This submission summarises research carried out by the Institute for Policy Research (IPR) at the University of Bath into the economic and political feasibility of universal basic income (UBI) in the context of technologically induced labour market change.
While mass technological unemployment may be a distant (indeed, unlikely) prospect, nevertheless labour markets are already witnessing important forms of dysfunction and disruption. These include frictional and long-term unemployment, skills redundancy, wage polarisation, and growing levels of insecure employment. These trends look set to become more widespread and more politically salient, as technological substitutes extend towards increasingly diverse (non-routine and cognitive) tasks – cutting across the skills and income profiles of the workforce – and as 'platform economies' become more prevalent.
Basic income has been proposed as a flexible way to cope with these diverse forthcoming threats. Furthermore, basic income is not only a 'crisis response' to labour market dysfunction; it may complement the emergent 'knowledge economy' by driving positive societal change. However, there appears to be some discord between basic income's potential value in the face of these emergent challenges, and its practical political prospects.
We find that despite a number of significant advantages for the emergent digitalised economy and apparently widespread public support across the political spectrum, basic income is subject to a number of profound feasibility constraints. These include:
In the UK as across the industrialised world, the 'automation' and 'digitalisation' of labour market tasks – also known as the 'Fourth Industrial Revolution' – continues apace. As some occupations disappear, others will expand, and jobs will also change in terms of the nature and composition of productive tasks. Individuals will need to be equipped for demands of the digitalised economy, and for the disruption these changes might bring. The Work and Pensions Select Committee's call for evidence includes questions about the nature and extent of the labour market effects of technological change, as well as the implications for public policy in terms of employment protection legislation, skills development, and – perhaps most crucially for the DWP – social protection. In particular, the inquiry is concerned with the prospects for universal basic income (UBI) as a solution to technologically-induced labour market change.
At the Institute for Policy Research (IPR) at the University of Bath, we have an established programme of research addressing precisely these issues. Last year we published a report titled Basic Income, Automation and Labour Market Change (Martinelli, 2019a), following 2017’s Assessing the Case for Universal Basic Income in the UK (Martinelli, 2017a). The former publication included an extensive literature review on the likely implications of technological change for labour market outcomes, and focused on UBI among potential public policy responses. Our research focuses on issues of UBI's political and economic feasibility, adopting a 'political economy' perspective. Microsimulation analysis illuminates patterns of gains and losses across different income and demographic groups, and exposes policy design trade-offs in relation to fiscal and distributional goals. Analysis of attitudinal survey data sheds light on the dynamics of support for and opposition to UBI, how these might be affected by ongoing labour market change, and the prospects for the emergence of robust coalitions in favour of UBI.
This submission summarises the main findings from this ongoing research programme, and will address the first three questions set by the inquiry: What are the main challenges that DWP faces as a result of the “Fourth Industrial Revolution”?, What do we know about the possible likely impact on the labour market?, and Is there a need to consider new, long-term approaches to addressing change in the labour market: for example, introducing a Universal Basic Income (UBI)?
Labour market change caused (in part at least) by the Fourth Industrial Revolution may present challenges to the DWP in several respects. As Millar and Whiteford (2020) suggest, changing patterns of work pose three major challenges to existing systems of social protection, relating to their inclusivity of people in a range of employment and financial circumstances, to the security of their funding mechanisms (especially in the context of systems that link entitlements to prior contributions, i.e. insurance-based systems), and how to deal with people with unstable or irregular employment. It is a distinct possibility that insecure contractual terms, low pay, and spells of unemployment will become more prevalent, although the certainty of adverse developments is often exaggerated.
Technological change may increase the need for social protection measures provided by the DWP, thus exacerbating budgetary constraints, if it leads to increased labour market polarisation and insecurity, as many analyses suggest. We return to the likely nature and magnitude of technological-induced labour market change in the next section. Although the UK system does not rely heavily on the contributory principle (compared to many other European countries, for example) – so that individuals will remain entitled to out-of-work benefits even if they have patchy contribution, and the depletion of social insurance funds is not an issue – nevertheless these developments might conceivably threaten the financial sustainability of measures financed through payroll taxation.
However, as discussed below, trajectories of labour market change are highly uncertain and contingent. Outcomes will doubtless depend on the wider policy and institutional context. The nature and extent of the challenges that DWP will face thus also depend on broader industrial development policies and institutions, and whether these are supportive of the creation of abundant, highly-paid jobs; on policies aimed at promoting human capital development and labour market transitions; on the nature of industrial relations and employment protection systems; and on global governance structures in relation to corporate regulation and tax. There should be no presumption that permanently adverse labour market conditions are inevitable, although some disruption will surely occur (as it has already). In particular, there may be a need to reform employment protections for non-standard workers (Taylor et al., 2017). Nevertheless, even if labour market disruption is purely short-term and frictional in nature, demand for DWP’s social security functions will increase.
