Pollution and Local Economies & Organic Farming
clw0dq8
WeekFiveOrganic_farming.pdf
2010, Vol 11, �o1 69
A comparison of the economic and environmental
performances of conventional and organic farming:
evidence from financial statements
Josep Mª. Argilés and �éstor Duch Brown ∗
Abstract
While conventional farming systems face serious problems of sustainability, organic
agriculture is seen as a more environmentally friendly system since it favours renew-
able resources, recycles nutrients, uses the environment’s own systems for controlling
pests and diseases, sustains ecosystems, protects soils, and reduces pollution. At the
same time organic farming promotes animal welfare, the use of natural foodstuffs,
product diversity and the avoidance of waste, among other practices. However, the fu-
ture of organic agriculture will depend on its economic viability and on the determina-
tion shown by governments to protect these practices. This paper performs panel re-
gressions with a sample of Catalan farms (Spain) to test the influence of organic farm-
ing on farm output, costs and incomes. It analyses the cost structures of both types of
farming and comments on their social and environmental performance.
Keywords: organic farming, conventional farming, social/environmental/financial
performance, social and environmental accounting
JEL Classification: Q01, Q12, Q51, M41
Introduction
Over the last few decades world agriculture has introduced increasing levels of mod-
ernization and productivity. Key factors in this evolution of modern, or conventional,
farming have been intensive capital endowments, farming specialization, the wide-scale
application of chemical fertilizers and nutrients and the selection of high-yield crops and
livestock, including genetically modified organisms in some countries.
In spite of these recent advances, intensive farming systems face a number of serious
problems: the declining effectiveness of land, pesticides and chemical fertilizers, the
ongoing loss of biodiversity, environmental and health risks, economic and social costs,
as well as various kinds of unpredictable future risks (Matson et al. 1997; Altieri 1998;
Boschma et al. 2001; Tilman 1998; Drinkwater et al. 1998). In particular, Dupraz
(1997), Mishra et al. (1999), Hornbaker et al. (1989), Kurosaki (1997), Popp and Rud-
strom (2000) and Omamo (1998) have highlighted the economic problems that arise
from specialization and monoculture. More specifically, Melfou and Papanagioutou
∗
Josep Mª. Argilés: University of Barcelona (Department of Accounting)
Néstor Duch Brown: University of Barcelona (Dep. Econometria, Estadística i Economia Espanyola)
and IEB
70 AGRICULTURAL ECO�OMICS REVIEW
(2003) measured the effect of nitrate pollution on the growth rate of total factor produc-
tivity in Greek agriculture, while Pretty et al. (2000, 2001) assessed a wide array of ex-
ternalities of modern agriculture in the UK, USA and Germany with aggregated data.
A growing interest in environmentally friendly goods and services has been ex-
pressed together with concerns for the risks, and broader environmental problems, asso-
ciated with intensive agriculture. These issues were all central concerns at the World
Summit on Sustainable Development held in Johannesburg in September 2002. In a
recent survey (European Commission 2005), citizens of the European Union (EU)
claimed that their main priorities for agricultural policy were, in order of importance:
ensuring stable and adequate incomes for farmers (36%), ensuring that agricultural
products are healthy and safe (30%), promoting respect for the environment (28%), fa-
vouring and improving life in the countryside (26%) and favouring organic production
(20%).
Organic agriculture is seen as the most environmentally friendly farming system. It
favours renewable resources, recycles nutrients, uses the environment’s own systems
for controlling pests and diseases, sustains ecosystems, protects soil, reduces pollution,
while at the same time it promotes animal welfare, the use of natural foodstuffs, product
diversity, avoidance of waste, etc. (European Commission 2002). Within the European
Union, environmental concerns form a major part of the Common Agricultural Policy
(CAP), which actively promotes organic agriculture.
Darnhofer et al. (2005) identified a cluster of committed organic farmers to whom
economic considerations were secondary. However, an increasing number of organic, or
potentially organic, farmers are tending to emphasize economic concerns (Lund et al.
2002). Rigby et al. (2001) suggest that the development of organic agriculture will de-
pend on its economic viability and the determination shown by the CAP to protect this
type of farming.
It is assumed that organic farming provides lower crop yields than intensive farming
(Trewavas 1999, 2001), that it is economically disadvantageous and that it requires
government financial support, because the premiums consumers are prepared to pay for
organic food (Gil et al. 2000; Mahenc 2007; Batte et al. 2007) are insufficient to ensure
that organic farming practices become more widespread (Rigby et al. 2001).
However, virtually no studies have been undertaken examining the economic viabil-
ity of organic farming. Tzouvelekas et al. (2001) found lower technical efficiency scores
in Greek organic cotton farms vis-à-vis their conventional counterparts, while Lansink
et al. (2001) found that Finnish organic farms are, on average, more efficient in relation
to their own technology, but that they use lower production technology than conven-
tional farms, thus resulting in approximately 40 per cent less productivity. Lansink and
Jensma (2003) found larger variable profit in organic than in conventional Dutch farms,
as well as interesting conclusions regarding the trends in organic farming practises. Un-
fortunately, they do not offer information about bottom line profits. Dima and Odero
(1997) studied a sample of Kenyan farms and found that maximum yields can be ob-
tained from a combination of organic manure and chemical fertilizers. Descriptive sta-
tistics presented by Offermann and Nieberg (2000), typically drawn from small sam-
ples, do not offer tests and inferences applicable to the population of farms. Dobbs and
Smolik (1996) found that a conventional corn and soybean farm was more profitable
than a corresponding organic farm during most years in an 8-year time period. Kerse-
2010, Vol 11, �o1 71
laers et al. (2007) reported the economic potential of converting to organic farming, but
due to a lack of available data their estimations had to be based on simulations. Pretty et
al. (2005) assessed the financial and environmental costs of conventional and organic
agriculture in the UK with aggregated data.
Our study constitutes an empirical analysis of the respective economic performances
of organic and conventional farming. Its main contribution to the literature lies in its
integration of five elements. Thus, first, the study operates at the individual farm level
rather than at the aggregated or average farm level and, secondly, it uses real data rather
than normative and simulation approaches. Third, unlike previous studies in the field, it
draws statistical inferences about the influence of organic and transitional farming on
financial costs, output and bottom line profit. Fourth, the study covers a broad spectrum
of farm productions and, finally, it analyses environmental and social issues on the basis
of available financial data.
