Evaluation of Large Scale RDI...
Transcript of Evaluation of Large Scale RDI...
![Page 1: Evaluation of Large Scale RDI Infrastructuresinstitute.eib.org/wp-content/uploads/2016/05/Florio-EIBpresentation... · Research Sponsorship Programme Proposal by University of Milan](https://reader034.fdocuments.us/reader034/viewer/2022042411/5f2a725b491ab42349178c17/html5/thumbnails/1.jpg)
Evaluation of Large Scale RDI
Infrastructures
Massimo Florio
Department of Economics, Management and Quantitative Methods
University of Milan
EIB INSTITUTE’S BROWN BAG SEMINARS
June 22, 2016
![Page 2: Evaluation of Large Scale RDI Infrastructuresinstitute.eib.org/wp-content/uploads/2016/05/Florio-EIBpresentation... · Research Sponsorship Programme Proposal by University of Milan](https://reader034.fdocuments.us/reader034/viewer/2022042411/5f2a725b491ab42349178c17/html5/thumbnails/2.jpg)
Background
June 2012
EIB Institute
Call for proposals
2012/C 162/10
EIBURS - EIB University
Research Sponsorship
Programme
Proposal by
University of Milan and CSIL
Cost-Benefit Analysis in
the Research, Development
and Innovation Sector
December 2012
December 2015
OBJECTIVE OF THE SEMINAR:
DISCUSSING THE
FINAL RESULTS OF THIS PROJECT
“(…)The current methodology within the Bank is
conservative and aligned to DG Regio’s ‘Guide to
cost/benefit analysis of investment projects’. However,
this document is itself deficient in its explanation of the
RDI sector.
(…) The aim of the project is to build on the current
internal review and produce a model which is
academically credible, robust but at the same time easy
to apply and easy to assess.”
![Page 3: Evaluation of Large Scale RDI Infrastructuresinstitute.eib.org/wp-content/uploads/2016/05/Florio-EIBpresentation... · Research Sponsorship Programme Proposal by University of Milan](https://reader034.fdocuments.us/reader034/viewer/2022042411/5f2a725b491ab42349178c17/html5/thumbnails/3.jpg)
![Page 4: Evaluation of Large Scale RDI Infrastructuresinstitute.eib.org/wp-content/uploads/2016/05/Florio-EIBpresentation... · Research Sponsorship Programme Proposal by University of Milan](https://reader034.fdocuments.us/reader034/viewer/2022042411/5f2a725b491ab42349178c17/html5/thumbnails/4.jpg)
History of CBA: France and USA
Jules Dupuit (1804 – 1866)
“De la mesure de l'utilité des travaux publics”
(Annales des Ponts et Chaussées, 1844)
![Page 5: Evaluation of Large Scale RDI Infrastructuresinstitute.eib.org/wp-content/uploads/2016/05/Florio-EIBpresentation... · Research Sponsorship Programme Proposal by University of Milan](https://reader034.fdocuments.us/reader034/viewer/2022042411/5f2a725b491ab42349178c17/html5/thumbnails/5.jpg)
History of CBA: France and USA
France, 1850's
• Ecole des Ponts et Chausseès
• Dupuit, Arsène Jules Étienne Juvénal, De la mesure de
l’utilité des travaux publics, Annales des Ponts et
Chaussées (1844)
USA, 1930's:
• Flood Control Act (1936)
USA, 1950's: attempts to codify the benefit-cost rules
• U.S. Army Corps of Engineers, Department of Agriculture,
Bureau of Reclamation, Federal Power Commission
• Proposed practices for economic analysis of river basin
projects - Green book (1950)
![Page 6: Evaluation of Large Scale RDI Infrastructuresinstitute.eib.org/wp-content/uploads/2016/05/Florio-EIBpresentation... · Research Sponsorship Programme Proposal by University of Milan](https://reader034.fdocuments.us/reader034/viewer/2022042411/5f2a725b491ab42349178c17/html5/thumbnails/6.jpg)
History of CBA:
international organisations and the UK
• Guidelines for project evaluation, UNIDO (1972)
• Project appraisal and planning for developing
countries by I.M.D. Little and J.A. Mirrlees, OECD
(1974)
• Economic analysis of projects by L. Squire and H.G.
van der Tak, World Bank (1975)
• UK, The Green Book – Appraisal and Evaluation in
Central Government (2003)
• OECD, Cost-Benefit Analysis and the Environment,
Pearce et al. (2006)
• EIB, The Economic Appraisal of Investment Projects
at the EIB (2013)
![Page 7: Evaluation of Large Scale RDI Infrastructuresinstitute.eib.org/wp-content/uploads/2016/05/Florio-EIBpresentation... · Research Sponsorship Programme Proposal by University of Milan](https://reader034.fdocuments.us/reader034/viewer/2022042411/5f2a725b491ab42349178c17/html5/thumbnails/7.jpg)
History of CBA: European Commission
1994: 1st edition, 28 pages
1997: 2nd edition, 84 pages
2008: 4th edition, 257 pages
2014: 5th edition, 364 pages
2002: 4th edition, 133 pages
A novelty of the fifth edition of
EC CBA Guide:
Ch. 7 - Research, development
and Innovation
![Page 8: Evaluation of Large Scale RDI Infrastructuresinstitute.eib.org/wp-content/uploads/2016/05/Florio-EIBpresentation... · Research Sponsorship Programme Proposal by University of Milan](https://reader034.fdocuments.us/reader034/viewer/2022042411/5f2a725b491ab42349178c17/html5/thumbnails/8.jpg)
8
Some information on CBA international practice are drawn from the results of a
survey conducted on selected OECD countries addressing the actual use, practice
and role of CBA in ex-ante project appraisal.
OECD, Government at glance
July 2015
http://www.oecd.org/gov/govataglance.htm
Rail (e.g. Austria, Denmark, Canada, Sweden,Netherlands).
Urban transport (e.g. New Zealand, Austria,Denmark, Canada, Sweden, Netherlands)
Airports, ports and waterways (e.g. Austria,Canada, Sweden, Netherlands, UK)
Water supply and wastewater (e.g. Canada, Netherlands)
Solid waste management (e.g. Canada, UK)
Other environmental projects: risk prevention and mitigation, natural asset conservation, etc. (e.g. Canada, Sweden, UK)
Energy: production, transmission anddistribution (e.g. Denmark, Canada, Sweden)
Education (e.g. Canada, UK)
Culture and leisure (e.g. NewZealand, Canada, UK)
ICT: telecommunications, broadband, ICTapplications to businesses and citizens(e.g. Canada, UK)
Health (e.g. Canada, Sweden)
Scientific research (e.g. Canada, UK)
Technological development andinnovation: science parks, technologicalparks, incubators, etc. (e.g. Canada, UK)
CBA for research infrastructures
![Page 9: Evaluation of Large Scale RDI Infrastructuresinstitute.eib.org/wp-content/uploads/2016/05/Florio-EIBpresentation... · Research Sponsorship Programme Proposal by University of Milan](https://reader034.fdocuments.us/reader034/viewer/2022042411/5f2a725b491ab42349178c17/html5/thumbnails/9.jpg)
The use of CBA in the practice: evidence from the OECD
Canada: All sectors,
including RDI
UK: All sectors,
including RDI
Sweden: Rail, Urban Transport, Airports, ports,
waterways, Energy, Health, Other environmental projects
The Netherlands: Rail, Urban Transport, Airports, ports,
waterways, Water supply
Denmark: Rail, Urban Transport, Energy
New Zealand: Urban Transport, Culture and
Leisure
Austria: Rail, Urban Transport, Airports,
ports, waterways
![Page 10: Evaluation of Large Scale RDI Infrastructuresinstitute.eib.org/wp-content/uploads/2016/05/Florio-EIBpresentation... · Research Sponsorship Programme Proposal by University of Milan](https://reader034.fdocuments.us/reader034/viewer/2022042411/5f2a725b491ab42349178c17/html5/thumbnails/10.jpg)
Key ingredients:
Long-run timeframe
Incremental approach
Shadow prices Risk assessement
![Page 11: Evaluation of Large Scale RDI Infrastructuresinstitute.eib.org/wp-content/uploads/2016/05/Florio-EIBpresentation... · Research Sponsorship Programme Proposal by University of Milan](https://reader034.fdocuments.us/reader034/viewer/2022042411/5f2a725b491ab42349178c17/html5/thumbnails/11.jpg)
Long-run timeframe Public infrastructure projects need long-term forecasts. For example,
o Railway: 30 years
o Energy: 15-25 years
o Broadband: 15-20 years
This makes long-term forecasts and intertemporal analysis crucial.
