Economic Insecurity and Political Trust in the United States

Abstract

Extant research demonstrates that citizens’ evaluations of national economic performance play an important role in determining trust in politicians and political institutions, whereas evaluations of their own economic situation play a lesser or even negligible role. Utilizing American National Election Studies data and more apposite measures of personal economic privation during an age of globalization and de-industrialization, this article finds that the extent to which citizens perceive themselves and their families to be economically insecure has a statistically significant and substantial negative effect on political trust. Indeed, the effect at least matches those of macro-economic evaluations and party identification. This article therefore adds a new dimension to our understanding of the economy–trust nexus and contributes to the small but growing body of scholarship on insecurity’s effects on political behavior.

Keywords

trust in government economic insecurity risk prospect theory perceptions

Washington, D.C., has become an island. The gap between our citizens and our government has never been so wide.

—President Jimmy Carter (July 15, 1979)

We face a deficit of trust—deep and corrosive doubts about how Washington works that have been growing for years.

—President Barack Obama, State of the Union Address (January 27, 2010)

For some time, it has been almost axiomatic that each new prospective president or prime minister in the oldest and most stable democracies will presage his or her term in office by declaring a crisis of confidence in the nation’s governing institutions, quickly followed by a plan to remedy the widespread malaise. Inevitably, political trust remains low, or, more often, falls even further. The next leader repeats the process, and so on ad infinitum. Low trust, then, is not new, and a rich seam of trust research has been mined by political scientists over several decades (Citrin, 1974; Miller, 1974). Little noticed by many academic colliers, however, is that the era of low trust has coincided with a growth in economic insecurity in the world’s most advanced economies. Regardless of the ebb and flow of national economic fortunes, citizens’ job tenure and health and retirement benefits are becoming increasingly insecure as globalization and other developments increase labor-market competition and corporations mothball generous health plans and final salary pension schemes. There has been a Great Risk Shift (Hacker, 2008) in the United States and possibly elsewhere from the broad shoulders of government and corporations onto the narrow shoulders of individuals. Despite these important developments, little social science work has focused on the consequences of the shift in risk for political behavior generally (Hacker, Rehm, & Schlesinger, 2013) and political trust specifically. This article represents one of the first attempts to explore these effects.

They are worth exploring because trust matters. Thinking about its effects in the round, Hetherington (2007) concludes that political distrust “has defined [the] American political landscape over the last several decades” (p. 1), from the election of outsider presidents to handicapping progressive domestic policies to increasing support for recent conservative foreign-policy interventions (Hetherington & Husser, 2012). Trust’s importance is also evident at a more micro level. A lack of it undermines presidential approval (Hetherington, 1998), increases voters’ support for non-incumbent and third-party candidates (Hetherington, 1999) and elite-challenging initiatives such as a term limits (Karp, 1995), but may, when combined with high levels of efficacy, increase political participation (Gamson, 1968). Distrust also leads citizens to think their political leaders are corrupt (Wroe, Allen, & Birch, 2013), and even influences the extent to which citizens obey the law (Tyler, 1990), pay their taxes (Scholz & Lubell, 1998), and trust each other (Brehm & Rahn, 1997).

This article contributes to the scholarly debate on the causes of political distrust by offering a more nuanced and detailed account of the role of personal economic privation. Research demonstrates that citizens’ trust in politicians and political institutions is in part driven by their evaluations of the performance of the national economy, but the conventional wisdom is that personal economic experiences and evaluations do not play much of a role (Dalton, 2004; Lawrence, 1997; Lipset & Schneider, 1983; McAllister, 1999; Mishler & Rose, 2001). A key claim of this article is that the conventional wisdom is wrong. It is so because it has relied on a few crude indicators that fail to capture how individuals really perceive their personal economic situation. In particular, the indicators generally used by political scientists fail to measure accurately an increasingly important concern of citizens in mature post-industrial democracies: that is, economic insecurity. Drawing on the innovative work of Jacob Hacker and his colleagues, this article utilizes some new insecurity measures to improve our understanding of the trust–economy nexus. Taking into account a range of other factors, it finds that citizens’ perceptions of their economic insecurity have a statistically significant and substantial negative effect on their levels of political trust.

There is compelling evidence that economic insecurity has increased steadily from at least the mid-1980s and probably from the 1970s, and is now an important concern for many individuals and families in the United States and elsewhere (Hacker, 2008; Hacker et al., 2012; Hacker et al., 2013; Kalleberg, 2013; Jacobs, 2007; Osberg, 2009; Rockefeller Foundation, 2007; Standing, 2011). It is, indeed, one of the defining economic and political issues of our age, and encompasses among other things insecurity in employment, in retirement, in health, in widowhood or as a single-parent, and in the family wealth that can guard against economic shocks.

Given its breadth, insecurity is unlikely to have a single cause. The Great Recession, which began in the United States in 2007 and spread around the world, has exacerbated the problem of insecurity, but insecurity was high before the recession set in and will persist after the recession is over (Hacker et al., 2012; Hacker et al., 2013). Indeed, insecurity has a number of interlinked and interacting causes that are deeply embedded in advanced post-industrial economies. De-industrialization (Iversen & Cusack, 2000), technological change (Rehm, 2010; Scheve & Slaughter, 2004), and de-unionization (Rehm, 2010) have all amplified job insecurity. These factors are sometimes lumped together under the title of globalization, but it is helpful to recognize that they are conceptually distinct from, albeit linked to, globalization. Globalization, which refers to the process of international economic integration (Scheve & Slaughter, 2004), can itself drive insecurity in employment by increasing the elasticity of demand for labor (Rehm, 2010; Scheve & Slaughter, 2004) and by exposing labor to exogenous shocks (Rodrik, 1998). The causes of other forms of insecurity—in health care and retirement, for example—are less obvious and perhaps more controversial. Hacker (2008) attributes the rise in insecurity in the United States in these domains to deliberate policy choices by governments and employers designed to privatize risk. Although the government and private-sector employers once pooled and spread risk via “social insurance” from the Great Depression-era onward, this has been steadily dismantled and replaced with private, individual provision in the name of “personal responsibility.” Globalization, de-unionization, technological change, and de-industrialization are likely to have exacerbated the privatization of risk, but the role of government and employers is central; not only was social insurance not reinforced in response to the threats to economic security posed by macro-economic changes, it was deliberately dismantled, argues Hacker (2008; see also Kalleberg, 2013; Standing, 2011).

Despite the contemporary prominence of economic insecurity, and compelling evidence of its growth, we probably know more about the causes of insecurity than its consequences.¹ What is known about the latter suggests that economic insecurity could be an important explanatory variable in political behavior generally and trust research specifically. It has been connected to a range of political attitudes, behavior, and outcomes. Insecurity, and specifically job insecurity, has been shown to influence party support, vote choices, and election results (Marx, 2014; Mughan, Bean, & McAllister, 2004; Mughan & Lacy, 2002), and is associated with the size of, and support for, the welfare state (Cusack, Iversen, & Rehm, 2006; Hacker et al., 2013; Rehm, 2009, 2010; Rehm, Hacker, & Schlesinger, 2012; Rodrik, 1998). Yet, in the round, “we know strikingly little about how the experience of major economic dislocations and worries about them affect . . . attitudes toward the economy and politics,” argue Hacker et al. (2013, p. 24).

This article contributes to this small but growing body of scholarship on the effects of economic insecurity on political phenomena. Its focus is political trust. It reassesses the link between citizens’ personal economic concerns and level of political trust in the face of existing research that has either dismissed the importance of such concerns or relegated them to secondary status behind evaluations of national economic performance. “The Economy and Trust” section of the article reviews what we know and do not know about the economy and trust, and explains why the standard measures of people’s personal economic experiences and worries need to be reconsidered. The “Theorizing and Measuring Economic Insecurity” and “Linking Economic Insecurity and Political Trust” sections briefly introduce some definitional issues and lay out the mechanisms via which insecurity may influence political trust. The article then presents the data and methods used to test the insecurity–trust relationship, before reporting the results of the empirical analysis. The final section discusses the importance, implications, and limitations of the findings.

The Economy and Trust

The connection between economics and trust seems intuitive and straightforward: Citizens trust government during economic good times and distrust it during bad ones. One causal explanation, rooted in democratic theory, is that citizens hold governments to account for their performance, and economic management is a key performance criteria; where economic performance is poor, or at least perceived to be poor, it is reasonable to expect that a government’s political support will decline (Alesina & Wacziarg, 2000; Clarke, Dutt, & Kornberg, 1993; Keele, 2007; Scharpf, 2000).

