Profit-Sharing in OECD Countries: a Review and Some Evidence

Business Strategy Review, 1997, Volume 8 Issue 4, pp 27-32 Profit-Sharing in OECD Countries: a Review and Some Evidence Saul Estrin, Virginie Pérotin, Andrew Robinson and Nick Wilson This article starts by describing the extent of profit-sharing schemes in OECD countries. It suggests that firms have increasingly introduced profit-sharing schemes not only because of the tax breaks they involve, but also because of accumulating information on the link between profit-sharing and productivity. It concludes with a summary of new research among UK manufacturing companies, which shows a productivity improvement of about 6% in cases where profit-sharing bonuses were of order 5-10% of market wages. There has been a rapid increase in the extent of profitsharing in Europe in the past decade. For example, in the United Kingdom in the late 1970s, according to estimates by the Wider Ownership Council, only 2% of employees benefited from profit-sharing; following tax relief in various Finance Acts and the introduction of profit-related pay (PRP) in 1987, Blanchflower and Oswald (1989) reported that some 40% of private manufacturing establishments operated some sort of scheme, while by 1993 PRP covered more than 10% of private sector employees in nearly 5,000 schemes. In France, Ministry of Labour figures indicate that the number of firms operating voluntary profit-sharing schemes grew from about 3,500 in the late 1970s to over 10,000, covering 2m employees, by 1992. After French legislation was simplified in 1986, the growth in the number of schemes markedly accelerated. Our primary purpose in this paper is to review the scope of profit-sharing in OECD economies, and to categorise countries according to the scale and extent of profit-sharing arrangements. We go on to discuss the expected benefits to companies from introducing profit-sharing, in terms of improved productivity. In the final section, we report empirical evidence from the UK which confirms that profit-sharing does raise company productivity. What is Profit-Sharing? Many companies use pay systems which they call profit-sharing. But to achieve the benefits we discuss below, schemes must conform with specific definitions. For our purposes, a profit-sharing scheme has some portion of all employees remuneration variable and is tied to overall company performance. The rules determining the share of profit to be allocated to employees and its distribution must be set and known in advance. The employees share of profits may be split equally among employees or individual bonuses may depend on wages, absenteeism, length of service, London Business School

28 Saul Estrin, Virginie Pérotin, Andrew Robinson and Nick Wilson etc. Bonuses may be paid in shares or cash and/or may be deferred for several years (as with participation in France or many profit-sharing plans serving as pension plans in the US). Two features distinguish profitsharing from other sharing schemes: it is a form of variable pay; it is a group incentive scheme related to the performance of the company as a whole. Thus, profit-sharing does not include fixed bonuses or employee share allocations unrelated to company performance. Nor does it include types of variable pay that depend on individual performance, such as piece rates. In practice, profit-sharing is often combined with employee share ownership and may be associated with other forms of financial participation (such as company savings plans) and with varying levels of employee information and participation in decisionmaking. We focus on schemes that have a profitsharing component, whether or not they also qualify as employee share ownership. We look at schemes that cover all or most of the employees and ignore executive schemes, which have different implications. Because managers have better information about firm performance and are able to make decisions that directly affect the firm (such as investment decisions), executive schemes must be analysed quite differently. Profit-Sharing in OECD Countries Profit-sharing schemes are widespread. In many countries, the number of schemes has recently grown fast, after stagnating for several decades. In some, growth has been spurred by legislative activity and tax incentives. In others, profit-sharing has spread despite the absence of favourable legislation or tax incentives. In this section, we briefly look at the incidence of profit-sharing schemes, their evolution in recent years and possible links with legislation. The incidence of profit-sharing in OECD countries is summarised in the table, which provides information from a variety of sources about the number of schemes in each country, the proportion of employees and firms covered and the nature of the benefits provided. We categorise countries into three groups by incidence of profit-sharing schemes. In the first category are countries where the incidence of profit-sharing is high (at least 15% of employees covered by a scheme). This group includes Japan, Mexico and France. Almost all Japanese firms employing more than thirty workers use establishment-wide bonus systems, but most of these are not related to profit. However, in 1983, 15% of firms with 30 employees or more operated profitsharing schemes. Profit-sharing is compulsory for most firms in Mexico. However, it seems that the law is unevenly enforced, and no estimate of the actual incidence is available. In France, profit-sharing is compulsory in firms with 50 or more employees and about one third of all employees are covered by a scheme (participation). The voluntary French scheme (intéressement) covers 10.4% of employees. In the second group, we