The Role of Economic Theory in Determining Systems of Labour Market Regulations

In recent years organisations such as the OECD, World Bank and the IMF have started to play a more active role in advocating the redesign of labour market institutions. Several similar themes can be found in all these proposals, primarily a belief in the existence of a 'natural' or 'NAIRU' rate of unemployment determined primarily by institutional arrangements such as unemployment benefits, minimum wage and job security legislation. The resultant policy prescriptions are based around the deregulation of labour market in an attempt to increase 'flexibility' and ensure that labour markets clear. This paper critically examines the conclusion that it is meaningful to refer to rising unemployment rates as 'natural' and that policy should focus on labour market flexibility. These policies are examined both from a micro and macroeconomic perspective. It is argued that such policy prescriptions tend to rely excessively on the short-run allocation of labour between alternate pursuits (especially labour and leisure) and alternative production techniques (especially the extent of capital intensity). Whilst such choices may play some role at the margins, they are not the only important issue in labour markets, nor do they have much effect on individuals or firms who have already made long term commitments (either to capital equipment or mortgages). The long term formation of skills; the interaction between educational, familial and occupational institutions; mechanisms for resolving asymmetries in bargaining power and information; and the geographic distribution of employment opportunities and suitably trained workers are all important considerations in the design of labour market institutions which are often ignored. At the macroeconomic level the long run determinants of the level of output and demand for labour will be discussed. It is argued that macro policies can be effective in the long run in increasing the level of output and reducing unemployment. This argument is based on the existence of path dependence in the level of output. A simple model is used to show how a period of contractionary macroeconomic policy can have a permanent adverse effect on the level of output and employment.. It is concluded that the design of labour market institutions can benefit substantially by incorporating a much wider view of the operation of the labour market, and its interaction with the product market, than by that offered by proponents of the 'natural' rate of unemployment. Introduction Unemployment remains as a major economic problem for nearly all countries, on all continents. For the last 25 years though, the dominant theoretical response to this problem has been, given that the economy is assumed to tend towards full employment in the long run, to avoid interventionist macro policy and instead to increase the ‘flexibility’ of the labour market. Greater flexibility, it is argued, will result in a more rapid return to full employment. Such a policy response has been regularly advocated in both the academic literature (see Layard et al 1991, Moore 1999), by individual governments (see Treasury 1996) and by international bodies such as the IMF (see Mauro et al 1999), the OECD (see OECD 1992, 1994, 1998). The existence of sustained periods of high unemployment is typically explained with reference to the Non Accelerating inflation rate of Unemployment (NAIRU). The NAIRU defines a rate of (equilibrium) unemployment at which the competing claims of income earners are consistent (Carlin and Soskice, 1990, p.136). It represents the minimum sustainable rate of unemployment. The size of the NAIRU is a manifestation of supply side imperfections, so that an unacceptably high equilibrium unemployment rate signals the need for supply side reform. Policies to reduce the supply side causes of the NAIRU typically involve the removal of ‘excessive’ welfare benefits, centralised wage fixing and excessive union bargaining power (see Mauro et al 1999). Despite 20 years of effort though, the deregulationist policy prescription has failed to restore low rates of unemployment (Burgess 1998). This paper will critically examine the theoretical arguments for a greater reliance on deregulation to achieve greater labour market flexibility to in turn provide a solution to the unemployment problem. Before doing so though, it is important to consider some questions that do not typically arise in a discussion of labour market deregulation. • Why were unemployment rates able to be kept so low throughout the 50’s and 60’s when labour markets in most Western economies were at their most rigid? • If the U.S. economy provides evidence of the virtues of a flexible labour market, why was its unemployment rate so much higher than those of much more regulated economies (such as Japan) in the 1980’s and early 1990’s? • If institutional factors are the main determinant of the level of unemployment why are regional disparities in unemployment so high across administrative areas that have identical laws and regulations? The answers to these questions are not unambiguous. The purpose of stating them at the outset is to highlight the complexity of both the unemployment problem and its solution. Issues such as the pace of technological change; changing demographic characteristics of the labour market; the business cycle; and immobile labour all combine to ensure that there is no simple answer to the problem of