The Material Convoy After Age 50

Abstract

Objectives.

Possessions constitute a dynamic “material convoy” that accumulates across adulthood to furnish role enactments and the development of the self. Following a familiar life course arc, older people should hypothetically release the possessions that equipped the daily lives that they no longer have.

Method.

We use new survey data on possession divestment from the 2010 Health and Retirement Study to assess activity on behalf of the material convoy after age 50.

Results.

After age 50, people are progressively less likely to divest themselves of belongings. After age 70, about 30% of persons say that they have done nothing in the past year to clean out, give away, or donate things, and over 80% have sold nothing. We tested whether divestments diminish with age because they do not seem necessary or because of health limitations, but the age pattern is robust, suggesting retention of the material convoy in later life.

Discussion.

Further research on this age pattern might consider housing, the construction of the self, and social networks as explanations for retention. Inertia toward the material convoy maintains the availability and comfort of things, but it may also lead to a predicament wherein the collection becomes a worry for self and others.

How people live with things has long been a research interest of anthropology and more recently of consumption studies. The question also has life course dimensions because material objects, as possessions, are long lasting, whereas human interests are changeable and lives are limited. Although people’s management of possessions is a theme at all stages of the life course, in this paper we focus on later life when individuals have likely accrued a larger body of possessions but also when their continued keeping can be a predicament for oneself and others.

Possessions constitute a “material convoy” that is borne from cradle to grave and from place to place (Ekerdt, in press; Smith & Ekerdt, 2011). Analogous to Kahn and Antonucci’s (1980) notion of a convoy of social support, the convoy of material support is likewise dynamic with a changing size and composition over time. Items assume a place in the convoy in order to support daily life and functioning, the enactment of social roles, and projects of self-development or presentation. Items also drop away over time (Ekerdt, 2009; Roster, 2001).

There is as yet no method by which to quantify the material convoy so as to trace it over time. The size or composition of an individual’s or household’s collection is difficult to characterize reliably because its items are practically uncountable. Although it is possible to self-report or enumerate large, singular objects (cars, televisions), other categories of belongings appear in sets and multiples (tools, jewelry, tax records, and cleaning supplies) of which people may not even be aware. There have been attempts by observers to quantify possessions in situ. Gosling, Craik, Martin, and Pryor (2005) needed 400 categories to code just the visible contents of single sleeping rooms. Arnold, Graesch, Ragazzini, and Ochs (2012) counted visible possessions in the homes of 32 nuclear families. In the first household that they analyzed, the team tallied 2,269 objects in two bedrooms and a living room. For certain kinds of objects, they were able to generate average counts per family (e.g., 438 books and magazines, 212 music CDs, 39 pairs of shoes, 52 objects affixed to the sides of refrigerators), but these, again, were only visible things. Some estimate of material convoy size might be had from the weights or volumes of material handled by moving companies (American Moving and Storage Assn., 2012), but households that hire movers are a selected group. Even if it were possible to quantify a person’s or household’s possessions at one point in time, empirical study of the dynamic nature of the material convoy would require follow-up assessments over years.

Nevertheless, we might expect that the totality of one’s possessions follows a temporal arc of more and then less, an iconic form in use in the West since the middle of the last millennium to depict a rising and falling trajectory of vigor and health (Cole, 1992). In the case of possessions, their expansion and contraction would hypothetically pace the life course density of social relationships (Cornwell, Laumann & Schumm, 2008) and the availability of financial resources (Bucks, Kennickell, Mach, & Moore, 2009).

