‘Consumption as Work’ - Questioning the Meaning of ‘Retirement’ in the Self-Service Economy

This article contributes to the establishment of a new perspective in the debate on how to respond to the economic and social challenges of ageing society. It re-examines the productive activity of older people through the lens of the self-service economy and initiates a discussion on how the self-service economy can be leveraged to help deal with some of the challenges created by an ageing population. The article’s main argument is built on the growing importance of self-service and the potential for older consumers to engage in such production/consumption. The substantial efficiency gains surrounding increased levels of self-service are illustrated. The potential of the retired population engaging in self-service and thus dampen the life cycle deficit is highlighted. Self-service activity by older consumers is presented as a type of productive activity and as forming part of older people’s contribution to society. It is argued that the growing self-service nature of consumption brings the meaning of ‘retirement’ even further into question. The policy implications of self-service are explored and a wide range of suggestions for further research areas related to self-service and the older consumer are put forward.


Introduction
The ageing of society refers to the tendency for populations to have increasing proportions of old people. Evidence confirms a strong global demographic trend towards the ageing of society. Researchers from a wide spectrum of disciplines have recognised the implications of an ageing population for society. Research addresses a wide array of issues from an ageing perspective, ranging from medical research through https://doi.org/10.1007/s12062-020-09317-z to the media's portrayal of the elderly. A fundamental ageing issue in more economically advanced societies concerns ensuring an 'acceptable' level of consumption for so-called 'post-production' members of society. Aspects include pensions, housing issues, health care, transportation infrastructure, and poverty amongst the elderly.
A key research theme in the study of ageing society is the economic challenge of an increasing proportion of the population being retired, i.e. of post-production age (Piggott and Woodland 2016). This paper seeks to nurture a fresh perspective on this theme, based on exploring the consequences of the increasing dominance of service consumption and the long-term trend towards customers themselves doing some of the work in producing services. Service firms are increasingly becoming facilitators of the active production role of service consumers. Classic examples of these include McDonald's, IKEA, and Ryanair. This trend started long before the ICT and internet revolution but has accelerated due to these developments. Management researchers have been examining the consumer's co-production role since the 1970s. The selfservice trend has, however, received relatively little attention from an ageing research perspective.
The shifting boundary between work and retirement necessitates that the meaning of retirement evolves (Ekerdt 2010). The central argument of the current paper is that the more service consumers are also co-producers of services the less dramatic, at least in terms of production, becomes the ending of formal work. In other words, retiring from work is only a partial stopping of a person's production role, and many self-service roles continue, or even expand, in retirement. Consequently, due to the 'hidden production' of self-service, the size of the life-cycle deficit of later life may be smaller than hitherto assumed. As self-service modes of production become more pervasive, understanding the relevance and extent of self-service amongst older consumers becomes increasingly important to assessing the consumption levels of the retired. Indeed, the use of self-service potentially has a similar logic to the substitution of home production for market purchased goods and services in retirement (Hurd and Rohwedder 2003). This will, however, depend on who acquires the productivity gains of self-service (see below: Directions for further research: conceptual research).
The paper starts by introducing frameworks for exploring the economic challenges of an ageing society. It then briefly outlines the rise of the self-service economy. Next it considers self-service in terms of the engagement of older consumers. Finally, the paper puts forward issues for further research which its perspective beckons.

The Economic Challenges of an Ageing Society
The populations of most countries in the world are ageing. Moreover, many of them are ageing far more rapidly than in the past. Many statistical measures of ageing paint a consistent picture. These measures include: the number of years a country takes to shift from having 7% of its population aged over 65 years to having 14%; the population share over a certain age (often 60 or 65 years); and the median age of a population.
What is significant, however, is not ageing itself but the increased rate of ageing. Moschis (2003) points out that during the twentieth century global life expectancy grew by 30 years, more than its growth in the last 5000 years. There exists a multitude of statistical evidence to support the position that the world's population is ageing rapidly.
Europe is expected to have nearly 35% of its population aged over 60 by 2050(UN 2017. In 2017 the percentage of the world's population aged 60 or over was 12.7%, compared to 8.5% in 1980, and is expected to rise to 21% by 2050 (UN 2017).
Population ageing does not necessarily represent an economic challenge for a country. Indeed, for many countries the ageing of their population is an economic bonus. In these countries, typically lower-income countries, population ageing means a growth in their working-age population (Lee and Mason 2011). Some countries, typically higher-income ones, are, however, experiencing a rapid ageing of their populations in a way which is markedly raising the proportion of people who are beyond traditional working age. These countries are experiencing a contraction in the relative labour resources in terms of the total population and this is generally not seen as economically favourable.
In many higher income countries there is much concern and debate about the economic challenges associated with their ageing populations. Economic issues include: (1) the need for a different infrastructure in the future (e.g. in transport, housing, and health), (2) an anticipated higher level of resources required for some sectors of the economy (e.g. health care), and (3) an increased share of public expenditure on pensions. These issues should be viewed as challenges and not problems. Our ancestors would have viewed the substantial increase in life-expectancy as a tremendous achievement. In some countries in the twentyfirst century most people will spend more time over the age of 50 years than the average person lived in the nineteenth century. In the words of Baroness Greengross (2005: 4) "population aging is the greatest legacy of the 20th century and the greatest challenge of the 21st century". As Harper (2009: 12) points out, part of this challenge is "to foster the contributions of the older generations, and to value and make use of their expertise". At the heart of the debate is the question of how to cope financially with a situation in which: (1) most people, upon retirement, will have a period of consumption of at least 20 years-far longer than their parents and grandparents, and (2) the retired population is a much higher proportion of the total population than in the past.
