Transcript Do we all have to be (social) - London South Bank University

How enterprising?!
Do we all have to be (social)
entrepreneurs now?
Mike Gordon
Department of Geography
University of Sheffield
[email protected]
4th Annual Social Enterprise Research Conference
LSBU, London, 4th-5th July 2007
Does social enterprise do what it says on the label?
• “A social enterprise is a business with primarily social
objectives whose surpluses are principally reinvested for that
purpose in the business or in the community, rather than
being driven by the need to maximise profit for shareholders
and owners.” (DTI, 2002)
• How enterprising are social enterprises? Where are they?
Are they the product of a new culture of enterprise, or of
something else? Are they really businesses? What evidence
is there that the sector is moving towards financial
sustainability? What trading approaches are being adopted?
Do we all have to be (social) entrepreneurs now?
The location of social enterprises: some
alternative theories
•
•
•
•
•
Social enterprises are just like any other enterprise, so they will be more common
in more entrepreneurial areas than in less entrepreneurial ones.
Social enterprises are more likely to occur where other opportunities are few, so
we would expect to find them in more deprived communities.
Social enterprises are more likely to occur in local authority areas which have
received EU Structural Funds, because the EU has long supported the social
economy and these local authorities will be more geared towards the Third Sector.
Social enterprises are more liable to be found in local authority areas which have
received certain types of UK funding, because the Government has supported
community and social enterprise for some years
Social enterprises will be more common in regions with pro-active Regional
Development Agencies than in regions where the RDA is less active.
What data could we use to test these theories?
•
•
•
•
•
•
•
•
CIC registrations from August 2005 to end November 2006 – proxy for social
enterprise (599 CICs, less 44 for Wales, Scotland and the CoL = 555)
VAT registration data 2005 – indicator of SME start-ups and proxy for general
entrepreneurialism
Five of the six measures of deprivation in the IMD 2004 (two converted to ratios) –
to show the extent of deprivation in each local authority area
Average house prices in county/unitary areas in March 2006 (calculated from Land
Registry data for March 2007) and/or percentage of home ownership in each
district (derived from the Census 2001) – general indicators of wealth and likely
ability to raise or access business finance
District eligible for EU Objective 1 or Objective 2 Structural Funds, 2000-6
District in receipt of SRB4 or SRB5 or SRB6 or Phoenix Development Fund, or
eligible for NRF 2000-6
Regional dummy variables – to measure for RDA effects
Note that this analysis of CICs is rough and ready; it uses reasonably accessible
publicly available data and creates proxies to represent some of the concepts
being explored. The aim is to obtain a broad picture of what is happening, but the
analysis is not exact or comprehensive and has to be treated with some caution.
Constructing a regression model: playing with
the variables
•
•
•
•
•
Omitted Wales and Scotland from the analysis – some data absent, incomplete or
organised on a different basis
Excluded City of London because the model confirmed it as unusual – a very
extreme leverage case
Used principal components analysis to carry out an exploratory factor analysis,
because of the evident multicollinearity of the five deprivation variables – all five
loaded very strongly on the same component and the resulting factor score was
saved as a new, composite deprivation variable
Calculated the mean number of CICs per 100,000 adult population by region to
help decide which of the regional dummy variables to leave out of the final model
– NE region had highest mean and was chosen, so it serves as the comparator for
the other regions
Correlation of all potential predictor (X) variables, except the regional dummy
variables, indicated a certain degree of significant multicollinearity between
variables, so it was decided to drop SRB4, SRB5, NRF anf home ownership from the
final analysis
Modelling start-up rates of CICs in English LAs and excluding the City
of London: table of coefficients
B
(Constant)
Sig.