Possessions accumulate across adulthood to furnish role enactments as workers, partners, parents, property owners, and cultural participants. The rising affluence that is typical of middle age facilitates the acquisition and accommodation of consumer goods. In addition, longer life experience yields more mementos and souvenirs that represent the self (Miller, 2008). Decades of birthdays, anniversaries, and seasonal occasions supply gifts that represent ties with others. Maturity may bring not only one’s own larger convoy but also the custody or inheritance of kin members’ heirlooms or property. At the same time, the flow is not wholly toward gain because living with things in daily life necessarily entails a certain amount of ridding, divestment, and disposal (Gregson, 2007). One reason for net accumulation rather than a steady state in the volume of possessions is that things get placed in “backstage” areas of the dwelling, perhaps in order to keep a tidy house, to evade the difficulty of disposal, or just because the space is available (Arnold et al., 2012; Gregson, 2007; Korosec-Serafaty, 1984). Gregson (2007) observes that what is out of sight generally goes unscrutinized for divestment. Storing contributes to keeping.

At some point later in adulthood, deaccumulation could be expected because the conditions for middle-age consumption typically lose force. Indeed, research on consumer expenditures in American households shows that spending on durable goods and clothing tends to peak around age 50 and then declines (Fernandez-Villaverde & Krueger, 2007; U.S. Bureau of Labor Statistics, 2013). People are no longer housing and furnishing growing children, nor, upon retirement, do they require work-related apparel or equipment. Widowhood leaves batches of personal possessions orphaned and candidates for disposal. Social connectedness is multifaceted, but age tends to see a reduction in network resources (Ajrouch, Blandon & Antonucci, 2005; Cornwell et al., 2008), thus reducing the need for cultural signifiers and goods that support sociality and hospitality (e.g., entertaining). Health and functional decline limit the ability to manage the further accommodation and care of possessions. Residential relocation, if it occurs, will commonly entail a smaller place (Banks, Blundell, Oldfield, & Smith, 2010) where possession downsizing will have been necessary. Finally, gathering awareness of shorter time horizons for accomplishing personal goals (Carstensen, 2006; Marshall, 1975) may prompt the distribution of possessions to others as a way to protect the things or project the self (Ekerdt, Luborsky, & Lysack, 2012). Although there are many exceptions to these several patterns, older people hypothetically should release quantities of the possessions that equipped the daily lives that they no longer have.

However, there are also reasons to predict continuity in the material convoy. Affluence can compensate any reduced functional ability to care for possessions and property; what was once do-it-yourself labor, such as cleaning or yard work, can be hired out. If there is no residential relocation, the dwelling can continue to contain the possessions of middle age. A strong rationale for retention of possessions lies in their function as an extension of the self and a scaffold for identity (Belk, 1988). The ongoing collection of belongings can secure continuity of the self in the face of aging and vulnerability. Finally, even if one would want to downsize and divest, waiting too long may compromise the intention if diminished ability limits the capacity to undertake the labor of dispossession or if shrinking social networks limit supportive assistance, how-to information, or actual human “conduits” for divestment and disposal (Luborsky, Lysack, and VanNuil, 2011).

So the stock and store of possessions could trend either way with advancing age: Toward divestment consistent with reduced role involvements, relationships, and capacity to accommodate things or toward retention. In the latter case, possessions can continue to comfort and gratify with their instrumental and symbolic properties, but their care and management can also become a worry to others (Smith & Ekerdt, 2011). Following the life course principle of “linked lives” (Elder, Johnson & Crosnoe, 2004), then the material convoy of one person or couple is ineluctably, wanted or not, the material convoy of their social counterparts and its disposition can become a shared concern.

Although the life course arc of possessions cannot be quantified, we can now assess activity on their behalf using experimental questions from the Health and Retirement Study (HRS), a national panel survey of Americans aged 50 and older. Questions to older adults about possessions have never appeared on any population survey that we know of. Heretofore, research on older people and possessions has been wholly qualitative and primarily focused on elders’ symbolic and emotional investments in things. The 2010 wave of the HRS carried a new series of four questions about activity in the past year: How often one has cleared out or reduced things in the home, and how many things one has sold, given away, or donated. These are questions about the totality of one’s things. If people are routinely divesting, we would generally expect to observe more activity or at least constant activity at older ages. Diminished activity would suggest a falloff in divestments and a retention of the material convoy.

