Scottish Government

CHAPTER TWO PROFILING RSA BENEFICIARIES IN SCOTLAND

Introduction

2.1 This chapter, as a prelude to the modelling of the impact of RSA in Scotland in Chapter 3, considers the characteristics of the beneficiaries of RSA financial assistance in Scotland compared to the control group of non-assisted firms. Our objective here is to identify the key differences between each group - highlighting differences in the characteristics of assisted firms - but also to identify differences which might influence differential performance between the two groups.

2.2 Throughout the chapter it is important to bear in mind the relatively small size of the Scottish sample - 157 RSA-assisted firms and the same number of non-assisted firms. This will limit the ability to disaggregate some of the variables for comparative purposes due to the constraints of the appropriate statistical tests. Further, it will lead to the need to undertake the econometric analysis in Chapter 3 for only the full sample of all RSA-assisted firms.

2.3 The chapter discusses in turn each of the following comparisons:

• The characteristics of assisted firms and the non-assisted group in terms of ownership, corporate structure and firm size and age, all factors which have been shown to be important in other studies of business growth;
• The characteristics of the partners and directors of assisted and non-assisted firms reflecting the quality of leadership in these enterprises and their openness to sharing equity as a means of promoting growth;
• The strategic direction of assisted and non-assisted firms and the extent of business planning in the enterprise. Again both factors have been linked to business growth in previous studies;
• The market situation and local market presence of assisted and non-assisted firms. In each case these factors will help to explain business performance. They are also important, however, in terms of shaping the multiplier or spillover effects from assisted firms;
• The extent of R&D and innovation, which will again contribute to our understanding of firm performance and the outcomes of assistance;
• Business growth - which will be the key focus of the analysis of Chapter 3 - and estimates of GVA measures derived from a series of questions on purchased inputs.

Characteristics of RSA beneficiaries

2.4 As Table 2.1 suggests, no statistically significant difference in ownership patterns between RSA recipients in Scotland and non-recipients was evident with around 85 per cent of RSA recipients being UK owned. Significant differences do emerge, however, between the proportion of assisted and non-assisted firms which are part of a group of multi-site firms. 69.9 per cent of RSA recipients in Scotland were members of multi-plant firms compared to only 53.5 per cent of the non-assisted group ¹⁰. The average vintage of businesses assisted by RSA in Scotland (21.1 years) was also significantly less than that of the control group (28.4. years) ¹¹.

Table 2.1: Nationality of ownership

Note: ? ² (2) = 2.026, ? = 0.363.

2.5 Business growth and the initial scale or size of the company is normally related, with larger firms generally tending to grow more slowly than smaller ones. Here two measures of size are considered: the overall size of the business being assisted (i.e. the whole company world-wide), and then the size of the assisted site itself. Table 2.2 summarises firms' responses in terms of the proportion of firms falling into various employment size bands with no significant difference in the profile of company employment in the RSA-assisted firms in Scotland and the non-assisted group. There is, however, evidence that RSA-assisted sites tend to be larger than non-assisted plants (Table 2.3).

Table 2.2: Business Size (all sites)

Note: There is no significant difference in the profile of company employment in RSA beneficiaries in Scotland and non-assisted businesses (? ² (7) =4.101, ? =0.768).

Table 2.3: Business Size in Scotland

Note: A significant difference is evident between the employment of RSA-assisted sites compared to non-beneficiaries ( ? ² (5) =16.064, ? =0.007)

Partners and directors among UK-owned single-site businesses

2.6 In this section we focus on a small number of indicators reflecting different dimensions of the leadership of RSA beneficiaries and non- beneficiaries. The choice of indicators here reflects our previous experience of modelling small business performance as well as more general studies which have emphasised the importance of the characteristics, background and attitudes of the Managing Director ( MD) of a business in shaping its growth performance ¹². Note, however, that this type of indicator is less relevant in multi-national organisations. We therefore restrict our comparisons here to the group of UK-owned single-site businesses in the group of Scottish RSA-assisted and non-assisted businesses. To set this group in context, Table 2.4 compares employment in UK-owned, single site RSA-assisted firms in Scotland and other RSA beneficiaries. UK-owned single-site RSA beneficiaries are markedly smaller than other RSA beneficiaries with only 2.0 per cent having employment greater than 250 compared to 14.5 per cent of all other RSA beneficiaries. Note also the relatively small size (101) of the Scottish group of UK-owned, single-site RSA beneficiaries.