The following section deals with the model specification and a description of our
sample, the third section presents and explains our results, while our concluding re-
marks are outlined in the final section.
Research design
Model specification
Our analysis emphasises the ability of farms to generate net revenue, in accordance
with their specific characteristics, from the range of activities that they practise. First, a
revenue function is defined, so that a farm’s output can be approximated and analysed.
Second, we define a cost function to study the respective efficiencies of organic and
conventional farming. Finally, we combine these two functions in a net revenue (or
profit) function to analyse jointly the characteristics generating profitability in the dif-
ferent types of farming.
Algebraically, gross revenue from agricultural production activity can be represented
in terms of variable inputs (x):
{ } )1(0);(max),( >∈= pxYqpqxpR q
where p is a m-dimensional vector of positive output prices and Y(x) refers to the pro-
ducible output set with q being physical output. Provided that certain conditions are
satisfied, this revenue function allows us to analyse whether different farm characteris-
tics have a differentiated impact on farm revenues.
Similarly, given the respective nature of conventional and organic farming, it seems
logical to assume that a further source of differences between them should be derived
from their different costs of production. Formally, the cost function can be defined as:
{ } )2(0);(min),( 0
>∈= ≥
wqVxwxqwC x
where w is a n-dimensional vector of positive input prices, wx is the inner product
(∑ i
ii xw ) and V(q) the input requirement set. Again, given the assumptions that under-
lie the existence of costs functions and the fulfilment of their properties, it is possible to
72 AGRICULTURAL ECO�OMICS REVIEW
study the set of characteristics that have most influence on costs of production for con-
ventional and organic farms.
Finally, the difference between revenues and costs gives some measure of profitabil-
ity. Profit maximization can be separated into two stages. The first stage, which can be
considered as the short run, involves maximizing profit for a given output. In the long
run (second stage), output has to be chosen to maximize profits. When output is fixed,
revenue is also fixed and profits are maximized by minimizing costs. Hence, fixed out-
put profit maximization yields the same input configuration as cost minimization. In a
multi-output setting such as this, it is possible to use the properties of the revenue func-
tion or the cost function to infer the properties of the profit function, but it is better to
link the properties of the profit function directly from the production possibility frontier
(T). Thus, from (1) and (2) we have:
{ } )3(0,;),(max),( ,
>∈−=Π pwTqxwxpqwp xq
These three equations form the basis of our empirical analysis. There are several
characteristics that should be controlled for, besides the fact that production units prac-
tise conventional or organic farming. These include unit size, type of crop, location, the
existence of irrigation systems, etcetera. In order not to impose more restrictions by
means of certain ad-hoc functional forms, our main interest is in detecting the character-
istics that lead to production, cost and profit differences between conventional and or-
ganic farms. Assuming further that farms are price takers in the product markets, that
factor markets are perfectly competitive and that farms are identical in all respects ex-
cept regarding whether they adopt organic or conventional farming practices, then the
gross revenue function we estimate below can be described by:
)4(),,,,(),( LFISORxpR =
where R represents gross revenues from the farming operation, which depends on the
fact that farms perform organic production systems (O), their size (S), irrigation of land
(I), the type of farming they perform (F) and, finally, their location (L). In the same
vein, assuming perfect competition in both factor and product markets, as well as iden-
tical input requirement sets for all firms, the cost function can be defined as:
)5(),,,,(),( LFISOCqwC =
Similarly, the profit function derived directly from the revenue and cost functions
can be expressed as:
)6(),,,,(),(),(),( LFISOPqwCxpRxpP =−=
Thus, the following reduced-form multiple regression model can be used to estimate
the influence of organic farming on farm output 1 , costs and profits:
0
1 1 1
ln ln ln
= = =
= + ◊ + ◊ + ◊ + ◊ + ◊ +Â Â Â jk m
i Ok ki S Ii Fji j Lm mi i
k j m
Y β β O β S β I β F β L ε (7)
The dependent variable Y represents the performance of farm i. The model seeks to
study the influence of organic farming on farm performance, controlling also for farm
2010, Vol 11, �o1 73
characteristics, such as size, irrigated area, type of production farming and geographical
location, which are all likely to affect empirical results when using a heterogeneous
sample of farm data.
Two dummy variables indicate whether a farm i performs, on the one hand, organic
farming (ORGA�IC) or, on the other, partly organic or in transition to organic farming
(ORGTRA�S), when their value is equal to one (or zero otherwise), while the default
variable corresponds to conventional farms.
As conventional farming is more intensive and not concerned with crop rotation and
land rest, it is expected to be more productive in terms of physical production. However,
organic farming tends to compensate for this through higher quality and the subse-
quently higher prices it can command. Consequently, no prior hypothesis can be formu-
lated with respect to the relation between organic farming and monetary output. In the
specific case of ORGTRA�S, lower output is expected for farms in transition to organic
farming, because, in line with European regulations, farms must cease production for
two to three years before they can label their produce as organic. However, the fact that
this category includes transitional and partly organic farming does not allow us to for-
mulate a definitive prior hypothesis.
Organic farming tends to avoid input waste and saturation, saving on chemicals, fer-
tilizers, medicines for livestock, etc. Indeed, it employs its own farm resources more
frequently than conventional farming. However, as it does not use resources intensively,
with the effect that yields are lower than in conventional farming, we would expect the
ratio of input-to-output to be higher. Additionally, higher costs would correspond to
higher product quality. Finally, organic farming usually requires more work than con-
ventional farming. Controlling for other variables, and specifically size, no clear hy-
pothesis can be formulated with respect to the influence of organic farming on costs.
Consequently, no prior hypothesis for profits can be proffered, though assumptions and
existing research seem to suggest that organic farms will record lower incomes or prof-
its than those reported by conventional farms. Farms in transition to organic farming are
required by existing EU regulations to implement 2-3 years of land rest. However, the
fact that our data base does not distinguish between transitional and partly performing
organic farming, no conclusive hypothesis can be made with respect to ORGTRA�S and
their costs and profits.
Size is an obvious control variable in the model, as we would expect bigger farms to
have a higher output. The European size unit (ESU) is the accepted, and widely used,
measure of size in EU agricultural statistics. ESU defines the economic size of an agri-
cultural holding on the basis of its potential gross added value. It is calculated by as-
signing predetermined values of gross added value to the different lines of farm produc-
tion. Since 1995 one ESU has been fixed at 1,200 ECU of standard gross margin. This
standardized measure of size is homogeneous for different types of farming.
