




































In ternationa l
Scholars
Journa ls

 

African Journal of Agricultural Marketing ISSN 2375-1061 Vol. 6 (7), pp. 001-015, July, 2018. Available online at 
www.internationalscholarsjournals.org © International Scholars Journals 

 

Author(s)                                           Author(s) retain the copyright of this article. 
 

 

Full Length Research Paper 

 

Institutional incentives and economic aspects of 
industrial relationships in the Italian agricultural sector 

 
Darina Zaimova 

 
European Research Institute on Cooperative and Social Enterprises, EURICSE, Via S. Giovanni 36 – 

38122 Trento (TN) – Italy. E-mail: dzaimova@gmail.com. Tel: 00359899807008. 
 

Accepted 05 March, 2018 
 
Policy interventions and private initiatives undertaken in the Italian agricultural sector have registered varying degrees 
of impact at the local level, due to the diversity of regional characteristics. The research paper aims to conceptualize 
the agricultural value-chain model and sheds light on the institutional incentives for product and supply organization in 
one integrated and interrelated process. The economic efficiency and competitiveness of strategic partnerships in the 
form of producer organizations, associations and cooperatives will be analysed. The first part of the paper investigates 
the theoretical and methodological framework for measuring the institutional effects on contract arrangements. The 
second part is devoted to assessing the structure of the value-chains in the agricultural sector in Italy and to 
operationalizing the concept of the behavioural model of strategic participants. The third part presents results of 
research, utilizing the applied econometric approach based on the stochastic frontier production models. 

 

Key words: Strategic partnerships in Italy, institutional framework, stochastic frontier production models. 
 
 
INTRODUCTION 

 
Increased concentration in the agricultural market has 
spurred controversial debates regarding the potential 
opportunities and adverse effects on the balance and 
distribution of market power across the value chain. The 
observed benefits and disadvantages are interpreted 
differently depending on whether this is done from an 
economic or political perspective. Hence, the lower costs 
of production, sufficient levels of aggregate efficiency and 
management gains are only part of the total market 
equation. Along with these outcomes is the prevailing 
dominance of a group of companies, which in principle is 
not in contradiction to the legal framework but it 
stimulates price levels that exceed the social cost of 
production and disrupts the equilibrium across the agri-
food chain. Development of the agricultural sector in Italy 
is not an exception to these common findings. Still, there 
are some distinguishing sector characteristics that need 
to be emphasized. Emergence of “cluster policy” at the 
sub-national level in Italy is supported by strong regional 
governance and producer initiative (Hospers and Sjoerd, 
2002). This provides for reliable legal framework and 
economic incentives for enterprises to integrate along the 
value-chain. 

 
 
 
 

 
An intriguing question is how small-sized companies 
succeed in being competitive in terms of quality, 
production costs and price. The answer is in their ability 
to behave as part of a group. They bind together on the 
basis of the product produced or geographical region and 
create their common competitive advantage through 
specialization and cooperation. An assertion such this 
calls for concrete proof of how institutional support and 
policy interventions translate into economic efficiency and 
productivity at the organizational level. The general 
interest in measurement has opened the prospect of 
micro-level approaches that develop new perspectives on 
how to define efficiency and productivity and how to 
calculate benchmark technology. The advantage of 
functional models that both account for agricultural and 
non-agricultural factors is mainly grounded on the 
assumption that there are other resources of efficiency 
that could be exploited by enterprises, but not modified or 
changed by them. An institutional explanation dominates 
over the other usual approaches concerning income 
differences, investment activity and productivity growth. 
Considering that informed management decisions require 
accurate information, a starting point is considering how 

file:///C:\Users\user\Documents\REPUBLICATION\AGRICULTURAL%20SCIENCES\AppData\Local\Temp\www.internationalscholarsjournals.org


 
 
 

 

to measure institutional effects. In Italy, the impact of the 
forthcoming institutional changes on producer organi-
zations, cooperatives and associations puts forward a 
unique organization of the value-chain model. The 
common feature of these organizations is their origin as 
bottom-up initiatives of producers and their dependence 
on the collective contribution and awareness. 
 

 

INSTITUTIONALIZATION OF AGRICULTURAL 
RELATIONS: THEORETICAL AND PRACTICAL 
SOLUTIONS 

 

Considering that “closing the gap between “know what” 
and “know how” in institutional reform is critical for 
achieving sustained growth” (Sobel and Coyne, 2010), it 
is difficult to unilaterally estimate the impact and results 
from the policy interventions. The main argument in favor 
of this assumption is that development of the structurally 
weak regions could not be addressed by generally 
accepted support mechanisms. For this reason, 
agricultural policy could not be assumed as a universal 
approach, but rather as a common framework of the 
development direction. Furthermore, the economic 
growth and productivity in the sector is not only assessed 
by the achieved economic results, but also by the 
flexibility and manners in the way in which the external 
environment enters into organizations (Powell and 
DiMaggio, 1991). It could be inferred that one of the most 
distinctive characteristics of the enterprises is the change 
rate of their external environment. They are compelled to 
implement certain policy regulations, to adopt “best-
practices” from their professional sector or to 
demonstrate leadership to ensue public interest. One 
trend in these adjustment activities is the isomorphic 
process among agricultural enterprises. They do not 
simply follow legal, economic, social, cultural or 
environmental prescripts. In order to create a stable 
equilibrium outcome, enterprises reflect the interests of 
the strongest coalition and relatively sub-coalitions that 
remain stable over certain periods (Aoki, 2001). This also 
corroborates the interpretation of the institutions as 
“persistent” social facts that regulate social behaviour and 
exert decisive influence over economic intensity. 
 

But it took rather a long time to formulate and appraise 
the significance of this adaptation process. The 
neoclassical comprehension for a competitive market 
considers its own ability to efficiently allocate production 
resources. The exchange of property rights is performed 
under assumption for precisely formulated contract 
agreements – well-defined responsibilities and under-
takings or “sharp in by clear agreement, sharp out by 
clear performance” (Macneil, 1974). Obviously neo-
classical theory has a clear concept on market exchange 
relations, but completely excludes their dynamic nature 
and the possibility for unexpected and abrupt changes. 
Moreover, the market seems to efficiently substitute 

 
  

 
 

 

the role of institutions, which in the real economic 
situation is excessively optimistic to rely on. The 
philosophy of the new institutional theory takes a step 
forward by introducing the role of transaction costs and 
by broadening the concept of property rights with the view 
of social norms and values. In that way, even the simple 
institutional interpretation of legislative prescriptions is 
elaborated (Bromley, 1989). The property itself is 
disclosed as a social relation, which defines and 
differentiates the owner of the asset from the other 
participants in the market. In addition, the more the asset 
becomes specialized the less the market exchange 
mechanisms are able to allocate it efficiently. In view of 
characteristics of the agricultural sector, this could result 
in disproportionate market power across the food-chain.  