The logic behind the above issues is usefully captured in two demographic/ economic concepts, namely: the old-age dependency ratio and the life cycle deficit. The old-age dependency ratio is defined by the UN as "the ratio of the population aged 65 or over to the population aged 15 to 64 expressed per 100 population". In Europe the old-age dependency ratio is currently 30 per 100, and it is projected to grow to 53 per 100 by 2050 (UN 2017).
The concept of the life-cycle deficit (LCD) usefully illustrates the economic challenge of an increasing old-age dependency ratio. The LCD conveys the reality that during the life span of an individual there are periods when their consumption exceeds their production, thus leaving gaps which are not financed real time by labour income. These deficits occur overwhelmingly during childhood and old age (Lee and Mason 2011). The LCD can also be applied at societal level. As a higher proportion of the population is retired the gap (or deficit) between labour production resources and consumption pressures will become greater. Approaches to tackling the economic challenges associated with an ageing population can usually be understood in terms of their intended impact on the LCD. For example, raising the retirement age is designed to lessen the later life LCD by delaying its onset. Lee and Mason (2011) highlight the issue of how societies fund old-age consumption as being one of their most important choices and proceed to enumerate four options: (1) delaying retirement; (2) relying more on the family; (3) increasing public transfers to the elderly; and (4) accumulating more assets. Delaying retirement by raising the retirement age, is a strategy being implemented in many OECD countries (OECD 2017). Politically, however, implementation is difficult. Lee and Mason (2011) highlight the difficulty of postponing retirement sufficiently to offset population ageing in countries experiencing rapid population ageing. They argue that some rapidly ageing countries would have to raise retirement ages by politically infeasible amounts. Germany, for example, would require a nine-year postponement of retirement by 2050.
Recent decades have witnessed an evolving discourse regarding the so-called 'burden' of a growing retired population (Taylor and Earl 2016). Some writers (Harper 2009;Kanji 2018;Thane 1987;Walker 1980Walker , 2018 argue that the association of ageing society, and an increasing proportion of the population being retired, with economic problems is an oversimplification. Thane (1987) questions the assumption that an ageing population means a growing burden and highlights the domestic services of many retired women; inter-generational support by elderly people; and voluntary and formal work. Harper (2009) discusses four roles of older people, namely: work, care, volunteering, and education. Thane states that "The importance of the shift in the socalled 'dependency ratio' is exaggerated because 'dependency' is unsatisfactorily defined" (p.373). Furthermore, she argues that as "the ageing of society is unavoidable we should seek ways to treat the elderly as a resource rather than viewing them as a burden" (p. 373). It is within this spirit that the current paper is positioned.
This paper follows the view of Thane, Harper, and others that old-age dependency is often an oversimplification and develops a fresh perspective to the economic challenge of an increasing old-age dependency ratio. It does so by questioning the meaning of retirement in the contemporary service economy. In terms of the life-cycle deficit this paper is arguing that how we measure consumption, combined with changes in the selfproduction element that increasing accompanies consumption, calls for reconsideration of the shape of the life-cycle consumption curve for older consumers.

The Rise of the Self-Service Economy
This section briefly outlines the long-term trend of the rise of the service economy in most countries and explores the growing importance of self-service. Next the implications of this in terms of dampening the life-cycle deficit are put forward. As services increasingly dominate GDP and employment, especially in countries with strong ageing profiles, to understand the production challenges in ageing societies is increasingly a matter of understanding the service production challenges in ageing societies.
Since the 1970s there has been increasing research attention to how services are produced and delivered. An important early realisation was that the consumer's role is often not a passive consumption role rather the service consumer is also a part-producer of services. That is, the service customer can usefully be conceptualised as an employee. Indeed, the concept and reality of self-service was well established before the coming of the so-called 'new economy' or e-services (e.g. Gartner and Reissman 1974;Lovelock and Young 1979). The digital economy has, however, accelerated this trend and enabled consumers to become more active in the production of services. Examples include e-banking; travel services; and education. Self-service is usefully defined by Castro, Atkinson and Ezell (2010: 4) as: "the process by which consumers engage in all or a portion of the provision of a service or product". The use of a self-service approach can transform the cost structure of a business and make the customer central to efficiency. As self-service is usually delivered through some form of technology, service management researchers often employ the term 'self-service technologies' (SSTs), defined as "technological interfaces that enable customers to produce a service independent of direct service employee involvement" (Meuter et al. 2000: 50). SST examples include internet banking, automated teller machines (ATMs), petrol pumps, fast-food restaurants, and supermarkets.
Two large service firms in which customers performing a sizeable amount of the work are IKEA and Ryanair. At IKEA, the role of the customer has been redefined to include work traditionally done by retail employees. Customers typically provide themselves with product information, collect items from the warehouse, and transport them home. To improve productivity, Ryanair's passengers perform several services which traditionally have been provided for them. For example: passengers book online; check-in online; and print their boarding card.