RDI infrastructure are not different.
Typical time horizon is often 25 years or
more.
Source: EC CBA Guide, 2014
Key ingredients
![Page 12: Evaluation of Large Scale RDI Infrastructuresinstitute.eib.org/wp-content/uploads/2016/05/Florio-EIBpresentation... · Research Sponsorship Programme Proposal by University of Milan](https://reader034.fdocuments.us/reader034/viewer/2022042411/5f2a725b491ab42349178c17/html5/thumbnails/12.jpg)
Key ingredients
Incremental approach
![Page 13: Evaluation of Large Scale RDI Infrastructuresinstitute.eib.org/wp-content/uploads/2016/05/Florio-EIBpresentation... · Research Sponsorship Programme Proposal by University of Milan](https://reader034.fdocuments.us/reader034/viewer/2022042411/5f2a725b491ab42349178c17/html5/thumbnails/13.jpg)
Markets do not price basic research: a public good not generating
profits.
applied research: imperfect markets because of asymmetric
information, externalities and risks.
applied research: imperfect markets because of asymmetric
information, externalities and risks.
Shadow prices are defined as the marginal social value of a good.
empirical proxies: marginal costs or willingness to pay, or a
combination of the two concepts.
Shadow prices
Key ingredients
![Page 14: Evaluation of Large Scale RDI Infrastructuresinstitute.eib.org/wp-content/uploads/2016/05/Florio-EIBpresentation... · Research Sponsorship Programme Proposal by University of Milan](https://reader034.fdocuments.us/reader034/viewer/2022042411/5f2a725b491ab42349178c17/html5/thumbnails/14.jpg)
Risk assessement
Key ingredients
Results of a Monte Carlo simulation
Examples of probability distribution and cumulated distribution of the ENPV
Values ENPV ERR
Base case 36.6 7.56%
Mean (expected value) 41.3 7.70%
Median 37.7 7.64%
Standard deviation 28.6 1.41%
Minimum value -25.9 3.65%
Central value 55.2 7.66%
Maximum value 136.3 11.66%
Probability of the ENPV being lower than zero or ERR being
lower than the reference discount rate
0.053 0.053
Probability distribution Cumulated distribution
![Page 15: Evaluation of Large Scale RDI Infrastructuresinstitute.eib.org/wp-content/uploads/2016/05/Florio-EIBpresentation... · Research Sponsorship Programme Proposal by University of Milan](https://reader034.fdocuments.us/reader034/viewer/2022042411/5f2a725b491ab42349178c17/html5/thumbnails/15.jpg)
The CBA model (1)
The expected economic net present value of the RDI infrastructure
[𝔼(𝐸𝑁𝑃𝑉𝑅𝐷𝐼 )] over the time horizon (T) is defined as the
difference between expected benefits and costs valued at shadow
prices and discounted at the social discount rate (r).
The model breaks down intertemporal benefits into two broad
classes – use and non-use benefits – and compares these benefits
with costs.
The expectation operator implies that all critical variables are
considered as stochastic.
All the benefits should be related to the main economic agents:
firms, consumers, employees, taxpayers.
![Page 16: Evaluation of Large Scale RDI Infrastructuresinstitute.eib.org/wp-content/uploads/2016/05/Florio-EIBpresentation... · Research Sponsorship Programme Proposal by University of Milan](https://reader034.fdocuments.us/reader034/viewer/2022042411/5f2a725b491ab42349178c17/html5/thumbnails/16.jpg)
The CBA model (2)
= Use benefits
= Non Use benefits
= Costs
![Page 17: Evaluation of Large Scale RDI Infrastructuresinstitute.eib.org/wp-content/uploads/2016/05/Florio-EIBpresentation... · Research Sponsorship Programme Proposal by University of Milan](https://reader034.fdocuments.us/reader034/viewer/2022042411/5f2a725b491ab42349178c17/html5/thumbnails/17.jpg)
)
The present value of COSTS is the sum of the:
• economic value of capital ( )
• labour cost of scientists ( )
• other administrative and technical staff ( )
• other operating costs ( )
• negative externalities if any ( ).
The CBA model: Costs and Benefits
5
The present value of
BENEFITS
is the sum of the:
• Firms ( )
• Employees ( )
• Users (A + S + C)
• Taxpayers (
(QOV + EXV)
5
![Page 18: Evaluation of Large Scale RDI Infrastructuresinstitute.eib.org/wp-content/uploads/2016/05/Florio-EIBpresentation... · Research Sponsorship Programme Proposal by University of Milan](https://reader034.fdocuments.us/reader034/viewer/2022042411/5f2a725b491ab42349178c17/html5/thumbnails/18.jpg)
It was built (1993-2008) by CERN.
It is located in a 27 km-long underground tunnel near Geneva.
In operation since 2009, it has discovered the Higgs Boson in 2013.
TIME HORIZON 33 years: 1993 - 2025
UNIT OF ANALYSIS the LHC and four detectors (collaborations)
SOCIAL DISCOUNT RATE 3% in real terms (adopted by the EC CBA Guide, 2014)
SHADOW PRICES Proxied by marginal WTP or marginal costs
COUNTERFACTUAL Business as usual scenario
QUASI-OPTION VALUE assumed 0
NEGATIVE EXTERNALITIES assumed 0
KEY PARAMETERS FOR THE CBA
Mont Blanc
Switzerland
France
The Large Hadron Collider (LHC)
Applications of the model
![Page 19: Evaluation of Large Scale RDI Infrastructuresinstitute.eib.org/wp-content/uploads/2016/05/Florio-EIBpresentation... · Research Sponsorship Programme Proposal by University of Milan](https://reader034.fdocuments.us/reader034/viewer/2022042411/5f2a725b491ab42349178c17/html5/thumbnails/19.jpg)
Consumers
Firms owners
Taxpayers
Employees
Benefits Customary partition of economic agents in the applied welfare economics literature:
Some evidence from literature: • Drèze, J. and Stern N. (1990). Policy reform, shadow prices and market prices, Journal of Public
Economics. • Johansson, P-O and Kriström, B. (2015). Cost-Benefit Analysis for Project Appraisal, Cambridge:
Cambridge University Press.
• Firms: profit maximization (producer surplus).
• Consumers: maximizing their utility (consumer surplus).
• Employees: maximizing their income for a given amount of efforts.
• Tax-payers: adjusting their decisions as a consequence of the existing fiscal constraints to minimize the burden of taxation.
