Self-employment in Wales: micro-business activity or the rise of the gig economy?

Introduction Across the UK selfemployment has been rising for some time and particularly since the onset of the global financial crisis. In fact, between 2007 and 2016 self-employment in the UK has risen from 3.6 million to 4.3 million, rising from around 12.5% of the workforce (aged 16 to 64) to over 14%. Self-employment is now at its highest level for 40 years. Although selfemployment rates in Wales are a little lower than in the UK as a whole and growth has not been as rapid, selfemployment here has risen over the same period from just over 12% to 13%. This growth in self-employment has not escaped the attention of politicians and Westminster policy makers (Deane, 2016; Mone, 2016; Taylor, 2017), nor of policy think tanks and bodies such as the Trades Union Congress (TUC). Some of this attention has been directed towards concerns about the transitory and precarious nature of selfemployment, the absence of social protection for the selfemployed in comparison to those in paid employment. This debate has crystalised around concerns about the rise in the so-called “gig economy”. Such new internet-based “platform” businesses are allegedly encouraging workers into self-employment status as a route toward increased labour flexibility and avoidance of potentially costly employment protection rights and other benefits which accrue to those in a paid employment relationship. Precise calculation of the size of the gig economy is problematic because of the lack of any agreed definition. The gig economy comprises both self-employed workers and those employed on zerohours contracts. It has been estimated that by 2016 there were approximately 1.3 million people (4% of all of those in employment) working in the gig economy in the UK, and commentators suggest that the gig economy will continue to grow. The Chartered Institute of Personnel and Development estimates that almost two-fifths of those engaged in gig-economy work are doing so to supplement more traditional ‘employed status’ jobs. Further indicative information is shown in changes in the proportion of the self-employed who work as freelancers or subcontractors, sometimes termed the “dependent” selfemployed. Data from the Understanding Society household panel survey suggests that the proportion of these dependent selfemployed has risen slightly from 23.5% in 2009/10 to 25.7% in 2014/15. Much of the policy discussion about self-employment, over recent years, has consequently been about the pattern of growth in emerging types of dependent selfemployment, not associated with business ownership.

Thi s v e r sio n is b ei n g m a d e a v ail a bl e in a c c o r d a n c e wit h p u blis h e r p olici e s. S e e h t t p://o r c a . cf. a c. u k/ p olici e s. h t ml fo r u s a g e p olici e s. Co py ri g h t a n d m o r al ri g h t s fo r p u blic a tio n s m a d e a v ail a bl e in ORCA a r e r e t ai n e d by t h e c o py ri g h t h ol d e r s .

Introduction
Across the UK selfemployment has been rising for some time and particularly since the onset of the global financial crisis. In fact, between 2007 and 2016 self-employment in the UK has risen from 3.6 million to 4.3 million, rising from around 12.5% of the workforce (aged 16 to 64) to over 14%. Self-employment is now at its highest level for 40 years. Although selfemployment rates in Wales are a little lower than in the UK as a whole and growth has not been as rapid, selfemployment here has risen over the same period from just over 12% to 13%. This growth in self-employment has not escaped the attention of politicians and Westminster policy makers (Deane, 2016;Mone, 2016;Taylor, 2017), nor of policy think tanks and bodies such as the Trades Union Congress (TUC). Some of this attention has been directed towards concerns about the transitory and precarious nature of selfemployment, the absence of social protection for the selfemployed in comparison to those in paid employment. This debate has crystalised around concerns about the rise in the so-called "gig economy".
Such new internet-based "platform" businesses are allegedly encouraging workers into self-employment status as a route toward increased labour flexibility and avoidance of potentially costly employment protection rights and other benefits which accrue to those in a paid employment relationship.
Precise calculation of the size of the gig economy is problematic because of the lack of any agreed definition. The gig economy comprises both self-employed workers and those employed on zerohours contracts. It has been estimated that by 2016 there were approximately 1.3 million people (4% of all of those in employment) working in the gig economy in the UK, and commentators suggest that the gig economy will continue to grow.
The Chartered Institute of Personnel and Development estimates that almost two-fifths of those engaged in gig-economy work are doing so to supplement more traditional 'employed status' jobs. 1 Further indicative information is shown in changes in the proportion of the self-employed who work as freelancers or subcontractors, sometimes termed the "dependent" selfemployed. Data from the Understanding Society household panel survey suggests that the proportion of these dependent selfemployed has risen slightly from 23.5% in 2009/10 to 25.7% in 2014/15. Much of the policy discussion about self-employment, over recent years, has consequently been about the pattern of growth in emerging types of dependent selfemployment, not associated with business ownership.
Despite the attention in Westminster, there has been little or no discussion in the Welsh context. The direction of policy in the Welsh Government remains largely focused on the growth of key sectors, the importance of leading "anchor" businesses and institutions in supporting SMEs in supply chains, and the promotion of city regions which aspire to be internationally competitive.
However, the macroeconomic significance of the rise in self-employment has remained a peripheral concern, and the implications for City Deals and other agglomeration and leading sector-focused strategies on selfemployment across Wales remain little understood. This paper, which draws substantially from a report undertaken by the authors for the Federation of Small Businesses Wales (Henley and Lang, 2017), seeks to describe and analyse selfemployment in Wales in greater detail, to identify considerable diversity in the prevalence of selfemployment across the subregions of Wales, and contextualise the "gig economy" versus "selfemployment-asentrepreneurship" debate to Wales.
It reports a secondary analysis of the data from the 5% random sample of the 2011 Census as made available to researchers. This provides the largest potential sample of observations on the selfemployed in Wales and thus the opportunity to produce tabulations on a range of individual characteristics at a local level. This analysis is supplemented with further more up-to-date information from analysis of the UK Annual Population Survey.

