Does a Different View Create Something New?
The Impact of Employee Diversity and Labor Mobility on Innovation
Bram Timmermans
DRUID/IKE, Department of Business and Management
Email: bram@business.aau.dk
METEOR Seminar (TIID)
About me
•
Postdoc in Innovation Studies
•
Department of Business and Management, Aalborg University, Denmark
•
Member of the IKE group (currently 11 seniors and 5 PhD. Students)
•
Member of DRUID (AAU, CBS, SDU)
•
Coordinator of the MIKE program (MSc in innovation and entrepreneurship)
•
2-year Master 20-30 students per year
•
Economics and Business students
•
Co-organizer of the DRUID Academy (Winter) conference
•
Conference for PhD Students
•
70-80 students, 6 invited keynotes, 17 DRUID faculty
•
Three days in mid-January in Cambridge: see www.DRUID.dk
•
Current research interest:
•
Labour Mobility and Performance
•
Offshoring and Changes in Employment Composition
•
Employee Diversity and Innovation
•
Entrepreneurial Team Composition
•
Dynamics of the (Global) Roller Coaster Industry
21 December 2011
METEOR Seminar (TIID)
Presentation Overview
•
•
•
•
•
•
•
Introduction
Basic Idea
•
Composition Study
•
Mobility Study
Issue of Diversity
•
Some problems in measuring diversity
•
Theories of employee diversity and innovation
•
Hypotheses
Issue of Mobility
•
Positive effects of mobility
•
Negative effects of mobility
•
Absorptive capacity
Data and Method
Results
Conclusion
21 December 2011
METEOR Seminar (TIID)
INTRODUCTION
21 December 2011
METEOR Seminar (TIID)
21 December 2011
Introduction-diversity on the labour market
•
Changes in the economy, industry structure, population, labor market and
work force during the last 40 years:
•
•
•
•
•
Women entered the labour market
Labour markets has become more international – Immigration
Increase in the educational level
New types of educations
Increasing labour mobility
•
Globalisation and technological change
•
International competition
•
Growing exports and imports
•
Changed production patterns (offshoring outsourcing)
•
Creates an increased focus on the composition of the workforce and
performance, both on innovation and other performance measures
METEOR Seminar (TIID)
21 December 2011
METEOR Seminar (TIID)
Composition study
21 December 2011
Mobility study
The effect of intra- and inter-regional labour mobility on
plant performance in Denmark: the significance of related
labour inflows
Bram Timmermans* and Ron Boschma**
* Aalborg University, Department of Business Studies, DRUID-IKE
** Utrecht University, Department of Economic Geography, Urban and Regional research
centre Utrecht
DRAFT
PLEASE DO NOT CITE WITHOUT PERMISSION OF THE AUTHORS
Abstract
This paper investigates the impact of different types of labour mobility on plant
performance, making use of the IDA-database that provides detailed information on all
individuals and plants for the whole of Denmark. Our study shows that the effect of labour
mobility can only be assessed when one accounts for the type of skills that flow into the
plant, and the degree to which these match the existing skills at the plant level. As
expected, we found that the inflow of skills that are related to skills in the plant impacts
positively on plant performance, while inflows of skills that are similar or unrelated to the
existing skills in the plant have a negative effect on plant performance. In these analyses,
we used a sophisticated indicator of revealed relatedness that measures the degree of skill
relatedness between each pair of sectors on the basis of the intensity of labour flows
between sectors. We also found that intra-regional skilled labour mobility had a negative
effect on plant performance in general, while the effect of inter-regional labour mobility
depended on the type of skills that flow into the plant.
Keywords: Labour Mobility, revealed relatedness, plant performance, geographical proximity
JEL classifications: J24, J61, J62, O18, R12
1. Introduction
Increasing attention is devoted to the meaning and significance of technological relatedness
for innovation and economic growth. With technological relatedness, we mean that economic
entities like firms or industries have a higher scope for interactive learning when there is
some degree but not too much cognitive proximity between firms and industries (Nooteboom,
2000). This basic idea has been used to explain a range of economic phenomena, like the
development of new technology systems (Carlsson and Stankiewicz, 1991), the economic
success of mergers and acquisitions (Ahuja and Katila, 2001) and the performance of
research collaboration networks (Gilsing et al., 2007; Leten et al., 2007). There is also
increasing awareness that relatedness between industries is a crucial factor to explain regional
phenomena, like regional economic growth (Frenken et al., 2007), the spatial clustering of
industries (Boschma and Wenting, 2007), and the process of diversification at the national
(Hidalgo et al., 2007) and regional level (Neffke et al., 2011).