![Page 20: Evaluation of Large Scale RDI Infrastructuresinstitute.eib.org/wp-content/uploads/2016/05/Florio-EIBpresentation... · Research Sponsorship Programme Proposal by University of Milan](https://reader034.fdocuments.us/reader034/viewer/2022042411/5f2a725b491ab42349178c17/html5/thumbnails/20.jpg)
The CBA model: Benefits
![Page 21: Evaluation of Large Scale RDI Infrastructuresinstitute.eib.org/wp-content/uploads/2016/05/Florio-EIBpresentation... · Research Sponsorship Programme Proposal by University of Milan](https://reader034.fdocuments.us/reader034/viewer/2022042411/5f2a725b491ab42349178c17/html5/thumbnails/21.jpg)
Technological Externalities
6
The value of presetn technological spillovers ( ) is given by:
• The discounted incremented social profits ( ) generated by companies
(j) of the RDI’s supply chain which have benefitted from a learning effect;
• And other externatilities.
tT
P jt
![Page 22: Evaluation of Large Scale RDI Infrastructuresinstitute.eib.org/wp-content/uploads/2016/05/Florio-EIBpresentation... · Research Sponsorship Programme Proposal by University of Milan](https://reader034.fdocuments.us/reader034/viewer/2022042411/5f2a725b491ab42349178c17/html5/thumbnails/22.jpg)
Benefits to suppliers Sample of 300 orders by purchase code compared with all LHC orders
LEGEND OF CERN ACTIVITY CODES 11 building work
12 roadworks
13 installation and supply of pipes
14 electrical installation work
15 heating and air-conditioning equipment (supply and installation)
16 hoisting gear
17 water supply and treatment
18 civil engineering and buildings
21 switch gear and switchboards
22 power transformers
23 power cables and conductors
24 control and communication cables
25 power supplies and converters
26 magnets
27 measurement and regulation
28 electrical engineering
29 electrical engineering components
31 active electronic components
32 passive electronic components
33 electronic measuring instruments
34 power supplies - transformers
35 functional modules & crates
36 rf and microwave components and equipment
37 circuit boards
38 electronics
39 electronic assembly and wiring work
41 computers and work-stations
42 storage systems
43 data-processing peripherals
44 interfaces (see also 35 series)
45 software
46 consumables items for data-processing
47 storage furniture (data-processing)
48 data communication
51 raw materials (supplies)
52 machine tools, workshop and quality control equipment
53 casting and moulding (manufacturing techniques)
54 forging (manufacturing techniques)
55 boiler metal work (manufacturing techniques)
56 sheet metal work (manufacturing techniques)
57 general machining work
58 precision machining work
59 specialised techniques
61 vacuum pumps
62 refrigeration equipment
63 gas-handling equipment
64 storage and transport of cryogens
65 measurement equipment (vacuum and low-temperature technology)
66 low-temperature materials
67 vacuum components & chambers
68 low-temperature components
69 vacuum and low-temperature technology
71 films and emulsions
72 scintillation counter components
73 wire chamber elements
74 special detector components
75 calorimeter elements
8a radiation protection
n.a. not available
STEP 1. IDENTIFICATION OF HIGH-TECH ORDERS
ACTIVITY CODES FOR HIGH-TECH ORDERS
POWER CABLES AND CONDUCTORS CASTING AND MOULDING (MANUFACTURING TECHNIQUES)
MAGNETS FORGING (MANUFACTURING TECHNIQUES)
MEASUREMENT AND REGULATION PRECISION MACHINING WORK
ELECTRICAL ENGINEERING VACUUM PUMPS
ELECTRICAL ENGINEERING COMPONENTS REFRIGERATION EQUIPMENT
ACTIVE ELECTRONIC COMPONENTS GAS-HANDLING EQUIPMENT
PASSIVE ELECTRONIC COMPONENTS STORAGE AND TRANSPORT OF CRYOGENS
ELECTRONIC MEASURING INSTRUMENTS MEASUREMENT EQUIPMENT (VACUUM AND LOW-TEMPERATURE TECHNOLOGY)
POWER SUPPLIERS - TRANSFORMERS LOW-TEMPERATURE MATERIALS
FUNCTIONAL MODULES & CRATES VACUUM COMPONENTS & CHAMBERS
RF AND MICROWAVE COMPONENTS AND EQUIPMENT LOW-TEMPERATURE COMPONENTS
CIRCUIT BOARDS VACUUM AND LOW-TEMPERATURE TECHNOLOGY
ELECTRONICS OPTICAL AND X-RAY EQUIPMENT
ELECTRONIC ASSEMBLY AND WIRING WORK
Technological Externalities
![Page 23: Evaluation of Large Scale RDI Infrastructuresinstitute.eib.org/wp-content/uploads/2016/05/Florio-EIBpresentation... · Research Sponsorship Programme Proposal by University of Milan](https://reader034.fdocuments.us/reader034/viewer/2022042411/5f2a725b491ab42349178c17/html5/thumbnails/23.jpg)
STEP 2. PROBABILITY DISTRIBUTION SHARE OF HIGH TECH PROCUREMENT- CERN
Empirical marginal Probability Density Function ; N = 10,0000
STEP 3. PROBABILITY DISTRIBUTION SHARE OF HIGH TECH PROCUREMENT- COLLABORATIONS
Empirical marginal Probability Density Function ; N = 10,0000
STEP 4. PROBABILITY DISTRIBUTION ECONOMIC UTILITY SALES RATIO
Empirical marginal Probability Density Function ; N = 10,0000
STEP 5. PROBABILITY DISTRIBUTION EBITDA MARGIN
Empirical marginal Probability Density Function ; N = 10,0000
Technological Externalities
![Page 24: Evaluation of Large Scale RDI Infrastructuresinstitute.eib.org/wp-content/uploads/2016/05/Florio-EIBpresentation... · Research Sponsorship Programme Proposal by University of Milan](https://reader034.fdocuments.us/reader034/viewer/2022042411/5f2a725b491ab42349178c17/html5/thumbnails/24.jpg)
TECHNOLOGICAL BENEFITS
ROOTTECHNOLOGICAL
TRANSFERGEANT4
•Multivariate analysis tool for very large datasets
•Available since1997
•External users: HEP community, industry
•Simulation software
•Available since1999
•External users: HEP community, spaceagencies, industry, hospitals
Licenses, start-ups, collaborationagreements
LHC
Benefits to software users
Technological Externalities
![Page 25: Evaluation of Large Scale RDI Infrastructuresinstitute.eib.org/wp-content/uploads/2016/05/Florio-EIBpresentation... · Research Sponsorship Programme Proposal by University of Milan](https://reader034.fdocuments.us/reader034/viewer/2022042411/5f2a725b491ab42349178c17/html5/thumbnails/25.jpg)
The present value of technological spillovers ( ) is given by:
• the discounted incremental social profits generated by companies ( ) of
the RI’s supply chain which have benefitted from a learning effect;
• and other externalities.
Technological Externalities
![Page 26: Evaluation of Large Scale RDI Infrastructuresinstitute.eib.org/wp-content/uploads/2016/05/Florio-EIBpresentation... · Research Sponsorship Programme Proposal by University of Milan](https://reader034.fdocuments.us/reader034/viewer/2022042411/5f2a725b491ab42349178c17/html5/thumbnails/26.jpg)
b) CERN external procurement -
commitment for total and high-tech orders a) Benefits to firms in the supply chain. Sample of 300 orders by purchase code
compared with all LHC orders
c) Distribution of EBITDA 2013 from ORBIS in firms at
NACE industry levels matched with CERN codes
d) ROOT download data
e) ENPV Cumulative distribution
function (Montecarlo simulation after
10,000 extractions) conditional to PDF
of critical variables (kEUR 2013)
Technological Externalities
![Page 27: Evaluation of Large Scale RDI Infrastructuresinstitute.eib.org/wp-content/uploads/2016/05/Florio-EIBpresentation... · Research Sponsorship Programme Proposal by University of Milan](https://reader034.fdocuments.us/reader034/viewer/2022042411/5f2a725b491ab42349178c17/html5/thumbnails/27.jpg)
12
• Objective: discovering whether there has been a long-term "learning
by doing" effect on CERN suppliers' profitability, beyond the initial
order.