Many writers underline the rapid growth of SST. Alcock and Millard (2007: 314), for example, write: "it is indisputable that self-service has been hugely successful, from cash dispensing ATMs to fast food counters". Rieder and Voß (2010) provide examples of how self-service is increasingly substituting traditional, employee-delivered, service. These include self-scanning checkouts at shops; self-check-in at airports; postal packing stations; and health monitoring stations. Increasing numbers of new SST applications are emerging, including hi-tech self-service restaurants; digital photograph printing; and reverse vending machines. Moreover, service delivery increasingly involves interaction with service robots, whether physical or virtual. Examples of service robots include chatbots providing information and advice, and physical robot receptionists. The rapidly falling cost of service robots suggests they will be widely applied (Wirtz et al. 2018). Service robots can considerably reduce the cost of serving customers. Researchers have started to identify the tasks suitable for delivery by service robots (e.g. simple, high volume tasks). Likewise, factors influencing robot acceptance have been identified. Several studies suggest that older people generally hold more negative attitudes towards robots (Belanche et al. 2020). As older consumers tend to have greater relational needs, they may be less likely to accept service robots (Čaić et al. 2018). Physical assistance technology (e.g. robotic exoskeletons) can, however, assist older consumers to live independently and to continue with self-service roles.
The use of self-service is expected to continue growing over coming decades (Rieder and Voß 2010). For example, the global self-checkout terminals market is forecast to grow at 19% annually during 2020-2024 (TechNavio 2020). The increased use of SSTs will be partly due to a growing use of existing SST applications, and partly due to the emergence of new technologies. As self-service becomes more predominant, more expertise in measuring the productivity and efficiency of service customers is likely to emerge.
As service consumers increasingly undertake service tasks, from a managerial perspective how the organisation works changes. Service firms increasingly becoming facilitators of the active production role of service consumers and treat customers as an extension of their human resources. This can include passing on cost reductions through reduced prices. From a wider economic perspective, 'grey' production is occurring. The work done by the service consumer classifies as 'unrecognised work'. More and more such work is being done, however, it is not registered by traditional national income accounting procedures. It happens widely but, over time, consumers increasingly take it for granted. This work is the work of serving petrol; shopping in self-service stores; self-check-in at airports etc.
To convey the vast size of the customer labour input, Rieder and Voß (2010) undertook rough calculations of the value of the work done by customers in selfservice settings. For example: assuming that clearing the table at McDonald's has a value of 0.10 Euros per customer and multiplying this out by some 58 million customers worldwide per day (2008 figure), Rieder and Voß arrive at a sum of 2.1 billion Euros annually as the value added by the customer from clearing tables at McDonald's. Alcock and Millard (2007: 313) emphasise that it costs service organisations "significantly less when customers can find information that they need or conduct transactions themselves, compared to when a human customer service advisor assists them". Indeed, it is possible to estimate the value of customer role in using SST systems. The lower price of SST products compared to employee-produced products can be considered. For example, the cost of conducting a bank transaction with a teller in a branch, compared to the cost of an e-banking transaction carried out by the customer. Castro et al. (2010) provide estimated cost data: the branch transaction costs around $4.25 and the online transaction around $0.20. Employee productivity improvement statistics also underline the enormous potential of customer SST participation for improving business efficiency. For example, when self-checkout is used in a store, one worker usually manages four checkout points (Dean 2008). Thus, the cost reduction benefits of introducing self-service technologies can be substantial.
The above discussion of the growing importance of self-service is highly relevant to the studying of ageing. This is because, in terms of the life-cycle deficit, the more that the retired are engaged in consuming SST-based services, then the less black and white the LCD becomes. The rise of co-production, including self-service technology and systems, makes the work/non-work divide less clear cut. As such, building on Harper (2009) and Thane (1987), this further changes and complicates the meaning of the word 'retirement'. It also adds to the call for a more contemporary definition of productivity (e.g. Dosman et al. 2006).
The more that older consumers participate or input to service production, the less will be the impact on service production of an ageing population. The more that co-production grows, the more demand for services is accompanied by its own supply. That is, co-production dampens the life cycle deficit because consumption brings some of its own resources with it. This could be very good news for ageing societies. Time budget researchers have drawn attention to the question of how people spend their growing number of post-retirement years and how these time-use patterns may vary by age, gender and country (Gauthier and Smeeding 2000;Vargha, Gál, and Crosby-Nagy 2017). Time budgeting techniques could be useful in generating more detailed insights into the extent of self-service activity of older people.

The Engagement of Older Consumers with Self-Service
The dual trends of the rise of a service-dominant economy and the ageing of the population structure creates an important context for strategy makers. A vital question is: what are the implications for providing services in markets where mature customers are abundant? More specifically, it is important to identify the competencies that service managers and service employees should possess to effectively deliver services to older service consumers. The issue of service provision in an ageing society has many aspects. These include the use of technology; the design of service facilities and the human interaction taking place in service settings.
Considering the growing role of SST, and its potential to enable the same level of consumption to be maintained for less monetary expenditure, self-service seems relevant to the study of the economic and social issues associated with an ageing population. This relevance is heightened when it is considered that older consumers are generally less engaged in the use of SSTs (Simon and Usunier 2007). For example, a study of Finnish consumers found that 18-to-35-year-olds were 1.85 times more likely to use mobile banking services than mature users (55+ years) (Laukkanen 2016). Indeed, one of the main motives for research into the use of SST by older consumers should be the widespread evidence that, generally, they are less likely to use selfservice delivery modes. Using SST in the narrower ICT-based SST sense, this lower use of ICT is sometimes referred to as part of the digital divide.
