Micro and small enterprises and employment creation: A case study of manufacturing micro and small enterprises in South Africa

Abstract

South Africa is characterised by extremely high unemployment, at a level that has the potential to derail its future political stability. This paper explores constraints on the job creation capabilities of micro and small enterprises, specifically focusing on the consequences of existing institutional and regulatory frameworks, looking at a panel of firms within the manufacturing sector. This study revealed that most of the small firms interviewed were locked into a ‘coping mode’, and were extremely reluctant to expand, which owners ascribed to inappropriate regulation, outright regulatory failures and/or poor labour relations. At the same time, many employers openly admitted to not adhering to at least some labour regulations. At a theoretical level, it is argued that there is much more to promoting enterprise than the formal protection of owner rights. The paper concludes with an assessment of the possibilities and challenges of the practicalities of reform.

Keywords:

• micro and small enterprises
• regulation
• job creation

1. Introduction

This paper explores the effects of institutions and labour regulation on the strategic choices made by owners of micro and small enterprises (MSEs). It focuses on the decisions of MSE owners whether or not to expand, and relates this to formal and informal regulation. Whilst South Africa is considered one of the most prosperous countries on the African continent, it remains highly unequal, with high unemployment and poverty rates. Micro and very small enterprises provide more than 55% of all jobs and 22% of gross domestic product (Kaufman, 2007). Given the fact that there has been a general trend toward leaner staffing in the South African larger firms (Wood & Glaister, 2008), it is largely up to South Africa's MSE sector to provide employment and a way out of poverty for those that have failed to materially benefit from the ending of apartheid.

More specifically, this study evaluates whether and how present regulation hinders the expansion and job-creation capabilities of MSEs in a context of high unemployment, inequality and poverty. It explores how labour legislation shapes the choices made by entrepreneurs, and how such choices could be optimised by reforms. Hence, we focus on competing perceptions as regards the impact and efficiency of labour regulation, the relationship between regulation and relations between key actors, and the limits and possible trajectories of regulatory reform. This paper is structured as follows. Firstly, we review the possible effects of institutions and the law on enterprise in theory, and present South African labour legislation. Secondly, we introduce the MSE sector in South Africa. Thirdly, based on a series of interviews with employers and other stakeholders, we explore employer attitudes to the law and how it moulds strategic choices made by employers in MSEs in three sectors – clothing, leather (footwear) and metal and engineering. Finally, we revisit the possibilities of policy reforms to promote enterprise in South Africa, and explore the theoretical implications of the empirical findings. A wide range of accounts blame South Africa's high unemployment rate on excessive labour-market regulation (Kesper, 2000; Rankin, 2006). Hence, the central proposition of this study: the reluctance of MSEs to expand represents the product of excessive labour regulation.

It could be argued that what defines an entrepreneur is an ability to secure growth of the enterprise in adverse circumstances. Hence, it could be argued that those firms that fail to grow exhibit a lack of entrepreneurial leadership (see Arbor & Quartey, 2010). However, it should be emphasised that the areas of manufacturing encompassed by this study – metal, clothing, textiles and leather – have had to face a very severe crisis of competitiveness, brought about by the gradual phasing out of protective tariffs, and the flood of very much cheaper competing goods from China (Van Eeden, 2009). This includes not only legal imports, but the illegal import of semi-finished and finished goods that have greatly undercut local firms; this is particularly the case with regard to clothing and textiles (Watson, 2011). The firms encompassed by the study are indeed remarkable, in that they have survived in industries with very high exit rates (see Van Eeden, 2009).

2. The institutional foundations of prosperity

There is a very influential body of theoretical literature that links the institutional foundations of property owner rights to enterprise and economic growth (North, 1990; La Porta et al., 1999; Pagano & Volpin, 2005). La Porta et al. (1999) argue that property rights ultimately depend on legal tradition, and associated statutory and case law. In later developments and extensions of their work, it is argued that employee rights under the law represent a key countervailing force, weakening owner rights (Botero et al., 2004). Should the rights of workers be relatively strong, property owner rights will necessarily be weak, and vice versa (Botero et al., 2004). Inspired by this body of theoretical work, the World Bank ‘Doing Business’ reports similarly suggest that employee rights under the law may undermine the ability of firms to make optimal choices, and hence promote enterprise (Cooney et al., 2010). A variation of this argument is that extensive labour protection legislation is a luxury that is unaffordable in the case of countries that are already underperforming economically, and where a large proportion of jobs are provided by small and vulnerable enterprises (see Chandra et al., 2001; Bhorat, 2006).