Table 2.4: Business Size: UK-owned, single-site firms and other RSA beneficiaries (Assisted site only)

Note: Significant differences are evident between the distribution of employment in UK-owned, single-sites and other RSA-assisted sites (? ² (5) = 22.138, ? < 0.000).

2.7 First, in terms of assisted and non-assisted businesses involvement of non-executive directors there is only a weak statistical difference between the proportion of RSA assisted firms with non-executive directors (19.2 per cent) and that for non-recipients (17.7 per cent). Similarly, no significant difference was evident between the ownership stake of the MDs of RSA beneficiaries in Scotland and the general population of similar firms
(Table 2.5). In terms of the MDs' involvement in other business start-ups and willingness to share equity, however, we find significant differences between the Scottish RSA-assisted and non-assisted groups.

Table 2.5: Non-executives and Role of Managing Director or Owner

Note: Table only applies to UK-owned, single-site companies. Test statistics were: for non-executive directors (t=0.250, ?=0.803); for ownership stake (t=-0.690,? =0.491); for involvement in other business start-ups (t=2.337, ? =0.021); and, in terms of willingness to share equity (t=4.471, ?<0.000),

2.8 In terms of other characteristics of company MDs, we find no significant difference in the qualification profile of MDs between assisted and non-assisted groups (Table 2.6). Similarly, we find no clear difference between the age distribution of the MDs of RSA beneficiaries in Scotland and those of non-beneficiaries (Table 2.7).

Table 2.6: Highest qualification level of Managing Director or Owner

Note: Table only applies to UK-owned, single-site companies. Difference between the distribution of highest qualification of MDs of RSA beneficiaries and non-beneficiaries is not significant (? ² (5)=5.889, ?=0.317).

Table 2.7: Age band of Managing Director or Owner

Note: Table only applies to UK-owned, single-site companies. Difference between the distribution of ages of MDs of RSA beneficiaries and non-beneficiaries is not significant (? ²(5)=7.299, ?=0.199).

Strategic direction and management among UK-owned, single-site businesses

2.9 We focus now on the strategic direction and management approach of assisted and non-assisted firms, but it is worth noting initially that 84 per cent of RSA beneficiaries had a business plan compared to 71.3 per cent of non-beneficiaries in Scotland ¹³. Firms were asked to identify their main business objective (with multiple responses being allowed) with a focus on whether their aim was maintaining or increasing their market share with current products or the development of new products or services. The percentage of firms giving each response is given in Table 2.8. RSA beneficiaries in Scotland were significantly more likely to emphasise increasing sales than non-beneficiaries but there was no statistical difference between the two groups in their likelihood to emphasise the development of new products or services as their key business objective (see notes to Table 2.8).

Table 2.8: Business objectives

Notes: Table only applies to UK-owned, single-site companies. Statistics are as follows for firms emphasising maintaining sales as their main business objective: t=-1.247, ? =0.213; for firms emphasising increasing sales as their main business objective: t=-2.401, ? =0.017; for firms emphasising the development of new products or services as their main business objective: t=-0.412, ? =0.681.

2.10 In terms of management style we asked firms to indicate which approaches most closely matched their management approach, again allowing multiple responses. Four alternatives were provided, intended to provide an indication of the extent to which managers were adopting a hierarchic or more consensual approach to managing and organising the business. No significant differences were identified between management practices in RSA beneficiaries in Scotland and the non-assisted group (see notes to
Table 2.9).

Table 2.9: Management approach

Notes: Table only applies to UK-owned, single-site companies. Statistics are as follows for firms emphasising teamworking: t=-1.293, ? = 0.197; for firms emphasising close supervision t=-1.570, ? =0.117; for firms emphasising the development of standardised working practices: t=-0.521, ? =0.603 ; for firms emphasising initial selection and staff development: t=1.092, ? =0.276.

Markets and local linkages ¹⁴

2.11 Comparing profiles of export intensity between groups highlights no significant differences between the group of all RSA beneficiaries in Scotland and the entire non-beneficiary group (Table 2.10). Nearly a fifth of RSA-assisted firms and around a quarter of non-assisted firms had few export sales (less than 5 per cent). In terms of the proportion of firms indicating that local sales accounted for an important segment of their sales, we find a more significant difference (Table 2.11). The impression created here, however, is that in terms of export intensity RSA beneficiaries were broadly in line with the non-assisted population.

Table 2.10: Export intensity

Note: There is no significant difference between the export intensity of RSA beneficiaries and non-beneficiaries (? ² (6)=5.448, ?=0.488).