Economic performance is, however, a broad term that incorporates many different aspects (Clarke, Stewart, & Zuk, 1989). Objective national-level measures of economic performance such as national output, unemployment, and inflation, and objective individual-level measures such as family or per capita income, changes in income, and employment or job status are often utilized in trust models. However, it is also common to use subjective indicators of performance, such as individuals’ perceptions of the wider national economy or their personal financial situation. The vast array of possible indicators complicates the economy–trust story. Which are more important and when?

According to some analyses, trust fell in advanced industrial democracies at the same time as some objective macro measures of economic performance (GDP, inflation, and unemployment, for example) were improving. That objective macro trends do not, on their face, appear to track with political trust suggests the two are not causally related (Dalton, 2004; McAllister, 1999). At the micro level, objective economic indicators also seem to be unimportant, with several scholars reporting that household income has an insignificant effect on political trust (Dalton, 2004; Hetherington, 1998; Lawrence, 1997; Lipset & Schneider, 1983; McAllister, 1999; but see Wroe, 2014).

The general thrust in the literature is that subjective economic evaluations are more important individual-level determinants of political trust than are objective criteria (Citrin, McClosky, Shanks, & Sniderman, 1975; Dalton, 2004; Lawrence, 1997; Lipset & Schneider, 1983; McAllister, 1999). And within the domain of evaluations, the conventional wisdom is that citizens’ perceptions of the performance of the wider economy matter more than citizens’ perceptions of their own or their families’ financial situation (Dalton, 2004; Mishler & Rose, 2001). However, it also appears to be the case that the economy–trust relationship may be asymmetrical. Hetherington and Rudolph’s (2008) time-series analysis demonstrates that trust declines during economically challenging times, but does not increase when perceptions of economic performance improve (see also Clarke & Kornberg, 1989). Overall, it would be incorrect to conclude that subjective assessments of one’s personal or household situation do not matter, but these effects appear to be modest at best (Dalton, 2004, pp. 64-5, 75; McAllister, 1999; Mishler & Rose, 2001).

There are, however, several potential problems with the conclusion that subjective personal economic criteria are unimportant to trust judgments, or of only moderate importance at best. As noted by Hacker et al. (2013), the first was highlighted nearly 30 years ago by Rosenstone, Hansen, and Kinder (1986) and remains relevant today. They argued that

the frail results produced by survey-based analyses reflect in part the frail measurement of changing personal economic well-being. While relatively trustworthy measures of aggregate economic conditions are readily available, the reliability and validity of individual measures of personal financial security are largely untested. (p. 177)

The key culprit in their measurement error story is the standard ANES question, which asks respondents whether they and their families are “financially . . . better off or worse off than you were a year ago.” As well as better off and worse off, respondents are allowed to report no change, resulting in a three-item measure that has been adopted as the standard measure of personal well-being by surveys around the world. However, as Rosenstone et al. (1986) conclude, “pinpointing the connection between economic circumstance and political preference based on this lone item is a perilous enterprise” (p. 177) because it restricts respondents’ response variance and thus underestimates the impact of changing economic circumstances.

A second problem is that these broad measures may fail to pick up the most prominent economic concerns of contemporary citizens. For example, an individual may well answer that she is “better off” financially than a year ago if she has recently been awarded a salary increase or if her partner has entered the workforce, but this same person may still feel economically insecure if her job is at risk or the value of her house has collapsed. As Mughan and Lacy (2002) note, the economy and politics interact in complex ways, and reducing “evaluations to short-term performance judgements alone risks missing important dimensions of people’s perceptions of the economy that help to explain facets of political behaviour for which the economy’s importance may go unnoticed” (p. 533). Indeed, such are the trends in advanced economies that family incomes are increasing at the same time as economic insecurity is growing (Stiglitz, Sen, & Fitoussi, 2009). The story that some scholars tell of an economically satisfied citizenry (Dalton, 2004; Inglehart, 1997a, 1997b) jars with a large and growing literature that points not to gains in wealth and security but to greater economic insecurity (Dynan, Elmendorf, & Sichel, 2012; Hacker, 2008; Hacker et al., 2012; Jacobs, 2007; Osberg, 2009; Rejda & Haley, 2004; Rockefeller Foundation, 2007; Stiglitz et al., 2009). The real economic story is that many people are hurting financially, primarily because they are increasingly insecure.

Theorizing and Measuring Economic Insecurity

As with many concepts in the social sciences, there is no settled definition of economic insecurity and thus no agreement on how it should be measured. The main definitional debate is whether economic insecurity should be conceived as a subjective or objective state. One group of scholars argues that insecurity is best rendered as the subjectively assessed risk of economic hardship (Dominitz & Manski, 1997; Scheve & Slaughter, 2004), whereas another conceives it as the extent to which individuals have actually experienced “hardship-causing economic losses” (Hacker et al., 2012, p. 5). The lack of consensus regarding the definition of economic insecurity has generated a plurality of different measures to capture it (Bossert & D’Ambrosio, 2013; Dynan et al., 2012; Hacker et al., 2012; Jacobs, 2007; Osberg, 2009; Osberg & Sharpe, 2009; Rejda & Haley, 2004; Rockefeller Foundation, 2007, 2008).

Researchers more interested in charting and explaining trends in the real economy generally utilize objective indicators such as income and income volatility (for a review, see Dynan et al., 2012), or more complex multidimensional measures, such as the Economic Security Index (Hacker et al., 2012), that weigh experienced losses in income against families’ ability to buffer the effects of such losses. Those interested in the individual-level psychological, sociological, and political effects of insecurity usually favor subjective risk measures (De Witte, 1999, 2005; Mughan et al., 2004; Mughan & Lacy, 2002; Slovic, 1999; Sverke, Hellgren, & Naswall, 2002, 2006), largely because different people will assess and respond to risk in different ways (Bossert & D’Ambrosio, 2013; Dominitz & Manski, 1997; Jacobs, 2007; Manski, 2004).

Because this study is interested in the political effects of citizens’ perceptions of their personal economic situation, insecurity is taken to be a subjective construct. The article therefore follows Dominitz & Manski (1997) and defines it as individuals’ “perceptions of the risk of economic misfortune” (p. 264). Future work may seek to explore the effect of experienced insecurity on political trust, but insecurity is operationalized here as subjectively perceived risk across a range of domains (employment, health, family, and wealth). It is, though, worth noting that despite the plurality of measures in the literature, four consistent and robust findings emerge regarding insecurity in the United States. Citizens in the United States are more insecure than their counterparts in other wealthy and mature democracies; economic insecurity in the United States increased in the late 20th and early 21st centuries; it did so at a faster rate than in other Organisation for Economic Co-Operation and Development (OECD) nations; and the Great Recession accelerated the increase (Hacker et al., 2012; Hacker et al., 2013; Osberg, 2009; Rockefeller Foundation, 2007, 2008).

Linking Economic Insecurity and Political Trust

One issue only touched on so far is the causal connection between insecurity and trust: Why should economic insecurity influence the extent to which individuals trust politicians and government? A very basic economy–trust mechanism, already established in the literature, is that citizens trust governments that perform well and meet expectations. The economy is a key performance barometer; thus, economic underperformance by government generates lower levels of trust (Alesina & Wacziarg, 2000; Clarke et al., 1993; Keele, 2007; Scharpf, 2000). However, this is a thin causal story. It offers little insights into individual-level cognitive processes, does not distinguish clearly between micro and macro aspects of performance, and ignores the possibility that citizens may make a different type of evaluation when thinking about their personal economic security compared with national conditions. Furthermore, it does not offer any insights as to why economic insecurity may be as or more important a determinant of trust than other types of underperformance, economic or otherwise.

Drawing on Kahneman and Tversky’s (1979) groundbreaking “prospect theory,” Hacker (2008, pp. 25-27) and Hacker et al. (2012, p. 4) have noted the deep psychological foundations of humans’ antipathy toward insecurity. Two aspects of prospect theory are particularly relevant to theorizing insecurity and its effects. First, individuals assess financial risks and outcomes from a particular reference point, which is usually their current position. It is likely therefore that people evaluate potential gains and losses in economic security, real and perceived, from this reference point. Thus, although wealth may act as a buffer to insecurity, everyone is susceptible to insecurity because threats and losses are assessed relative to present circumstances.