A general, age-related trend will of course encompass all manner of variation in individual behavior. The pace and motivations for divestment will arise from the multiple reasons that people have for both keeping and disposing of things (Ekerdt, 2009). Just as one might have many different kinds of relations and even ambivalent relations with members of the social convoy (Fingerman & Hay, 2004), so too the meaning of one’s possessions can be multilayered and temporally fluid, enough so to counsel caution in generalizing why people keep things and release things. The net result, nonetheless, of individual actions is a material convoy that grows or diminishes and that must be continually accommodated.

The actual divestment pattern, as it turns out, is less activity at older ages (results to be presented below). Although many possible reasons for this pattern come to mind, here we hypothesize that a major reason for less possession management is the rising risk of poor health at older ages that can limit the capacity to carry out the cognitive, physical, social, and emotional tasks of divestment. Ridding the household of possessions is ordinarily laborious at any stage of the life course (Gregson, 2007; Gregson, Metcalfe & Crewe, 2007; Hetherington, 2004) but perhaps even more so in later life.

We must also recognize that any observed falloff in the frequency of possession management could well be an artifact of these cross-sectional data. Changes in the decade-by-decade composition of the population after age 50 (i.e., progressive selection) may account for the pattern. Most obviously, people may have already divested (perhaps by having moved) and so are content with their collection of belongings. At older ages, they may occupy more rental units and their households may be less populated, and so enclose comparatively fewer things to manage. At the same time, there is one characteristic of the older part of this age range that should predict more possession management: Women predominate, and women more often manage the household (England, 2010; Mannino & Deutsch, 2007).

In the analysis that follows, we describe age differences in possession management activity and test a health explanation for such differences. We also control for relevant characteristics that may confound the observation of age differences.

Method

Data Source and Measures

Data are drawn from the HRS, a nationally representative panel survey of Americans aged 50 and older that collects economic, health, and psychosocial information. Panel members have been resurveyed biennially since 1992 with new birth cohorts added over time. Details about sampling, measures, and response rates can be found at http://hrsonline.isr.umich.edu. This analysis is based on data from the 2010 wave of the HRS. The 2010 wave was administered to 22,042 persons in 15,286 households. Each survey wave of the HRS includes 10 experimental question modules, one of which is randomly assigned to respondents who also answered the core survey. Experimental modules are strictly limited to a few minutes in length. A 12-item module on possessions was completed by 1,898 persons, of whom 1,814 were aged 50+ (a 90% response rate). Specific modules are not assigned to both spouses in the same household, and so the dyadic “household structure” that typifies the HRS data does not apply for this analysis.

The survey items are specified in Table 1 along with descriptive statistics. As dependent variables for the analysis, activity toward possessions (in the last year) was measured by four questions with responses on a 1–3 scale. A general question asked about the frequency of activity “to clean out or reduce the number of things that you have” (not at all, a few times, many times). Three further questions asked about how many things had been sold by various means, given to family or friends, or donated to organizations (none, a few things, many things). These three divestment strategies—sales, gifts, and donations—are most frequently mentioned in downsizing advice manuals for older adults and by elders themselves in reporting their activities (Ekerdt, Sergeant, Dingel, & Bowen, 2004; Smith & Ekerdt, 2011). As phrased, the survey items asked respondents what “you” did with “your” belongings. It is possible, even likely, that married people answered collectively—on behalf of the household—thinking about what “we” did with “our” possessions. Because spouses hold most possessions jointly, such replies are nonetheless valid and true to the experience of possession management.

Table 1.