Table 2.11: Proportion of firms indicating that sales are local (i.e. within 20 miles)

Note: Local market orientation of RSA beneficiaries is weaker than that for non-beneficiaries (? ² (4) = 10.847, ? = 0.013).

2.12 Some more significant differences were also evident in terms of firms' assessments of the cross and own price sensitivity of the markets in which they were operating. Specifically, in the survey firms were asked by how much the demand for their own product would change if they, or their main competitor, were able to reduce their price by 10 per cent. In terms of the cross price sensitivity, very similar impacts were anticipated by RSA beneficiaries in Scotland and non-beneficiaries. In terms of own-price sensitivity there were statistically significant differences between RSA beneficiaries and non-beneficiaries but these suggest no very clear difference in overall price sensitivity (see notes to Table 2.12). We also find no significant difference in the proportion of firms (68-74 per cent) in each group which regard themselves as facing intense or very intense competition in their main markets ¹⁵.

Table 2.12: Demand response to own and competitor price changes

Notes: Statistics are as follows for firms' cross-price sensitivity: ? ²(4) =0.809, ? =0.937; for firms' own-price sensitivity: ? ² (4)=11.090, ?=0.039.

2.13 As part of the survey we also asked firms about their purchasing patterns asking businesses to identify the location of their suppliers. Purchasing patterns with local suppliers were broadly similar across the groups with RSA-assisted firms having a similar profile of national and international suppliers to non-assisted firms. The only significant difference between RSA-assisted and non-assisted businesses in Scotland was that RSA beneficiaries were less likely to be sourcing from other parts of their parent group
(Table 2.13). There was also no clear difference in the distribution of the proportion of purchases being made locally by RSA beneficiaries in Scotland and the non-beneficiary group (Table 2.14).

Table 2.13: Sources of Inputs

Notes: Statistics are as follows for firms reporting purchases from other group firms: t=-1.948, ? =0.053; for firms reporting purchases from other local suppliers: t=-0.113, ? =0.910; for firms reporting purchases from other regional suppliers: RSA beneficiaries , t=-0.564, ? =0.573; for firms reporting purchases from other UK firms outside their home region: t=1.256, ? =0.210; for firms reporting international purchases: t=0.339, ? =0.735.

Table 2.14: Proportion of firms indicating local purchasing intensity

Note: Comparison of the distribution of the proportion of purchases which were local : (? ²(3)=5.101, ?=0.165).

R&D and innovation

2.14 More significant was the difference between the proportion of RSA recipients and non-recipients undertaking R&D in Scotland (47.8 per cent of RSA recipients but only 29.0 per cent of non-recipients). This difference seems to be dominated by the different focus of R&D and development activity in RSA beneficiaries in Scotland with a stronger emphasis on product-oriented research but a similar emphasis on process related R&D to the non-assisted group (Table 2.15). The difference in the proportion of RSA beneficiaries and non-assisted firms undertaking R&D is reflected in higher proportions of both product innovators and process innovators among RSA beneficiaries (Table 2.15).

Table 2.15: R&D and Innovation Activity

Note: Statistics were: for the proportion of firms undertaking R&D: t=3.457, ?=0.001; basic product research t=2.252, ?= 0.025; applied product research t=2.957, ?=0.003; basic process research t=0.790, ?=0.430; applied process research, t=1.384, ?= 0.167; product innovation, t=3.866, ?=0.000; and, process innovation, t=3.743, ?=0.000.

Business growth

2.15 The aim in this section is to compare the growth rates of RSA beneficiaries and non-beneficiaries before, during and after the period of assistance. This is complicated somewhat, however, because newer start-up firms may have particularly fast percentage growth rates and may therefore distort group averages. To avoid this problem we focus our comparisons on firms in each group which existed in 2000. Table 2.16 below summarises the key growth comparisons for RSA-assisted firms and the non-beneficiary group. In each period both employment and turnover growth of the RSA beneficiaries is more rapid than that of non-beneficiaries. The suggestion is that prior to receiving grant support and afterwards RSA beneficiaries were outperforming non-beneficiaries.

2.16 The difference in growth performance is statistically significant for each variable and period with the exception of employment growth in the 2000 to 2002 period. Additional non-parametric tests, designed to allow for a sample which may not be normally distributed, also emphasise the very different distributions of growth rates among RSA beneficiaries and non-beneficiaries in each period and for both employment and sales growth (see notes to Table 2.16). Both mean and median values are presented to illustrate the extent to which the samples are not normally distributed. In each case, the median values are lower than the mean values indicating that the distribution of the samples is skewed. The non-parametric statistical tests allow for this and the differences are still significant.