1
METEOR Seminar (TIID)
THE BASIC IDEA
21 December 2011
METEOR Seminar (TIID)
21 December 2011
The Basic Idea-Composition study
•
Our knowledge and actions are affected by our experiences
•
Categorization – related to instincts - we do not like differences and the
unknown
•
Cognitive models - psychology: our knowledge, world-view and the way we
react depends on our experience, education, social class, gender and age
•
Culture (anthropology)
•
Innovation as an interactive process that involves interaction and communication
between various levels of the firm (Lundvall)
•
Differences in the knowledge base in a firm creates opportunities for learning and
new combinations
•
Employee diversity might create a broader search space, make firms more open
towards new ideas and be more creative
•
Intuitive idea that employee diversity is good for innovation
METEOR Seminar (TIID)
21 December 2011
The Basic Idea-Mobility study
• Increasing attention on the significance of (technological)
relatedness for innovation and growth.
• Relatedness is positively associated with interactive
learning of economic entities.
• Diversity helps learning but it should not be too
diverse nor too similar
• Mergers, acquisition and collaboration
• Explanation in regional economic development and
spatial clustering of industries
• Recently this concept is used in labour market studies
METEOR Seminar (TIID)
21 December 2011
Inside or Outside
•
Cohen and Levinthal (1989, 1990, 1994), Joshi and Jackson (2003), and many others
•
Two knowledge dimensions of effects of employee diversity
1.
Within firm
•
•
1.
Broad knowledge base
Cross fertilization and creation of new ideas
Outside links
•
•
•
Broad outside search and networks
Increased external sourcing
Absorptive capacity – use of external knowledge
•
In the composition study we neglect outside links – focus on within firm effects.
•
In the mobility study we link to the outside links and relation to knowledge and skills
from particular geographical regions.
METEOR Seminar (TIID)
THE ISSUE OF DIVERSITY
21 December 2011
METEOR Seminar (TIID)
21 December 2011
What is diversity and how should it be
measured
•
Employee diversity:
•
Everybody is unique, but there are many similarities
•
Measure depends on which factors you look at and the purpose of the
analysis
•
Employees differ along a wide range of dimensions (demography, experiences,
knowledge base, cognitive models, attitudes, values, norms) that are generated
through complex processes
•
We need measurable characteristics and the diversity measures need to be
workable
•
Visible or invisible differences
•
Ascribed and achieved characteristics (Rueff et al. 2003)
•
Ascribed: demographic attributes such as gender, age, and ethnicity
•
Achieved characteristics: educational background, functional background,
work experience
METEOR Seminar (TIID)
21 December 2011
What is diversity and some problems in
measuring diversity
•
Diversity has three dimensions (Stirling, 2007)
•
Variety (number of different types/groups)
•
Balance (shares of the different groups)
•
Disparity (the distance between the different groups)
•
Different characteristics: gender, age, education, ethnicity …
•
but every employee fits into multiple categories
•
We define employee diversity as the distribution of differences among
the employees of the firm with respect to a common attribute
•
Unit-level compositional construct (Harrison and Klein, 2007)
•
Diversity should always been seen in context
•
Diversity is not always good or bad
METEOR Seminar (TIID)
21 December 2011
Diversity and Performance
•
There is a long tradition for studying diversity and performance
•
“It is the heterogeneity of the productive services available or potentially available
from its resources that gives each firm its unique character” (Penrose, 1959).
•
Evolutionary economics
•
Firms with diversity in knowledge, experience and skills among their
employees benefits from complementarities that can foster development in
other fields
•
… they have broader organisational routines (Nelson&Winter)
•
… they have a broader search for new solutions (Dosi)
•
… they are better to gain and exploit external knowledge (Cohen& Levinthal)
•
… they are better to exploit internal knowledge through interaction and
learning (Lundvall)
•
Empirical studies show that technologically diverse firms survive longer and
are more innovative
•
Diversity is positive for innovation, but most studies focus is on technology
not human capital
METEOR Seminar (TIID)
21 December 2011
Diversity and Performance-Top
Management Teams
•
Upper echelon theories (Finkelstein and Hambrick)
•
Education, experience and demographic characteristics affects the
managements interpretation of problems and their strategy, which has an
effect on firm performance.