Survey Items in the Analysis With Descriptive Statistics by Age Decade

Possession management	Age, 50–59 (N = 635)	Age, 60–69 (N = 501)	Age, 70–79 (N = 461)	Age, 80+ (N = 213)
In the last year, how often have you gone through your home or other storage areas to clean out or reduce the number of things that you have? (column percent)
• Not at all	18	20	28	37
• A few times	59	63	62	50
• Many times	23	17	10	13
In the last year, how many of your belongings have you sold in a yard, garage, or estate sale, a community sale, on consignment or on the Internet? (column percent)
• None	76	77	81	85
• A few things	19	19	17	14
• Many things	5	4	2	1
In the last year, how many of your belongings have you given away to family or friends? (column percent)
• None	29	31	35	39
• A few things	50	52	54	47
• Many things	21	17	11	14
In the last year, how many of your belongings have you donated to charity, a church, or a community group? (column percent)
• None	27	22	31	32
• A few things	42	49	48	49
• Many things	31	29	21	19
Independent and control variables
Age, range 50–96, Mean (±SD)	54.7 (2.6)	64.2 (3.0)	74.0 (2.8)	84.6 (3.9)
Self- or spouse-reported health as good/very good/ excellent (vs poor or fair), %	60	62	64	64
Functional difficulties of self or spouse, range 0–10, Mean (±SD)	2.6 (2.9)	3.3 (2.9)	3.7 (2.7)	4.2 (2.7)
Appraisal of possession volume: More things than you need (vs fewer things/right amount), %.	56	61	62	53
Moved residence in the last 2 years, %	6	9	7	12
Home ownership, %	69	78	81	69
Marital status, Column %:
• Married	61	64	57	37
• Divorced/Separated	24	18	13	6
• Widowed	5	13	25	55
• Never married	11	5	6	2
Number of persons in household, Mean (±SD)	2.6 (1.5)	2.2 (1.3)	2.0 (1.0)	1.7 (0.9)
Female, %	56	61	60	64

Note. All items drawn from the 2010 Health and Retirement Study, N = 1,810.

To test explanations for any age pattern, we include two measures of health: Self-reported health status as a dichotomy, comparing excellent/very good/good health to fair/poor health; and a summed scale of difficulties with functional activities based on 12 yes/no questions about mobility, strength, and motor skills (Fonda & Herzog, 2004). If the respondent had a spouse as a HRS panel member who also responded to the 2010 core survey, the health of the “least-well” partner is taken to stand for the household, reasoning that it might be the health difficulties of either partner that determines behavior toward possessions.

We include control variables to account for the possibility that the composition of the analytic sample differs between its younger and older members. These include two measures of the possibility that divestment is not necessary. Appraisal of possession volume is measured by a question from the same experimental module as the dependent variables: whether the respondent claimed to have “more things than you need, fewer things than you need, or just the right amount.” Only a small percentage (6%) claimed “fewer things” than needed, so a binary variable (coded 0, 1) compares “more than you need” with “fewer” or the “right amount.” People who moved residence in the last two years (0, 1) may also be less likely to divest, having recently downsized for the move. Three variables point to possibly fuller houses and a greater need to divest but also, by contrast, more ability to maintain a material convoy: home ownership (0, 1); marital status as married, widowed, or divorced compared with having never been married; and number of persons in the household. As a final covariate, being female (0, 1) is more likely at older ages and a status that potentially points to a greater inclination toward household or possession management.

Analysis Strategy

The four survey items on possession management were not combined or scaled and so were analyzed as separate dependent variables. Response options for these questions were not all the same, and keeping them separate preserves detail about the different divestment strategies. The items were examined in cross-classification by age decades. Then they were dichotomized to indicate activity on behalf of possessions—cleaning out “a few, many times,” disposing “a few or many things”—and regressed on age as a continuous variable in logit models (Long, 1997). Further models controlled for factors that would contribute to levels of possession activity across age, including indicators of having already divested, living in a more- or less-populated household, and gender. Finally, models including control variables were then composed to test whether levels of health might explain the age effect on possession activity.

Missing data across the set of variables were negligible—1% or less for any survey item. Cases with missing information were deleted from the logit models, which were based on a range of 1,784–1,787 cases.

