presentation of the results of the FLINT project on collecting sustainability data for the US Academy of Science CNSTAT committee, UC Davis, May 217

1. FLINT: Farm Level Indicators for New Topics
Krijn Poppe | Wageningen Economic Research
Thanks to Hans Vrolijk and the FLINT partners

2. Context: Sustainable Development Goals
2

3. Incentivise farmers on sustainability (PPP)
0
20
40
60
80
100
Cost price
per 100
kg milk
Income per
Family
Labour unit
solvability
(%)
Energy use
per euro output
Water use per euro output
Pesticide use
per hectare
Grazing days
Education
Surplus of
Phosphate per
hectare
Surplus of
Nitrogen per
hectare
PEOPLE
PROFIT
<< PLANET >>

4. Context: country initiatives (example)
4

5. Context: company initiatives (examples)
5

6. Context: industry initiatives (example)
6

7. Challenge: the need for common indicators
7

8. 8
Even bigger challenge: data collection
The FLINT project
(Farm Level
Indicators on New
Topics) tested an
infrastructure to
collect data on
farms in a
harmonised way,
linked to the
FADN

9. 9
Building upon the FADN
And that is
already in some
member states
used to collect
additional
sustainability
data (e.g. NL,
Ireland, Flanders)
Farm
Accountancy
Data
Network
(FADN)
85.000 farms
that are
monitored by EU
on economics and
income

10. Need for an Integrated data set
• Measurement of different sustainability indicators on
the same set of farms is essential
• Allows the analysis of the full chain from: Policy
objective -> policy measure -> impact on farm -> farm
management decisions -> up to: sustainability
performance of farms
• Trade-off and jointness of performance on different
sustainability measures as a consequence of policy
measures
– (for example is the economic performance at the expense of environmental performance,
sustainability performance of large farms etc.)

11. FLINT data collection

12. 12
FLINT Indicator list (economic data already in FADN)
S1: Advisory service
S2: Education and
training
S3: Ownership
management
S4: Social engagement S5: Working conditions S6: Quality of life
S7: Social
diversification
E3: Semi-natural areas
E6: Soil organic matter E10: Nitrate leaching
E11: Soil erosion E12: Use of legumes
E4: Pesticide usage
E5: Nutrient balance E7: Indirect energy use
E8: Direct energy
usage
E9: On-farm RE prod.
E1: Greening
E16: Water usage,
storage
E17: Irrigation
practices
E14: GHG calculation
EI1: Innovation
EI2: Producing under
label
EI3: Market outlet
EI4: Farm duration
EI5: Efficiency field
parcel
EI7: Insurance
EI8: Marketing
contracts
EI9: Risk exposure EI6: Modernization
Environmental
Economic,
innovation
Social
sustainability

13. 13
Number of farms in the test of data collection
# of farms FLINT
data collected
Finland 49
Germany 52
Ireland 63
Hungary 102
Greece 124
Poland 146
NL 155
Spain 128
France 280
Total 1099

14. Example of results: dairy farms in 6 MS.
14
Variable Unit ES FI FR IE NL PL
Dairy cows heads 74.4 42.8 50.0 75.0 99.2 22.4
Gross farm income Euro 78,630 113,319 58,943 82,901 117,555 22,511
Farm net value added Euro 38,999 47,680 28,611 70,587 70,044 18,567
Farm net income Euro 27,205 33,057 12,643 61,299 18,692 16,719
GHG emission tonnes CO2 eq. 585.7 335.1 NA 624.0 597.3 130.8
GHG/cow tonnes CO2 per cow 7.88 7.84 NA 8.32 6.02 5.84
N-balance kg N per ha 431.1 162.4 NA 323.1 408.5 127.9
NUE output/input 6.2 6.6 NA 4.7 6.8 7.3
Advisory contacts number 25.0 13.0 21.0 11.0 24.5 21.0
Job satisfaction max 10 6.5 8.0 7.0 8.0 8.0 6.0
Quality of life max 10 5.9 8.0 6.2 7.8 7.2 6.0
Pesticides 2.1 0.4 1.6 NA 0.8 0.7
PS For illustration purposes only. These results (2015, France:
2014) are based on relatively small groups of farms which are not
necessarily representative for the whole country.