• Access to CERN database on procurement data for LHC between
1995 and 2008. 1360 suppliers from 35 countries with least one order
of over CHF 10,000. A total of 11,969 orders. Out of the original list of
LHC suppliers, 1,060 companies have been identified in AMADEUS and
ORBIS. Companies (in total 500) - for which core financial indicators
over the time span 1991-2013 - were selected for a detailed analysis.
• Methodology entailed the collection of 23-year long time series of
financial data (1991-2013) for a large sample of companies, controlling
for time trends, firm-level and country-level possible confounding
factors.
Florio, M., Castelnovo, P., Sirtori, E., Rossi, L., Forte, S.
The economic impact of CERN procurement: evidence from
the Large Hadron Collider
Technological Externalities
![Page 28: Evaluation of Large Scale RDI Infrastructuresinstitute.eib.org/wp-content/uploads/2016/05/Florio-EIBpresentation... · Research Sponsorship Programme Proposal by University of Milan](https://reader034.fdocuments.us/reader034/viewer/2022042411/5f2a725b491ab42349178c17/html5/thumbnails/28.jpg)
13
• Research question: Is there any evidence in the data that the
profit margin of suppliers of the LHC increased over time
after the initial contract with CERN?
• Most interesting performance variable is the profit margin,
defined as the ratio of EBIT (Earnings Before Interest and
Taxes) to operating revenues. It is around 4.6% on average in the
sample.
• A dynamic panel data model was built. The following variables
were included amongst the controls: profit margins (lagged one
year), the firms’ total assets (control for size of the supplier) and
yearly GDP change in the country where the firm was located
and in the OECD area, a time trend variable, year fixed
effects.
Technological Externalities
![Page 29: Evaluation of Large Scale RDI Infrastructuresinstitute.eib.org/wp-content/uploads/2016/05/Florio-EIBpresentation... · Research Sponsorship Programme Proposal by University of Milan](https://reader034.fdocuments.us/reader034/viewer/2022042411/5f2a725b491ab42349178c17/html5/thumbnails/29.jpg)
Total CERN orders each year is the sum of the number of all the orders above CHF 10,000 in that year, including
multiple orders to individual firms. First time orders are the number of orders that were agreed with a firm for the
first time in that year. (First orders of 1995=2; 1996=17; 2008=4). Source: Processing of CERN data.
Total CERN orders for LHC (left scale) 5 44 443 646 818 866 1212 1068 2621 1504 1355 929 403 55
First time orders (right scale) 2 17 128 83 100 89 94 85 112 99 89 75 25 4
Technological Externalities
![Page 30: Evaluation of Large Scale RDI Infrastructuresinstitute.eib.org/wp-content/uploads/2016/05/Florio-EIBpresentation... · Research Sponsorship Programme Proposal by University of Milan](https://reader034.fdocuments.us/reader034/viewer/2022042411/5f2a725b491ab42349178c17/html5/thumbnails/30.jpg)
A: M
agn
ets
B: S
witc
h g
ea
r and
sw
itch
boa
rds
C: P
ow
er
cable
s a
nd c
ond
ucto
rs
D: C
ontr
ol an
d c
om
mun
ica
tion
cable
s
E: P
ow
er
supp
liers
an
dconve
rters
F:
Pa
ssi
ve e
lect
ronic
co
mpo
ne
nts
G:
Pow
er
sup
plie
rs -
tra
nsfo
rmers
H: R
F a
nd
mic
row
ave
com
pone
nts
I: R
aw
mate
rials
(su
pp
lies)
J: M
ach
ine t
oo
ls,
an
d q
ualit
ycontr
ol
K: C
asting
and m
ould
ing
L:
She
et m
eta
l w
ork
M:
Gen
era
l m
achin
ing w
ork
N: P
reci
sio
n m
achin
ing w
ork
O:
Spe
cia
lise
d te
chniq
ues
P: R
efr
igera
tion
equ
ipm
en
t
Q:
Sto
rage a
nd t
ranspo
rt o
fcry
og
ens
R: V
acuu
m a
nd
low
-tem
pera
ture
S: L
ow
-tem
pera
ture
mate
rials
T:
Va
cuum
com
pon
ents
&cham
bers
U: S
pe
cia
l dete
cto
rs c
om
pone
nts
0
200
400
600
800
1000
1200
1400
1600
1800
2000
Percent share of number of orders. Minor items omitted. Source: CERN data
.
Technological Externalities
![Page 31: Evaluation of Large Scale RDI Infrastructuresinstitute.eib.org/wp-content/uploads/2016/05/Florio-EIBpresentation... · Research Sponsorship Programme Proposal by University of Milan](https://reader034.fdocuments.us/reader034/viewer/2022042411/5f2a725b491ab42349178c17/html5/thumbnails/31.jpg)
Distribution of LHC suppliers across countries
AT, 15
BE, 29
BG, 3 CA, 1
CZ, 8
DK, 9
FI, 14
FR, 401
DE, 148
GR, 5
HU, 3
IN, 3
IL, 3
IT, 101
JP, 5
LU, 3
NL, 33NO, 4
PL, 11
PT, 7RU, 15
SK, 4
ES, 37
SE, 15
CH, 371
UK, 77
US, 32
Others, 8
TOTAL (
SAMPLE
AT, 14
BE, 24
BG, 1CA, 1
CZ, 8 DK, 6
FI, 13FR, 272
DE, 138
GR, 5HU, 3
IN, 0
IL, 3
IT, 82
JP, 5LU, 3
NL, 33NO, 3
PL, 5
PT, 5RU, 8
SK, 3ES, 30SE, 14
CH, 279
UK, 71
US, 25
Others, 4
≅ 1000)
(≅ 500)
Technological Externalities
![Page 32: Evaluation of Large Scale RDI Infrastructuresinstitute.eib.org/wp-content/uploads/2016/05/Florio-EIBpresentation... · Research Sponsorship Programme Proposal by University of Milan](https://reader034.fdocuments.us/reader034/viewer/2022042411/5f2a725b491ab42349178c17/html5/thumbnails/32.jpg)
LHC suppliers distribution by NACE code
0 50 100 150 200 250
20. Manufacture of chemicals and chemical products
22. Manufacture of rubber and plastic products
23. Manufacture of other non-metallic mineral products
24. Manufacture of basic meta ls
25. Manufacture of fabr icated metal products, except machinery
26. Manufacture of computer, electronic and optical products
27. Manufacture of e lectrical equipment
28. Manufacture of machinery and equipment n.e.c.
30. Manufacture of o ther transpor t equipment
32. Other manufacturing
33. Repair and insta llation of machinery and equipment
35. Electricity, gas, steam and air conditioning supply
41. Construction of build ings
42. Civil engineering
43. Specialised construction activities
45. Wholesale and retail trade and repai r of motor vehicles
46. Wholesale trade, except of motor vehicles and motorcycles
47. Retail trade, except of motor vehicles and motorcycles
61. Telecommunications
62. Computer programming, consultancy
64. Financial service activities, except insurance
70. Activities of head offices; management consultancy
71. Architectural and engineer ing activities; technical testing
72. Scientific research and development
77. Rental and leasing activities
82. Office support and other business support activities
85. Education
Other (1)
n. of firms
Technological Externalities
![Page 33: Evaluation of Large Scale RDI Infrastructuresinstitute.eib.org/wp-content/uploads/2016/05/Florio-EIBpresentation... · Research Sponsorship Programme Proposal by University of Milan](https://reader034.fdocuments.us/reader034/viewer/2022042411/5f2a725b491ab42349178c17/html5/thumbnails/33.jpg)
average std. dev. 25th median 75th
Total sample 4.58% 12.94 1.21% 4.26% 8.45%
High-tech 4.62% 13.44 1.35% 4.46% 8.53%
Non-hi-tech 4.52% 11.98 0.95% 3.84% 8.15%
EBIT margin descriptive statistics
EBIT margin = EBIT/Operating Revenues. 25th and 75th are percentiles.