One way to counteract this is through collective 
bargaining and competitive inter-sector relations in the 
form of vertically-integrated economic and social 
structures. The flexibility of these structures provides for 
further market specification, environmental and food 
safety improvement, as well as responses to service 
access issues. The integration process also changes the 
consumption model - consumers demand quality of the 
products, retailers introduce traceability of these products, 
processors use specific inputs to meet environmental 
requirements. The equilibrium between market demand 
and the capacity of the supply side is a matter of long-
term strategy. Several studies apply the “fixed asset” 
principle (Williamson, 1981) to underline some of the most 
important benefits and potential of integrated relations in 
agriculture. The term “fixed asset” represents the asset, 
whose present marginal value is neither justified by 
acquisition of more of it nor by its disposition. This concept 
is used by Hathaway (1963) to differentiate between the 
individual economic agents and the forms of joint 
economic performance. An important peculiarity is that the 
salvage value of the fixed asset, which is the asset‟s value 
when it is sold instead of used in production, is 
substantially higher in the agricultural sector. A situation 
where market price falls would not necessarily lead to 
reduction in production quantities. A reasonable solution is 
to transfer assets to a more profitable type of production 
or activity along the value-chain. Considering the opposite 
situation where the marginal output value of the asset 
exceeds its acquisition costs due to an increase in the 
market price, then more of the input would be added, 
since it enhances total productivity. In the case of a fixed 
asset with low salvage value, when market prices return to 
their initial level, the level of the production volumes would 
not be able to be adjusted automatically. This is defined 
as a “productivity trap” (Hathaway, 1963). The immobility 
degree of the fixed asset complicates production 
decisions even more at the farm-level. This assets‟ 
characteristic is envisaged as an impediment for efficient 
functioning of the factor markets. It leads also to higher 
cost related expenditures, especially considering the 
cases of individual production choices. 



 
 
 

 

Another consideration is the parallel accomplishment of 
some of the activities in the agricultural production 
process. The rational producer is expected to ignore 
activities that are more time demanding and cost 
consuming and to focus his efforts on the more profitable 
ones. An opportunity in this direction is provided by 
various organizational types in the agricultural sector 
such as: cooperatives, second-level cooperatives 
(consortia) or producer organizations. This observation 
can be viewed as simply an application of the “equal 
compensation principle” in the meaning that: “if a 
regulated firm‟s allocation of effort to two different 
objectives cannot be monitored by the regulator, then 
either the power of the incentive for each objective is 
equal, or the objective with the lower power receives no 
effort” (Milgrom and Roberts, 1992).  

Decision to enter into a contract relations is well-
founded, only if the expected income is at least equal to 
that one from an alternative production decision. This 
could be defined as “rational restriction”. In order to 
maximize its profit, the enterprise would try to offer a 
contract that satisfies „shareholders‟ or „members‟ 
conditions. Milgrom and Roberts (1992) go on to 
emphasize that: “the equal compensation principle 
imposes a serious constraint on the incentive regulation 
mechanisms that can be effective in practice. In particular, 
if a regulated firm is expected to devote some effort in the 
pursuit of an objective for which performance cannot be 
measured at all, then incentive regulation cannot be 
effectively used for any of the objectives that the regulator 
might wish to pursue. The use of straightforward rate-of-
return regulation can often be justified on these grounds”. 
 
 

 
STRUCTURING THE AGRO-INDUSTRIAL RELATIONS 
IN THE ITALIAN REGIONS 

 

Common agricultural policy (CAP) is a classical example 
of an intervention policy, considering its legal provisions 
and mechanisms for product support, production quotas, 
aid schemes and direct payments. Early retirement 
schemes, grants for setting up young farmers and 
diversification initiatives promote intra-sector structural 
change and enhance the mobility of production resources. 
Indeed, these measures are often assumed as an 
impediment for efficient allocation of resources because 
they are aimed at intervention in the primary production 
sectors. Even direct transfer payments are not qualified as 
an efficient instrument for employment promotion as they 
have short-term effects. The difficulty with the policy 
application is associated with the characteristics of 
members within the target group. Agricultural producers 
are not very receptive to significant changes in the farm 
development because of their age, risk aversion or 
insufficient capital resources. The most efficient way to 
maintain and develop small farms is through diversi-
fication and off-farm employment. 

 
 
 
 

 

Hence, it is necessary to improve and give publicity to 
the attractiveness of rural areas to the other economic 
sectors. One of the main concerns is how to enhance 
producers‟ positions in the supply process and to 
guarantee visibility of the products' quality. This issue is 
addressed and operationalized in the legal forms of 
cooperatives, producer organizations, and associations. 
They provide for a wide range of tools for production and 
crisis management, market flexibility, and simultaneously 
maintain close relationships among the members and 
coordinate their individual interests. The Italian agricultural 
sector is particularly dynamic in development and 
organization of cooperatives and associations, as well as 
in the concentration of production and market supply. 
Policy mechanisms and interventions are complicated and 
vary because of the heterogeneity of the Italian regions. In 
the first place they fall into different European funding 
categories. At the same time, not all regions possess the 
same special forms and conditions of autonomy pursuant 
to the special statutes adopted by Italian constitutional law 
1
 . Strong regional identity is preserved in fruit, wine and 

cheese production. Local production systems emerge in 
these sectors, which are based on small-scale production 
– the total number of the registered producer 

organizations all over the country is 235 
2
 . This is 

observed as an opportunity to maintain small-scale and 
semi-subsistence farming; and to integrate through a large 
set of cooperative arrangements. From one point of view, 
this fact could be interpreted as rather beneficial, because 
concentration of production and distribution processes 
ensures its continuity and imposes high quality standards 
for the final product. Nevertheless, this “modernization” 
process of the agricultural sector causes its restructuring 
in a manner less favourable for market participants in 
certain regions. The imbalanced relationship between 
small producers and their customers – processors or 
retailers, as well as the high level of fragmentation and 
low level of cooperation additionally complicates the retail 
system in most sectors of agricultural production. 
 

One of the main characteristic of the Italian fruit and 
vegetable sector is the presence of strong concentration 
of production and initially well-structured and organized 

cooperatives and associations
3
. Producer organizations 

play an important role for stability of the market and  

 
1
 Constitution of the Italian Republic, Art.116, “Friuili-Venezia Giulia, 

Sardinia, Sicily, Trentino-Alto Adige/ Südtirol and Valle d’Aosta/ Vallee 
d’Aoste have special forms and conditions of autonomy pursuant to the special 
statutes adopted by constitutional law.  
2 Decreto N.85/TRAV Attuazione del decreto legislativo 27 maggio 2005, n. 
102, sulla regolazione dei mercati, a norma dell’art. 1, comma 2, lettera c), 
della legge 7 marzo 2003, n. 38, recante i requisiti minimi per il riconoscimento 
delle organizzazioni di produttori, le modalità per il controllo e per la vigilanza 
delle organizzazioni dei produttori, al fine di accertare il rispetto dei requisiti 
per il riconoscimento nonché le modalità per la revoca del 
riconoscimento.(integrato con le modifiche alla Tabella 1 introdotte dal D.M. 
prot. n. 121/traV del 10 marzo 2008)

 
 

3
 Terms and conditions of the scheme of aid for producer 

organizations in the fruit and vegetables sector (2009)
 



 
 
 

 

efficient relationships across the agri-food chain. 
Membership in these organizations is perceived as an 
important competitive factor and long-term strategy in 
improving supply chain relations through adjustment to 
the market criteria (logistics, quality management, 
financial capacity, price and cost competitiveness). Five 
years after the first introduction of the Council Regulation 
2200/1996, 130 producer organizations were registered 
all over the country. These organizations concentrate 
approximately 20% of marketed production with 
significant difference in the various regions. The 
organizational rate in the fruit and vegetable sector is 
100% in Trentino Alto Adige, 65% in Emilia Romagna, 
while in Sicily the organizational rate is about 7% and in 
Puglia 5%. The turnover of the dairy sector represents 
15% of the total turnover of the food industry. The sector 
is represented by a group of big enterprises, and by a 
great number of small firms. Production of fresh milk and 
innovated products is very concentrated within a number 
of mergers and significantly vertically integrated strategic 
groups. The most important division of the operating 
structures in the sector is as follows: 

 
1) Strategic groups – enterprises, mainly cooperatives 
(Abit, Lattira Soresina) that operate predominantly on the 
national market;  
2) In Parmigiano Reggiano area there are two 
classification types - degree of vertical integration and 
type of market strategy (trade or product oriented) of 
organizations;  
3) Assolatte (Associazione Italiana Lattiero Casearia) – 
represents the industries of milk transformation (fresh 
milk, cheese, yogurt, butter). 

