esi economic security index report...shaky ground, rocked by economic tremors whose consequences...

standing on shaky GroundAMERICANS'�EXPERIENCES�WITH��ECONOMIC�INSECURITY

December 2010

ESI EconomIc SEcurIty IndEx

Jacob S. HackerPhilipp RehmMark Schlesinger

With support from The Rockefeller Foundation

The�Economic�Security�Index�(ESI),�developed�by�political�scientist��

Jacob�Hacker�and�a�multi-disciplinary�research�team�with�support�from�

the�Rockefeller�Foundation,�is�designed�to�provide�a�meaningful,�succinct�

measure�of�Americans’�economic�security.�Professor�Hacker�is�based�at�the�

Institution�for�Social�and�Policy�Studies�at�Yale�University,�which�aims�to�

facilitate�interdisciplinary�inquiry�in�the�social�sciences�and�research�into�

important�public�policy�arenas.

The�ESI�is�part�of�the�“Campaign�for�American�Workers”�initiative�of�

the�Rockefeller�Foundation.�The�initiative�strives�to�improve�economic�

security�among�American�workers�and�their�families,�in�part�by�improving�

knowledge�and�understanding�among�policymakers�and�thought�leaders�of�

the�dimensions�of�American�economic�security.

The�ESI�research�team�has�been�guided�by�a�technical�committee�retained�

by�the�Rockefeller�Foundation�to�provide�oversight�and�to�reinforce�the�

intellectual�and�analytical�integrity�of�the�resulting�work.�Chaired�by��

Brookings�Institution�economist�Henry�Aaron,�the�technical�committee�is��

comprised�of�seven�leading�experts�on�economic�security:

�� Henry�Aaron�(BROOkINgS�INSTITUTION)

�� gary�Burtless�(BROOkINgS�INSTITUTION)

�� Henry�Farber�(PRINCETON�UNIvERSITY)

�� Robert�greenstein�(PRESIDENT,�CENTER�ON�BUDgET�AND�POLICY�PRIORITIES)

�� Larry�Mishel�(DIRECTOR,�ECONOMIC�POLICY�INSTITUTE)

�� Alicia�Munnell�(DIRECTOR,�BOSTON�COLLEgE�CENTER�ON�RETIREMENT�RESEARCH)

�� Robert�Solow�(NOBEL�PRIzE�IN�ECONOMICS,�1987)


standing on shaky ground 1

Executive SummaryThis report draws on two new surveys fielded in the spring and fall of 2009 that paint a comprehensive and sobering portrait of Americans’ experience with economic insecurity and their capacity to cope with economic instability.

Collectively termed the Survey of Economic Risk Perceptions and Insecurity (SERPI), these surveys focus on four domains of economic life: employment, health care, family, and wealth. In each, they identify how often households experience various economic disruptions, how frequently those disruptions coincide across the four domains and persist over time, and what imprint those disruptions leave on Americans’ expectations, concerns, and ability to meet basic economic needs. Because parts of the survey were modeled after a separate 2007 poll, the SERPI also shows how Americans’ outlooks changed in response to the recent economic downturn.

These repeated snapshots convey a powerful picture of Americans standing on shaky ground, rocked by economic tremors whose consequences include not just worry and anxiety but severe economic hardship. Economic shocks were strikingly widespread in 2009.

n In the 18 months from March 2008 to September 2009, fully 93 percent of households experienced at least one substantial decline in their wealth or earnings or substantial increase in nondiscretionary spending, most often for medical needs or assistance to family members.

n Nearly seven in ten households saw their earnings substantially fall or their nondiscretionary expenses substantially rise.

n During this 18-month period, 23 percent of households reported a drop of at least a quarter of their annual household income. This confirms the findings of the Economic Security Index (ESI), an integrated measure of economic security based on publicly available statistics. Projections based on the ESI show that the share of Americans experiencing large income losses was higher in 2009 than at any point in the last quarter century.

Though intensified by the downturn, Americans’ economic insecurity has been growing for years, and it appears to have little diminished since 2009.

n While public concerns about job security rose dramatically as the economy weakened, worries about other risks to economic security—debt, retirement savings, medical costs, health insurance, and even housing stability— were already as common in 2007 as they were in the depths of the recession.

n According to separate opinion surveys, concerns about retirement savings and medical costs did not diminish at all between the summers of 2009 and 2010, and concerns about the job market declined only slightly.

2

Economic instability leads not just to uncertainty but to anxiety and economic hardship. This hardship is experienced not just by those at the bottom of the economic ladder but also by those squarely in the middle class.

n By the spring of 2009, 78 percent of Americans were quite worried about at least one risk to their overall economic security.

n Households experiencing major economic dislocations are, on average, three to four times more likely than otherwise comparable households to report being unable to meet multiple basic needs, such as food, shelter, and medical care.

n More than half of families with incomes between $60,000 and $100,000 that experience employment or medical disruptions report being unable to meet at least one basic economic need.

n Households with dependent children appear to be more at risk of experiencing problems in the face of economic instability than do households without children.

Looking forward, Americans appear extremely vulnerable to future economic shocks, in part because of the wearing down of their basic household “buffers” against economic risks, such as personal wealth and the potential to borrow from family and friends.

n By the fall of 2009, roughly three in ten Americans appeared highly vulnerable to additional shocks; perhaps as many as half appeared at least partially vulnerable, in the sense that their buffers against economic instability were limited.

n Buffers against economic instability are eroded by persisting and clustered economic shocks, depleting the security of even previously prepared economic households.

n While economic shocks are broad-based, the private buffers that households have against economic risks are much weaker for less affluent and less educated households than for higher-income and well-educated households.

Economic instability is so disruptive because shocks frequently persist over time, come clustered together, and occur at roughly the same time in multiple domains (employment, health care, family, and wealth).

n About half of all the economic shocks experienced in 2008 reoccured in the same households in 2009; these “persisting” shocks are associated with higher levels of unmet need.

n In a given domain, households often experienced repeated shocks in close succession. For example, more than a third of households that experienced a shock in employment or medical expenses experienced multiple shocks in the same area.

n Of those Americans who reported persisting disruptions of employment, three-quarters also experienced persisting shocks in at least one of the other three domains of economic life.

economic security project

3standing on shaky ground

Yet strikingly little is known about how Americans perceive their economic security, how those perceptions relate to their economic experiences, or how experienced instability affects households, well-being.

This report fills this gap. It draws on two new surveys fielded in the spring and fall of 2009, collectively termed the Survey of Economic Risk Perceptions and Insecurity (SERPI). The SERPI focuses on four major domains of Americans’ economic lives: (1) employment; (2) wealth (housing, retirement savings, and other asset holdings); (3) health care; and (4) families.

In each area, the SERPI examines not just the economic “shocks” that Americans experienced—whether loss of resources or increases in nondiscretionary spending, such as medical costs—but also the effects of those shocks on Americans’ lives, expectations, and concerns for the future. Because parts of the survey were modeled after a poll on economic security conducted just before the downturn, the SERPI also shows how Americans’ outlooks changed in response to the recession.

As suggested by the title of this report, many Americans are—and see themselves to be—standing on shaky ground. The recent economic downturn represented an especially powerful quake, emanating from two distinctive epicenters: declining employment and disrupted financial markets. Yet Americans faced jarring economic shocks even before the downturn, and continue to do so today. The SERPI and related evidence suggest that economic insecurity has become the rule, not the exception, for many Americans—even in good times.

Economic shocks, like an earthquake, do not consist of a single tremor. Households experiencing an economic shock are also more likely to experience that same shock again in the near future, to face other shocks in the same domain, and to be hit simultaneously by shocks in other domains. This persistence and clustering means that even the most prudent households can lose their financial footing. Although poor households are most vulnerable because they often lack the financial resources to buffer economic disruptions, the security and well-being of even more affluent households are at risk.

The damage is often severe. Households experiencing major economic dislocations are, on average, three to four times more likely than otherwise comparable households to report being unable to meet multiple basic needs, such as food, shelter, and medical care. Strikingly, these effects are not limited

Standing on Shaky Groundamericans’ experiences with economic insecurity

Three years after the onset of the recession, public concerns about the economy remain high.

4 economic security project

to those at the bottom end of the income distribution: more than half of families with incomes between $60,000 and $100,000 that experience persisting medical or employment shocks report being unable to meet at least one basic economic need.

This pattern of repeated shocks underscores how crucial basic safeguards against economic insecurity are. Yet the SERPI suggests that Americans are not adequately protected against future economic instability. By the fall of 2009, between a third and a half of all Americans appeared vulnerable to future instability, in that they lacked substantial financial reserves, could not pay off their current debt, and/or believed they could not borrow more than $5,000 from family and friends. This is in part the result of past instability: households facing repeated economic shocks are about twice as likely to report having inadequate buffers to deal with future economic uncertainties.

The remainder of this report lays out the basis for these conclusions, starting with a description of the survey. The first part of the report examines the main dimensions of American economic security: how often families experience shocks; how they perceive the economic risks they face, including whether and when they feel anxiety about them; and how these shocks affect their economic well-being. The second part of the report considers how these experiences and perceptions differ across major socio-demographic groups in American society. The concluding section takes stock of these findings and considers their implications.

The Survey of Economic Risk Perceptions and Insecurity Long before the onset of the recent downturn, Americans worried about the stability of their finances. Yet surveys assessing Americans’ perceptions of economic security were both rare and limited in scope. Although researchers documented varied forms of increasing economic risk, efforts to assess public views and experiences were quite narrow and focused almost entirely on risks related to job loss.1 Recent surveys have filled in some of the gaps, but they too have generally asked questions about only one or two domains of risk.2 More important, these assessments of public attitudes have looked almost exclusively at Americans’ expressed worries, with little attention to their actual experiences or to their capacity to buffer economic shocks. 3 The SERPI in Brief

The Survey of Economic Risk Perceptions and Insecurity was designed to fill this void. Fielded twice in 2009, it measured a wide range of events and perceptions that could leave families feeling insecure, focusing on four broad domains: employment, medical care, wealth, and familial arrangements.4 The survey gauged Americans’ worries using a consistent set of questions that allows for comparison and ranking within and across these areas, as well as over time. The survey also included an extensive set of questions about people’s


encounters with unstable economic circumstances in both the recent and the more distant past. Another cluster of questions measured the capacity of households to safeguard themselves against economic risks or to buffer the financial shocks that they experience. Yet another set of questions assessed both the psychological consequences of insecurity and its relationship to households’ ability to meet basic needs involving food, housing, and essential medical care.

The SERPI was incorporated as an independent part of the 2008–2009 Panel Survey of the American National Election Studies, a survey funded by the National Science Foundation for the past half century. Panel surveys interview the same individuals repeatedly over time. With

the financial support of the Rockefeller Foundation, the SERPI was fielded as the sole questionnaire for the ANES panel in March (Wave 15) and September (Wave 21) of 2009. The panel was constructed to provide a representative sample of the American population aged 18 and older as of November 4, 2008.5 Data collected on economic experiences, perceptions, and expectations during these two waves were merged with socio-demographic and other personal characteristics collected from respondents in other waves of the survey.

Ensuring Comparable, Accurate Measures of Perceptions

To allow comparison of pre- and post-recession responses, parts of the SERPI were modeled after a poll sponsored by the Rockefeller Foundation in the spring of 2007. The core of this 2007 survey was a question about “your family’s economic security” defined in general terms.6 The survey also asked about concerns related to more specific economic risks, including losing one’s job, facing large out-of-pocket medical spending, and lacking adequate income to pay for a comfortable and secure retirement.7 All told, the survey asked about fifteen distinct economic risks. In each case, respondents were asked if they were “very,” “fairly,” “somewhat,” or “not at all” worried about each risk.8

To assess the frequency with which worries coincided with bad experiences, a matched set of questions was asked about respondents’ experiences with economic instability (referred to below as “experiencing an economic shock”). For example, worries about employment were matched with questions regarding whether any adults in the household had lost their jobs. Respondents in the March 2009 SERPI were asked about whether they had experienced the event

The SERPI was designed to complement the Economic Security Index (ESI), an integrated measure of economic security based on publicly available statistics. The ESI represents the share of Americans who experience a 25 percent or greater year-to-year decline in their available household resources—their income minus their medical costs and debt service—and who lack an adequate financial safety net to cushion the fall. According to the ESI, economic insecurity has worsened since the mid-1980s, with a projected one-in-five Americans experiencing 25 percent or greater losses in 2009. Yet the ESI provides limited information about the dimensions and causes of these large losses, or how Americans subjectively perceive these objective economic events—all of which the SERPI uniquely illuminates. For more on the ESI, visit economicsecurityindex.org.


in the prior twelve months, and those responding on the September 2009 wave were asked if they had experienced the event in the past six months. For those completing both waves of the survey, therefore, the SERPI provides a unique 18-month history from March 2008 to September 2009 of Americans’ experiences with economic risks during the recession. The appendix to this report describes the survey in greater detail and discusses some of the more technical issues raised by its implementation and interpretation.

Americans’ Experiences of Economic InsecurityAlmost every American’s economic security was disrupted in the 18 months between March 2008 and September 2009. These disruptions are evident in the frequency, character, and consequences of the economic shocks experienced by American households.

Economic Instability Touches Almost Every American

The SERPI included measures of eleven distinct shocks related to household earnings or nondiscretionary spending (see text box). During the 18 months preceding the fall of 2009, over 90 percent of respondents reported experiencing at least one shock in the domain of employment (two measures), wealth (three measures), health (three measures), or the family (three measures). While it is not surprising that the simultaneous downturn in the housing and stock markets destabilized wealth for many Americans who might otherwise have felt secure, almost seven in ten (68 percent) of all Americans experienced a shock related to domains other than wealth, either an unexpected loss of earnings or an increase in nondiscretionary spending.