As well as affecting the prevalence of circumstances that require support, technological change may also affect policy design and the type of support required, with challenges in relation to the administration of means-tested, conditional benefits in the presence of increasingly volatile income fluctuations, more frequent employment transitions, and growing in-work poverty.
Conventional social security systems, such as DWP’s legacy system are based on a clear in-work / out-of-work dichotomy and the assumption of ‘regular’ (if not high) hours and wages in work. They are therefore ill-equipped for intermittent and insecure working patterns. Claimants moving in and out of work would have to make regular reapplications, thus introducing a ‘bureaucracy trap’ that could discourage claimants from taking job opportunities, for fear of losing benefits. There have also been issues to do with assessing irregular incomes for means-testing purposes. If earnings are irregular or ‘lumpy’, means testing can be problematic; the administration could close a claim based on temporarily high earnings, when the claimant has a low average income.
In principle, Universal Credit (UC) improves upon the legacy system in these respects, being paid in- and out-of-work. By design, UC ensures that claimants face positive incentives to do any amount of work, and that they face constant, transparent withdrawal rates that fall within acceptable bounds. Claimants do not have to re-apply as their employment circumstances change, but rather simply update their record to reflect their fluctuating earnings. In theory, UC should act as an automatic safety net, with payments adjusted to ensure an adequate level of income when earnings are low. Indeed, UC’s core features – consolidation of numerous working age benefits, and the harmonisation of administrative procedures and conditions for working and out-of-work claimants – are motivated by many of the same arguments that support BI.
However, although Universal Credit is arguably better equipped to cope with irregular work and fluctuating incomes, there are still difficulties assessing people accurately without error or inadvertently causing hardship. Problems include the excessive waiting period and payment volatility (due to ‘real-time’ response to changes in wages and other circumstances) which mean that UC payments and income shortfalls may not be in step, and intrusive conditionality (Millar, 2019). While not related to irregular employment and income patterns per se, many commentators have also criticised the intrusive and punitive nature of UC’s conditionality regime (Dwyer, 2019).
Another aspect of the Fourth Industrial Revolution that might present a challenge to the DWP relates to the human capital requirements of the emerging knowledge economy, and the greater need people will have to engage in lifelong learning. This might present challenges in relation to the provision of financial support for these specific circumstances.
There are two distinct manifestations of the Fourth Industrial Revolution that have profoundly influenced labour market structures and outcomes: ‘automation’ and ‘digitalisation’ (Martinelli, 2019a). The former term has typically referred to the mechanisation of tasks previously carried out manually, but increasingly embraces the use of digital technologies to carry out cognitive tasks as well. In this way, automation overlaps with the concept of ‘digitalisation’, which describes the transition to a new and distinct phase of capitalism, characterised by expansion in the use of computers and digital technologies and associated changes to socio-economic structures. Digitalisation invokes a broader range of phenomena to automation, indicating a generalised structural shift towards high-tech and knowledge-intensive activities in the economy. Thus, we can distinguish two main ways that technological change is affecting labour market outcomes: by providing technological substitutes for human labour (automation) and through the enablement of organisational change and the appropriation of cumulative technological gains through the exploitation (and creation) of high market entry barriers (digitalisation). Both phenomena appear to have polarising effects on the labour market.
In the most sensational accounts – following influential studies such as Frey and Osborne (2013) and Brynjolfsson and McAfee (2014) – technological advances will eventually give rise to mass unemployment by rendering human labour redundant at any (tolerable) wage rate. Automation will provide viable, more cost-effective substitutes for an increasingly wide array of complex (non-routine and cognitive) tasks – with technological unemployment as an inevitable consequence. There have been many attempts to estimate the potential impact of automation for job losses, many of which report concerning figures (see Winick, 2018 for an overview).
Estimations about the number of jobs that could plausibly be automated are not the end of the story, of course. Societal preferences for human labour may discourage automation in relation to certain tasks (Arntz et al., 2016). There may also be legal, ethical and institutional barriers to automation (PWC, 2018). It is also crucial to remember that alongside job destruction, digitisation is expected to lead to the creation of new jobs[1].