 

Weaknesses of the dairy sector refer to: retail system, 
which is not visible enough in terms of participating high 
quality producers; imbalanced relationships between 
small producers and their customers – processors and 
retailers; high level of fragmentation and low level of 
cooperation among producers, especially in Southern 
Italy; lack of cooperation centers (mountain areas), where 
the link between producers and other operators is more 
difficult and complicated. Concentration of the wine 
production has created an important framework and 
conditions for innovation and knowledge sharing, 
institutional support and small-scale producers‟ support. 
The most important organizations in the “wine industry 
value-chain” are “Consortia Volontari di Tutela delle  
Denominazioni di Origine and FEDERDOC 
(Confederazione dei Consorzi Volontari per la tutela delle 
denominazioni d‟origine)”. Financial crisis had overall 
negative effects on economy, creating unemployment, 
cash-flow problems, and lowered consumption. For the 
wine sector, these consequences meant limited demand 
and over-supply. Current policy in the sector is focused 
on limitation of overproduction, requesting that wine 
producing countries rip out a certain amount of vineyard 

 
 
 

 

acreage, lowering and then gradually eliminating public 
subsidies for destruction of stocks, and allocating funds 
for promoting the consumption of wine in non-European 
countries. In light of these considerations, the following 
key-aspects were proposed by leading Italian wine 
organizations: promotion of wine consumption, especially 
in the area with concentrated tourist activity; short-term 
and medium-to-long-term programmes to enhance 
export; re-distribution of subsidies for promotion of Italian 
wine from domestic to foreign markets; investments 
aimed at professional training.  

A large part of agricultural production is produced and 
realized through a relatively stable number of 
cooperatives (Table 1). Cooperatives operate in every 
sector of the economy – agriculture, banking, industry 
and services. The gross value added (GVA) of the 
agricultural cooperatives for 2007 represents 5% of the 
total GVA of the economy. As a comparison, the GVA in 
the industrial sector is 3.0%; in the services sector – 
6.8%. The organizational process in the agricultural 
sector is further developed in the form of producer 
organizations, second degree cooperatives – consortia 
and associations. Development in the sector could also 
be described by “localization through intensified 
interaction and cooperation” (Brunori et al., 2002). 
Nevertheless, observations so far envisage that the large-
scale industry is more favoured, since it receives a 
considerable percentage of the European funds. At the 
same time, administration costs are pushed up because 
of the minute payments to small – scale farmers. The 
solution is imposed by the fixed limits of the payments 
that farmers could receive under the single payment 
scheme. The policy implementation relies on more 
market conformity and less direct payments. But what is 
the consequent effect on employment, income and 
consumers in the sector? The marginal cost of producing 
goods, which use land could exceed the social cost of 
production. Consequently, the market prices would 
influence the service sector, regardless to the products‟ 
true value. Besides this, a decrease in the total 
employment in the agricultural sector probably cannot be 
avoided, as it follows the extent to which production 
responds to price fluctuation. 
 

 

MEASURING INSTITUTIONAL INFLUENCE OVER  
ECONOMIC EFFICIENCY OF AGRICULTURAL 
ENTERPRISES 

 

If the government policy and the factors of institutional 
environment possess significant and even decisive role 
over income accumulation and comparison (Rodriket al., 
2002), how can these particular effects be measured? As 
it is stated: “when it comes to reforms, economists know 
what institutions and policies are necessary to ignite 
economic growth but they know much less about how to 
go about getting those institutions and policies” (Sobel 



 
 
 

 
Table 1. Number of registered agricultural cooperatives.  

 
 

Region 
2006  2007  2008  

 

 

Number % Number % Number % 
 

  
 

 Piedmont 288 5.57 286 5.47 296 5.63 
 

 Valle d'Aosta/Vallée d'Aoste 45 0.87 44 0.84 43 0.82 
 

 Liguria 61 1.18 64 1.22 65 1.24 
 

 Lombardy 320 6.19 320 6.12 310 5.89 
 

 Trentino Alto Adige 227 4.39 220 4.21 210 3.99 
 

 Veneto 401 7.76 398 7.61 401 7.63 
 

 Friuli-Venezia Giulia 158 3.06 161 3.08 159 3.02 
 

 Emilia-Romagna 671 12.98 653 12.48 638 12.13 
 

 Tuscany 171 3.31 174 3.33 170 3.23 
 

 Umbria 111 2.15 112 2.14 113 2.15 
 

 Marches 125 2.42 128 2.45 124 2.36 
 

 Lazio 324 6.27 329 6.29 334 6.35 
 

 Abruzzo 143 2.77 145 2.77 146 2.78 
 

 Molise 52 1 48 0.92 49 0.93 
 

 Campania 420 8.12 439 8.39 447 8.5 
 

 Puglia 561 10.85 575 10.99 589 11.2 
 

 Basilicata 96 1.86 100 1.91 103 1.96 
 

 Calabria 324 6.27 337 6.44 341 6.48 
 

 Sicily 518 10 543 10.38 565 10.74 
 

 Sardinia 154 2.98 155 2.96 156 2.97 
 

 Total 5170 100 5231 100 5259 100 
 

 
Source: Italian National Institute of Statistics, Istat.it. 

 
 

 

and Coyne,  2008).  Rodrik  (2000)  identifies 5 types of 
institutions that permit adequacy of the market: 
institutions for property rights, regulatory institutions, 

institutions of macroeconomic stability, institutions for 
social insurance,  institutions  for  conflict  management. 
The studied relation between aggregate efficiency and 

economic freedom shows that lack of economic freedom 

results in lower aggregate efficiency (Adkins et al., 2002). 
The agricultural sector, its participants and their typical 
features provide a fruitful research field for studying the 
relationships among efficient performance, market  
competition and policy-making. An interesting perspective 

is given by several studies that attempt to quantify 

institutional influence over economic efficiency. 

Stochastic frontier analysis (SFA) is applied in order to 

measure how divergence in quality of institutions, 

including: control of corruption, strength of the law, quality 

of the regulatory framework; explains cross-country 

differences in aggregate efficiency (Meon and Weill, 
2004). The relationship between the foreign direct 

investment  and  the  rate  of  growth  of  gross  domestic 
product is also developed through the quantitative and  
comprehensive results obtained from the same analysis 

(Wijeweera et al., 2010). According to them, the flows of 
foreign direct investment exerts a positive impact over the 

economic growth only in the presence of a highly skilled 

labor force; accordingly open trade policy gains efficiency, 

 
 
 

 

but at the same time corruption practices have a negative 
impact. The aforementioned analysis‟ application and 
suggested results are only part of the existing research 
experience (Appendix A). Although it is not possible to 
describe them all, it is important to consider the 
opportunities that frontier methodology provides in 
studying the various factors that influence efficiency.  

The main objective in applying stochastic frontier 
analysis in the present case study is to measure the 
efficiency levels of cooperatives, partnerships and 
producer organizations in the agricultural sector in the 
context of the influence of institutional factors on their 
productive choices. The formulation of the production 
function requires the definition of two types of variables:  
1) the output of agricultural enterprises and 2) the inputs, 

utilized in the production process
4
. The parameters of the 

constructed translog production function in the present 
case study are represented in Table 2. Instead of the 
physical quantities of output, the gross margin and the 
gross value added are used as measurement tools. This 
decision is based on the fact that the higher physical 
output of the more intensive enterprises is not 
comparable to the lower output of smaller enterprises.  