Second, and perhaps most important to our understanding of how humans think about financial changes and evaluate risks, is the idea of “loss aversion” (Kahneman, 2011; Kahneman & Tversky, 1979). Of course, the idea that humans are risk averse is hardly new, but Kahneman and Tversky were the first to integrate it successfully in a challenge to the dominant marginal utility theory paradigm. A key premise of marginal utility theory is that people assess gains and losses equally: A dollar loss generates the same unit change (−1) in utility as a dollar gained (+1). Kahneman and Tversky discovered that people weigh losses more heavily than gains. This asymmetry, or negativity bias, may be an evolutionary consequence of natural selection, because “organisms that treat threats as more urgent than opportunities have a better chance to survive and reproduce” (Kahneman, 2011, p. 282; see also McDermott, Fowler, & Smirnov, 2008). Moreover, people quickly grow very attached to items they own, which Kahneman, Knetsch, and Thaler (1990, 1991) labeled an endowment effect. In this thinking, economic insecurity threatens the core human instinct to avert loses and preserve what we already have (Hacker, 2008; Hacker et al., 2012).

Negativity bias has been a prominent feature in the work of psychologists and behavioral economists (Baumesiter, Bratslavsky, Finkenauer, & Vohs, 2001; Rozin & Royzman, 2001) and is gaining more traction among political scientists (for reviews, see Levy, 2003; McDermott, 2004; Mercer, 2005; Soroka, 2014; Wilson, 2011). Attitudes, vote choices, and turnout are swayed more by the negative traits of presidential candidates and political parties than by their positive ones (Holbrook, Krosnick, Visser, Gardner, & Cacioppo, 2001; Kernell, 1977; Lau, 1982), and negative economic conditions play a larger role than positive conditions in determining votes for the incumbent president’s party in U.S. House elections (Bloom & Price, 1975). In part, this may be connected to media coverage of politics and the economy, which is asymmetrically focused on negative events and attributes, but a poorly performing economy has an important and asymmetrical effect on citizens’ sociotropic economic evaluations independent of media coverage (Soroka, 2006). These asymmetries help explain Hetherington and Rudolph’s (2008) finding that political trust falls when economic performance is poor but does not recover when the economy improves. They also dovetail with a new research paradigm that locates political behavior in evolutionary theory (Alford & Hibbing, 2004; Arceneaux, 2007; Smith, Larimer, Littvay, & Hibbing, 2007), and particularly the idea that people do not want to be suckers (Hibbing & Alford, 2004).

So, prospect theory and the wider negativity bias literature provide some important insights as to the reasons for insecurity’s importance, most notably that humans are psychologically programmed to fear it. However, why would Americans blame government for their insecurity, especially given their strong support for small government and economic individualism? One part of the answer is that they advocate these ideas in the abstract, while concomitantly supporting a whole raft of government programs designed to alleviate the insecurities associated with free markets and the general precariousness of the human condition. Americans are, in the words of Free and Cantril (1968), ideological conservatives and operational liberals (see also Feldman & Zaller, 1992).

The second part is that government once provided Americans with a greater level of protection against insecurity, but withdrew many protections at the same time as insecurity was itself increasing due to structural changes in the international economy (Hacker, 2008; Kalleberg, 2013; Standing, 2011). In this thinking, Americans should therefore blame government for their precarity because it stepped aside when it was needed most (Hacker, 2008). It was not inevitable that an “age of anxiety” would replace the “age of security” (Kalleberg, 2013, pp. 24, 86), but a result of deliberative policy choices. It is, then, not surprising that government’s incremental withdrawal from the post-war social contract would generate considerable distrust given the unyielding attachments people develop to social and welfare programs that benefit them (Kahneman et al., 1990, 1991; Pierson, 1994).

Data and Method

To recap, the aim of the article is to test the individual-level linkage between economic insecurity and political trust. A negative relationship is hypothesized: Individuals with high levels of insecurity will have low levels of political trust. The reason is that insecurity activates deep-seated psychological biases, rooted in human evolution and survival, to avert loses and preserve the status quo. Government and the political class are blamed for insecurity, because economic precariousness violates citizens’ expectations about the role of the state.

Data are from the American National Election Studies 2008-2009 panel study. Particularly useful is the Survey of Economic Risk Perceptions and Insecurity, designed by Jacob Hacker, Philipp Rehm, and Mark Schlesinger, which appeared in two waves of the ANES panel (March 2009 and September 2009). It includes extensive questions about respondents’ subjective assessment of the risk of economic insecurity in four specific domains: employment, health care, family, and wealth. Other waves of the survey probe the same respondents’ political attitudes, including political trust, and ask the standard range of demographic, political, and economic questions (see DeBell, Krosnick, & Lupia, 2010, and Hacker et al., 2013, for more details).

Variables and Models

The dependent variable, political trust, is constructed from responses to the following question asked in May 2009 (Wave 17): “How often does the federal government do what most Americans want it to do?” The response options are always, most of the time, about half the time, once in a while, and never. This 5-point scale represents an improvement over the standard ANES trust-in-government question, which has been criticized for its amorphous, unbalanced categories (Cook & Gronke, 2005; Gershtenson & Plane, 2007).² Another way in which the panel question differs from the standard but unasked ANES question is that it does not specifically mention the word trust, but instead taps into one of the central dimensions of trust, which is that government should be responsive to the citizenry (Citrin, 1974; Hetherington, 1998, 2007; Miller, 1974; Miller & Listhaug, 1990). Understanding trust as responsiveness has a long history. In his early work, Miller (1974) conceptualized trust as “the belief that the government is not functioning and producing outcomes in accord with individual expectations” (p. 952), and later concluded that “an expression of trust in government . . . is a summary judgment that the system is responsive and will do what is right even in the absence of constant scrutiny” (Miller & Listhaug, 1990, p. 358). More recently, Hetherington (2007) defined political trust as “the degree to which people perceive that government is producing outcomes consistent with their expectations” (p. 9). These definitions all tap directly into the idea that trust is at least in part about the responsiveness of government.

Some scholars, however, regard responsiveness as a key element of political efficacy, and specifically, external efficacy. To be sure, distinguishing between trust and efficacy, definitionally and operationally, is a long-standing problem for political scientists (Citrin & Muste, 1990). However, efforts to do so sometimes coalesce around seeing efficacy as a system input, and specifically as citizens’ perceptions of their (individual or collective) influence on the system, and trust as a system output, specifically citizens’ perceptions that political agents are responsive to the community’s demands and interests (Gamson, 1968; Mason, House, & Martin, 1985; Reef & Knoke, 1990). Indeed, one of the discipline’s most careful empirical analyses of the trust versus efficacy problem concluded that “the responsiveness of incumbent authorities” to citizens’ demands should be taken as a measure of trust, not external efficacy (Craig, Niemi, & Silver, 1990). In short, and following a rich research tradition, trust as defined and operationalized in this article is about citizens’ perceptions of the responsiveness of the federal government.

The trust variable is coded 1 if the respondent agreed that the federal government does what most Americans want it to do at least half the time (i.e., about half the time, most of the time, or always = 43%) and 0 if the respondent said less than half the time (i.e., once in a while or never = 57%). The trusting category therefore includes the midpoint. The about half the time and once in a while categories together constitute 90% of responses. Only 6% selected most of the time or always, and only 4% said never.³

The key explanatory variables tap respondents’ perceived levels of economic insecurity. This article follows Hacker et al. (2013) and disaggregates perceived insecurity into the four domains of employment, health care, family, and wealth. Wave 15 (March 2009) of the ANES 2008-2009 panel survey asked respondents whether they were very worried, fairly worried, slightly worried, or not at all worried about a range of negative economic outcomes across the four domains.⁴ Four indexes were created to measure the extent to which each respondent was worried about future-oriented insecurity in each domain. Each index is itself constructed from respondents’ average responses to four separate questions, with scores ranging from 1 for the least worried to 4 for the most worried. A fifth, composite index aggregates and averages each individual’s scores from all the questions across the four domains.⁵ Table 1 reports the questions used to construct the indexes, the question means, and index means, the Ns, and Cronbach’s alphas.⁶ The high alpha scores indicate each index is internally consistent and tapping a unidimensional insecurity construct. The distributions of the indexes are presented in Figure 1. The expectation is that higher levels of insecurity will be associated with lower levels of political trust.

Table 1.

