The Topic: Segmentation

Developed countries throughout the world are experiencing population aging characterized by unprecedented national proportions of older persons (see Table 3.1). This phenomenon has brought to the fore various policy reforms regarding benefits for older persons that have long been provided by established old-age welfare states (see Kohli and Arza 2010). It has also increasingly brought the attention of politicians to strategies for marketing themselves to older voters in the context of both policy decisions and election campaigns.

This chapter focuses on political marketing in aging democracies in the contexts of what we know about older voters and campaigns to attract them, in the context of the special emphasis on segmentation within political marketing theory. It begins with a brief exposition regarding segmentation in political marketing and an assessment of the strengths and weaknesses of the ‘senior power’ model of voting. It then reviews the recent research literature on segmentation and marketing, with attention to issues related to older age groups. Next, it presents case studies of segmenting and marketing to older voters in the US and Britain. There follows a discussion of the importance of attention to life stages, the life cycle, and concomitant values in marketing to older populations. Three final sections suggest: first, some implications of this chapter for practitioners; second, the impact on politics in general of efforts to market to senior voters; and third, future considerations for those who segment and market politically to older voters.

Previous Research on Segmentation

Parties and candidates segment the electorate – the process of defining and targeting identified sub-sections of voters – in their search for competitive advantages over their opponents. Segmentation allows a more efficient allocation of communication resources and is an increasingly sophisticated dimension to the most basic of campaign objectives – the need to identify target audiences and then get them out to vote. As such segmentation represents a key element in the wider adoption of strategic communications and marketing in the campaigns and elections process, the incorporation of these principles has been recorded and theorized widely (Kavanagh 1995; Baines 1999; Newman 1999; Smith and Hirst 2001; Wring 2005).

Increasingly, campaigns bypass first-order segmentation such as gender and class and concentrate on second-order variables such as media habits or lifestyle choices (Butler and Harris 2009). While there may be many different approaches to segmenting a market, Bannon (2004) identified two overarching approaches. First, a priori, which in this context is the utilization of prior political intelligence, such as the knowledge that seniors are more likely to vote than younger voters. This historical intelligence is then combined with cluster analysis techniques that search for common traits such as behavior or attitudes that are shared by sub-groups, but may cut across more general demographic categorizations. Within these processes Bannon then identified four common methods for identifying sub-segments of the electorate:

• Geographic: Voters with similar characteristics tend to congregate in the same geographic location.
• Demographic: Age, gender and family status are all variables that could indicate potential political preferences.
• Behavioral: Grouping voters based on their actual actions, such as the benefits they may seek from a government.
• Psychographic: The development of segments through combining data on lifestyle choices such as leisure pursuits, media habits, etc., with social attitudes and dispositions towards candidates and parties.

Mattinson has described how in 2009 the British Labour Party adopted a segmentation process that began first with identifying voters who were most likely to swing to or from the party, and then analyzed the demographics, attitudes and lifestyles of these voters to ascertain who were the ‘winnable’ or most ‘persuadable’ segments (Mattinson 2010: 16). In the US the parties are combing their proprietary information in combination with consumer or demographic data from companies such as Experian or Claritas to refine their segments into smaller units for micro-targeting (Johnson 2011).

These trends have raised a number of concerns that segmented groups of voters achieve a privileged status within the political process at the expense of politics representing the public as a whole. Savigny (2005) argues that segmentation elevates the potential of a minority of voters to disproportionately influence political actors, and that this process takes place through research instruments such as focus groups rather than through open public engagement. Lilleker (2005) has described this process as creating a division between those to whom politics belongs, and those it has abandoned, and can be responsible for alienating voters, including those who once considered themselves partisan loyalists.

Demographic trends in combination with the increasing use of segmentation have influenced the tone and direction of debate on the impact of the growing number of older voters. Implicit in concerns about ‘the gray peril’ is a ‘senior power’ model for interpreting the politics of aging. The model rests on the assumption – influenced by traditional economic theory – that older persons are likely to vote to preserve or enhance their material self-interests. The model also assumes: that older people constitute a numerically important proportion of the electorate; that all or most of them perceive their stakes in old-age benefits similarly (regardless of their diverse economic and social situations); and that because of material self-interest older people are homogeneous in political attitudes and voting behavior and will thereby clash sharply with younger age groups in the electoral process. The senior power model also includes the notion that interest groups that purport to represent older people are influential forces that can ‘swing’ the votes of older persons and thereby ‘intimidate’ politicians (for example, see Pratt 1976).