unemployment. Increased labour market flexibility may be a mechanism for facilitating the matching of labour with job vacancies in some circumstances. However, it may also act to reduce the dynamic efficiency of the labour market by reducing incentives to invest in training and by providing subsidised access to labour for low productivity firms. Finally, it will be argued that whilst the matching efficiency of the labour market is important (with labour market flexibility playing a more or less important role depending on labour market conditions) the overall level of employment will be determined outside the labour market. Whilst the view that the economy automatically tends towards the full employment level of output is the dominant one today the conditions for such an optimistic view are also outlined. It is argued that in an economy with rising unemployment macro policy can play an important role in reducing the long run level of unemployment. What is an efficient labour market? In order to determine the desirability, or otherwise, of any change in policy it is essential to determine exactly what outcomes are actually being pursued. That is, it is essential to determine exactly what is meant by an ‘efficient’ labour market. In so doing it is important that this definition is not tautological. It is meaningless to define flexible labour markets as efficient on the assumption that efficiency requires flexibility. It is also important to determine the level (micro or macro economic) at which the term ‘efficiency’ is to be applied. Actions that may be efficient for individual agents may not be efficient for the labour market as a whole, or for the broader economy. That is, market failure may exist in the labour market even if competitive conditions are satisfied. For the purposes of the following analysis the notion of efficiency in the labour market will be a broad one. Therefore, changes in labour market policies that result in improvements in the wages or profits of individual agents will only be considered to be unambiguously associated with an increase in labour market efficiency in the absence of any market failures. Whilst any attempt to define ‘efficiency’ in the labour market is likely to be controversial, it is a necessary step towards any policy analysis. It is also essential to make such an attempt to illustrate the multi-faceted nature of efficiency in the labour market. Finally, it is essential to define labour market efficiency in order to escape the default definition which has emerged, the tautology of defining efficiency in terms of flexibility, when flexibility is not the objective of labour market policy but one possible means of achieving the goals stated below. For the purposes of this analysis an efficient labour market is defined as one which facilitates the co-ordination process of labour and capital whilst it simultaneously: • rewards effort and ensures that income distribution is commensurate with the distribution of risk and the economic agents capacity to absorb risk. • provides opportunities to earn stable and sufficient income (subject to sufficient macroeconomic circumstances) • encourages both workers and employers to invest in training, providing the potential for dynamic efficiency gains • allows disputes regarding wages and conditions to be resolved in a way which does not just reflect uneven bargaining power and without imposing excessive costs on other agents in the economy through income loss, inconvenience etc An improvement in labour market flexibility may, or may not, lead to the above outcomes. The objective of labour market policy should be to facilitate the efficient allocation of labour resources between alternative uses in the short term and in the long term. ‘Flexibility’ is one means of achieving this objective, but it is not an end in itself. The purpose of the following sections is to explore the effectiveness of the means of flexibility in achieving the end of efficiency. Does Increased Labour Market Flexibility Lead to Increased Labour Market Efficiency? All market transactions take place in a regulated environment. The issue for policy makers is not to regulate or not, but to determine the nature and extent of regulation. Attempts to increase flexibility typically entail measures to reduce the extent of regulations on centralised wage fixing, holiday and sick pay, and protection of parties involved in industrial disputes. Similarly, there are often calls for reductions in the restrictions on the use of casual labour, the minimum shifts that can be worked by casual labour, and the extent of penalty rates which historically were payable for the use of such labour (see King 1993,Trinca 1998, Norington 1997; Campbell and Burgess, 1997). A separate, but related trend has been the increased use of external labour. Contracting out, outsourcing, and increased reliance on temporary employment agencies are all examples of this approach.