Since before Adam Smith, labour markets have been known to differ greatly from other markets. Workers cannot be separated from the services they provide, and their motivation and beliefs critically affect how they work. Labour is not exchanged on a labour market: instead, people contract to supply labour in a form to be specified in detail (and within certain limits) only after they have been hired. The employment relation is often of considerable duration. In 1991, the percentage of workers with their current firms for ten years or more was UK, 29 per cent; US, 27 per cent; Germany, 41 per cent; and Japan, 43 per cent (OECD 1993). As many workers change firms only occasionally, it is clear that many employment problems are the result of a complex mix of market and organizational causes. Which of these we believe is the predominant cause will condition whether we start from price theory, stressing the analogy with other markets, as would most neo-classical economists, or from a more organizational or institutional approach. To explore the differences between these I shall look at labour allocation and work incentives.
The majority, ‘neo-classical’, tradition emphasizes competition and flexible prices, believing that these will, in the long run, provide the best overall outcomes for the greatest number. Because firms compete for workers, those with dangerous or unpleasant working conditions have to pay more to attract them. If they need workers with particular skills, then by offering higher wages, they provide an incentive for people to undertake the necessary training. The price mechanism works better than government educational planning based on surveys of skill needs because it forces firms to say not only what skills they would like, but also how much they would be worth to them. In the context of unemployment, with flexible wages and no restrictions on laying workers off, firms would hire extra labour until workers no longer thought the wages offered worthwhile.
Faced with bad working conditions, skill shortages or unemployment, neo-classical economists would generally look first for failures of, or obstacles to, the price mechanism. Is it better to seek to improve dangerous or unhealthy working conditions by means of government regulation which is costly to run and is often easily evaded, or to rely on the price mechanism? The latter means that employers have to pay more to attract workers, and so have a financial incentive to invest in technology to improve conditions. If accident rates remain high, is it because employers enjoy monopsonistic power in their local labour markets, or because of some perversity in the price mechanism? For example, employers may fear that if they are the first to invest in safer equipment they will attract the most accident-prone workers (Elliot 1991).
If an industry faces persistent problems of skill shortages, are the incentives for people to train right? Do the rewards for skilled jobs adequately compensate for the outlays involved? Are employers failing to get any return on training they provide because workers are leaving once their training is complete? If the skills are transferable, should the trainees bear most of the cost, perhaps as low trainee allowances? In the case of unemployment among low-skilled workers, are minimum wage laws or collective agreements setting pay levels which make it uneconomic to employ them—‘pricing them out of jobs’?
Treating wages as if they were market prices, and labour as a freely traded commodity, arguably comes closest to reality when looking at the markets for casual labour, where job durations are typically rather short, and for craft and professional occupations where skills are transferable between firms.
However, the clear analytical prescriptions of the theory become clouded when looking at relations within the enterprise, and when looking at the institutions which characterize modern labour markets such as trade unions and training organizations. A great number of job changes occur within firms; through promotions and job reassignments, as indeed do interregional job moves (about 40 per cent of such moves in the UK in the late 1980s). It has been common to contrast the allocation of labour between firms, on ‘external’ or ‘local’ labour markets, with that within the firm, on ‘internal’ labour markets (ILMs). Whereas the former involves movement between firms, or between unemployment and jobs, the latter involves movement between jobs or employment locations within firms.
Whether wages act in the same way on ILMs as they are reputed to do on local labour markets is hotly debated (e.g. Rosenberg 1989). Some argue that the force of competition among workers is weak compared with inter-group power relations, and once workers have been hired, management’s primary task is to organize and motivate staff, not negotiate market prices.
The reasons that firms develop ILMs, and hence their overall significance for economic theory, are also under debate. Building on price theory, the neoclassical approach seeks to explain ILMs as a response to certain kinds of failure in the market system. For example, ILMs may develop in firms where the necessary skills are lacking on their local labour markets; or firms may develop ILMs to encourage workers to take a long-term interest in their firm’s success. In the first case, many firms have their own unique production methods, or a particular style of service, and then develop their own training programmes linking these to job progression, thus creating an ILM (Doeringer and Piore 1971). In the second case, because workers have much better knowledge of their work and the production process than management, they have a strong incentive to manipulate information in order to negotiate a more favourable effort-bargain, but which reduces productivity. By creating an ILM, the employer gives workers a long-term stake in their firm’s success so that they will wish to raise productivity (Williamson 1975). In both explanations, the ILM remains nested within the general framework of price theory.
