Training and the acquisition of skills have become increasingly urgent policy issues in the industrial world. There is widespread concern that many employees are insufficiently skilled – a deficiency with serious economic consequences: unemployment is concentrated among unskilled workers while skill shortages can lead to inflationary pressures. What is more, productivity growth, innovation and product quality all rest critically in the hands of skilled employees and will suffer in their absence.

For many years, mainstream economists believed that the ‘invisible hand’ would invariably lead people to acquire skills as long as the resulting benefit to society exceeded the associated cost. Over the past decade, however, it has become clear that much of the time the free market fails to provide adequate incentives to train. A recent CEPR publication explores these issues in a systematic account of the causes, consequences and policy implications of market failures in training provision and skills acquisition.

A central premise of the book is that since the market for training tends to be characterized by imperfect competition and imperfect information, people generally do not receive compensation for the training they acquire and/or provide. For example, when there is a chance that skilled employees will be poached by employers who did not train them, the trainers and trainees fail to appropriate all the returns from training. Hence the free market will not generate sufficient training.

This market failure can be amplified through complementarities between labour and capital: employees who acquire skills make more effective use of the capital equipment they work with and enable firms to introduce more sophisticated and productive machines. In reverse, this process implies that deficient investment in human capital reduces the productivity of physical capital, leading to deficient capital investment and poor economic growth.

Paradoxically, the greater the proportion of trained people in an economy, the stronger the possibility of future underinvestment in skills. In these circumstances, firms can attract skilled employees more quickly in the job-matching process, raising their market power in wage negotiations and reducing workers’ returns from training. At the opposite end of the spectrum is what’s known as the ‘low skills, bad job trap’, where firms create few skilled vacancies because there are few skilled workers available and few workers acquire skills because there are few skilled vacancies.

Credit constraints, arising from the fact that human capital cannot be used as collateral against loan default, can also prevent employees from acquiring sufficient training. In addition, there is the problematic interaction between skills and innovative performance: when successful innovation requires highly trained workers, economies can get stuck in a vicious cycle in which firms do not innovate sufficiently because the workforce is insufficiently skilled and workers do not train sufficiently because there is insufficient demand for them from innovating firms.

The predominant strategic choices made by employers reflect the extent to which a skilled workforce is available or easily developed. But they also affect the supply of skills by signalling to individuals the value of investment in vocational training and education.

These arguments suggest that market failure produces ‘skills gaps’ or deficiencies in the availability of trained employees. This is an issue of great concern in the 1990s since the composition of labour demand has shifted over the previous decade in favour of skilled workers. The shift is particularly strong in workplaces that have a relative intensity in research and development, lending some credence to the view that labour-saving technological change has played an important role in the restructuring of employment.

The book explores the implications of skills gaps for product quality and export performance. In the UK, for example, skill shortages are higher on average and more variable over the business cycle than in comparable economies. At the same time, shortages of unskilled labour are comparatively rare.

Relative to Germany, the UK’s main deficiency is in craft and technical skills, which leads to the UK exporting relatively poor quality products. In general, the less skilled a country’s workforce, the greater the tendency to produce non-traded commodities, such as some kinds of services, rather than traded ones, such as manufactured goods. This is because non-traded commodities are often more shielded from competition. The result is a relatively poor export performance.

Of course, it would be naive to suppose that whenever the market fails, the government can put it right: ‘government failure’ is just as likely in the provision and finance of training. For example, there is no reason to expect the state to be a more efficient provider of training than private sector firms. Firms’ skill requirements are highly idiosyncratic, while government officials cannot be expected to know the millions of required skills, let alone provide them. And the concentration of government control over training programmes in a central agency may insulate decision-makers from local labour market conditions and needs.

Since market economies differ markedly in their political institutions and the size and function of their public sectors, training programmes that entail massive government failures in one country may entail little in another. As a result, inter-country comparisons should be made with caution. Institutions that have successfully promoted training in one country – such as the apprenticeship system in Germany – may not be easily exportable to other countries, at least without a careful consideration of other institutional changes that may be required.

Moreover, since countries also differ in the lines they draw between private sector and government activities, as well as in their social norms governing competition, what is a market failure in one country may well not be in another. For example, the serious market failure that arises when firms poach trained employees from one another is much more likely to be a problem in countries like the UK and the US than in Japan, where the norms and principles governing competition among firms are far more circumscribed.