Table 2.16: Growth comparison - assisted and non-assisted

Notes

1. Growth comparisons between RSA beneficiaries and non-beneficiaries are as follows: employment growth 2000 to 2002, t=2.548, ? =0.012; employment growth 2002 to 2004, t=3.717, ? =0.000; employment growth 2004 to 2006, t=4.643, ? =0.000; turnover growth 2000 to 2002, t=0.897, ? =0.871; turnover growth 2002 to 2004, t=2.439, ? <0.016; turnover growth 2004 to 2006, t=3.625, ? =0.000.
2. Non-parametric test Mann-Whitney tests are as follows: employment growth 2000 to 2002, Z=-4.443, ? < 0.000; employment growth 2002 to 2004, Z= -5.061, ? < 0.000; employment growth 2004 to 2006, Z=-5.571, ? < 0.000; sales growth 2000 to 2002, Z=-1.924, ? < 0.000; sales growth 2002 to 2004, Z= -3.312, ? < 0.000; sales growth 2004 to 2006, Z=-3.523, ? < 0.000.

GVA comparisons

2.17 As discussed in Chapter 1 we have attempted to estimate the impact of assistance on the value added at firm or plant level by using a proxy based on the percentage of annual turnover taken up by purchased inputs. This proxy has been developed in recognition of the difficulty in obtaining GVA information directly from the respondent business through the telephone survey. We present below the outcome of the analysis of the proxy variable for 2004 and 2006 (Table 2.17). Again, we concentrate on only those firms/plants that were in existence in 2000 in order to control for the effects of newer firms and plants growing faster.

Table 2.17: GVA Comparison - assisted and non-assisted

Notes

1. Growth comparisons between RSA beneficiaries and non-beneficiaries are as follows: GVA growth 2004 to 2006, t=3.00, ? =0.003; GVA per employee growth 2004 to 2006, t= -1.135, ? =0.259.
2. Non-parametric Mann-Whitney U tests are as follows: GVA growth 2004 to 2006, Z=-3.103, ? =0.002; GVA per employee growth 2004 to 2006, Z=-0.700, ? =0.484.

2.18 Over the period 2004 and 2006 growth in GVA of the RSA beneficiaries is more rapid than that of non-beneficiaries both in terms of mean and median growth rates. The difference in growth performance is statistically significant. However, with respect to GVA per employee growth is higher among the non-assisted group in terms of both the mean and median growth rates although this difference is not statistically significant. The same picture is suggested by the non-parametric tests which suggest a marked difference in growth in GVA but no significant difference in the growth of GVA per employee.

2.19 A final point to note is that the number of respondents in the sample providing even this proxy information is much reduced and will constrain its usefulness as a dependent variable in the econometric modelling of the effects of the assistance. However, the GVA per employee information derived here will be used later as one of the estimates in the calculation of the net benefit of the RSA Scheme in Scotland.

Summary

2.20 The previous descriptive analysis has highlighted a number of significant differences between the group of RSA-assisted and non-assisted firms in Scotland. This is important as it underlines the value of a multivariate analysis of firm growth and the impact of RSA assistance while controlling for selection effects. In this section we summarise the key differences between the RSA-assisted firms and those in the non-assisted group.

2.21 The key contrasts between the characteristics and performance of RSA beneficiaries and non-beneficiaries are:

• RSA beneficiaries grew faster than non-beneficiaries both before and after receiving assistance.
• RSA beneficiaries in Scotland are more likely to be part of multi-site groups than non-beneficiaries and tend to be larger than non-beneficiaries.
• RSA beneficiaries are broadly similar to non-beneficiaries in terms of their export focus as well as their assessment of the impact of a fall in their or a main competitor's prices on the demand for their products and services (i.e. own and cross price elasticities). That is to say, they are no more likely than non-beneficiaries to be responsive to such price changes. For both groups, the degree of responsiveness to such price changes is relatively small and from this we conclude that both groups do not compete primarily on the basis of price. As a result, for the assisted group of firms, we can infer that the displacement effects may be low.
• RSA beneficiaries are less likely to be selling to the public sector and individual consumers than firms in the general population.
• RSA beneficiaries are more likely to be undertaking R&D and product and process innovation than non-beneficiaries.
• Finally, RSA beneficiaries have faster GVA growth than non-beneficiaries although there is no robust difference between growth rates of GVA per employee.