Critics have disputed the relative incompatibility of owner and worker rights (Goergen et al., 2009). This would suggest that synergies are possible within and beyond the firm, with the potential for bargained cooperation to provide a durable basis for prosperity and growth. In the case of small enterprises, it has also been argued that, within a range of contexts, employment laws may only have limited impact on actual practice (Psychogios & Wood, 2011). In other words, a focus on the formal institutional foundations of private property rights may lead to a neglect of how institutions actually work in practice, and the extent to which whole areas of the economy may be at least partially decoupled from formal regulatory structures (Lane & Wood, 2009).

It is easier to monitor compliance by a single large employer than by many small ones that, between them, provide the same number of jobs. This is particularly the case as, given lower levels of unionisation, the labour movement is less likely to be able to actively promote compliance, especially as fragmentation precludes economies of scale in monitoring activities. In any event, in the most marginal enterprises, employees are likely to be particularly vulnerable and may conspire with owners to evade the law, should they believe the viability of the enterprise and their jobs are at stake. Hence, when unemployment is high, labour law may not necessarily represent a serious curb on owner power. On the one hand, it could be argued that an ability to evade the law means that MSEs can escape its negative effects. On the other, a strategy of evasion brings with it costs of its own, in imparting an unpredictability to proceedings and raising the transaction costs of interactions between owners and employees (Marsden, 1999).

3. The South African context and micro and small enterprises

Within South Africa, the MSE sector extends over a large range of business activity (Berry et al., 2002), including from the completely non-regulated to the entirely regulated. It involves many sub-sectors or branches of the economy, from street traders to small capital intensive manufacturing firms.

The South African labour movement is generally considered one of the most effective in the world (Wood & Glaister, 2008). Nonetheless, it has failed to make significant inroads in the MSE sector, although this has not deterred periodic organisational drives in this direction (cf. Chandra et al., 2001). Under high apartheid, the overwhelming majority of Africans were confined to poorly-paid unskilled jobs, workforce discipline being enforced through a wide range of repressive legislation and the direct actions of the security forces. However, as Webster (1986) notes, control does not equate with cooperation; productivity was generally low, with many employers compensating for this through overstaffing.

Within larger firms, the decline and fall of apartheid lead to new partnerships between firms and the independent unions that had emerged in the 1970s (Wood & Glaister, 2008). In practice, this led to major increases in productivity, and declining strike action; however, this also enabled firms to have much leaner staffing (Barnes, 2000). Such partnerships remain uncommon in the MSE sector. Although within post-apartheid South Africa some areas of manufacturing – most notably, automotive industries – have enjoyed a renaissance, others – notably clothing and footwear – have had to contend with rapid decline in the face of cheap foreign imports, following on the gradual phasing out of protective tariffs (Kaplan, 2004). The effects on smaller manufacturing firms have been particularly severe.

3.1 Micro and small enterprises and the law in South Africa

There are two parts to the debate about the effect of law on enterprise on MSEs in South Africa. The first is whether the regulation should apply to MSEs, and, if so, how it should apply; in other words, do worker rights detract from competitiveness, preclude owners from making the right choices, and are MSEs particularly vulnerable to enhancements in worker rights? Critics have suggested that present labour standards discourage employment (Kesper, 2000; Rankin, 2006). This would include excessive red tape, the ‘hassle factor’, and the difficulty of dismissing employees, all of which may deter employers from adopting strategies centring on creating new jobs.

The second relates to the compliance by MSEs to the labour legislation. In other words, as suggested by the literature, is the compliance problem particularly pronounced in the case of MSEs in South Africa? Does government policy have little real impact? A general limitation is that there is little research-based evidence about the constraints to doing business in South Africa, other than general comparative analysis, and ‘the evidence presented is often patchy, overly reliant on anecdote, sometimes contradictory, sometimes politicized’ (Hudson, 2004:7). There is similarly limited information on the extent of systematic non-compliance with labour law among MSEs, although existing research evidence would point to evasion and generally poor labour standards, especially in the informal sector (Webster, 2004).

3.2 The changes in government policy and labour legislation pertaining to MSEs

Under the 1995 Labour Relations Act, South Africa has industry-specific collective bargaining bodies, Bargaining Councils, which have the power to reach legally binding agreements (Department of Labour, 2011). Whilst the Councils are supposed to represent all employers and workers in an industry, in practice they are dominated by the larger unions and employers. The latter may have a shared interest in working together to prevent smaller employers from undercutting larger, heavily unionised competitors. The Councils also administer provident funds; these and general Council infrastructures are funded through compulsory contributions from employers, calculated according to the size of workforces. The Basic Conditions of Employment Act (BCEA) regulates employment contacts, leave conditions, pay, deductions, working hours and termination on a nationwide basis (Department of Labour, 2011). Bargaining Council agreements must at least comply with the BCEA, and typically supplement it.