These shocks were often quite substantial. During these 18 months, almost a quarter of all households reported a decline in earnings totaling 25 percent or more of annual income. This confirms the findings of the Economic Security

Economic Shocks Measured by the SERPIEmploymentn unemployed not by personal

choice (respondent or immediate family member)

n lost more than a month of work due to serious illness or injury

Health Caren lost health insurance n major out-of-pocket medical

expenses as the result of serious illness or injury (respondent or immediate family member)

n paid a lot more for health insurance than expected

Wealthn retirement benefits at work cut

substantiallyn value of investments or retirement

funds declined substantiallyn value of house declined

substantially

Familyn divorced or separated from

spousen spent a substantial sum helping

out extended familyn spouse/partner passed away


index, an integrated measure of economic security based on publicly available statistics that suggests that the share of Americans experiencing large income losses was higher in 2009 than at any point in the last quarter century. (For more information, see the prior report in this series, “Economic Security at Risk: Findings from the Economic Security Index,” July 2010.)

As Figure 1 shows, the most common form of economic disruption involved losses of household wealth, which were experienced by more than three-quarters of all households. Shocks related to medical costs were the second most common, experienced by about half of all households. A third of all households reported some disruption of employment; shocks caused by family needs were about as frequent.

Economic Shocks Often Persist over Time

Economic shocks may be isolated—a one-time or infrequent event in a family’s life. However, once experienced, a shock may reoccur persistently over time or coincide with other economic shocks. Because the SERPI assessed experiences at two points in time, it is possible to determine the proportion of those who experienced a shock in the year prior to March 2009 who also reported that same shock in the subsequent six months.

As it turns out, such “persisting” shocks are close to the norm among those experiencing an economic shock: About half of all respondents who experienced

Fig. 1 Prevalance of Economic ShocksSpring 2008 through Fall 2009

70%

80%

70%

80%

0%

10%

20%

30%

40%

50%

60%

70%

80%

Any Wealth Shock Housing Value Down Stock Value Down Retirement Benefits Cut

0%

10%

20%

30%

40%

50%

60%

70%

80%

Any Family Shock Financial Aid to Family Divorce Partner/Spouse Die

0%

10%

20%

30%

40%

50%

60%

70%

80%


0%

10%

20%

30%

40%

50%

60%

70%

80%


‐ ‐

80%80%

80%80%

80%80%

70%70%

70%70%

70%70%

60%60%60%60%60%60%

50% 50%50% 50%50% 50%

40%40%40%40%40%40%

30% 30%30% 30%30% 30%

20% 20%20% 20%20% 20%

10% 10%10% 10%10% 10%

0% 0%0% 0%

Any Employment Shock Lost Job Too Sick To Work Any Medical Cost Out‐of‐Pocket Cost Premium Cost Up Lost InsuranceAny Employment Shock Lost Job Too Sick To Work Any Medical Cost Out of Pocket Cost Premium Cost Up Lost InsuranceShockShock

0%

10%

20%

30%

40%

50%

60%

70%

80%

Any Employment Shock Lost Job Too Sick To Work

0%

10%

20%

30%

40%

50%

60%

70%

80%

Any Medical Cost Shock

Out‐of‐Pocket Cost Premium Cost Up Lost Insurance

70%

80%

70%

80%

0%

10%

20%

30%

40%

50%

60%

70%

80%


0%

10%

20%

30%

40%

50%

60%

70%

80%


0%

10%

20%

30%

40%

50%

60%

70%

80%


0%

10%

20%

30%

40%

50%

60%

70%

80%


‐ ‐

80%80%

80%80%

80%80%

70%70%

70%70%

70%70%

60%60%60%60%60%60%

50% 50%50% 50%50% 50%

40%40%40%40%40%40%

30% 30%30% 30%30% 30%

20% 20%20% 20%20% 20%

10% 10%10% 10%10% 10%

0% 0%0% 0%

Any Employment Shock Lost Job Too Sick To Work Any Medical Cost Out‐of‐Pocket Cost Premium Cost Up Lost InsuranceAny Employment Shock Lost Job Too Sick To Work Any Medical Cost Out of Pocket Cost Premium Cost Up Lost InsuranceShockShock

0%

10%

20%

30%

40%

50%

60%

70%

80%

Any Employment Shock Lost Job Too Sick To Work

0%

10%

20%

30%

40%

50%

60%

70%

80%

Any Medical Cost Shock

Out‐of‐Pocket Cost Premium Cost Up Lost Insurance

Employment Family Needs

Household Wealth Medical Costs

Perc

enta

ge o

f Am

eric

ans

Expe

rien

cing

Sho

cks

Too sick to Work for 1+ month

Major Out-of-Pocket Cost

Retirement Benefits Cut


a shock in the first year reported that it reoccurred in the next six months. Persisting shocks are most common in the wealth domain and least common in the family domain. As documented below, persisting shocks are more likely to produce negative effects for the households that experience them.

Multiple Shocks Are Common

Many households experience multiple shocks in the same domain. As Figure 2 shows, these are most common for declines in wealth: More than half of all Americans who experienced at least one wealth shock (such as a decline in stock value) reported other wealth shocks (such as declines in housing value and reduced retirement benefits) over the same time period. A multiplicity of shocks is least common for family disruption (only 10 percent who experienced a family shock reported multiple shocks). Employment and medical domains fall in between these extremes.

The most frequent form of clustering, however, occurs across risk domains rather than within them. Figure 3 indicates that households experiencing economic instability in one domain more often than not also experience shocks in other domains at the same time. For instance, almost three in four Americans who experienced a persisting shock to employment between the beginning of

Fig. 3 Persisting Shocks across Domains

Persisting Employment Shocks Only

Persisting Shocks in Employment + One Other Domain

Persisting Shocks in Employment + Two Other Domains

Persisting Shocks in All Four Domains





6.7%

23.3%

35.7%

34.3%

















0%10%20%30%40%50%60%70%80%

Employment Medical Cost Wealth Family

Frequency of Multiple Shocks in Each Domain

Only Single Shock Multiple Shocks

Fig. 2 Frequency of Multiple Shocks in Each Domain

Perc

enta

ge E

xper

ienc

ing

Shoc

ks

0%10%20%30%40%50%60%70%80%

Employment Medical Cost Wealth Family

Frequency of Multiple Shocks in Each Domain

Only Single Shock Multiple Shocks


2008 and the fall of 2009 also reported a persisting shock in one of the other domains (wealth, health, and family), and often in several of them at once. About a third of those with persisting employment shocks experienced a persisting shock in just one of the other three domains. But almost half (44 percent) experienced persisting shocks in three or more domains over these 18 months.

As these figures reveal, Americans’ lives are not simply disrupted by the occasional unfortunate happenstance. Many experience a series of economic shocks, the combined impact of which is much larger than any of the individual shocks. When economic shocks both cluster and persist, the lives of even the most prudent and careful households can be deeply disrupted and those

households’ expectations for the future can be profoundly unsettled.

Widespread Worries about Economic Insecurity

The deep economic downturn of 2009 gave rise to widespread concerns among Americans. Prior to the onset of the recession, about half of the American public worried somewhat about their economic security considered in the most general

terms, with a quarter “very” or “fairly” worried. By the spring of 2009, more than half (53 percent) of all Americans were very or fairly worried, with only about one in ten “not at all” concerned. By the fall, concerns had slightly abated as the public grew more sanguine about the stock market and the viability of their plans for retirement.

figure 4

Americans’ lives are not simply disrupted by the occasional unfortunate happenstance. Many experience a series of economic shocks.

Fig. 4 Rising concerns About Economic Insecurity2007 to 2009

2009

20%cen

er

2009 20 0%

20%

Sp Sp Sp Sp Sp Sp Sp Sp Sp Sp Sp Sp Sp Sp

Percen

tage

er

Sources of Insecurity

ll 2009 20 8% 18 1%

FIGURE 4

Intensity of Concern

Very Worried Fairly Worried Rising Concerns About Economic InsecurityRising Concerns About Economic InsecurityRising Concerns About Economic InsecurityEconomic Security Spring 2007 15.1% 12.5% 2007 to 20092007 to 20092007 to 2009

Spring 2009 24.3% 28.5%60%60%60%

F ll 2009Fall 19 7% 26 2%19.7% 26.2%

Employment Spring 2007 12.8% 6.1% 50%50%

Spring 2009 19.2% 19.7% ss

Fall 2009 21.6% 16.4% 40%ican

40%ican

Housing Stability Spring 2007 10.2% 6.2%

of Am

ofAm

Spring 2009 12.4% 11.6% 30% 30%

Fall 2009 12.5% 14.0% tage

Debt Spring 2007 17.5% 8.4% Per

Spring 2009 20.6% 18.6% 10%10%F ll 2009Fall 20 7% 17 0%.7% 17.

Retirement Spring 2007 28.5% 12.8% 0%

Spring 2009 30.1% 24.1% 2007

2009

2009

2007

2009

2009

2007

2009

2009

2007

2009

2009

2007

2009

2009

2007

2009

2009

2007

2009

2009

Fall 2009 28.9% 24.6%

ring

ring

Fall

ring

ring

Fall

ring

ring

Fall

ring

ring

Fall

ring

ring

Fall

ring

ring

Fall

ring

ring

Fall

Medical Cost Spring 2007 20.8% 9.7%Spring 2009 19.0% 22.1% Economic Security Employment Housing Stability Debt Retirement Medical Cost Health

Fall 2009 21.6% 23.9% Insurance

Health Insurance Spring 2007 26.3% 7.7% Very Worried Fairly WorriedSpring 2009 17.3% 16.7%

Fall 2009Fa 20 8% 18 1%. .

0%

10%

20%

30%

40%

50%

60%

Spring

2007

Spring

2009

Fall 2009

Spring

2007

Spring

2009

Fall 2009

Spring

2007

Spring

2009

Fall 2009

Spring

2007

Spring

2009

Fall 2009

Spring

2007

Spring

2009

Fall 2009

Spring

2007

Spring

2009

Fall 2009

Spring

2007

Spring

2009

Fall 2009

Economic Security Employment Housing Stability Debt Retirement Medical Cost Health Insurance

Percen

tage

of A

merican

s


Rising Concerns About Economic Insecurity2007 to 2009

Very Worried Fairly Worried

2009

20%enng

2009 20 0%

20%


Percen

tage

ng


ll 2009 20 8% 18 1%

FIGURE 4



Spring 2009 24.3% 28.5%60%60%60%

F ll 2009Fall 19 7% 26 2%19.7% 26.2%


Spring 2009 19.2% 19.7% rry

rry

Fall 2009 21.6% 16.4% 40% Wo

40%Wo

Housing Value Spring 2007 10.2% 6.2%

Repo

rti

Repo

rti

Spring 2009 12.4% 11.6% 30%

30%

Fall 2009 12.5% 14.0%

tage

Debt Spring 2007 17.5% 8.4%

Perc

Spring 2009 20.6% 18.6% 10%10%F ll 2009Fall 20 7% 17 0%.7% 17.


Spring 2009 30.1% 24.1% 2007

2009

2009

2007

2009

2009

2007

2009

2009

2007

2009

2009

2007

2009

2009

2007

2009

2009

2007

2009

2009

Fall 2009 28.9% 24.6%

ring

ring

Fall

ring

ring

Fall

ring

ring

Fall

ring

ring

Fall

ring

ring

Fall

ring

ring

Fall

ring

ring

Fall

Medical Costs Spring 2007 20.8% 9.7%Spring 2009 19.0% 22.1% Economic Security Employment Housing Value Debt Retirement Medical Costs Health



Fall 2009Fa 20 8% 18 1%. .

0%

10%

20%

30%

40%

50%

60%

Spring

200

7

Spring

200

9

Fall 20

09

Spring

200

7

Spring

200

9

Fall 20

09

Spring

200

7

Spring

200

9

Fall 20

09

Spring

200

7

Spring

200

9

Fall 20

09

Spring

200

7

Spring

200

9

Fall 20

09

Spring

200

7

Spring

200

9

Fall 20

09

Spring

200

7

Spring

200

9

Fall 20

09

Economic Security Employment Housing Value Debt Retirement Medical Costs Health Insurance

Percen

tage

Rep

orting

Worry




2009

20%enng

2009 20 0%

20%


Percen

tage

ng


ll 2009 20 8% 18 1%

FIGURE 4



Spring 2009 24.3% 28.5%60%60%60%

F ll 2009Fall 19 7% 26 2%19.7% 26.2%


Spring 2009 19.2% 19.7% rry

rry

Fall 2009 21.6% 16.4% 40% Wo

40%Wo

Housing Value Spring 2007 10.2% 6.2%

Repo

rti

Repo

rti

Spring 2009 12.4% 11.6% 30%

30%

Fall 2009 12.5% 14.0%

tage

Debt Spring 2007 17.5% 8.4%

Perc

Spring 2009 20.6% 18.6% 10%10%F ll 2009Fall 20 7% 17 0%.7% 17.


Spring 2009 30.1% 24.1% 2007

2009

2009

2007

2009

2009

2007

2009

2009

2007

2009

2009

2007

2009

2009

2007

2009

2009

2007

2009

2009

Fall 2009 28.9% 24.6%

ring

ring

Fall

ring

ring

Fall

ring

ring

Fall

ring

ring

Fall

ring

ring

Fall

ring

ring

Fall

ring

ring

Fall

Medical Costs Spring 2007 20.8% 9.7%Spring 2009 19.0% 22.1% Economic Security Employment Housing Value Debt Retirement Medical Costs Health



Fall 2009Fa 20 8% 18 1%. .

0%

10%

20%

30%

40%

50%

60%

Spring

200

7

Spring

200

9

Fall 20

09

Spring

200

7

Spring

200

9

Fall 20

09

Spring

200

7

Spring

200

9

Fall 20

09

Spring

200

7

Spring

200

9

Fall 20

09

Spring

200

7

Spring

200

9

Fall 20

09

Spring

200

7

Spring

200

9

Fall 20

09

Spring

200

7

Spring

200

9

Fall 20

09

Economic Security Employment Housing Value Debt Retirement Medical Costs Health Insurance

Percen

tage

Rep

orting

Worry





Just as the recession amplified general concerns about economic security, it also intensified worries about job loss. Employment concerns rose dramatically as the unemployment rate rose from 4.5 percent in the spring of 2007 to 9.3 percent in the spring of 2009.

Yet worries about a host of other risks to economic security—debt, retirement savings, medical costs, health insurance, and even housing stability—increased more modestly between 2007 and 2009. More than 40 percent of Americans, for example, were already very or fairly worried about “having enough money to retire on” prior to the onset of the recession. The share very or fairly worried about medical costs and health insurance also remained relatively stable (and, in the case of health insurance, the share very worried actually dropped between 2007 and 2009). In these areas, concerns were already notably high before the downturn.

