Extra - DIME workshop September 29-30, 2006 1 Laurent Bach, Nicolas Carayol, Patrick Llerena BETA- L. Pasteur University of Strasbourg and CNRS (UMR 7522)

1Extra - DIME workshop September 29-30, 2006

Laurent Bach, Nicolas Carayol, Patrick LlerenaBETA- L. Pasteur University of Strasbourg and CNRS (UMR 7522)*[email protected]

Assessing the performance of French University

transfert offices: preliminary results


• Presentation of the survey content

• First main results and remarks:- on research contracts and patents- Performance indicators and their interpretation

• Presentation of ongoing work: SFE estimation

OUTLINES


STUDY ON UNIVERSITY TT : Presentation

• Performed : 2005-2006 BETA Univ. L. Pasteur + French TTOs network CURIE

• On the behalf of university presidents (CPU) and the French Ministry of Education, Higher Education and Research (MENESR)

• survey by e-questionnaire filled out by universities' TTOs - General organisation and functionning of TTOs

(status, expend. and resources, policy, forms of contracts, personal, etc)

- Measurement of activities of TTOs

• situation as end 2004 + time-based variables 2000-2004first of a regular evaluation

sent to all universities (88, incl. INP) + 11 members of CURIE; base : 2003 Ministry data

The study on TT activities of universities


STUDY ON UNIVERSITY TT : Main results

Sample and representativity :74 answers (66 univ. + 3 INP + 5 others)Rate of return : 74.7 %

78% of the universities + INPbetween 71% and 100% according to the type of universitiesbetween 8 and 10 out of the 10 Top universities (18 out of Top 20) according to Ministry data on research-oriented resources, number of teacher-researchers, industrial contracts, IPR revenuessample mean ≈ poulation mean for those criteria

Not 100% answers on 100% questions => following results on sub-sample of respondants (not detailed here)

First results of the study



Creation of TT activities : more than 50% of univ. since 1999 (law on innovation), some very old ones

Size of TTOs : 263 Full-Time Equivalent; 3.9 FTE per university; 1.2 TTOs staff for 100 teacher-researchers (exl. PROS staff of university labs)Total 2004 expenditures 9.049 million €; 148 K€ per university; 2% of 2003 research-oriented resources

Organisation :in 86% of the case TTOs = department or internal service from the university (1/5 : the 1999 born SAIC)31 universities have at least two TTOs (2nd = very often external entity)54% of the TTOs employees are civil servantsHead : engineer (57% of univ), half have experience in private business.



Research contracts and services :2000-2004 : 766.8 M€ Average per university per year : 2.6 M€ (research contracts = 10 x service contracts) / 73 contracts

Results on "outputs"

0

50 000 000

100 000 000

150 000 000

200 000 000

250 000 000

2000 2001 2002 2003 2004





Patent (applying – French priority/EPO) :2000-2004 : 1104 patents = 3.3 patent per university per year(≈ 20% of univ.: no patent)

Patent (ownership – French + extension)End 2004 : 464 patents = 9.9 patents per university58% fully owned - 240 are extensions (European patent or national ones)

0

50

100

150

200

250

300

2000 2001 2002 2003 2004


0

10

20

30

40

50

60

70

80

90

100

min : 1

max : 95

3 own more than 80 patents

50% own less than 10 patentsmean : 18.6

asymetric distribution of performance




Licensing 2000-2004 : 351 licenses awarded (1.1 license per year per university50% on patent and related know-how30% awarded to less than 3 years old university spin-off

Revenues from license 2000-2004 : 22.1 M€ ((90% on patent & related know-how) )average : 0.2 M€ per university per year

Other results on creation of firms, spin-off shareholding, "entrepreneurship" of civil servants, etc

End 2004 : 243 "active" licenses (5.2 per university)126 licenses are generating money (2.7 per univiversity)




Performance indicators : some results and remarks

per year on 2000-2004 ("flow" data) :•number of patent claims•number of copyright on software•number of licenses awarded per year•income from licenses per year (€)•amount of industrial contracts (€)2004 ("stock data")•number of patents owned•number of current licenses•number of licenses generating incomes

TT performance indicators from the study :


•patent indicators : what about patent claim property regime ? - 1


Performance as effectiveness : coverage of indicators

2000-2004 : on 1091 patent claims, 79% (859 patents) full or co-ownership

ownership waived





0

50

100

150

200

250

2000 2001 2002 2003 2004

ownership co-ownership ownership waived Total





0

20

40

60

80

100

120

140

160

2000 2001 2002 2003 2004

co-ownership with PROs/university co-ownership with firmsownership waived to PROs/university ownership waived to firmsclaim related to firms


•Contracting activities indicators : what about distribution according to the different sources ? - 1



0%

5%

10%

15%

20%

25%

30%

35%

40%

2000 2001 2002 2003 2004

Ministry PROs Local authoritiesForeign (EC…) Associations Firms


•Contracting activities indicators : what about distribution according to the different sources ? - 2



Performance indicators : industrial contracts compared to research-oriented

resources : decreasing share ? distribution of different resources ?