 
4
Information about the quantitative characteristics of the output and production 

inputs is derived from Eurostat, Istat and the Farm accountancy data network of 
the European Commission for the period from 2003 to 2007. 



 
 
 

 
Table 2. Parameters in translog production function.  
 

Output (Y) Input (X)   
Standard The total production in mil 
gross euro minus variable costs 
margin 

  
B1 -  Utilized  agricultural In hectares per each region (cereals, vegetables in 
area (UAA) open field, industrial crop, vegetables in greenhouses) 

B2 - Permanent crops (PC) In hectares, in relation to the received subsidies and 
  the impact they exert upon efficiency and productivity 
  (fresh fruits, vineyards and wine, olive plantations) 
 

Gross Output at  market  prices B3  - Intermediate Measured  by  cumulative  costs  of  raw  material 
 

added minus intermediate consumption (IC) consumption and service procurement 
 

value consumption at purchaser      
 

 prices  B4 - Annual working  unit Corresponds to the work performed by one person  

   
 

   (AWU)  who is occupied on agricultural enterprises for each 
 

       region on a full-time basis 
 

   B5 - Employed   annual Employed on a regular basis, including group holders 
 

   working unit (EAWU)  
 

 

 
Table 3. Parameters in the inefficiency model.  
 

Group 1: General information C1 - Training level Refers to the ratio of trained managers and employees to all employees 
about agricultural enterprises  in agricultural enterprises. 

 C2   -  Specialized  mixed Corresponds  to  the  output  orientation  of  the  agricultural  enterprises 
 farming specialized  in  a  particular  activity  (crop  production)  that  provides  a 
  standard gross margin of at least 2/3 of the total standard gross margin of 
  the enterprises. 

Group  2:  Characteristics  of C3-    Protection    of Rank provided by the component “Protection of property rights” in Area 2 
the institutional framework property rights “Legal structure and security of property rights” measured by Economic 

  Freedom of the World index. 

 C4   -  Legal  enforcement Rank provided by the component “Legal enforcement of contracts” in 
 of contracts Area  2  “Legal  structure and security of property rights” measured  by 
  Economic Freedom of the World index. 

 C5 - Starting a business Rank  provided  by  the  component  “Starting  a  business”  in  Area  5 
  “Regulation  of  credit,  labour  and  business”  measured  by  Economic 
  Freedom of the World index. 

 C6 – Cooperatives Number of the registered agricultural cooperatives 
 
 

 

Additionally, the gross margin variable includes subsidies 
as payment received for the fixed production factors. 
Therefore, the area under permanent crops is also 

included as an input variable
5
. Three types of permanent 

crops are eligible to receive subsidy payments partially or 
completely based on area: olive trees, vineyards and 
more recently nuts. Decision-making units are also 
compared on the basis of the relationship between  
 
5
Since 1992, the role of remote sensing and geomatics in the management and 

control of the Common Agricultural Policy has become significantly important 
in terms of implementation of the Land Parcel Identification System (LPIS) for 
identification of all parcels, for which area-based subsidies are claimed. 

 
 

 

annual working unit and employed annual working unit. In 
the analysis, annual working unit corresponds to the total 
labour input, including family labour. Variables in the 
inefficiency model are grouped in two categories that 
attempt to explain the level of inefficiency in respect to 
the observed institutional influence. The first one is 
regarding the general information about agricultural 
enterprises; the second category refers to characteristics 
of the opportunities and restrictions of the institutional 
framework (Table 3).  

The training level variable denotes the ratio of the 
professional and trained managers and employees to the 



 
 
 

 

total number of employees 
6
 . The specialized mixed 

farming variable indicates output orientation and product 
diversification in the enterprises. The variables in the 
second group attempt to describe institutional 
characteristics closely related to the economic 
performance of the agricultural enterprises. Information 
about property rights protection, enforcement of contract 
arrangements and incentives for starting a new business 
is derived from the Economic Freedom of the World 

(EFW index)
7
 . The index measures the consistency of 

institutions and policies within the concept of ownership 
and business activity. The main incentive to include these 
particular variables stems from the supposition that 
enterprises regardless of their legal forms, face the same 
market pressures, compete through adoption of similar 
strategies, and aim at higher levels of efficient business 
performance. Considering the impact of cooperatives‟ 
economic activity on overall regional development, an 
additional variable for 2007 was included: the number of 

agricultural cooperatives 
8
 . Cooperatives have the 

incentive to be an equivalent competitor in the market 
along with other investment oriented companies, as long 
as this position would secure their financial stability and 
maintain the loyalty of their members. The null 
hypotheses states that there is no technical inefficiency in 
the structured models: 
 

H0: hi (θ) = 0 against H1: hi (θ) ≠ 0 (1) 

 

The vector of estimated parameters is represented by θ. 
In order to determine the lower and upper bounds, the 
Kodde and Palm‟s Wald test for jointly testing non-linear 

equality and inequality constraints either under H0 or H1 

is used (Kodde and Palm, 1986). The null hypothesis H0 
is rejected when the estimated value of LR-tests exceeds 

the upper bound value, and H0 is accepted when the LR-
tests value is smaller than the lower bound value. The 

parameter γ = σu
2
/ σv

2
 + σu

2
 is the variance ratio, which 

explains the total variation in the output from the frontier 
level attributed to technical efficiency. When γ=0, then 
there is no technical inefficiency observed in the data set 
and all decision-making units belong to the optimal 
production frontier. According to the first hypothesis, 
there are no inefficiency components in the constructed 
Cobb-Douglas function: 
 

Ln (Y) = β0  + β1 Ln (В1) + β2 Ln (В2) + β3 Ln (В3) + β4 Ln 

(В4) + β5 Ln (В5) (2)  

 
6
The training levels of farm holders are indicated by: IRENA 06, the level of 

agricultural training of managers of agricultural holdings; IRENA 06a, the 
training in agri environmental issues; IRENA 01, the area under agri-
environment support; IRENA 02, regional levels of good farming practice; 
IRENA 13, cropping/livestock patterns; IRENA 14, farm management 
practices; IRENA 15, intensification/ Eextensification; IRENA 16, 
diversification/specialization 
7
Economic Freedom of the World: 2010 Annual Report, p.1  

8 The information about the number of cooperative is taken from the 
cooperative register of the Economic Development Ministry

 

 
 
 
 

 

The “second hypothesis” also states that the value of βi 

parameters is zero and the formulated translog function 

is: 
 

Ln(Y) = β0 + β1 Ln (В1) + β2 Ln (В2) + β3 Ln (В3) + β4 Ln(В4) 

+ β5 Ln (В5) + β6 Ln (В6) + ½[β7Ln (В7)
2
 + β8 Ln (В8)

2
 + β9 

Ln   (В9)
2
    +   β10    Ln   (В10)

2
 +   β11    Ln   (В11)

2
]   + 

β12 Ln (В12) + β13 Ln (В13) + β14 Ln(В14) + β15 Ln (В15) + β16 

Ln (В16) + β17 Ln (В17) +β18 Ln(В18) + β19 Ln (В19) + β20 Ln 

(В20) + β21 Ln (В21) (3) 
 

The “third hypothesis” states that values of δi parameters 
in the inefficiency model are zero: 
 

Uit = δ0 + δ1 Ln(С1) + δ2 Ln(С2) + δ3 Ln(С3) + δ4 Ln(С4) + δ5 

Ln(С5) + δ6 Ln(С6) (4) 
 
The results provided in Table 4 signify the likelihood ratio 
test of the three null hypotheses against the general 
model, which assumes that there are no inefficiency 
components in the structured production functions. The 
first null hypothesis states that the Cobb-Douglas 
production function is preferable to the translog 
production function. According to the results of the LR-
test, the null hypothesis is strongly rejected at the 5% 
level. The second null hypothesis states that each 
decision-making unit in the analysis operates on the 
technical efficiency frontier. The LR-test results also 
reject this assumption and suggest that there are 
technical inefficiency components in the structured 
production function. Following this, the focus is towards 
the joint effect of the selected variables and the possibility 
for optimization in terms of cost reduction. According to 
the third null hypothesis, the inefficiency effect is not a 
function of the two groups of explanatory variables. The 
results also reject this hypothesis, which confirms the 
supposition that these variables have a considerable 
effect on the technical efficiency of the decision-making 
units. 
 