An alternative, ‘institutionalist’ view of ILMs treats them as the normal pattern for firms to adopt unless there already exists a structure of occupational markets for qualified labour (OLMs). Inter-firm markets for labour with transferable skills are in fact like public goods and so depend upon a high degree of co-operation among employers to control ‘free-rider’ behaviour. Without this support, they are inherently unstable (Marsden 1986). Although in theory the trainee should bear the cost of wholly transferable training (Becker 1975), there are strong theoretical and practical reasons why employers commonly bear much of the cost.
The ‘cost-sharing’ mechanisms predicted by Becker tend to break down for a number of reasons. Relying on low trainee pay rates limits investments in transferable skills to what working-class families can afford to subsidize. Adult skilled workers may be leery of low trainee rates, fearing that trainees could be exploited as a form of cheap, substitute labour, and therefore either seek to restrict numbers, or to raise trainee pay rates. Employers may not be able to enforce apprenticeship contracts and recoup their training costs—the cause of its decline in the USA.
Hence, employers have an incentive to cut their own training expenditures and to ‘poach’ skilled labour from their competitors. This increases the costs for employers who continue to train, and thus further encourages poaching. Out of the resulting vicious circle the OLM atrophies and breaks up, and firms resort to ILM practices in order to be sure to capture the returns on their investment in skill formation.
Yet despite these problems, we observe a certain number of large OLMs for occupational skills in Germany, in parts of UK industry, and in many countries in the artisan sector. Their survival has depended largely upon a strong institutional infra-structure capable of controlling free-rider activities by employers, and of reassuring skilled workers that cost-sharing would not lead to their being substituted by cheap apprentice labour. In Germany, the local chambers of industry and commerce, of which all local employers have to be members, provide powerful channels for peer group pressures against free-riding, and strong supervisory powers exercised by the works councils, supported by the unions, have ensured that apprentices get their full training. In the absence of such structures, firms will generally adopt ILM strategies.
Thus to contrast the two approaches on labour allocation, neo-classical economists generally look for blockages in the price mechanism as potential causes of many of the serious labour market problems, and commonly identify institutional factors as one of the prime culprits. The institutionalist approach treats the institutional framework as a precondition for an effective price mechanism. Thus, in the case of skills training, the former stresses the need for low trainee rates of pay and high rewards for skilled workers, whereas the latter stresses the need to create the institutional structures under which adult workers will accept the cost-sharing solution, and employers will refrain from poaching (Marsden and Ryan 1990)
Turning to incentives within the firm, a revolution has taken place in economic thinking on wages. Neoclassical theorists have abandoned the idea that the market sets a single rate of pay for a given job as irrelevant to many jobs. Instead, they have argued that workers are attracted to jobs by the whole package of benefits on offer, be they short-term employment at a fixed wage or the prospect of career employment and salary progression. Employers adopt one or other model depending on the kind of worker they wish to recruit. Thus fast-food firms might have relatively simple incentive structures with a rate for the job, whereas retail banks might offer complex structures with career jobs, age or performance incremental pay scales, and so on, depending on the kind of recruit they want.
The kind of performance that employers require from their staff has also been examined (Akerlof and Yellen 1986). In many types of job, sub-standard performance by the employee does not become apparent for some time. By providing career employment, the employer has time to monitor performance, and to sanction good or bad results. Employees might be encouraged to stay by taking a starting salary below the value of their output in the expectation that, later in their career, they would be paid in excess of it. By offering such profiles, employers can attract employees who are looking for a long-term job. In addition, the penalty of dismissal for poor performance increases with seniority (since alternative jobs would pay them only the value of their output) so the incentive to work well increases. Some other theories look at how different kinds of salary and promotion systems encourage good rather than perfunctory performance. Competition for promotion can be organized as a ‘tournament’ to ensure high performance, thus top salaries may reflect less the value of the work done by top managers than that of those competing for the top jobs. Another theory looks at pay as involving a ‘partial gift exchange’ and reciprocity whereby employers offer better than average career conditions to their staff, and, in return, employees feel that they should reciprocate by giving better performance (Akerlof and Yellen 1986).