Dry weather and water shortages handicap farming, especially in Mediterranean
countries, as they tend to limit farms to just a few types of farming and to reduce farm
productivity. Irrigation yields more productive crops and is usually devoted to more
intensive, productive crops. Thus, the percentage of irrigated utilized agricultural area
(PERCIRRIGUAA) is hypothesized as being associated with greater output and higher
costs - because such land is typically used for more intensive farming, and higher profits
- because it allows most profitable farming opportunities to be chosen. The variable
74 AGRICULTURAL ECO�OMICS REVIEW
cannot be transformed into logarithms because of zero values.
According to the methodology of the Farm Accountancy Data Network (FADN),
seven dummy variables can be employed to indicate that a farm operates the corre-
sponding type of production farming when these variables are equal to one, and zero
otherwise: FIELD for farms predominantly specialized in cereals, general field exten-
sive or mixed crops, HORTICULTURE for farms specialized in horticulture, PERMA-
�E�T for farms predominantly specialized in fruits, citrus, olives, wine or combined
permanent crops, MILK for farms specialized in dairying, GRAZING for farms special-
ized in rearing or fattening cattle, sheep, goats and other grazing livestock,
GRA�IVORE for farms predominantly specialized in pigs and poultry, while mixed
livestock and/or crop type of farming (sometimes combining with various livestock and
crops) is the default category. In the geographical context of our sample, where water
shortages and dry weather are frequent, agricultural land is very scarce and livestock is
usually reared on capital-intensive farms. Mixed livestock farms are expected to face
higher costs (and production) than farms with predominantly field and permanent crops
and those with extensive grazing livestock, while mixed farms should face lower costs
(and production) than those specialized in more intensive agriculture, such as horticul-
ture, dairy and granivores. No conclusive hypothesis could be formulated with respect
to profits by production type.
Two dummy variables indicate whether a farm is located in less-favoured (LESSFA-
ZO�E) or mountain zones (MOU�TZO�E) when their values equal one (and zero oth-
erwise), while the default category applies to farms located in what are labelled “usual
zones”. The former are usually located at some distance from consumer and purchasing
markets and have lower technological, infrastructure and service endowments. Farming
in such locations is usually handicapped by climate conditions and location opportuni-
ties. Higher outputs are expected from farms located in “usual zones” than from those in
mountain or less-favoured zones. No conclusive hypothesis can be formulated with re-
spect to costs because, on the one hand, less-favoured and mountain-located farms enjoy
lower prices for some inputs (work, land rent, etc.), whereas, on the other, they have
more restricted access to services and technological facilities.
Equation (7) is expressed in the following full equation that tests the influence of
organic farming on farm output (OUTPUT):
[ ] [ ]i 0 1 i 2 i 3 i 4 i 5 i 6 i 7 i
8 i 9 i 10 i 11 i
12 i i
ln OUTPUT β β ORGA�IC β ORGTRA�S β ln ESU
β PERCIRRIGUAA β FIELD β HORTICULTURE β PERMA�E�T
β MILK β GRAZI�G β GRA�IVOR β MOU�TZO�E
β LESSFAZO�E ε ( 8 )
= + + + +
+ + + +
+ + +
+ +
In order to test the influence of organic farming on costs we took data for registered
costs (REGCOST) from the FADN for farm i. PROFITREG indicates the difference
between output and registered costs for farm i. However, Schmitt (1991) recognised that
agriculture is still predominantly centred around family farms in advanced western
economies, and consequently family work constitutes an important share of total work
on farms. Various authors (e.g. Hopkins and Heady 1982; Bublot 1990) have discussed
the need, therefore, to include family work in farm costs, and have suggested a number
of valuation methods. FADN provides data about the amount of work expended on the
2010, Vol 11, �o1 75
farm (expressed in annual work units), distinguishing that proportion which corresponds
to the work put in by the members of the family, but it considers only those costs that
correspond to non-family work. Thus, although the need to include family work in cost
valuation is widely recognized, FADN usually fails to do so. Each year the Spanish
Ministry for Agriculture publishes the reference income that a farmer would earn in an
alternative job. In this way, we calculated the opportunity cost of the work put in by the
family and added it to the registered costs so as to obtain the total costs of the farm
(TOTALCOST) and the subsequent income in absolute (PROFITTOTALCOST).
In the traditional model, cost behaviour is dependent on activity. As output is the most
common measure for activity, costs can be expected to be positively influenced by out-
put. Costs are described as being either fixed or variable with respect to changes in ac-
tivity. It is widely assumed that variable costs change in proportion to changes in activ-
ity, while fixed costs, which remain invariable in the short term, are also related to
changes in activity in the long term. Thus, we can expect costs, and profits, to be posi-
tively influenced by output.
When the dependent variable is costs or profits, the full model of equation (7) can be
expressed as:
[ ]i 0 1 i 2 i 3 i 4 i
5 i 6 i 7 i 8 i
9 i 10 i 11 i 12 i i
lnφ β β ORGA�IC β ORGTRA�S β ln OUTPUT β PERCIRRIGUAA
β FIELD β HORTICULTURE β PERMA�E�T β MILK
β GRAZI�G β GRA�IVOR β MOU�TZO�E β LESSFAZO�E ε (9 )
= + + + + +
+ + + +
+ + + +
where the dependent variable φ symbolises performance with respect to costs and prof-
its.
Here, we also perform an analysis of the cost structure of organic and conventional
farms, drawing conclusions about their respective social and environmental perform-
ances. FADN classifies costs as: specific, farming overheads, depreciation, external
factors and taxes. The European Commission (1997, 1998) provides a detailed classifi-
cation of these costs. Specific costs include seeds and seedlings, fertilizers, crop protec-
tion products, feed and feedstuffs for livestock, medicines, veterinary fees and other
specific crop, livestock and forestry costs. Farming overheads correspond to supply
costs linked to productive activity but not linked to specific lines of production. They
include energy, machinery and building current costs, as well as costs linked to work
carried out by contractors and to the hire of machinery, water, insurance, accountants’
fees, telephone charges, etc.
However, because of their impact on the environment we decided to separate energy
costs from other overhead costs. Depreciation is determined on the basis of the replace-
ment value and is concerned with plantations of permanent crops, farm buildings and
fixed equipment, land improvements, machinery and equipment and forest plantations.