At the aggregate level, productivity effects countervail against displacement effects in several ways. Optimists point to historical evidence regarding the impacts of past episodes of technological change on aggregate employment trends (Autor, 2015). In particular, despite decades of declining employment within routine-intensive occupations, there are no clear signs that aggregate employment levels have suffered. Thus, when considering automation’s labour market effects we should not just focus on negative effects on overall employment levels.
A more immediate concern is how disruptive effects will be distributed and how they might contribute to inequality with respect to labour market outcomes. The first step in conceptualising the labour market effects is to understand that technological change will have differential impacts across different spheres of activity – different tasks, jobs, occupations, and sectors. According to the skill-biased technological change (SBTC) hypothesis, while demand for highly skilled labour – employed in the creation and marketisation of new technologies but also in professions to which new technologies are complements rather than substitutes – increases in line with technological adoption, pushing up employment levels and putting pressure on wages to rise. Meanwhile, demand for unskilled manual labour falls due to the availability of technological substitutes as well as lower relative levels of demand for ‘low-tech’ manufactures and commodities. Furthermore, increased supply of educated workers motivates employers to engage in technological upgrading according to a logic of cumulative causation (Lauder et al., 2018).
But this is not the end of the story. Recent studies have shown that technological change is biased in favour of non-routine cognitive and manual tasks. Routine occupations include a raft of mid-skilled and relatively well-paid jobs (bookkeeping, manually operating machinery and so forth). Thus, so-called ‘routine biased technological change’ (RBTC) has had the effect of hollowing out ‘middling’ occupations, leading to occupational polarisation through the relative growth of ‘lovely’ and ‘lousy’ jobs (Autor and Dorn, 2013; Goos et al., 2014). Due to the interplay of supply and demand conditions, substitution of technology for human labour raises the risk of unemployment for the relatively unskilled – especially those in routine-intensive occupations – and places downward pressure on wages at the bottom (and middle) of the income distribution.
The losers from the disappearance of low-skilled and routine jobs, and the nature of their loss, justify a particular focus. We do not need to assume that digitisation leads to the permanent disappearance of jobs without compensatory job creation to be concerned about the effects of the former; frictional unemployment, skills redundancy and distributional effects are all still important public policy issues irrespective of potentially benign long-term net employment effects. Alongside labour market transformation – whether skill- or routine-biased – certain competences will be subject to reduced demand, and “some people will not have the qualifications for those jobs that are or will be available… Even if they can find a job, this might be at a very low wage level” (Greve, 2017: 2). IPPR (2019) also find that the automation is likely to affect women disproportionately, given gendered occupational patterns.
Besides automation, we note that technological change could plausibly contribute to labour market dysfunction through several other channels, which we characterise here as aspects of ‘digitalisation’ – a generalised structural shift towards high-tech and knowledge-intensive activities in the economy.
One mechanism is that technological change facilitates new forms of digital organisation of working practices, thus enabling employers to extend their control over subordinate labour processes, for example through the development of ‘online platforms’ (known variously as the ‘gig economy’ or ‘Uberisation’). These developments may be positive for workers; advantages include greater flexibility over where and when they can work, the ability to identify and reach new customers and suppliers, and the emergence of opportunities to contract out or automate ‘undesirable’ tasks (OECD, 2016). In principle, these developments could boost employment levels, wages and conditions for some. But a number of studies (Degryse, 2016; Valenduc and Vendramin, 2016; Taylor et al., 2017) have concluded that the further expansion of online platforms is likely to have detrimental effects on employment security, as well as limiting workers’ opportunities for progression and providing low rates of job satisfaction.
More generally, digitalisation also enables the proliferation of the ‘platform’ economy and ‘Superstar’ firms (and related increases in market concentration) (Autor et al., 2017). Digitisation also enables firms to externalise subordinate labour processes via outsourcing and offshoring – thereby enabling globalisation, by intensifying processes of agglomeration and fragmentation in production networks (Lall et al., 2004). The intrinsic properties of technology give rise to numerous and pervasive market failures (Lall, 1992; Lucarelli and Fumagalli, 2008) which enable firms with various advantages (including transnationality, economies of scale and scope, and proprietary intellectual property) to dominate high-tech sectors, leading to increasing market concentration. This affects the relative bargaining positions of capital and labour, and permits firms to exploit their enhanced leverage to drive down wages and conditions (Autor et al., 2017). Concerns about technology’s potentially detrimental effects on labour market outcomes are bolstered by empirical evidence that the labour share of value-added has exhibited long-term decline (Dao et al., 2017).