Findings

Table 1 presents responses on possession management items by age decade. Of general efforts to “clean out or reduce” belongings, claims to have done so “many times” decline across age decades (from 23% to 13%), whereas claims to have done this “not at all” are higher at older ages. Over age 70, the prevalence of inactivity (not at all) compared with frequent activity (many times) is nearly 3 to 1. The age pattern of less activity at older ages can also be seen for the three divestment strategies—selling, giving, and donating. Among the three specific strategies of divestment, donating is the most common, followed by gifts to family and friends, with sales as far less frequent. The hierarchy of strategies—apparent in all decades—is not surprising. Selling things requires skills, know-how, and special effort. Gifts to family and friends often require an occasion, premise, transition, or acknowledged turning point in the elder’s life (Ekerdt, Addington, & Hayter, 2011). Both selling and giving are social transactions that risk disappointment with the outcome. But donating to charities and agencies is easier because these recipients have regular hours and the organizations generally accept most goods.

The age pattern was confirmed in logit models predicting some level of possession activity using age as a continuous variable (Table 2). In all cases (coefficients under Model A), the estimates for age are negative, which points to greater inactivity at older ages. The negative coefficients remain (Model B) even after controlling for factors on which the sample may be progressively selected across age. The relationships between these covariates and possession management are worth noting. People who claim to have “more things than you need” have significantly more activity for all outcomes, and most strongly so for donations. Having more things than one needs entails more management of them. (Interestingly, having “more things” than needed is not a sentiment that falls off until later, being 56% in one’s 50s, 61% in one’s 60s, 62% in one’s 70s, but 53% among those 80 and older.) People who have moved recently (in the last two years) report more cleaning out, selling, and giving, not less, perhaps as concomitants of downsizing to move into smaller homes. Home owners are more likely to clean out and to make donations of things, and married persons are more likely to be active in all strategies. Divorced and widowed persons are also more active in donating things, perhaps winding up the material traces of former marital roles. Households with more persons, however, tend to do less cleaning out, consistent with a need to continue to furnish daily life. As expected, women are more active than men on behalf of possessions. None of these circumstances, however, reduces the negative effect of age on possession activity across the four types of behavior.

Table 2.

Unstandardized Coefficients From Logit Models That Regress Possession Activity (“A Few or Many Times, Things”) on Age, Control Variables, and Health Variables

	Clean out or reduce (N = 1,784)			Sold things (N = 1,786)			Given things (N = 1,787)			Donated things (N = 1,786)
	Model A	Model B	Model C	Model A	Model B	Model C	Model A	Model B	Model C	Model A	Model B	Model C
Age	−0.034*** (0.005)	−0.042*** (0.006)	−0.042*** (0.006)	−0.018*** (0.006)	−0.019** (0.006)	−0.018** (0.007)	−0.013** (0.005)	−0.018*** (0.006)	−0.019*** (0.006)	−0.010* (0.005)	−0.015* (0.006)	−0.012* (0.006)
Appraisal of volume (more than need)		0.268* (0.121)	0.245* (0.122)		0.310* (0.126)	0.281* (0.127)		0.584*** (0.107)	0.561*** (0.108)		0.742*** (0.114)	0.690*** (0.116)
Moved last 2 years		0.385+ (0.227)	0.401+ (0.228)		0.862*** (0.199)	0.891*** (0.200)		0.450* (0.206)	0.459* (0.206)		0.080 (0.205)	0.123 (0.207)
Home ownership		0.419** (0.141)	0.379** (0.142)		0.055 (0.152)	−0.003 (0.154)		0.096 (0.127)	0.075 (0.129)		0.612*** (0.130)	0.507*** (0.132)
Marital statusa:
Married		0.482* (0.240)	0.496* (0.241)		0.641* (0.276)	0.659* (0.277)		0.469* (0.214)	0.467* (0.214)		1.027*** (0.218)	1.127*** (0.221)
Divorced/Separated		0.103 (0.252)	0.096 (0.253)		0.299 (0.290)	0.293 (0.291)		0.180 (0.225)	0.177 (0.225)		0.571* (0.227)	0.574* (0.230)
Widowed		0.205 (0.262)	0.193 (0.262)		0.104 (0.311)	0.078 (0.313)		0.176 (0.235)	0.161 (0.235)		0.555* (0.240)	0.589* (0.243)
Number in the household		−0.146** (0.049)	−0.135** (0.049)		−0.049 (0.054)	−0.035 (0.054)		−0.098* (0.044)	−0.087+ (0.045)		−0.093* (0.047)	−0.075 (0.048)
Female		0.638*** (0.122)	0.645*** (0.123)		−0.070 (0.121)	−0.056 (0.122)		0.178+ (0.108)	0.173 (0.109)		0.373*** (0.117)	0.424*** (0.119)
Self-reported health as good to excellent			0.266+ (0.140)			0.274+ (0.148)			0.317* (0.126)			0.391** (0.134)
Functional difficulties			−0.009 (0.024)			−0.019 (0.025)			0.018 (0.022)			−0.083*** (0.023)