find countries with a relatively high incidence of schemes (at least 5% of employees covered). This group contains Canada, the United States, the United Kingdom, Germany, Netherlands, Finland, Denmark, New Zealand, and possibly also Italy. Australia, Ireland and Belgium form the third group, with the possible addition of Switzerland. These countries are known to have a number of profit-sharing schemes, but less than 5% of employees are covered. Insufficient information is available on profit-sharing in the remaining countries to allow us to give even a reliable order of magnitude of the actual incidence of these schemes. A few comments on figures in the table are in order. Differences in sources imply that the coverage of the surveys greatly varies across countries. Thus, US data, for example, have been collected primarily by private agencies and include discretionary schemes (up to 46% of the plans according to North American authors) which are in principle out of our definition. In contrast, the figures we present for Britain and France come from government sources and cover only schemes that qualify for tax concessions, excluding executive or discretionary schemes and, in the case of France, schemes that have not been negotiated with the workforce. Other variations come from differences in what are considered relevant categories of financial participation. Italian estimates of the incidence of profit-sharing tend to include all forms of variable pay schemes, including small group bonuses and piece rates. Similarly, figures given for Spain could include very few all-employee schemes. In several countries the number of schemes and employees covered has grown, sometimes quite dramatically, in the last two decades. Examples of this trend include France, UK, New Zealand, Netherlands, US, Canada, Belgium and Italy. In the UK and in France, the observed growth clearly coincided with favourable revisions of the legislation simplifying administrative procedures or offering tax concessions Business Strategy Review

Profit-Sharing in OECD Countries: a Review and Some Evidence 29 for the establishment of profit-sharing schemes. It seems likely that many of the early British schemes represented conversions of pre-existing profit-sharing plans that had not hitherto been eligible for tax benefit. But later rapid growth is perhaps due to an accumulation of information on the advantages reaching new firms. In addition, in both countries, cosmetic schemes may have been set up exclusively to take advantage of tax concessions, with effectively fixed bonuses. However, these factors alone are unlikely to explain such high rates of growth. In France, both intéressement and participation (introduced in 1959 and 1967 respectively) attract tax concessions, but intéressement is voluntary whereas participation is mandatory in firms above a certain size, and voluntary in smaller firms. The reform of the legislation in 1986 made it easier for firms to set up an intéressement scheme and in 1990 the obligation to have participation, which had applied to firms with 100 employees or more, was extended to firms with 50-99 employees. The number of firms with voluntary schemes has risen from around 4,000 in 1980 to more than 10,000 in 1992. This growth first came from voluntary participation. Later on, especially after 1986, intéressement spread much faster, though partly at the expense of voluntary participation. (Some of the apparent decrease in the number of voluntary schemes observed after 1990 is due to the extension of the obligation to have a participation scheme. In other cases, firms that were newly obliged to have a participation scheme discontinued their interéssement one.) However, in the 1980s profit-sharing has also been observed to spread reasonably fast in Italy and Canada, where the legislation is not particularly favourable and no tax incentive is provided. In Canada, one type of regulated plan stopped spreading after the tax treatment became less advantageous in 1983, but other plans multiplied despite the absence of subsidy. Once again, this suggests that the growth is explained by the gradual spread of information about the benefits of such schemes. We find that outside Japan and Britain, profitsharing represents, on average, 3-5% of pay in profitsharing firms (but eg over 20% of profits on average in the years a bonus is paid out in France). The proportion of pay is much higher in Japan and has recently grown in Britain. In contrast, the level has remained stable in France. Individual bonuses can vary considerably across employees and firms and from year to year. (In order to limit the risk to employees, legislation in both Britain and France imposes a cap on the proportion of pay that may take the form of profit-sharing in plans eligible for tax benefits.) How Profit-Sharing Affects Company Performance There is a large body of research suggesting that profitsharing can improve company performance on three fronts: employee effort, productive skills eg through longer service, and organisational efficiency. Group incentives are clearly not the same as individual incentives in the effect they have on employee effort. Relative to personal performance schemes, profitsharing may suffer from free rider problems because each person may receive personally only a small fraction of their own contribution to increased profits. But group arrangements may permit reduced monitoring costs and sharper incentives in situations where individual productivity is costly to measure, and may stimulate positive peer group pressure to increase work intensity in situations when the productivity of individuals is inter-related. The second issue relates to the