Table 1 provides further detail on key aspects of labour legislation relevant to MSEs, and exemptions allowed for under existing law.

Table 1: Summary of mechanism in labour law catering for MSEs

Item	Mechanism in labour statute that specifically caters for MSEs	Reference
1	A streamlined and simplified code of practice with regard to unfair dismissals relating to misconduct and incapacity, including probationary employees	Schedule 8 of the Labour Relations Act (LRA) 1995
2	A simpler process for dismissals for operational requirements	Section 189 of the LRA (as opposed to section 189A)
3	Simple procedures for the resolution of disputes for small businesses	Item 3(1) of schedule 8 of the LRA
4	Specific provisions regulating the representation of small businesses on bargaining and statutory councils	Sections 29(11)(b)(ii) and (iii), 30(1)(b), 40(4)(b) and 41(3)(b)(ii) of the LRA
5	Exemption procedures for small and other enterprises from collective bargaining agreements that are extended to non-parties	Sections 32(3)(e) and (f) of the LRA
6	Making certain provisions of the BCEA not applicable to businesses with less than five employees. For example:	Sections 29(1)(n), (o) and (p), 30, 31 and 33 of the BCEA
	• Written particulars of employment – sections 29(1)(n), (o) and (p)
	• Displaying summary of the BCEA in workplace – section 30
	• Keeping employees' records – section 31
	• Written information with pay – section 33
7	The Small Business Variation of the BCEA, which allows businesses with less than 10 employees, to extend weekly overtime from 10 to 15 hours, to reduce the rate of overtime pay from ‘time and a half’ to ‘time and a third’, to average working hours for up to four months by written individual agreement instead of collective agreement, and to include employee's entitlement to three family responsibility leaves in their annual leave	BCEA; GenN R1295 in GG 20587 of 5 November 1999
8	The affirmative action provisions of the Employment Equity Act are only applicable to business employing more than 50 employees and with a turnover above a specified threshold	Section 1 (definition of designated employer) and Schedule 4 of the Employment Equity Act
9	The exemption of certain small businesses from the Skills Development Levies Act 9 of 1999	Section 4(b)

4. Method

This paper is based on in-depth interviews with stakeholders and a panel of MSEs. The choice of qualitative methods was chosen in order to explore the meanings and interpretations of key actors to objective phenomenon (cf. Rynes & Gephart, 2004). It further is argued that such a process helps ‘rehumanize’ empirical findings and theory (Rynes & Gephart, 2004). Owner-managers were interviewed in three key areas of South African manufacturing – namely, clothing, leather (footwear) and metal and engineering industries – involving a total of 14 enterprises. These sectors were selected owing to the historical prominence of centralised bargaining in these industries; in other words, these industries are those where the effects of regulation are likely to be the most pronounced. Moreover, in choosing to formally register with the Bargaining Council – as opposed to, as many employers do, refusing to do so – they have indicated some willingness to engage with the system; it is therefore likely that such employers would be more open minded as to the benefits and disadvantages of the system, and more receptive to policy innovations. The study did not cover the service sector, and it is recognised that research on the latter area may generate different results, and would represent a fertile area for future research. As the research centres on a limited number of in-depth interviews, the findings cannot be generalised, although they do afford key insights into strategic choices made at firm level, given existing public policy, and the attitudes of MSE owners towards regulation and competitiveness.

Thereafter, 30 key further respondents were interviewed, representing a cross-section of stakeholder interests: trade union officials, labour law consultants, government officials and Bargaining Council officials. Given that owner-managers may have been reluctant to fully elaborate on the extent to which they did not follow the law and to gain insights as to the ‘other side of the story’, we sought to gain additional data from the principal other party stakeholders (cf. Rynes & Gephart, 2004).