Building the Economic Insecurity Indexes.

Index	Question: Are you very worried, fairly worried, slightly worried, or not at all worried about^a	Q. M	Q. N^b	Index M^c	Index α	Index N
Job	J1. Losing your job?	2.13	1,615	2.18	.77	1,188
	J2. Having your retirement benefits cut substantially at your main job?	2.04	2,516
	J3. Having health care coverage substantially cut or its costs substantially increased by your employer?	2.42	1,494
	J4. Becoming unable to work for the rest of your life as the result of a disability?	1.89	2,513
Health	H1. Losing your health care coverage?	2.12	2,327	2.07	.81	2,307
	H2. Having to go into a nursing home when older?	2.06	2,522
	H3. Getting seriously ill and not being able to figure out what your insurance will pay for?	2.15	2,524
	H4. Getting seriously ill and not being able to find the best doctors?	2.02	2,531
Family	F1. Paying your rent/mortgage?	2.12	1,748	2.08	.77	1,195
	F2. Needing to help out a member of your extended family if they get in financial trouble?	2.13	2,521
	F3. Having enough money to put food on the table?	1.84	2,519
	F4. Getting by without your spouse’s/partner’s income if they were no longer around due to death, divorce, or other circumstances?	2.25	1,691
Wealth	W1. Getting out of debt?	2.15	2,513	2.51	.77	875
	W2. Having enough money to retire on?	2.68	2,516
	W3. Paying for your children’s education?	2.50	1,105
	W4. Getting by without your spouse’s/partner’s income if they were no longer around due to death, divorce, or other circumstances?	2.25	1,691
Composite^d	(Job Index + Health Index + Family Index + Wealth Index) / 4			2.42	.92	480

Answers scored 4 if very worried, 3 if fairly worried, 2 if slightly worried, and 1 if not at all worried.

A total of 2,561 respondents started the survey; 2,498 completed successfully.

Index mean is the average of the four questions’ means. Only respondents who answered all four questions are included in the index mean.

The composite index excludes W4 so as not to double count responses, which are already incorporated via the family index.

Figure 1.

Distributions of insecurity indexes.

A number of control variables are also included in the model. Most important are three economic variables. The first (labeled Pktbk Retro in Table 2) taps individuals’ retrospective assessment of their personal economic situation and is based on responses to the following question asked in Wave 19: “During the last 3 months, how difficult has your personal economic situation been—extremely difficult, very difficult, moderately difficult, slightly difficult, or not at all difficult?” This question may represent an improvement on the standard personal finances question criticized by Rosenstone et al. (1986), in part because it offers five rather than three response options but also because it asks more broadly about one’s personal economic situation rather than the narrower financial situation. The second economic variable (labeled Socio Pros) is based on a variant of the standard question about the future health of the wider economy. Respondents were asked the following question in May 2009 (Wave 17): “What about 12 months from now? Do you think the economy, in the country as a whole, will be better, about the same, or worse in 12 months?” As noted above, prospective evaluations of the general economy have been shown to be important predictors of a wide range of political behavior. A third economic variable (labeled Socio Retro) also taps evaluations of the wider economy, but picks up on perceptions of how it performed in the past. It is built from the following question asked in May 2009: “Now thinking about the economy in the country as a whole, would you say that as compared to January 2009, the nation’s economy is now better, about the same, or worse?” In the case of all three economic variables, the hypothesis is that perceived improvements in economic performance, whether pocketbook or sociotropic, will be associated with higher levels of political trust. Moreover, prior trust research suggests that prospective evaluations of the wider economy should be the most powerful predictor and retrospective evaluations of personal economic conditions the least powerful. Retrospective sociotropic evaluations should fall in the middle.

Table 2.

Explaining Political Trust With Perceptions of Economic Insecurity.

	Job insecurity model	Health insecurity model	Family insecurity model	Wealth insecurity model	Composite insecurity model
Insecurity Index (low to high, 4-point scale)	−.294** (.136)	−.158* (.098)	−.515*** (.145)	−.396*** (.152)	−.924** (.415)
Economic: Pktbk Retro (not at all difficult to extremely difficult, 5-point scale)	.072 (.093)	.019 (.069)	.085 (.098)	.065 (.106)	−.040 (.130)
Economic: Socio Pros (better to worse, 3-point scale)	−.389*** (.147)	−.399*** (.112)	−.289** (.137)	−.067 (.155)	−.026 (.215)
Economic: Socio Retro (better to worse, 3-point scale)	−.196 (.149)	−.213* (.111)	−.328** (.136)	−.330** (.160)	−.204 (.210)
Party ID (strong Democrat to strong Republican, 7-point scale)	−.209*** (.063)	−.214*** (.046)	−.115* (.062)	−.111 (.074)	−.201** (.097)
Ideology (extremely liberal to extremely conservative, 7-point scale)	−.104 (.074)	−.069 (.056)	−.103 (.073)	−.099 (.088)	−.047 (.107)
Income (low to high, 19-point scale)	−.010 (.041)	.013 (.025)	−.000 (.039)	.038 (.046)	.036 (.075)
Education (low to high, 5-point scale)	.209** (.109)	.190** (.078)	−.017 (.100)	.158 (.121)	.228 (.169)
Religion (ref = Prot)
Catholic	−.037 (.232)	−.310 (.191)	.042 (.227)	.090 (.254)	.308 (.314)
Jew	.368 (.435)	.141 (.347)	.572 (.568)	.580 (.480)	.892 (.631)
Other	−.325 (.359)	−.412 (.264)	.041 (.324)	.350 (.301)	.167 (.493)
None	−.328 (.272)	−.148 (.240)	.146 (.327)	.310 (.397)	.000 (.467)
Race (ref = White)
Black	−.571 (.399)	−.287 (.329)	−.517 (.457)	.405 (.460)	.032 (.569)
Hispanic	−.034 (.422)	−.126 (.355)	−.210 (.489)	.162 (.525)	.227 (.744)
Other	.484 (.460)	.437 (.458)	.709 (.441)	1.146** (.530)	.978 (.618)
Gender (male)	.257 (.197)	.198 (.148)	.265 (194)	.179 (.232)	.382 (.293)
Age (in years)	.007 (.009)	.005 (.006)	−.004 (.008)	−.007 (.011)	−.004 (.018)
Constant	1.255 (.931)	.938 (.731)	2.514*** (.959)	.966 (1.063)	1.573 (1.963)
n	910	1,726	893	633	364
Pseudo R²	.12	.11	.10	.10	.12
Log likelihood	−483.18	−946.54	−550.06	−381.89	−221.22
Prob > χ²	.000	.000	.000	.000	.001

Note. All models estimated with logistic regression. Coefficients are log-odds (logit) regression coefficients. Standard errors are in parentheses. Dependent variable scored 0-1. See main text for more details of coding schemes for all variables. All equations include weights, but reported n is the unweighted number of cases.

p ≤ .1. **p ≤ .05. ***p ≤ .01 (two-tailed tests).

Although much extant research has suggested that income may not play a significant role in determining political trust (Dalton, 2004; Hetherington, 1998; Lawrence, 1997; Lipset & Schneider, 1983; McAllister, 1999), more recent work has found a relationship (Wroe, 2014). It is anyway important to include income in the present analysis because it can act as a buffer to insecurity, and may thus moderate insecurity’s effect on political trust. Ceteris paribus, higher wages and, by implication, greater wealth help protect citizens against the vagaries of risk (Bossert & D’Ambrosio, 2013; Hacker et al., 2012; Hacker et al., 2013; Rehm, 2009, 2010; Rehm et al., 2012; Rejda & Haley, 2004). Similarly, education should also act as a buffer to insecurity, because it gives more choices and options not available to the uneducated (Sverke et al., 2006; Walter, 2010). Moreover, education has been shown to be an important independent predictor of political trust, with citizens with higher levels of education more trusting than those with lower levels (Claes, Hooghe, & Marien, 2012; D. King, 1997; Schoon, Cheng, Gale, Batty, & Deary, 2010; Wroe, 2014), although this relationship may only hold in mature democracies and the causal mechanisms are complicated and contested (Hooghe, Dassonneville, & Marien, 2015; Van Elsas, 2014). Education is scored on a 5-point scale and income on a 19-point scale.