Note. Standard errors in parentheses. Health and Retirement Study, 2010.

^aCompared with “never married.”

⁺ p < .10; *p < .05; **p < .01; ***p < .001 (two-tailed tests).

To test the role of health as an explanation for the age pattern, we included measures of the household’s self-reported health and functional difficulties in further models. Health affects possession management in expected ways (Table 2, Model C). Good self-reported health is indeed associated with a greater likelihood of all types of activity; by the same token, functional difficulties inhibit the effort to donate things. Controlling for health and ability, however, leaves intact the negative effect of age on activity. Thus, the age pattern is robust with covariates in the models that might account for diminished possession activity across age.

To illustrate the age pattern, we calculated the predicted probabilities of activity at ages 55, 65, 75, and 85 at mean values or population proportions of the covariates. The results (Figure 1) depict lower levels of activity across age. After age 75, over one quarter of the respondents were predicted to do nothing to clean out, give, or donate things. The figure also illustrates how infrequently people sell things in later life.

Figure 1.

Predicted probabilities of possession activity “in the last year,” at selected ages: Clean out or reduce “a few or many times,” and given, sold and donated “a few or many things.”

It is possible that some of the oldest members of the sample have already divested and so, being content with the volume of their possessions, have lower activity in this regard. In an additional analysis, we created a cross-product term between age and appraisal of possession volume and added it to the models reported in Table 2. However, having “too many things” vs the “right amount” was not shown to affect divestments differentially by age (results available from authors).

We noted earlier that married persons may have answered the experimental survey items about possessions on behalf of both partners. Perhaps these “household” replies are somehow qualitatively different than the replies of single people, and this could confound the observation of an age pattern. We therefore estimated the full models (excepting the covariates for marital status and number of persons in the household) among the 738 single respondents. The coefficients for age in these results were negative and nearly identical to those shown under the four versions of Model C in Table 2: −0.045, −0.019, −0.024, and −0.012, respectively.

Discussion

The focus of our paper, the material convoy, lies at the intersection between the life course and people’s consumption practices of possession and disposal (Hetherington, 2004). In the temporal arcs of each, lives change but material goods can endure, raising questions about how well material environments suit aging individuals (Golant, 2011). We find that after age 50, people are progressively less likely to divest themselves of belongings. After age 70, about 30% of persons say that they have done nothing in the past year to clean out, give away, or donate things, and more than 80% have sold nothing. Nevertheless, smaller proportions claim to have divested many things, but that share declines across age as well. Perhaps divestments diminish because they do not seem necessary: one is content with the volume of belongings, has recently downsized for a residential move, is no longer storing things as a home owner, or is no longer sharing the household with a spouse. These circumstances were indeed shown to affect possession activity, but not the age pattern. Perhaps failing health impedes possession activity. Again, it does, but age remains a predictor for doing less.