External factors represent the remuneration of inputs (work, land and capital) that are
not the property of the holder. Here, we chose to analyse wages separately from rent and
interest. “Taxes” refers to the value added tax (VAT) balance on current operations,
when the special agricultural VAT system applies, as well as farm taxes and other
charges on land and buildings. It does not include taxes on farm profits.
76 AGRICULTURAL ECO�OMICS REVIEW
Data collection and sample.
The farm accountancy data network (FADN) was created in 1965 by Regulation
(EEC) 79/65 of the Council under the Common Agricultural Policy (CAP). Today,
FADN collects accounting information at the level of individual farms, and every year it
gathers data from a rotating sample of professional farms across all member states.
FADN data is collected through a questionnaire, called the “Farm Return”, which is
filled out by the farms with the assistance of specialised local accounting offices. The
information obtained through the Farm Return is coded and transmitted to the European
Commission. The information is summarised in reports similar to balance sheets and
income statements and published by the European Commission in aggregated terms.
The European Commission (1997, 1998) provides detailed information about its pro-
cedures and methodology. FADN was conceived as a complementary source of statisti-
cal information about farm income for policy makers, and the sample of farms from
which the data is obtained should be representative of a range of characteristics and
types of farming in European agriculture. Since 2000 data on organic farming in the
European Union have been collected. Every participating farm must present information
according to one of three possible codes: partly organic or in transition to organic farm-
ing (code 3), exclusively organic farming (code 2) and non-organic farming (code 1).
As can be seen, no distinction is drawn between farms in transition to organic and farms
performing partly organic and conventional farming.
The Catalan Government provided data from its Xarxa Comptable Agrària de Cata-
lunya (XCAC), the Catalan subsidiary of the FADN, for the year 2000, the first year in
which data on organic farming was available, to 2003. From the overall unbalanced
records of 1,556 farm-years, 1,414 practised non-organic farming, 97 were partly or-
ganic or in transition to organic farming and only 45 farms were exclusively engaged in
organic farming. This proportion of organic farms, however, is even larger than that
found in Spanish agriculture 2 . We deflated these data to 2000 values using a gross do-
mestic product deflator.
Table 1 shows our descriptive data sample. As can be seen from the univariate analy-
sis, organic farms obtain a higher output, generate more costs and are larger. At the
same time they were found to use a smaller agricultural area, but recorded a higher per-
centage of irrigated area. The significantly lower amount of subsidies available for or-
ganic farming indicates that Spanish authorities are not fully committed to organic agri-
culture, and that there are more important targets than organic agriculture for subsidies.
Table 2 shows a low Pearson’s correlation between the continuous independent vari-
ables, giving an initial indication that collinearity does probably not affect estimations.
Empirical results
Variance inflation factors, condition indexes and variance proportions of variables
suggest that multicollinearity is unlikely to affect estimations. As our sample presents
the typical autocorrelation pattern for independent variables throughout the period stud-
ied, we performed various panel regression estimations correcting for autocorrelation
disturbances. Thus, the estimation method assumes disturbances to be heteroscedastic
and contemporaneously correlated across panels. The commonly used Hausman test
(Hsiao 2005) rejected the null hypothesis of no correlation between individual effects
2010, Vol 11, �o1 77
and explanatory variables. As individual effects are correlated with the regressors in all
estimations, the random effects estimator is inconsistent, while the fixed effects estima-
tor is consistent and efficient. We therefore performed panel data estimations with fixed
effects to test the influence of organic farming on performance.
Table 1. Descriptive statistics: mean values (monetary values in €)
Conventional
farming
Only organic farming
Transitional or partly organic
farming Total
Number of farm-year observations 1,414.00 45.00 97.00 1,556.00 Farm output (OUTPUT) 74,097.79 102,210.90 49,159.71 73,356.21 *** Registered costs (REGCOST) 61,288.44 82,305.59 38,708.99 60,488.67 *** Profit with registered costs (PROFITREGCOST) 12,809.35 19,905.29 10,450.72 12,867.53 Total costs including family work (TOTALCOSTS) 85,423.17 105,364.00 58,571.26 84,325.94 *** Profit with total costs (PROFITTOTALCOST) -11,325.38 -3,153.10 -9,411.54 -10,969.73 Family farm income (PROFITREGCOST + subsidies) 22,376.44 25,218.06 17,444.35 22,151.16 PROFITTOTALCOST + subsidies -1,758.30 2,159.67 -2,417.92 -1,686.11 Current subsidies 8,624.63 5,312.76 6,471.10 8,394.60 *** Investment subsidies 942.46 0.00 522.53 889.02 Livestock units 102.71 31.43 86.37 99.63 Utilized agricultural area (ha.) 36.07 28.47 28.39 35.37 * Percent of irrigated area (PERCIRRIGUAA) 39.76 64.39 67.43 42.20 *** Economic Size Units (ESU) 29.69 45.38 33.64 30.39 ***
Notes: Significance levels: *p<0.1, **p<0.05 and ***p<0.01
Table 2. Pearson correlations between continuous independent variables
PERCIRRIGUAA ln[ESU] 0.1738 ESU 0.1476 OUTPUT -0.0487 ln[OUTPUT] 0.0077
Table 3 displays these results. Estimations in column A, corresponding to farm out-
put, show significant expected coefficients for farm size and field crops with p<0.01,
and for horticulture and permanent crops with p<0.1. A significant positive sign for or-
ganic farming suggests that farmers obtain a premium price from the market which fully
exceeds the lower amounts of physical output. The negative sign for farms in transition
or partly performing organic farming is not significant with p<0.1.
78 AGRICULTURAL ECO�OMICS REVIEW
Table 3. Estimations relating organic farming to output, costs and income (t-statistics in
parentheses).