It is also important to control for party identification and political ideology, acknowledged predictors of trust (Miller, 1974). Republicans and conservatives are generally more hostile to government than Democrats and liberals, but the sign of the coefficients is sensitive to the occupant of the White House and the majority party in Congress (Anderson & Guillory, 1997; Anderson & LoTempio, 2002; Keele, 2005). Citizens are more trusting when their partisan and ideological soul mates are in charge. As responses to this study’s key attitudinal variables in the 2008-2009 panel were sampled during a time of Democratic hegemony in Washington, D.C., the strong expectation is that conservatives and Republicans will be less trusting. Both variables are scored on 7-point scales from strong Democrat to strong Republican and extremely liberal to extremely conservative. Independents and moderates are coded at the scales’ midpoints.⁷

The model also controls for the usual range of demographic variables, including religion, race, gender, and age. Religion is scored as a series of dummies, with Protestant excluded as the reference category. Race is also scored as dummy variables, with White as the reference. Gender is scored 1 for male and 0 for female, and age is coded in years.

In sum, there are five trust models, each testing one aspect of the proposition that political trust is sensitive to citizens’ perceptions of their level of economic insecurity. Although each of the models includes a unique insecurity index variable, all models include identical control variables.⁸

Analysis

Because the dependent variable is dichotomous, a non-linear probability model (logit) is used to estimate the equations. Table 2 presents the log-odds regression estimates and associated statistics. The log likelihoods and chi-square statistics demonstrate that all the models are statistically significant.

It is immediately apparent that perceptions of future-oriented insecurity are correlated with political trust. The insecurity indexes are statistically significant across all five perceptions models. In two models (family insecurity and wealth insecurity), the indexes are significant at p ≤ .01, and in two others at p ≤ .05 (job insecurity and the composite index). The signs are in the hypothesized direction. Americans with high levels of perceived economic insecurity are significantly less trusting than more secure individuals. No other variable reaches statistical significance across all five models.

The size of insecurity’s coefficients differs across the models. It is largest in the composite model, and notably smaller in the health model. The smaller effect for health insecurity may be a consequence of the febrile partisan debate over President Obama’s Affordable Care Act, which coincided with the collection of these data. The suggested mechanism is that partisan polarization on the health care issue crowds out health insecurity’s effect on trust, and partisanship becomes relatively more important. This finding is supported both by the smaller-than-average coefficient for health insecurity and the larger-than-average coefficient for party identification in the health model. The wider point, of course, is that insecurity’s effect on trust is sensitive to different political, social, and economic contexts.

Prospective evaluations of the wider economy are significant in the job, health, and family models but not the wealth or composite models. Retrospective evaluations of the general economy are also significant in three models (health, family, and wealth). Still, it is important to recognize that evaluations of the macro economy’s performance do matter in a majority of models. People who are positive about the wider economy, whether regarding its past or likely future performance, are more trusting than those who are pessimistic. In contrast, citizens’ evaluations of their personal economic situation do not influence political trust in any of the five models. And income, theorized as a buffer to insecurity, is also insignificant across all the models.

Party identification is also a statistically significant predictor of political trust in four of five models, and the sign is in the hypothesized direction in each case. Unsurprisingly, Republicans were more likely than Democrats to be suspicious of the political class at a time of Democratic hegemony in Washington. Surprisingly, ideology is not correlated with political trust at standard levels of statistical significance—although the signs are in the hypothesized direction. Party identification trumps ideology, in these models at least.

Education is statistically significant in two of the models (job insecurity and health). Controlling for other factors, better educated Americans are more trusting than their less educated peers. With only one exception—the vague “other” race category in the wealth model—none of the other demographic variables is statistically significant in any model.

One problem with the log-odds coefficients generated by non-linear regression techniques is that the magnitude of the effects is not easily interpreted. A secondary analysis was therefore conducted using the Clarify add-on to Stata (G. King, Tomz, & Wittenberg, 2000; Tomz, Wittenberg, & King, 2003). The analysis simulated first differences in the expected value of the dependent variable, trust, holding all other variables in the model constant at their means. Each explanatory variable of interest was changed, in the first instance, from its lowest to highest value and the effect of the change on Pr(Y = 0) estimated—that is, the probability of moving from a trusting to an untrusting classification on the dependent variable given the value change on the explanatory variable.

Figure 2 plots the magnitude of the first differences effects and also reports the precise size of the coefficients. The primary variables of interest are the five insecurity indexes in the first bloc of variables on the left-hand side of the figure and accompanying table. Changing the score on, for example, the family insecurity index from its lowest (1) to highest value (4) increases the likelihood of a distrusting outcome by 32%, holding all other variables constant at their means. Put differently, if Americans’ perceptions of their family’s economic situation shifted from very secure to very insecure, political trust would fall by a third. The first difference scores for all the insecurity indexes are large, with none dropping below 10%, and ranging up to 36% on the composite index. Equally notable is that the magnitude of the first differences effects of four of the five insecurity indexes (the health index is the one exception) is larger than the effects of all the retrospective and prospective evaluations of the macro economy, when these change from better to worse. Moreover, the insecurity effects dwarf the (incorrectly signed and statistically insignificant) effects of personal economic evaluations.

Figure 2.

First differences, given change in X from lowest to highest category.

However, a critic could argue that it is unlikely in the real world that individuals would move from being very worried to not at all worried about some aspect of their security, or from being a strong Democrat to a strong Republican. An alternative, more realistic scenario is to explore the effect of a smaller change. One which allows a robust comparison between variables is the effect on Pr(Y = 0) of moving from −1 standard deviation below the mean to +1 standard deviation above the mean for each X of interest. Figure 3 reports and plots these first difference effects. The insecurity indexes’ effects are, of course, smaller than those reported in Figure 2, but remain important in and of themselves and relative to the other explanatory variables. Insecurity has the largest effect of any explanatory variable in three of the five models (family, wealth, and the composite), again outgunning the other economic variables and even party identification. Its effect is smaller in the job and health models, but important nonetheless. Of all the economic variables in the job insecurity model, only prospective evaluations of the macro economy have a larger effect on trust, and in the health insecurity model, its effect is similar to that of retrospective sociotropic evaluations. Moreover, insecurity’s effect dwarfs that of personal economic evaluations in all models.

Figure 3.

First differences, given change in X from −1 SD to +1 SD.

In sum, the results are stark. The extent to which Americans perceive their economic situation to be insecure, across a range of domains, has an important and large effect on their trust in the political class. This effect is, in the round, larger for economic insecurity than the three other economic variables. The only other variable to match economic insecurity in terms of the magnitude of its effect on political trust is party identification, an unsurprising result given that the Democrats controlled both the White House and Capitol Hill and a deeply partisan and polarized atmosphere dominated DC’s political proceedings (Mann & Ornstein, 2012) when the trust responses were solicited in the panel survey. However, despite this febrile Washington atmosphere, perceptions of economic insecurity are at least as important as party identification in determining Americans’ level of trust.

Discussion

Americans who perceive themselves as economically insecure have lower levels of political trust than their more secure peers. And the difference is large. The first differences analysis—whether moving from the lowest to highest value or from −1 standard deviation below the mean to 1 standard deviation above the mean on each explanatory variable—shows that insecurity has the largest effect on trust of any variable in three of the five models. Insecurity at least matches, and by some measures trumps, evaluations of the wider economy’s future performance and party identification. Insecurity is considerably more important than citizens’ self-assessment of their recent economic circumstances. The latter result is perhaps not surprising given the findings of prior trust research and the criticisms of the questions used to tap citizens’ personal finances (Rosenstone et al., 1986), although the present analysis utilizes a broader question that asked about individuals’ economic, not financial, situation and allowed a more fine-grained response. The inference, however, is not that personal economic evaluations do not matter—clearly they do, because perceived insecurity is one example of a personal economic evaluation—but that the traditional questions used by political scientists do not measure well the basic economic concerns of citizens in advanced democracies. We need to think more clearly about the different types of egocentric economic factors that may plausibly influence political behavior and outcomes, and then devise carefully tailored instruments to better measure the concepts we want them to measure. Without better instruments, political scientists will not be able to tell an accurate and convincing story about how our part of the world works. Recent improvements in statistical techniques and computing power are rendered redundant in the face of basic measurement error.