In sum, there are three major forms of labour market dysfunction that appear likely to be exacerbated by technological change: (frictional) unemployment, low wages, and insecure employment. In combination, these outcomes connect intimately to the notion of dualisation, a process which grants ‘insiders’ access to stable, well-paid employment (and associated social protections) and excludes outsiders (Rovny and Rovny, 2017: 163). Although explanations for these phenomena usually focus on their institutional determinants, the latter clearly interact with broader structural factors such as technological change: automation and digitalisation contribute to the erosion of ‘insider’ jobs in routine occupations such as manufacturing, and increasing the prevalence of insecure (and poorly paid) service sector work in its place (Greve, 2017).
Arguably, technological and labour market change strengthen arguments in favour of UBI in numerous respects. However, such arguments are not uncontentious. There are normative objections and more pragmatic concerns around cost, affordability and labour market effects. These counter arguments are important in their own right, but also impose political barriers which are vital to consideration of UBI’s feasibility. Public policies create winners and losers in material terms, and may also be supported or opposed on ethical or 'cultural' grounds; even if UBI generates numerous important advantages and represents an financial improvement for most, this would not ensure that the policy would generate sufficient support would overcome entrenched opposition.
As the preceding literature review suggests, automation and digitalisation are already giving rise to profound occupation restructuring; frictional unemployment and skills redundancy in blue and white-collar routine occupations; the proliferation of poorly paid and insecure work; Being unconditional, UBI could provide more effective protection against income fluctuations than existing traditional provisions, whether based on targeting or contributory principles.
UBI might also provide more comprehensive de facto coverage and more adequate provision in relation to the alleviation of in-work poverty. It also has the advantage that, unlike means-tested benefits and those conditional on circumstances such as unemployment, it is not withdrawn as recipients increase their earnings or enter employment. This implies that under UBI, individuals face positive financial incentives to take any form of work, including 'mini jobs' and short-term and irregular contracts. UBI should also avoid so-called 'bureaucracy traps' – in which claimants are reluctant to enter employment due to risk aversion regarding job security and concern regarding delays in reapplying for benefit.
UBI may reduce the material and psychological burdens on recipients due to factors relating to labour market conditionality, bureaucratic effort, stigma, and the imposition of sanctions. Bureaucratic burdens such as these, as well as associated stigma, are likely to reduce take-up (Van Parijs, 2004) compared to UBI, which should approach substantive universality provided efforts are taken to ensure the coverage of marginalised groups (De Wispelaere and Stirton, 2013). Bureaucratic eligibility tests (especially means tests) are also very costly for the state (Van Oorschot, 2002), and look likely to become more so as labour market transitions, fluctuations in earnings – and thus, changes to qualification for benefit and the level of payment to which recipients are entitled – occur more and frequently.
Technological change also raises new opportunities within the knowledge-intensive IT, creative and professional service sectors, and leads to new forms of flexible working practice. How well does UBI suit these emerging opportunities?
People will increasingly need to engage in lifelong education and skills upgrading to participate in emerging dynamic sectors of the economy; UBI could facilitate that by allowing people to take time off work or reduce their hours. High-tech IT sectors such as software design and development require long periods of research – which represent large sunk costs – but could have very low marginal costs associated with their dissemination and use. In this context, a UBI could encourage people to undertake innovative activities, underwritten by the income security that a UBI provides. A similar case can be made with respect to engagement in entrepreneurial self-employment – a phenomenon which has been increasing steadily in the era of the knowledge economy – and in creative and artistic work. Such work, like the 'gig economy', is usually atypical in terms of contractual arrangements, and involves a high degree of risk. In these ways, we observe complementarities between UBI and the need of the digital economy for a flexible, high-skill workforce.
Other justifications for UBI in the context of technological change are based on macroeconomic concerns, in relation to a declining labour share and the emergence of 'winner takes all' capitalism. A declining labour share of value is problematic not only because for equity reasons, but also because of the relative propensities of labour and capital to spend money in the economy, and thus contribute to demand multiplier effects. Unemployment and low wages matter instrumentally, to the extent that the capitalist system relies on maintenance of workers' purchasing power (Crocker, 2017). Furthermore, inequality feeds also through into financial instability through a variety of mechanisms (Michell, 2014). According to Crocker (2017), the consequence of a mismatch between GDP (productivity) growth and household incomes – itself partially a consequence of the declining labour share attributed to technological change – is that consumption has only been sustained at acceptable levels through growing levels of private debt. Far better, then, to fund a UBI through “'overt money funding' or 'helicopter money', i.e. an annual deficit which is simply written off, as current recurring deficits essentially are written off to ever ascending but entirely notional national debt”. A UBI would inject demand into the economy in much the same way as recent episodes of 'quantitative easing', only more effectively and in far more egalitarian fashion (Standing, 2018).
