In ‘Fighting for Birds’, and also on his blog, Mark
Avery expresses frustration at a reluctance amongst farmers to take advantage
of funding for skylark patches ( 16m2 areas of arable land left
unfarmed) under the Entry Level Stewardship Scheme. Having recently read the fantastic ‘Thinking,
fast and slow’ by Daniel Kahneman, ‘Nudge’ (Thaler and Sunstein) and
‘Predictably irrational’ (Dan Ariely) I am convinced that conservation can
learn from behavioural economics, the study of the ways in which individuals
act irrationally (from the perspective of classical economics). In this blog I will outline a series of
experiments to identify whether farmers are more concerned with maximising
their profits or the size of their farm’s yield. I will also present an idea for a way to
address individuals’ overestimations of the size of skylark patches (if this is
a problem as Mark Avery suggests).
Behavioural economics is based on the observation that
humans are not economically rational. An
economically rational individual would always act to maximise their utility (or
overall wellbeing). In the case of the
farmer, making the assumption that farmers attribute no benefit to increased on
farm biodiversity resulting from the incorporation of skylark patches, a
rational farmer would incorporate skylark patches if they increased his (or
her) (predicted) profits. It has been
shown that skylark patches do increase profits, see Mark Avery’s
blog. So why, when presented with
this evidence, do all farmers not rush to incorporate skylark patches?
The first possibility is that farmers are not interested in
maximising their profits. Instead they
may be
interested in either A)maximising the area of land which they use to
produce crops or B) they may evaluate the success of their farming efforts by
calculating the total yield of the farm each year. If farming is a vocation and farmers care
about producing crops then this would make sense. To test between these hypothesises the
following scenarios would be used; each scenario concerns a different fictional
farm.
|
Scenario
|
Area (hectares)
|
Yield (tonnes Per Hectare)
|
Total Yield
|
Profit per tonne (£)
|
Total Profit (£)
|
|
1
|
40
|
8
|
320
|
400
|
128000
|
|
2
|
40
|
9
|
360
|
350
|
126000
|
|
3
|
45
|
8
|
360
|
350
|
126000
|
|
4
|
35
|
8
|
280
|
450
|
126000
|
|
5
|
40
|
7
|
280
|
450
|
126000
|
|
6
|
35
|
9
|
315
|
400
|
126000
|
|
7
|
45
|
7
|
315
|
400
|
126000
|
Farmers would be shown scenarios in pairs or threes (with and
without the ‘Total Profit’ column, removed in different experiments) and would
be asked which of the farms they would prefer to be theirs. It would be explained that they will not be
able to alter the yield, which reflects the quality of the land or the profit
per tonnes which reflects a deal with a local crop buyer which cannot be
altered. An economically rational
farmer, if presented with all scenarios would prefer scenario 1 to every other
scenario but would not have no preference between other scenarios.
Experiment A: the farmer would be offered scenario 1 and 2
(in a randomised order e.g. scenario 1 presented first 50% of the time). If the farmer chose scenario two then this
would signal that they would rather run a farm with higher yields. Next scenario 1 and 3 would be
presented. If a farmer chose option 3
then this would signal that they would rather run a farm which had a larger area
producing crops. Both options 2 and 3
offer the same yield which is greater than that offered by scenario 1 but offer
a lower total profit. If farmers attempt
to maximise the quantity of crop they produce they should prefer scenarios 2
and 3 over scenario 1.
Experiment B and C: individuals would be presented with
scenario 1, 4 and 5 and then 1, 6 and 7.
This time individuals would be asked to rank the scenarios in order of
preference (with the option of 2 or more scenarios being equally
preferable). If individual preferred 4
to 5 and 6 to 7 then it would be possible to conclude that individuals would
rather maximise their yield than the area of their farm and vice versa for
individuals who preferred 5 over 4 and 7 over 6.
Experiment D: farmers would be given the option of the
following 4 farms and would be asked to rank them from highest to lowest preference
with the option of applying equal preference to 2 or more scenarios.
|
Scenario
|
Area (hectares)
|
Yield (tonnes Per Hectare)
|
Total yield
|
Profit per tonne
|
Total Profit
|
|
A
|
40
|
8
|
320
|
400
|
128000
|
|
B
|
40
|
8
|
320
|
450
|
144000
|
|
C
|
40
|
9
|
360
|
400
|
144000
|
|
D
|
45
|
8
|
360
|
400
|
144000
|
These two sets of experiments would make it possible to see
if farmers prefer to maximise the quantity of crops they produce when total
profits are equal and, if this is the case, if farmers prefer to do this by
maximising the area of land they farm or by maximising their yield per
area. If farmers do prefer to maximise
the area they farm then this will help to explain reluctance to take up skylark
patches. The inclusion of skylark
patches means a reduction in area farmed.
These experiments could be extended to offer scenarios in which the
total area farmed, or yield per area, were increased but total profit decreases
due to a decrease in profit per unit of crop.
If farmers do want to maximise the quantity of crops produced
then conservation can learn from Thaler and Benartzi’s ‘Save
more tomorrow’ scheme. Thaler and
Benartzi recognised that individuals should be saving more for their pensions
than they were but did not like to see a drop in their paychecks. The authors trialled an arrangement whereby
upon receiving a payrise workers increased their monthly contribution to their
pension (without seeing their paychecks shrink). The scheme was highly successful but how does
it apply to farmers? Once the currency
that farmers use to evaluate the wellbeing they gain from farming (area of
farm, yield per area, profit per tonne of crop or total profit) has been
identified then skylark patch scheme can advertised so that individuals
increase the number of skylark patches when this does not result in a decrease
to the relevant aspect of their farming.
For example, if farmers are most concerned about the size of their farm
then skylark patch uptake may be highest when targeted at individuals
increasing the total area of their farm.
Mark
Avery also writes that the patches look as if they occupy more space than they
do. I suggest offering farmers the
feedback they need to improve their estimation of the area of skylark
patches. This could be achieved as
follows: Using aerial photographs or the
following type of diagrams (with labelled axis), ask individuals to estimate
the size of the red squares and the proportion of the total (blue) area which
they occupy:
If mark is right then individuals will over estimate the
proportion of the total area occupied by the red shapes, if I am right then
repeated feedback (providing individuals with correct answers instantly after they
make their estimations) will result in individuals’ skills increasing. Subsequently, individuals will be better able
to imagine the difference that 0.5, 1 or 5% of their land being used as skylark
patches will really look like.
Other thoughts:
People don’t deal with percentages rationally and
overestimate small percentages.
Individuals should be expected (on the basis of current research) to be
more in favour of incorporating skylark patches when they are advertised as
‘leaving 98% of farm land area as arable' than as ‘taking only 2% of total farm land area').