These theories rely mainly upon individual performance and individual incentives, and there is minimal reference to supporting institutional structures. An alternative approach is to look at collective incentives: at what is needed in order to promote co-operative exchange within the workplace. It can be argued that co-operative exchange leads to high productivity because it encourages flexible working and information sharing. Such practices are hard to obtain in many organizations because they affect small group power relations (Brown 1973). Since co-operation usually involves one party placing itself in a weak position vis-à-vis others (for example, by sharing information it controls) it exposes itself to exploitation. Therefore, parties will not normally adopt a co-operative stance unless they are confident the others will do likewise. The rise of ‘lean production’ and the increasing awareness of the greater job flexibility to be found in German and Japanese workplaces, as compared with the job demarcation rules common in Britain or the seniority rules common in the USA, have brought these issues to centre stage (Womack et al. 1990).
Game theory stresses the difficulty of obtaining stable co-operative exchange unless there is some kind of trust, and thus some kind of framework to support it (Dasgupta 1988). Two important conditions should be met: the parties need to know something about the motivations and intentions of the others, and they need some framework of guarantees to give them redress, and to enable them to exchange information so that genuine cases when force majeure prevents reciprocity can be distinguished from simple opportunism. Concrete examples of such frameworks can be found in the current practice of German co-determination, and in the relations between large Japanese firms, enterprise unions and their federations (Aoki 1988).
As for policy, the individualistic approach normally stresses the value of greater inequalities to reward higher levels of individual performance. Against this, if performance depends on co-operative team working, it could be argued that pay inequalities should be smaller. Since responsibility and decision making are shared in team working, there is less reason to focus incentives on key individuals. Strong individual incentives may be incompatible with co-operative exchange because they encourage people to conceal information which enables them to perform better than their colleagues.
Whatever the outcome of this debate, the idea of there being a single entity called the ‘labour market’, and that it sets prices for single units of work, is clearly too simplistic. There are many different kinds of employment relations. At the low paid-low skill end of the market, something approximating casual labour markets with short-term employment and competitive wage rates tied to individual jobs may exist. In the middle, where qualified blue and white collar skills are used, jobs tend to be fairly long term, and management are more concerned about motivation and efficient deployment of staff within their organizations than with the haggling of the market-place. Collective bargaining remains influential in many countries, and acts as a powerful force for shaping pay structures according to workers’ views of fairness. Among management and higher professionals, collective bargaining is much weaker, and many workers have their own individual reputations which often have a market value. The policy problems raised by each of these are different, and, in the field of incentives, are potentially contradictory.
David Marsden
London School of Economics and Political Science
References
Akerlof, G. and Yellen, J. (eds) (1986) Efficiency Wage Models of the Labour Market, Cambridge, UK.
Aoki, M. (1988) Information, Incentives, and Bargaining in the Japanese Economy, Cambridge, UK.
Becker, G.S. (1975) Human Capital: A Theoretical and Empirical Analysis, with Special Reference to Education, Chicago.
Brown, W.E. (1973) Piecework Bargaining, London.
Dasgupta, P. (1988) ‘Trust as a commodity’, in D.Gambetta (ed.) Trust: Making and Breaking Cooperative Relations, Oxford.
Doeringer, P.B., and Piore, M.J. (1971) International Labor Markets and Manpower Analysis, Lexington, MA.
Elliott, R.F. (1991) Labor Economics: A Comparative Text, New York.
Marsden, D.W. (1986) The End of Economic Man? Custom and Competition in Labour Markets, Brighton.
Marsden, D.W. and Ryan, P. (1990) ‘Institutional aspects of youth employment and training policy in Britain’, British Journal of Industrial Relations 28(3).
OECD (1993) Employment Outlook 1993, Paris.
Rosenberg, S. (ed.) (1989) The State and the Labor Market, New York.
Williamson, O.E. (1975) Markets and Hierarchies: Analysis and Antitrust Implications, New York.
Womack, J., Jones, D.T. and Roos, D. (1990) The Machine that Changed the World, New York.