What Sources of Economic Instability Cause the Most Worry?

The SERPI provides a more comprehensive portrait of economic concerns in 2009. Looking across the 15 different sources of potential worry, more than three-quarters of all Americans reported that they were very or fairly worried about at least one of these economic risks. Worries about wealth were the most frequent cause of economic unease, though concerns about medical costs were a close second.

Perhaps the most surprising finding is the prevalence of worries related to the family. While employment concerns have received considerable attention, insecurity related to the family realm is largely absent from the agenda of the media and public officials. Yet Americans worry about family-induced insecurities roughly as much as they worry about insecurities due to

Fig. 5 Scope of Concerns about Economic SecuritySpring 2009

Scope of Concerns About Economic SecurityScope of Concerns About Economic SecuritySpring 2009Spring 2009

80%80%

70%70%

60%

50%

40%

30%

20%

10%

0%

k ) t n st ts

c Ris

d Job

Illne

ss

Divorce

Family

men

nsio

Deb

t

Stability

t Co

verage

Cu lear

Care

omi

/Fin

to

/ lp

tire Pe

Pocke

Co rage Unc

Term

con

Lose

Due

eath He

Re

using

f ose

ove

erage

ng‐

C

0%

10%

20%

30%

40%

50%

60%

70%

80%

Any Econo

mic Risk

Lose/Find Job

Lost W

ork Due

to Illne

ss

Lose Spo

use (Death/Divorce)

Help Family

Retirem

ent

Pension

Deb

t

Hou

sing

Stability

Out of P

ocket C

ost

Lose Coverage

Coverage Cuts

Coverage Unclear

Long

‐Term Care

Employment Family Wealth Health

Scope of Concerns About Economic SecuritySpring 2009


Scope of Concerns About Economic SecurityScope of Concerns About Economic SecuritySpring 2009Spring 2009

80%80%

70%70%

60%

50%

40%

30%

20%

10%

0%

k ) t n st ts

c Ris

d Job

Illne

ss

Divorce

Family

men

nsio

Deb

t

Stability

t Co

verage

Cu lear

Care

omi

/Fin

to

/ lp

tire Pe

Pocke

Co rage Unc

Term

con

Lose

Due

eath He

Re

using

f ose

ove

erage

ng‐

C

0%

10%

20%

30%

40%

50%

60%

70%

80%

Any Econo

mic Risk

Lose/Find Job

Lost W

ork Due

to Illne

ss

Lose Spo

use (Death/Divorce)

Help Family

Retirem

ent

Pension

Deb

t

Hou

sing

Stability

Out of P

ocket C

ost

Lose Coverage

Coverage Cuts

Coverage Unclear

Long

‐Term Care

Employment Family Wealth Health

Scope of Concerns About Economic SecuritySpring 2009


Perc

enta

ge R

epor

ting

Wor

ry


employment: In the midst of a dramatic economic downturn about 39 percent of Americans reported being very or fairly worried about losing their job (or, if out of work, finding employment) —almost exactly the same proportion who reported being very of fairly worried about losing their spouse/partner.

In light of the recent debate over health care reform, it is not surprising that Americans were deeply worried about unstable insurance enrollment, rising premiums, and gaps in coverage. Yet during that debate, there was remarkably little public discussion of the financial effects of time away from work due to illness. Thus it is striking that about a third of the public was very or fairly worried about losing several months of earnings due to an extended illness. Worries about forgone income due to illness are just slightly less common than concerns about losing one’s job and just as prevalent as concerns about losing one’s health insurance.9

Are Americans’ Worries Reasonable?

Comparing the prevalence of worries and the frequency of shocks, Americans appear to accurately assess the likelihood of some common risks. In the domain of employment, for example, the frequency of shocks and the proportion of the

public worried about these events match relatively closely. However, this close match does not hold for all shocks. About as many Americans worry about losing their partner or spouse as about substantial out-of-pocket medical expenses, and many more do so than are worried about needing to assist family members in financial need. Yet divorce or death of a spouse or partner is actually far less common than either of these latter sources of nondiscretionary spending. Similarly, worries about losing insurance coverage are almost as common as are worries about high out-of-pocket medical spending. Yet losses of insurance coverage

actually occur only about half as often as do high out-of-pocket medical costs.

These divergences may reflect the public’s misperception of certain risks. People tend to inflate the chance of unpredictable events that induce great fear or anxiety.10 Yet the heightened worry associated with certain risks may also reflect the expected severity of the resulting economic losses. An intuitive

Worries about forgone income due to illness are just slightly less common than concerns about losing one’s job and just as prevalent as concerns about losing one’s health insurance.


measure of expected severity is the estimated length of time it takes for a household’s financial circumstances to return to the level enjoyed before the event occurred. When asked to estimate how long recovery would typically take, Americans predicted that family disruptions (death of spouse or divorce) would have the longest-lasting consequences, with nearly eight in ten (78 percent) saying it would take more than six months to recover economically. The consequences of illness (either out-of-pocket spending or time lost from work) came close in terms of their anticipated duration. Unemployment and investment losses, by contrast, were seen as having the most transitory effects.11

Substantial Anxiety about Economic Insecurity

While the previous measures concern people’s worries about economic insecurity, worries may not affect the quality of individuals’ lives substantially. Thinking about negative events is rarely pleasant, but worries are not always associated with intense negative feelings.12 Extreme anxiety, by contrast, can induce physiological stress responses. If anxiety is chronic, these stressors can lead to declines in both physical and mental health.13

According to the spring and fall 2009 waves of the SERPI, roughly 17 to 19 percent of the public reported themselves to be very or extremely anxious on each wave of the survey when thinking about economic insecurity. A total of 28 percent of all Americans were either very or extremely anxious about their economic prospects on at least one of the two surveys; 9 percent had persisting high anxiety across both surveys.

Not surprisingly, those who were most worried about specific forms of economic uncertainty proved to be most anxious as well: Of those who reported themselves to be very worried about unstable employment, medical costs, fluctuating wealth, or family-related shocks, roughly 40 percent were very or extremely anxious about economic insecurity. By contrast, among those who reported no intense worries about these specific economic risks, less than 2 percent felt intense anxiety about economic insecurity.

A similar pattern is evident for those who experienced shocks, as is evident in Figure 6. Among the households that had been free of any of the eleven shocks measured in the SERPI, only about 5 percent reported themselves to be very or extremely anxious. Among households that had experienced one or more shocks, the frequency of this intense anxiety jumped to between 15 and 25 percent depending on the domain of risk. And for all four domains, when households experienced shocks that persisted across the two time periods in the survey, they reported even more extreme anxiety. After experiencing persisting shocks unrelated to wealth, about a third of all Americans reported themselves to be very or extremely anxious about economic uncertainty.

A total of 28 percent of all Americans were either very or extremely anxious about their economic prospects on at least one of the two surveys.


The Impact of Insecurity on Economic Well-BeingThe frequency of anxiety and the degree to which it occurs in tandem with economic shocks strongly suggest that Americans fear the potentially serious economic dislocations these shocks create. The SERPI allows us to examine this implication more directly. A series of questions prompted respondents to indicate if they or members of their household had unmet needs for food, housing stability, or medical care—for example if they had gone hungry, lost their home, or failed to see a doctor (see the Technical Appendix for the specific questions asked). Because these unmet needs are measured on each wave of the survey, respondents could report up to six unmet basic needs over the 18 months covered by the SERPI.

Economic Shocks Are Associated with Unmet Needs

Some households that have not experienced any economic shocks nonetheless report unmet needs, because their earnings are modest and savings are limited. Indeed, roughly one in five households that have experienced no economic instability (no shocks in the past 18 months) still report having at least one unmet basic need.14 But households that have experienced economic shocks—especially persisting shocks—report much higher levels of unmet need. As Figure 7 shows, this is particularly true of employment and medical shocks:

Fig. 6 Economic Shocks and Anxiety about Economic Instability2008–2009

About Economic Security

30%t

15%

30%

Sh Sh

Recen

Pas

ting

Recen

Pas

ting

Recen

Pas

ting

Recen

Pas

ting

Very Anxious

l k

In Past Six Months, but not in Past 7‐18 months = In Past 7‐18 Months, Not Reoccuring

FIGURE 6Reported Anxiety

Very Anxious Extremely Anxious

No Shocks At All 4.0% 0.8%Wealth Shocks Only 8.4% 2.7% 35%35%

E lEmploym t R t Shen Recent kShock 15 4% 7 0%15.4% 7.0%Past Shock 15.5% 10.8% 25%

Persisting Shocks 19.1% 13.4% 20%

Health Recent Shock 11.5% 8.3%Past Shock 9.0% 7.4% 10%

Persisting Shocks 17.0% 14.2% 5% Extremely Anxious

Wealth Recent Shock 8.4% 8.2%0%

Past Shock 13.2% 4.9%

At A

ll

Only

ock

ock

ocks

ock

ock

ocks

ock

ock

ocks

ock

ock

ocks

Persisting Shocks 12.3% 8.6%

ocks

ocks

t Sh

t Sh

Sh

t Sh

t Sh

Sh

t Sh

t Sh

Sh

t Sh

t Sh

Sh

F ilFami y R t ShRecent Shock 10 0% 8 1% o lth

ersis

ersis

ersis

ersis

10.0% 8.1%Past Shock

N

12.6% 12.5%Persisting Shocks 15.5% 14.0%

Recent=PastPersisting+ In Past 6 Months AND in Past 7‐18 Months

0%

5%

10%

15%

20%

25%

30%

35%

No Shocks At A

ll

Wealth

Sho

cks Only

Recent Sho

ck

Past Sho

ck

Persistin

g Shocks

Recent Sho

ck

Past Sho

ck

Persistin

g Shocks

Recent Sho

ck

Past Sho

ck

Persistin

g Shocks

Recent Sho

ck

Past Sho

ck

Persistin

g Shocks

Employment Health Wealth Family

Extremely Anxious

Very Anxious

About Economic Security

30%t

15%

30%

Sh Sh

Recen

Pas

ting

Recen

Pas

ting

Recen

Pas

ting

Recen

Pas

ting

Very Anxious

l k

In Past Six Months, but not in Past 7‐18 months = In Past 7‐18 Months, Not Reoccuring

FIGURE 6Reported Anxiety

Very Anxious Extremely Anxious

No Shocks At All 4.0% 0.8%Wealth Shocks Only 8.4% 2.7% 35%35%

E lEmploym t R t Shen Recent kShock 15 4% 7 0%15.4% 7.0%Past Shock 15.5% 10.8% 25%

Persisting Shocks 19.1% 13.4% 20%

Health Recent Shock 11.5% 8.3%Past Shock 9.0% 7.4% 10%

Persisting Shocks 17.0% 14.2% 5% Extremely Anxious

Wealth Recent Shock 8.4% 8.2%0%

Past Shock 13.2% 4.9%

At A

ll

Only

ock

ock

ocks

ock

ock

ocks

ock

ock

ocks

ock

ock

ocks

Persisting Shocks 12.3% 8.6%

ocks

ocks

t Sh

t Sh

Sh

t Sh

t Sh

Sh

t Sh

t Sh

Sh

t Sh

t Sh

Sh

F ilFami y R t ShRecent Shock 10 0% 8 1% o lth

ersis

ersis

ersis

ersis

10.0% 8.1%Past Shock

N

12.6% 12.5%Persisting Shocks 15.5% 14.0%

Recent=PastPersisting+ In Past 6 Months AND in Past 7‐18 Months

0%

5%

10%

15%

20%

25%

30%

35%

No Shocks At A

ll

Wealth

Sho

cks Only

Recent Sho

ck

Past Sho

ck

Persistin

g Shocks

Recent Sho

ck

Past Sho

ck

Persistin

g Shocks

Recent Sho

ck

Past Sho

ck

Persistin

g Shocks

Recent Sho

ck

Past Sho

ck

Persistin

g Shocks

Employment Health Wealth Family

Extremely Anxious

Very Anxious

Recent = In Past Six Months, but Not in Past 7-18 MonthsPast = In Past 7-18 Months, Not ReoccuringPersisiting = In Past 6 Months AND in Past 7-18 Months


70%70%70%

60%60%60%

50%

Nee

ds

50%

Nee

dset

et

40%Unm 40%Unm

ng

ng

30%30%

Repo

rti

Repo

rti

ent

ent

20%

Perc 20%

Perc

10%10%

0%

At A

ll

ock

ocks ock

ocks ock

ocks ock

ocks

t Sch

Sh

st Sh

Sh Sh

Sh Sh

Sh

Shocks

men

men

t

Co

Cost alth

alth

Family

Family

No oy oy alth

alth We

We

mpl

mpl He

He Any

ting Any

ting

E E Any

ting rsis rsis

Any

istin

g

is Pe Pe

rs Pers

Pe

0%

10%

20%

30%

40%

50%

60%

70%

No Shocks At A

ll

Any

Employmen

t Schock

Persistin

g Em

ploymen

t Sho

cks

Any

Health

Cost Shock

Persistin

g Health

Cost S

hocks

Any

Wealth

Sho

ck

Persistin

g Wealth

Sho

cks

Any

Fam

ily Sho

ck

Persistin

g Family Sho

cks

Percen

t Rep

orting

Unm

et Nee

ds

One Unmet Need Multiple Unmet Needs

Fig. 7 Economic Shocks and Unmet Basic Needs2008–2009

‐

‐

Economic Shocks and Unmet Basic NeedsEconomic Shocks and Unmet Basic NeedsEconomic Shocks and Unmet Basic Needs2008‐20092008 20092008 2009

70.0%70.0%70.0%

60 0%60.0%60.0%

50.0%

Nee

ds

50.0%

Nee

dset

et

40.0%Unm 40.0%Unm

ng

ng

30.0%30.0%

Repo

rti

Repo

rti

ent

ent

20.0%

Perc 20.0%

Perc

10.0%10.0%

0.0%

At A

ll

ock

ocks ock

ocks ock

ocks ock

ocks

t Sch

Sh

st Sh

Sh Sh

Sh Sh

Sh

Shocks

men

men

t

Co

Cost alth

alth

Family

Family

No oy oy alth

alth We

We

mpl

mpl He

He Any

ting Any

ting

E E Any

ting rsis rsis

Any

istin

g

is Pe Pe

rs Pers

Pe

0.0%

10.0%

20.0%

30.0%

40.0%

50.0%

60.0%

70.0%No Shocks At A

ll

Any

Employmen

t Schock

Persistin

g Em

ploymen

t Sho

cks

Any

Health

Cost Shock

Persistin

g Health

Cost S

hocks

Any

Wealth

Sho

ck

Persistin

g Wealth

Sho

cks

Any

Fam

ily Sho

ck

Persistin

g Family Sho

cks

Percen

t Rep

orting

Unm

et Nee

ds

Economic Shocks and Unmet Basic Needs2008‐2009


Shocks to wealth and family are associated with a doubling of households reporting any unmet need and a fivefold increase in households reporting multiple unmet needs, whereas shocks to employment and medical spending are associated with a tripling of households reporting any unmet needs and a sevenfold increase in the proportion of households reporting multiple unmet needs. For the most part, then, basic needs seem to be far more threatened by disruptions of employment or by unexpected medical spending than by emergencies in the family or by unanticipated drops in wealth.15

Are the Effects of Insecurity Distinct from Economic Disadvantage?

