HOW DO WE MEASURE ENTREPRENEURSHIP FROM A
GENDER PERSPECTIVE? OVERVIEW OF CURRENT
APPROACHES AND EXISTING DATA SOURCES
EDGE Technical Meeting
New York, 5-6 December 2013
Mario Piacentini,
OECD Statistics Directorate
Objective: measure gender differences
in entrepreneurship
Identify smart
indicators of
gender
differences in
entrepreneurship
Define the
population of
interest
Find sustainable
processes and
methodologies
to collect the
data
Start from a review of issues,
options and existing data
2
Defining the entrepreneurs
• Who is an entrepreneur?
–
–
–
–
The founder of a start-up?
A board member of a publicly listed company?
The owner of a small, home-based business?
An innovating manager?
• Little convergence on definitions among researchers
• The typical traits of ‘entrepreneurs’ – willingness to take
risk, innovativeness, problem-solving – are difficult to
observe and measure.
3
Definition of entrepreneur in the EIP
The OECD/Eurostat Entrepreneurship Programme (EIP) defines
entrepreneurs as:
“Entrepreneurs are those persons (business owners) who seek to generate
value, through the creation or expansion of economic activity, by
identifying and exploiting new products, processes or markets.”
The entrepreneurs are business owners who:
1) Make an investment to put in place and manage an activity involving a
degree of risk and uncertainty;
2) the outcome of their activity needs to be ‘novel’;
3) the innovation embodied in the activity needs to generate economic value.
Ownership
Management
4
From a conceptual to an operational
definition: some open questions
Size thresholds for the business
Minimum
size: Only
employer
enterprises?
• + : comparability; ‘casual’
businesses are excluded
• - : exclusion of a relevant
population; possible gender bias
?
?
Maximum • +: defining the gender of large
corporations is problematic
size: only
SMEs?
• - : upper bound is difficult to define
Specific
legal forms? in practice
5
Definition: Open questions
Mode of acquisition of ownership
Only business
founders?
• + : more homogeneous
population
• - : possible to be
entrepreneurial no matter
the mode of acquisition
?
6
POPULATION-BASED DATA
7
Self-employment data from labour force
or general household surveys
Advantages:
• high quality;
• timely;
• good international coverage.
Shortcomings:
• not all the self-employed are entrepreneurs;
• no information on the business other than its size
and sector of activity;
• Comparability issues (treatment of incorporated
self-employed).
8
Use of self-employment data from
labour force or household surveys
With current data
With a new
entrepreneurship
module
Number of
employers and own
account workers
Cleaner identification:
distinguish free
professionals, current and
discontinued owners
Characteristics and
sector of activity of
the self-employed
Entrepreneurial
motivations, conditions at
start, business
characteristics, etc.
9
Basic indicator of participation: Share of
women employers
Share of employers (self-employed with employees) who
are women, 2000-2010
%
50
2000
2005
2010
45
40
35
30
25
20
15
10
5
0
Brazil
Canada
Chile
EU 27
Japan
Korea
Mexico
United Australia Indonesia Russia
States
The share of women among employers has only
marginally grown over the last decade in most OECD
and G20 countries
10
Trends in ‘own-account’ and ‘employer’ series
Women, EU27
own-account workers
employers
7 800
7 700
2 450
7 600
2 400
7 500
7 400
2 350
2 300
2008
Q2
2 250
Diverging trends for number of women own-account
workers and employers during the crisis
7 300
7 200
7 100
Self-employed persons without employees
Self-employed persons with employees
(employers)
2 500
11
Indicators on characteristics of self-employed
women and men
Percentage of self-employed who completed tertiary education,2010
%
Women
Men
60
50
40
30
20
10
0
Self-employed women have higher educational attainment
than men
12
Measuring profits from household surveys
Gender gap in self-employment earnings
2010
70
60
• Gender gap ranges from
10 to over 60%
• Accounting for both
benefits and losses
50
40
30
20
10
0
Source: Entrepreneurship at a Glance (2012)
• Measure is not fully
harmonized; available
only for the
unincorporated selfemployed; profits can
be difficult to estimate
for respondents
• Other measures: hourly
earnings? drawings vs.
profits
13
Models for survey’s design and questions: mixedsurveys of micro-business owners
Mixed entrepreneur-enterprise surveys
• use an existing household-level data collection as
sampling frame;
• provide estimates of the size and economic performance
of micro-businesses, registered or not;
• include modules on capital and intermediates, expenses
and earnings;
• include detailed information on micro-entrepreneurs:
reasons for starting, employment history, time spent on
the business.
14
Surveys of micro-business owners and
informality
Percentage of small and micro-enterprises owned by women, Mexico
Informal
%
45
Formal
40
35
30
25
20
1992
1994
1996
1998
2002
2008
Source: Closing the Gender Gap: Act now (2012)
• The methodology allows to capture data on gender issues in the informal
sector.
• In the Mexican case (ENAMIN Survey), the share of female owners is
higher for non-registered than for registered enterprises
15
Models for survey’s questions: Surveys of
entrepreneurial attitudes and activities
• ‘Unofficial’ surveys conducted by research consortia or
research companies.
• Phone or face-to-face interviews of randomized
individuals.