.470
.422
O1 funding 2000-2006
1.240
.010**
O2 funding 2000-2006
.593
.016**
SRB6 funding
.278
.151
-.299
.154
5.47E-.006
.081
SME per 100K population
.001
.331
Deprivation factor score
.453
.000**
NW dummy
-1.295
.002**
YH dummy
-.999
.036**
EM dummy
-.937
.028**
WM dummy
-1.413
.001**
EE dummy
-.567
.198
-1.289
.032**
SE dummy
-.567
.217
SW dummy
-.002
.996
Area with local Phoenix Development Fund project
Average house price (£) for LA area as at March 2006
London dummy
R²
.186
Adjusted R²
.149
F
Durbin-Watson
Dependent Variable: CICs per 100K population
.000**
1.832
* or ** significant at p<0.05
Key conclusions from the regression analysis 1
• There is a significant relationship between CICs and deprivation; all else
being equal, the more deprived a local authority area is, the more CICs per
100,000 adult population it will have (or, conversely, the more affluent the
local authority area, the fewer CICs per 100,000 adult population). One
implication of this is that Government policy towards social enterprise and
enterprise in deprived areas may be having some effect. However, this
has to be treated with caution and needs further exploration.
• Given that deprivation indicators in essence measure underlying poverty
in an area, it is not surprising that – with a significant relationship
between CIC formation rates and deprivation – there is no significant
relationship between CIC start-up rates and house prices. The latter were
used as a proxy for cost of capital, because they represent wealth and the
likelihood that people in areas with higher house prices should be able to
raise or otherwise access business funding.
Key conclusions from the regression analysis 2
• There is no significant relationship between the formation of CICs and
SMEs, so the rate of CIC start-ups appears to have nothing to do with
enterprise or a general entrepreneurial culture. The model implies that
CIC start-ups, and thus social enterprise more generally, are not a product
of local entrepreneurialism, nor are they a substitute for a lack of local
entrepreneurialism: the two things are unrelated.
• There is a significant relationship between the rate of CIC start-ups and
both EU Objective 1 and EU Objective 2 funding. The latter are dummy
variables, and the comparators are local authorities without the relevant
funding, so, for instance, the Objective 1 coefficient tells us that, other
things being equal, local authorities with Objective 1 funding had, on
average, 1.2 more CIC start-ups per 100,000 adult population than local
authorities without such funding. Taken with the previous point, this
suggests that CIC formation may have more to do with the availability of
certain types of funding, and the policies which accompany that funding,
than with enterprise culture or activity.
Key conclusions from the regression analysis 3
• Five of the eight regional dummy variables have a significant relationship
with the rates of formation of CICs. In all these cases, the relationship is a
reverse (negative) one. Remembering that the interpretation of these
dummy variable relationships has to be seen with respect to the omitted
dummy variable for the North East, we can say that, all other things being
equal, a local authority in the North West (for instance) can expect there
to be 1.3 fewer start-ups than an equivalent local authority in the North
East. In turn, a local authority in Yorkshire and the Humber can expect
there to be 1 fewer start-ups than an equivalent local authority in the
North East, and so on. However, there is no significant relationship
between the start-up rate of CICs and the regional variables for either the
East of England, the South East, or the South West regions. The variations
in rates between those regional variables which show a significant
relationship between themselves and CIC formation rates may suggest
differences in policy between those regions, but this would also require
further explanation.
Regression residuals for the South Yorkshire
local authorities
South Yorkshire
No. of
CICs/100K
Standardized
district
CICs
pop
residuals
Barnsley
0
0
-1.29
Doncaster
4
1.7
-0.05
Rotherham
0
0
-1.08
Sheffield
4
0.9
-0.33
The residuals are the difference between the actual and predicted scores from
the regression; the negative scores show that all four South Yorkshire districts
underperformed slightly against expectations when compared to the results for
England as a whole. That said, none of the South Yorkshire local authorities
were significantly underpredicted, and the model came close to predicting the
number of CICs in Sheffield and (especially) Doncaster pretty accurately. Both
authorities had roughly the same number of CICs present as national trends
would lead us to expect.