Households that report unstable economic circumstances are far more likely to experience both heightened anxiety and unmet needs. But families that face unstable economic circumstances might have other attributes that make them vulnerable to these bad outcomes as well—most notably, lower incomes. Less affluent households are more likely to have unmet needs and also to have less stable employment and less comprehensive health insurance. Thus, it is crucial to examine whether economic insecurity is independently associated with hardship, even after household income is taken into account.

To do that, it is helpful to group respondents into household income quartiles—four equally sized groups ranked from poorest to richest. Comparing within

70%70%70%

60%60%60%

50%

Nee

ds

50%

Nee

dset

et

40%Unm 40%Unm

ng

ng

30%30%

Repo

rti

Repo

rti

ent

ent

20%

Perc 20%

Perc

10%10%

0%

At A

ll

ock

ocks ock

ocks ock

ocks ock

ocks

t Sch

Sh

st Sh

Sh Sh

Sh Sh

Sh

Shocks

men

men

t

Co

Cost alth

alth

Family

Family

No oy oy alth

alth We

We

mpl

mpl He

He Any

ting Any

ting

E E Any

ting rsis rsis

Any

istin

g

is Pe Pe

rs Pers

Pe

0%

10%

20%

30%

40%

50%

60%

70%No Shocks At A

ll

Any

Employmen

t Schock

Persistin

g Em

ploymen

t Sho

cks

Any

Health

Cost Shock

Persistin

g Health

Cost S

hocks

Any

Wealth

Sho

ck

Persistin

g Wealth

Sho

cks

Any

Fam

ily Sho

ck

Persistin

g Family Sho

cks

Percen

t Rep

orting

Unm

et Nee

ds


Perc

enta

ge R

epor

ting

Unm

et N

eeds


each quartile, it becomes clear that even among the top 25 percent of Americans—and, indeed, within every income quartile—economic instability is associated with an increased level of unmet basic needs (Figure 8). A quarter of families in the lowest income quartile (those with annual incomes of less than $35,000) reported some unmet needs even if they had experienced no economic instability in the previous 18 months. But among those that had faced economic instability, unmet needs were far more common. Among those who had experienced persisting employment shocks, for example, almost 90 percent reported some unmet basic needs, and 56 percent reported multiple unmet needs.

Strikingly, the same relationship holds among families in the third quartile of household income (annual incomes between $60,000 and $100,000). Among those that experienced no economic instability, unmet basic needs were about half as common as in the lowest income quartile, reported by about 12 percent of all households. For those who encountered persisting economic shocks, the level of unmet needs was far more common—strikingly so, considering that these households are relatively well-off. About half of households in this quartile that had experienced persisting shocks to employment or medical costs reported some unmet need; almost 40 percent reported multiple unmet needs. Even among solidly middle-class families, in short, economic instability is associated with much higher levels of unmet basic needs.

Fig. 8 Unmet Needs Associated with Economic Instability,for Selected Income Strata

Unmet Needs Associated with Economic Instability,Unmet Needs Associated with Economic Instability,Unmet Needs Associated with Economic Instability,For Selected Income StrataFor Selected Income StrataFor Selected Income Strata

100%100%

90%90%

80%80%

70%70%

60%60%

50%50%

40%40%

30%30%

20%

10%

0%

All

All

All

All

At ocks

ocks

ocks

ocks

At ocks

ocks

ocks

ocks

At ocks

ocks

ocks

ocks

At ocks

ocks

ocks

ocks

ocks

t Sh

s Sh Sh

Sh

ocks

t Sh

s Sh Sh

Sh

ocks

t Sh

s Sh Sh

Sh

ocks

t Sh

s Sh Sh

Sh

Sh men

Cost alth

Family

Sh men

Cost alth

Family

Sh men

Cost alth

Family

Sh men

Cost alth

Family

No loy

We

No loy

We

No loy

We

No loy

We

mp

Med

ical

ting

isting

mp

Med

ical

ting

isting

mp

Med

ical

ting

isting

mp

Med

ical

ting

isting

g E

rsis

Pers g E

rsis

Pers g E

rsis

Pers g E

rsis

Pers

sistin

sting

Pe

sistin

sting

Pe

sistin

sting

Pe

sistin

sting

Pe

0%

10%

20%

30%

40%

50%

60%

70%

80%

90%

100%

No Shocks At All

Persisting

Employmen

t Shocks

Persisting

Med

ical Costs Sho

cks

Persisting

Wealth Shocks

Persisting

Fam

ily Sho

cks

No Shocks At All

Persisting

Employmen

t Shocks

Persisting

Med

ical Costs Sho

cks

Persisting

Wealth Shocks

Persisting

Fam

ily Sho

cks

No Shocks At All

Persisting

Employmen

t Shocks

Persisting

Med

ical Costs Sho

cks

Persisting

Wealth Shocks

Persisting

Fam

ily Sho

cks

No Shocks At All

Persisting

Employmen

t Shocks

Persisting

Med

ical Costs Sho

cks

Persisting

Wealth Shocks

Persisting

Fam

ily Sho

cks

Lowest Income Quartile Second Quartile Third Income Quartile Highest Income Quartile

Unmet Needs Associated with Economic Instability,For Selected Income Strata

Multiple Unmet Needs One Unmet Need

Unmet Needs Associated with Economic Instability,Unmet Needs Associated with Economic Instability,Unmet Needs Associated with Economic Instability,For Selected Income StrataFor Selected Income StrataFor Selected Income Strata

100%100%

90%90%

80%80%

70%70%

60%60%

50%50%

40%40%

30%30%

20%

10%

0%

All

nt s

All

nt s

All

nt s nt s

At e

Cost alth

Family

At e

Cost alth

Family

At e

Cost alth

Family

All* e

Cost alth

Family

ocks

ploym

We

ocks

ploym

We

ocks

ploym

We

At

ploym

We

Sh Em Med

ical

ocks

ocks

Sh Em Med

ical

ocks

ocks

Sh Em Med

ical

ocks

ocks

ocks

Em Med

ical

ocks

ocks

No

ocks Sh Sh

No

ocks Sh Sh

No

ocks Sh Sh

No Sh

ocks Sh Sh

Sh ocks

ting

istin

g

Sh ocks

ting

istin

g

Sh ocks

ting

istin

g

Sh ocks

ting

istin

g

g Sh

rsis

Pers g Sh

rsis

Pers g Sh

rsis

Pers g Sh

rsis

Pers

sistin

sting

Pe

sistin

sting

Pe

sistin

sting

Pe

sistin

sting

Pe

0%

10%

20%

30%

40%

50%

60%

70%

80%

90%

100%

No Shocks At A

ll

Persistin

g Shocks Employmen

t

Persistin

g Shocks M

edical Costs

Persistin

g Shocks W

ealth

Persistin

g Shocks Fam

ily

No Shocks At A

ll

Persistin

g Shocks Employmen

t

Persistin

g Shocks M

edical Costs

Persistin

g Shocks W

ealth

Persistin

g Shocks Fam

ily

No Shocks At A

ll

Persistin

g Shocks Employmen

t

Persistin

g Shocks M

edical Costs

Persistin

g Shocks W

ealth

Persistin

g Shocks Fam

ily

No Shocks At A

ll*

Persistin

g Shocks Employmen

t

Persistin

g Shocks M

edical Costs

Persistin

g Shocks W

ealth

Persistin

g Shocks Fam

ily

Lowest Income Quartile Second Quartile Third Income Quartile Highest Income Quartile

Unmet Needs Associated with Economic Instability,For Selected Income Strata

Multiple Unmet Needs One Unmet Need

* No respondents in this income strata reported unmet needs in the absence of economic shocks

Perc

enta

ge R

epor

ting

Unm

et N

eeds

*


Three to six months

EXHIBIT 14

Time Without Income Before Hardship Less than a week 9.0%One to two weeks 10.4%Three to four weeks 9.2%O t t th 18 7%One to two months 18.7%Three to four months 14.5% Less than two weeks

Five to six months 8.9% Three to four weeks

Six months or more 29.3% One to two months

Time Without Income Before Hardship Six months or more

Less than two weeks 19.4%Three to four weeks 9.2%One to two months 18.7%Three to six months 23.4%Six months or more 29.3%

Less than two weeks

Three to four weeks

One to two months

Three to six months

Six months or more

Incomplete Buffers against the Impact of Economic Shocks

The impact of some economic shocks, such as high health care costs and partial disability from work, can be lessened by purchasing insurance. Others, such as unemployment and a complete inability to work, are at least partially insured by public policies. But for many risks (including loss of a spouse, declining value of investments, urgent financial assistance to one’s extended family, and lost earnings apart from formal unemployment or disability), insurance protection is generally either unavailable or beyond the means of the average citizen.

Households can guard against uninsured or partially insured risks only by saving, borrowing from family and friends, or drawing on the equity of their assets, such as their home or retirement account. The SERPI included several questions about these informal risk buffers.16

At the core of these measures is the household’s self-assessed capacity to get by without hardship should their income be interrupted for an extended length of time. As Figure 9 shows, just over 29 percent of Americans reported that their household could go six months or longer without experiencing hardship if their earnings were to stop tomorrow. Yet nearly half of households could go no longer than two months without hardship setting in; one in five could last no more than two weeks. If the capacity to go at least three months without income is treated as a threshold of adequate reserves (recall that almost a quarter of all household experienced a drop of 25 percent or more in their annual income in the 18 months preceding the fall of 2009), then more than half the American population appears to lack adequate buffers.

Three additional measures of buffers in the SERPI are related to the capacity to borrow: in financial markets (i.e., taking out a loan), drawing against equity (i.e., home equity loans), and through more informal mechanisms (i.e., from family and friends). Combined with our measure of financial reserves, these offer a multifaceted assessment of the ability of households to deal with unexpected economic shocks. In the aggregate, they do not make Americans appear much

Fig. 9 How Long Can Household Go without IncomeBefore Hardship Sets In?

29.3%19.4%

18.7%23.4%

9.2%

Three to six months

EXHIBIT 14

Time Without Income Before Hardship Less than a week 9.0%One to two weeks 10.4%Three to four weeks 9.2%O t t th 18 7%One to two months 18.7%Three to four months 14.5% Less than two weeks

Five to six months 8.9% Three to four weeks

Six months or more 29.3% One to two months

Time Without Income Before Hardship Six months or more

Less than two weeks 19.4%Three to four weeks 9.2%One to two months 18.7%Three to six months 23.4%Six months or more 29.3%

Less than two weeks

Three to four weeks

One to two months

Three to six months

Six months or more


more secure. To be sure, about a third of respondents anticipated that they could borrow $5,000 or more from family or friends if facing some urgent financial need. However, 44 percent expected not to be able to borrow at all, and one in five considered themselves to be so deep in debt that they doubted that they could ever pay it off.

Why Household Buffers Are Weak

Why does such a large share of households have limited savings, high levels of debt, and limited capacity to borrow in times of need from friends and family? Although some Americans lack foresight or prudence, two other causes also come into play: prior economic shocks and a basic lack of resources. Households’ buffers are frequently depleted by shocks to earnings or nondiscretionary spending. Comparing respondents who experienced a drop in income or loss of wealth between the first and second waves of the survey (and had not experienced

a prior shock), those that did not experience an economic shock between the two surveys saw a modest improvement in their risk buffers over the intervening six months. By contrast, those who did experience a shock reported degradation in all four measures of buffer resilience.

Lack of buffers is also strongly associated with economic disadvantage, suggesting that resource constraints are one important reason why some families are ill-prepared against economic risks.

Understanding why such a large share of households have limited economic buffers thus requires examining more closely who is most likely to experience economic shocks without adequate protections—the question taken up in the next section.

The Uneven Experience of Economic InsecurityThe SERPI documents that insecurity and instability are commonplace in Americans’ economic lives, with consequences that are frequently severe for even middle-class families. Nonetheless, the experience of insecurity varies substantially across American society—across income and educational groups, family types, and other important demographic characteristics. This is not

Nearly half of households could go no longer than two months without hardship setting in. One in five could last no more than two weeks.


simply because the chance of major economic shocks (and especially clusters of major shocks) is different for these various groups. The greatest disparities in the experience of economic instability emerge when we consider the ways in which households differ in their capacity to buffer the shocks that occur.

