Introduction: technology, jobs and skills.

Riley, Rebecca

Throughout history technological advance has been a source of increasing prosperity for many. There is also little doubt that advances in technology have affected the way we work and the skills we employ in our jobs. With the spread of the modern computer, the past decade has witnessed a large body of research focussing on the relationships between technical change and the labour market, addressing issues such as the effects of recent technologies and the diffusion of computers on workplace organisation, the organisation of production, income distribution and the demand for skills, and in turn the way in which the skill structure affects technology adoption. This issue of the Review brings together five articles providing new insights and perspectives on the relationship between technology, jobs and skills.

The question of whether and by how much computer use affects pay has been much debated. The first article, by Francis Green et al., reviews the literature on this topic, which ranges from studies that suggest computers are associated with significant pay premia for workers who use them in their jobs or who work in firms where computer use is relatively widespread, to studies that suggest there is no impact of computer use on pay, once other attributes that appear to be correlated with computer use are taken into account. Next, based on data collected through a series of surveys, the most recent of which is the 2006 Skills Survey, this article provides an up-to-date picture of the diffusion of computers in British jobs and how this compares to the situation only ten years ago. Since then, the proportion of employees who describe computing as central to their job has risen from 1 in 3 to 1 in 2. The proportion of employees who report using computers at a complex or advanced level has risen from fewer than 1 in 6 to more than 1 in 5. Although these shifts over time are visible within age, gender and education groups, the strong positive correlation between computer skills and qualification levels persists and has in many instances strengthened. Interestingly, this article shows that alongside the rise in computer skills employed in British jobs has been a rise in a set of skills described as influence skills. These capture amongst other factors the importance in jobs of persuading and influencing others, analysing complex problems in depth and planning the activities of others. Analysing wages, the authors find evidence to suggest that, independently of a range of other skills, computer skills do carry a premium in the British labour market today. One new insight coming from this analysis is that the computer pay premium occurs in combination with influence skills, which, the authors hypothesise, are likely to be necessary for the efficient deployment of computers. Further, the analysis in this article suggests that this interaction between influence skills and computers has gained in significance over time, becoming important only recently.

The second article, by Kirby and Riley, investigates the relationship between industry investment in information and communication technologies (ICT) and the premia paid to two other dimensions of skill or human capital--experience accumulated on-the-job with a particular employer and years spent in full-time education. Based on an analysis of the Labour Force Survey for Great Britain over the period 1994-2001, the study finds that those industries where schooling investments achieved the highest returns or where these were rising over time were the same industries where ICT investments expanded most substantially over the period considered, a pattern which is picked up in the statistical analysis. Interestingly, the analysis also suggests that in these same industries, notably the business services sector, the communications sector, and the manufacture of electrical and optical equipment industries, the return to job-specific experience or tenure was declining. The results of this analysis suggest that a 10 percentage point increase in the share of ICT capital in total capital is associated with a reduction in the wage premia associated with investing in an additional year of job-specific experience of 0.1-0.4 percentage points and an increase in the differential to investing in an additional year of schooling versus an additional year of job-specific experience of approximately I percentage point. One interpretation of these linkages in the data is that the introduction or diffusion of new technologies has been associated with the redundancy or obsolesence of some job-specific skills.

Another trend affecting jobs and the demand for skills is the globalisation of trade and the consequent offshoring of production, which with recent developments in technology affect an increasing number of sectors and jobs. Adopting a novel approach, Hijzen and Swaim assess the effects of offshoring on labour demand within industry sectors for seventeen high income OECD countries, conditional on the level of output and, separately, for given output prices. This distinction enables them to measure the direct effect on jobs held domestically that occurs when part of the industry production process is relocated abroad, and the indirect effects that occur when, as a consequence, the reduction in unit costs allows output to expand, creating new employment opportunities. They find evidence of job destruction associated with offshoring within the industry, but, taking into account the productivity gains associated with offshoring, the authors find no adverse and indeed a small positive effect of offshoring on industry employment.

Spitz-Oener examines the way in which the East German labour market made the leap from socialist economy, using largely out of date production techniques, to modern market economy. One question she asks is whether the skills acquired through the East German education system and employment under the socialist regime were transferable to the new economic order and production technology. Analysing data from the German Qualification and Career Survey she finds a remarkable similarity in job tasks, categorised according to their substitutability or complementarity with computers, of employees in East and West Germany in 1991, soon after the collapse of the German Democratic Republic. She also finds that changes in job tasks in East Germany, between 1991 and 1999, were very similar to those observed in West Germany, with a shift towards analytic, interactive and non-routine manual tasks away from routine cognitive and routine manual tasks. Interestingly, she finds that these shifts are not explained by cohort effects, which one might have expected to be influential given the historical experience of East Germany.

In the last article of this issue of the Review, Eve Caroli puts forth an argument as to why the diffusion of ICT may have resulted in a change in human resource management within firms away from the use of internal promotion strategies towards strategies based on external labour flexibility. She suggests that firms' reliance on internal versus external labour markets depends to a large extent on the way in which firms' competences, knowledge and skills are organised, i.e. the degree to which these are embodied within firms or in individuals. When firms' competences are embodied in individuals, it becomes more important for firms to retain these individuals. She argues that knowledge cannot be embodied in the organisation, independently of individuals, unless it is largely codifiable in nature, where tacit knowledge is by definition embodied in individuals. With the exception of R&D occupations, she suggests that ICT is likely to have shifted the balance away from tacit and towards codifiable knowledge. By logical conclusion this wound imply that firms need rely less on internal promotions. This line of reasoning provides an alternative rationale for the patterns in the British data described in Kirby and Riley in this Review. If skills and competences are increasingly codifiable and thus embodied in the firm, rather than the individual, it is easy to see that this might devalue the returns to individuals' job-specific experience.