The process of selecting respondents was in line with theoretical sampling, aimed not just at uncovering categories and trends, but also what further data should be gathered from whom (Gill & Johnson, 2010:176), with the sample being expanded via the snowballing method (Gill & Johnson, 2010). The latter was guided by emerging insights as to the informal/formal regulation nexus, and by the fact that the regulatory difficulties experienced by MSEs were not simply a product of inappropriate formal regulation. Such an approach is not, of course, statistically representative, but affords key insights into a specific situation, and seeks to contribute to theoretical illumination and development. The use of quantitative tools – including data analysis software – would arguably have obscured this reality (Rynes & Gephart, 2004:455). As part of a process of grounded theorising, the views of players were contrasted (Rynes & Gephart, 2004). Findings were clustered together through an open-ended coding process, with excerpts from interviews being used to illustrate key points (Gill & Johnson, 2010:176). An extensive survey of the literature resulted in the identification of core sensitising concepts – that is, broad issues – giving ‘a general sense of reference and guidelines in approaching [the] empirical’ (Gill & Johnson, 2010:175); such concepts provided the framework for core questions, highlighting core issues where there was likely to be a variance of opinion. Related information from the different, transcribed, interviews was grouped together into conceptual categories (Gill & Johnson, 2010:175). Although these categories were partially derived from the existing theoretical and empirical literature, they were also adjusted according to the incoming flow of information from the earliest interviews, in an open-ended coding process (Gill & Johnson, 2010:176). As part of a process of grounded theorising, the views of different groups were compared ‘on the basis of theoretical similarities and differences’ (Rynes & Gephart, 2004). Whilst the research focuses on a distinct area that constitutes ‘its own analytical unit’, the concerns and issues emerging may serve as a basis for future comparative analysis (Eisenhardt & Graebner, 2007:25). As such, it can be argued the findings are both context specific yet theoretically relevant.

Table 2 provides a summary of the firms encompassed by the study. The interviews of MSE owners followed a semi-structured questionnaire. The outcome from responses to set questions are summarised in Tables 3 and 4.

Table 2: Case descriptions

Clothing manufacturing industry	Leather footwear industry	Metal and engineering industries
Case Two: ladies apparel fashion clothing, Marketing and Design House, Johannesburg CBD (small enterprise 50 workers)	Case Four: cut–make–trim (cut and close uppers), Pietermaritzburg Kwa-Zulu/Natal (small enterprise 32 workers)	Case One: specialised machinery engineering (fasteners) labour supplier, father–wife–daughter firm, Johannesburg CBD (very small, nine workers)
Case Four: knitwear manufacturer (party to two Bargaining Councils), Johannesburg CBD (small enterprise – 50 workers including 10 seasonal)	Case Five: redundant materials (off cuts) shoemaker, father–wife–son business, Manchester Road, Pietermaritzburg, Kwa-Zulu/Natal (small enterprise 25 workers)	Case Two: engineering (hospital electronic monitoring systems) contractors, Johannesburg South (small enterprise 30 workers)
Case Five: embroidery manufacturer (caps, and t-shirts), Johannesburg CBD (very small enterprise 10 workers)	Case Two: leather shoe and sandal footwear manufacturer, semi-formal company, Manchester Road, Pietermaritzburg, Kwa-Zulu/Natal (small enterprise 30 to 50 workers (seasonal)	Case Three: fabrication and engineering (rollers), Alberton (very small enterprise, 10 workers)
Case One: corporate clothing cut–make–trim, Johannesburg CBD (small enterprise 50 workers, many foreign)	Case Three: leather sandals, ladies shoes, school shoes and safety shoe manufacturer, Isipingo (Durban), Kwa-Zulu/Natal (small enterprise 38 workers)	Case Four: engineering, Randfontein (very small, 10 workers)
Case Three: school sock cut–make–trim, Roodepoort CBD (very small enterprise 10 workers mostly foreign)	Case One: men's sandals' leather footwear manufacturer, Chatsworth, Kwa-Zulu/Natal (small enterprise 20 to 50 workers (seasonal)

Table 3: Key issues emerging from interviews

More than two-thirds of the MSEs in this sample are closed corporations

The majority of the owners are male

More than three-quarters are older than age 45

Almost two-thirds of owners started their business before 2001

More than one-half of the owners started off as micro enterprises – employing one to five employees

There is a mixture of employment relationships within the enterprises but one-half of employees are in permanent full-time employment (with very limited use of labour brokers); the other half are temporary workers

One-half of the MSEs outsource certain activities, usually the labour-intensive part

One-fifth of owners classify their employees as skilled and semi-skilled with a mixture of skills

One-fifth of MSE owners who have applied for exemptions from Bargaining Council provisions were registered with the Council when they did so, were successful in their application and applied for a mixture of exemptions from the Bargaining Council, but almost four-fifths have never applied for exemptions

More than two-thirds of MSE owners do not belong to employer organisations

MSE employees work the legally prescribed number of hours and days per week as per Bargaining Council regulations

All MSEs have workers working overtime at some point

The leave requirements should be as stipulated by the Bargaining Council – those who do not follow the Council rules give their employees annual leave only, but this is the minority of cases. The owners may not have mentioned all of the leave employees may be entitled to but these were specifically mentioned

More than three-quarters of MSE owners believe that employees are aware of the labour law protection

MSE owners said that in the majority of cases they paid at the legal minimum rate or well above it; however, a significant minority admitted they did not do so