Crowding out or in effects ? In terms of funding ? Of activities ?



• Towards a first measurement of a "TT chain-link "PATENT CLAIM

Patent claim before 2000

patent claim 2000-04

1104 (220,8/year)

no decision

patents owned 2000-04

patents owned 31/12/04

464

OWNERSHIP

licenses awarded 2000-04

185 (37/year)

current licenses

31/12/04158

LICENSES AWARDED

licenses generating revenues 31/12/04

66

LICENSES GENERATING REVENUES

revenues from licenses 2000-04

REVENUES

22 055 634 €(4 411 426€/year)

5 131 224 €

2004

(a)

(b)

(A)

(B)

(B)

(B)



"average university" is far behind "frontier university"

Top universities perform differently on different activities

0%

10%

20%

patents owned95

current licenses43

licenses withrevenues

18

software /year6

patent claims/year19.4

licensesawarded/year

9.8

revenues fromlicenses/year

3.137 M€

industrialcontracts/year

4.246 M€


• Two articles on TTO performance: on UK (Chapple, Locket, Siegel and Wright (R.P, 2005) and on US data (Siegel, Waldman and Link (R.P. 2003)).

• We tried to start with, an estimation using very similar technics, the SFE

• We add some new variables (more appropriate to the french case)

SFE (stochastic frontier estimation): first trials



In their work, Siegel, Waldman and Link (2003) used the following model:εδδδδδδ ββββββ ++++++= −++++++= PUBLICINDOUTINDRDAGEMEDSCHu uvUBMTLEGALSTAFFINVDISCY 654210 543210 lnlnlnlnlnln SThe data we gathered allows us to build a similar model, expressed as follow:εδδδδδδδ βββββ +++++++= −+++++= RDPUBRDINTGDPAGEYAGEXMEDSCHu uvLEGALSTAFFSUBMTY 42413222110 3212110 lnlnlnlnln

Where Y is either NOLIC or LICINC.

For a better understanding, let us name this second comparison: comparison number 2.

*** In their work, Chapple, Locket, Siegel and Wright (2005) used the following model:εδδδδδ ββββ +++++= −++++= DRGDPAGEMEDSCHu uvLEGALSTAFFTRESINCY &lnlnlnln 43210 3210The data we gathered allows us to build a similar model, expressed as follow:εδδδδδδδ βββββ +++++++= −+++++= RDINTRDINTGDPAGEYAGEXMEDSCHu uvLEGALSTAFFSUBMTY 42413222110 3212110 lnlnlnlnln

Where Y is either NOLIC or LICINC.

For a better understanding, let us name this first comparison: comparison number 1.


• Number of licences: very similar results1. on production:(+) staff TTO; research income (UK-U)/subv (F)Legal exp.: (-) US/UK; (+) F : institutional particularities2. on inefficiency: (-) MedicalSchool; (-) LocPub spending, (+) internal R&D

• Lic. Income: poor results (in both cases)(+) Staff TTO more important in F.

• Number of patents + additional variables:1. on production:(+) staff TTO; legal exp. and dummy for Sc Uni. and INP2. on inefficiency: (-) LocPub spending, (+) internal R&D

• Lic. Income + additional variables: poor resultson prod: (+) Staff TTO and legal exp OR dummyon inefficiency: emergence of Agex ie the age of TT activities



THANK YOU FOR YOUR ATTENTION






LOLF ‘performance’ indicators : Are the incentives and governance structure right ?

What about the effectiveness of both ? In particular in terms of technology transfer to firms ?


The LOLF

LOLF : Loi Organique sur la Loi de FinancePublic Budgeting procedures / Parliementary levelAdopted 2001 - first implemented 2005 for 2006 budgetObjectives : control of public money /actions

multi-year strategic planning more flexibility of funds allocation

•Gvtal action : very broad missions->programmes->actions•Performance targets and related evaluation system : objectives / performance indicators•Budget year N+1 voted by Parliament depending on evaluation of success in year N (+ 5 years horizon)



• Towards a first measurement of a "TT chain-link "

2000-2004, one license => 119 K€ as revenue

As for 2004one patent owned => 0.34 license 42% of the licenses are generating revenues, ie

one patent => 0.14 license generating revenues revenue : 78 K€ per license = 11 K€ per patent

owned

Extra - DIME workshop September 29-30, 2006 1 Laurent Bach, Nicolas Carayol, Patrick Llerena BETA- L. Pasteur University of Strasbourg and CNRS (UMR 7522)

Documents

university tt

extra dime workshop

study slide

main results patent

main results sample

tt activities of universities

following results

preliminary results