We applied these data from the HRS to inform a question about the arc of the material convoy. Our findings support the view that some proportion of people habitually divest in later life, but larger numbers do not, suggesting a tendency to maintain collections of possessions across later life. But, absent some way to quantify possessions, it is hard to be sure. It could happen that divestment is a periodic activity, occurring in bursts, rather than a routine chore, and via these episodes, the convoy depopulates over the course of years. At the same time, our findings are consistent with anecdotal evidence about the overfull homes of older Americans and with an entire genre of trade books, manuals, and blogs that offers advice about elders and downsizing (Smith & Ekerdt, 2011; Stum, Bublz & Althoff, 2011). Major newspapers take up this topic with regularity (Brody, 2011; Greene, 2012; Span, 2012), and there are now service providers—senior move managers—who can assist with divestment and downsizing (www.nasmm.com). And, consistent with possession accumulation, a majority of the HRS respondents studied here volunteered that they have “more things” than they need.

The data resources of the HRS do not presently allow us to pursue full analyses of the age-related inertia toward possessions. Nonetheless, we can suggest three directions for further investigation. First, the nature of elders’ housing may facilitate the continued accommodation of the material convoy. The majority of older Americans are home owners and have a firm intention not to move (AARP, 2000). As long as one ages in place—the encouragement of which is doctrine in gerontology (Golant, 2008)—there is little practical reason to address the collection of belongings. Second, possession retention may be a story about the self. There are many resources for self-construction (Callero, 2003) and these include possessions, which people come to regard as parts of themselves (Belk, 1988; Dittmar, 1992). Numerous studies have confirmed the personal meaningfulness of selected possessions in later life (Kroger & Adair, 2008; Price, Arnould & Curasi, 2000; Rubinstein, 1987; Shenk, Kuwahara & Zablotsky, 2004). Considerable significance attaches to things that symbolize identity and ties to others. In the face of age-related losses and limitations, possessions can be held as real evidence of an unchanging self. Third, the declining size and intensity of social networks may play a role (Ajrouch, Blandon & Antonucci, 2005), reducing information and know-how about methods of divestment and narrowing potential recipients for gifts and donations. At the same time, increased contact with a reduced circle of close family members and friends may bring more instrumental help and advice for household management, but as a form of caring work.

Looking to the future, techniques for assessing the material convoy could help address a persistent question about cohort influences on possession retention. There is a firm conviction—often heard in casual conversation—that older adults who came through Depression and world war learned frugality from that experience and therefore keep things. They will eventually be succeeded by Baby Boom elders who grew up in an economy of relative abundance (Gilleard & Higgs, 2005). Will the postwar generation, too, have overprovisioned homes in later life, if for different reasons? As yet, current data sources have not allowed confirmation of such cohort effects on possession attachment and practices.

The optimal amount, quantity, volume, or store of possessions in later life is difficult to prescribe; one person’s clutter is another person’s comfort (Miller, 2008). One’s belongings can be a resource, achievement, and delight, but they may also by turns be a burden. There are good reasons to maintain the material convoy past age 50, including continued work and family responsibilities, leisure enjoyment, the protection of property, and intentions to pass things to others. Nonetheless, how to manage a household and its contents and the fate of those contents are questions that will press to the fore as the risk of vulnerability rises. To the extent that possessions—not just single, cherished items but in their totality—create emotional and environmental drag, individuals will be less adaptive should they need to make changes in place or even change places (Wald, Iwarsson & Oswald, 2012). The material convoy is not innocuous, mere stuff; its disposition must be undertaken sooner or later by someone. The realization of this is why the material convoy—personal property—becomes an intergenerational or collective matter (Ekerdt & Sergeant, 2006). For those with a stake in the safety and security of elders, possessions are one more way that the private becomes insistently public.

Funding

This work was supported by the National Institute on Aging at the National Institutes of Health (AG19978, AG30477). The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Institute on Aging or the National Institutes of Health.