Spatial aspects of selfemployment growth
A commonly expressed idea is that self-employment is counter-cyclical, and tends to rise at times and in places of reduced paid labour market opportunity. However, recent research (Henley, 2017) demonstrates that there is little evidence since 2008 of a negative 'recession-push' on self-employment, but, strong evidence for local demand 'pull', and that this is particularly true for selfemployed women. This suggests that people are more likely to choose selfemployment when local unemployment reduces and earnings improve.
The longer someone has been unemployed and higher the local unemployment rate, the less likely they are to move into self-employment. This supports the evidence that the majority of the increase in selfemployment, since 2008, has been because of entrepreneurial 'pull' decisions, rather than a dependency 'push'. One implication of these findings is that peripheral regions, such as Wales, have not seen such strong growth in opportunity selfemployment as has been observed closer to London and south east England. For these reasons, policies aimed at supporting the unemployed into entrepreneurship may increase spatial inequalities, as they are likely to have more impact in places with stronger existing local economies. The spatial differences experienced in selfemployment can therefore be understood as resulting from the interplay of a range of contextual factors, including in particular the existing strength of local economies and their historic sectoral structure. Economic policy to stimulate local demand or outputs for new businesses, rather than supply of entrepreneurs is, therefore, more likely to be effective in raising the overall number of selfemployed, whilst at the same time helping to overcome spatial economic inequalities.
Other place-based factors can also have a significant impact on the pattern and experience of selfemployment. The factors include existing and historical local entrepreneurial cultures, although these are often difficult to capture as measurable constructs. The size of the existing selfemployment base can be a strong indicator of likely future growth, as it can be indicative of an entrepreneurial culture, stronger business networks, and communities that are more supportive of the selfemployed (Goetz and Rupasingha;. The effects might be observed in terms of spatial distance from strong centres of business and innovative activity. Although caution needs to be exercised over causes and effects, this does appear to reinforce the view that the self-employed respond to economic signals, including the financial risk and benefits of self-employment. The lack of access to capital does not necessarily constrain the expansion of selfemployment, as selfemployment earnings and educational attainment have a major impact on selfemployment choices regardless of access to capital. Nevertheless, past research has shown that self-employment tends to correlate with housing wealth, as the value of home ownership is thought to provide a ready signal of credit-worthiness to lenders (Reuschke, 2016).

Other influences on micro-business activity
Although self-employment levels have been increasing in the UK since 2008, this does not appear to correlate strongly with job creation for others.
The factors that influence business growth do not appear to be necessarily the same as those that influence decisions on whether to become self-employed (Henley, 2016). Most selfemployed people do not establish businesses with the objective of growing them, and not all business will create jobs. Part of the explanation may be in the gradual convergence in UK average self-employment rates for men and women over this period. Female business founders, for a variety of reasons, appear to be statistically less likely to employ people than male business owners. One of the key policy challenges is not only to remove structural barriers to female selfemployment, but also from those women who wish to grow their businesses or employ others. Personal factors tend to influence self-employment career choices more than any other, including underlying economic conditions. A recent review of international evidence (Simoes et al., 2016) identified 12 critical individual factors, grouped into seven categories, that help to explain selfemployment career decision-making. These included: basic characteristics (gender, age, marital status, children); family background (parents and spouse); personality characteristics; human capital (education and experience); health condition; nationality and ethnicity; and access to financial resources. The propensity to enter selfemployment is generally greater for men, but, also greater for women who need more flexibility.
It also increases with age and experience, and for people with fewer financial constraints, but this tails off as people also tend to become more risk averse when they pass a certain age threshold. The propensity to become selfemployed is greater for people who are married, as well as for people who need more flexibility because of childcare. It is also greater when people have had at least one parent who had self-employment experience and, where benefits of a self-employed lifestyle are perceived directly from a partner's experience of being self-employed, but, conversely, higher where a partner is employed as it diversifies the risk. However, the effects of education and health conditions upon the propensity to become selfemployed remain uncertain.
A UK-wide survey conducted by the Federation of Small Businesses (FSB), of 1,600 of its self-employed members during 2015, sought to identify some of the key benefits and challenges to selfemployment as perceived by its members (Federation of Small Businesses, 2016). In common with previous research (Dawson et al., 2014), the survey found that independence and personal fulfilment were key attractors, but that income insecurity, lack of protection in the event of illness and the difficulties of finding new business were perceived as the most significant challenges.
Financial success tends not to figure as the main attractor for the self-employment, particularly in the case of women. This all serves to reinforce an argument that the career choices of the self-employed arise from a complex interaction of influences, many positive and some negative. This complexity is beyond those negative factors used to explain the growth in dependent selfemployment.