In sum, proponents argue that UBI is a ‘minimally presumptuous’ welfare policy that copes well with the vagaries of contemporary labour market dynamics and facilitates the adoption of any conceivable pattern of employment, providing a secure income floor to which one can add income from intermittent and poorly paid work. However, there are numerous and convincing countervailing arguments, each of which exacerbates UBI’s political barriers.
UBI simply does not function very effectively at consumption smoothing, because for relatively well-paid individuals it would still leave very significant income shortfalls in comparison to previous income. This reduces the effectiveness of UBI as a short-term response to labour market shocks. At the very least, it suggests that insurance-based, earnings-related benefits would also be required to supplement the income floor provided by a UBI. It also gives rise to political barriers. UBI’s flat-rate structure may discourage support from those who already have generous social insurance entitlements, either because of the threat that existing provisions will be replaced by a less generous UBI payment, or because UBI threatens to erode their privileged access to welfare (Martinelli and Chrisp, 2020). Thus, although left-leaning individuals, those with cosmopolitan cultural preferences, and those subject to more pronounced labour market risks are more likely to favour UBI, these preferences are moderated by ‘insiderness’, which drives opposition to UBI. The findings cast doubt on the possibility that technological and associated labour market change will augment (and broaden) basic income’s support base, thus improving the prospects that the policy will receive party support, ascend the legislative agenda, and achieve implementation.
Another objection to UBI is that its unconditional nature permits idleness and encourages labour market exit, thus undermining goals in relation to the promotion of a dynamic market economy. This relates to practical objections that a generous UBI would replace employment income and discourage labour market participation, thus becoming rapidly unsustainable. For some critics, UBI is “capitulation to deregulation and exploitation, not a solution to it” (Coote and Yazici, 2019: 4). For example, Mestrum (2018) suggests that UBI would act as a wage subsidy for low paying firms. In this case, UBI does not simply respond to labour market polarisation and dualisation, but actively encourages it. Opponents suggest that the absence of labour market conditions would lead to the entrenchment of labour market disadvantage and the emergence of “a new underclass stuck at UBI level and an economic elite that would reap the greatest benefits” (Mayer, 2018). It is worth noting that many of UBI’s critics endorse a ‘social investment’ model of welfare state reform; rather than providing passive compensation, social investment reforms actively promote labour market integration, by enhancing human capital and facilitating labour market and life course transitions (Hemerijck, 2017). Arguably, such a model may be more effective at ‘upskilling’ the workforce for the emerging challenges of digitalisation than UBI’s ‘passive’ approach. The debate for and against labour market conditions is difficult to resolve, since there is limited evidence about how well UBI compares to conventional policies.[2] Whether low-skilled individuals would use their UBIs to retrain and invest in human capital and entrepreneurship in the context of occupational decline is also not clear. In any case, there is considerable normative opposition to UBI’s unconditional nature, and the way that it departs from accepted notions of fairness, embodied by the primacy of the principles of reciprocity and need in the provision of social welfare (Martinelli, 2017a).
Perhaps the most significant objection and political barrier relates to UBI’s significant opportunity cost, especially in relation to functional equivalents. The various functions of UBI may be carried by alternative instruments. This means that UBI must be assessed in relation to other hypothetical alternatives, not just the status quo ante, which may come at lower fiscal cost, and may be closer to existing institutional arrangements.
UBI probably requires a combination of large tax increases and/or significant cuts to existing expenditure. This generates political barriers as awareness of trade-offs – specifically, the likely implications for tax rates and cuts to other welfare provisions – inevitably reduces support (Ipsos Mori, 2017; Busemeyer and Garritzman, 2017). Any reconfiguration of existing social insurance and assistance benefits may engender opposition from vested interests.
UBI’s suitability as a solution to various labour market issues, as well as its political viability, are complicated by the diversity of forms the policy can take. UBI refers to an overarching family of policies representing a wide range of specific schemes. UBI schemes vary with respect to a number of important parameters, but perhaps most significantly with respect to their payment level, their interactions with the wider welfare system, and their funding mechanisms. UBI is usually understood to replace existing welfare transfers at least partially. The manner in which other payments are replaced can vary in nature and extent. Related to this, ‘full’ UBI schemes – pitched at levels that permit most other payments to be eliminated wholesale – may be contrasted with ‘partial’ schemes, which are pitched at levels designed to replace fewer benefits and/or may leave them in payment, with the levels of transfer adjusted downwards to take into account additional UBI payments (Kela, 2016).