Variables (A)
ln[OUTPUT] (B)
ln[REGCOST]
(C)1 PROFITREG-
COST
(D) ln[TOTALCOS
TS]
(E)1 PROFITTO- TALCOST
Constant 9.164686 (48.90)
*** 6.899933 (25.06)
*** -21891.61 (-5.21)
*** 8.525886 (50.37)
*** -46112.85 (-10.90)
***
ORGA�IC .3086983 (2.19)
** .0720551 (0.61)
-1201.013 (-0.21)
-.0023101 (-0.03)
2643.53 (0.45)
ORGTRA�S -.0053656 (-0.11)
.0013193 (0.03)
1240.564 (0.62)
-.0128886 (-0.52)
1925.885 (0.95)
Control
variables:
ln[ESU] .4734719 (9.67)
***
ln[OUTPUT] .3376585 (13.75)
*** .4114774 (28.21)
*** .2417884 (16.01)
*** .4079846 (27.75)
***
PERCIRRIGUAA .0018293 (1.26)
-.0005486 (-0.46)
9.294989 (0.16)
-.0002824 (-0.38)
8.880331 (0.15)
FIELD -.1781938 (-2.81)
*** -.2218203 (-4.28)
*** 6135.572 (2.40)
** -.1553832 (-4.87)
*** 6886.17 (2.67)
***
HORTICULTURE -.3745642 (-1.78)
* -.2253169 (-1.32)
11335.47 (1.34)
-.1240358 (-1.18)
9535.014 (1.12)
PERMA�E�T -.1620507 (-1.91)
* -.2629831 (-3.72)
*** 6590.053 (1.89)
* -.1444926 (-3.33)
*** 6513.575 (1.86)
*
MILK -.0654425 (-0.46)
.1099518 (0.94)
6213.646 (1.07)
.0694591 (0.96)
5062.994 (0.87)
GRAZI�G -.1052164 (-0.86)
.0483861 (0.48)
-4852.8 (-0.97)
.0500469 (0.80)
-5448.338 (-1.08)
GRA�IVOR .0533478 (0.66)
-.0586958 (-0.89)
-1749.237 (-0.53)
-.0084159 (-0.21)
-1695.235 (-0.51)
MOU�TZO�E .2704225 (1.30)
-.1135267 (-0.66)
10648.65 (1.25)
-.1214467 (-1.15)
11861.25 (1.38)
LESSFAZO�E .0314435 (0.28)
.096176 (1.03)
-3526.773 (-0.77)
.0350932 (0.61)
-3075.574 (-0.66)
R-square
(overall):
0.5753 0.7499 0.3166 0.7656 0.1683
F 12.20 *** 20.57 *** 68.66 *** 27.35 *** 66.63 ***
Notes:
Significance levels: *p<0.1, **p<0.05 and ***p<0.01
1. Untransformed dependent and independent variable OUTPUT, because logarithms cannot be
calculated for negative values of I�COME.
2010, Vol 11, �o1 79
Columns (B) and (D) display estimations for farm costs, while columns (C) and (E)
do the same for profits. These coefficients indicate that organic and transitional farming
do not significantly influence farm costs or profits. Estimations suggest that the higher
charges associated with organic farming, compared to those associated with conven-
tional farming, are balanced by input savings. OUTPUT, the most influential variable,
and dummies for farms specialised in field and permanent crops present the expected
significant signs with p<0.01 for costs. As expected, OUTPUT influences profits posi-
tively and significantly with p<0.01. Farms specialized in permanent crops have a sig-
nificant influence on higher profits with p<0.1, as do those specialised in field crops,
with p<0.05 and p<0.01 with respect to profits with registered costs and total costs re-
spectively. The remaining dummies for farm specialization and location, as well as the
variable for the percentage of irrigated land, do not present significant signs with p<0.1.
The fact that no significant sign was found in the case of output, costs, or profits for
transitional farms suggests that farmers attempt to make a gradual conversion to organic
farming, and initially combine organic practices with conventional farming.
We also performed regressions for profits including subsidies, calculated both with reg-
istered costs, and including opportunity costs. These results (not shown here) were simi-
lar to those included in columns (C) and (E) of Table 3. While output is the most influ-
ential variable, organic and transitional and partly organic farming do not significantly
influence farm incomes (subsidies included). Subsidies are mainly influenced by geo-
graphical location and the type of farming production. In terms of support for organic
agriculture, European policies should complement those initiated by national govern-
ments. Though the CAP seeks to promote organic agriculture, unlike other European
governments, Spain does not emphasize measures that protect sustainable agriculture.
The only remaining significant (with p<0.05) control variable was the dummy for farms
specialised in field crops.
It might be argued that the sample of conventional farms includes a number of small,
backward farms with ageing farmers and/or farmers with no expectations of continuing
operations in the near future. We would expect these farms to be poor performers, and
so any comparisons between organic and conventional farms should not include these
farms. Regressions performed excluding the 5 th
and the 10 th
percentiles of the smallest
farms yielded very similar results (not shown here) to those in Table 3, thereby demon-
strating that our results are not biased by the small, backward non-viable farms.
The cost structure of farms is shown in Table 4. Although Table 3 does not display
any significant influence of organic farming on farm costs, there are significant differ-
ences in cost structure between conventional, organic and transitional or partly organic
farming. When the three farming types are considered, significant differences are found
for energy, other overheads, depreciation, salaries and rent and interests. Overlooking
the special circumstances of transitional farming, it is interesting to note that organic
farming has significantly higher wage costs than those reported for conventional farm-
ing, but significantly lower specific and energy costs. Organic farming relies less on
chemical and mechanical procedures than conventional farming but, by contrast, it uses
more labour and generates more employment. Likewise as its operations are less de-
pendent on machines, it consumes less diesel oil. As it recycles nutrients and uses the
environment’s own systems for controlling pests and diseases, it spends less on fungi-
cides, insecticides, chemical-based fertilizers and crop protectors, purchased feedstuff
80 AGRICULTURAL ECO�OMICS REVIEW
and medicines for livestock. Consequently, specific costs and energy consumption are
lower in organic farming, while wages paid are higher.
Table 5 displays more detailed information about specific costs, work use and energy
consumption. In order to avoid any misleading information caused by extreme values,
we also performed tests with median values. In this instance, organic farming almost
doubled its mean value of annual work with respect to that of conventional farming. The
difference was even more marked for hired work. Although no significant differences
were found for the mean values of the percentage of energy and specific costs to output
Table 4. Cost structure of registered costs for conventional, organic and transitional or
partly organic farming type (in percent of total registered costs)
C o
st s
tr u
c tu
re
C o
n v
e n
ti o
n a l
fa rm
- in
g
O n
ly o
rg a n
ic f
a rm
- in
g
T ra
n si
ti o
n a l
o r
p a rt
ly o
rg a n
ic f
a rm
- in
g
T o
ta l
S ig
n if
ic a n t
d if
fe r-
e n
c e s
b e tw
e e n
t h
e
th re
e f
a rm
in g
t y
p e s
S ig
n if
ic a n t
d if
fe r-
e n
c e s
b e tw
e e n
c o
n -
v e n
ti o
n a l
a n
d o
r- g
a n
ic f
a rm
in g
Specific cost 37.18 30.52 35.80 36.90 ** Energy 7.91 5.11 10.97 8.02 *** *** Other overhead costs 20.94 19.11 24.56 21.11 *** Depreciation 23.83 23.52 17.47 23.42 *** Wages paid 4.80 17.76 4.13 5.13 *** *** Rent and interests 6.77 4.60 9.09 6.85 *** VAT balance and taxes -1.43 -0.63 -2.02 -1.44 Total registered costs 100.00 100.00 100.00 100.00
Notes:Significance levels: *p<0.1, **p<0.05 and ***p<0.01
Table 5. Mean and median values of work units, specific and energy costs.