The results obtained for the variance parameter γ 
indicate the proportion of the one-sided error component 
in the total variance of the composed error term 
(Appendix B). The average variation of the estimated 
output from the optimal frontier level, which is attributed 
to technical inefficiency is 0.6509. According to the 
estimated variances, output variability is mainly due to 
technical inefficiency rather than to statistical noise. For 
the period 2003 to 2007, 13 coefficients out of 21 total 
coefficients in the translog function are statistically 
significant at the 5% level. This leads to the conclusion 
for interaction and non-linearity among the included 
variables. The parameter β1 that corresponds to the 
utilized agricultural area (UAA) appears to be significant 
at the 5% level for the last 2 years. This confirms that the 
larger size of the enterprises entails better labour and 
capital endowments; they obtain higher efficiency levels 
and achieve better economic performance. The following 
is supported by the positive effect of the interaction 



 
 
 

 
Table 4. LR-test results.  

 

Test Null hypothesis 
Loglikelihood  

Value λ* Critical value** Decision  

function 
 

 

      
 

   2003   
 

1 H0 : βi = 0 -16,0378 70,0171 25,689 Reject H0 
 

2 H0 : βij = 0 -6,1317 3,1885 2,706 Reject H0 
 

3 H0 : γ = δi = 0 28,1062 20,0499 17,670 Reject H0 
 

   2005   
 

1 H0 : βi = 0 -11,9286 92,059 25,689 Reject H0 
 

2 H0 : βij = 0 18,8789 8,1042 2,706 Reject H0 
 

3 H0 : γ = δi = 0 23,6858 17,7180 17,670 Reject H0 
 

   2007   
 

1 H0 : βi = 0 -6,7947 75,821 25,689 Reject H0 
 

2 H0 : βij = 0 25,0111 19,2112 2,706 Reject H0 
 

3 H0 : γ = δi = 0 17,0140 32,169 17,670 Reject H0 
 

 
*λ – is the value of the likelihood ratio test of the null hypothesis associated with each of the three models against the 
alternative general model. This test has 16 degree of freedom. **0.005 significance level. Source: Own calculations. 

 

 

between utilized land and the variables: permanent crops 

(β12), intermediate consumption (β13), annual working 

unit (β14) and employed annual working unit (β15). The 
coefficients obtained for the period suggests theexistence 
of scale economies. Since in the analysis the agricultural 
enterprises vary in terms of land size, it is reasonable to 
consider what is the relationship between utilized area 
and permanent crops? From the results, it could be 
assumed that a specialization in permanent crops is 
preferable in small-sized enterprises, especially if there 
are insufficient investment funds and capital. The variable 
annual working unit (AWU) refers to the total labour input 
in the enterprise.  

Furthermore, the variable employed annual working 
unit (EAWU) is included in the analysis, which represents 
employees on a regular basis. The first reason for that 
choice is that it is difficult to obtain information about the 
family labour component or the “implicit costs”. These 
costs include non-distributed income from own labour in 
the farm, entrepreneurs income, income from own land 
and from own capital included in production. The second 
reason is that even evaluated, implicit costs do not 
account for market demand conditions. Finally, when 
family labour is to be considered, it is better to be 
described as a distinct input variable and not to be 
included in the hired labour variable. The results from 
translog function calculation show that utilization of the 
two variables (AWU, EAWU) in production process has 
reached satisfactory levels for 2005 and 2007. The most 
efficient enterprises use labour more rationally due to the 
more intensive use of other production resources, such 
as machinery or any technology equipment. Based on the 
differences in the obtained parameters‟ coefficients, it is 
appropriate to focus on the extent to which institutional 

 
 

 

framework may influence some enterprises and their 
respective regions to achieve relatively high efficiency 
scores compared to other apparently less efficient 
enterprises and regions. The inefficiency function 
provides some explanations of this effect (Figure 1). The 
results vary over the period analyzed. In 2003, the 

variables training level (δ1), specialized mixed farming 

(δ2) and starting new business (δ5) possess negative 
signs. The training level coefficient is statistically 
significant with a value different from 0, and a negative 
sign. This suggests that the higher education and training 
of managers and employees has a positive effect over 
the technical efficiency of agricultural enterprises. 
Another possible conclusion is related to the age of 
farmers, which is not considered as a variable in this 
analysis. The age of producers has increased over the 
years; they have many years of experience in the 
agricultural sector with satisfactory production results. By 
choosing to be a member of a cooperative or producer 
organization, they become involved in new production 
technologies and methods, and this decision keeps them 
in step with other, more innovative enterprises.  

There is also a significant relationship between 
production specialization and the obtained efficiency 
levels. Specialized strategies contribute to better 
allocation of production resources and their appropriate 
utilization in the production process. The results are also 
consistent with the estimated significant relationship 
between the agricultural area and the cumulative costs of 
raw material consumption and service procurement. 
Although specialization differs over the regions studied, it 
infers that most enterprises have managed to exploit the 
benefits of their particular locations.  

The coefficients in the inefficiency model  acknowledge 



  
 
 

 
0.8000 

 
0.6000 

 
0.4000 

 
0.2000 

 
0.0000 

 
-0.2000 

 
-0.4000 

 
-0.6000 

 
-0.8000 

 
-1.0000  

 
-1.2000 

 
 
 

 
 

          0.5733          
 

                            

          

0.3458 
      

0.3826 
       

                      
 

                         
 

                       
 

 
0.2641 0.2423           0.2008        

 

            0.1037              
 

                           
 

     2003      2005     2007     
 

                          
 

                         

                  -0.2213   
  

 

             

0.3030 
      

                         
 

                     

--0.3348   

             

-0.3307    
 

          

 

           

 

- -0.4592 
              

 

                      
 

                            

-0.6378                          
 

                           
 

                   -0.7365    
 

                           
 

                           
 

      -0.9564                  
 

                           
 

   Training level (TrLevel)  Specialized mixed farming (SMF) Property right protection (PRP) 
   

 Legal enforcement of contract (LEC)  Starting new business (SNB) 
  

Figure 1. Efficiency determinants patterns (source: Own calculations). 
 
 

 
Table 5. Estimates of elasticity of complementarity and substitution between pairs of inputs*.  

 

Year 
b12 b13 b14 b15 b23 b24 b25 b34 b35 b45 SCE 

 

   

p-complements > 0; p-substitutes < 0 
   

 

       
 

2003 0.204 -0.743 0.000 0.238 0.569 0.673 0.279 0.668 -0.537 0.403 0.429 
 

2005 -0.476 -0.241 0.209 0.000 0.563 -0.469 0.667 -0.220 -0.726 0.216 2.444 
 

2007 -0.927 -0.280 0.390 0.000 0.232 -0.120 0.523 -0.973 -0.221 0.378 2.002 
 

Sample mean -0.400 -0.421 0.299 0.079 0.455 0.028 0.489 -0.175 -0.495 0.332 1.625 
  

b12 is elasticity of substitution between AUU and PC, b13 is elasticity of substitution between UAA and IC, b14 is elasticity of substitution between 

UAA and AWU, b15 is elasticity of substitution between UAA and EAWU, b23 is elasticity of substitution between PC and IC, b24 is elasticity of 

substitution between PC and AWU, b25 is elasticity of substitution between PC and EAWU, b34 is elasticity of substitution between IC and AWU, 

b35 is elasticity of substitution between IC and EAWU, b45 is elasticity of substitution between AWU and EAWU. Source: Own calculation. 
 