Various analysts use different age ranges when defining older voters. US analysts customarily use those aged 65 and older because 66 is the age at which individuals become eligible for full retirement benefits under social security and 65 the age for getting Medicare, the national health insurance program for older people. In contrast, many European analysts include persons aged in their 50s. This is justified on several grounds. First, people in their 50s start to personally experience the many manifestations of age discrimination in society, most critically in employment. Moreover, persons aged 55 and older have reached a stage in their life courses where they have to consider retirement and ‘old-age’ as issues requiring practical, sometimes urgent, personal attention. In addition, the family structure of many in their late 50s and early 60s is likely to include parents who are in their 70s or 80s, heightening awareness among the former regarding issues involving the quality of health and social care provision for those in later life.

Regardless of which age categories are used, many of the senior power model’s assumptions regarding older voters are contradicted by the following facts and observations: Although older persons participate in elections at a higher rate than younger voters, they are not necessarily the largest age group in the electorate. In the 2008 US presidential election, for example, Americans aged 45–64 cast 38 percent of the vote and those aged 25–44 accounted for 36 percent, compared with only 16 percent by people aged 65 and older (Campbell and Binstock 2011). Despite election campaign efforts to target older voters with ‘senior issues’ and ‘senior desks,’ old-age benefit issues do not seem to have much impact on their electoral choices; as shown in Figure 3.1, in the last 10 US presidential elections, all age groups except the youngest (aged 18–29) distributed their votes among candidates in roughly the same proportions.

Figure 3.1   Percent Voting for Republican US Presidential Candidates, by Age Groups, 1972–2008

Old age is only one of many personal characteristics of older people with which they may identify themselves; there is little reason to expect that a birth cohort – diverse in economic and social status, labor force participation, gender, race, ethnicity, religion, education, health status, family status, residential locale, political party attachments and every other characteristic in society – would suddenly become homogenized in self-interests and political behavior when it reaches the old-age category.

Candidates are on the ballot, not old-age benefit policies; candidates usually identify themselves with their political parties, as well as a broad range of issue positions of which old-age benefits may be only one of many. Older voters, like all voters, respond to a variety of candidate traits, such as their personalities, appearances, career backgrounds, performances and even religions, ethnicity and race. In the 2008 US presidential election, for instance, all age groups of whites aged 30 and older voted heavily in favor of John McCain over Barack Obama, in contrast with African-Americans and Hispanics (Binstock 2009). For fuller discussions and documentation of these and related matters involving the voting behavior of older persons, see Campbell and Binstock (2011).

Nonetheless, for several reasons the image of older persons as bloc voters swayed by ‘senior issues’ persists. First, it helps journalists to reduce the intricate complexities of politics. Second, and more important, politicians share the widespread perception that there is a huge, monolithic senior citizen army of voters (Peterson and Somit 1994). This perception is reinforced by the fact that a great many older citizens are active in making their views known to members of their legislatures, especially when proposals arise for cutting back old-age benefits (Campbell 2003). Hence, politicians are wary of ‘waking a sleeping giant’ of angry older voters. They strive to position themselves in a fashion that they think will appeal to the self-interests of older voters, and usually take care that their opponents do not gain an advantage in this arena. So even though older persons do not vote as a bloc, they do have an impact on election campaign strategies and often lead incumbents to be concerned about how their actions in the governing process, such as votes in Congress, can be portrayed to older voters in subsequent re-election campaigns. Third, the image of a senior voting bloc is marketed by the leaders of old age-based interest groups. These organizations have a strong incentive to inflate perceptions of the voting power of the constituency for which they purport to speak.

Yet, as Walker concluded in summarizing an overview of political participation and representation of older people in European nations, ‘[old] age per se is not a sound basis for political mobilization’ (Walker 1999: 7). Similarly, regarding US politics, Heclo concluded, ‘The elderly is really a category created by policy analysts, pension officials, and mechanical models of interest group politics’ (Heclo 1988: 393).

New Research

The quality of the segmentation process will be a key determinant in deciding which campaigns successfully build bonds of trust with the aging electorate. With seniors aged in their 50s and older likely to count for approaching, or even exceeding half of turnout in many contests, a campaign plan that groups together all older voters into one segment is unsophisticated in the extreme and likely to fail.

The process of segmentation will require parties to adopt a process of ongoing adjustments of their positioning in order to maximize the benefits of their own perceived strengths as well as the weaknesses of their opponents. In practical terms, this requires brand adjustments and, informed by market research, the selection of issues, images, language and policies to foreground in their campaigning.