The Surprising Commonality of Economic Instability

During the recession, differences in exposure to economic disruptions were remarkably modest, as Figure 10 suggests. Even putting aside wealth losses—which are, not surprisingly, more frequent among more affluent Americans—middle-income and well-educated Americans often experienced major economic shocks. Employment-related instability was less frequent for more advantaged Americans, but job loss was sufficiently common that it presented a realistic and not uncommon worry: 16 percent of college graduates and 19

Fig. 10 Shocks and Worries by Sociodemographic Groups

Education

Race/Ethnicity

Quartile Hihest

Quartile Hihest

70% 80%70% 80%70% 80%70% 80%

50%

60%Education

60%

70%Income Strata50%

60%Education

60%

70%Income Strata50%

60%Education

60%

70%Income Strata50%

60%Education

60%

70%Income Strata

40%50%

40%50%

40%50%

40%50%

30%40%

30%40%

30%40%

30%40%

20%20%20%20%

10% 30%10% 30%10% 30%10% 30%

0% "Very Worried" about 20%0% "Very Worried" about 20%0% "Very Worried" about 20%

t

sts

alth t

sts

alth t

sts

alth at Least One Risk 10%

"Very Worried" about t

sts

alth t

sts

alth t

sts


"Very Worried" aboutt

sts

alth t

sts

alth t

sts


"Very Worried" about

ymen

Co

We

Family

ymen

Co

We

Family

ymen

Co

We

Family

0%at Least One Risk

ymen

Co

We

Family

ymen

Co

We

Family

ymen

Co

We

Family

0%at Least One Risk

ymen

Co

We

Family

ymen

Co

We

Family

ymen

Co

We

Family

0%at Least One Risk

plo

plo

plo

Persisting Shocks t s t s t s t splo

plo

plo


plo

plo

Persisting Shocks t s t s t s t s

Em Med

ical

Em Med

ical

Em Med

ical

men

Cost alth

Family

men

Cost alth

Family

men

Cost alth

Family

men

Cost alth

Family Persisting Shocks

Em Med

ical

Em Med

ical

Em Med

ical

men

Cost alth

Family

men

Cost alth

Family

men

Cost alth

Family

men

Cost alth


Em Med

ical

Em Med

ical

Em Med

ical

men

Cost alth

Family

men

Cost alth

Family

men

Cost alth

Family

men

Cost alth


loy We

loy We

loy We

loy We

loy We

loy We

loy We

loy We

loy We

loy We

loy We

loy We

High School Some College College Graduate mp

Med

ical

mp

Med

ical

mp

Med

ical

mp

Med

ical


Med

ical

mp

Med

ical

mp

Med

ical

mp

Med

ical


Med

ical

mp

Med

ical

mp

Med

ical

mp

Med

ical

E E E EE E E EE E E E

Lowest Quartile Second Quartile Third Quartile Hihest QuartileLowest Quartile Second Quartile Third QuartileLowest Quartile Second Quartile Third Quartile

60% 60%60% 60%

50%50%50%Child DependentsEducation

50%Child DependentsEducation

40%40%

40%40%

30%30%30% 20%30% 20%

20% 10%20% 10%

10% "Very Worried" about 0% "Very Worried" about 10% "Very Worried" about 0% "Very Worried" aboutat Least One Risk

men

t

sts

alth

Family

men

t

sts

alth

Family at Least One Riskat Least One Risk ent

sts

alth

Family ent

sts

alth

Family at Least One Risk

0%

oy al Co

We

oy al Co

We0%

oym Co

We

oym Co

We

ent

osts

alth

mily ent

osts

alth

mily ent

osts

alth

mily Persisting Shocks

mpl dic

mpl dic Persisting Shocks

men

t

sts

alth

Family

men

t

sts

alth

Family

men

t

sts

alth

Family Persisting Shocks pl

Med

ical

pl

Med

ical

Persisting Shocks

oy

Co

We

oy

Co

We

oy

Co

We Em Em

mpl

Med

ical

mpl

Med

ical

mpl

Med

ical

E E E Child Dependents No Children

White African‐American Latino

0%

10%

20%

30%

40%

50%

60%

70%

Employmen

t

Med

ical Costs

Wealth

Family

Employmen

t

Med

ical Costs

Wealth

Family

Employmen

t

Med

ical Costs

Wealth

Family

High School Some College College Graduate

"Very Worried" about at Least One Risk

Persisting Shocks

Education

0%

10%

20%

30%

40%

50%

60%

Employmen

t

Med

ical Costs

Wealth

Family

Employmen

t

Med

ical Costs

Wealth

Family

Employmen

t

Med

ical Costs

Wealth

Family



Persisting Shocks

0%

10%

20%

30%

40%

50%

60%

70%

80%

Employmen

t

Med

ical Costs

Wealth

Family

Employmen

t

Med

ical Costs

Wealth

Family

Employmen

t

Med

ical Costs

Wealth

Family

Employmen

t

Med

ical Costs

Wealth

Family

Lowest Quartile Second Quartile Third Quartile Hihest Quartile


Persisting Shocks

Income Strata

0%

10%

20%

30%

40%

50%

60%

Employmen

t

Med

ical Costs

Wealth

Family

Employmen

t

Med

ical Costs

Wealth

Family

Child Dependents No Children


Persisting Shocks

Child DependentsEducation

Quartile Hihest

Quartile Hihest

70% 80%70% 80%70% 80%70% 80%

50%

60%Education

60%

70%Income Strata50%

60%Education

60%

70%Income Strata50%

60%Education

60%

70%Income Strata50%

60%Education

60%

70%Income Strata

40%50%

40%50%

40%50%

40%50%

30%40%

30%40%

30%40%

30%40%

20%20%20%20%

10% 30%10% 30%10% 30%10% 30%


t

sts

alth t

sts

alth t

sts



sts

alth t

sts

alth t

sts



sts

alth t

sts

alth t

sts



ymen

Co

We

Family

ymen

Co

We

Family

ymen

Co

We

Family

0%at Least One Risk

ymen

Co

We

Family

ymen

Co

We

Family

ymen

Co

We

Family

0%at Least One Risk

ymen

Co

We

Family

ymen

Co

We

Family

ymen

Co

We

Family

0%at Least One Risk

plo

plo

plo


plo

plo


plo

plo


Em Med

ical

Em Med

ical

Em Med

ical

men

Cost alth

Family

men

Cost alth

Family

men

Cost alth

Family

men

Cost alth


Em Med

ical

Em Med

ical

Em Med

ical

men

Cost alth

Family

men

Cost alth

Family

men

Cost alth

Family

men

Cost alth


Em Med

ical

Em Med

ical

Em Med

ical

men

Cost alth

Family

men

Cost alth

Family

men

Cost alth

Family

men

Cost alth


loy We

loy We

loy We

loy We

loy We

loy We

loy We

loy We

loy We

loy We

loy We

loy We


Med

ical

mp

Med

ical

mp

Med

ical

mp

Med

ical


Med

ical

mp

Med

ical

mp

Med

ical

mp

Med

ical


Med

ical

mp

Med

ical

mp

Med

ical

mp

Med

ical



60% 60%60% 60%



40%40%

40%40%

30%30%30% 20%30% 20%

20% 10%20% 10%


men

t

sts

alth

Family

men

t

sts

alth


sts

alth

Family ent

sts

alth


0%

oy al Co

We

oy al Co

We0%

oym Co

We

oym Co

We

ent

osts

alth

mily ent

osts

alth

mily ent

osts

alth


mpl dic


men

t

sts

alth

Family

men

t

sts

alth

Family

men

t

sts

alth


Med

ical

pl

Med

ical

Persisting Shocks

oy

Co

We

oy

Co

We

oy

Co

We Em Em

mpl

Med

ical

mpl

Med

ical

mpl

Med

ical



0%

10%

20%

30%

40%

50%

60%

70%

Employmen

t

Med

ical Costs

Wealth

Family

Employmen

t

Med

ical Costs

Wealth

Family

Employmen

t

Med

ical Costs

Wealth

Family



Persisting Shocks

Education

0%

10%

20%

30%

40%

50%

60%

Employmen

t

Med

ical Costs

Wealth

Family

Employmen

t

Med

ical Costs

Wealth

Family

Employmen

t

Med

ical Costs

Wealth

Family



Persisting Shocks

0%

10%

20%

30%

40%

50%

60%

70%

80%

Employmen

t

Med

ical Costs

Wealth

Family

Employmen

t

Med

ical Costs

Wealth

Family

Employmen

t

Med

ical Costs

Wealth

Family

Employmen

t

Med

ical Costs

Wealth

Family



Persisting Shocks

Income Strata

0%

10%

20%

30%

40%

50%

60%

Employmen

t

Med

ical Costs

Wealth

Family

Employmen

t

Med

ical Costs

Wealth

Family



Persisting Shocks


Perc

enta

ge o

f Gro

upPe

rcen

tage

of G

roup


Child Dependents

Quartile Hihest

Quartile Hihest

70% 80%70% 80%70% 80%70% 80%

50%

60%Education

60%

70%Income Strata50%

60%Education

60%

70%Income Strata50%

60%Education

60%

70%Income Strata50%

60%Education

60%

70%Income Strata

40%50%

40%50%

40%50%

40%50%

30%40%

30%40%

30%40%

30%40%

20%20%20%20%

10% 30%10% 30%10% 30%10% 30%


t

sts

alth t

sts

alth t

sts



sts

alth t

sts

alth t

sts



sts

alth t

sts

alth t

sts



ymen

Co

We

Family

ymen

Co

We

Family

ymen

Co

We

Family

0%at Least One Risk

ymen

Co

We

Family

ymen

Co

We

Family

ymen

Co

We

Family

0%at Least One Risk

ymen

Co

We

Family

ymen

Co

We

Family

ymen

Co

We

Family

0%at Least One Risk

plo

plo

plo


plo

plo


plo

plo


Em Med

ical

Em Med

ical

Em Med

ical

men

Cost alth

Family

men

Cost alth

Family

men

Cost alth

Family

men

Cost alth


Em Med

ical

Em Med

ical

Em Med

ical

men

Cost alth

Family

men

Cost alth

Family

men

Cost alth

Family

men

Cost alth


Em Med

ical

Em Med

ical

Em Med

ical

men

Cost alth

Family

men

Cost alth

Family

men

Cost alth

Family

men

Cost alth


loy We

loy We

loy We

loy We

loy We

loy We

loy We

loy We

loy We

loy We

loy We

loy We


Med

ical

mp

Med

ical

mp

Med

ical

mp

Med

ical


Med

ical

mp

Med

ical

mp

Med

ical

mp

Med

ical


Med

ical

mp

Med

ical

mp

Med

ical

mp

Med

ical



60% 60%60% 60%



40%40%

40%40%

30%30%30% 20%30% 20%

20% 10%20% 10%


men

t

sts

alth

Family

men

t

sts

alth


sts

alth

Family ent

sts

alth


0%

oy al Co

We

oy al Co

We0%

oym Co

We

oym Co

We

ent

osts

alth

mily ent

osts

alth

mily ent

osts

alth


mpl dic


men

t

sts

alth

Family

men

t

sts

alth

Family

men

t

sts

alth


Med

ical

pl

Med

ical

Persisting Shocks

oy

Co

We

oy

Co

We

oy

Co

We Em Em

mpl

Med

ical

mpl

Med

ical

mpl

Med

ical



0%

10%

20%

30%

40%

50%

60%

70%

Employmen

t

Med

ical Costs

Wealth

Family

Employmen

t

Med

ical Costs

Wealth

Family

Employmen

t

Med

ical Costs

Wealth

Family



Persisting Shocks

Education

0%

10%

20%

30%

40%

50%

60%

Employmen

t

Med

ical Costs

Wealth

Family

Employmen

t

Med

ical Costs

Wealth

Family

Employmen

t

Med

ical Costs

Wealth

Family



Persisting Shocks

0%

10%

20%

30%

40%

50%

60%

70%

80%

Employmen

t

Med

ical Costs

Wealth

Family

Employmen

t

Med

ical Costs

Wealth

Family

Employmen

t

Med

ical Costs

Wealth

Family

Employmen

t

Med

ical Costs

Wealth

Family



Persisting Shocks

Income Strata

0%

10%

20%

30%

40%

50%

60%

Employmen

t

Med

ical Costs

Wealth

Family

Employmen

t

Med

ical Costs

Wealth

Family



Persisting Shocks


Perc

enta

ge o

f Gro

upPe

rcen

tage

of G

roup

70%70%

60%60%Adults in HouseholdAdults in Household

50%50%

40%40%

30%

20%

"Very Worried" about 10%at Least One Risk

0%

t

sts

alth t

sts

alth Persisting Shocks

men

Co Family

men

Co Family

ploy

We

ploy

We

Em Med

ical

Em Med

ical

Coupled Single

0%

10%

20%

30%

40%

50%

60%

70%

Employmen

t

Med

ical Costs

Wealth

Family

Employmen

t

Med

ical Costs

Wealth

Family

Coupled Single


Persisting Shocks

Adults in Household

Adults in Household

Perc

enta

ge o

f Gro

up

Couple

Perc

enta

ge o

f Gro

up

Income Strata

Quartile Hihest

Quartile Hihest

70% 80%70% 80%70% 80%70% 80%

50%

60%Education

60%

70%Income Strata50%

60%Education

60%

70%Income Strata50%

60%Education

60%

70%Income Strata50%

60%Education

60%

70%Income Strata

40%50%

40%50%

40%50%

40%50%

30%40%

30%40%

30%40%

30%40%

20%20%20%20%

10% 30%10% 30%10% 30%10% 30%


t

sts

alth t

sts

alth t

sts



sts

alth t

sts

alth t

sts



sts

alth t

sts

alth t

sts



ymen

Co

We

Family

ymen

Co

We

Family

ymen

Co

We

Family

0%at Least One Risk

ymen

Co

We

Family

ymen

Co

We

Family

ymen

Co

We

Family

0%at Least One Risk

ymen

Co

We

Family

ymen

Co

We

Family

ymen

Co

We

Family

0%at Least One Risk

plo

plo

plo


plo

plo


plo

plo


Em Med

ical

Em Med

ical

Em Med

ical

men

Cost alth

Family

men

Cost alth

Family

men

Cost alth

Family

men

Cost alth


Em Med

ical

Em Med

ical

Em Med

ical

men

Cost alth

Family

men

Cost alth

Family

men

Cost alth

Family

men

Cost alth


Em Med

ical

Em Med

ical

Em Med

ical

men

Cost alth

Family

men

Cost alth

Family

men

Cost alth

Family

men

Cost alth


loy We

loy We

loy We

loy We

loy We

loy We

loy We

loy We

loy We

loy We

loy We

loy We


Med

ical

mp

Med

ical

mp

Med

ical

mp

Med

ical


Med

ical

mp

Med

ical

mp

Med

ical

mp

Med

ical


Med

ical

mp

Med

ical

mp

Med

ical

mp

Med

ical



60% 60%60% 60%



40%40%

40%40%

30%30%30% 20%30% 20%

20% 10%20% 10%


men

t

sts

alth

Family

men

t

sts

alth


sts

alth

Family ent

sts

alth


0%

oy al Co

We

oy al Co

We0%

oym Co

We

oym Co

We

ent

osts

alth

mily ent

osts

alth

mily ent

osts

alth


mpl dic


men

t

sts

alth

Family

men

t

sts

alth

Family

men

t

sts

alth


Med

ical

pl

Med

ical

Persisting Shocks

oy

Co

We

oy

Co

We

oy

Co

We Em Em

mpl

Med

ical

mpl

Med

ical

mpl

Med

ical



0%

10%

20%

30%

40%

50%

60%

70%

Employmen

t

Med

ical Costs

Wealth

Family

Employmen

t

Med

ical Costs

Wealth

Family

Employmen

t

Med

ical Costs

Wealth

Family



Persisting Shocks

Education

0%

10%

20%

30%

40%

50%

60%

Employmen

t

Med

ical Costs

Wealth

Family

Employmen

t

Med

ical Costs

Wealth

Family

Employmen

t

Med

ical Costs

Wealth

Family



Persisting Shocks

0%

10%

20%

30%

40%

50%

60%

70%

80%

Employmen

t

Med

ical Costs

Wealth

Family

Employmen

t

Med

ical Costs

Wealth

Family

Employmen

t

Med

ical Costs

Wealth

Family

Employmen

t

Med

ical Costs

Wealth

Family



Persisting Shocks

Income Strata

0%

10%

20%

30%

40%

50%

60%

Employmen

t

Med

ical Costs

Wealth

Family

Employmen

t

Med

ical Costs

Wealth

Family



Persisting Shocks


Highest Quartile


percent of households in the highest income quartile reported themselves to be very worried about at least one threat to continued employment. Economic uncertainty in other domains—notably, medical care—was even more common: About a quarter of the most economically advantaged households reported themselves to be very worried about these risks.