The growth of the SST-based service encounters has raised a range of issues related to older service consumers and their use of SST. Researchers have sought to gain an older consumer perspective on self-service technology. Issues examined include the degree of openness of older consumers to SST, and the reasons for lower use of SST options by older consumers. There is some research evidence that decreased physical and/or cognitive ability, often associated with ageing, has an influence on self-service usage. Age-related physical impairments, such as impaired vision or hearing and lack of mobility, can constrain the ability of older consumers to perform physical tasks associated with self-service (Kim and Jang 2019). For example, Tomazelli et al. (2017) identify a range of physiologically based constraints on the capacity of older consumers to effectively serve themselves in supermarkets. This section reviews a selection of research findings on the attitudes and behaviours of older consumers regarding SST. It attempts to assess the general trend in their SST usage. It is important to understand the older consumer perspective on self-service technology. This perspective, with its own reasoning and logic, offers insights into how the motivation to adopt new modes of service consumption may vary with consumer age.
Before examining the insights offered by a range of studies relating to self-service and the older consumer it is helpful to consider the types of self-service. There is a long tradition of classifying services according to the nature of service delivery (e.g. Lovelock, 1983). Self-services can usefully be distinguished according to the locus of their delivery. Types include self-service in 'bricks and mortar' settings with staff available (or unstaffed); self-service accessed via the internet from home; self-service accessed via smart phones away from home; and peer-to-peer problem solving (P3) communities, i.e. internet-enabled consumer communities.

ICT-Based Self-Service
Much of the growth in self-service in recent years has been based on ICT. Examples of such technologies include ATMs, self-check-in and checkout, self-scanning, ticket machines and information kiosks. Table 1 (below) outlines and illustrates four themes that are particularly relevant to understanding the perception of ICT-based self-service from the perspective of the older consumers. These are: (1) the social contact implications; (2) ability and confidence to use such services; (3) ergonomic issues using ICT; and (4) privacy and security concerns.
An important theme in discussion of the older consumer's use of ICT-based selfservice concerns the implications for social contact. Studies reveal insights into the influence of preferences for human interaction and contact on some older consumers in their decision on whether to adopt SST modes of service consumption. McPhail and Fogarty (2004), in a study of Australian financial services consumers, found consumers over the age of 50 years the least likely to adopt self-service banking technologies (SSBTs). They observed that most non-users of SSBTs were older consumers living alone. This group was found to place greater value on the social interaction of face-toface banking. Likewise, studying of service preferences in banks, post offices, and rail stations, Simon and Usunier (2007) found age to have "a strong and direct negative effect on preferences for SST over personnel-in-contact" (p. 170). More recent research has similar findings. For example, Rahman and Yu (2019) found a social preference for shopping even though most respondents had online purchasing skills.
Technological confidence has also been found to influence older consumers use of ICT-based self-service. Dean (2008) investigated the effect of age on attitude toward and use of self-service technology in a US retailing context. The study found consumer age to have a highly significant influence on several variables associated with selfservice technology. The older respondents (49+ years) were found to have use experience with fewer self-service technologies; a lower level of confidence in using SSTs; a higher degree of missing human interaction; and a lower likelihood of opting for self- Some studies within the innovation resistance literature place more emphasis on the ergonomic functionality aspects of ICT-based self-service, rather than on older consumers' willingness to engage with them. Rieder and Voß (2010) describe the findings of an exploratory study into the implications for daily life of seniors of the spread of SSTs in Switzerland. The study found that SST played a major role in seniors' everyday lives. The use of ATMs, ticket machines and vending machines was common, and some seniors had used e-services. The research identified several age-related usability barriers connected to self-service technology. For example: low contrast on display screens and small lettering. Other studies (e.g. Rahman and Yu 2019) contain similar findings.
Some research highlights the role of privacy and security concerns as barriers to the use of ICT-based self-service by older consumers. Laukkanen et al. (2007) set aside the myth that mature consumers do not try products based on new technology, and state that "mature consumers are becoming familiar with technology such as computers, the internet and mobile phones" (p. 420), and emphasise the importance of service providers understanding the barriers preventing the elderly from adopting service innovations. These barriers may include a risk barrier with older consumers having greater security concerns and perception of risk than younger consumers (Laukkanen 2016).
The locus of ICT-based technologies can also be relevant for older consumers attitudes towards them and adoption of them. Such technologies may be used in physical service settings, at home or on the move. In physical service settings issues include the option of using employee service, the possibility of assistance from others in using ICTs, and the social interaction aspect. The use of ICTs at home involves issues such as the desire for continued independent living, the ability to use the technology, and privacy concerns. The use of ICT-based self-service on the move often relates to the use of a smart phone. Space prohibits an extensive discussion of these issues here, but a few relevant themes are briefly outlined for each of these three loci.
The decision to use ICT-based self-service, such as ATMs and self-checkout, in physical service settings involves the social contact preference issue discussed above. Studies of SST adoption have found older consumers less likely to adopt SST due to a desire for social interaction (Leng and Wee 2017). SST adoption is also influenced by the extent to which training and assistance in using the technology is available (Yagil, Cohen and Beer 2016). The perception of risk is also a relevant factor. A study of selfcheckout usage in supermarkets found that for older consumers perceived risk was a significant barrier to the adoption of the SST (Lee and Lyu 2019).