Drawing heavily on the innovative work of Kahneman and Tversky and Hacker and colleagues, the article sketched out a psychologically grounded causal story to explain why economic insecurity may influence the way in, and extent to, which citizens evaluate their leaders. To be sure, the empirical analysis, like much social science work, was not able to test these mechanisms directly (Hedstrom, 2008), but the causal story is consistent with the empirical results. Key to the story is that economic insecurity activates deep-seated psychological biases to avert loses and maintain the status quo. Politicians are blamed by citizens for their insecurity, because insecurity violates citizens’ expectations that the state will provide security. Insecure individuals do not trust government. Most extant theory connecting trust to economic factors has relied on a broad-brush story about government performance. The theory outlined here provides a more nuanced causal mechanism, which identifies why economic insecurity specifically, as opposed to other types of economic hardship such as inequality or recession, may drive political distrust. It also fits into an increasingly important theme of political science research that stresses the importance of negativity bias (Soroka, 2014).

Although the article did not seek to assess the contribution of economic insecurity to the ebb and flow of aggregate levels of political trust over time, instead concentrating on individual-level trust, it is at least plausible that insecurity may account for some of the temporal variation. Future work may wish to explore this in more detail. Although the downward trend in trust observed in the 1960s precedes the rise in insecurity, the post-1960s trend broadly fits. Indeed, insecurity appears, at least on its face, to fit better than the usual aggregate economic measures such as GDP, unemployment, and consumer sentiment.

The article, however, is limited in some respects, as are all articles. One potential limitation is that the categorization of the indexes is somewhat arbitrary, despite the high alphas indicating internal consistency. It would be possible to construct a different set of indexes picking up on different aspects of insecurity, which may generate somewhat different results. In the event, the analysis followed so far as it could the structure and measurement properties set out in Hacker et al.’s (2013) article and Hacker’s earlier single-authored and joint work, to generate continuity in the growing but still small economic insecurity research portfolio. Still, other researchers may disagree with this categorization of economic insecurity and offer alternative ways of operationalizing its various elements. This should be welcomed, for it will increase our understanding of the concept of economic insecurity and its political importance. Another limitation is that the article focused on perceived insecurity and did not explore the effect of experienced or realized insecurity. Above and beyond the limitations of space, one reason for not doing so was that it was not possible to construct internally consistent and unidimensional indexes with the available data. Future work may seek to address this omission.

A further potential limitation is that the article relies on data sampled during a period of great economic turmoil. The asymmetry hypothesis suggests that insecurity’s effect will atrophy as the economy improves. However, although it is true that the Great Recession encapsulates a period of great economic uncertainty and although insecurity increased during the recession, insecurity was high before the recession’s onset and was very probably increasing from the 1970s onward (Hacker et al., 2012; Hacker et al., 2013; Jacobs, 2007; Kalleberg, 2013; Rockefeller Foundation, 2007). That economic insecurity has been deeply embedded in American society for some time and likely will be for some time to come suggests strongly that the results generated by the empirical analysis are not sui generis but are instead the new normal in the socioeconomic and political life of the United States. Nonetheless, additional work is required to examine insecurity’s effects in a more buoyant economy and over an extended period of time.

It would also be instructive to explore its effects during periods of unified Republican control of the presidency and Congress and during divided government. The analysis indicated that health insecurity’s effect on trust was moderated by the partisan rancor surrounding the passage of the Affordable Care Act; different political and social contexts may also moderate insecurity’s influence in interesting ways. In addition, future work may seek to estimate economic insecurity’s effect on other, more diffuse aspects of political support (Easton, 1965, 1975; Norris, 1999). We know that job insecurity in Europe is correlated with lower levels of trust in the judiciary and satisfaction with the operation of democracy (Wroe, 2014), but we do not know whether these findings hold in the United States and with other types of insecurity. Future work could also extend the analysis to cover other types of economic concerns, with inequality a primary candidate.

A final but related limitation is that the results here may not be applicable outside the United States. Many observers have commented on America’s exceptional nature (Hartz, 1955; Lipset, 1997; Rose, 1989; Tocqueville, 2000). One pertinent aspect of this is the relatively small size of its government and, relatedly, the presence of only the most basic government-funded welfare safety net compared with other mature Western democracies with advanced industrial and post-industrial economies. In this thinking, American citizens are more insecure and, by extension, more distrusting because government does not protect them from the vagaries of life. Comparative empirical work has confirmed that Americans are more, and becoming more, insecure than citizens in other OECD democracies (Osberg, 2009; Osberg & Sharpe, 2009). Moreover, the United States’s unique political structure and limited welfare and social provisions suggest that at least some of this article’s results will not travel. Health is an obvious example. Despite recent reforms, many millions of Americans lack any health insurance and millions more are poorly and/or expensively insured. It is difficult to imagine that health insecurity’s effect would be as large in, say, the United Kingdom or France, where citizens enjoy more comprehensive protection. Conversely, globalization and the Great Recession have increased job insecurity across many Western democracies, not just in the United States. A priori, then, we would expect smaller cross-national differences on jobs than on health. However, that the insecurity–trust relationship is robust even in the exceptional, hard case of the United States, where citizens revere rugged individualism, suggests that it may be robust elsewhere too. However, more comparative research is needed to explore these and other relationships; we know surprisingly little hitherto.

In sum, and despite these limitations, the article offers some unambiguous and noteworthy conclusions regarding the importance of perceived economic insecurity on citizens’ political trust. This finding improves our understanding of how trust is generated and lost, but it also supports those scholars who are calling for a reassessment of the personal in the political. For many years, political scientists have downplayed the influence of personal economic evaluations and circumstances on political phenomena. Hacker and his colleagues (2013) are right that it is time to bring them back in: “Practitioners of American politics know that the search for security is a powerful motive, whether harnessed for good or ill. Scholars of American politics have been much slower to recognize this reality. This neglect, one regrettable, is now untenable” (p. 45).

Footnotes

Acknowledgements

The author would like to thank participants at the U.K. Political Studies Association American Politics Group 2014 annual conference at the University of Oxford and members of the University of Kent Comparative Politics Seminar for their helpful comments on earlier drafts of this article. Particular thanks are also due to Amanda Gosling and the APR’s editor and anonymous reviewers for their incisive and pertinent comments and helpful suggestions.

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) received no financial support for the research, authorship, and/or publication of this article.

Notes

Author Biography

Andrew Wroe is a lecturer in American politics in the School of Politics and International Relations at the University of Kent in the United Kindgom. Current research interests include the causes and consequences of political trust, the political effects of economic insecurity and inequality, and political polarization in the USA. His other recent contributions on political trust have appeared in the Journal of Trust Research and the European Political Science Review.

References

Alesina

Wacziarg

(2000). The economics of civic trust. In Pharr

S. J.

Putnam

R. D.

(Eds.), Disaffected democracies: What’s troubling the trilateral countries? (pp. 149-170). Princeton, NJ: Princeton University Press.

Alford

J. R.

Hibbing

J. R.

(2004). The origin of politics: An evolutionary theory of political behavior. American Political Science Review, 2, 707-723.

American National Election Studies. (2008-2009). Panel Study (dataset). Stanford University and the University of Michigan (producers and distributors). Available from www.electionstudies.org

Anderson

C. J.

Guillory

C. A.

(1997). Political institutions and satisfaction with democracy: A cross-national analysis of consensus and majoritarian systems. American Political Science Review, 91, 66-81.

Anderson

C. J.

LoTempio

A. J.

(2002). Winning, losing and political trust in America. British Journal of Political Science, 32, 335-351.

Arceneaux

(2007). Cognitive biases and the strength of political arguments. American Journal of Political Science, 56, 271-285.

Ashford

S. J.

Lee

Bobko

(1989). Content, causes, and consequences of job insecurity: A theory-based measure and substantive test. Academy of Management Journal, 32, 803-829.

Baumesiter

R. F.

Bratslavsky

Finkenauer

Vohs

K. D.

(2001). Bad is stronger than good. Review of General Psychology, 5, 323-370.

Bloom

H. S.

Price

H. D.

(1975). Voter response to short-run economic conditions: The asymmetric effect of prosperity and recession. American Political Science Review, 69, 1240-1254.

10.

Bossert

D’Ambrosio

(2013). Measuring economic insecurity. International Economic Review, 54, 1017-1030.

11.

Brehm

Rahn

(1997). Individual-level evidence for the causes and consequences of social capital. American Journal of Political Science, 41, 999-1023.

12.

Chung

Mau

(2014). Subjective insecurity and the role of institutions. Journal of European Social Policy, 24, 303-318.

13.

Citrin

(1974). Comment: The political relevance of trust in government. American Political Science Review, 68, 973-988.

