicher observed that 2007 was the peak year for housing prices and the most recent year for the two major surveys relevant to this topic: the American Housing Survey and the Survey of Consumer Finances. As of 2007, approximately 80 percent of elderly households were homeowners, and roughly 55 percent owned their homes free and clear. About 25 percent had mortgages or home equity lines of credit, with very few having anything that looked like a subprime mortgage. In general, elderly households with mortgages have held them for a while5 and do not owe a lot on them.6 Home values have fallen by approximately 13 percent since 2007, with larger declines in the “sand states”—Arizona, California, Florida, and Nevada—in which a disproportionately large share of elderly owners live.7 The S&P 500, by comparison, fell by roughly 25 percent. Although some of the elderly are likely to be in trouble, especially in the sand states, elderly homeowners were on the whole less affected by the recession than other demographic groups. Weicher indicated that this is fortunate, since home equity is a large share of elderly households’ net worth (as measured in the Survey of Consumer Finances). In 2007, the median home value for the elderly ($170,000) represented approximately three-quarters of their median household wealth ($225,000). Retirement accounts, owned by 45 percent of the elderly, are the next most common asset; in 2007, the median value among those with accounts was $55,000. The typical elderly homeowner, who has a house that is worth more than five times his or her income ($30,000), is “house poor” compared with the typical younger homeowner, whose home value ($200,000) is about three times his or her income ($60,000). This has led to a growth in reverse mortgages, which allow elderly owners to turn home equity into current income. In his presentation, Joseph Tracy described the Federal Reserve Bank of New York (FRBNY) Credit Panel, a data set that can provide real-time information on the housing sector. The FRBNY recently entered into a partnership with the credit agency Equifax, in which Equifax draws 5 In 2007, the median origination year for mortgages held by elderly households was 2001 (with 17 median years remaining on the mortgage). Weicher indicated that although this may seem fairly recent—and in a normal cycle it might be—the first 6 years of this decade were years with very large volumes of mortgages and very large volumes of home purchases. For younger households, the median origination year was 2003 (with 23 median years remaining on the mortgage); prices appreciated substantially in nearly all markets between 2001 and 2003. 6 According to Weicher, the median loan-to-value ratio for the elderly is 32 percent, and the median outstanding balance is $59,000—about half of what these figures are for younger households. 7 Since 2007, house prices were down 50 percent in Nevada, 40 percent in Arizona, 40 percent in Florida, and 35 percent in California.
OCR for page 31
Assessing the Impact of Severe Economic Recession on the Elderly: Summary of a Workshop a 5 percent random sample of households, pulls together all of their credit files, and updates the information on a quarterly basis. This allows households to be tracked through time and across geographic locations (something that is not possible with mortgage data); the data set will be refreshed so that it remains a random sample. Equifax also created 10 years of history that allow the events leading up to the current crisis to be tracked by the FRBNY Credit Panel. One of the challenges of establishing loan-to-value ratios for housing has been knowing the combined loan-to-value ratio for people who have multiple liens on a home; the FRBNY Credit Panel allows these to be linked up and therefore provides a more complete picture than typical data sets on housing. The FRBNY is also attempting to integrate information about valuation because it has information on the debts on homes but not on the value of the house itself; it is working with First American and also using its other loan-level data in order to do so. FRBNY Credit Panel data indicate that there is a sizeable gap between rates of official and effective homeownership after subtracting negative equity. Although mortgage debt was less of an issue for older people, those who did have mortgages experienced similar increases in 90-day delinquency rates. The FRBNY Credit Panel does not contain demographic information other than age; Home Mortgage Disclosure Act data, if they could be merged with the Equifax data, would broaden the demographic information available. Several participants suggested that the age profile of bankruptcies could be interesting to look at. On one hand, older people may have more resources on which to draw; on the other hand, if they have already bought their homes, they may be less worried about how they would appear to future creditors if they were to default. A graph of credit ratings by age before and after the economic crisis might also be informative. It was also suggested during the general discussion that the median U.S. household has held onto its housing and equity assets throughout the market boom and bust period and therefore suffered minimal losses in well-being; a small number of households even came out ahead because they sold their assets at the right time. The core of the problem consists of the losses experienced by the small number of households that bought and sold their assets at precisely the wrong time. There were probably not many older homeowners in that negative tail, although older people who do want to sell their homes (and move into a retirement home) may now find it difficult to do so at a preferred price. Older people also do not have the luxury of adjusting their retirement date and will not be able to make up their losses in housing wealth to the same extent as younger people. Moreover, given the role of the family as a potential provider of insurance
OCR for page 32
Assessing the Impact of Severe Economic Recession on the Elderly: Summary of a Workshop against financial shocks, older people could be indirectly affected by the housing losses of their children. GENERAL DISCUSSION: PRIORITIES FOR DATA COLLECTION In the general discussion on priorities for data collection, one topic that emerged had to do with the frequency with which surveys should be carried out. Given the impracticalities of having high-frequency surveys in the field on a continuous basis, several participants suggested that it might be useful to target a subsample of a major survey (such as the HRS) for more frequent interviews, perhaps identifying vulnerable populations that were experiencing or were at risk of experiencing certain events and then following up with them at short notice for more detailed interviews. The success of such an effort would hinge on the ability to establish protocols and procedures for going into the field at precisely the right time. One participant suggested that the experience of the ALP, which came about as part of the project on Internet interviewing and the HRS, may be instructive. One of the purposes of the project was to make the HRS instrument suitable for Internet interviewing, and all HRS modules are now being administered to ALP respondents. Roughly two surveys are administered per month, and they are often fairly short (around 20 minutes) and do not seem to be onerous; response rates are around 80 percent. The ambition is to expand the size of the sample (currently around 2,500), which will make it easier to experiment and do other things with the survey. The importance of communication and collaboration among different surveys, as well as the need for flexibility and innovation (without necessarily having such experimentation being tied to particular projects), was also noted. For example, something like the MINYVan survey could be used as a test bed for questions that ultimately end up in a special module of the HRS. Or there could be an ongoing test base study that receives input from multiple contributors, encourages collaboration and experimentation, and contains ample room for error. One participant commented that time-use data are fairly sensitive measures of what happens during a recession (whether people spend more time watching television, looking for a job, etc.) and could be useful if they were reported frequently enough. Participants also discussed the potential for merging data sets with proprietary information such as mortgage and credit records. Informed consent was mentioned as a serious issue, as people could provide approval but then forget that they did so; very general consents also might not hold up for more specific purposes. Scandal in one data set can very easily spread to others, causing response rates to plummet. It was suggested that these issues might best be approached through pilot surveys that are separate and distinct from other ongoing surveys. Two other issues
OCR for page 33
Assessing the Impact of Severe Economic Recession on the Elderly: Summary of a Workshop that were raised relating to merging data sets with proprietary information were the operational risks of data getting lost or misplaced during transfers and the problematic nature of geography as a potential identifier. One participant brought attention to two data systems that were not mentioned during the workshop sessions. First, the 2007 wave of the Survey of Consumer Finances has been made into a panel, with reinterview information collected in 2009; the panel data will be made publicly available in early 2011. Second, the 2008 panel of SIPP began in fall 2008 to collect detailed information from the previous 4 months. Because of supplemental funding from SSA, SIPP administered topical modules on pension coverage and retirement accounts in summer 2009 and assets and liabilities in fall 2009.
OCR for page 34
Assessing the Impact of Severe Economic Recession on the Elderly: Summary of a Workshop This page intentionally left blank.