Alternative forms of UBI give rise to a specific pattern of winners and losers, and are likely to appeal to (or dissuade) different coalitions of political actors. Microsimulation analysis for the UK (Martinelli, 2017b) shows that schemes that aim to replace the myriad means-tested supports (‘full schemes’) either lead to unacceptable household losses (including some disadvantaged groups falling deeper into poverty) or simply cost too much. On the other hand, it is possible to design an affordable and adequate ‘partial’ UBI scheme. However, this comes at the cost of (at least partly) forfeiting some key advantages that motivate interest in UBI in the first place, since these only arise as a consequence of the elimination of means testing and related conditionality from the welfare system. The advantages which may not materialise for partial schemes include the provision of genuine income security, drastic reductions in bureaucratic complexity and the minimisation of poverty and unemployment traps. Thus, UBI proponents face a three-way trade-off in policy design between affordability, adequacy, and securing the full advantages of UBI as a radical simplification of existing welfare policy.
Martinelli and Chrisp (2019) argue that while UBI embodies aspects of both left and right simultaneously, depending on policy design features relating to funding and payment level, specific UBI schemes are more appropriately described as conforming to progressive or conservative versions. Polling commissioned by the IPR (Ipsos Mori, 2017) shows that when respondents were asked to consider how the UBI would be financed (tax rises, benefit cuts or a combination of both), support fell dramatically. Furthermore, different groups of voters have differing levels of enthusiasm for UBI. For example, 40% of conservative-leaning respondents support UBI in general terms, compared to 63% of Labour-leaning respondents, with 41 and 17% respectively opposing the idea. Even more crucially, among supporters of UBI, ideological opponents support different types of scheme. Support among Conservative-leaning respondents grows to 49% when asked to consider a UBI funded through benefit cuts, but falls to 22% for schemes funded through tax rises. While support among Labour-leaning respondents drops when asked to consider any concrete funding model, this group has a clear preference for schemes which rely on tax increases (43% in favour) – particularly on wealth as opposed to income – compared to those that require benefit cuts (34% in favour). Thus, it is likely that apparently broad-based support will not survive the specification of a concrete direction for reform.
Arntz, M., Gregory, T. and Zierahn, U. (2016), “The Risk of Automation for Jobs in OECD Countries: A Comparative Analysis”, OECD Social, Employment and Migration Working Papers no. 189, OECD Publishing: Paris. http://dx.doi.org/10.1787/5jlz9h56dvq7-en [accessed 1/6/2019].
Autor, D. (2015). “Why Are There Still So Many Jobs? The History and Future of Workplace Automation.”
Autor, D., and Dorn, D. (2013). “The growth of low-skill service jobs and the polarization of the US labor market.” American Economic Review, 103(5), 1553-97.
Autor, D., Dorn, D., Katz, L. F., Patterson, C., and Van Reenen, J. (2017). “The fall of the labor share and the rise of superstar firms.” NBER Working Papers no. 23396. National Bureau of Economic Research.
Brynjolfsson, E., and McAfee, A. (2014). The second machine age: Work, progress, and prosperity in a time of brilliant technologies. WW Norton and Company.
Busemeyer, M., and Garritzmann, J. (2017). "Public opinion on policy and budgetary trade-offs in European welfare states: evidence from a new comparative survey." Journal of European Public Policy, 24(6), 871-889.
Coote, A., and Yazici, E. (2019). Universal Basic Income: A Union Perspective. London: New Economics Foundation.
Crocker, G. (2017). “The Affordability of Basic Income: A radical heterodox economics view.” IPR blog. http://blogs.bath.ac.uk/iprblog/2017/10/10/the-affordability-of-basic-income-a-radical-heterodox-economics-view/ [accessed 1/6/2019].
Dao, M., Das, M., Koczan, Z., and Lian, W. (2017). “Why is Labor Receiving a Smaller Share of Global Income? Theory and Empirical Evidence.” IMF Working Paper.
De Wispelaere, J. and Stirton, L. (2013). “The Politics of Unconditional Basic Income: Bringing Bureaucracy Back In.” Political Studies, 61(4): 915-932.
Degryse, C. (2016). Digitalisation of the Economy and its Impact on Labour Markets. ETUI Research Paper, Working Paper no. 2016/02.
Dwyer, P. (ed.) (2019). Dealing with welfare conditionality: Implementation and effects. Policy Press.