The use of ICT-based self-service by older consumers at home includes technologies enabling internet banking, healthcare, and food ordering. Some studies have stressed the value placed on the possibilities for maintaining independent living that such technologies can provide (e.g. Harrefors et al. 2010). On the other hand, the concern that using such modes of service delivery may reduce home visits by employees is also relevant to some elderly people (e.g. Sallinen, Hentonen and Teeri 2020). Likewise, the availability of others to train and support older consumers in the use of ICT-based self-service in the home is also a factor in its adoption and continued use (Rieder and Voß 2010).
The use of ICT-based self-service 'on the move' refers mainly to service occurring beyond traditional service settings, such as banks and shops, and often outside the home. Examples of such self-service include using smart devices to pay for parking, app-based banking services, and to hire mobility solutions. Often such service is accessed through devices such as smartphones and tablets. Smartphone adoption by older consumers, especially the younger old, is relatively common but often limited to basic functions with older consumers rarely having someone to assist them in learning (Choudrie, Pheeraphuttranghkoon and Davari 2020). If older users cannot utilise smartphones easily, they will have difficulties in accessing information, shopping and communicating in on-line communities (Hong, Trimi and Kim 2016). Research draws attention to the functionality usage barriers that mobile devices might have for older consumers. For example, the small size of devices can make readability and text entry more difficult than on a PC/laptop (Laukkanen 2016). Likewise, concerns may exist about issues such as the limited battery life and connectivity of mobile devices. Another important concern about mobile-based service is the perception of greater privacy and security risks (Chen 2013).
It is important to understand that SSTs extend beyond ICT and the internet. Older consumers engage with self-service in 'bricks and mortar' service settings such as shops, banks and public transport. The remainder of this section will explore selfservice in 'bricks and mortar' contexts from the perspective of older consumers and their capacity to continue to engage in this widespread form of self-service.

Traditional Self-Service
Whilst ICT developments have given rise to a range of new self-service options, much self-service occurs in traditional service settings, and often without ICT. This section explores some of the literature that relates to the older consumers desire and ability to engage with self-service in traditional (or 'bricks and mortar') service settings such as limited service restaurants, supermarkets, and public transport. Research identifies a range of ways in which the older consumer service experience may differ to younger consumers. Some of this research contains insights into the older consumer's capacity and desire for self-service. Research also draws attention to how service environments can be designed or re-designed to facilitate self-service by older consumers. Whilst it is not the purpose of this paper to conduct a systematic literature review of older consumers and their self-service engagement, Table 2 (below) presents a selective review of the more significant literature organised into main themes.
A significant body of research exists into the social context of older consumers visits to physical service settings, especially retail and leisure settings. With 80% of consumers over the age of 50 being frequent supermarket shoppers (Arnold and Krancioch 2008), grocery retailing, has attracted ageing-related research attention. The social context is relevant as it influences the older consumer's motivation to continue shopping and thus engaging in self-service. Viewing shopping from a social perspective includes the direct interactions of going shopping with friends, engaging with employees and other customers, and the indirect interaction of having other people around. Older consumers have been frequently found to seek social contact in their service consumption experiences. Myers and Lumbers (2008: 296) conclude that many older consumers view shopping as "a means of being with and interacting with other people". Likewise, Kim, Kang and Kim (2005) found that for many older consumers retail stores often fulfil the function of reducing loneliness. Research from contexts beyond retailing also confirms that the desire for social contact as a motivator for many older consumers to visit service settings. Example include hospitality (Cheang 2002;Rosenbaum 2006;Song et al. 2018); leisure (Grove and Fisk 1997) and travel (Hudson 2010). Older consumers will, however, vary in the personal importance they attach to social contact in service settings (Nicholls and Gad Mohsen 2015). Likewise, this importance may also vary with cultural values such as the level of family orientation (Song et al. 2018).
Connected to the social contact desire aspect of physical service settings is the issue of social comfort. Research shows that older consumers are more inclined to engage in selfservice when they feel comfortable with those around them. Some research provides evidence of older consumers experiencing social discomfort when they perceive the service setting to be dominated by young consumers (e.g. Angell et al. 2012;Hu and Jasper 2007;Kwortnik 2008;Myers and Lumbers 2008). Given that self-service settings are more likely to require asking other customers for help, social comfort is important as some older consumers report feeling embarrassed at needing to ask for help. So, whilst many older consumers may find the social contact of continuing to engage with service motivational, it  • Public transport ease of use is important that they feel comfortable with that contact. Furthermore, some social situations will be inherently embarrassing, and are best solved by improved service design. For example, Tomazelli et al. (2017) mention the social awkwardness of an older consumer needing to request assistance to reach a geriatric diaper placed on a high shelf.