Source: Processing of CERN and AMADEUS-ORBIS data
Technological Externalities
![Page 34: Evaluation of Large Scale RDI Infrastructuresinstitute.eib.org/wp-content/uploads/2016/05/Florio-EIBpresentation... · Research Sponsorship Programme Proposal by University of Milan](https://reader034.fdocuments.us/reader034/viewer/2022042411/5f2a725b491ab42349178c17/html5/thumbnails/34.jpg)
Regression
results
EBIT MARGIN (%)
Full Sample High-Tech Firms Non High-Tech Firms
CERN Effect 0.673** 1.007*** 0.642
p-value (0.032) (0.001) (0.256)
Firm Level Controls
∆Total Assets1 YES YES YES**
∆EBIT margin, 1-year lagged
value YES*** YES*** YES***
Macroeconomic Controls
∆GDP growth rate, firm’s
country YES** YES* YES***
∆GDP growth rate, OECD Area YES YES YES**
Year fixed effect YES*** YES*** YES***
Time trend YES*** YES*** YES***
Constant YES*** YES YES***
Number of observations 4856 3017 1676
Number of Firms 333 203 118
Standard Errors are
clustered by country.
* p < 0.1, ** p < 0.05, *** p
< 0.01
Technological Externalities
![Page 35: Evaluation of Large Scale RDI Infrastructuresinstitute.eib.org/wp-content/uploads/2016/05/Florio-EIBpresentation... · Research Sponsorship Programme Proposal by University of Milan](https://reader034.fdocuments.us/reader034/viewer/2022042411/5f2a725b491ab42349178c17/html5/thumbnails/35.jpg)
Full Sample
0= Year of the 1st order
-2.0
-1.5
-1.0
-0.5
0.0
0.5
1.0
1.5
2.0
2.5
-15 -14 -13 -12 -11 -10 -9 -8 -7 -6 -5 -4 -3 -2 -1 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16
EBIT margin mean
high tech
-2.5
-2.0
-1.5
-1.0
-0.5
0.0
0.5
1.0
1.5
2.0
2.5
3.0
-15 -14 -13 -12 -11 -10 -9 -8 -7 -6 -5 -4 -3 -2 -1 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16
EBIT margin media
non-high-tech
-1.0
-0.5
0.0
0.5
1.0
1.5
2.0
2.5
3.0
3.5
-15 -14 -13 -12 -11 -10 -9 -8 -7 -6 -5 -4 -3 -2 -1 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15
EBIT margin media
Technological Externalities
![Page 36: Evaluation of Large Scale RDI Infrastructuresinstitute.eib.org/wp-content/uploads/2016/05/Florio-EIBpresentation... · Research Sponsorship Programme Proposal by University of Milan](https://reader034.fdocuments.us/reader034/viewer/2022042411/5f2a725b491ab42349178c17/html5/thumbnails/36.jpg)
Focus on Technological impact
What can we learn from the LHC about the
impact on firms?
• Earlier studies at CERN were based on interviews
to managers. They concluded that for each Euro
of procurement there will be 3 Euro of sales to
other customers (or cost savings).
• Problems: not updated, selection bias, and
sales per se are not a social benefit.
• Lessons for the FCC Study.
![Page 37: Evaluation of Large Scale RDI Infrastructuresinstitute.eib.org/wp-content/uploads/2016/05/Florio-EIBpresentation... · Research Sponsorship Programme Proposal by University of Milan](https://reader034.fdocuments.us/reader034/viewer/2022042411/5f2a725b491ab42349178c17/html5/thumbnails/37.jpg)
Human Capital Formation
6
Human capital formation benefits (H) are valued as increased earnings 𝐈 gained by
RI’s students and former employees 𝐳 , since the moment 𝛗 they leave the
project, against counterfactual scenario.
![Page 38: Evaluation of Large Scale RDI Infrastructuresinstitute.eib.org/wp-content/uploads/2016/05/Florio-EIBpresentation... · Research Sponsorship Programme Proposal by University of Milan](https://reader034.fdocuments.us/reader034/viewer/2022042411/5f2a725b491ab42349178c17/html5/thumbnails/38.jpg)
Sector CERN
fellows
CERN
technical
students
CERN
doctoral
students
User-
students and
post-docs
Industry 20% 45% 20% 20%
Others
(computing,
finance, public
administration, …)
20% 45% 20% 20%
Research centres 30% 5% 30% 30%
Academia 30% 5% 30% 30%
TOTAL 100% 100% 100% 100%
Variable
Number over the 1993-2025 period
Average staying at CERN
CERN fellows working on LHC 5,873 2 years
CERN technical students working on LHC
3,940 1 year
CERN doctoral students working on LHC
1,332 3 years
User-students working on LHC 14,225 3 years
Post-doc researchers (users) working on LHC
11,301 2 years
TOTAL 36,671
Sources: - CERN personnel statistics; - Interviews to CERN staff
Main assumptions: - Future number of beneficiaries; - Number of users-students and post-docs among users (assumed based on their age group); - Incoming number of user-students and post docs
TYPES AND QUANTITIES OF PEOPLE
BENEFITTING FROM TRAINING
TYPES AND NUMBER OF PEOPLE
BENEFITTING FROM TRAINING
Post-docs (users 31-35 yrs old)
User-students (<30 yrs old)
Fellows
Technical students
Doctoral students
ESTIMATION OF FUTURE AVERAGE SALARIES DETERMINING THE RETURN TO SALARY DUE TO LHC TRAINING
SALARY EFFECT (1)
SALARY BONUS
FOR JOB EFFECT (2)
Sector CERN fellows,
doctoral students,
user students, post-
docs
CERN technical
students
Research centres
9.3% 2.5% Academia
Industry
Others (computing, financial, …)
(1) Survey to 192 former LHC students (out of a total survey to 385 students and former students): declared salary impact of the experience at LHC on their current salary
(2) Own assumption based on survey results and Payscale salaries
Main source: Findings from the survey to LHC current and former students
Main assumptions: • Same economic return regardless of the professional sector and type of student • Same return over the entire work career (40 yrs)
ASSUMED DISTRIBUTION OF FORMER LHC
STUDENTS BY PROFESSIONAL SECTOR
Industry: y = 12731ln(x) + 31792
Others: y = 14180ln(x) + 36165
Research, Academia: y = 9685.8ln(x) + 32575
0
10,000
20,000
30,000
40,000
50,000
60,000
70,000
80,000
90,000
100,000
0 10 20 30 40
EU
R
Years of career
Industry
Others
Research centres
Academia
Log. (Industry)
Log. (Others)
Log. (Academia)
Human Capital Formation
![Page 39: Evaluation of Large Scale RDI Infrastructuresinstitute.eib.org/wp-content/uploads/2016/05/Florio-EIBpresentation... · Research Sponsorship Programme Proposal by University of Milan](https://reader034.fdocuments.us/reader034/viewer/2022042411/5f2a725b491ab42349178c17/html5/thumbnails/39.jpg)
SHARE OF RESPONDENTS BY EXPERIMENT
SKILLS IMPROVED THANKS TO THE LHC
EXPERIENCE. AVERAGE JUDGEMENT
AN OVERVIEW OF CURRENT EMPLOYMENT
SECTOR. SHARE OF RESPONDENTS
AVERAGE SALARY EVOLUTION: A COMPARISON BETWEEN THE TWO
GROUPS OF RESPONDENTS (THOUSAND EUR)
37.9
64.7
84.7
38.3
75.8
94.4
0
10
20
30
40
50
60
70
80
90
100
Entry Salary Mid-Career Salary End-Career Salary
Respondentswho are currentlystudying orunemployed
Respondentswho are currentlyworking
THE IMPACT OF LHC EXPERIENCE ON SALARY (%)
ICT sector
(e.g. computing)
9%
ALICE5%
ATLAS22%
CMS65%
LHCb7%
Other1% 4.29
4.17
3.93
4.053.38
3.42
3.81
0.0
1.0
2.0
3.0
4.0
5.0Scientific skills
Technical skills
Communication skills
Problem-solving capacity
Team/project leadership
Developing, maintaining and
using networks of collaborations
Independent thinking/critical
analysis/creativity
Human Capital Formation
![Page 40: Evaluation of Large Scale RDI Infrastructuresinstitute.eib.org/wp-content/uploads/2016/05/Florio-EIBpresentation... · Research Sponsorship Programme Proposal by University of Milan](https://reader034.fdocuments.us/reader034/viewer/2022042411/5f2a725b491ab42349178c17/html5/thumbnails/40.jpg)
Knowledge output
6
• The sum of the present value of papers signed by RDI’s scientists (P0t) and the value of subsequentflows of papers produced by other scientists that se or elaborate the RDI scientist’s results.