•
Diversity of a larger management proved to be a predictor for firm
performance.
•
Top Management Team Literature and firm performance (Murray 1989, Bantel and
Jackson 1989, Kilduff et al. 2000).
•
Top management does not reflect composition of the entire firm.
•
Firms knowledge in the form of human capital is important in explaining performance
(Laursen et al. 2005).
•
Diversity in the composition of employees contributes to diversity in the
knowledge base.
METEOR Seminar (TIID)
21 December 2011
Diversity and Performance-Critical
Perspectives
•
Decision making (Priem et al, 1995):
•
To make good decisions when facing complex problems a degree of cognitive conflicts
and different view points are needed to avoid premature consensus.
•
Too much diversity could create conflict.
•
Group think (Baron, 2005)
•
Too much cohesion in groups hinders good evaluation of ideas.
•
Social identity (Joshi and Jackson 2003):
•
Diversity in groups creates competition and conflict which reduce cooperation and
internal communication.
•
Use of information (Dahlin et al, 2005)
•
Diversity increase the breath of information collection and use of information, but too
much diversity hinders diffusion.
•
Communities of practice and learning (Wenger)
•
Interaction between different competencies and experiences creates learning if the
disparity is not too large.
METEOR Seminar (TIID)
21 December 2011
Positive or Negative?
•
William & O’Reilly (1998) 40 years of studies of demography and diversity in
organisations:
•
Diversity has both direct and indirect effects on processes and group
performance
•
These effects can be negative and positive.
•
But there are also many studies that find no effect
•
More recent studies find similar results
•
Positive effects: openness, creativity, learning, flexibility, broader search space, better
problem solving, increased absorptive capacity and new combinations of knowledge
•
Negative effects: distrust, conflict, dissatisfaction and increased transaction costs
(interaction and communication between two different knowledge bases and groups
might be difficult )
•
Group membership, social interactions, power relations, organisational framework
METEOR Seminar (TIID)
21 December 2011
Innovation
What is the relation between diversity
and innovation?
Diversity
METEOR Seminar (TIID)
21 December 2011
The relation between employee diversity
and innovation
•
Performance measures:
•
Productivity
•
Turnover or profits
•
Patents
•
Innovation
•
Innovation is different (human capital is more important)
•
An invention and creativity phase
•
An implementation phase
•
Selection effects
•
Hiring policies
•
Self selection
•
Organisational structure
•
Only a few studies of innovation and they focus on TMT or a single firm and use
innovation proxies, but generally finds some positive effects
METEOR Seminar (TIID)
21 December 2011
Hypotheses
•
•
Hypothesis 1: There is a positive relation between employee diversity and the
likelihood that firms innovate.
•
Hypothesis 1a: There is a positive relation between gender diversity and
the likelihood that firms innovate.
•
Hypothesis 1b: There is a negative or neutral relation between age
diversity and the likelihood that firms innovate.
•
Hypothesis 1c: There is a positive relation between diversity in ethnicity
and the likelihood that firms innovate.
•
Hypothesis 1d: There is a positive relation between educational diversity
and the likelihood that firms innovate.
Hypothesis 2: The likelihood that firms innovate decreases for high levels of
employee diversity.
METEOR Seminar (TIID)
THE ISSUE OF LABOUR
MOBILITY
21 December 2011
METEOR Seminar (TIID)
21 December 2011
Labour Mobility- the positive effects
•
Labour mobility is often regarded as a mechanism that
enhances the competitiveness of firms (and regions)
• Labour mobility is considered an important mechanism for the
dissemination of (tacit) knowledge (e.g. Almeida and Kogut, 1999;
Pinch and Henry, 1999).
•
Overall, the benefits of labour mobility exceed the
negative effects.
•
These studies assume that new employees get integrated
without major frictions.
• Contribute to internal learning processes and the well being of
the firm.