Frey, C., and Osborne, M. (2013). The Future of Employment: How Susceptible are Jobs to Computerization? University of Oxford.
Goos, M., Manning, A., and Salomons, A. (2014). “Explaining job polarization: Routine-biased technological change and offshoring.” American Economic Review, 104(8): 2509-26.
Gray, A. (2017). Behavioural Effects of a Citizen’s Income on wages, job security and labour supply. Citizen’s Income Trust. http://citizensincome.org/research-analysis/behavioural-effects-of-a-citizens-income-on-wages-job-security-and-labour-supply/ [accessed 1/6/2020].
Greve, B. (2017). Technology and the future of work: The impact on labour markets and welfare states. Cheltenham: Edward Elgar Publishing.
Hemerijck, A. (ed.) (2017). The Uses of Social Investment. Oxford: Oxford University Press.
Institute for Public Policy Research (IPPR) (2019). The future is ours: Women, automation and equality in the digital age. IPPR Centre for Economic Justice. https://www.ippr.org/files/2019-07/the-future-is-ours-women-automation-equality-july19.pdf [accessed 1/6/2019].
Ipsos Mori (2017). Polling into Public Attitudes to Basic Income. Questionnaire written and developed by Joe Chrisp, and commissioned by the Institute for Policy Research (IPR). https://www.ipsos.com/ipsos-mori/en-uk/half-uk-adults-would-support-universal-basic-income-principle [accessed 1/6/2019].
Kela (2016). From Idea to Experiment: Report on universal basic income experiment in Finland. Kela Working Paper 106/2016. Helsinki: Kela.
Lall, S. (1992). “Technological capabilities and industrialization.” World development 20(2): 165-186.
Lall, S., M. Albaladejo and J. Zhang (2004). “Mapping fragmentation: electronics and automobiles in East Asia and Latin America.” Oxford Development Studies, 32(3): 407-432.
Lauder, H., Brown, P. and Cheung, S. Y. (2018). “Fractures in the education–economy relationship: The end of the skill bias technological change research programme?” Oxford Review of Economic Policy, 34(3): 495-515.
Lucarelli, S. and Fumagalli, A. (2008). “Basic Income and Productivity in Cognitive Capitalism.” Review of Social Economy, 66(1): 71-92.
Manyika, J., Lund, S., Chui, M., Bughin, J., Woetzel, J., Batra, P., Ko, R. and Sanghvi, S. (2017). Jobs lost, jobs gained: Workforce transitions in a time of automation. McKinsey Global Institute.
Martinelli, L. (2017a). Assessing the Case for Universal Basic Income in the UK. IPR Policy Brief, September 2017. Bath: Institute for Policy Research, University of Bath.
— (2017b). Exploring the distributional and work incentive effects of plausible illustrative basic income schemes. IPR Report, May 2017. Bath: Institute for Policy Research, University of Bath.
— (2019a). Basic income, automation and labour market change. IPR Report, September 2019. Bath: Institute for Policy Research, University of Bath.
— (2019b). “A basic income trilemma: affordability, adequacy, and the advantages of universal and unconditional welfare.” Journal of Social Policy, DOI: https://doi.org/10.1017/S0047279419000424.
Martinelli, L., and Chrisp, J. (2019). “Neither Left nor Right” in Torry, M. (ed.) The International Handbook of Basic Income, Palgrave Macmillan.
— (2020). Individual preferences for basic income in the context of labour market change: Automation risk, precarity, and the moderating roles of higher education, union membership and income. Unpublished manuscript, submitted to Journal of European Social Policy.
Mayer, H. (2018). “No Need For Basic Income – Five Policies To Deal With The Threat of Technological Unemployment.” in Van Parijs (ed.) Basic Income and the Left: A European Debate.
Mestrum, F. (2018) “Why basic income can never be a progressive solution – A response to Van Parijs” in Van Parijs (ed.) Basic Income and the Left: A European Debate.
Michell, J. (2014). “Factors generating and transmitting the crisis: Functional distribution of income.” Financialisation, Economy, Society and Sustainable Development (FESSUD) Working Paper no. 41. Leeds: FESSUD.
Millar, J. (2019). “Universal Credit, means-testing and social security” Social Policy Association 50th Anniversary Blog, no. 15. http://www.social-policy.org.uk/50-for-50/universal-credit-means-testing/ [accessed 1/6/2020].
Millar, J. and Whiteford, P. (2020). “Timing it right or timing it wrong: how should income-tested benefits deal with changes in circumstances?” Journal of Poverty and Social Justice, 28(1), pp.3-20.
