Robots and the Workplace of the Future effective use of technology will increasingly depend on skilled workers with deep, company-specific process knowledge. If investment in automation is to be effective, it must be accompanied by an investment in human capital. As we have discussed in Section 4., competitive advantage will come from the successful implementation of human-machine collaborative processes. In many industries, this requires workers who combine technical /digital skills with deep process knowledge, much of which is company-specific. Company-specific process knowledge not only improves return on capital spending, it is also shown to improve corporate innovation. One study found that 75% of innovations are new processes and products in traditional manufacturing (D'Acunto 2014). Another found that temporary contracts lead to reduced investment in research and development, particularly in companies that depend on a historically accumulated, company-specific knowledge base (Kleinknecht, van Schaik and Zhou 2014). The question is who will pay - government, employers, employees, or a combination of all three? The European Political Strategy Centre (ESPC), the European Commission’s in- house think-tank, argues that ‘the increased likelihood of job changes, contractual and on- demand work, means that ‘new arrangements will have to be crafted to effectively share the burden of training between individual workers, employers, and the state. Innovative public– private partnerships between schools, universities and private training providers, should be encouraged’ (European Political Strategy Centre 2016). A study by Accenture, Harvard Business School and Burning Glass, on the other hand, argues this investment must come from companies, particularly for middle-skills requirements, though SMEs will require government support and incentives to fund training programmes. (Accenture 2016). The evidence on corporate investment in human capital is mixed. Whilst investment in corporate training appears to be static, or increasing, only half to two thirds of companies offer vocational training to employees30 . Meanwhile, the percentage of employment accounted for by agency, temporary, part‑time and self‑employed workers - typically do not have access to corporate training programmes - is increasing31. Specific recommendations on how to fund skills training are beyond the scope of this paper. However, evidence would indicate that, whatever the role of the state and educational 30 Corporate spending on training has increased in the U.S over the last 6 years according to (Bersin by Deloitte 2015) and (Training magazine 2017). In EU 28 countries, spending on corporate vocational training remained constant between 2005 and 2010, when figures were last reported (Eurostat n.d.). However, whilst an estimated two thirds (66 %) of EU 28 companies provided continuous vocational training to their employees in 2010 – up from 60% in 2005 (Eurostat n.d.), a U.S survey by Accenture and Harvard Business School reports only 45% of companies offering internships or apprenticeships for middle-skills jobs (Accenture 2016). 31 Temporary and part‑time work and self‑employment accounted for about a third of total employment in OECD countries in 2015 (OECD 2015). Economists Lawrence Katz and Alan Krueger claim all of the net employment growth in the U.S. economy from 2005 to 2015 – 9.4 million jobs – has been in alternative work arrangements (Katz and Krueger 2017). A survey by Harvard Business School of its alumni showed that 49% of respondents said their firms prefer to outsource work when possible rather than hire additional employees and, when hiring, are increasing their use of part-time workers (Accenture 2016).. One out of three workers in Japan – the traditional bastion of ‘a-job-for- life’- are employed on fixed-term contracts according to one source (Japan Industry News 2016). March 2018 Positioning Paper 26