(See Borland 1994, Paddon and Thanki 1995, Industry Commission 1996; Denniss 1997; Burgess and Macdonald, 1999). The implicit assumption in the work of many authors who have considered this trend is that such increases in numeric flexibility are associated with an increase in efficiency (CAI, 1991). Such an assumption is only valid under a strict set of conditions; conditions that are unlikely to exist in a modern industrial economy. The simple reason that an increase in numeric flexibility is unlikely to unambiguously result in an increase in efficiency is that is that there are likely to be substantial sources of market failure in the labour market. These market failures are present at both the micro and macro economic level. When market failure is present market outcomes will not result in an efficient allocation of resources. In such circumstances to rely more heavily on market mechanisms, and less heavily on regulation, will result in a reduction in economic efficiency. The following paragraphs outline conditions which, if they exist, imply that an increase in the flexibility of the labour market will result in a reduction in economic efficiency • Macroeconomic Externalities The existence of macroeconomic externalities imply that individual agents, bargaining in isolation, may arrive at decisions, or implement bargaining strategies, which will impose costs (or benefits) on the performance of the macroeconomy. Potential macroeconomic externalities arise from the following: • Inflationary pressures. It has been argued (see Gittins 1997, Allard 1997, Norigton 1997) that workers and employers do not consider the macroeconomic consequences of their actions, specifically with regard to inflationary impacts. Whilst this is quite likely it seems perverse to blame such actions on ‘greedy unions’ when the underlying premise of a deregulated wage fixing system is that individual agents should be free to bargain in their own self interest. If governments do not wish to control wages through wages policy then it would seem unreasonable to expect individual agents to implement such a policy on a voluntary basis, especially when the benefits of such community minded action accrue to society as a whole. Substantial scope for free riding exists in such a voluntary system, such as has occurred with executive salaries (see for example Mychasuk 1994). • Increased cyclical volatility A further macro economic externality associated with an increased reliance on casual and external labour is related to the impact of job shedding on aggregate demand. Historically the practice of labour hoarding, particularly of skilled labour, was the norm in times of economic slowdown (Oi 1962, Okun 1981). If the benefits to individual firms of increased flexibility are that they can more accurately match labour with available work then the implication is less stable levels of employment, and in turn, less stable consumption expenditure. Increased cyclical volatility of employment and labour incomes will result in a less stable macroeconomic climate. The existence of such externalities implies that there are macroeconomic efficiency arguments for reducing the extent of some forms of labour market flexibility. However, there is also a range of microeconomic market failures that raise doubts as to the efficacy of persisting with the assumption that increased flexibility equals increase labour market efficiency. Microeconomic Coordination Problems • Excessive reliance on strike action Individual firms and their workers usually have a limited right to participate in strikes and lockouts to resolve industrial disputes. It can be assumed that both parties will consider the expected costs and benefits of their actions in order to determine their optimal strategy. Whilst such an approach may result in an outcome which is desirable for the relevant parties, it will not be socially optimal if the dispute process imposed costs on third parties. In such a case it is in societies interest to resolve the dispute with less reliance on disputation and place a greater reliance on arbitration. • Need for inter firm coordination When multiple firms need to act in a coordinated manner, for example multiple contractors on a construction site, it is important that the employment conditions negotiated between each employer and their employees are compatible. For example, if multiple firms have different conditions for length or timing of workdays or number and length of work breaks then the production process will be less efficient. • Imperfect information Efficient market outcomes require the existence, and availability, of good information. Such information may not be sufficiently available to workers, or firms, to ensure the desirability of their individual decisions. In order for individuals to bargain effectively with their employers they require a large amount of information, especially in relation to the earnings and conditions of similar workers in other workplaces. If gaining this information is not costless then it is unlikely that individuals will have the resources to acquire it. Alternatively, if all individuals undertake such search then the costs would be much greater than if it was collected centrally and disseminated. One of two types of inefficiency must therefore occur. Firstly, decisions are made on the basis of incomplete information. Or secondly, the amount spent on collecting information is much greater than what it would be under a centralised regime. Under such circumstances it will be more difficult to maintain wage equalisation because it relies on the providers of labour (both workers and the unemployed) to have full information and the ability to assess that information. A key element in the individual contract framework is often the secrecy of the contract. Hence it is difficult for co-workers to know each other’s remuneration packages, let alone those of a similar worker in another firm. Furthermore as work practices become more flexible it is increasingly difficult for a current or new worker to determine just who is a similar worker.