• Divided by the number of reference they contain (Pit / Kit wih i=1,…n) and the value of citation each paper receives, as a proxy of the social recognition that the scientific community ackonwledges to the paper (Qit with i=0,…n) .
The social value of knowledge output is measured by:
![Page 41: Evaluation of Large Scale RDI Infrastructuresinstitute.eib.org/wp-content/uploads/2016/05/Florio-EIBpresentation... · Research Sponsorship Programme Proposal by University of Milan](https://reader034.fdocuments.us/reader034/viewer/2022042411/5f2a725b491ab42349178c17/html5/thumbnails/41.jpg)
Knowledge output
6
41
The social value of knowledge output is measured by:
the sum of the present value of papers signed by RI’s scientists and the value of
subsequent flows of papers produced by other scientists that use or elaborate of the RI’s
scientists’ results
divided by the number of references they contain and the value of
citations each paper receives, as a proxy of the social recognition that the scientific
community acknowledges to the paper
![Page 42: Evaluation of Large Scale RDI Infrastructuresinstitute.eib.org/wp-content/uploads/2016/05/Florio-EIBpresentation... · Research Sponsorship Programme Proposal by University of Milan](https://reader034.fdocuments.us/reader034/viewer/2022042411/5f2a725b491ab42349178c17/html5/thumbnails/42.jpg)
Knowledge output
6
42
PAPERS PRODUCED BY LHC USERS (L0) PAPERS PRODUCED BY NON-LHC USERS (L1 & L2)
DOWNLOADS OF LHC PAPERS (D1)
0
500
1,000
1,500
2,000
2,500
19
93
19
95
19
97
19
99
20
01
20
03
20
05
20
07
20
09
20
11
20
13
20
15
20
17
20
19
20
21
20
23
20
25
Number of papers L0
L0, 2013-2025
L0, 1993-2012
Forecast
0
20,000
40,000
60,000
80,000
100,000
120,000
140,000
160,000
180,000
199
3
199
5
199
7
199
9
200
1
200
3
200
5
200
7
200
9
201
1
201
3
201
5
201
7
201
9
202
1
202
3
202
5
202
7
202
9
203
1
203
3
203
5
203
7
203
9
204
1
204
3
204
5
204
7
204
9
Number of papers L0, L1 and L2
L0, 1993-2025 L1, 1993-2050 L2, 1993-2050
Forecast
0
10
20
30
40
50
60
70
80
90
1994
1995
1996
1997
1998
1999
2000
2001
2002
2003
2004
2005
2006
2007
2008
2009
2010
2011
2012
2013
Number of downloads per paper (ArXiv, field HEP)
0
20,000
40,000
60,000
80,000
100,000
120,000
140,000
160,000
180,0001
99
3
19
96
19
99
20
02
20
05
20
08
20
11
20
14
20
17
20
20
20
23
20
26
20
29
20
32
20
35
20
38
20
41
20
44
20
47
20
50
Number of papers L0, L1 and L2 and downloads D1
L0, 1993-2025 L1, 1993-2050
L2, 1993-2050 D1, 1993-2050
Forecast
OUR RESULTS
VALUATION
TRACKING THE KNOWLEDGE OUTPUTS
Quantification of citations L1 Quantification of citations L2
Future number of citations
L2 per paper L0 = 4
Source: Preliminary scientometric analysis of INSPIRE database of papers and citations
42
Present value of papers L1
Present value of citations L1
Present value of citations L2
Present value of downloads
Unit economic value of papers L1 Value Source
Number of references in
paper L1 35
Own assumption, based on an analysis of 41
research journals by Abt and Garfield (2002)
Share of time dedicated
to research 65%
Own assumption. The remainder is for
teaching and other non scientific activities
Number of paper
(published
and non) per year
3.5
Own assumption. It represents the number of
papers to wich a scientist gives a real
contribution
Average annual gross
salary 59,289 €
Own elaboration based on PayScale data. It is
the average salary for a scientists working in
research centres and academia in the USA
Unit production cost per
paper L1 315 € = (59,289 € *
65%/3.5/35)
Own estimation, based on the approach
suggested by Florio and Sirtori (2014)
Unit economic value of citations and downloads Value Source
Working hours per year 1,800 = 225 working days *
8 hours/day Own assumption
Average hourly gross salary 33 € = 59,289/1,800 Own estimation
Hours per citation 3 Own assumption
Hours per download 3 Own assumption
Value of one citation L1 and
L2 99 € = 33 € * 3
Own estimation, based on Florio
and Sirtori (2014)
Value of one L0 paper
downloaded but non cited 99 € = 33 € * 3 Own estimation, based on Florio
and Sirtori (2014)
Except
![Page 43: Evaluation of Large Scale RDI Infrastructuresinstitute.eib.org/wp-content/uploads/2016/05/Florio-EIBpresentation... · Research Sponsorship Programme Proposal by University of Milan](https://reader034.fdocuments.us/reader034/viewer/2022042411/5f2a725b491ab42349178c17/html5/thumbnails/43.jpg)
Cultural Effects
Outreach activities carried out by RI produce cultural effects on the general
public 𝑔 , which can be valued by estimating the willingness to pay of the
general public for such activities.
![Page 44: Evaluation of Large Scale RDI Infrastructuresinstitute.eib.org/wp-content/uploads/2016/05/Florio-EIBpresentation... · Research Sponsorship Programme Proposal by University of Milan](https://reader034.fdocuments.us/reader034/viewer/2022042411/5f2a725b491ab42349178c17/html5/thumbnails/44.jpg)
Cultural Effects
Outreach activities carried out by RI produce cultural effects on the
general public 𝑔 , which can be valued by estimating the willingness to
pay of the general public for such activities.