14.

Citrin

McClosky

Shanks

J. M.

Sniderman

P. M.

(1975). Personal and political sources of political alienation. British Journal of Political Science, 5, 1-31.

15.

Citrin

Muste

(1990). Trust in government. In Robinson

J. P.

Shaver

P. R.

Wrightsman

L. S.

(Eds.), Measures of Social Psychological Attitudes: Vol. 2. Measures of political attitudes (pp. 465-532). San Diego, CA: Academic Press.

16.

Claes

Hooghe

Marien

(2012). A two-year panel study among Belgian late adolescents on the impact of school environment characteristics on political trust. International Journal of Public Opinion Research, 24, 208-224.

17.

Clarke

H. D.

Dutt

Kornberg

(1993). The political economy of attitudes toward polity and society in western European democracies. Journal of Politics, 55, 998-1021.

18.

Clarke

H. D.

Kornberg

(1989). Public reactions to economic performance and political support in contemporary liberal democracies: The case of Canada. In Clarke

H. D.

Stewart

M. C.

Zuk

(Eds.), Economic decline and political change (pp. 253-279). Pittsburgh, PA: University of Pittsburgh Press.

19.

Clarke

H. D.

Stewart

M. C.

Zuk

(1989). Economic decline and political change: Canada, Great Britain, the United States. Pittsburgh, PA: University of Pittsburgh Press.

20.

Cook

T. E.

Gronke

(2005). The skeptical American: Revisiting the meanings of trust in government and confidence in institutions. Journal of Politics, 67, 784-803.

21.

Craig

S. C.

Niemi

R. G.

Silver

G. E.

(1990). Political efficacy and trust: A report on the NES pilot study items. Political Behavior, 12, 289-314.

22.

Cusack

Iversen

Rehm

(2006). Risks at work: The demand and supply sides of government redistribution. Oxford Review of Economic Policy, 22, 365-389.

23.

Dalton

R. J.

(2004). Democratic challenges, democratic choices: The erosion of political support in advanced industrial democracies. Oxford, UK: Oxford University Press.

24.

DeBell

Krosnick

J. A.

Lupia

(2010). Methodology report and user’s guide for the 2008-2009 ANES panel study. Available from www.electionstudies.org

25.

De Witte

(1999). Job insecurity and psychological well-being: Review of the literature and exploration of some unresolved issues. European Journal of Work & Organizational Psychology, 8, 155-177.

26.

De Witte

(2005). Job insecurity: Review of the international literature on definitions, prevalence, antecedents and consequences. SA: Journal of Industrial Psychology, 31(4), 1-6.

27.

Dominitz

Manski

C. F.

(1997). Perceptions of economic insecurity: Evidence from the survey of economic expectations. The Public Opinion Quarterly, 61, 261-287.

28.

Dynan

K. E.

Elmendorf

D. W.

Sichel

D. E.

(2012). The evolution of household income volatility. Retrieved from http://ssrn.com/abstract=2138990

29.

Easton

(1965). A framework for political analysis. Englewood Cliffs, NJ: Prentice Hall.

30.

Easton

(1975). A reassessment of the concept of political support. British Journal of Political Science, 5, 435-457.

31.

Feldman

Zaller

(1992). The political culture of ambivalence: Ideological responses to the welfare state. American Journal of Political Science, 36, 268-307.

32.

Free

L. A.

Cantril

(1968). The political beliefs of Americans: A study of public opinion. New York, NY: Simon & Shuster.

33.

Gamson

W. A.

(1968). Power and discontent. Homewood, IL: Dorsey.

34.

Gershtenson

Plane

D. L.

(2007, April 10). Trust in government: 2006 American National Election Studies pilot report. Retrieved from http://www.electionstudies.org/resources/papers/Pilot2006/nes011890.pdf.

35.

Hacker

J. S.

(2008). The great risk shift: The new economic insecurity and the decline of the American dream. New York, NY: Oxford University Press.

36.

Hacker

J. S.

Huber

G. A.

Nichols

Rehm

Schlesinger

Valletta

R. G.

Craig

(2012). The Economic Security Index: A new measure for research and policy analysis (Federal Reserve Bank of San Francisco Working Paper Series, Working Paper 2012-21). Retrieved from http://www.frbsf.org/economic-research/files/wp12-21bk.pdf.

37.

Hacker

J. S.

Rehm

Schlesinger

(2013). The insecure American: Economic experiences, financial worries, and policy attitudes. Perspectives on Politics, 11, 23-49.

38.

Hartz

(1955). The liberal tradition in America: An interpretation of American political thought since the revolution. San Diego, CA: Harcourt, Brace.

39.

Hedstrom

(2008). Studying mechanisms to strengthen causal inferences in quantitative research. In Box-Steffensmeier

J. M.

Brady

H. E.

Collier

(Eds.), The Oxford handbook of political methodology (pp. 319-337). Oxford, UK: Oxford University Press.

40.

Hetherington

M. J.

(1998). The political relevance of political trust. American Political Science Review, 92, 791-808.

41.

Hetherington

M. J.

(1999). The effect of political trust on the presidential vote, 1968-96. American Political Science Review, 93, 311-326.

42.

Hetherington

M. J.

(2007). Why trust matters: Declining political trust and the demise of American liberalism. Princeton, NJ: Princeton University Press.

43.

Hetherington

M. J.

Husser

J. A.

(2012). How trust matters: The changing relevance of political trust. American Journal of Political Science, 56, 312-325.

44.

Hetherington

M. J.

Rudolph

T. J.

(2008). Priming, performance, and the dynamics of political trust. Journal of Politics, 70, 498-512.

45.

Hibbing

J. R.

Alford

J. R.

(2004). Accepting authoritative decisions: Humans as wary cooperators. American Journal of Political Science, 48, 62-76.

46.

Holbrook

A. L.

Krosnick

J. A.

Visser

P. S.

Gardner

W. L.

Cacioppo

J. T.

(2001). Attitudes toward presidential candidates and political parties: Initial optimism, inertial first impressions, and a focus on flaws. American Journal of Political Science, 45, 930-950.

47.

Hooghe

Dassonneville

Marien

(2015). The impact of education on the development of political trust: Results from a five-year panel study among late adolescents and young adults in Belgium. Political Studies, 68, 123-141.

48.

Inglehart

(1997a). Modernization and postmodernization: Cultural, economic, and political change in 43 societies. Princeton, NJ: Princeton University Press.

49.

Inglehart

(1997b). Postmaterialist values and the erosion of institutional authority. In Nye

J. S.

Zelikow

P. D.

King

D. C.

(Eds.), Why people don’t trust government (pp. 217-236). Cambridge, MA: Harvard University Press.

50.

Iversen

Cusack

T. R.

(2000). The causes of welfare state expansion: Deindustrialization or globalization? World Politics, 52, 313-349.

51.

Jacobs

(2007). The politics of economic insecurity (Issues in Governance Studies, No. 10). Washington, DC: The Brookings Institution.

52.

Kahneman

(2011). Thinking fast and slow. London, England: Penguin Books.

53.

Kahneman

Knetsch

J. L.

Thaler

R. H.

(1990). Experimental tests of the endowment effect and the Coase theorem. Journal of Political Economy, 98, 1325-1348.

54.

Kahneman

Knetsch

J. L.

Thaler

R. H.

(1991). Anomalies: The endowment effect, loss aversion, and the status quo bias. Journal of Economic Perspective, 5, 193-206.

55.

Kahneman

Tversky

(1979). Prospect theory: An analysis of decision under risk. Econometrica, 47, 263-291.

56.

Kalleberg

A. L.

(2013). Good jobs, bad jobs: The rise of polarized and precarious employment systems in the United States, 1970s to 2000s. New York, NY: Russell Sage.

57.

Karp

J. A.

(1995). Explaining public support for legislative term limits. The Public Opinion Quarterly, 59, 373-391.

58.

Keele

(2005). The authorities really do matter: Party control and trust in government. Journal of Politics, 67, 873-886.

59.

Keele

(2007). Social capital and the dynamics of trust in government. American Journal of Political Science, 51, 241-254.

60.

Kernell

(1977). Presidential popularity and negative voting: An alternative explanation of the midterm congressional decline of the President’s party. American Political Science Review, 71, 44-66.

61.

King

(1997). The polarization of American parties and mistrust of government. In Nye

J. S.

Zelikow

P. D.

King

D. C.

(Eds.), Why people don’t trust government (pp. 155-178). Cambridge, MA: Harvard University Press.