 

 

that institutions and institutional arrangements have a 
direct and positive influence over the economic efficiency 
of the observed units. During the first period, institutional 
incentives and legislative provisions have stimulated new 
enterprises to enter the agricultural sector. This result is 
supported by the statistical data that in 2005 the total 
number of registered agricultural enterprises was about 
four times more than their primary number at the 

beginning of the period in 2003
10

.  
After 2005, the coefficients of the variables property 

rights protection (δ3) and legal enforcement of the 

contracts (δ4) confirm the positive effect of contractual 
arrangements over the efficiency level of decision-making 
units. In 2007, the effect of legal initiatives for starting 
business activity also contributed to attained efficiency 
levels. The coefficient of the variable of the total number 

of registered cooperatives by regions (δ6) is statistically 
significant and possesses a negative sign. Cooperatives 
significantly influence the overall sector performance.  

A substantial source of the cooperatives‟ impact is 
found in the common organization of production in terms 
of quality standards and demand-based quantities, as 
well as cost minimization and scale economies. 

 
 
 

 

The pairwise elasticity of inputs substitution is calculated 
for further interpretation of the results and in order to 
isolate each input‟s effect on the output (Table 5). The 
theoretical explanation in the case of pairs of inputs 
considers that “there is a simple correspondence between 
the cost function setting and the production function 
setting, since the elasticity of substitution is then equal to 

the inverse of the elasticity of complemen-tarity”
9
(Kohli, 

2011). If the elasticity of complementarity between pairs of 
inputs is positive, then the conclusion is that these inputs 
both contribute to the increase of the output level. In case 
the estimated value is negative the two inputs are 
substitutes. The values of the elasticity between inputs for 

utilized agricultural area and annual working unit (b14), 

permanent crops and intermediate consumption (b23), 

permanent crops and employed annual working unit (b25), 
and annual working unit and employed annual working 

unit (b45) are calculated to be higher than 0 and suggest 
positive cross elasticity of  
 

 
9 Kohli, Ul. (2010), “Labour productivity: Average versus Marginal”, Ch. 6, pp. 
103-132 in W.E. Diewert; B.M. Balk, D. Fixler, K. J. Fox and A. O. Nakamura 
(2010), “Price and productivity measurement”, vol. 6, Trafford Press

 



 
 
 

 
Table 6. Mean efficiency value (2003 – 2007).  

 
    Efficiency results   

 Regions Cobb-Douglas Translog Inefficiency model 

  Average Std.dev Average Std.dev Average Std.dev 

 Piemond 0.689 0.059 0.924 0.038 0.638 0.408 

 Valle d'Aosta 0.596 0.209 0.875 0.119 0.559 0.143 

 Liguria 0.783 0.188 0.925 0.071 0.403 0.232 

 Lombardy 0.856 0.124 0.969 0.036 0.972 0.023 

 Provincia Autonoma Bolzano 0.795 0.106 0.881 0.085 0.512 0.147 

 Provincia Autonoma Trento 0.549 0.240 0.878 0.077 0.551 0.085 

 Veneto 0.801 0.209 0.974 0.003 0.765 0.320 

 Friuli-Venezia Giulia 0.865 0.031 0.958 0.047 0.361 0.420 

 Emilia-Romagna 0.682 0.187 0.958 0.034 0.926 0.086 

 Tuscany 0.721 0.184 0.812 0.024 0.631 0.145 

 Umbria 0.545 0.291 0.966 0.025 0.613 0.399 

 Marches 0.576 0.268 0.947 0.055 0.531 0.355 

 Lazio 0.704 0.128 0.828 0.142 0.590 0.359 

 Abruzzo 0.681 0.209 0.958 0.053 0.871 0.177 

 Molise 0.667 0.210 0.802 0.247 0.552 0.411 

 Campania 0.902 0.082 0.909 0.140 0.626 0.326 

 Puglia 0.597 0.229 0.906 0.098 0.689 0.436 

 Basilicata 0.588 0.268 0.858 0.113 0.513 0.148 

 Calabria 0.690 0.024 0.869 0.131 0.567 0.433 

 Sicily 0.754 0.132 0.889 0.121 0.694 0.284 

 Sardinia 0.594 0.198 0.917 0.112 0.594 0.415 
 

Source: Own calculations. 
 

 
Table 7. Mean efficiency coefficients.  

 
Production function 2003 2005 2007 Average Std.dev 

Cobb-Douglas 0.782 0.667 0.642 0.697 0.075 

Translog 0.942 0.928 0.862 0.905 0.035 

Inefficiency model 0.37 0.731 0.779 0.627 0.07 
 

Source: Own calculations. 
 

 

demand. It should be noted that the estimated results are 
positive but less than unity. This means that increase in 
the quantity of the first input would increase utility of the 
other input in the pair and thereby improve the marginal 
product of the decision-making unit. According to the 
calculations, the positive joint contribution of the inputs 
utilized area and annual working units is represent by the 
0.29% increase of the final outputs. In the case of 
permanent crops and intermediate consumption - their 
pair would increase the output level by 0.45%. The same 
relationship is estimated for the joint contribution of 
permanent crops and employed annual working unit, and 
annual working unit and employed annual working unit, 
which contribute for the output increase – 0.49 and 0.33% 
respectively. The values of the elasticity between the 

utilized agricultural area and permanent crops (b12), 
utilized agricultural area and intermediate consumption 

 
 
 

(b13), intermediate consumption and annual working unit 

(b34) are less than 0, which suggests that they are 
substitute inputs. The mean technical efficiency of the 21 
regions in Italy is estimated to be 69.7% (Table 6).  

During the observed period, agricultural enterprises 
produced 70% of the maximum attainable output. 
Estimates of technical efficiencies based on the frontier 
production function show a relatively high efficiency level 
(Table 7). The results imply that more than 90% of 
agricultural enterprises operate close to the efficient 
production frontier. Taking into consideration institutional 
influence and included variables in the inefficiency model, 
the mean efficiency results have also undergone positive 
trends from the lowest level in 2003 (0.370) to the highest 
level in 2007 (0.779). The contribution of efficiency 
changes to total factor productivity results in increased 
productivity growth. 



 
 
 

 

SUMMARY AND RECOMMENDATIONS 

 
Two-sectored models that distinguish between 
agricultural and non-agricultural production functions was 
developed in the present analysis. Variables included in 
the first production function represent inputs that are 
related directly to the final product and are utilized during 
the production process. Non-agricultural production 
function includes institutional factors that exert different 
effects over technical efficiency coefficients of enterprises 
in the data set. Production functions are structured under 
the assumption for non-linear relation among their value 
and the value of the technical efficiency. The equilibrium 
occurs when: 
 
1) Prices solve enterprises‟ profit optimization problem;  
2) Prices and profits solve enterprises‟ shareholders or 
members utility optimization problem;  
3) Shareholders or members are indifferent in quitting 
their enterprise and starting new business activity. 

 

The main results from the performed stochastic frontier 
analysis lead to several conclusions regarding the 
organization of the production process. Balanced 
productivity growth in the 21 regions is supported by the 
contribution of efficiency change to the total factor 
productivity. Specialization in the agricultural sector 
appears to reduce production costs and specialized 
mixed farming improves land utilization. Geographical 
clustering additionally enhances relationships between 
producers, their cooperatives, and the final customer. 
Distribution of the labor input also has an underlying 
effect on efficiency growth. Nevertheless, there are some 
internal organizational lapses towards employment on a 
regular basis. The allocation of the labor in specialized 
production is not entirely consistent with the inter-firm 
utilization.  