Diversity of selfemployment in Wales
The effect of place on selfemployment in Wales is significant. It is possible to divide local authorities in Wales into three groups, more of less equally sized in number: those with selfemployment rates of around 20%, those between 10%-20% and those under 10% respectively. These might be described as "rural", "urban and urban periphery" and "Valleys and deprived urban".
These local contexts appear to have a significant impact on local variations and types of selfemployment. 2 Although not entirely consistent with the data, the characterisation generally works well. The self-employment data for Newport, for example, looks more like a "Valleys" area than other South Wales urban centres. The available Census categorisation amalgamates Torfaen and Monmouthshire, the first of which may be a "Valleys and deprived urban" area and the second "rural". Nevertheless, the differences in the averages between the three groupings are statistically significant. High rural self-employment is not solely the result of a preponderance of farmers. Stripping out the primary sector only reduces the overall self-employment rate for the rural grouping from 18.4% to 16.3%, still well above other areas of Wales. Overall female selfemployment is less than half the rate for men. However, it is particularly low compared to men in the "Valleys" areas. In Rhondda-Cynon-Taff the female rate is a third of the male rate, and the male rate is below average. In rural Wales female selfemployment is absolutely and proportionately higher.  Selfemployment has not risen as fast elsewhere and, in some areas, has fallen slightly, including in rural Wales.
However, the starting point remains one of significantly higher self-employment in rural Wales.
The significant diversity in self-employment across Wales suggest that a one size fits all approach to supporting the selfemployed and, if desirable, the growth of self-employment, is likely to be inappropriate. Welsh Government is seeking to divide the economic governance of Wales into four regions: Cardiff Capital Region, Swansea Bay, Mid Wales, and North Wales (Welsh Government, 2017). The four regions approach to economic development currently being pursued however, does not appear to entirely fit with the patterns of selfemployment across Wales. This is important because the regional economic priorities identified by these new structures might cut across the variations found in self-employment, which, as has been identified, accounts for 38% of jobs growth in Wales since 2007.

Figure 1: Change in self-employment rate, Wales 2009 -2016
Source: Annual Population Survey via Nomisweb In all areas of Wales female selfemployment rates are well below the equivalent male rates, and can be as low as only a third of the male rate. The gap between female and male self-employment is highest in the Valleys and deprived urban areas of Wales, despite low overall self-employment in these areas.

Self-employment demographics across Wales
Here, there appears to be a double need to support self-employment and to address the gender gap. Although, at the UK level, there is some evidence of a closing gender gap, any previous policy initiatives in Wales to encourage women into selfemployment appear to have had limited success.
Self-employment rates increase significantly with age, and are typically a lot higher for over 45 year olds, and in most areas are even higher for over 60s. Selfemployment typically rises with age, mainly because of accumulated experience and other financial and social capital resources. On average across Wales 37.4% of the active workforce over 60 are selfemployed, but only 7.8% of the under 30s. Young self-employment tends to be higher in Cardiff and Swansea, but is proportionately lower, on average, in the Valleys, pointing to the need for further targeted policy intervention to encourage business start-up amongst the young. This is also true in rural areas, although the lower rate of young selfemployment is offset by higher absolute numbers. Other concerns, for example business succession and business leadership opportunities for younger people, may be an issue in rural areas.
It is possible in the Census data to distinguish between  of selfemployed people who employ others, but these proportions are of a smaller overall self-employment base. Therefore, areas with lower overall selfemployment tend to have a higher proportion of selfemployed employers; whereas those areas with the highest rates of selfemployment tend to have proportionately lower numbers of self-employed employers.
The percentage of selfemployed people who employ others increases with age in all areas of Wales.
In rural Wales, almost 80% of employers are over 45 years of age. This suggests that support is needed for younger selfemployed people in the skills and resources required to create jobs for others. So, in the context of low overall selfemployment in the Valleys, the key policy questions may be better focused on why opportunities for soletrading are particularly weak in these areas.
Compared to all English regions, Wales has the highest proportion of the self-employed with no educational qualifications and, correspondingly, the lowest proportion of graduate self-employed. In many areas of Wales there is a bimodal pattern of higher self-employment amongst those with no educational qualifications, as well as those with higher qualifications.
In urban Wales, however, there is less concentration of selfemployment amongst those with no qualifications and, significantly higher graduate level selfemployment. Generally, high rates of selfemployment among those with little or no qualifications should be a concern for policy and support programmes. Blanket support programmes for the selfemployed, available to graduates and nongraduates alike, may be inefficient and ineffective.