The aging electorates of the 21st century will see the continuation of social trends that began in the previous century, which have combined to increase the levels of voter volatility. Citizens across all age groups are increasingly disloyal to political brands (Dalton 2002) and more likely to switch their preferences across first- and second-order elections (Carrubba and Timpone 2005). This decline in partisan allegiances means that campaign managers cannot make the kind of firm assumptions of voter support based on first-order variables such as class or ethnicity that were common in the middle of the 20th century. A further challenge to strategists will be presented by building the capacity to research and understand senior voters as a sub-group of the electorate because senior cohorts will change profoundly from one general election to the next. As Butler and Stokes famously noted in 1969, every year millions of voters die and leave the electorate, millions join the ranks of the retired and millions enter the electorate for the first time. The electorate as a whole changes every five years, as does the composition of senior voters, as cohorts join and slowly leave this category. As such, generational replacement can be argued as a key variable behind social political changes in the electoral market (Hooghe 2004).

The following are examples of where governmental and commercial organizations have deployed research in order to develop segmentation categories of older people.

In research commissioned by the UK Central Office of Information (Darnton 2005), to assist in the refinement of their communication with older people, social and market research knowledge was synthesized to produce a list of key life stages or events that are highly relevant to the segmentation process. These are listed in Table 3.2.

Snyder (2002) argues that attitudes and opinions will constantly change during a voter’s lifetime, but that values developed over time through accumulated personal experience and confirmed in interactions with peers and family members are less likely to change and act as guiding rules for living and decision-making, and form the basis of personal identity. Snyder’s value-based segmentation research breaks older voters into eight value ‘portraits’:

• True-blue Believers: good health, fun, faith.
• Hearth and Homemaker: good health, relationships, active in community.
• Fiscal Conservative: worried about financial security and health.
• Intense Individualists: possessions, travel, independence.
• Active Achievers: active, online, intellectual.
• Liberal Loners: healthy, lower income, value social equality.
• In-Charge Intellectuals: intellectual, affluent, physically active.
• Woeful Worriers: lower income, financial security worries, family.

The Understanding Fifties and Over (UFO) project in 2004 was a collaboration between media agency OMD and businesses such as the Daily Telegraph newspaper and Peugeot cars. Its research identified seven segments of older persons based on lifestyle and attitudes in areas such as media, brands and politics:

• Live Wires: busy, health-conscious, income to spend on holidays and cars.
• Happy and Fulfilled: financially secure, against change.
• Super Troopers: positive outlook, often lost a partner, high TV viewing.
• Rat Race Junkies: still working, but may have financial worries.
• Living Day to Day: low income but interested in buying brands.
• Unfulfilled Dreamers: often with loans or reliant on social security benefits.
• Anchored in the Past: risk-averse, traditional outlooks, lack interest in new trends.

Although all of these approaches represent valuable attempts to go beyond clumsy segmentation by age alone, it should be noted that there is good evidence to suggest that voters will reject attempts to target them purely on the basis of their age. Indeed, many seniors actively resent being targeted because of their age (RHC Advantage 2010). People age at different rates, and a voter’s cognitive age – how they feel and see themselves – will usually be younger than their chronological age. They will ignore or actively resist any message that explicitly states you should agree because you are old.

The lesson for political marketers is that segmentation by chronological age alone is unlikely to gain a competitive edge for their candidates. A more intelligent development of age-related voter segments based on variables such as life stage, values, generational identities and media consumption habits will be required to achieve strategic advancement. There will be no easy short cuts in the process. The necessary research will require financial and time resources to ultimately develop the ostensibly age-neutral political brands that, nonetheless, resonate strongly with older voters.

This can be illustrated by two case studies, one in the UK and one in the US.

Advice for Practitioners

The relationship between birth certificates and ballot can be a surprisingly complex variable for understanding political behavior. Age is simultaneously a fixed chronological value, a relative concept, a probability indicator of morbidity, and a shared as well as a highly individualized personal experience. Any given individual’s attitudes and behaviors are likely to be forged by the dominant influences in childhood, the main political cleavages experienced as a young adult, and the impact of social trends during the life span. Also important are the cumulative impact of advantages and disadvantages experienced throughout the life course (Dannefer 2003).

To understand how political strategists can research, create and target segments of older voters the role of differing generational characteristics and the influence of an individual’s progress through the life cycle needs to be carefully considered. Historical events or social changes frequently leave lasting impressions on significant sections of society. Such changes would include the Second World War, stark economic recessions or social movements such as feminism. A consistent theme in the debates on political generations is the notion that effects that take place when voters are younger tend to be profound and long-lasting. This is because youth is seen as a formative period in a person’s life, when they are relatively more open to new ideas (Mannheim 1952), as opposed to middle-aged and older voters who reflect new experiences through a much more defined prism of existing views and experiences. However, it would be a mistake to assume that voters from particular generations hold fixed party political allegiances. Van der Brug (2010) argues that people do not get ‘stuck in their ways’ in terms of party preference, but rather there are small but discernible differences in the criteria in how different generations evaluate parties and candidates.