One possible solution is a combination between 
individual responsibility and division of labor for each 
performed operation or task. The results allow for some 
generalizations towards the salary expenditures. The 
level of wages is sufficient enough to eliminate the 
incentive for the employee to leave enterprise for another 
employment opportunity. Consequently, it could be 
inferred that the average wage is responsive to market 
wage. Considering specific qualifications which work in 
the agricultural sector, requires the last assumption which 
provides that offered wages stimulate a flow of seasonal 
workers from regions with lower income. Legal 
enforcement of contracts contributes to enterprises‟ 
empowerment and collective action. However, this result 
should be taken with precaution in relation to land input. 
Contract farming does not benefit the poorest part of the 
rural population, but rather absentee landlords and large-
scale producers. Cooperatives prove to be dynamic and 
influential organizational structures with stable relative 
share among the other legal entities in the agricultural 
sector. Their contribution to the overall technical 

 
 
 
 

 

efficiency is through the balancing of market demand and 
producers' production choice, by implementing fair pricing 
and quality standards. 
 

 

Conclusion 

 

This paper draws attention to the opportunity to evaluate 
the influence of certain institutional factors and their 
contribution to economic efficiency. Two main 
conclusions are reached from the analysis. The first one 
focuses on those efficiency resources that could be 
exploited by the enterprises, but not modified or changed 
by them. They refer to the institutional environment and 
its attributes – legislation, enforcement regimes and 
ensuing contracts arrangements, and the social 
characteristics of the informal relations. To this effect, the 
second conclusion suggests that the interaction between 
the actors of this bilateral process “strategic partnerships-
institutions” has a broader impact, in addition to improving 
and benefiting economic performance of the agricultural 
enterprises. Cooperatives and producer organizations 
contribute to sustainability of the region in which they are 
established. They influence various aspects of the 
agricultural value chain through balanced and competitive 
relationships; concentration of the primary agricultural 
production, higher economic effi-ciency, and development 
of rural areas. 
 

 
REFERENCES 

 
Adkins LC, Moomaw R, Savvides A (1998). Institutions, freedom and 

technical efficiency. Southern Econ. J., 69: 92-108.  
Aoki M (2001). Toward a Comparative Institutional Analysis. Cambridge, 

MA: The MIT Press, pp. 35-181  
Legge regionale 24 gennaio (2006). n. 1., Modifiche alla L.R. 16 aprile 

2003, n. 5 "Provvedimenti per favorire lo sviluppo della 
cooperazione" (B.U.R. n. 13 del 02.02.2006 - Errata corrige B.U.R. n. 
17 del 09.02.2006)  

Bromley DW (1989). Property Relations and Economic Development: 
The Other Land Reform. World Dev., 17(9): 867-877.  

Brunori G, Cerutti R, Medeot S, Rossi A, Vanni F (2002). Marketing 
sustainable agriculture: an analysis of the potential role of the new 
food supply chains unsustainable rural development; Macro-level 
analysis of food supply chain dynamics and diversity, National report  
– Italy, pp. 5-41  

Decreto n.85/TRAV Attuazione del decreto legislativo 27 maggio 2005, n. 

102, sulla regolazione dei mercati, a norma dell‟art. 1, comma 2, lettera 

c), della legge 7 marzo 2003, n. 38, recante i requisiti minimi per il 

riconoscimento delle organizzazioni di produttori, le modalità per il 

controllo e per la vigilanza delle organizzazioni dei produttori, al fine di 

accertare il rispetto dei requisiti per il riconoscimento nonché le modalità 

per la revoca del riconoscimento.(integrato con le modifiche alla Tabella 1 

introdotte dal D.M. prot. n. 121/traV del 10 marzo 2008) Economic 

Freedom of the World (2010). Annu. Rep., p. 1.  
Hathaway DE (1963). Government and Agriculture:Public Policy in a 

Democratic Society.New York: Macmillan Company. 1-180.  
Hospers G, Beugelsdijk S (2002). Regional Cluster Policies: Learning 

by Comparing?KYKLOS,ISTAT,Information Syst. Agric. Livest., 5: 
381-402.  

Kodde DA, Palm FC (1986). Notes and Comments:Wald Criteria for 
Jointly Testing Inequality Restrictions. Econometrica, 54(5): 1243-
1248. 



 
 
 

 
Kohli Ul (2010). Labour productivity: Average versus Marginal, Ch. in 

W.E. Diewert; B.M. Balk, D. Fixler, K. J. Fox and A. O. Nakamura 
(2010). “Price and productivity measurement”, Trafford Press, 6: 103-
132.  

Macneil IR (1974). The many futures of contracts. Southern California 
Law Rev., 47: 691-816.  

Milgrom P, Roberts J (1992) Economics, Organization and 
Management, Prentice Hall, p. 185.  

Powell WW, DiMaggio PJ (1991). The new institutionalism in 
organizational analysis. University of Chicago Press (Chicago, Il) 
pp.1-38 
 

Rodrik D, Subramanian A, Trebbi F (2002). Institutions rule: the primacy 
of institutions over geography and integration in economic 
development. NBERWorking, p. 930.                                    

 
 
 
 

 
Rodrik D (2000). Institutions for high-quality growth: What they are and 

how to acquire them. NBER Working, p. 7540.  
Rural development in Italy and its evolution,National Institute for 

Agricultural Economics,Options Mediterraneennes, Ser.A /number 71, 
2006. 

 
Sobel S, Coyne Ch.J (2010). How are institutions related? In Economic 

Freedom of the World: 2010 Annual Report,163-174.  
Terms and conditions of the scheme of aid for producer organizations in 

the fruit and vegetables sector (2009).  
Wijeweera A, Villano R, Dollery B (2010).Economic Growth and FDI 

Inflows: A Stochastic Frontier Analysis, J.Dev. Areas, 43(2):143-158. 
Williamson Ol  (1981).  The Modern Corporation:  Origins, Evolution,  

Attributes. J. Econ. Lite., 19: 1537-1568. 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 



 
 
 

 

APPENDIX 

 
Appendix A. Application of frontier models in studying institutional influence over economic development.  

 
Author Analysis and Results  
 
Johnson (2006) 
 
 
Blonigen (2005) 
 

 

Meon and Weill (2005) 
 
 

 

Wijeweera et al. (2004) 
 
 

 

Fulginiti et al. (2004) 
 
 
 
Adkins et al. (2002) 
 
 
 
Rodric (2000) 
 
 

 

Sedik et al. (1999) 
 

 

De Mello (1999) 
 

 

Bauer et al. (1998) 
 

 

Borensztein et al. (1998) 
 
 
Dowson (1998) 
 

 

Edwards (1998) 
 

 

Rodric (1997) 
 

 

Moroney and Lovel (1997) 

  
Foreign direct investment should exert positive effects on economic growth in developing countries which 
suffer from low productivity and capital stock deficiencies. 

 

Foreign direct investments (FDI) lead to an increase rate of economic growth. 

 

Applied Stochastic Frontier Analysis (SFA) to study how difference in quality of institutions (control of 
corruption, strength of the law, quality of the regulatory framework) may explain cross-country differences in 
aggregate efficiency. 

 

Applied SFA to study the relationship between FDI and the rate of growth of GDP. The result show that FDI 
inflows exert a positive impact on economic growth only in presence of a highly skilled labor force, 
corruption has negative impact; trade openness gains efficiency. 