No group in American society appears fully insulated from economic instability. Among college graduates, for example, more than 63 percent experienced persisting shocks to their wealth, almost 21 percent experienced persisting shocks due to medical costs, more than 12 percent experienced persisting shocks due to employment instability, and 11 percent experienced family-related persisting shocks. By comparison, the prevalence of persisting shocks among households headed by those with, at most, a high school degree was more than 45 percent for wealth shocks, nearly 26 percent for medical shocks, nearly 24 percent for employment shocks, and just over 13 percent for family-related shocks—higher for every category but wealth, but only modestly so for every domain but employment.

Starker Disparities in the Strength of Buffers

While exposure to risk during the recession did not vary as dramatically as might be supposed, the strength of protections against those risks did. Thus the significance of the uneven exposure to economic risks depends in large part on the distribution of “buffers” against economic risks—insurance, savings, family risk-sharing, and other resources that can be tapped in response to economic instability.

We might expect that those whose lives are most affected by economic turbulence would have the greatest incentive to prepare. What the SERPI shows, however, is that those households that have the weakest buffers also face the greatest risk. Because these risks are concentrated among households

Fig. 11 Uneven Financial Buffers against Economic Uncertainy

Socio-Economic Status, 2009

0

Percen

tage

0

Percen

tage

= 3 Income Account ‐

80808080

nces

60ta 60

umstan

ces

60

umstan

ces

60

umstan

ces

ums

cCirc

Circ

Circ

40al

40al 40alna

nci

nanci

nanci

FiFiFi

20ese 20ese 20ese

g Th

gTh

gTh

nnn

0

Repo

rti

Repo

rti

Repo

rti

20tage

2020

cen

Per

4040

Can Borrow $5000+ from Family/Friends Financial Reserves = 3 Months+ Income Drained Retirement Account to Pay Bills Debt Too Large to Pay‐Off Can Borrow $5000+ from Family/Friends Financial Reserves Months+ Drained Retirement to Pay Bills Debt Too Large to Pay Off

40

20

0

20

40

60

80

Percen

tage

Rep

orting

The

se Finan

cial Circumstan

ces

Can Borrow $5000+ from Family/Friends Financial Reserves = 3 Months+ Income Drained Retirement Account to Pay Bills Debt Too Large to Pay‐Off

%

%

%

%

%

%

%

High School Some College College Graduate Lowest Quartile Second Quartile Third Quartile Fourth Quartile

0

Percen

tage

0

Percen

tage


80808080

nces

60ta 60

umstan

ces

60

umstan

ces

60

umstan

ces

ums

cCirc

Circ

Circ

40al

40al 40alna

nci

nanci

nanci

FiFiFi

20ese 20ese 20ese

g Th

gTh

gTh

nnn

0

Repo

rti

Repo

rti

Repo

rti

20tage

2020

cen

Per

4040


40

20

0

20

40

60

80

Percen

tage

Rep

orting

The

se Finan

cial Circumstan

ces



that have faced longstanding economic disadvantages, households face serious constraints in preparing against economic dislocations.

Which Americans are better prepared to cope with risks because they can buffer them? Some of the differences charted in Figure 11 are unsurprising: Risk buffers are more resilient for households with advantaged socio-economic characteristics, whether measured in terms of educational attainment or household income.

Family Composition, 2009

0

Percen

tage

0

Percen

tage

= 3 Income

Account ‐

60606060

40

cumstan

ces

40

umstan

ces

40

umstan

ces

40

umstan

ces

cccrl Cir

lCir

lCir

20

ncia 20

ncia 20

ncia

ina

ina

ina

0

FFFTh

ese

These

These

ng

ngng

202020

Repo

rti

Repo

rti

Repo

rti

tage

40

en

4040

cPe

r

606060

Can Borrow $5000+ from Family/Friends Financial Reserves = 3 Months+ IncomeCan Borrow $5000+ from Family/Friends Financial Reserves Months+

Drained Retirement Account to Pay Bills Debt Too Large to Pay‐Off Drained Retirement to Pay Bills Debt Too Large to Pay Off

7070

6060

nces

50ta 50

umstan

ces

ms

40rcial Ci

30

nc

20Fina

ese

10 Th

ng

0

Repo

rti

10 e

20

ercentag

P

40

30

20

10

0

10

20

30

40

50

60

70

Percen

tage

Rep

orting

The

se Finan

cial Circumstan

ces

Can Borrow $5000+ from Family/Friends Financial Reserves = 3 Months+ Income

Drained Retirement Account to Pay Bills Debt Too Large to Pay‐Off

60

40

20

0

20

40

60

Percen

tage

Rep

orting

The

se Finan

cial Circumstan

ces



Race and Ethnicity, 2009

White African-American Latino

0

Percen

tage

0

Percen

tage

= 3 Income

Account ‐

60606060

40cumstan

ces

40um

stan

ces

40um

stan

ces

40um

stan

ces

cccrl Cir

lCir

lCir

20

ncia 20

ncia 20

ncia

ina

ina

ina

0

FFFTh

ese

These

These

ng

ngng

202020

Repo

rti

Repo

rti

Repo

rti

tage

40

en

4040

cPe

r

606060

Can Borrow $5000+ from Family/Friends Financial Reserves = 3 Months+ IncomeCan Borrow $5000+ from Family/Friends Financial Reserves Months+

Drained Retirement Account to Pay Bills Debt Too Large to Pay‐Off Drained Retirement to Pay Bills Debt Too Large to Pay Off

7070

6060

nces

50ta 50

umstan

ces

ms

40rcial Ci

30

nc

20Fina

ese

10 Th

ng

0

Repo

rti

10 e

20

ercentag

P

40

30

20

10

0

10

20

30

40

50

60

70

Percen

tage

Rep

orting

The

se Finan

cial Circumstan

ces



60

40

20

0

20

40

60

Percen

tage

Rep

orting

The

se Finan

cial Circumstan

ces



Couple Single Dependent Children No Dependent Children

%

%

%

%

%

%

%

%

%

%

%

%

%

%

%

%

%

%

0

Percen

tage

0

Percen

tage


80808080

nces

60ta 60

umstan

ces

60

umstan

ces

60

umstan

ces

ums

cCirc

Circ

Circ

40al

40al 40alna

nci

nanci

nanci

FiFiFi

20ese 20ese 20ese

g Th

gTh

gTh

nnn

0

Repo

rti

Repo

rti

Repo

rti

20tage

2020

cen

Per

4040


40

20

0

20

40

60

80

Percen

tage

Rep

orting

The

se Finan

cial Circumstan

ces


0

Percen

tage

0

Percen

tage


80808080

nces

60ta 60

umstan

ces

60

umstan

ces

60

umstan

ces

ums

cCirc

Circ

Circ

40al

40al 40alna

nci

nanci

nanci

FiFiFi

20ese 20ese 20ese

g Th

gTh

gTh

nnn

0

Repo

rti

Repo

rti

Repo

rti

20tage

2020

cen

Per

4040


40

20

0

20

40

60

80

Percen

tage

Rep

orting

The

se Finan

cial Circumstan

ces



Other patterns, however, are less predictable. Although African-Americans and Latinos have roughly similar economic circumstances, risk buffers for the former are distinctly weaker, particularly the balance between debt and financial reserves and the very limited capacity to borrow from family and friends—a difference that has been identified in prior research for wealth holding, but not for these other forms of risk buffers.17

Also noteworthy is that households with children appear to be less adequately buffered against risk than do other households. Their greater capacity to borrow from family and friends appears more than offset by weaknesses evident in the other three measures of buffer resiliency.

The Unequal Consequences of Economic Insecurity

How do unequal exposure to and protections against economic risk influence households’ financial well-being? Figure 12 tells the story. Because respondents from higher socio-economic status are both (modestly) less likely to experience

Unmet Needs by Education, Race/Ethnicity(With and Without Shocks)

Multiple Multiple Multiple


Unmet Needs By Family CompositionUnmet Needs By Family CompositionUnmet Needs By Family CompositionUnmet Needs By Family Composition(With and Without Shocks)(With and Without Shocks)(With and Without Shocks)(With and Without Shocks)

60%60%60%60%

50%50%50%50%

40%40%40%40%

30%30%30%

20%20%20%

10%10%10%

0%0%0%

Single Multiple Single Multiple Single Multiple No Children No Children No ChildrenSingle Single Single No Children No Children No ChildrenSingle Single Single No Children No Children No ChildrenAdults Adults Adults Adults Adults Adults Children Children ChildrenAdults Adults Adults Adults Adults Adults Children Children ChildrenAdults Adults Adults Adults Adults Adults Children Children Children

No Shocks Any Shocks Nonwealth Shocks No Shocks Any Shocks Nonwealth Shocks No Shocks Any Shocks Nonwealth Shocks No Shocks Any Shocks Nonwealth ShocksNo Shocks Any Shocks Nonwealth Shocks No Shocks Any Shocks Nonwealth Shocks

Unmet Needs By Education, Race/EthnicityUnmet Needs By Education, Race/Ethnicity(With and Without Shocks)(With and Without Shocks)

70%70%

60%60%

50%50%

40%40%

30%30%

20%

10%

0%

Scho

ol

College

aduate

Scho

ol

College

aduate

Scho

ol

College

aduate

White

Latin

o

White

Latin

o

White

Latin

o

r r r

American

American

American

High

Some G

High

Some G

High

Some G ‐ ‐ ‐

College

College

College

African

African

African

No Shocks Any Shocks Nonwealth Shocks No Shocks Any Shocks Nonwealth Shocks

0%

10%

20%

30%

40%

50%

60%

Single Adults

Multiple Adults

Single Adults

Multiple Adults

Single Adults

Multiple Adults

No Children

Children No Children


Children


Unmet Needs By Family Composition(With and Without Shocks)

0%

10%

20%

30%

40%

50%

60%

70%

High Scho

ol

Some Co

llege

College Gradu

ate

High Scho

ol

Some Co

llege

College Gradu

ate

High Scho

ol

Some Co

llege

College Gradu

ate

White

African

‐American

Latin

o

White

African

‐American

Latin

o

White

African

‐American

Latin

o


Unmet Needs By Education, Race/Ethnicity(With and Without Shocks)

Perc

enta

ge w

ith

Unm

et N

eeds


shocks that might induce unmet needs and (substantially) more capable of buffering those shocks when they occur, the consequences of economic shocks for economic well-being are notably more negative for less advantaged Americans, measured in terms of education.

Differences by race and ethnicity are also striking. Given that African-Americans have more risk exposure and less resilient buffers, it is not surprising that they report very high levels of unmet needs when they experience economic shocks. Latinos report comparable levels of unmet needs, and both groups report much

higher levels of unmet needs than whites. Yet, more surprising, Latinos who do not experience economic shocks report lower levels of unmet needs than either African-Americans or whites.

Family composition has mixed implications. On the one hand, two-adult households are much less likely to experience unmet needs when they face

no economic instability. On the other hand, among households experiencing economic shocks, the gap in unmet needs between single- and multiple- adult households is much smaller. This may partly reflect the higher baseline level of unmet needs for single adults, but it suggests, at a minimum, that multiple adult households are powerfully affected by shocks. This is in large part because many of these shocks are transmitted through employment; multi-worker households thus get doubly exposed to unemployment, benefit cuts and the like, even though their dual-earner capacity also allows them to more readily buffer some of these shocks.

The presence of children in the household yields a more consistent pattern. Although households with children report modestly greater unmet needs compared to childless households when there is no economic instability, both

Households with children appear to be less adequately buffered against risk than do other households.

