Economic Efficiency and Land Rights

--A Stochastic Frontier Analysis of Agricultural

Production in China

Yanjie Zhang

Department für Agrarökonomie und Rurale Entwicklung, Georg-August-

Universität Göttingen
Platz der Göttinger Sieben 5 D-37073 Göttingen, Germany
Tel: +49 551 394821
Fax: +49 551 3912177
Email: yzhang1@gwdg.de or zhang@iamo.de

1
1. Introduction
The rural reform started at late 1970s improved farmers’ incentives and has
had great impacts on China’s agricultural production and productivity growth.
Interested in China’s dramatic agricultural development since the reform,
numerous authors have made efforts to explain changes in productivity and
explore the growth sources behind it, including Lin (1992), Fan (1997, 1999),
Lambert and Parker (1998), Brümmer et al. (2006) and so on. These studies
convey some core messages: While the productivity improvement and
technological progress are spectacular over the last two decades, the
performance in terms of efficiency change is not so inspiring.

The negligible changes in allocative efficiency during the whole

observation period might be caused by frequent adjustments of the market
conditions, and missing land transfer rights. These unstable and uncertain
conditions lead to increasing adjustment costs which in turn might hinder
farmers to discover economically efficient production plans (Brümmer et al.,
2006). There are also consensuses that technical efficiency has improved
greatly over the whole period of the 80s due to a series of institutional
arrangements, especially the implementation of household responsibility
system1 (HRS). Entering the 1990s, the situation of technical efficiency has
deteriorated gradually and the reasons behind it may be more involved.
Compared with the former period of the reform, there seems to be lack of
greater institutional incentives to arouse farmers’ enthusiasm for agricultural
production.

In transition countries, there are evidences of the institutional

incompleteness or failures and even the lack of complementary institutions
accompanying the reform course. Such as market uncertainties and
incompleteness, poorly developed credit markets, incomplete legal system,
information asymmetries and path-dependencies influencing the efficiency of
the privatization process, have been found to have significant efficiency and

1
Known as baogandaohu (literally “contracting everything to the household”), a system that
ultimately allowed farmers more control over both the crops they produced and the surplus
produced by their household over the government procurement quotas (Bhattacharyya and
Parker, 1999).

2
growth implications (see Brandt et al., 2002; Curtiss, 2002). Considering the
imperfect development of land issues in China, the remaining ambiguity over
land tenure rights seems closely relevant to the deterioration of technical
efficiency after 1990 as well as negligible allocative efficiency. One event from
last year highlights the necessity to study the relations between land rights
and efficiency change. After eight years’ controversy and reading, the
Property Law of the People’s Republic of China was approved on March 16,
2007 and will go into effect on October 1, 2007. The law stipulates equal
protection of state, collective and private property rights. As for the aspects of
rural land issues, it says: In order to give farmers long-term, guaranteed land-
use right, after expiry of the contract of arable land, grassland and woodland,
those who own the contracted land management rights continue to contract in
accordance with relevant state regulations.

It should be useful to examine the institutional environment and reform

policy effects on land tenure and related property rights in the past. Identifying
the contribution of insecure land tenure system to efficiency change will be of
great importance to productivity improvement, and also the future course of
China’s rural reform.

This study will estimate the productivity change and efficiency change of
China’s agricultural production since the reform, from an institutional
environment perspective. More interestingly, we are trying to explore how
effective the existing land tenure and related property rights systems have
been in providing households with incentives to ensure the development of
agricultural production and productivity progress, to what extent they have
impacted the farmers’ efficiency.

2. Theoretical framework
2.1 Economic efficiency

2.1.1 Leibenstein’s theory of X-efficiency

Leibenstein (1966) originally introduced the concept of X-efficiency from the

psychological aspects of human nature, which is defined as the difference

3
between the minimal and the actual costs of a production, or as the difference
between the actual and the maximal (potential) output (Leibenstein, 1975).
This notion exposes the important inefficiency source of an inability and/or
unwillingness to achieve fully-specified organizational (contractual)
relationships and obligations, rights and responsibilities, co-ordination and
focus, as well as technical incompetence (Dean and Perlman, 1998).
Leibenstein (1966) proposed three determinants of X-efficiency: intra-plant
motivational efficiency, external motivational efficiency, and non-market input
efficiency. More specifically, those could be the situation as contracts for labor
are incomplete, the production function is not completely specified or known,
and not all inputs are marketed or, if marketed, they are not available on equal
terms to all buyers. Thus the common assumption of cost-minimization for all
firms does not hold and X-efficiency has an important impact on the level of
unit cost. The motivational factors for the measurement of X-efficiency, such
as competitive pressure, are hence considered to partially explain the residual
in economic growth (Leibenstein, 1966).