C o
n v
e n
ti o
n a l
fa rm
in g
O n
ly o
rg a n
ic
fa rm
in g
T o
ta l
fa rm
s
S ig
n if
ic a n t
d if
- fe
re n
c e s
b e tw
e e n
c o
n v
e n
ti o
n a l
a n
d
o rg
a n
ic f
a rm
in g
mean 1.60 3.10 1.64 *** Total annual work units
median 1.25 1.75 1.25 *** mean 0.26 1.81 0.31 ***
Hired work units median 0.00 0.54 0.00 *** mean 6.79 4.11 6.80
Percent of energy costs to output median 4.93 3.02 4.97 *** mean 32.09 22.35 31.49
Percent of specific costs to output median 23.61 21.11 23.56 **
Notes: Significance levels: *p<0.1, **p<0.05 and ***p<0.01
2010, Vol 11, �o1 81
between organic and conventional farming, significant differences with p<0.01 and
p<0.05 were found for median values.
Wackernagel and Rees (1996) proposed a method for measuring the human impact
on the earth by calculating the ecological footprint (EF). The EF appraises the total bio-
productive area needed to sustain society’s activities, accounting for resource supply,
waste absorption and the space occupied by human infrastructure (Haberl et al. 2004).
In spite of its limitations (Ayres 2000; Opschoor 2000; van Kooten and Bulte 2000), it
provides meaningful comparisons between nations as to the demands they place on na-
ture to sustain human activities and their respective biocapacity (Monfreda et al. 2004;
Deutsch et al. 2000). Human consumption of energy is an important component of the
EF (Stöglehner 2003). Specific data on energy consumption from our sample allow us
to assess the incremental environmental impact of conventional farming with respect to
organic farming in terms of EF. From the three EF calculations available for converting
energy consumption into its corresponding land area, the forest area needed to sequester
the CO2 emitted from burning fossil fuel is the most commonly used and accepted, even
though it gives the smallest EF measurement (Wackernagel and Rees 1996, p. 72-74).
However, all three approaches have been found to give similar results (Wackernagel
and Rees 1996, p. 72; Stöglehner 2003), albeit that they tend to underestimate the real
spatial impact on the biosphere (Wackernagel and Silverstein 2000). The XCAC pro-
vided us with detailed data about the fuel and electricity consumption for each farm. On
average, the conventional farms in our sample spent 6.10% of total output on fuel and
0.69% on electricity over the period studied, while organic farms spent 3.74% and
0.37% respectively. According to these data, the EF of the energy spent by conventional
farms is on average 5.32 hectares, 14.75% of their mean utilized agricultural area,
which means an incremental EF of 2.08 hectares with respect to organic farming, or
5.77% of their mean utilized agricultural area 3 , thus providing additional evidence of the
lower environmental impact of organic farming.
Discussion and Conclusions
This paper conducts an empirical analysis of output, costs and income in organic
farming. Organic farming is the most environmentally friendly farming system available
today and the citizens of the European Union have identified it as one of the main pri-
orities within the region’s agricultural policy. Likewise, a small group of organic farm-
ers are highly committed to safeguarding the environment. However, its future will de-
pend on the economic viability of its practices and the support it receives from the CAP.
Our results indicate that organic agriculture has a significant influence on raising
financial output, suggesting that organic farmers obtain a market premium that reflects
the consumer’s willingness to pay for healthier and environmentally friendly food. Yet,
no significant influence was found in farm costs and bottom line profits when calculat-
ing the two with registered financial costs and adding the opportunity costs of the work
put in by the family. No influence was found either when subsidies were included as
part of farm profits. Our results suggest that subsidies are mainly driven by factors other
than organic farming.
Surprisingly, estimations for farms in transition to, or partly performing, organic
farming did not show any significant influence on output, costs and profits. It seems
82 AGRICULTURAL ECO�OMICS REVIEW
probable that farms convert gradually to organic farming, combining both conventional
and organic farming with a tiny proportion of their business in transition.
Although we found no significant differences in total costs between the two types of
farming, their composition did differ. Wages accounted for a greater share of costs in
organic than in conventional farming, while energy and specific costs accounted for
smaller shares.
Total costs or the bottom line profits provide biased information about the economic
and social performance of organic farming with respect to that of conventional farming.
Detailed information about costs showed that organic farming generates more employ-
ment and consumes less energy, insecticides, fungicides, chemical-based fertilizers and
crop protectors, as well as less purchased feedstuff and medicines for livestock, thus
contributing to alleviate the environmental impact of agriculture.
The financial data available provide homogenous values that allowed us to compare
various situations. However, they also hide inherently different facts and can be mis-
leading. While the impact on the profit and loss statement of 1€ of energy was the same
as that of 1€ of wages, both expenses differ markedly in terms of their social and envi-
ronmental impact. There are crucial transactions that are not marketed, registered and
valued, but yet yield social and environmental profits and costs. In the specific case of
agriculture, conventional farming is reaching a point of saturation that heralds many
present and future environmental risks and problems. The issue is too important to be
solved purely in terms of financial viability. Rather there is a need to examine nitrate
pollution, biodiversity, food safety, soil protection, etc. when assessing agricultural de-
cisions. Financial accounting values cannot be considered reliable when disclosing the
social and environmental costs of individual farms. The International Accounting Stan-
dard 41 did not attempt to include social and environmental data, although their inclu-
sion is essential if they are to be given adequate weight in the decision-making process.