Fig. 12 Unmet Needs by Family Composition(With and Without Shocks)



Unmet Needs By Family CompositionUnmet Needs By Family CompositionUnmet Needs By Family CompositionUnmet Needs By Family Composition(With and Without Shocks)(With and Without Shocks)(With and Without Shocks)(With and Without Shocks)

60%60%60%60%

50%50%50%50%

40%40%40%40%

30%30%30%

20%20%20%

10%10%10%

0%0%0%

Single Multiple Single Multiple Single Multiple No Children No Children No ChildrenSingle Single Single No Children No Children No ChildrenSingle Single Single No Children No Children No ChildrenAdults Adults Adults Adults Adults Adults Children Children ChildrenAdults Adults Adults Adults Adults Adults Children Children ChildrenAdults Adults Adults Adults Adults Adults Children Children Children

No Shocks Any Shocks Nonwealth Shocks No Shocks Any Shocks Nonwealth Shocks No Shocks Any Shocks Nonwealth Shocks No Shocks Any Shocks Nonwealth ShocksNo Shocks Any Shocks Nonwealth Shocks No Shocks Any Shocks Nonwealth Shocks

Unmet Needs By Education, Race/EthnicityUnmet Needs By Education, Race/Ethnicity(With and Without Shocks)(With and Without Shocks)

70%70%

60%60%

50%50%

40%40%

30%30%

20%

10%

0%

Scho

ol

College

aduate

Scho

ol

College

aduate

Scho

ol

College

aduate

White

Latin

o

White

Latin

o

White

Latin

o

r r r

American

American

American

High

Some G

High

Some G

High

Some G ‐ ‐ ‐

College

College

College

African

African

African


0%

10%

20%

30%

40%

50%

60%

Single Adults

Multiple Adults

Single Adults

Multiple Adults

Single Adults

Multiple Adults

No Children



Children


Unmet Needs By Family Composition(With and Without Shocks)

0%

10%

20%

30%

40%

50%

60%

70%

High Scho

ol

Some Co

llege

College Gradu

ate

High Scho

ol

Some Co

llege

College Gradu

ate

High Scho

ol

Some Co

llege

College Gradu

ate

White

African

‐American

Latin

o

White

African

‐American

Latin

o

White

African

‐American

Latin

o


Unmet Needs By Education, Race/Ethnicity(With and Without Shocks)

Perc

enta

ge w

ith

Unm

et N

eeds


differentials get strikingly larger when the households experience economic shocks. The gap in unmet needs between households with children and without children grows by 10 percentage points in the face of economic instability and appears equally large whether the shocks involve wealth or other risk domains. It remains unclear why households with children are less capable of coping with economic uncertainty. Children themselves can be a source of unexpected costs. In addition, adults in households with children may be constrained by child care or other responsibilities, limiting their ability to adapt to shifting work expectations, medical needs, or other changing circumstances.

Economic Insecurity through Americans’ EyesInsecurity has become a dominant motif in Americans’ economic lives. Strikingly, however, no regularly funded national survey tracks even a small share of the factors that shape Americans’ economic security.18 Drawing on a unique two-wave survey that comprehensively examines these factors during the recession (and allows comparison back to 2007), this report has painted a broad portrait of how Americans perceive their economic security.

This portrait shows that Americans are standing on shaky ground. During the recent downturn, few have not been touched by economic uncertainties. Yet even before the recession, more than half of all Americans expressed worries about their economic prospects. As economic shocks have multiplied and instability persisted, these worries have transformed into broader and more persistent anxiety driven by the widespread experience of economic loss.

To be sure, these shocks are not felt equally. Americans differ in their exposure to economic instability and, even more starkly, in their capacity to cope with it. Less advantaged Americans and racial and ethnic minorities look substantially more vulnerable to economic shocks than do Americans as a whole—primarily because their basic economic buffers are so weak. Households with children appear more deeply affected, perhaps because the adults in these households have less flexibility to adapt to changing economic circumstances.

Yet economic uncertainties touch even those who are relatively well-off: About half of households with between $60,000 and $100,000 in annual income who experienced persisting shocks to employment or medical costs reported some unmet basic needs; almost 40 percent reported multiple unmet needs. Altogether, somewhere between a quarter and a third of all Americans reported unmet needs associated with economic shocks.

While less advantaged Americans and racial and ethnic minorities look substantially more vulnerable to economic shocks, economic uncertainties touch even those who are relatively well-off.


The SERPI assessed Americans’ experiences and perceptions of economic insecurity in the depths of the recession, which might lead to the concern that its findings are highly time-bound. Yet, as Figure 13 indicates, general poll data suggest that 2009 was not particularly unusual.19

Worries about job loss did jump in 2009. By the summer of 2010, however, these concerns had abated only modestly. The increase in reported worries during the downturn was surprisingly modest for housing stability, in part because Americans’ concerns were already high; nor did they decline much after 2009. For wealth (adequate retirement savings) and medical costs, concerns were high even before 2009 and have not diminished since. Americans have been standing on shaky ground for some time, and the ground still seems extremely unstable.

Whether the ground on which Americans stand is more stable in the future depends not just on the pace of economic recovery, but also on whether we learn from their longstanding experiences with economic insecurity and work to reduce its frequency, severity, and impact in the future.

Tracking Public Concern About Some Aspects Of Economic SecurityTracking Public Concern About Some Aspects Of Economic SecurityTracking Public Concern About Some Aspects Of Economic SecurityPercent Very or Somewhat Worried, 2005 to 2010Percent Very or Somewhat Worried, 2005 to 2010Percent Very or Somewhat Worried, 2005 to 2010

90%90%

80%80%

70%70%

ans

ans

60%

ric 60%

ric

50%Ame

50%Ame

of

of

40%ge

40%

tage

enta

en

30%

Perc

20%

10%

0%

Summer 2005 Summer 2007 Summer 2009 Summer 2010

Categories of Insecurity

Losing Job Losing Home Medical Costs

0%

10%

20%

30%

40%

50%

60%

70%

80%

90%

Summer 2005 Summer 2007 Summer 2009 Summer 2010

Percen

tage

of A

merican

s

Categories of Insecurity

Tracking Public Concern About Some Aspects Of Economic SecurityPercent Very or Somewhat Worried, 2005 to 2010

Losing Job Losing Home Medical Costs

Fig. 13 Tracking Public Concern about Aspects of Economic SecurityPercent “Very” or “Somewhat” Worried, 2005 to 2010

Source: Kaiser Family Foundation, Kaiser Health Tracking Poll


Technical AppendixThis appendix describes in more detail the Survey of Economic Risk Perceptions and Insecurity.

The SERPI was incorporated as a part of the 2008-2009 Panel Survey of the American National Election Studies (ANES), a survey funded by the National Science Foundation for the past half century. The Panel Survey was an on-line survey of a nationally representative sample of Americans who agreed to be interviewed monthly between January 2008 and October 2009. Thirteen of the monthly waves were available to other researchers. With financial support from the Rockefeller Foundation, the SERPI was fielded as the sole questions for the panel in March (Wave 15) and September (Wave 21) of 2009.

The ANES panel was constructed to provide a representative sample of the American population aged 18 and older as of November 4, 2008. Although not all eligible panelists completed the two waves involved in the SERPI, all the results reported below are weighted to replicate a nationally representative distribution of respondents. Data collected on economic experiences, perceptions, and expectations during these two waves were merged with socio-demographic and other personal characteristics collected from respondents in other waves of the ANES survey.

Measures of Economic Insecurity

To accurately assess the prevalence and consequences of insecurity, it was essential to measure insecurity in a more complete manner than had been done previously. At the same time, the SERPI was designed to allow comparison of perceptions of insecurity measured prior to the economic downturn. In the spring of 2007, the Rockefeller Foundation sponsored a survey under the auspices of its New American Worker initiative that asked three thousand Americans about their economic security. The core of this survey was a question about “your family’s economic security” defined for respondents as “being able to keep your job, maintain your income, have health insurance coverage, and retire comfortably.” The survey also asked about concerns related to more specific economic risks, including losing one’s job, facing large out-of-pocket medical spending, and lacking adequate income to pay for a comfortable and secure retirement.

To allow comparisons with this 2007 survey, the two waves of the SERPI incorporated an identically worded set of questions about economic security in general, as well as about some of the more specific forms of economic risks asked about in the 2007 survey.20 However, the SERPI substantially augmented these earlier questions. In the employment domain, the expanded scope included questions about (1) losing one’s job (if currently employed), (2) finding a job (if currently unemployed but still in the workforce), and (3) losing several months from work due to serious illness. In the health domain, the


risks included (1) large out-of-pocket medical expenses, (2) losing insurance coverage due to cost increases, (3) having coverage cut, (4) being unable to determine what types of services were actually covered, and (5) future nursing home costs. In the wealth domain, risks included (1) inadequate savings to support an adequate retirement, (2) being unable to cover the costs of current housing arrangements, (3) cuts in pension benefits, and (4) having debt so large that it could never be paid off. Risks in the family domain included (1) urgent financial assistance to members of one’s extended family, (2) losing one’s partner/spouse due to divorce, and (3) losing one’s partner/spouse due to death.

To ensure comparability across surveys, the same response scale was used as in the 2007 American Worker survey: respondents were asked if they were very, fairly, somewhat, or not at all worried about each risk. Although framing responses in terms of “worry” may mean that some respondents will be more willing than others to express their concerns, past studies suggest that worry is a reliable measure of concerns involving uncertain prospects.21

To assess the frequency with which these worries were realized, a matched set of questions were asked about whether respondents had experienced the corresponding unexpected events. For example, worries about employment correspond with questions regarding whether any adults in the household had lost their job. These negative events are described collectively in the report as economic “shocks.” Respondents on the March 2009 SERPI were asked about whether they had experienced specific shocks in the past year, those responding on the September 2009 wave were asked if they had experienced specific shocks in the past six months. For those completing both waves of the survey, it is therefore possible to construct an 18-month history of their experiences, ranging from March 2008 to September 2009.

Specifically, respondents were asked about two shocks in the employment domain (whether any worker in the household had “been unemployed not by personal choice” or had “lost more than a month from work due to serious illness or injury”), four shocks related to medical expenses (whether they had “lost their health insurance,” “had problems getting their insurance to pay for major medical expenses,” “had out-of-pocket major medical expenses as the result of serious illness or injury in their immediate family,” or “had to pay a lot more for their health insurance than expected”), three shocks related to wealth (“had their retirement benefits at work cut substantially,” “had the value of their investments or retirement funds decline substantially,” or “had the value of their house decline substantially”) and three shocks related to family (whether they had “spent a substantial sum helping out their extended family,” “been divorced or separated from their spouse,” or “had their spouse/partner pass away”).

Past research suggests that particular risks will feel more distressing if they are relatively hard to anticipate or expected to have a relatively large impact on household well-being.22 As part of the SERPI, respondents were asked, for a subset of economic risks, to assess whether the events in question were “ones that people can plan for, ones that come unexpectedly, or a mix of predictable


and unpredictable circumstances?” 23 And if those unfortunate events came to pass, how long it might be, “for the typical person like yourself, before the household’s financial situation returns to how it was before the event occurred.”

Buffers against Economic Risks

The extent to which economic uncertainty feels threatening depends in large part on a household’s capacity to buffer financial shocks should they occur. This in turn depends on the household’s savings and ability to tap into the equity of its illiquid assets, balanced against the carrying costs for the household’s debt.24 Beyond these household resources, risk can also be buffered by drawing upon the financial resources inherent in broader family connections and social networks.25

The value of households’ equity is difficult to assess with great reliability; the liquidity of these assets is even more difficult to measure. The SERPI therefore included indirect, but more readily answered, questions. These included: (1) an estimate of how many weeks/months the household could last, without hardship, if its current earnings vanished, (2) whether it had recently tapped into the equity of retirement accounts in order to pay expenses, and (3) whether the household had accumulated debt so substantial that the respondent anticipated difficulty in being able to pay it off. To assess the availability of financial resources through social networks, respondents were asked how much they could borrow from “family members and close friends” in a time of need.

Consequences of Economic Insecurity

High levels of insecurity may be consequential either because they cause distress or because they interfere with households’ ability to meet other basic needs. The SERPI measured distress through a combination of questions regarding how frequently respondents reported thinking about their households’ economic insecurity and the extent of anxiety that they experienced when they did so. The SERPI measured households’ ability to meet basic needs through a combination of questions regarding food security (had members of the household gone hungry because there was not enough money to pay for groceries), housing instability (losing one’s home through eviction or inability to pay the mortgage), and unmet medical needs (not going to the doctor for a known problem out of concern for costs).

Unlike the measures of psychological distress, which are directly attributed by respondents to perceptions of economic insecurity, these measures of unmet basic needs do not require households to attribute unmet needs to specific causes. To assess their relationship with insecurity, therefore, requires comparing unmet needs for households that report higher and lower levels of perceived insecurity or experienced instability but that otherwise look similar. In the text, this is done through comparison of households in similar demographic groups (for example, the third income quartile). To ensure that


the higher levels of unmet need that are associated with reported economic instability were not in fact a consequence of some other factor that occurred in conjunction with economic instability, a set of multivariate regression models that identified the association of unmet needs and economic shocks was estimated. These models controlled statistically for the household’s annual income in the year prior to the economic shocks; the value of its stock holdings; household size, race and ethnicity; the educational attainment of the primary respondent on the survey; and gender and age.

The results from these models revealed that while instability in all four economic domains was associated with higher levels of unmet need for the household, only for shocks associated with employment and medical expenses was this relationship statistically significant, controlling for shocks in other domains and the above-mentioned socio-economic characteristics of the household. Conversely, all four buffers were associated with unmet needs in a statistically significant manner, again controlling for the presence of economic shocks and other socio-economic characteristics of the household. The probability of unmet need was about half as large for persisting employment shocks, compared with those that occurred but once in the previous 18 months; for persisting shocks associated with medical expenses, the probability of unmet need was twice as high as for non-reoccurring shocks.


1 Jacob Hacker, The Great Risk Shift: The New Economic Insecurity and the Decline of the American Dream, rev. and exp. (New York: Oxford University Press, 2008); Peter Gosselin and Seth Zimmerman, Trends in Income Volatility and Risk, 1970–2004, Urban Institute Work-ing Paper (Washington, DC: Urban Institute, May 2008); Thomas Cusack, Torben Iversen, and Philipp Rehm, “Risks at Work: The Demand and Supply Sides of Government Redistribution,” Oxford Review Economic Policy 22, no. 3 (2006):365–389; Philipp Rehm, “Risks and Redistribution: An Individual-Level Analysis,” Comparative Political Studies 42, no. 7 (2009):855–881; Charles F. Manski, “Measuring Expectations,” Econometrica 72, no. 5 (2004):1329–1376; J. Dominitz and C. Manski, “Perceptions of Economic Insecu-rity: Evidence from the Survey of Economic Expectations,” Public Opinion Quarterly 61 (1997):261–287.