Research into the service experience of older consumers reveals a wide range of ways in which the design of physical service settings, such as hotels, restaurants, and shops, can present difficulties to some older users. Many studies of older consumers' perceptions of retail services identify a range of ergonomic issues relevant to customer satisfaction and service accessibility (e.g. Angell et al. 2014;Omar, Tjandra and Ensor 2014;Pettigrew et al. 2005;Yin et al. 2013). The ability of older consumers to access self-service retail environments is influenced by factors such as print size on price tags; the adequacy of store lighting; and trolley functionality. Similar factors are mentioned in studies of limited service restaurants. Almanza et al. (2017), for example, identify older consumer experiencing barriers to service due to menu font size, inadequate lighting, and loud background music. They also identify ergonomic issues for older consumers regarding narrow aisle width, restroom accessibility and excessive depth of self-service areas. Likewise, research into older consumers use of hotel facilities identifies design issues, such as the use of low glare lighting and grab bars in bathrooms, that are important to address to ensure the facilities can be used (Marvel 1999). Some studies focus on specific aspects of the older consumer service experience. For example, Seo and Fiore (2016) found that fitting room design was a significant aspect of the shopping experience for older females. There are many other ergonomic insights into the barriers to engaging in self-service facing older service consumers in physical service settings. For the reader interested in more detail useful reviews of this literature exist (e.g. Yin et al. 2013).
Linked to the design of physical service settings is the issue of their transport accessibility. Research shows older consumers' service use to be affected by the inability to drive (Rahman and Yu 2019). Studies of older consumers have found that many respondents, especially the older old, rely strongly on public transport to get to self-service settings such as supermarkets (e.g. Angell et al. 2012;Meneely Burns and Strugnell 2009). Older consumers, however, face many physical challenges on public transport that often make them hesitant to use it (Yuan et al. 2019), thus constraining their use of self-service settings. For those using cars the availability of car parking spaces close to the physical service setting is important to continue using self-service facilities (Angell et al. 2012). From both a societal and business perspective it is important that older consumers can get to self-service facilities in order to perform as co-producers of services. New forms of mobility, such as ridesharing, may contribute to improving older consumer access to self-service facilities, but research suggests that a better understanding of the ridesharing needs of older adults is required (Payyanadan and Lee 2018).
Some research is particularly relevant to the focus of the current paper on how to facilitate self-service labour during retirement. Specifically, the challenge is how to enable older consumers to keep using retail and service facilities to a greater age. Petermans and Van Cleempoel (2010: 27) address the question of "how can the design of retail stores be adapted to meet older consumers' needs and wants?". Their case study describes and evaluates two examples of "senior supermarket" chains operating in Germany/Austria. These supermarkets have been designed around the physical and social needs of the 50+ segment.
The studies discussed in this section highlight several issues. They provide empirical evidence that older consumers are generally less engaged in the use of some SSTs. They also provide evidence that SST usage amongst older consumers is a changing picture and a complex one. Furthermore, the studies provide insights into how the motivation of older consumers to use self-service options may differ to younger consumers.
Based on this selective literature review several fundamental questions connected to social policy emerge: & How will the SST usage of older consumers evolve over the next 30 years? & To what extent will future generations of older consumers, used to disruptive innovation and embracing change throughout their working lives, continue to adopt new SSTs in retirement? & What can be done now to influence SST usage of older consumers and future generations of older consumers?
The following section takes up these, and other, such questions and develops them into potential research themes which would contribute to our understanding of the challenges and opportunities available to societies looking for resources and approaches with which to restructure their production potential because of substantial shifts in their age structures.

Directions for Further Research into Self-Service and Older Consumers
This paper makes a conceptual contribution to the debate of how society copes economically with a rising post-retirement (i.e. upper end) dependency ratio. It highlights the impact of older people as users of self-service systems in their consumption of services. Further research in this field is critical given the possibilities that selfservice technologies provide for reducing resource scarcity in societies with rising oldage dependency ratios. If a sizeable proportion of the 'retired' population became more actively engaged as co-producers of services (i.e. 'partial employees') then the 'dependent' and 'working age' distinction would become less significant. This would have an important impact on the entire ageing debate, and carry major policy implications at government, corporate and individual levels. The scale of these implications is illustrated by the following simplified hypothetical example: if in a country with an average retirement age of 65 years the retired population aged between 65 and 80 years engaged in an additional 7 h per week of self-service, this equates to raising the average retirement age to 68 years (assuming constant productivity). 1 In other words, an age management social policy which raised the self-service engagement of the retired would de facto raise the retirement age. The assumption of increased time being used for economic activity in retirement is supported by empirical research. For example, Lürhmann (2009), using German time-use 1 Note on calculation: a 35-h working week is assumed, so 7 h of self-service work represents 20% of 35 h (or 1 day per week). Thus 20% of 15 years (the number of 'younger old' years between 65 and 80) is 3 years. Adding these 3 years of self-service work to the age of 65 years gives 68 years. The purpose of this illustrative calculation is to demonstrate how the gradual increase of self-service by the active retired population could in effect accomplish the equivalent of the politically contentious raising of the retirement age by 3 years. data, found that households with a retired head spent about 90 min more daily on home production activities compared to non-retired ones.
This section identifies topics where research attention could build upon the older consumer self-service perspective put forward in this paper. These research topics include: further conceptual work considering how self-service work is perceived by individuals and society; the older consumer perspective of self-service, including the factors influencing older consumers SST adoption; the role of family and friends in supporting older consumer self-service adoption; prediction of how older consumers attitudes to self-service will evolve over the coming decades; the development of an active ageing perspective on the self-service involvement of older consumers; and the organisational perspective on older consumers and self-service.