15. Experiences in data collection
• Some countries already collecting more data than
required by FADN – experienced
• Perceived importance of sustainability varies by MS
• Many FLINT variables already indirectly available in
FADN information flow (eg. quantities on invoices):
environmental accounting
– reduces information collected from farmers
• Problematic variables tend to be country specific
• Role of trusted data collector (and his know-how)
important
• Data collectors attitude changed from hesitant to more enthusiastic

16. • Policy researchers need to understand relation
between policy measure and farm management with
exact relation between inputs, outputs and income.
• Collecting these data on the same set of farms is
conceptually and empirically superior to a solution of
separate panels (as illustrated in some show cases)
• Collecting environmental data very often also
depends on systematic recording of flows:
environmental accounting is based on documents
also used in financial accounting. Reduces
administrative burden and increases quality
Basic recommendations

17. • (Environmental) statistics:
– Do not show relation between inputs, (bad) outputs and
income at the same farms. However, relation between
policies and farm management in impact analysis is needed
– Can have quality problems if not based on systematic
recording cross checked with financial accounts.
• Separate panel on environmental accounting (and
social indicators):
– Same quality problems
– Conceptually and empirically less good than 1 (FADN) panel
– Higher costs and total administrative burden, as much more
farmers are visited.
• Fits in Basic act of FADN
This leads to preference for FADN

18. 18
FLINT proposal to DG Agri: adapt FADN
1. CAP Reform and other
policies demand better data
for policy evaluation
2. Reduce current FADN from
85.000 farms to 75.000 to
cover costs of new data
3. Collect sustainability data on
15.000 farms

19. Detailed recommendations (1)
• Start collecting FLINT data
• Including FLINT data on all FADN farms would
increase total running costs with 40%.
• More feasible option to collect FLINT data on a
subsample of farms.
• Create FADN sub-sample of 15.000 farms on which
sustainability data are collected
• Distribution of 15.000 over member states based on
optimal allocation over the member states

21. Recommendations (2)
• Would increase operating costs of FADN (to be paid by
MS or EU?)
• Alternative solution within current budget limitations:
– Reduction of current sample of about 85.000 to
75.000 farms
• impact on quality of estimates (at EU and MS level)
of economic indicators very limited.
– Large differences between MS depending on level
of costs of current FADN data collection and
estimated costs of FLINT data collection
• Negotiation needed for countries like FR
and D where this applies

23. Recommendations (3)
• Setting up FLINT data collection requires investments
(software, instructions)
• DG-AGRI could support exchange of experiences and best
practises
• Recommendation to start a FLINT-2 project.
– Could start soon
– Make use of existing data in MS for policy analysis in 2018 (also from
FLINT partners who will continue their data collection)
– Transfer of FLINT knowledge to other MS, start testing
And:
– Connect with developments in IT and private sustainability schemes
– Share best practises and legal arrangements in use of administrative and
commercial data

24. In conclusion: where do we stand now ?
• Data on new indicators is clearly needed for impact assessment
of policies, to defend and improve the CAP
• Collecting the data in FADN has clear advantages over other
options (environmental statistics, separate panel)
• FLINT proved that data collection is feasible.
• FLINT showed how such data improves policy analysis.
• FADN can keep its relevance by innovating its data collection
• Next steps:
– Detailed plan with FADN managers on how many farms per country
depending on finance / acceptable reduction in number of FADN farms
– Adapt Farm Return
– Start data exchange (existing data) and data collection.

25. More
information
http://www.flint-fp7.eu/index.html
25
HansVrolijk:
hans.vrolijk@wur.nl
Krijn Poppe:
krijn.poppe@wur.nl

26. Relevance for Food Industry
• Sustainability reporting and sustainability management of
sourcing by the Food industry and Retail would benefit from
micro-level data
• Should data gathering and indicator standardisation be the
sole responsibility of the EuropeanCommission in FADN or
makes it sense to set up public-private initiatives or could the
private sector take the lead in sustainability reporting with
micro-data at a European level?
– creating a monitoring targeted to farm- and industry needs
• Would it make sense to involve research partners like
Wageningen Economic Research for know how and a third
trusted party between companies (and companies and
farmers) ?

27. Key message
The results of the FLINT
project provide an
excellent opportunity
for food industries like
dairy to install a
European monitoring of
their sustainability, to
report to the public and
governments,
benchmark companies
and incentivize farmers
27