• The Global Entrepreneurship Monitor (GEM) is the best
known example.
• Very rich information on attitudes towards
entrepreneurship, risk tolerance, expectations, access and
use of networks, etc.
16
Measuring gender differences in
entrepreneurial attitudes
Self-assessed feasibility of self-employment, 2012
Women
Men
80
70
60
50
40
30
20
10
0
Source; Flash Eurobarometer on Entrepreneurship, 2012
Survey question: ‘would it be feasible for you to become
self-employed within the next five years?’
17
FIRM-LEVEL DATA
18
Use of firm-level data: advantages and
limitations
Advantage:
• more suited than population surveys to the
analysis of differences in the performance of firms
owned and controlled by women and men.
Issues:
• Limited availability of comparable business
surveys with information on owners;
• limited availability of linked business and
population registers;
• very small businesses might not be covered.
19
Identifying ‘women-run’ enterprises
• Women-run enterprises are those enterprises where
one or more women control the majority of shareholding
and management.
• In practice, identification is easy only for the enterprises
with a single owner
• When there is more than one owner
Surveys
Define a robust set of
survey questions on
distribution of
ownership/management
Administrative data
Integrate information
on
shareholding/declared
revenues of owners in
business registers
20
Questions to identify the gender of
firm’s owners in World Bank data
• Core Enterprise Survey questionnaire:
-
Female Participation in Ownership: ‘are any of the
owners female?’
• In the 2013 Manufacturing Module:
- Percentage of female ownership: ‘What percentage of
the firm is owned by females?‘
• Informal Surveys:
– Gender of largest owner: ‘Is the largest owner (the
person most active in the operation of the firm) female?’
• African Indicator Surveys:
– Gender of majority of the owners: ‘Are the majority of
the owners female?
21
Definition in the U.S. Survey of
Business Owners
• Questionnaire asks for the gender and percentage of ownership for
up to four owners
• A firm is classified as women-owned if one or more women own
more than 50% of the business
Changes in number and employment of enterprises
by gender of owner, 2002-07, United States
%
25
Women-owned enterprises
Men-owned enterprises
20
15
10
5
0
-5
Source: Survey of Business Owners 2007
-10
% change number all firms
% change number employer firms
% change employees
22
Surveys on enterprises founders
• Examples: SINE
(France), Eurostat FOBS
Survival of female-founders by place of birth, France
France
Other EU
Other country
0.95
• Why interesting?
– Homogeneous,
policy-relevant
population
– sample selected from
business registers
– Possible longitudinal
design
0.90
0.85
0.80
0.75
0.70
0.65
0.60
0.55
0.50
1 year
2 years
3 years
23
OECD experience on use of business
registers
• Pilot project conducted in 2012 within the
framework of the OECD Gender Initiative;
• Objective: test the feasibility of disaggregating
EIP structural (size, industry) and
demography (births, deaths, survival)
indicators by gender;
• Coverage: 10 countries, only sole-proprietor
enterprises.
24
Demography indicators from business registers by
gender
3-Year survival rates of sole-proprietor enterprises
Women owned
Men owned
0.7
0.65
0.6
0.55
0.5
0.45
0.4
0.35
0.3
Source: Entrepreneurship at a Glance (2012)
25
Demography indicators from business
registers by gender
3-Year Employment growth rates
Women
Men
1.6
1.5
1.4
1.3
1.2
1.1
1.0
0.9
0.8
Spain
Netherlands
Slovak
Republic
Finland
Poland
Switzerland
Italy
New Zealand
Source: Entrepreneurship at a Glance (2012)
France
26
Lessons learnt from this project
• It is still difficult for most OECD countries to link
business registers with data on individuals.
• Extensive processing is often needed, and moving
beyond single-owner businesses is demanding on
data.
• Simpler indicators from administrative sources
should be considered (e.g. business registrations
by gender).
27
TAKING STOCK AND
MOVING FORWARD
28
A step-wise process with some open
questions
Policy
Question
Population
Available
tools
Indicators
• Participation gaps?
• Performance gaps?
• All business owners? Only founders?
• Use only existing data? New entrepreneurship
module in population surveys?
• Which indicators?
• Which ‘headline’ indicators?
29
Keep it realistic and sustainable
• Build as much as possible on existing data
– Explore further the feasibility of using administrative
sources
– Expand existing surveys or use existing survey frames
• Evaluate response-burden and collection costs
when considering different solutions
– are detailed asset and expenditure modules
necessary?
– should demographic information on all owners be
collected?
30
Extra slides
31
From questions to indicators
Issue
Indicators
Source
Entrepreneurial
Participation
Share of employers
Labour Force Surveys
Registered businesses, by gender
Administrative records
Entrepreneurial
Outcomes
Gender gap in business earnings
Household surveys
Share of exported sales
Business Survey/
Entrepreneurship module
Share of external credit in start-up
finances
Business Survey/
Entrepreneurship module
Hours worked for the business, by
presence of children
Entrepreneurship module
Entrepreneurial
resources and
constraints
32
Contextual indicators
Access to
finance
Work-family
balance
Entrepreneurial
and financial
education
Access to social
security
Informality
Senior
management
33