![Page 45: Evaluation of Large Scale RDI Infrastructuresinstitute.eib.org/wp-content/uploads/2016/05/Florio-EIBpresentation... · Research Sponsorship Programme Proposal by University of Milan](https://reader034.fdocuments.us/reader034/viewer/2022042411/5f2a725b491ab42349178c17/html5/thumbnails/45.jpg)
Cultural Effects TRAVEL ZONES CONSIDERED
VALUATION THROUGH THE
TRAVEL COST METHOD
Origin
zone
Radius distance
from CERN
Share of
visitors
Source/
Assumption
Zone 1 500 km 24% CERN
Zone 2 500-1,500 km 50% Own assumption
Zone 3 Beyond 1,500 km 26% Own assumption
Main assumption: • % of visitors by mode of transport • Travel cost by zone
Source: HEATCO values of travel time by modes of transport
45
Zone 1
Zone 2
Zone 3
BENEFIT FOR SOCIAL MEDIA USERS
Main source: CERN staff
Main assumption: Future number of visitors
BENEFITS TO PERSONAL VISITORS:
QUANTIFICATION OF VISITORS
BENEFIT FOR WEBSITE VISITORS
Main assumption: Benefit = value of time spent on social media: approximate 2 minutes/hit
Main assumption: Benefit = value of time spent on social media
MASS MEDIA BENEFITS: NEWS BY MEDIA CHART
BENEFIT FOR VOLUNTEER COMPUTING
0
2,000
4,000
6,000
8,000
10,000
12,000
20
07
20
08
20
09
20
10
20
11
20
12
20
13
20
14
20
15
20
16
20
17
20
18
20
19
20
20
20
21
20
22
20
23
20
24
20
25
Number of volunteers - Six track
Number of volunteers - Test4 Theory
Main assumption: Benefit = Value of time spent to download, on forum.
SHARE OF BENEFITS BY TYPE OF OUTREACH ACTIVITY
![Page 46: Evaluation of Large Scale RDI Infrastructuresinstitute.eib.org/wp-content/uploads/2016/05/Florio-EIBpresentation... · Research Sponsorship Programme Proposal by University of Milan](https://reader034.fdocuments.us/reader034/viewer/2022042411/5f2a725b491ab42349178c17/html5/thumbnails/46.jpg)
The Non-use Benefit
where
• QOV is intrinsically uncertain and therefore not measurable, simply
assumed to be non-negative and then skipped;
• EXV, on the other hand, can be proxied by stated or revealed willingness to
pay for scientific research, and/or through benefit transfer, borrowing ideas
from CBA of the environment.
Bn captures two types of benefit related to the social value of discovery: the quasi-option value (QOV) and the existence value (EXV)
![Page 47: Evaluation of Large Scale RDI Infrastructuresinstitute.eib.org/wp-content/uploads/2016/05/Florio-EIBpresentation... · Research Sponsorship Programme Proposal by University of Milan](https://reader034.fdocuments.us/reader034/viewer/2022042411/5f2a725b491ab42349178c17/html5/thumbnails/47.jpg)
The Non-use Benefit 47
captures two types of benefits related to the social value of discovery:
the quasi-option value (QOV) and the existence value (EXV):
where
• QOV is intrinsically uncertain and therefore not measurable, simply assumed to be non-
negative and then skipped;
• EXV, on the other hand, can be proxied by stated or revealed willingness to pay for scientific
research, and/or through benefit transfer, borrowing ideas from CBA of the environment.
![Page 48: Evaluation of Large Scale RDI Infrastructuresinstitute.eib.org/wp-content/uploads/2016/05/Florio-EIBpresentation... · Research Sponsorship Programme Proposal by University of Milan](https://reader034.fdocuments.us/reader034/viewer/2022042411/5f2a725b491ab42349178c17/html5/thumbnails/48.jpg)
The Non-use Benefit 48
0%
10%
20%
30%
40%
50%
60%
70%
80%
90%
100%
ES FR IT UK
Humanistic Mixed Scientific n.a.
RESPONDENTS BY UNIVERSITY DEGREE RESPONDENTS BY LEVEL OF HOUSEHOLD INCOME GENDER Number
Female 581
Male 446
Total 1027
COUNTRY Number Italy 422
Spain 204
France 201
UK 200
Total 1027
YEARS Number 19-25 years 875
26-30 years 95
31-35 years 34
Over 35 years 20
n.a. 3
Total 1027
0%
20%
40%
60%
80%
100%
ES FR IT UK
up to EUR 1000 EUR 1000-3000 EUR 3000-5000
higher than EUR 5000 n.a.
RATING THE IMPORTANCE TO FINANCE RDI
Essential
38%
Important
46%
Rather
Important
14%
It is
insignificant
1%
It is not
necessary
1%
WHAT IS THE UTILITY OF THE LHC
0%
20%
40%
60%
80%
100%
ES FR IT UK
NO I DON'T KNOW YES
WILLINGNESS TO PAY FOR LHC
27%
8%
22%
43%
0%
5%
10%
15%
20%
25%
30%
35%
40%
45%
50%
0 0.5 1 2
EUR per person per year
AVERAGE ANNUAL WTP WTP TO PAY UNA TANTUM
0%
5%
10%
15%
20%
25%
30%
Average = 18%
SHARE OF ADULT POPULATION (18-74 YEARS OLD) WITH AT
LEAST TERTIARY EDUCATION
NO
49.1%
I DON'T
KNOW
36.6%
YES
14.3%
0
200
400
600
800
1000
It is a worthless infrastructurewhose construction could
have been avoided
It is an infrastructure ofinterest for physicists
It is an infrastructuredangerous because of the risk
of nuclear accidents
It is an infrastructure useful forthe production of the energy
It is a useful infrastructure toexperience accelerations
between protons that can beused for many purposes
![Page 49: Evaluation of Large Scale RDI Infrastructuresinstitute.eib.org/wp-content/uploads/2016/05/Florio-EIBpresentation... · Research Sponsorship Programme Proposal by University of Milan](https://reader034.fdocuments.us/reader034/viewer/2022042411/5f2a725b491ab42349178c17/html5/thumbnails/49.jpg)
LHC: CBA in a nutshell (1)
●
TOTAL MEASURED BENEFITS
LHC: summary of costs and benefits (Billion, EUR)
COSTS: 13.5 ± 0.4
USE BENEFITS:
Knowledge Formation 0.3 ± 0.1
Human Capital 5.5 ± 0.3
Technological Spillovers 5.3 ±1.7
Cultural 2.1 ± 0.5
NON-USE BENEFITS:
Existence Value 3.2 ± 1.0
Human capital, technological spillovers,
cultural + existence value each give
about 33% of benefits (publications are
negligible)
Uncertainty largest on technological
spillovers
More than 90% chance of positive NPV
Scientific publications 2%
Human capital formation 33%
Technological spillovers 32%
Cultural effects 13%
Existence value 20%
![Page 50: Evaluation of Large Scale RDI Infrastructuresinstitute.eib.org/wp-content/uploads/2016/05/Florio-EIBpresentation... · Research Sponsorship Programme Proposal by University of Milan](https://reader034.fdocuments.us/reader034/viewer/2022042411/5f2a725b491ab42349178c17/html5/thumbnails/50.jpg)
LHC: CBA in a nutshell (2)
●
ESTIMATED PARAMETERS OF
DISTRIBUTION
Mean 2,855,528
Median 2,825,860
Standard
Deviation
2,134,763
Minimum -6,220,259
Maximum 11,573,387
Estimated probabilities
Pr. ENPV ≤ 0 0.086
Montecarlo error
3 ϭ 10,000 0.02
PROBABILITY DENSITY FUNCTION
![Page 51: Evaluation of Large Scale RDI Infrastructuresinstitute.eib.org/wp-content/uploads/2016/05/Florio-EIBpresentation... · Research Sponsorship Programme Proposal by University of Milan](https://reader034.fdocuments.us/reader034/viewer/2022042411/5f2a725b491ab42349178c17/html5/thumbnails/51.jpg)
Provision of Services Social benefits of RDI services for target groups of consumers
Some Examples:
• NASA Glenn Research Center
• European Synchrotron Radiation Facility
• High Magnetic Field Laboratory in The
Netherlands
• Laserlab-Europe
Some RDI infrastructures provide services to external users. They may pay a fee for accessing and using the infrastructure’s equipment and/or specific services offered.