Political behaviors can also be expected to evolve as the individual leaves home for full-time education, enters the labor market, develops adult relationships and/or starts new family units, re-locates into new communities, retires and eventually enters later life. Each successive stage in the life cycle produces different networks and economic contexts (Norris 2003). So both younger and older voters will not only be from different generations, but they will also be in distinctly different life stages. One life cycle effect that sharply differentiates the young and the old is that the average retired person is on a lower income than the average younger worker in their 30s who is in full-time employment; the incomes of the retired are also more likely to be fixed in the form of state or secondary pension payments (Blundell and Tanner 1999; Hills 2006) unless they are adjusted for inflation.

Impact on Politics

The increasing practice of segmenting older voters means that political parties and individual office holders are increasingly conscious during the governing process (not just during election campaigns) of the potential reaction of seniors to policy decisions. This goes beyond traditional concerns with pensions and healthcare and applies to a wider range of policy issues than in the past.

A sign of the increased importance of older voters may not be evidenced through high-profile clashes between the parties on senior issues during the last weeks of any given campaign, but rather through a strategic imperative to ensure before any formal campaign that no valence opportunity on an aging issue will be presented for exploitation by opponents. A perception that one party has the best chance of delivering policies that are generally considered by most older voters to be important, would be highly significant. This strategic imperative is likely to intensify as the proportion of older voters in the electorate grows.

As society takes on a demographic profile never seen before in human history, segmentation can help politics understand the needs of this diverse and growing section of older citizens. It can serve as a tool to open up a dialogue about the meaning of retirement and later life, negotiating a response to the transformation of older voters from excluded minority to a position where politics and government delivers a socially equitable response to the new policy challenges.

However, the danger remains that segmentation will further exaggerate inequalities amongst older people, if only the more literate and more active are attended to. Any perceived disregard of the needs of some sections of older voters as part of a process of privileging others may result in alienating from future politics significant sections of the senior vote. Seniors currently show the healthiest levels of civic engagement, but if their participation rates were to fall towards those of younger voters, this would represent a considerable blow to democratic legitimacy.

Another threat may come from wilder media narratives about ‘greedy geezers’, ‘selfish boomers’ and ‘the gray peril’, as they will hamper the tone and quality of public policy debates. Aging issues such as social care and pensions often require long-term solutions, strong cross-party consensuses and multi-generational support.

The Way Forward

All the major democracies are going through a prolonged period of population aging. There will be an additional 32 million Americans aged 65 and over in the 20 years from 2010 to 2030 (US Census Bureau 2010). In Britain a large number of parliamentary constituencies will see a majority of turnout coming from voters aged 55 and over (Davidson 2010). The significance and potential of voter segmentation for democracies that are now experiencing an age transformation will not be limited to the application of scientific persuasion and the selling of policy programs. The normative application of segmentation will see a process of discovery of the political needs and aspirations of the aging electorate, and provide the evidence base for the communicative and policy responses from governments and parties.

It is now a feature of modern campaigns for commentators to proclaim seniors as one of the pivotal battlegrounds in determining the final outcome. Certainly, it has been the variation in age-group turnout rates internationally that has accelerated the impact of population aging. For example, in the UK younger age groups in the 1970s showed lower turnout rates, but in subsequent elections, as they grew older, their turnout increased. However, this trend seems to have been broken in the 1990s and first-time voters in 2001 maintained their low participation rates in 2005 (Phelps 2005).

Older people are also more likely to vote, join campaigns and contact elected representatives. They have high levels of political literacy and are more likely to follow the campaign closely in the mainstream media. However, as demonstrated in this chapter, the senior power model – and the overly simplistic rational choice-based predictions of ‘gerontocracy’ (Sinn and Uebelmesser 2002) – hold only limited value and are ultimately flawed. They downplay the diversity of older voters and falsely assume that they vote as a single bloc that perceives a single shared economic interest. This model also ignores older voter concerns regarding the prospects for their own children and grandchildren and how they are divided by hugely varying personal social and economic circumstances.

That said, it is clear that there are issues that particularly impact on the quality of life for older voters. If gray voters were to perceive one party to be discernibly stronger, or weaker, on those issues, this is likely to be significant. Any candidate that performs poorly with seniors is going to have to do remarkably well with younger age groups to compensate. For strategists there is a clear choice: to either be the beneficiaries or the victims of long-term demographic change.

The lesson for political marketers is that segmentation by chronological age is crude and unlikely to gain a competitive edge for their candidates. A more intelligent development of age-related voter segments based on variables such as life stage, values, generational identities and media consumption habits will be required to achieve significant strategic advancement. There will be no easy short cuts in the process. The necessary research will require financial and time resources to ultimately develop the ostensibly age-neutral political brands that, nonetheless, resonate strongly with older voters.

Bibliography