Few studies in economics have considered non-marketed outputs and costs. Con-
stanza et al. (1997) estimated the current economic value of the world’s ecological sys-
tems and its natural capital. In the specific case of agriculture, Pretty et al. (2000) as-
sessed a wide array of external costs of agriculture, none of which are available from
either agricultural financial statements or from the FADN, one of the cornerstones of the
CAP. From within accounting circles demands are being made concerning the necessity
of broadening the field covered by accounting in order to include social and environ-
mental data (e.g. Mathews 1997; Bebbington 1997). A worthy and fruitful result of such
calls can be found in the Global Reporting Initiative, but this remains a voluntary initia-
tive and is not at all suited to the agricultural sector.
This paper has performed an empirical analysis comparing organic and conventional
farming practices and has sought to draw environmental and social conclusions from the
limited financial accounting information available for our sample of individual farms.
We found no significant differences in financial performance between organic and con-
ventional farms, although the former recorded a significantly lower environmental im-
pact and created more employment opportunities.
Future research is needed to analyse broader aspects of the environmental and social
impacts of the two types of farming. The inclusion of social and environmental issues in
agricultural accounting and/or the FADN should make this easier. Future research is
also needed in order to identify the kind of data that can provide a more appropriate
2010, Vol 11, �o1 83
assessment of sustainability (Edwards-Jones and Howells 2001; Rigby and Cáceres
2001). In our opinion, the EF analysis provides an interesting and comprehensive
framework in which to build these studies.
�otes
1. The Farm Accountancy Data Network (FADN) recognises revenue with production
valued at market price, and labels it as output. Consequently with the data used in
this study, hereinafter, we employ this term.
2. According to data from the Spanish Ministry for Agriculture, 1.19% of Spanish
farms practised organic farming exclusively in 1999 (similar to the 1.39% of Catalan
farms), while in our sample they account for 2.9%.
3. Calculations were performed based on the following data:
Energy conversion factors: 28.38095 litres of gas oil/diesel per Gigajoule and 277.77
kWh per Gigajoule (British Petroleum 2007).
Specific energy footprint global average in Gigajoules/hectare per year: 55 for coal,
71 for liquid fossil fuel, 93 for fossil gas, 71 for nuclear energy and 1000 for hydro-
electric energy (Wackernagel et al. 1999).
Prices for electricity and agricultural gas oil in 2000 in Spain: 12.96 pts./kWh (Span-
ish Ministry of Economics 2001) and 74.7075 pts./litre (COAG 2004) respectively.
Sources of electricity in 2000 in Spain: 15.73% from hydraulic, 35.19% from nu-
clear, 43.27% from coal and 5.81% from fuel-gas (Spanish Ministry of Economics
2001).
Acknowledgements
The authors would like to thank the Xarxa Comptable Agrària de Catalunya for providing
the data that made this paper possible, and the Spanish Ministerio de Educación y Ciencia
(SEJ2005-04037/ECON) and the Generalitat de Catalunya Research Group
2005SGR00285 for funding this research.
References
Altieri, M.A. (1998). Ecological impacts of industrial agriculture and the possibilities for truly sustainable farming. Mothly Labor Review 50 (3): 60-72.
Ayres, R.U. (2000). Commentary on the utility of the ecological footprint concept. Ecol. Econ. 32 (3): 347-349.
Batte, M.T., Hooker, N.H., Haab, T.C. and Beaverson, J. (2007). Putting their money where their mouths are: consumer willingness to pay for multi-ingredient, processed organic food products. Food Policy 32(2): 145-159.
Bebbington, J. (1997). Engagement, education and sustainability: a review of environmental accounting. Accounting, Auditing and Accountability Journal 10(3): 365-381.
Boschma, M., Joaris, A. and Vidal, C. (2001). Agriculture and Environment. Concentration of livestock production. European Commission, Brussels.
Britis Petroleum. (2007). Statistical review of world energy 2007
84 AGRICULTURAL ECO�OMICS REVIEW
(http://www.bp.com/productlanding.do?categoryId=6848&contentId=7033471) Bublot, G. (1990). Économie et Gestion de l'Exploitation Agricole. Volumes I et II. Université
Cathólique de Louvain, Louvain. COAG. (2004). Informe gasóleo agrícola: situación económica y propuestas de COAG
(http://www.portalbesana.es/estaticas/servicios/documentacion/gasoleo_agricola.html). Darnhofer, I., Schneeberger, W., Freyer, B. (2005). Converting or not converting to organic
farming in Austria: farmer types and their rationale. Agriculture and Human Values 22: 39- 52.
Deutsch, L., Janson, A., Troell, M., Rönnbäck, P., Folke, C. and Kautsky, N. (2000). The eco- logical footprint communicating human dependence on nature’s work. Ecol Econ 32 (3): 351-355.
Dima, S.J. and Odero, A.N. (1997). Organic farming for sustainable agricultural production. A brief theoretical review and preliminary empirical evidence. Environ Resource Econ 10: 177-188.
Dobbs, ThL. and Smolik, J.D. (1996). Productivity and profitability of conventional and alterna- tive farming systems: a long-term on-farm paired comparison. Journal of Sustainable Agri- culture 91(1): 63-79.
Drinkwater, L.E., Wagoner, P. and Sarrantonio, M. (1998). Legume-based cropping systems have reduced carbon and nitrogen losses. �ature 396: 262-265.
Dupraz, P. (1997). La spécialisation des exploitations agricoles: changements techniques et prix des facteurs. Cahiers d’Économie et Sociologie Rurales 45: 93-122.
Edwards-Jones, G. and Howells, O. (2001). The origin of inputs to crop protection in organic farming systems: are they sustainable?. Agricultural Systems 67: 31-47.
European Commission. (1997). Data Definitions and Instructions for the FADN Farm Return. Community Committee for the Accountancy data Network, RI/CC 1256 Prov.
European Commission (1998) Definition of Variables Used in FADN Standard Results. Com- munity Committee for the Accountancy data Network, RI/CC 882 Rev. 6.
European Commission. (2002). Analysis of the possibilities of a European Action plan for or- ganic food and farming. Commission Staff Working Paper, SEC1368.
European Commission. (2005). Europeans and the Common Agricultural Policy. Special Euro- barometer, 221/Wave 622 – TNS Opinion and Social.
Gil, J.M., Gracia, A. and Sánchez, M. (2000). Market segmentation and willingness to pay for organic products in Spain. International Food and Agribusiness Management Review 3: 207-226.