2 The MetLife’s Study of the American Dream: Against the Backdrop of the Financial Burden Shift (New York, NY: Met-ropolitan Life Insurance Company, 2007). Families USA, “Too Great A Burden: America’s Families at Risk,” Publica-tion No. 07-113 (Washington, DC: Families USA Foundation, 2007); Kaiser Family Foundation, Health Security Watch (Menlo Park, CA;: Kaiser Family Foundation, 2007); Paul Fronstin, “Savings Needed to Fund Health Insurance and Health Care Expenses in Retirement,” EBRI Issue Brief no. 295 (Employee Benefit Research Institute, July 2006). One exception to these more narrowly focused surveys was a survey to which we will return later in the report: The Rockefeller Foundation, American Worker Survey (New York: Rockefeller Foundation, 2007) [fielded by Yankelovich Inc, February 6-19, 2007].

3 Lars Osberg and Andrew Sharpe, “How Should We Measure the Economic Aspects of Well-Being,” Review of Income and Wealth 51, no. 2 (2005):311–336; E. Ligon and L. Schechter, “Measuring Vulnerability,” Economic Journal 113, no. 486 (2003):C95–C102.

4 There is an extensive literature relating each of these domains to instability of economic circumstances and (less extensively) to perceptions of economic insecurity. For some illustrative examples: Links between unemploy-ment and subsequent earnings trajectories and perceived insecurity were explored in Dominitz and Manski, “Per-ceptions of Economic Insecurity;” and Liliana Winkelman and Ranier Winkelman, “Why Are the Unemployed So Unhappy? Evidence from Panel Data,” Economica 65 (1998):1–15. The consequences of medical expenditures on economic security were documented most extensively in Sara R. Collins, Jennifer L. Kriss, Michelle M. Doty, and

Sheila D. Rustgi, Losing Ground: How The Loss of Adequate Health Insurance Is Burdening Working Families (New York: Commonwealth Fund, 2008); see also Peter J. Cun-ningham, “The Growing Financial Burden of Health Care: National and State Trends, 2001 – 2006,” Health Affairs 2, no.5 (2010):1037–1044; Changing distribution of wealth and its consequences for risk buffering was explored in Karen E. Dynan and Donald L. Kohn, The Rise in U.S. House-hold Indebtedness: Causes and Consequences, Discussion Paper, 2007-37, Finance and Economics Discussion Series Divisions of Research & Statistics and Monetary Affairs Federal Reserve Board (Washington, DC, 2007); and Jona-than Gruber, “The Wealth of the Unemployed,” Industrial and Labor Relations Review 55, no. 1 (2001):79–94. Illustra-tions of the impact of family configurations on economic security was examined in Marianne E. Page and Ann Huff Stevens, “The Economic Consequences of Absent Parents,” The Journal of Human Resources 39, no. 1 (2004):80–107; and Paul Attewell, “The Impact of Family on Job Displace-ment and Recovery,” Annals of the American Academy of Political and Social Science 562 (1999):66–82.

5 Although not all eligible panelists completed the two waves involved in the SERPI, all the findings reported are weighted to produce nationally representative results. Of the 3,657 active panelists in March 2009, 2,493 (68.2 percent of the eligibles) completed the survey; of the 2,527 active panelists in September 2009, 2,203 (62.5 percent of the eligibles) completed the survey. A total of 2,084 respondents completed both waves of the survey.

6 “Economic security” was defined as “being able to keep your job, maintain your income, have health insurance coverage, and retire comfortably.”

7 Not every respondent was asked about all of these risks: those who were retired were not asked about losing their job; those who had no younger children were not asked about the costs of child care or education; those who had no health insurance were not asked about the risk of unexpected premium increases or otherwise losing insurance coverage.

8 Worry is a generally robust metric for assessing inse-curity. See Louie Rivers and Joseph Arvai, “Win Some, Lose Some: The Effect of Chronic Losses on Decision Making Under Risk,” Journal of Risk Research 10, no. 8 (2007):1085–1099; V. Ricciardi, “The Financial Psychology of Worry and Women,” Social Science Research Network (SSRN) Working Paper Series, 2008. [(http://papers.ssrn.com/sol3/papers.cfm?abstract_id=1093351].

Notes


9 Collins, Kriss, Doty, and Rustgi, Losing Ground; Families USA, Too Great a Burden; Kaiser Family Foundation, Health Security Watch.

10 Jennifer Lerner and Dacher Keltner, “Fear, Anger and Risk,” Journal of Personality and Social Psychology 81, no. 1 (2001):146-159; Jennifer Lerner, Roxana Gonzalez, Deborah Small, and Baruch Fischhoff, “Effects of Fear and Anger on Perceived Risks of Terrorism: A Natural Experiment,” Psychological Science 14, no.2 (2003):144–150.

11 Comparing expectations between people who have actually experienced these events and those who have not, it appears that those who are subject to an economic shock are more pessimistic about the duration of their impact, perhaps reflecting greater awareness of the actual consequences or greater anxiety based on prior experience.

12 Talya Miron-Shatz, “Am I Going To Be Happy And Financially Stable?” How American Women Feel When They Think About Financial Security,” Judgment and Decision Making 4, no. 1 (2009):102–112.

13 The literature on the linkage between stress, anxiety and health is vast. See, for example, Terrence D. Hill, Cath-erine E. Ross, and Ronald J. Angel, “Neighborhood Disorder, Psychophysiological Distress, and Health,” Journal of Health and Social Behavior 46, no. 2 (2005):170–186, and E. B. Faragher, M. Cass, and C. L. Cooper, “The Relationship between Job Satisfaction and Health: A Meta-Analysis,” Occupational and Environmental Medicine 62, no. 2 (2005):105–112.

14 Most commonly, this involves not going to the doctor for needed care because the respondent was concerned about the potential costs of that visit.

15 There is one exception to this general pattern. When there are multiple shocks in the family domain (most of-ten, when the loss of a partner/spouse coincides with the need for financial assistance from one’s extended family), the level of unmet need is more than twice as high as for households that experience a single family-related shock. Multiple shocks in the family domain seemingly induce more unmet needs than for multiple shocks in any other domain of economic life.

16 The value of households’ equity is difficult to assess with great reliability; the liquidity of these assets is even more difficult to measure. The SERPI therefore included indirect, but more readily answered, questions. These included: (a) an estimate of how many weeks/months the household could last without hardship if its current earn-ings vanished, (b) whether it had recently tapped into the

equity of retirement accounts in order to pay expenses, and (c) whether the household had accumulated debt so substantial that the respondent anticipated difficulty in being able to pay it off. To assess the availability of financial resources through social networks, respondents were asked how much they could borrow from “family members and close friends” in a time of need. Because the aggregate distribution of responses was very similar across the two waves of the survey, only the fall results are presented here. As noted above, because the stock market had rallied somewhat by the fall of 2009 and some aspects of the national economy were at least tem-porarily appearing a bit more robust, these fall estimates are slightly more optimistic about the capacity of risk buffers than would be the estimated based on the spring survey data.

17 Lauren J. Krivo and Robert L. Kaufman, “Housing and Wealth Inequality: Racial-Ethnic Differences in Home Equity in the United States,” Demography 41, no. 3 (2004):585–605; N. Anders Klevmarken, Joseph P. Lupton, and Frank P. Stafford, “Wealth Dynamics in the 1980s and 1990s: Sweden and the United States,” The Journal of Human Resources 38, no. 2 (2003): 322–353; Michael Hurd and Arie Kapteyn, “Health, Wealth, and the Role of Institutions,” The Journal of Human Resources 38, no. 2 (2003):386–415; Seymour Spilerman,“Wealth and Stratification Processes,” Annual Review of Sociol-ogy 26 (2000):497–524; Lisa A. Keister and Stephanie Moller,“Wealth Inequality in the United States,” Annual Review of Sociology 26 (2000):63–81.

18 The Health Security poll that has been fielded by the Kaiser Family Foundation since 2004 incorporates a broad array of questions about Americans worries related to medical expenses, as well as a few non-medical sources of economic insecurity. But it is not sufficiently comprehen-sive regarding the latter and collects no information at all about experienced economic shocks, risk buffers, or other attributes of the household essential for understanding how it is coping with economic uncertainty.

19 Kaiser Family Foundation, Kaiser Health Tracking Poll: July 2010. Publication No. 8084-T (Menlo Park, CA: Kaiser Family Foundation).

20 In some cases the wording of the items from the American Worker Survey was altered to more effectively measure the risks in question. For example, the 2007 survey asked about concerns about providing one’s parents with financial assistance, whereas the two 2009 waves broadened the wording to include financial assistance for all members of one’s extended family. Of the 15 specific risks identified in the text above, 7 retained identical wording between the 2007 and 2009 versions.


21 C. F. Manski, “Measuring Expectations;” Louie Rivers and Joseph Arvai, “Win Some, Lose Some: The Effect of Chronic Losses on Decision Making Under Risk,” Journal of Risk Research 10, no. 8 (2007):1085–1099; V. Ricciardi, “The Financial Psychology of Worry and Women,” Social Science Research Network (SSRN) Working Paper Series, 2008. [http://papers.ssrn.com/sol3/papers.cfm?abstract_id=1093351]. To test the reliability of the worry measure, for a subset of a half dozen risks the SERPI also included an alternative method for measuring concern in terms of the frequency with which people reported thinking about the risk in emotionally charged terms. (For a rationale for this alternative method, see Talya Miron-Shatz, “Am I Going To Be Happy and Financially Stable?” Although there was some variation across the type of risk, the worry measures captured about 85-90% of those who reported that they were fretting about these risks (that is, thinking about them often in ways that evoked strong negative emotions).

22 L. Osberg and A. Sharpe, “How Should We Measure the ‘Economic’ Aspects of Well-Being;” C.F. Manski, “Measuring Expectations”; Carol Graham and Stefano Pettinato, Happiness and Hardship: Opportunity and Insecurity in New Market Economies (Washington, DC: The Brookings Institution Press, 2002); Paul Slovic, “Trust, Emotion, Sex, Politics and Science: Surveying the Risk-Assessment Battlefield,” Risk Analysis 19, no. 4 (1999):689–702.

23 Two risks were assessed in each of the four domains that were the focus of the SERPI: For employment, the risk of being laid-off or becoming disabled; for medical costs the risk of large out-of-pocket medical expenses or of losing one’s health insurance; for wealth, the risk of having insufficient income during retirement or of losing one’s home; and for family, the risk of divorce or having a member of one’s extended family require substantial financial assistance.

24 Austin Nichols and Melissa M. Favreault, The Impact of Changing Earnings Volatility on Retirement Wealth (Washington, DC: Urban Institute, 2009); Christopher D. Carroll, Karen E. Dynan, and Spencer D. Krane, “Unemployment Risk and Precautionary Wealth: Evidence from Households’ Balance Sheets,” The Review of Economics and Statistics 85, no.3 (2003):586–604.

25 Samuel Bentolila and Andrea Ichino, “Unemployment and Consumption Near and Far Away from the Mediterranean,” Journal of Population Economics 21, no. 2 (2008):255–280; R. Schoeni, “Does Unemployment Insurance Displace Familial Assistance?” Public Choice 110, no. 1–2(2002):99–119.


About�the�Economic�Security�Index�Research�Team

Jacob S. Hacker (Ph.d., Yale University) is Stanley Resor Professor of Political Science at Yale University, where he is also a Resident Fellow of the Center for the Study of American Politics and the Institution for Social and Policy Studies. An expert on the politics of U.S. health and social policy in cross-national perspective, he is the author of five books, numer-ous journal articles, and a wide range of popular writings on American politics and public policy, with a focus on health and economic security.

Gregory Huber (Ph.d., Princeton University) is Associate Professor of Political Science at Yale University and a Resident Fellow of the Institution for Social and Policy Studies and the Center for the Study of American Politics. Prior to coming to Yale, he held the Robert Hartley Fellowship in Governmental Studies at the Brookings Institution. His research, which has been funded by the national Science Foundation, has appeared in the American Political Science Review, Quarterly Journal of Political Science, American Journal of Political Science, and Journal of Law, Economics, and Organization. His is also the author of The Craft of Bureaucrat-ic Neutrality (Cambridge University Press, 2007), which examines the conditions under which external political actors are able to influence how bureaucratic agencies enforce the law.

Philipp Rehm (Ph.d., duke University) is Assistant Professor of Political Science at Ohio State University; previous posts include the Postdoctoral Prize Research Fellowship at nuff-ield College, Oxford University. His work is located at the intersection of political economy and political behavior. In particular, he is interested in the causes and consequences of in-come dynamics (such as income loss, income volatility, and risk exposure). At the micro-level, his research explores how income dynamics shape individual preferences for redistribution, social policies, and parties. At the macro-level, his work analyzes the impact of labor market and income dynamics on polarization, electoral majorities, and coalitions underpinning social policy.

Mark Schlesinger (Ph.d., University of Wisconsin) is Professor of Health Policy and a Fellow of the Institution for Social and Policy Studies at Yale University and the most recent past editor of the Journal of Health Policy, Politics and Law. His research explores the determinants of public opinion about health and social policy, the influence of bounded rationality on medical consumers, and the role of nonprofit organizations in American medicine. He has consulted to a half dozen federal agencies, several dozen state and local governments, and more than a score of nonprofit organizations.

Future reports�� Who�Are�America’s�Insecure?��

Results�from�the�Economic�Security�Index

�� Medical�Care�as�a�Threat�to�American�Economic�Security

�� Wealth,�Debt,�and�American�Economic�Security

�� The�Causes,�Severity,�and�Persistence�of�Family�Income�Losses

�� American�Economic�Security�in�Cross-National�Perspective

www.economicsecurityindex.org

Index

Future reports�� Who�Are�America’s�Insecure?��

Results�from�the�Economic�Security�Index

�� Medical�Care�as�a�Threat�to�American�Economic�Security

�� Wealth,�Debt,�and�American�Economic�Security

�� The�Causes,�Severity,�and�Persistence�of�Family�Income�Losses

�� American�Economic�Security�in�Cross-National�Perspective

www.economicsecurityindex.org


With support from The Rockefeller Foundation

esi economic security index report...shaky ground, rocked by economic tremors whose consequences...

Documents