Other factors
One in four of the selfemployed in Wales have no fixed place of work and 44% work from home. The only English region with a comparably high level of home-based selfemployment is the South West. Over half of all selfemployment in rural Wales is home-based. For the home-based self-employed in rural areas, policy responses may need to address the lack of appropriate business premises, as well as issues relating to access to, and ability to capitalise on, high speed broadband services.
Wales has the highest rate of self-employment amongst outright home owners (i.e. without a mortgage) when compared with the English regions, with very high rates in rural Wales. Conversely, Wales has the lowest rate of selfemployment among private renters compared with the English regions. The rate of self-employed employers also tends to be higher amongst home-owners, especially those with no mortgage, and are again particularly high in rural Wales. To make sense of this it should be noted that outright ownership correlates strongly with age and this, perhaps, explains more strongly the regional variation. However, the possession of housing wealth is clearly a strong supporting factor for selfemployment.
Another financial supporting factor is diversified income within the household. Across Wales, 49% of selfemployed men live in couple households, and a further 20% are women who live in couple households. Only a very small proportion of the selfemployed in all Welsh areas live in single households, and this may be connected to the older age profile of the selfemployed generally. A sizeable minority of the selfemployed appear to be young people living with parents, particularly in rural Wales. In Ceredigion and Pembrokeshire, for example, 24% of the selfemployed are young people living with their parents.
The rural-urban split in selfemployment in Wales is also apparent in the breakdown of industrial classification of business activity. Note that location is based on place of residence and not place of work.
Unsurprisingly primary sector activity (agriculture, fisheries and forestry) is higher in rural Wales. Hotel and catering activity is also higher in rural Wales. In urban Wales self-employment is more concentrated in businessrelated service activity and health and educational services. In the Valleys there is notably higher construction sector selfemployment activity, although the place of work of these self-employed may vary from week to week. In fact, in the Valleys areas just over 1 in 3 of the selfemployed report having no fixed place of work, somewhat higher than in urban or rural Wales. For these self-employed with no fixed place of work, local transport infrastructure may be important to facilitate access to business opportunities. For some, improved access to appropriate, affordable business premises to facilitate business growth may also help.
Of all the self-employed in Wales, 1 in 3 also report working more than 48 hours per week. By comparison for the employed, where the European Working Time Directive applies, only 9% reported working more than 48 hours in the week of the Census.
A long-hours culture is clearly prevalent amongst the self-employment, and in Wales, particularly in rural Wales, the incidence of long hours is somewhat higher than in other English regions.

Discussion and conclusions
The rise in self-employment has accounted for 38% of total jobs growth in Wales between 2007 and 2016. However, a casual observer of Welsh Government economic strategy statements and commentary might be forgiven for thinking that this growth has barely registered on the policy radar.
Growing selfemployment is an important feature of the Welsh labour market and it seems unlikely that there will be any reversal in this trend soon. This impacts on a range of Welsh economic policy topics including macroeconomic conditions and the demand for labour, support for entrepreneurship and local economic development, business succession, networking and collaboration amongst selfemployed business owners, skills development for the self-employed, housing and transport policies, and microbusiness finance and the provision of business premises. Taken together, this list ought to form the basis of an urgent policy review in Wales. As the analysis here has shown, self-employment across Wales is far from uniform.
It varies considerably between rural and urban Wales, and low levels of self-employment in Valleys areas may arise as a consequence of particular barriers and challenges. A spatially uniform approach to self-employment and microbusiness support is highly unlikely to deliver the fine grain variation in support, across these various dimensions, needed to help the selfemployed and the prospective self-employed build businesses that will meet their aspirations and provide some measure of secure well-being and job satisfaction.
Much commentary on the recent UK-wide growth in self-employment has focused on the development of the "gig economy".
While not denying that the emergence of internet platform-based businesses such as Uber and Deliveroo has resulted in pressure on some to work on an insecure self-employed basis, the gig economy is only one part of the story, particularly in Wales. The regulatory policy responses required by this development are very different to those needed to support self-employed business owners, who continue to form the majority of the selfemployed both across the UK and in Wales in particular, and who remain an important element of what has become known as the foundational economy.