Conceptual Research
This article has introduced to the ageing literature a conversation on how the self-service economy can be leveraged to help deal with some of the challenges created by an ageing population. Its basic premise is that the rise of the selfservice economy has increased the potential for older consumers to engage in production as they consume and is changing the meaning of 'retirement'. Further conceptual research is needed into the consequences of increased selfservice by older consumers and to clarify the mechanisms through which increased SST engagement by older individuals leads to lower burden. Issues include: 1) To what extent are consumers rewarded for self-service work? If self-service productivity gains are not passed on to consumers there may be a dampening of the societal life cycle deficit but not of individuals' deficits. 2) If a self-service is widely adopted by all age groups does it not simply become a new norm and not perceived as production? For example, like direct dialling or taking petrol. If so, one consequence would be that it would not, unlike increased domestic production in retirement (e.g. Dosman et al. 2006;Stancanelli and Soest 2012), change the relative income of the retired compared to those in work. 3) How can economists monetise the production that takes place whilst self-service is performed? 4) To what extent are there different economic, social and political outcomes depending on how the savings of the enhanced productivity of older consumer selfservice are shared?
In connection to such issues, a greater awareness of how much service is self-service would be a fruitful area for research. Such information would contribute to improving our knowledge of the tempo and the scale of the shift to self-service. It would also allow a greater assessment of the risk of human service routes being removed or marginalized. A key issue here is that as service becomes increasingly SST-based, human-based service can become more expensive and/or less available. If older consumers are less likely than average to adopt SST-formats, then they may actually be worse off than when the service was typically delivered via employees. This is a fruitful area for poverty researchers to investigate.

The Older Consumer Perspective on Self-Service
An important research area is furthering of our understanding of the older consumer perspective on SST. There is quite an extensive literature on the adoption and use of technologies, such as healthcare tools (Mostaghel 2016), that are specifically intended for the elderly. However, the literature on the adoption and use of self-service intended for the general population requires more focus on researching how older consumers view and engage with these systems and technologies. This research should include how the value of SST is assessed by older consumers and what specific concerns they may have. A detailed assessment of the extent to which older consumers engage in different types of SST, compared to younger consumers, would provide a useful starting point. Future research should include the use on non-ICT based self-service, for example, shopping at supermarkets. The emphasis here will be on the influence of factors which may cause the elderly to discontinue their use of such services. This includes in-store ergonomic factors, but also extends to housing policy and barriers to age-friendly housing with local services which can be accessed independently until later in life. Non-ICT based self-service also needs further research to better understand the positive motivation that older consumers possess to continue using them and how service providers can provide environments that rewards this motivation. A key area for future research here is studying the social rewards that older consumers obtain by continuing to visit physical self-service settings. Such research would build on the older consumers and third places literature stream (e.g. Cheang 2002;Rosenbaum 2006;Song et al. 2018) and address questions such as How can service organisations maintain authentic social contact in increasingly automated service settings?
Further research is needed on the feasibility of older consumers following SSTbased consumption. Greater understanding is required of the factors that influence older consumers SST adoption. Given the age-range and diversity of the older consumer group, insights into ways of segmenting this group, both demographically and psychographically, would be useful. An important area for research is into how to design more age-friendly interfaces. Of particular interest is how to actively engage with older consumers as innosumers (Peine, Rollwagen and Neven 2014). Likewise, given the growing importance of smart devices as a gateway for accessing many self-services, future research needs to further investigate the digital divide facing older consumers and explore it in terms of the access divide, the capability divide, and the utilisation divide (Hong et al. 2016). Similarly, service robot design (e.g. Belanche et al. 2020) needs to be researched from an older consumer perspective.

Older Consumer Support and Motivation to Embrace Self-Service
Another aspect of managing the use of SST by the elderly is for service providers to understand how they manage to use it. Research has found that older consumers may rely on the skills of others, such as relatives, to use self-service options (Rieder and Voß 2010). A greater understanding is needed of the role of family and friends in enabling older consumers to use, or continue to use, self-service options. For example: a 50year-old daughter/son might assist a 75-year-old parent with SSTs such as e-banking and e-shopping. The growth of SSTs, especially online ones, provides a new avenue for research into inter-generational support. Moreover, given the context in some countries of adult children often living substantial geographical distances from parents, research into how e-service options can facilitate care for elderly relatives is extremely relevant. Greater understanding of the routes by which older consumers adopt SST, for example via their children, would be valuable for service providers.
Further research should investigate how the perspective of older consumers may evolve in the future. A vital research area concerns the extent to which future older consumers, or Next Generation Older Consumers (NGOCs), will embrace self-service. Some evidence and theory support the view that today's younger consumers will view SSTs as normal when they are older, suggesting higher rates of SST use among future older consumers. This may not automatically happen, due to factors such as technology continuing to progress and the age-related deterioration of perceptive, motor, and cognitive skills (Wandke, Sengpiel and Sönksen 2012). Systematic research is needed into NGOC e-service/SST usage and predicted usage. Policy needs to address the question of how to significantly reduce any gap between the NGOCs' usage of SST and the usage of generations younger than them.
Successfully addressing research issues connected with NGOC before they themselves become older consumers facing the same current dilemmas is clearly a strategic priority. Given strong SST trends, the research can identify what the main issues are and proposed solutions to avoid having a significant SST generational divide when today's 40-60-year olds are 60-80-year olds. The key notion is that for today's older generation it is more 'difficult' to heavily adopt SSTs, but for the generation behind them there is more time to work on raising adoption levels. Factors such as relatively higher ICT awareness and motivation of the NGOC segment should make raising eservice usage relatively easier but, on the other hand, it would be a real lost opportunity if significant numbers of NGOCs 'missed the e-services boat'. Accordingly, the monitoring of SST participation rates by age, gender, class, employment status, and geographic location, becomes highly relevant.