Haberl, H., Wackernagel, M., Krausmann, F., Erb, K. and Monfreda, Ch. (2004). Ecological footprints and human appropriation of net primary production: a comparison. Land Use and Policy 21: 279-288.
Hopkins, J.A. and Heady, E.O. (1982). Contabilidad y Control de Explotaciones Agrícolas. Reverté, Barcelona.
Hornbaker, R.H., Dixon, B.L. and Sonka, S.T. (1989). Estimating production activity costs for multioutput firms with a random coefficient regression model. American Journal of Agri- cultural Economics 71 (1): 167-177.
Hsiao, Ch. (2005). Analysis of panel data. Cambridge University Press, Cambridge. Kerselaers, E., De Cock, L., Lauwers, L. and Van Huylenbroeck, G. (2007). Modelling farm-
level economic potential for conversion to organic farming. Agricultural Systems 94: 671- 682.
Kurosaki, T. (1997). Production risk and advantages of mixed farming in the Pakistan Punjab.
2010, Vol 11, �o1 85
The Developing Economics XXXV (1): 28-47. Lansink, A.O. and Jensma, K. (2003). Analysing profits and economic behaviour of organic and
conventional Dutch arable farms. Agricultural Economics Review 4(2): 19-31. Lansink, A.O., Pietola, K. and Bäckman, S. (2002). Efficiency and productivity of conventional
and organic farms in Finland 1994-1997. European Review of Agricultural Economics 29(1): 51-65.
Lund, V., Hemlin, S. and Lockeretz, W. (2002). Organic livestock production as viewed by Swedish farmers and organic initiators. Agriculture and Human Values 19: 255-268.
Mahenc, Ph. (2007). Are green products over-priced?. Environ Resource Econ 38: 461-473. Mathews, M..R. (1997). Twenty-five years of social and environmental accounting research. Is
there a silver jubilee to celebrate?. Accounting, Auditing & Accountability Journal 10 (4): 481-531.
Matson, P.A., Parton, W.J., Power, A.G. and Swift, M.J. (1997). Agricultural intensification and ecosystems properties. Science 277: 504-509.
Melfou, K. and Papanagioutou, E. (2003). Total factor productivity adjusted for a detrimental input. Agricultural Economics Review 4(2): 5-18.
Mishra, A.K., El-Osta, H.S. and Steele, Ch.J .(1999) Factors affecting the profitability of limited resource and other small farms. Agricultural Finance Review 59 (0): 77-91.
Monfreda, C., Wackernagel, M. and Deumling, D. (2004). Establishing national natural capital accounts based on detailed Ecological Footprint and biological capacity assessments. Land and Use Policy 21: 231-246.
Offermann, F. and Nieberg, H. (2000). Economic performance of organic farms in Europe. Uni- versity of Hohenheim, Stuttgart.
Omamo, S.W. (1998). Farm-to-market transaction costs and specialisation in small-scale agri- culture: explorations with a non-separable household model. Journal of Development Stud- ies 35 (2): 152-163.
Opschoor, H. (2000). The ecological footprint: measuring rod or metaphor?. Ecol. Econ. 32 (3): 363-365.
Popp, M,. and Rudstrom, M. (2000). Crop enterprise diversification and specialty crops. Agri- cultural Finance Review 60 (0): 85-98.
Pretty, J.N., Brett, C., Gee, D., Hine, R.E., Mason, C.F., Morison, J.I.L., Raven, H., Rayment, M.D. and van der Bijl, G. (2000). An assessment of the total external costs of UK agricul- ture. Agricultural Systems 65: 113-136.
Pretty, J.N., Brett, C., Gee, D., Hine, R.E., Mason, C.F., Morison, J.I.L., Rayment, M.D., van der Bijl, G. and Dobbs, T. (2001). Policy challenges and priorities for internalizing the ex- ternalities of modern agriculture. Journal of Environmental Planning and Management 44(2): 263-283.
Pretty, J.N., Ball, A.S., Lang, T., Morison, J.I.L. (2005) Farm costs and food miles: an assess- ment of the full cost of the UK weekly food basket. Food Policy, 30: 1-19.
Rigby, D. and Cáceres, D. (2001). Organic farming and the sustainability of agricultural sys- tems. Agricultural Systems 68: 21-40.
Rigby, D., Young, T., Burton, M. (2001). The development of and prospects for organic farm- ing in the UK. Food Policy 26: 599-613.
Schmitt, G. (1991). Why is the agriculture of advanced western economies still organized by family farms? Will this continue to be so in the future?. European Review of Agricultural Economics 18: 443-458.
Spanish Ministry of Economics. (2001). Boletín Estadístico de Energía Eléctrica 34
86 AGRICULTURAL ECO�OMICS REVIEW
(http://www.ree.es/apps/index_dinamico.asp?menu=/cap07/menu_sis.htm&principal=/cap0 7/estadistico.htm)
Stöglehner, G. (2001). Ecological footprint – a tool for assessing sustainable energy supplies. Journal of Cleaner Production 11: 267-277.
Tilman, D. (1998). The greening of the green revolution. Nature 396 (19th november): 211-212. Trewavas, A. (1999). Much food, many problems. �ature 402 (18): 231-232. Trewavas, A. (2001). Urban miths of organic farming. �ature 410 (22): 409-410. Tzouvelekas, V. and Pantzios, C.J. (2001). Economic Efficiency in Organic Farming: Evidence
From Cotton Farms in Viotia, Greece. Journal of Agricultural & Applied Economics 33 (1): 35-48.
Van Kooten, G.C. and Bulte, E.H. (2000). The ecological footprint: useful science or politics?. Ecol. Econ. 32 (3): 385-389.
Wackernagel, M. and Rees, W. (1996). Our ecological footprint, reducing human impact on the Earth. New Society, Gabriola Island.
Wackernagel, M. and Silverstein, J. (2000). Big things first: focusing on the scale imperative with the ecological footprint. Ecol. Econ. 32 (3): 391-394.
Wackernagel, M., Onisto, L., Bello, P., Callejas, A., López, I.S., Méndez, J., Suárez, A.I. and Suárez, M.G. (1999). Nacional natural capital accounting with the ecological footprint con- cept. Ecol. Econ. 29 (3): 375-390.
Copyright of Agricultural Economics Review is the property of Agricultural Economics Review and its content
may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express
written permission. However, users may print, download, or email articles for individual use.