For the majority of workers, the benefits are as stipulated by the Council and are the contributions made by the employer and the employee. Those MSEs who do not register the employee do not pay the employee benefits that they are supposed to do so

Table 4: MSEs and compliance

Two-fifths of MSE owners have avoided complying with the provisions of the main agreement at some point, whilst the majority have always complied

There is a mixture of overtime provision pay, especially on a Sunday; almost one-third say that they pay double the daily wage rate, but others say that they pay one and one-half times only – this is below the threshold set by the agreement

The main areas of non-compliance include the non-payment of the minimum wage, not registering all the employees with the Council and paying over the benefits' contributions to the Council late or irregularly

More than four-fifths of MSE owners now comply with the law. They do so because they have always complied or, when they could not, they applied for an exemption and were able to ‘fix their problems’ and after a while, and when the exemption had expired, they could start complying again

Of those MSE owners who still do not comply, they will only do so if they see another compliance order from the Council

More than two-thirds of MSE owners have been visited by a Bargaining Council agent within the last year

More than three-quarters have not seen a Department of Labour inspector but over one-half have seen a Bargaining Council agent recently

As the Bargaining Councils have jurisdiction, and are accredited to perform dispute resolution services, one-half of disputes over the minimum conditions of employment within the MSEs were resolved by the Council

It is difficult for MSEs to quantify in percentage terms to cost complying with the contributions to the benefits – ranges from 0.5% of turnover to as much as 10%

Almost two-thirds of MSEs pay over their contributions to the Council, but a significant one-third do not

Almost two-thirds of the MSEs did not find the administration of the benefits funds by the Council as too helpful – this negative attitude can partly be attributed to the fact that many of them find the levels of administrative services provided by the Council too cumbersome

Almost two-thirds of MSE owners do not make use of any of the Council services and over two-thirds are not interested in making use of the Council services

5. Findings

More than two-thirds of the MSEs covered by the study are closed corporations. The majority of the owners are male, with over three-quarters older than age 45. Within the informal sector, the proportion of female-owned enterprises is considerably larger, thus constituting an important difference between MSEs in the formal and informal sectors. Almost two-thirds of owners started their business before 2001. Most small business fall within a short period of inception, and the firms encompassed by the study can be considered mature enterprises and survivors. The average workforce of firms interviewed was 26 employees. A range of different types of employment contracts were encountered, with one-half of employees being in permanent full-time employment. One-fifth of owners in our study classify the bulk of their employees as skilled and semi-skilled; most were using unskilled labour, a choice that would allow for immediate cost savings but makes it harder to move over to higher value-added strategies.

Significantly, over two-thirds of MSE owners interviewed did not belong to employer organisations; in turn, this would mean that the interests of a large proportion of MSEs are not properly represented on Bargaining Councils. Indeed, more than two-thirds of MSE owners interviewed believed that they drew no benefits from the Bargaining Council system. However, this did not deter many employers from making use of Bargaining Council services when it suited them, especially when it came to dispute resolution. More than one-half of industrial disputes over minimum conditions of employment within the MSEs studied were resolved by the Council.

5.1 Selective compliance with the law

Most employers complied reluctantly with labour law and were convinced that it inhibited the expansion and growth of their businesses. It was argued that:

… if we look at the labour laws, they are biased against the employer. The view is that the law is against employers and that the unions have too much power. The employer has to prove everything whilst the employee does not have to do so. It is impossible for employers to run a profitable business … the job losses are due to the high cost of labour. (Respondent 19)

Non-compliance was notable in four key areas. Firstly, almost one-half of MSEs paid below the minimum wage stipulated for the industry. Again, almost one-third of respondents admitted breaching the law, when it came to overtime pay. Most of the employers interviewed did not register all of their employees with the Bargaining Councils (statutory industry-specific centralised bargaining bodies), legal obligations notwithstanding, so as not to pay compulsory contributions to the Council and/or to better evade regulation. Underlying this partial disengagement was a view that the Bargaining Councils do not have MSE interests at heart and that the labour laws have been imposed on them. The main areas of non-compliance included paying less than the minimum wage, not registering employees with the Council, and, commonly, not adhering to overtime regulations. Stakeholder interviews suggested that non-compliance was particularly high among the unknown number of employers who chose not to register with the Council at all.

All employers reported having made use of overtime work at some or other stage. Two-fifths of employers interviewed admitted that they have not complied with the provisions of Bargaining Council Agreements when it came to overtime. Many admitted paying wages below the threshold commonly set by Bargaining Council Agreements.

