Transcript Agricultural Labour in the EU - uni

Agricultural Labour in the EU
• Dyna-SPAT – continuation of CAPRI
(1997-2000) and CAP-STRAT (2001-2004)
• Galway team – continuation of two strands
of work: a) labour/holder projections from
CAPRI and b) estimations using FADN
data, from CAP-STRAT
CAPRI
• Main Papers to be followed up from:
• 1) A Regional Analysis of the Changing
Structure of Agricultural Land Holdings; by
Steele and Gaffney (98-05)
• 2) Notes on Agricultural Labour; by Steele
and Garvey (00-01)
1. WP 98-05
• Main Focus on Methodology of Cohort
Analysis e.g.
C 25 (1993)  C 24 (1992)2425  NA2425
Cohort Analysis gives Autonous
Change
• It is possible to then model NA, the nonautonomous change (the difference between
expected and actual changes)
• Some examples of results follow….
CAP-STRAT: also 2 papers
• 1) Estimation of Input Allocation
from EU Farm Accounting Data
using Generalized Maximum Entropy;
Britz and Garvey, 2002 (Valencia)
• 2) Integration of input estimates into the
CAPRI data base;
Britz, Adenäuer, Pfaffenzeller, Garvey,
McInerney, 2004 (Leuven-la-neuve)
Paper 1
• Implemented an Entropy Approach to Input
Estimations, incorporating a number of
FOC’s and constraints.
• Included Labour Coefficients.
• Unfinished business……
Paper 2
• Straightforward ‘non-system/univariate’
econometric estimations using the same
FADN data
• Estimates found for a number of sometimes
fairly aggregated inputs, not including
unpaid family labour (though wages in real
terms are included)
Reconciliation with EAA’s
• These econometric estimates (at national
level) are then reconciled with EAA’s, using
additional sources of information (in
particular, standard gross margins).
• Regionalization is possible based on (some
of) the econometric coefficients and the
regional SGMs. However, there is not full
econometric regionalization.
Next Steps…..
• 1) More detailed FADN estimations, with
emphasis on labour
• 2) Consider accession countries…….
• 3) Revisit Paper 1 of CAP-STRAT
• 4) Development of Cohort calculations –
but how to tie into CAPRI?
1) New FADN Estimations
• A) Incorporate regional slope coefficients
• B) Incorporate Labour (different types?),
with farm size multiplicative effects
• C) Include more disaggregated inputs where
possible
• D) Use more recent extraneous info. in
EAA reconciliations (e.g. SGMs)
2) Accession Countries’ Farm
Labour
• The double problem – projections and input
estimations. Presumably use can be made of
FADN type data for the latter.
• For the former, long term trends may have
set in by now – maybe not…….
3) Revisit Paper 1) CAP-STRAT
• Possible – but unlikely to have
time/resources. Nevertheless, some sort of
system approach might be possible. Issue of
incorporating economic behaviour into the
estimations is interesting.
4) Cohort Calculations, Other
Sectors and CAPRI
• Cohort projections give autonomous
changes in the number of holders. The
difference between this change and the
actual change ex post can be put down to
Economic factors mainly. These can be
estimated at some fairly aggregated level.
Another possibility ……
Some sort of dynamic Markov Chain
analysis
In either case the important thing is to
model a link between the wider
Economy and Farm Entry/Exit, and –
by implication – farm size.
Linking the two tasks…..
• Using some sort of link (probably farm size)
the wider economic model can link with
labour inputs e.g. non-Ag. GDP rises,
expected returns to non-Ag. work rise,
people leave agriculture, average farm size
rises. This means – linking to L inputs - that
labour used per activity (almost certainly)
falls and farm income per capita rises.
That’s it!