A discrete choice experiment to explain
knowledge acquisition strategies of SMEs
Frank van Rijnsoever
[email protected]
Sander Kempkes
Copernicus Institute of Sustainable Development
Introduction
•
Internal or external R&D?
Aim:
“to compare the influence of attributes
of an innovation project on the choice
for a knowledge acquisition strategy for
different latent classes of firms”
Copernicus Institute of Sustainable Development
How does this work?
Copernicus Institute of Sustainable Development
Knowledge acquisition strategies
Internal R&D
Buy
Collaborate
Copernicus Institute of Sustainable Development
Characteristic
Explanation
Attributes
Level
Speed to market Are you ahead of your competitors? 1. First mover
2. Second mover
Risk
The likeliness the project will fail,
1. Low risk more than 50% chance succes
does not yield the expected results 2. High risk less than 50% chance succes
in the end.
Development
time
The timescale of the project.
Knowledge
source
Where do you get the knowledge to 1. None
develop your idea to a proof of
2. Supplier
concept from?
3. Competing firm
4. Non-competing firm
5. Buyer
6. University or public research institute
7. Private research institute
1. Short time: two times as fast
2. Long time: normal time
Your intellectual How is your IP ownership arranged? 1. Free IP
property (IP)
2. Shared IP
3. License IP
4. Full ownership
Copernicus Institute of Sustainable Development
Characteristic
Explanation
Level
Access to a
network
Access to new contacts you gain by
choosing a specific strategy to other
partners.
1.
2.
3.
4.
Access to
Free access to an office and/or
1.
Office/Specialized specialized equipment (e.g. laboratory, 2.
equipment
field test facilities)
3.
4.
No
Scientific
Business
Scientific & Business
No
Office
Specialized equipment
Office & Specialized equipment
Access to
Training/Coach
Free access to a coach and/or training
such as master classes and workshops.
Coaching includes advisory boards,
mentors and other coaches.
1.
2.
3.
4.
No
Coach
Training
Coach & Training
Funding model
The way the project is financed.
1.
2.
3.
4.
5.
6.
Subsidy from the government
Loan from a bank
Crowd funding
Investor
Family & Friends
Own investment
Copernicus Institute of Sustainable Development
Example choice task
Copernicus Institute of Sustainable Development
Discrete choice experiment
Respondents receive 9 choice tasks

Experimental design
Data collection

427 entrepreneurs that founded or owned in
manufacturing SMEs in the UK (284) or Germany
(143)