One MSE owner in the clothing industry admitted that:

Non-compliance is rampant and the norm. In MSEs, non-compliance goes beyond wages: the small firms work 234 days per annum but will work on Saturdays and Sundays too in their busy periods. This is a responsible way of organizing working time but the Council sees any work on a Saturday or Sunday as overtime and want workers to be paid or else (in theory) the enterprise will be fined for being non-compliant. (Respondent 2)

Among the one-fifth who admitted that they still did not comply with the law at all, there was a general sentiment that they would only consider doing so if they received a compliance order from the relevant Bargaining Council.

A general concern voiced was the cost of some of the legal rights enjoyed by workers. For example, we were told of abuse of the sick-leave system where doctors' certificates are fraudulently obtained and employers are expected to pay for a full day's work; this led to suggestions that rights to sick leave needed tightening up. As Respondent 6 remarked:

… labour laws are nice for the worker but that there are huge fines that hang over the employer's heads for not complying and all the employer wants to do is close the company down as a result of this threat.

In addition to direct financial costs, a major factor behind non-compliance with Bargaining Council Agreements appeared to be the complex and lengthy nature of the collective bargaining agreements in each sector, making compliance difficult for a MSE with limited resources. Reflecting these sentiments, another respondent argued that:

I think that there should be as little regulation as possible for small business and I should be able to hire and fire at will and there should be no regulation of overtime hours. (Respondent 4)

The majority consensus amongst the 30 stakeholders interviewed was that MSEs often avoided registering at least some of their workers with the Councils. In part, this may be due to the growing number of – often illegal – foreign immigrants being employed in labour-intensive industries in South Africa from the Southern African region, as well as the upfront costs. Many MSE owners interviewed admitted that they only registered a small proportion of their workers with the Council, owing to the additional contributions and the associated red tape.

Also, as one owner manager remarked:

I only have 1 worker registered with the council but I have a workforce of 10 and most are foreign – registering all ten with the council would be too costly. (Respondent 3)

The latter interviewee did not say whether his foreign workers were legal or illegal immigrants, which may have underlain his decision. South Africa has a very large number of illegal immigrants (some estimates put the number as high as eight million or 20% of the overall population), who often undercut local jobseekers and are likely to acquiesce with lack of adherence to the law. Despite this non-compliance, three-quarters of respondents believed that their workers were aware of their rights under labour law. If this is correct, this would highlight a widespread inability by workers to enforce their rights, given their poor bargaining position.

In other words, what this information reveals is that employers appear able to evade those aspects of legal regulation not to their liking. In other words, formal regulation does not appear to be a direct constraint, although it may have indirect effects; for example, expansion would make a firm more conspicuous and hence more likely to be the target of any enforcement drive.

5.2 Compliance and benefit funds

Benefit funds (or industry-specific non-government-administered social security) are a significant aspect of Bargaining Council activity in South Africa (Godfrey et al., 2006:172). They encompass sick, provident, HIV/AIDS, holiday pay, and funeral or death benefit funds. However, the funds do add to the wage costs and are seen as onerous by some businesses. Again, many employers complained about the administrative burden imposed. It was also argued that many older workers would be unlikely to be able to utilise all of the benefits administered by the Council.

Respondents felt that it was difficult to quantify in percentage terms the of cost complying with the contributions to the various benefit funds administered by Bargaining Councils; with estimates ranging from 0.5% of turnover to as much as 10%. One-third of interviewees admitted that they did not pay their due contributions to their Bargaining Council. Almost two-thirds of the MSEs did not find the administration of the benefits funds particularly useful. Almost two-thirds of MSE owners interviewed were not interested in making use of the council services. Many felt that MSEs think that applying for exemptions from the Council is difficult or impossible to do because it involves too much paperwork.

5.3 Weak enforcement of the law

More than two-thirds of MSE owner respondents reported being visited by a Bargaining Council agent within the last year. However, over three-quarters of employers interviewed have never seen a Department of Labour inspector; this would suggest that official pressures to comply with legally binding agreements originate from the relevant Bargaining Council. This does, however, raise a question regarding the efficacy of law enforcement in terms of general labour relations legislation (as adverse to non-compliance with collective agreements reached at Bargaining Councils). Several owner-managers felt that inspectors were simply not competent. As one employer in the leather and footwear industry remarked:

The Department of Labour inspector carried out an Occupational Health and Safety (OHS) inspection only – he looked only at why workers did not have gloves on – he did not look for hazardous chemicals, the electric wiring etc … so his inspection was lacking. (Respondent 2)

One employer claimed that weak enforcement undermined the position of more principled employers, and that more should be done to make sure that all complied with regulatory provisions:

I know a lot of non-compliant employers – they keep their expenses down by paying below the wage rates – these are the micro enterprises who run the factory in a garage or an office and have 4–5 people; they do not pay taxes, they close down when there is no business – they are really small scale operators; they also tend to employ the migrant workers … they do have an unfair advantage. (Respondent 1)

Again, this highlights the extent to which employers are able to evade the law: however, what is clear is that the smallest operators have the most room for manoeuvre in this regard.