productive skills of the labour force, and their willingness to remain with the firm. For example, if profit-sharing improves group awareness and assists in conflict resolution, it may reduce labour turnover and increase firm-specific human capital. Once again, complementarities in employee productivity may be relevant; many productive skills are learnt with the help of fellow workers, who will be more motivated to assist under profit-sharing schemes. Finally, there is improved organisational efficiency. To quote Cable and FitzRoy (1980, p166), participatory firms - with or without profit-sharing - will produce better outcomes than traditional firms if the negative collusion to maximise one party s area... can be replaced with positive collusion to maximise joint wealth. There are many ways that improved employee-manager relations could raise company performance. If workers have an increased sense of responsibility, they will be less likely to strike or undertake damaging industrial disputes. Improved employee morale will also show up in lower rates of absenteeism, greater willingness to work together and improved labour flexibility with respect to work practices. To test for this effect, we used a British data set Winter 1997

30 Saul Estrin, Virginie Pérotin, Andrew Robinson and Nick Wilson Table 1 Profit-sharing in Practice in Selected OECD Countries High Incidence of Profit-Sharing Legislative stance Proportion of all Proportion of Performance Eligibility/Allocation Proportion of employees covered firms with scheme indicator indicator earnings in firms France Encourages 19% of all non- 0.7% of all firms, Linked to profit Primarily wages, On average 2% of Interessement des governmental 10% of those with in 78% of cases. absenteeism or a gross wages for salariés (CPS) employees, 2.7m > 50 employees, combination of firms operating 1959 employees (1992). 9,500 firms (1992). both. schemes, 3% for those paying bonuses (1993). Participation aux Roughly a third of Compulsory for firms Specified by law. Normally in In firms that paid fruits de all non-governmental > 50 employees, proportion to wages. out bonuses 4-4.5% l expansion (DPS) employees, 5m voluntary otherwise. of gross wages. 1967 employees (1993). Still in transition but covers 1.4% of all firms (18,700). Japan Permits 14.7% of firms with 30 employees or more, Linked to profit. n.a. n.a. Profit-sharing varying by no. of employees: 8.3% of firms (formal scheme) with > 1,000 employees; 19.9% between 100 and 999 employees; and 12.9% of firms with between 30 and 99 employees (1983). (All employees are covered by such schemes). Relatively High Incidence of Profit-Sharing Canada Permits Roughly 12% in 14.7% of firms with Generally profits. n.a. n.a. Cash profit-sharing early 1990s. 20 or more (informal schemes) employees (1990). Deferred Profit- Under 8% in 4.3% of firms with Profits. Pay levels, no. of Around 3% of total Sharing Plans the early 1990s. 20 or more years service, or payroll including employees (1990). some combination other types of of both. bonuses. Germany Permits Approximately 6% Approximately 1% Various, including Various including 5 to 10% of gross Cash profit-sharing employee coverage of all firms (1993). sales, value added, gross earnings, earnings. (informal schemes) (1993). pre-tax profits plus tenure and other more firm/ contracted work. hours specific. Share and capital 4.3% of employees 0.5% of all firms Commonly linked to tenure, per capita. 0.3 to 0.5% of net based profit-sharing (1993). (1993). value-added, pre-tax earnings. profits or dividends. United Kingdom Encourages Just over 5% of 0.2% of all firms, Profit, or changes Usually base salary, On average, 2 to Profit-related pay employees, 1.17m 4,615 schemes in profit over the occasionally also 5%. 1993 (CPS) workers (1993). (1993). period. seniority. Approved profit- Just over 3% of Total of 1,015 Mainly profits - Base salary in 80% In 1991/92 the sharing scheme employees (1992). schemes approved but only 60% of of cases. average share (APS) (SPS) since 1979. schemes use pre- appropriation was determined formula. 450. Business Strategy Review

Profit-Sharing in OECD Countries: a Review and Some Evidence 31 Legislative stance Proportion of all Proportion of Performance Eligibility/Allocation Proportion of employees covered firms with scheme indicator indicator earnings in firms United States Encourages/ n.a 3% of firms with If discretionary, Occasional n.a. Cash-based profit- Permits >100 employees determined by restrictions by sharing (CPS) have CPS (includes Board of Directors. tenure or to 2% which have CPS If by formula, permanent full-time and DPS). determined by pre- employees. Allocation tax profits adjusted by earnings, for unusual/extraordinary items. sometimes weighted by tenure or performance. Deferred profit- Approximately 20m 15% of firms with As above. As above. n.a. sharing (DPS). employees (early >100 employees 1990s). have DPS (includes 2% which have DPS and CPS). Finland Permits Approximately 7% Just under 1% of all Linked to operating Mainly hours of On average around Personnel funds of employees (1994). firms, 40 firms margin. work, sometimes 5% of gross wages. and profit bonus (1994). additionally related system 1990 (DPS) to wages. Netherlands Permits Evidence in early 5% of firms with In order of Varies between Median of 4.2% of Various informal 1990s indicates 1-9 employees; importance: profits firms. total gross earnings mechanisms between 10 and 20% 10.5% of firms with (52%), fiscal (1992). Higher for (information on of business sector 10-99 employees; profits (20%), managers than for Vermeend- employees. 