METEOR Seminar (TIID)
21 December 2011
Labour Mobility- the negative effects
•
There are negative impacts of labour
mobility
• Studies have shown that high level of labour
mobilit have a negative impact on firm
performance (Faggian and McCann, 2006; Boschma
2009).
• Challenges with integration of people with a
very different profile
• There is little attention on the type of
knowledge and skills that are transferred as a
result of labour mobility.
METEOR Seminar (TIID)
21 December 2011
Labour Mobility- Absorptive Capacity
• Firms require absorptive capacity to understand external
knowledge.
• Not only the level of absorptive capacity is important but also
the proximity of this external knowledge compared to the
existing knowledge base is important, i.e. not too far nor too
close.
• This has been applied to a Swedish study on labour mobility
(Boschma, 2009).
• This mobility study focuses, in addition to the level of
relatedness, also on the geographical location where the
worker has obtained the knowledge and skills.
METEOR Seminar (TIID)
DATA AND METHOD
21 December 2011
METEOR Seminar (TIID)
Database
•
The Danish Integrated
Database for Labour Market
Research (IDA)
•
detailed information on all
Danish firms and all
individuals on the labour
market from 1980 onwards
•
information on the
characteristics of individuals
•
Information on the
performance and
characteristics of firms and
plants
21 December 2011
Innovation survey
METEOR Seminar (TIID)
21 December 2011
Sample
Composition study
Mobility study
•
•
•
Disko4 innovation questionnaire
survey on organisations, employees
and research and development
strategies in Danish firms
Innovation activity in the period
2003-2005
• sent to stratified sample of 4.136
companies,
• 1.775 answers
• response rate of 42.9 percent
• we use 1.648 observations
•
Plants in the period 1999-2003 that
experienced an inflow of skilled
labour (N=16,709)
Human characteristics
•
•
•
•
•
•
Mean taxable income
25 yrs of age
at least 20 hours a week
registered to change plant
hold a university degree of belong
to the top 20 percent wage
earners.
Plant characteristics
•
•
•
plants active in manufacturing and
services
firms with accounting data
No new plants/start-ups
METEOR Seminar (TIID)
21 December 2011
Dependent Variable
Composition study
• Innovation
• Innovation is defined as
whether the firm has
introduced a new product
or service during the period
2003-2005, excluding minor
improvements on already
existing products and
service.
Mobility study
• Labour productivity growth on
the plant level
•
Value added per worker
• Data not available on the plant
level
•
•
6,402 (38.31 percent)
multiple plant firms
Determining the value added
per plant based on the
distribution of wages
• two year lag to determine
labour productivity growth
METEOR Seminar (TIID)
21 December 2011
Control Variable
Composition study
• Industry: Two digit NACE
industries
• Size
• Firm age
• Organisational change
• Diversity policy
• Dummy variables for high
intensity collaboration along
the value chain
Mobility study
• Industry: Two digit NACE
industries
• Size
• Firm age
• Employment growth
• Growth in fixed assets
• High education ratio
• Region
• Year
METEOR Seminar (TIID)
21 December 2011
Independent Variable
Composition study
Mobility study
• Entropy measures
Gender
Ethnicity
Education
• Standard error
• Age
• Categories
• Gender
• Academics
• Share
• Academics
• Identify the high-skilled inflow
workers
• Identify the industry experience
of the workers in the last 5 yrs.
• Take the most related industry
experience
• Measure the degree of similar,
related and unrelated variety
• Make a distinction between intraand interregional experience
METEOR Seminar (TIID)
21 December 2011
Diversity in more detail
•
Gender diversity
•
Entropy
•
•
•
Gendergroups (1–5) based on the share of the most represented gender: Group 1:
90–100% of the same gender, Group 2: 80–90% of the same gender, Group 3: 70–
80% of the same gender, Group 4: 60–70% of the same gender and Group 5: 50–
60% of the same gender
Ethnicity (EthnDiv) we use the individual’s country of origin.
•
divided into six different groups: Danish, Nordic, EU15 and Swiss, other
Europeans, other western countries, and the rest of the world
Education
•
16 different higher education categories making a distinction between Bachelor’s,
Master’s and Ph.D. degrees
•
social sciences, humanities, food and health science, engineering, and natural
sciences, high-school teachers and officers in the army, navy, and air force
•
dummy variable on the presence of at least one highly educated employee
(HighEducDummy)
•
Share of share of highly educated employees in the firm (HighEducShare).