5.4 Exemptions from the law

Although MSEs may apply for exemptions from regulations on account of their size (see Table 1), respondents found it difficult to apply for exemptions on account of the burdensome paperwork. Almost four-fifths of respondents had never applied for exemptions. One clothing-manufacturing MSE owner stated that he was interested in applying for an exemption from the council but, ‘when I got a 12 page fax and I saw that I had to provide financial statements, I lost interest’.

It was not only the bureaucracy that deterred employers. For example, one MSE owner in the metal and engineering industry said that ‘applying for exemptions does not work as workers will not be happy with this’; any attempt to be seen to gain concessions from the system would be likely to be destabilising. Hence, informal evasion would be appear a more attractive option, rather than public attempts to seek formal concessions from the system.

5.5 Reluctance to hire

Even in the case of mature firms, there was a general reluctance to take on full-time workers. This can be linked to the fear amongst MSE owners of dismissing workers because the procedures to be followed are very complicated to follow. Hence, one employer (Respondent 4) argued that an ‘area of bother are the rules around retrenchment or dismissal – as employers are too scared to dismiss they land up [sic] not employing’. This reflected a general risk aversion when it came to dealing with labour matters among almost all owners interviewed. One-half of the MSEs interviewed outsourced certain activities, usually the labour-intensive part. Moreover, only one-half of the employees of interviewed owner-managers were in permanent employment.

5.6 Attitudes towards unions

A large number of interviewees were hostile to unions. One employer felt that ‘they eat your blood’. His workforce is not unionised, and he believes that unions are incapable of helping workers because they never helped him when he was a worker: ‘Unions do not have a vision of how business works and have unrealistic views on wages. The shop stewards help workers lose their jobs and they are not strategic enough’. In addition, it was argued that union officials need better training, and need to become more professional, improve their education and not work on threats alone. It was also felt that, unlike small businesses, unions have considerable power with government and regularly resort to the Department of Labour when there is problem. Again, it was argued that trade unions have ‘destroyed the industry and under a centralised system, the industry now has no power with the political union’ (Respondent 6).

This reluctance to engage with unions also extended to a reluctance to join employer associations: over two-thirds of MSE owners interviewed did not belong to one. As one owner manager argued:

… membership is very expensive and as a small business I see no benefit in being affiliated. (Respondent 7)

5.7 Further limitations of regulatory capacity

Although almost all of the MSE owner-managers interviewed complained about excessive regulation of work and employment relations, the majority also complained about inadequate and ineffective regulation in other areas. More specifically, there was much dissatisfaction with the apparent inability of the revenue authorities to control illegal imports and dumping. As one interviewee remarked, there was an urgent need to:

… stop the imports of cheaper goods and under-valued goods coming in [illegal imports], beef up the borders and customs as imported [products] negatively affects the industry. (Respondent 1)

Many interviewees complained that influxes of cheap manufactured goods – above all from China – made it extremely difficult to plan for the future. Illegal importers did not have to worry about the costs of complying with South African legislation in a range of areas from taxation to health and safety, both of the products and the workforce. Hence, competition was on an unequal footing, making it difficult to plan for the future and invest with confidence. Whilst the study concentrated on the major urban centre of the Gauteng region (encompassing Johannesburg), it is worth noting that there has been a proliferation of small-scale textile manufacturers across the South Africa's rural periphery, many of whom are little more than a front for the re-labelling of illegal exports.

6. The possibility of compromise?

The study revealed that many MSEs were partially disengaged from the wider employment relations system. This included not only a (perhaps predictable) hostility to unions, but also a reluctance to participate in employer associations, on account of high fees, and a perceived lack of responsiveness to the needs of small business. What appears to be the case is that excessive regulation does not always directly hamper MSEs. Rather, MSEs appear able to defy key aspects of formal regulation with relative impunity, and indeed selectively disengage from key industrial relations institutions and players. At the same time, this non-compliance may become more difficult as firms become larger and more conspicuous.

