The document explores the role of public procurement of innovation (PPI) in driving innovation in firms, using data from the German innovation surveys. It identifies various external sources of innovation, analyzes their significance, and assesses the impact of PPI on firms' innovation performance. Furthermore, it discusses the strengths and limitations of the methodology used to gather innovation-related data.

1. 1
Using Business Innovation Surveys to Learn about Procurement and PPI - some Experience from Germany Christian Rammer Centre for European Economic Research (ZEW), Mannheim OECD Expert Workshop Measurement of Public Procurement of Innovation Paris, February 4th, 2013

2. 2
Content
(1)„Sources of Innovation“ as a way to identify the role of customers/demand (incl. customers/demand from the public sector) for innovation
(2)Significance of PPI: results from the German Innovation Survey
(3)Impact of PPI on firms„ innovation performance
(4)Strengths and limitations of this approach

3. 3
Sources of Innovation
- Mansfield (1991): innovations that could not have been developed in the absence of recent academic research
- Extending to other potential external sources that were indispensible for developing and introducing product or process innovation: - customers/demand - suppliers - competitors - regulation - public science
- Firms had to assess the significance of each source for their product and process innovations separately
- For each innovation source (except regulation), data on sector and location of sources was collected (free text)
 2-page question, included in 1999 and 2003 surveys

4. 4

5. 5
Public Procurement Innovation
- PPI: purchase of innovative product the development of which has been explicitly demanded by a public authority
- Simple purchase of innovative products by public authorities is not PPI
- Public authorities also include public services

6. 6
- Sectors were assigned to NACE (rev. 1.1) 3-digit
-For each innovation source, sales with new products (for product innovation) and cost savings (for process innovation) triggered by that source were calculated using weights: IOk = Si (Si aik wi) IO innovation output S total sales / total costs a sales share of new products / cost savings share triggered by source k w weight of firm i in total firm population - A sector‘s contribution to innovation output was calculated by weighting IOk by the (approximate) share of sector j in all innovation impulses received from source k - Public administration and public services identified through corresponding NACE (rev. 1.1) sectors (75, 80, 85, 40, 41, 60, 70, 73, 90)
Data Analysis

7. 7
Significance of Innovation Sources
(1998 / 2002 average)
0
10
20
30
40
50
60
70
80
Customers/demand
Suppliers
Competitors Public science
Regulation
Others*
0
5
10
15
20
25
Product innovation
Process innovation
* include in-house R&D and other creative work and other external sources
share in total sales with new products
1996-1998 and 2000-2002 (%)
share in total cost savings due to process
innovation 1996-1998 and 2000-2002 (%)
16
4 4
1
3
71
20
22
17
11
15
15
Source: Mannheim Innovation Panel (German CIS), surveys 1999 and 2003

8. 8
Significance of Public Procurement
as Innovation Source (1998 / 2002)
Source: Mannheim Innovation Panel (German CIS), surveys 1999 and 2003
NACE rev 1.1
product innovation
process innovation
product innovation
process innovation
75/80
2.8
0.5
0.5
0.1
85
1.2
0.1
0.2
0.0
73*
0.2
0.7
0.0
0.1
90*
0.1
0.0
0.0
0.0
60*
2.4
0.3
0.4
0.1
70*
0.4
0.2
0.1
0.0
40/41*
0.4
0.4
0.1
0.1
Total
7.5
2.2
1.2
0.4
Regulation
3.5
15.4
Public science
1.5
10.8
* share of innovation impulses from public organisations estimated
Share in all innovation impulses by customers (%) Share in total innovation output (%)
Public customers from

9. 9
Innovativeness of Demand I (1998 / 2002)
Source: Mannheim Innovation Panel (German CIS), surveys 1999 and 2003 – Federal Statistical Office of Germany: Input-Output-Tables
0
5
10
15
20
25
30
35
34
95
60_63 65_67
29
50/52
22
75/80
15_16
31
24*
70_71
32
45
01_05
51
64
74
85 33
35
36_37
24.4
72
28
17_19
25
40_41
20_21
30
26
27
55
73
23 10_14
90_93
innovation impulses by customers
domestic demand
share in total innovation impulses by customers/demand / share in total domestic intermediate consumption/final use/capital expenditure (%)
Customer sector
NACE rev 1.1
* 24 excl.
24.4
Share in total

10. 10
Innovativeness of Demand II (1998 / 2002)
Source: Mannheim Innovation Panel (German CIS), surveys 1999 and 2003 – Federal Statistical Office of Germany: Input-Output-Tables
* 24 excl.
24.4
share of sales of innovative products
based on innovation impulses
from customers
in total domestic intermediate consumption, final use and capital expenditure
of the customer sector
(export share of innovators estimated)
share of demanded innovation in total domestic demand (%)
0
5 10
15
20
25
30
35
40
45
34
24.4
65_67
50, 52
36_37
22
60_63
32 29
35
72
30
01_05
17_19
31
33
23
73
64
15_16
Total
51
25
10_14 85
26
24*
28
45
40_41
20_21
55
74
95
70
27
75_80
90
Customer sector
NACE rev 1.1

11. 11 Impact of PP on Innovation Performance
Aschhoff & Sofka (“Innovation on Demand – Can Public Procurement Drive Market Success of Innovations?” Research Policy 38, 1235-1247, 2009):
- PPI contributes to higher sales with new products (which is true for customer/demand impulses in general)
- no such effect found for defence demand, however
- SMEs, services and firms from eastern Germany profit most from PPI
Beise & Rammer (“Local User-Producer Interaction in Innovation and Export Performance of Firms”, Small Business Economics 27(2-3), 207- 222, 2006):
- PPI tend to limit export activities of firms, particularly for service firms that use PPI from the health sector
- „ideosyncratic‟ demand evident for private households

12. 12
Strengths and Limitations of the Approach
-Collecting data on various external sources for innovation helps to provide a balanced picture of the drivers of innovation
-Information on sectors and location of innovation sources offers a great potential for analysis
-Focussing on only one potential source (e.g. PP) risks to overestimate the role of PP for industrial innovation
-Collecting full information on innovation sources requires substantial questionnaire space and puts high burden on respondents (and on statistical offices for coding texts) - Limited opportunity to add additional questions on PP & innovation

13. 13
Thank you for
your attention!
rammer@zew.de