Some RDI infrastructures are expected to use new knowledge to deliver innovative services and products addressing specific societal needs. Benefits arise to users who are better off by the delivery of the innovative service or product.
Some Examples:
• Reduction of GHG and air pollutant emissions
• Improved energy efficiency
• Reduction in vulnerability and exposure to
natural hazards
• Improved health conditions
Social benefits to consumers of services
![Page 52: Evaluation of Large Scale RDI Infrastructuresinstitute.eib.org/wp-content/uploads/2016/05/Florio-EIBpresentation... · Research Sponsorship Programme Proposal by University of Milan](https://reader034.fdocuments.us/reader034/viewer/2022042411/5f2a725b491ab42349178c17/html5/thumbnails/52.jpg)
Applications of the model
CNAO - National Hadrontherapy Center for Cancer Treatment
TIME HORIZON 30 years: 2001 - 2031
UNIT OF ANALYSIS
The hall hosting the particle accelerators and the other areas functional to the proper functioning of the clinical facility
SOCIAL DISCOUNT RATE 3% in real terms (adopted by the EC CBA Guide, 2014)
SHADOW PRICES Proxied by marginal WTP or marginal costs
COUNTERFACTUAL Do-nothing
Non-use benefits assumed 0
NEGATIVE EXTERNALITIES assumed 0
KEY PARAMETERS FOR THE CBA
It is an applied research facility specialised in hadrontherapy, an advanced oncological
treatment showing (in some cases) clinical advantages as compared to traditional radiotherapy.
It comprises two broad distinct areas: the high technology components, made of a set of
accelerators and transport lines of particle beams, and a clinical ‘day hospital’ facility,
comprising reception desks, waiting and changing rooms.
![Page 53: Evaluation of Large Scale RDI Infrastructuresinstitute.eib.org/wp-content/uploads/2016/05/Florio-EIBpresentation... · Research Sponsorship Programme Proposal by University of Milan](https://reader034.fdocuments.us/reader034/viewer/2022042411/5f2a725b491ab42349178c17/html5/thumbnails/53.jpg)
ENPV
Past investment costs
The ingredients of the CBA of CNAO
Past operating costs
Future operating costs
Future investment costs
Knowledge output
Technological spillovers
Human capital formation
Cultural effects
Health benefits
Decommissioning costs Provision of Services
Applications of the model
![Page 54: Evaluation of Large Scale RDI Infrastructuresinstitute.eib.org/wp-content/uploads/2016/05/Florio-EIBpresentation... · Research Sponsorship Programme Proposal by University of Milan](https://reader034.fdocuments.us/reader034/viewer/2022042411/5f2a725b491ab42349178c17/html5/thumbnails/54.jpg)
# of protocol Clinical alternative Marginal percentage of patients who fully recover compared to the
counterfactual situation
1 No alternative 73%
2 No alternative 73%
3 No alternative 33%
9 Surgery + photon therapy 45%
10 Surgey 21%
11 No alterative* 45%
12 No alterative* 14%
15 Surgery + photon therapy 30%
16 Photon therapy 43%
13 No alterative* 33%
19 Photontherapy 36%
Type 1 – FULL RECOVERY Marginal benefit by protocols
# of
protocol Clinical alternative
Marginal percentage of patients
who fully recover compared to the
counterfactual
Number of life years
gained with respect to
the counterfactual
6 No alternative for advanced tumours 15% 5
8 No alterative 43% 3
14 No alterative* 68% 0.5
18 Palliative chemotherapy 40% 2
20 No alterative 43% 3
22 Surgey + photon therapy 10% 5
23 Photontherapy* 35% 5
Type 2 – INCREASE IN LIFE EXPECTANCY Marginal benefit by protocols
Type 3 – BETTER QUALITY OF LIFE Marginal benefit by protocols
# of
protocol Clinical alternative
Marginal percentage of
patients who fully recover
compared to the
counterfactual
Number of life years gained
with respect to the
counterfactual
Quality of life
adjustment factor*
7 No alterative 100% 1 0.3
21 Surgey 100% 15 0.3
Probability distribution of applied research benefits on patients (Euros)
Estimated parameters of the distribution
Mean 2,028,626,666
Median 1,984,699,763
Standard deviation 495,675,860
Minimum 935,508,430
Maximum 4,061,318,078
Estimated probabilities
Pr. EPV ≤ base value 0.480
Pr. EPV ≤ 0 0.000
CNAO: Estimation of health benefits Applications of the model
![Page 55: Evaluation of Large Scale RDI Infrastructuresinstitute.eib.org/wp-content/uploads/2016/05/Florio-EIBpresentation... · Research Sponsorship Programme Proposal by University of Milan](https://reader034.fdocuments.us/reader034/viewer/2022042411/5f2a725b491ab42349178c17/html5/thumbnails/55.jpg)
CNAO: CBA Results PROBABILITY DISTRIBUTION OF
THE CNAO NET PRESENT VALUE
Own estimate of the Present Value PDF resulting from a Monte Carlo simulation (10,000 random extractions)
ESTIMATED PARAMETERS OF DISTRIBUTION
Mean 1,658,358
Median 1,615,046
Standard Deviation 499,225
Minimum 498,433
Maximum 3,686,989
Estimated probabilities
Pr. ENPV ≤ 0 0.000
0.00
0.01
0.02
0.03
0.04
0.05
0.06
0.07
0.08
0.09
ENPV Probability Density Function
0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1
498,433,509 1,136,144,790 1,773,856,072 2,411,567,353 3,049,278,634 3,686,989,916
ENPV
ENPV Cumulative Distribution Function
Cumulated probability CBA reference value
Mean Median
Std. Dev. from mean
Carbon Ion Therapy
Proton Therapy
Revenues
Benefit of Technological Spillovers
Benefit of Human Capital Generation
Benefit of Knowledge Creation
Benefit of Cultural Outreach
74.2
20.9%
2.2%
1.1%
0.7%
0.6%
0.3%
Health
benefits
55/0
Values in Thousands EUR, 2013
Applications of the model
![Page 56: Evaluation of Large Scale RDI Infrastructuresinstitute.eib.org/wp-content/uploads/2016/05/Florio-EIBpresentation... · Research Sponsorship Programme Proposal by University of Milan](https://reader034.fdocuments.us/reader034/viewer/2022042411/5f2a725b491ab42349178c17/html5/thumbnails/56.jpg)
Conclusions: Future Research
8
Testing the model on other Research Infrastructures
Forecasting technological spillovers with a control group
of firms (non-CERN suppliers)
Estimating human capital effects with econometric
‘treatment’ techniques
Developing a forecasting model for media impact of
outreach
Expanding the contingent valuation of willingness to pay
for discoveries
Applying the model to the Future Circular Collider
![Page 57: Evaluation of Large Scale RDI Infrastructuresinstitute.eib.org/wp-content/uploads/2016/05/Florio-EIBpresentation... · Research Sponsorship Programme Proposal by University of Milan](https://reader034.fdocuments.us/reader034/viewer/2022042411/5f2a725b491ab42349178c17/html5/thumbnails/57.jpg)
THANK YOU
Florio, M., Forte, S. and Sirtori, E.
2016
Forecasting the Social Impact
of the Large Hadron Collider:
A Cost-Benefit Analysis to
2025 and Beyond.
Available at
http://arxiv.org/abs/1603.00886
Forthcoming on Technological
Forecasting & Social Change,
Special Issue on the Social
Impact of Research
Infrastructures at the Frontier of Science and Technology