What this study further highlights is the extent of divisions in a country that combines pockets of prosperity with large-scale unemployment and poverty. On the one hand, MSE owners are generally highly critical of the state's inability to protect them against competitors who regularly evade import tariffs and other measures, and reward compliance in these areas of over-burdensome regulation of employment relations. On the other, the study revealed considerable evidence of hostility towards unions, and indicators of poor relationships with rank-and-file workers, notably in terms of absence. Whilst the present Bargaining Council arrangements are seen as dysfunctional for many MSEs, blanket exemption from their coverage is likely to worsen already poor relations with unions, and, it appears, in some instances, employees as well. A further risk is that a ‘bonfire of regulations’ may encourage further competitive rounds of cost-cutting, discouraging investment in higher value-added production paradigms, whilst still failing to provide a durable competitive advantage over ultra-low-cost competitors from abroad. In applied terms, the central proposition that the reluctance of MSEs to expand is simply a product of too much regulation is disproved: for MSEs to grow requires not just dispensing with inappropriate rules, but also creating effective rules that all parties to the employment relationship actively support.

The mistrust between employers, unions and the representatives of government, at least in part, probably reflects the bitter social divisions that characterised the apartheid era. As the literature on historical institutionalism alerts us, formative historical experiences may cast a long shadow over both formal rules and social relationships (Streeck, 2009). Hence, it could be argued, similar to South Africa's negotiated political settlement, there need to be similar negotiations and trade-offs between smaller employers, regulatory authorities, stakeholders, employees and their representatives in order to move towards a mutually acceptable modus operandi. The latter would be one whereby MSEs feel more confident to grow and create jobs, appropriate regulations provide a framework for consistency and basic trust, and employees and unions are confident that short-term concessions will genuinely lead to longer term benefits for the present and future workforce. However, such compromises will require considerable will on the side of all parties, and the ability and willingness of the state to provide resources to reward concessions and to support deals reached, given the great resource constraints faced by both MSE owners and their employees. A workable compromise would seek to provide positive complementarities through working to find a middle ground between wage restraint and job creation, flexibility in adjusting workforce sizes and managing performance against predictability and continuity in employment relations, and lighter, more effective and better regulatory coverage.

7. Conclusion

This study has revealed that most of the small firms interviewed were locked into a ‘coping mode’, being extremely reluctant to expand, for fear of attracting the attentions of the regulatory authorities and unions. More specifically, despite very high unemployment, they were generally reluctant to hire labour. Whilst many openly admitted to not complying with labour law, they were similarly critical of the lack of effective regulation – and enforcement of the law – in other areas, above all in protecting them from competition from the Far East. This highlights the limitations of theoretical accounts that focus exclusively on the formal institutional foundations of property rights, rather than their operation in practice (cf. La Porta et al., 1999). More specifically, it is evident that lighter regulation alone would not provide more favourable circumstances for small businesses to expand: what is more important is better and more focused regulation, and one that enjoys greater legitimacy amongst key players.

As noted earlier, this study only covers a relatively small cross-section of employers within the formal component of the South African MSE sector, and the views of a limited number of stakeholders. However, the findings raise a number of issues of broader relevance. MSEs appear to have been marginalised by the official employment relations system; compliance is reluctant, and most employers appear at least partially disengaged from the Bargaining Councils system, which is seen as serving larger employers and unions. Whilst there is much room for debate as to the desirable scope and scale for exempting MSEs from complying with basic labour standards, it is evident that the system is dysfunctional even in terms of accessibility; the interviews revealed that many MSEs appear to have been deterred from engaging with Bargaining Councils by its apparent complexity and bureaucratic requirements.

Whilst there is insufficient evidence to suggest that employment relations in all South African MSEs are characterised by very low trust, it appears evident that a meaningful basis of consent has yet to be established in the cross-section of firms encompassed by this study, in contrast to what is the case in many of their larger counterparts (Wood & Glaister, 2008). Such low trust imposes clear costs on employees. They cannot count on employment security, advancement, or, even, in some instances, equitable treatment. However, this also imposes costs on employers, ranging from difficulties in managing absenteeism through to the threat of overt collective action (Cohen, 1994). Low trust would discourage the accumulation of organisation specific human capital, and weaken the personal ties that would otherwise make working for small business more attractive. On the one hand, MSEs in South Africa will have to make significant adjustments and accommodations both to contribute to job creation and to secure their future prosperity. On the other, this will also require meaningful compromises by organised labour and employees, and compensatory action by the state. The latter would entail not just more flexible employment regulation, but also better enforcement of the laws that would protect MSEs from unfair competition. Such an accommodation could unlock new complementarities, building on existing strengths (above all, resilience), whilst compensating for the weaknesses imposed by the legacies of the past. However, any compromise will only work if past mistrusts are transcended, and immediate struggles for resources and survival are placed in the context of sustainability.

Acknowledgements

The authors are indebted to the constructive and extremely helpful comments and suggestions from the anonymous referees.