13.3% of firms with dividends paid ordinary emloyees. Vreugdenhil Law, >100 employees (14%), other enacted in 1994, (1992). concept (12%) unavailable). (1975). Low Incidence of Profit-Sharing Australia Permits Approximately 3% 4% of private sector n.a. n.a. n.a. Informal schemes coverage. firms with 5 or more only employees. Belgium Constrains n.a. Survey evidence Wide variety, Variety including Approximately 1.5 Various types of CPS suggests 10% of including profits salary, rank and to 6% skewed to and DPS, including large firms and return on specifically for PSC, lower end of this profit-sharing ( 2 / 3 CPS, 1 / 3 PSC). capital. merit and job distribution. certificates (PSC). evaluation. covering a random sample of 93 manufacturing firms observed over four years from 1988-91. Among these firms, 40% operated a cash-based profit-sharing scheme over the period considered, which is representative of the proportion of profit-sharing firms in manufacturing as a whole in Britain. Profit-sharing bonuses were of the order of 5-10% of market wages. In addition to accounting data, the data set includes detailed information on the environment, technology, workforce characteristics and industrial relations. In this study, profit-sharing increased productivity by about 6%. (For summary details of the methodology, see Appendix.) Conclusions Profit-sharing was widely discussed in the 1980s as a panacea for economic problems, and in many OECD countries (including the UK), legislation was introduced to encourage it. However, academic analysis was sceptical of the economic benefits of profit-sharing (see eg Blanchflower and Oswald 1988). In fact, the nature and scope of profit-sharing has increased enormously in the past decade throughout OECD countries, and not only in those which had introduced permissive or tax-advantageous legislation. This suggests firms had begun to learn that profitsharing might have positive benefits for the bottom Winter 1997

32 Saul Estrin, Virginie Pérotin, Andrew Robinson and Nick Wilson line. To explore this, we have undertaken a preliminary study of UK firms. Our empirical work, though preliminary, suggests that firms which have introduced profit-sharing schemes are more efficient, in the sense that they produce more output from a given set of factor inputs (see also Weitzman and Kruse 1990). Further research will be needed to explore whether this finding is true more generally. This paper was produced as part of the IPSE network which is supported by a grant from the European Commission s Human Capital and Mobility Programme. Virginie Pérotin writes in a personal capacity - the views expressed here do not necessarily reflect those of the International Labour Office References Blanchflower, David G. and Oswald, Andrew J. (1988) Profit-Related Pay: Prose Discovered? Economic Journal 98, 392, pp720-730. Cable, John R., and FitzRoy, Felix R. (1980) Cooperation and Productivity: Some Evidence from West German Experience, Economic Analysis and Workers Management 14, 2, pp163-180. Weitzman, Martin L., and Kruse, Douglas (1990) Profit-Sharing and Productivity, in Alan S.Blinder, ed: Paying for Productivity: A Look at the Evidence Washington DC: Brookings Institute, pp95-140. Appendix: Profit-sharing and Productivity In our study of 93 UK firms in 1988-91, we estimated a production function linking factor inputs - labour and capital - to production in each firm. A production function provides a way to describe the technical relationship between inputs and outputs, but is not behavioural. Theory therefore offers few insights into the appropriate way to specify the relationship, but experience suggests that a loglinear functional form tends to yield the most robust estimates. The unexplained residual in the estimated equation is so-called technical efficiency ; that part of output which cannot be explained by factor inputs alone. If profit-sharing increases productivity in the three ways we outline on page xx, profit-sharing firms should show higher technical efficiency, and therefore higher output (holding all inputs constant). To test this, we estimated a loglinear production function on the data including a [0-1] dummy variable to control for firms which were profit-sharing. If this variable were positive and significant, it would imply that profit-sharing increases productivity. Since the functional form is loglinear, the estimated coefficient is an elasticity, telling us how much productivity will be changed. When the equation was estimated on our UK data set, we found: ln(output)=7.24***+0.72ln(employment)+0.26ln(capital)+0.06(profit-sharing dummy) Other variables were included in the equation to control for labour supply skills, technology and capacity utilisation. The equation implies that the firms have no economies of scale but, rather, approximately constant returns (eg the coefficient on labour and capital sum to approximately unity). Moreover, the share of labour in value added is estimated to be more than 70%, and of capital around one quarter. We confirm profit-sharing firms to be more productive than non-profit-sharing firms and the estimate is statistically significant at the 99% level. Our results suggest that firms which use profit-sharing obtain levels of factor productivity about 6% higher on average than firms which do not use such schemes. Saul Estrin is a Professor of Economics at London Business School and Director of the CIS - Middle Europe Centre Virginie Pérotin is a Senior Research Economist at the International Labour Office Andrew Robinson is a Lecturer in Financial Economics at the University of Bradford Management Centre Nicholas Wilson is ICM Professor of Credit Management at the University of Bradford Management Centre Business Strategy Review