METEOR Seminar (TIID)
21 December 2011
Relatedness in more detail
•
Similar
•
Most related experience in the same 4 digit NACE industry class
•
inflow (count)
•
Relatedness
•
Revealed Relatedness vs. NACE relatedness
•
NACE relatedness measure makes a distinction based on the 2 and 4
digit industry classes
•
Revealed Relatedness method described in Neffke and Henning (2009)
makes a distinction based on Swedish labour mobility patterns.
•
Unrelated Variety
•
All the inflows that are not similar or related
•
We make a distinction on whether these inflows are from the same labour
market region or not.
METEOR Seminar (TIID)
21 December 2011
Danish Labour Market Regions
METEOR Seminar (TIID)
CLMA vs. Rest of Denmark
21 December 2011
METEOR Seminar (TIID)
21 December 2011
METEOR Seminar (TIID)
21 December 2011
METEOR Seminar (TIID)
RESULTS
21 December 2011
METEOR Seminar (TIID)
21 December 2011
Composition study
METEOR Seminar (TIID)
21 December 2011
METEOR Seminar (TIID)
21 December 2011
Results-Composition Study
•
Hypothesis 1a. There is a positive relation between gender diversity and the likelihood that firms innovate.
•
Positive effect
•
Generally overlooked in the innovation literature
•
Hypothesis 1b. There is a negative or neutral relation between age diversity and the likelihood that firms
innovate.
•
Negative relation
•
Hypothesis 1c. There is a positive relation between diversity in ethnicity and the likelihood that firms
innovate.
•
No significant relation
•
Labour market dominated by Danes
•
Routine work
•
Hypothesis 1d. There is a positive relation between educational diversity and the likelihood that firms
innovate.
•
Positive relation
•
Most important diversity dimension
•
Human capital
•
Professional identity
•
Hypothesis 2. The likelihood that firms innovate decreases for high levels of employee diversity.
•
Not supported
•
Self selection, HRM policies, generally no extreme diversity
METEOR Seminar (TIID)
21 December 2011
Issues
• What about related variety?
• What about tenure (length of service) ,
experience and positions
• How to deal with mobility
• Who is actually involved in the innovation
process
• Self reported data
• Sample bias
• Causality
METEOR Seminar (TIID)
21 December 2011
Mobility study
METEOR Seminar (TIID)
21 December 2011
Mobility study
METEOR Seminar (TIID)
CONCLUSION
21 December 2011
METEOR Seminar (TIID)
21 December 2011
Composition Study
Conclusion
•
The study of 1,648 Danish firms shows that
employee based on the characteristics of all
employees have a effect on their likelihood to
innovate
•
Employee diversity matters for firms’ innovative
performance
•
Employee diversity in terms of gender, age,
education has an effect on the likelihood that firms
innovate with controls for other factors
•
Strongest effect of education
•
Followed by gender
•
Negative effect for diverse age distribution
•
No effect on ethnicity
•
No indication for curvilinear effects
•
Similar findings in other innovation-diversity studies
(education and age).
Future Research
•
Consider other factors that make the human
capital composition of a firm to a success - not
only at the demographic composition
•
Diversity management and management
culture
•
Better measure of teams involved in innovation
•
Persistent innovators and diversity
•
Causality for the link between diversity
and innovation
•
Including various types of work organisation,
innovation modes (DUI/STI) and strategies
•
Types of jobs and mobility
•
Regional differences and regional
characteristics
METEOR Seminar (TIID)
21 December 2011
Results-Mobility Study
Conclusion
Future research
• Related variety has a
positive impact on labour
producitivy growth
• Similarity has a negative
impact on firm performance
• Does not apply to the
Copenhagen area
•
The particular findings of the
Copenhagen region require attention
•
How does the issue of related variety
impact particular sector and on different
stages in the industry life-cycle?
•
How do different regions react on
similarity, related and unrelated variety.
•
Investigate whether regions with high
labour mobility enhance regional growth
in general?
•
Investigate the black box of interaction
within firms between employees with
similar, related and unrelated skills.
•
•
Might be the more
dominant role of service
Statistical artifact since
services are more broadly
defined.
• Lots of variety within the
same industry
Slut
bram@business.aau.dk