62.

King

Tomz

Wittenberg

(2000). Making the most of statistical analyses: Improving interpretation and presentation. American Journal of Political Science, 44, 347-361.

63.

Lau

(1982). Negativity in political perception. Political Behavior, 4, 353-377.

64.

Lawrence

R. Z.

(1997). Is it really the economy, stupid? In Nye

J. S.

Zelikow

P. D.

King

D. C.

(Eds.), Why people don’t trust government (pp. 111-132). Cambridge, MA: Harvard University Press.

65.

Levy

J. S.

(2003). Applications of prospect theory to political science. Synthese, 135, 215-241.

66.

Lipset

S. M.

(1997). American exceptionalism: A double-edged sword. New York, NY: W.W. Norton.

67.

Lipset

S. M.

Schneider

(1983). Confidence gap: Business, labor, and government in the public mind. New York, NY: The Free Press.

68.

Mann

T. E.

Ornstein

N. J.

(2012). It’s even worse than it looks: How the American constitutional system collided with the new politics of extremism. New York, NY: Basic Books.

69.

Manski

C. F.

(2004). Measuring expectations. Econometrica, 72, 1329-1376.

70.

Marx

(2014). Labour market risks and political preferences: The case of temporary employment. European Journal of Political Research, 53, 136-159.

71.

Mason

W. M.

House

J. S.

Martin

S. S.

(1985). On the dimensions of political alienation in America. American Sociological Methodology, 15, 111-151.

72.

McAllister

(1999). The economic performance of governments. In Norris

(Ed.), Critical citizens: Global support for democratic governance (pp. 188-203). New York, NY: Oxford University Press.

73.

McDermott

(2004). Prospect theory in political science: Gains and losses from the first decade. Political Psychology, 25, 289-312.

74.

McDermott

Fowler

J. H.

Smirnov

(2008). On the evolutionary origin of prospect theory preferences. Journal of Politics, 70, 335-350.

75.

Mercer

(2005). Prospect theory and political science. Annual Review of Political Science, 8, 1-21.

76.

Miller

A. H.

(1974). Political issues and trust in government: 1964-70. American Political Science Review, 68, 951-972.

77.

Miller

A. H.

Listhaug

(1990). Political parties and confidence in government: A comparison of Norway, Sweden and the United States. British Journal of Political Science, 20, 357-386.

78.

Mishler

Rose

(2001). What are the origins of political trust? Testing institutional and cultural theories in post-communist societies. Comparative Political Studies, 34, 30-62.

79.

Mughan

Bean

McAllister

(2004). Economic globalization, job insecurity and the populist reaction. Electoral Studies, 22, 617-633.

80.

Mughan

Lacy

(2002). Economic performance, job insecurity and electoral choice. British Journal of Political Science, 32, 513-533.

81.

Norris

(1999). Critical citizens: Global support for democratic governance. New York, NY: Oxford University Press.

82.

Osberg

(2009). Measuring economic security in insecure times: New perspective, new events, and the index of economic well-being (CSLS Research Report 2009-12). Ottawa, Ontario: Centre for the Study of Living Standards.

83.

Osberg

Sharpe

(2009). New estimates of the index of economic well-being for selected OECD countries, 1980–2007 (CSLS Research Report 2009-11). Ottawa, Ontario: Centre for the Study of Living Standards.

84.

Pierson

(1994). Dismantling the welfare state? Reagan, Thatcher, and the politics of retrenchment. Cambridge, UK: Cambridge University Press.

85.

Reef

M. J.

Knoke

(1990). Political alienation and efficacy. In Robinson

J. P.

Shaver

P. R.

Wrightsman

L. S.

(Eds.), Measures of Social Psychological Attitudes: Vol. 2. Measures of political attitudes (pp. 413-461). San Diego, CA: Academic Press.

86.

Rehm

(2009). Risks and redistribution: An individual-level analysis. Comparative Political Studies, 42, 855-881.

87.

Rehm

(2010). Risk inequality and the polarized American electorate. British Journal of Political Science, 41, 363-387.

88.

Rehm

Hacker

J. S.

Schlesinger

(2012). Insecure alliances: Risk, inequality, and support for the welfare state. American Political Science Review, 106, 386-406.

89.

Rejda

Haley

(2004). Construction of an economic index to measure the causes of economic insecurity. Forum for Social Economics, 34(1-2), 9-30

90.

Rockefeller Foundation. (2007). American workers survey: Complete results. Available from www.rockefellerfoundation.org

91.

Rockefeller Foundation. (2008). The Rockefeller Foundation/Time campaign for American workers survey: Executive summary. Available from www.rockefellerfoundation.org

92.

Rodrik

(1998). Why do more open economies have bigger governments? Journal of Political Economy, 106, 997-1032.

93.

Rose

(1989). How exceptional is the American political economy? Political Science Quarterly, 104, 91-115.

94.

Rosenstone

S. J.

Hansen

J. M.

Kinder

D. R.

(1986). Measuring change in personal well-being. The Public Opinion Quarterly, 50, 176-192.

95.

Rozin

Royzman

E. B.

(2001). Negativity bias, negativity dominance, and contagion. Personality and Social Psychology Review, 5, 296-320.

96.

Scharpf

F. W.

(2000). Interdependence and democratic legitimation. In Pharr

S. J.

Putnam

R. D.

(Eds.), Disaffected democracies: What’s troubling the trilateral countries? (pp. 101-120). Princeton, NJ: Princeton University Press.

97.

Scheve

Slaughter

M. J.

(2004). Economic insecurity and the globalization of production. American Journal of Political Science, 48, 662-674.

98.

Scholz

J. T.

Lubell

(1998). Trust and taxpaying: Testing the heuristic approach to collective action. American Journal of Political Science, 42, 398-417.

99.

Schoon

Cheng

Gale

C. R.

Batty

G. D.

Deary

I. J.

(2010). Social status, cognitive ability, and educational attainment as predictors of liberal social attitudes and political trust. Intelligence, 38, 144-150.

100.

Slovic

(1999). Trust, emotion, sex, politics, and science: Surveying the risk-assessment battlefield. Risk Analysis, 19, 689-701.

101.

Smith

K. B.

Larimer

C. W.

Littvay

Hibbing

J. R.

(2007). Evolutionary theory and political leadership: Why certain people do not trust decision makers. Journal of Politics, 69, 285-299.

102.

Soroka

(2006). Good news and bad news: Asymmetric responses to economic information. Journal of Politics, 68, 372-385.

103.

Soroka

(2014). Negativity in democratic politics: Causes and consequences. New York, NY: Cambridge University Press.

104.

Standing

(2011). The precariat: The new dangerous class. London, England: Bloomsbury Academic.

105.

Stiglitz

Sen

Fitoussi

(2009). Report by the commission on the measurement of economic performance and social progress. Available from www.stiglitz-sen-fitoussi.fr

106.

Sverke

Hellgren

Naswall

(2002). No security: A meta-analysis and review of job insecurity and its consequences. Journal of Occupational Health Psychology, 7, 242-264.

107.

Sverke

Hellgren

Naswall

(2006). Job security: A literature review. Stockholm, Sweden: National Institute for Working Life.

108.

Tocqueville

(2000). Democracy in America. Chicago, IL: University of Chicago Press.

109.

Tomz

Wittenberg

King

(2003). Clarify: Software for interpreting and presenting statistical results. Available from http://gking.harvard.edu

110.

Tyler

T. R.

(1990). Why people obey the law. New Haven, CT: Yale University Press.

111.

Van Elsas

(2014). Political trust as a rational attitude: A comparison of the nature of political trust across different levels of education. Political Studies. Retrieved from http://onlinelibrary.wiley.com/doi/10.1111/1467-9248.12148/full.

112.

Walter

(2010, November 12-13). Globalization and the demand-side of politics: How globalization shapes individual perceptions of labor market risk and policy preferences. Paper presented to the 2010 IPES Conference, Harvard University, Cambridge, MA.

113.

Wilson

R. K.

(2011). The contribution of behavioral economics to political science. Annual Review of Political Science, 14, 201-223.

114.

Wroe

(2014). Political trust and job insecurity in 18 European polities. Journal of Trust Research, 4, 90-112.

115.

Wroe

Allen

Birch

(2013). The role of political trust in conditioning perceptions of corruption. European Political Science Review, 5, 175-195.