ICT may also indirectly influence SST adoption by future older consumers. They are likely to be more at ease with e-services, and more engaged with social media. As such they may be less inclined to view human service interaction, such as shopping, in terms of social interaction rewards. Therefore, older consumer resistance to change based on the loss of social contact, reported in many studies (see Tables 1 and 2), may become less significant. Accordingly, an important research question is: Will future older consumers attach less importance to the social exchange surrounding face-to-face service encounters?

Active Ageing Perspective
The self-service perspective on older consumers also offers potential for enriching conceptual understanding of active ageing. Hamblin (2010) discusses the emergence of the term 'active ageing' and draws attention to the various forces, including economic and health, influencing its evolution. Some writers have focused more on active ageing as a labour force participation issue, whilst others have emphasised the health implications of greater activity (Hamblin 2010). The self-service labour of the retired can be used to further develop understanding of active ageing. Indeed, it raises several interrelated questions such as: How may self-service activity in retirement influence health? To what extent is the self-service activity of the retired a form of labour force participation? How may early exit from the formal labour market influence selfservice activity in retirement?
A connected active ageing research issue concerns whether remaining in work provides an enhanced capacity for self-service activity during retirement. Given evidence from the expansion enhancement perspective in the Work-Life Balance literature that work-family facilitation can occur (Marston et al. 2019;Wayne et al. 2007), it would be fruitful to research how skills and attitudes gained at work may provide resources for self-service. Specifically, a fertile field for investigation is the extent to which delayed retirement, including longer periods of work due to increasing the attractiveness of phased retirement (Allen, Clark and Ghent 2004), develops potential for using internet-based SST. For example, ICT skills acquired at work may facilitate an older consumer's ability to manage their health as studies have shown strong linkages between low internet literacy and low health literacy (Xie 2012). Given the growth of bridge employment (Beehr and Bennett 2015), i.e. paid employment after retirement from a career job, it would be useful to research how ICT skills are maintained in bridge employment contexts compared to career employment ones. Future research could also examine approaches for imparting ICT skills to retired people, for example, the role of libraries in introducing technologies to older consumers (Kesselman 2019).

The Organisational Perspective
Another recommended area for future research is the acquisition of an organisational perspective, both managerial and front-line, on older consumer engagement with selfservice. One priority is developing a managerial perspective on the challenges service organisations face in raising older consumer SST participation rates, and the strategies and tactics organisations have adopted. Specific themes include approaches for encouraging older consumers to adopt self-service modes of production; methods for monitoring older consumer SST engagement; and managing employee support to older consumers using SSTs. One strategic aspect that has been suggested for providing a social experience for older consumers is the inclusion of older employees in the customer contact staff (e.g. Song et al. 2018). A second area for future research is how to design physical service settings that blend self-service functionality with interactive spaces for social contact opportunities. A third area for research is to examine the extent to which organisations follow the recommendations of consumer behaviour researchers (e.g. Bae, Jo and Lee 2020; Chaouali and Souiden 2019; Kim and Jang 2019) regarding the importance of segmenting the older consumer market and the limitations of a chronological age approach to segmentation, and employ differentiated communication strategies to encourage older consumer sub-segments to increase their use of self-service.
Research that acquires a front-line employee perspective on the older consumer's interaction with self-service is also needed. Services management research has recognised the ability of front-line employees to provide insights into the service experience of customers (e.g. Nicholls and Gad Mohsen 2019; Verleye, Gemmel and Rangarajan 2016). Likewise, research into the use of healthcare technologies by the elderly has found employees to be a useful source of service improvement suggestions (Mostaghel 2016). Further research could seek employee insights into areas such as how older consumers interact with self-service; how employees offer and provide assistance to older consumers and how consumers respond to such assistance. Such research would also assist in the identification of good practice. For example, writing about retail self-checkouts, Dean (2008) suggests using older employees to supervise the self-checkout zone. More generally, having older 'helpers' at self-service kiosks uses the power of older consumers having a reference group to encourage them to try the technology.
This section has enumerated a diversity of research themes connected to the older consumer and self-service and has demonstrated the abundance of research opportunities available.

Conclusion
Over the next half century, creating a society in which the vast majority of the young elderly (say 60-80-year olds) are active consumers of services, engaging heavily with self-service technology, may be easier to achieve than raising the pension age by up to 10 years. Indeed, making citizens aware of the trade-offs involved may assist in the transition itself.
The more service consumers are also co-producers of services the less dramatic, at least in terms of production, becomes the ending of formal work. In other words, retiring from work is only a partial stopping of a person's production role, and many self-service roles continue. The higher the self-service component of economic activity, the more economically active those who are 'retired' remain. Thus, the size of the lifecycle deficit of later life may be smaller than previously thought. As self-service modes of production become more pervasive, understanding the relevance and extent of selfservice amongst older consumers becomes increasingly important to assessing the consumption levels of those who are retired.
More fundamentally this paper contributes to the perspective that population ageing is not automatically a problem. The paper, through its exploration of self-service, broadens and enriches our understanding of the economic contribution of older people. It also outlines a service management perspective and context which offers a unifying reference point to accommodate, and facilitate cross-learning from, some existing ageing research and initiatives.
Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/.