Millar, J., and Bennett, F. (2017). “Universal Credit: Assumptions, Contradictions and Virtual Reality.” Social Policy and Society, 16(2), 169-182.
OECD (2016). Automation and Independent Work in a Digital Economy. Policy Brief on the Future of Work. Paris: OECD.
Price Waterhouse Coopers (PWC) (2018). Will robots really steal our jobs? An international analysis of the potential long term impact of automation. https://www.pwc.co.uk/economic-services/assets/international-impact-of-automation-feb-2018.pdf [accessed 1/6/2019].
Rovny, A. and Rovny, J. (2017). “Outsiders at the ballot box: operationalizations and political consequences of the insider-outsider dualism.” Socio-Economic Review, 15(1): 161-185.
Standing, G. (2018). “Basic income pilots: a better option than quantitative easing” in Van Parijs (ed.) Basic Income and the Left: A European Debate.
Taylor, M., Marsh, G., Nicol, D., and Broadbent, P. (2017). Good work: the Taylor review of modern working practices. London: Department for Business, Energy and Industrial Strategy.
Valenduc, G. and Vendramin, P. (2016). “Work in the digital economy: sorting the old from the new.” ETUI Research Paper, Working Paper no. 2016.03.
Van Oorschot, W. (2002). “Targeting welfare: on the functions and dysfunctions of means testing in social policy”, in Townsend, P. and Gordon D. (eds.) World Poverty: New Policies to Defeat an Old Enemy. Bristol, Policy Press.
Van Parijs, P. (2004) “Basic Income: a simple and powerful idea for the twenty-first century.” Politics and Society, 32(1): 7-39.
Widerquist, K. (2005). “A failure to communicate: what (if anything) can we learn from the negative income tax experiments?” The Journal of Socio-Economics, 34(2005): 49-81.
Winick, E. (2018). “Every study we could find on what automation will do to jobs, in one chart.” MIT Technology Review, January 25th 2018. https://www.technologyreview.com/s/610005/every-study-we-could-find-on-what-automation-will-do-to-jobs-in-one-chart/ [accessed 1/6/2019].
World Economic Forum (2016). The Future of Jobs Employment, Skills and Workforce Strategy for the Fourth Industrial Revolution. http://www3.weforum.org/docs/WEF_Future_of_Jobs.pdf [accessed 1/6/2019].
[1] Based on a survey methodology, World Economic Forum (2016) estimates that jobs lost will outnumber jobs created, leading to a net loss of over 5 million jobs – mainly concentrated in routine white collar (admin) jobs – across the 15 economies in their sample. However, other reports (e.g. Manyika et al., 2017) suggest that there will be net employment gains, while others still are agnostic on the net effects of digitisation on employment levels (PWC, 2018).
[2] UBI would have contradictory effects on labour supply, the net effect of which would be hard to predict (Martinelli, 2017a, 2017b). Specifically, effects would be determined by variation at the level of the individual (e.g. their preferences, earning potential, and financial relationships with family members) and in the features of the UBI (especially with respect to generosity). In Gray’s (2017) concise depiction, “the higher the UBI in relation to the individual’s hourly wage, the greater would be the likely reduction in labour supply from people already in paid work”. Empirical evidence is rather limited, and mainly drawn from two sources: a series of negative income tax [a close cognate policy to UBI] experiments across the US and Canada between 1968 and 1980, and, more recently, UBI trials in Namibia (2008-10), India (2011-13), and Finland (2017-18). In summary, the US and Canadian experiments data suggest that the provision of an unconditional income floor was a marginal disincentive to work. On average across the US studies, men reduced their working hours by 6%, married women by 19%, and single women by 15% (Hum and Simpson, 1993). The reduction in labour market effort was even smaller for the Canadian study: 1%, 3% and 5% respectively for the same categories (ibid.). However, as Widerquist (2005) points out, it is by no means clear that even the larger reductions in labour supply would damage the case for UBI. Firstly, the labour market exodus predicted by UBI’s fiercest critics clearly did not materialise; and secondly, if the UBI permitted people to balance their work and home lives more effectively – especially women overburdened by a combination of domestic and paid work – the reduction in work effort could conceivably have contributed to increased societal welfare. This relates to arguments that UBI scholars have made in favour of the provision of an ‘exit option’ to strengthen the bargaining position of workers and to enable them to engage in unpaid but socially-valuable activities. The recently concluded Finnish experiment looks likely to paint a similarly ambiguous picture, with UBI recipients neither more nor less likely to enter employment than the control group (Martinelli, 2019c).