 

Used panel data on output and conventional agricultural inputs for 41 SSA countries to study productivity 
during political conflicts and wars for the period 1961-1999. The used institutional variables are: colonial 
heritage, independence; armed conflict; political rights/civil liberties. 

 
Estimate production frontier and study to extend economical and political institutions contribute to the 
technical inefficiency, output growth, TFP growth. 

 

Identified 5 types of institutions that permit adequacy of the market: institutions for property rights; 
regulatory institutions; institutions of macroeconomic stability; institutions for social insurance; institutions 
for conflict management. 

 

Took different approach to study farm restructuring in Russia for the period 1991-1995 by concentrating on 
technical efficiency. Efficiency scores are explained by economic and institutional factors: farm size, 
softness of budget constraint; deterioration in terms of trade and region level specialization. 

 

Foreign direct investments‟ (FDI) contribution depends on host country characteristics (skilled labor). 

 

Set consistency conditions for regulatory analysis of financial institutions which include: efficiency levels, 
ranking, and identification of best and worst firms, time, competitive market conditions, and standard non-
frontier measures of performance. Applied four approaches – DEA, SFA, TFA and DFA 

 

Foreign direct investment (FDI) has positive impact on GDP and is a function of the human capital. 

 
Economic growth is associated with economic freedom, because the latter has positive effect on investment 
and total factor productivity (TFP). 

 
Determinants of the TFP growth are the initial per capital GDP, the initial level of human capital and the 
degree of openness. 

 
Political factors affect economic performance, democracy is associated with stable long-run growth rates; 
better short-run stability; ability to deal with adverse shock; higher wages. 

 
First applied SFA to compare productive performance of planned and market economies. EE Countries 
were no more that 76% as efficient as the western European economies during 1978-1980. 
 

 
Bergson (1987, 1989, 1991)  Planned economic tend to use capital and land less efficiently (CRS, dummy variable technical efficiency)  



Zaimova 461 
 
 

 
Appendix B. Loglikelihood results of translog production function and inefficiency model.  
 
 

Variable Parameter 

 2003   2005   2007  
 

 Coefficient St. error T-ratio Coefficient St. error T-ratio Coefficient St. error T-ratio 
 

 Stochastic frontier model           
 

 Constant β0 -0.7908* 1.0000 -0.7908 0.6725 0.8839 0.7608 -0.2029* 0.6432 -0.3155 
 

 Utilized agricultural area (UAA) β1 0.0000 0.2298 0.0000 -0.1172* 0.1206 -0.9722 -0.1238* 0.1809 -0.6845 
 

 Permanent Crops (PC) β2 0.8247 1.0000 0.8247 0.9692 0.1595 0.6075 0.1031 0.9108 0.1132 
 

 Intermediate consumption (IC) β3 0.8303 0.3969 0.2091 0.9025 0.4833 0.1867 0.8123 0.4748 0.1710 
 

 Annual Working Unit (AWU) β4 0.2616 0.1000 0.2616 0.9134 0.2588 0.3528 0.3913 0.1208 0.3236 
 

 Emloyed annual working unit (EAWU) β5 0.2589 0.2721 0.9515 0.1012 0.2633 0.3844 -0.7683* 0.3692 -0.2080 
 

 Time β6 0.1466 0.1000 0.1466 0.5647 0.1176 0.4801 -0.8315* 0.7819 -0.1063 
 

 0,5*(UAA)
2
 β7 0.4705 0.8233 0.5715 0.3828 0.1189 0.3218 -0.9134* 0.1243 -0.7346 

 

 0,5*(PM)
2
 β8 0.1815 0,1000 0.1815 0.2735 0.2323 0.1177 0.1151 0.5962 0.1931 

 

 0,5*(IC)
2
 β9 -0.1377* 0.9337 -0.1475 0.4670 0.2609 0.1789 0.9566 0.7573 0.1263 

 

 0,5*(AWU)
2
 β10 0.1335 0.1000 0.1335 -0.1071* 0.2022 -0.5298 -0.5040* 0.4867 -0.1035 

 

 0,5*(EAWU)
2
 β11 0.3775 0.3999 0.9438 -0.8212* 0.2204 -0.3725 -0.6061* 0.8856 -0.6843 

 

 (UAA)*(PC) β12 0.2043 0.1000 0.2043 -0.4764* 0.5137 -0.9273 -0.5889* 0.1021 -0.5767 
 

 (UAA)*(IC) β13 -0.7427* 0.5337 -0.1391 -0,2406* 0.8597 -0.2798 -0.1773* 0.4611 -0.3846 
 

 (UAA)*(AWU) β14 0.0000 0.1000 0.0000 0,2093 0.5368 0.3899 0.0000 0.1000 0.0000 
 

 (UAA)*(EAWU) β15 0.2376 0.1000 0.2376 0,0000 0.1000 0.0000 -0.2910* 0.5672 -0.5130 
 

 (PC)*(IC) β16 0.5685 0.7191 0.7905 0,5632 0.2423 0,2324 -0.2454* 0.2914 -0.8421 
 

 (PC)*(AWU) β17 0.6730 0.1000 0.6730 -0,4686* 0.3908 -0,1199 0.7666 0.1292 0.5932 
 

 (PC)*(EAWU) β18 0.2785 0.6693 0.4161 0,6672 0,1276 0,5227 0.1459 0.1341 0.1087 
 

 (IC)*(AWU) β19 0.6684 0.1000 0.6684 -0,2201* 0,2263 -0,9729 -0.9165* 0.5501 -0.1665 
 

 (IC)*(EAWU) β20 -0.5370* 0.9312 -0.5766 -0,7264* 0,3292 -0,2206 -0.2433* 0.3078 -0.7903 
 

 (AWU)*(EAWU) β21 0.4032 0.1000 0.4032 0,2157 0,5706 0.3780 0.6075 0.5817 0.1044 
 

 Variance parameters           
 

  s² 0.1262 1.0000 0.1262 0.0128 0.0474 0.2693 0.0063 0.0566 0.1114 
 

  γ 0.5000 1.0000 0.0500 0.7348* 0.1030 1.3086 0.7181 1.5305 0.1163 
 

 Loglikelihood function   3.1885   8,1042   19.2112  
 

 Inefficiency effects model           
 

 Training level (TrLevel) δ1 -0.6378* 0.5959 -0.1070 0,3458 0.6609 0.5232 0.2008 0.1421 0.1412 
 

 Specialized mixed farming (SMF) δ2 -0.4592* 0.1594 -0.2879 0,1037 0.1241 0.8360 0.3826 0.1144 0.3344 
 

 Property right protection (PRP) δ3 0.2641 0.8656 0.3051 -0.3030* 0.6375 -0.4753 -0.2213* 0.1734 -0.1276 
 

 Legal enforcement of contract (LEC) δ4 0.2423 0.1403 0.1727 -0.3307* 0.1132 -0.2921 -0.7365* 0.1230 -0.5987 
 



 
462 J. Dev. Agric. Econ.           

Appendix B. Contd.           
            

 Starting new business (SNB) δ5 -0.9564* 0.1386 -0.6899 0.5733 0.5134 0.1116 -0.3348* 0,2461 -0.1360 

 Number of cooperatives (C) δ6 - - - - - - -0.1872* 0.8617 -0.2173 

 Variance parameters           
   s² 0.0106 0.0036 2.9043 0,0078 0.0024 3.2043 0.0134 0,0041 3.2485 

   γ 1.0000 0.0147 67.8640 1,0000 0.0293 34.1114 0.8354 0,0005 0.6465 

 Loglikelihood function   20.0499   17.7180   3.2169  
 
*Signifies that the estimated parameters in bold can be accepted at 5% significance level (Source: Own calculations). 