Distribution across South Yorkshire local authorities of social
enterprises responding to questionnaire survey
Local authority
Numbers of social
enterprises
Resident population
mid-1991
VAT registrations
2005
Barnsley
15 (12.2%)
223,200 (17.3%)
425 (16.0%)
Doncaster
16 (13.0%)
291,700 (22.6%)
575 (21.6%)
Rotherham
13 (10.6%)
253,700 (19.7%)
495 (18.6%)
Sheffield
79 (64.2%)
520,100 (40.4%)
1165 (43.8%)
Total: 123 enterprises
Although 40% of the county’s resident population live in Sheffield, 64% of South
Yorkshire’s social enterprises are located there. But Doncaster and Rotherham
have only 58% and 54% respectively as many social enterprises as their share of
the county population might lead one to expect. Nor is this simply a reflection of
different rates of entrepreneurialism between the authorities: the proportion of
the county’s VAT registration rates for 2005 in each local authority is roughly the
same as its share of the population. The level of social enterprise activity in
Sheffield is doubly unusual, therefore.
South Yorkshire social enterprises: trading
income
Approximate percentage of total income from trading
Valid
100%
75-99%
50-74%
25-49%
1-24%
0%
Total
Frequency
20
23
23
17
28
12
123
Percent
16.3
18.7
18.7
13.8
22.8
9.8
100.0
Cumulative
Percent
16.3
35.0
53.7
67.5
90.2
100.0
South Yorkshire social enterprises: financial
sustainability
Frequency
Valid
already 100%
financially self-sufficient
aiming for complete
financial self-sufficiency
aiming for 75-99% selfsufficiency
aiming for 50-74% selfsufficiency
aiming for <50% selfsufficiency
Total
Missing
Total
non-response
Percent
Valid
Cumulative
Percent
Percent
30
24.4
25.2
25.2
26
21.1
21.8
47.1
30
24.4
25.2
72.3
15
12.2
12.6
84.9
18
14.6
15.1
100.0
119
96.7
100.0
4
3.3
123
100.0
Trading income and financial self-sufficiency
• The DTI’s 2005 survey of UK social enterprises found that 82%
(£14.8 billion) of the nearly £18 billion turnover of CLGs and
IPSs was from trading revenues, and that 88% of those
surveyed generated 50% or more of their income through
trading (IFF, 2005, p.3, 15), but the UK survey said nothing
about the actual financial sustainability of the enterprises.
• In South Yorkshire, just 16% of questionnaire respondents
derived 100% of their income from trading, whilst 54%
obtained more than 50%. Just 25% said their social enterprise
was already 100% financially self-sufficient, with a further
22% aiming for that.
South Yorkshire social enterprises: trading
approaches
Intend to use**
Ranked #1
(No. and %)
(No. and %)
95 (84%)
56 (51%)
33 (29%)
12 (11%)
74 (66%)
38 (35%)
22 (20%)
7 (6%)
113 (100%)
110 (100%)
Trading approach
Income from selling goods and/or services in
the open market
Asset-based development, deriving rental
income from land and/or buildings
Income from public procurement/public
sector contracts/service level agreements
Other trading income
Total responding to question
** N.B. These categories are not mutually exclusive
Social enterprise: prospects as businesses?
•
•
•
•
•
•
For social enterprises, income from open market sales or from procurement are
both in competition with the private sector, but with the added disadvantage for
social enterprises of having to expend resources to meet social aims.
There are any number of suggested processes and methodologies for assessing the
added value of social enterprises, but apparently relatively little in the way of
formal requirements, commitment, acceptance, or resources to implement these.
Recent reports (e.g. NAO, 2007x2, DSC, 2007) do not seem to be optimistic about
progress with procurement; there are problems with the size and capacity of
voluntary sector organisations, the size of contracts, full cost recovery, profit
margins, and other barriers in dealing with public bodies.
The potential of asset-based development is recognised (Quirk, 2007), but much
more needs to be done; there can also be problems, such as heavy repair and
maintenance costs, and marginal returns.
Do social enterprises really have good prospects as businesses, or should they stick
to being ‘businesslike’?
If social enterprises are businesses with a social purpose, what are the real
distinctions between them and companies with a strong commitment to Corporate
Social Responsibility?