Online panel, 20 minute questionnaire
Copernicus Institute of Sustainable Development
Attribute
Knowledge acquisition
strategy
Wald χ2
53,66
Sig.
76,85
***
Risk
133,85
***
Development time
32,38
***
Knowledge source
96,24
***
Network
Office & specialized
equipment
Training & coach
Funding model
86,06
48,37
49,10
49,91
207,07
β class 1
Sig.
β class 2
Sig.
β class 3
**
**
16,93
-17,61
0,68
***
***
Collaborate
-8,08
3,95
4,12
-1,61
-8,16
9,77
First mover
Second mover
Small
Large
Short
Long
Buyer
Competing firm
Private research institute
Non-competing firm
Supplier
University or public research institute
No IP
Full ownership
License IP
Shared IP
None
Business
Scientific
Scientific & business
None
Office
Specialized equipment
Office & specialized equipment
None
Training
Coach
Training & coach
Own investment
Family & friends
-0,33
0,33
0,82
-0,82
0,00
0,00
0,16
-0,80
-0,23
0,77
0,17
-0,08
-0,16
0,01
0,15
0,00
-0,14
0,38
-0,23
-0,01
0,00
0,34
0,02
-0,36
0,33
-0,28
0,12
-0,17
0,71
-0,84
***
***
***
***
Investor
Loan (bank)
Crowd funding
Subsidy (government)
0,44
**
0,37
-0,77
***
Copernicus
0,09
*** Internal R&D
Buy
Speed to market
IP
Level
***
***
***
***
***
**
***
***
**
**
**
**
*
***
***
0,03
-0,03
0,13
-0,13
-0,12
0,12
-0,26
0,08
0,22
0,44
0,11
-0,59
-0,20
0,96
-0,38
-0,38
-0,10
0,19
0,10
-0,19
0,22
0,12
-0,30
-0,04
-0,25
0,16
0,32
-0,22
3,02
0,09
-0,47
-1,16
-0,97
Institute
-0,51
**
**
***
**
**
*
*
*
***
2,23
-2,23
2,24
-2,24
0,85
-0,85
0,28
-0,93
0,62
1,30
-0,43
-0,84
0,35
1,53
-1,54
-0,35
0,96
-1,82
-0,31
1,18
-0,17
-0,40
0,93
-0,36
-0,97
1,43
-0,26
-0,21
0,66
0,90
Sig.
***
***
***
***
***
***
***
***
**
*
***
*
***
***
**
***
***
***
**
***
*
*
β class 4 Sig.
0,75
-0,15
-0,60
0,23
-0,23
-0,05
0,05
0,51
-0,51
0,15
0,04
-0,46
-0,27
0,33
0,22
0,18
-0,33
0,25
-0,10
-0,36
0,39
-0,05
0,01
-0,50
0,07
-0,24
0,66
-0,20
-0,19
0,47
-0,09
-0,27
-0,06
***
***
**
**
*
*
*
**
***
**
**
0,43
-0,19
***
-2,06
***
-0,42
**
***
-2,51
***
0,72
***
of Sustainable Development
**
2,59
***
0,22
Results: latent class analysis
Class 1 (24%)
Class 2 (28%)
Class 3 (22%)
Class 4 (26%)
KAS
Collaborate,
Buy
Internal R&D
Collaboration
Internal R&D
Change in
KAS?
From
collaborate to
buy
No
No
Yes!
Market
speed
Second mover
First mover
First mover
First mover
Copernicus Institute of Sustainable Development
Firm characteristic
Wald χ2
Sig.
Intercept
R&D
Patent stock
12,12
16,61
20,07
**
***
*
Product innovations
20,31
a
Past use of KAS: suppliers
16,41
*
Past use of KAS: competing firm
14,50
*
Past use of KAS: non-competing firm
14,21
*
Past use of KAS: customers
11,27
Past use of KAS: Universities & Public
knowledge institutes
8,71
Past use of KAS: Private knowledge
institutes
6,88
Size
McFadden R2
24,13
0,15
a
**
Category
β class 1
Sig.
β class 2
Sig.
β class 3
Sig.
β class 4 Sig.
External R&D
1,48
**
-0,80
***
0,21
0,16
0,52
*
-0,90
*
-0,11
-0,06
-0,35
0,37
0,15
-0,17
0,41
**
-0,25
0,68
*
-0,23
Internal R&D
-0,55
0,40
*
-0,23
0,32
-0,53
a
0,44
0,12
-0,31
a
0,36
a
-0,60
*
0,42
a
0,14
-0,01
-0,13
0,22
-0,41
*
Collaborate
-1,28
*
0,07
0,51
*
-0,42
-0,04
-0,05
0,10
0,02
-0,27
0,50
*
-0,35
-0,05
0,09
-0,04
-0,92
0,48
Flexible
0,35
0,33
a
-0,49
**
-0,06
0,04
0,51
-0,11
0,35
*
0,27
-0,27
-0,23
0,08
-0,49 ***
0,42
*
0,01
0,15
Not used
Buy
-0,46
0,34
0,18
0,43
0,44
-0,55
-0,16
-0,23
Collaborate
Not used
Buy
Collaborate
-0,26
-0,08
0,05
-0,09
-0,48
0,04
0,01
0,14
Not used
Buy
Collaborate
Not used
Buy
0,03
0,17
-0,01
-0,16
0,23
-0,14
-0,47
-0,09
0,56
-0,11
Collaborate
Not used
Micro
Small
Medium
Medium-large
-0,28
0,05
-0,27
0,47
0,62
-0,82
0,31
-0,20
0,31
-0,08
-0,04
-0,20
Yes
0
1-2
3-10
>10
0
1
2
3
>3
Buy
Collaborate
Not used
Buy
Collaborate
*
a
*
**
0,50
0,04
-0,61
0,43
a
a
0,23
0,00
0,55
-0,48
0,17
0,17
0,10
-0,27
0,05
-0,07
0,12
0,00
-0,13
-0,17
-0,23
0,18
-0,12
0,04
-0,91
0,99
0,20
-0,03
0,08
-0,43
0,33
0,03
**
a
*
**
a
Copernicus Institute of Sustainable Development
Results: latent class analysis
Class 1 (24%)
Class 2 (28%)
Class 3 (22%)
Class 4 (26%)
KAS
Collaborate, Buy
Internal R&D
Collaboration
Internal R&D
Change in KAS?
From collaborate
to buy
No
No
Yes!
Market speed
Second mover
First mover
First mover
First Mover
R&D
No
Yes
No
Yes
Patents
Some
Yes
Some
Yes
Innovative
Not that much
Yes
Yes
A bit
Interesting for
collaboration?
Mwa…
Yes, but
unlikely
Yes, and doing so
Yes, and you can
get them to do
so!
Copernicus Institute of Sustainable Development
Conclusions
•
Label, risk and funding are important
• Knowledge institutes are unpopular partners
•
Heterogeneity matters!
•
•
•
•
•
Four latent classes
Difference in preferences and likelihood to change
Preferences are grounded in past behavior (routines?)
22% already collaborates
26% percent can be ‘seduced’ into collaboration
Copernicus Institute of Sustainable Development
Practical implications
•
Resource endowments like subsidies, mentorship,
coaches, laboratories, office, etc. have a limited
effect.

•
But can still be useful!
Since firm history is matters, it is important to
create firms that are externally oriented from the
start.

Focus on collaboration in entrepreneurship education and
acceleration programs
Copernicus Institute of Sustainable Development
Questions
[email protected]
Copernicus Institute of Sustainable Development