2.1.2 Farrell’s measures of economic efficiency vs. X-efficiency theory

Farrell proposed the measure of economic efficiency in 1957. Conceptually,

Farrell’s measure of economic efficiency and X-efficiency theory are identical,
both seeking to explain why firms may not reach their optimization of
production. Whereas there are important distinctions between them: as
Leibenstein (1977) himself stated, X-efficiency deals with problems that are
intrinsic to the nature of human organization both within and outside of the
firm, whose work in this sense has challenged and also enlarged upon the
basic assumptions of neoclassical economics (Perlman, 1990); whereas
Farrell’s work is concerned with the empirical measurement of productive
efficiency within an essentially well-established neoclassical economic
modeling framework (Button and Weyman-Jones, 1992). Farrell’s production
frontier (potentially optimal production) is estimated under the assumption of
perfect market, which is usually not consistent with the ideal one according to
Leibenstein’s definition. That is because the best farms under Farrell’s
estimation are conceived to be affected inevitably by some of the X-efficiency
determinants, such as the principal-agent problems, inertia or selective

4
rationality. Hence a loss of empirical evidence identifying the X-inefficiency
sources by using Farrell’s (1957) economic efficiency measures should be
expected (Curtiss, 2002).

2.1.3 Institutional efficiency

Generally speaking, Institutions refer to different types of organizations,

markets, contracts, cultural rules, informal rules and formal laws defining the
rights of access to goods and services and the management of natural
resources (Hatzius, 2000). Initial from dealing with a firm’s production as
combination of output and input factors in neoclassical economics, Coase
(1937) and Williamson (1985) extended the meaning of economic activity of
production which incorporates market and non-market co-ordination as well
as firm internal co-ordination. Therefore, besides neoclassical production
factors, principal-agent relationships within the firm and market and non-
market institutions are also decisive for the production outcome (Curtiss,
2002). Thus the institutional environment factors are included as explanation
to the motivational or incentive X-inefficiency sources connected with
Leibenstein’s theory.

Taking into account market and non-market institutions’ failures or

incompleteness, the concept of institutional efficiency is introduced to capture
inefficiencies resulting from missing or distorted information, or the absence of
competitive pressure (Curtiss, 2002). The existence of institutional inefficiency
is obvious in transition economies, especially in the agricultural sector in
general. The high trust and confidence required for the well functioning
market-led economy are not always taken for granted in transition economies.
As Hobbs et al. (1997) state, the fostering of market institutions to reduce
transaction costs is pivot to the development of market-oriented economy in
transition countries. The effects of market and non-market institutions on
production activity of course should be considered and incorporated in
traditional measures of economic efficiency. Curtiss (2002) has done some
attempts as specifying these institutional effects into the three components of
economic efficiency: technical efficiency, scale efficiency and allocative
efficiency. These components constitute total institutional efficiency and are
closely connected with the particular main institution representatives, factor

5
market efficiency, product market efficiency and non-market institutional
efficiency. Hence the institutional efficiency provides us an approach to solve
the compatibility problem of Farrell’s measures of economic efficiency with
Leibenstein’s theory of X-efficiency.

2.2 Land tenure and property rights in transition economies

Property rights to land is assigned to a class of institutional arrangements,

according to Feder and Feeny (1991)’s classification of three basic categories
of institutions2. They state further, in developing countries that are undergoing
evolution in all three categories of institutions there is the potential for a lack
of congruence among the three types of institutions. Hence the economic
analyses of land rights issues in developing countries can not omit the
institutions pillar, as the usual presumption of exclusive, transferable,
alienable, and enforceable private property rights to land is frequently
inaccurate and potentially misleading (Feder and Feeny, 1991). Ho and Spoor
(2006) also point out, that the choice and timing of successfully creating new
institutions should be consistent with the development level of a special
society characterized by its societal, economic, political and cultural factors.

Economists of the property rights school argue that, a clear assignment

of property rights is a prerequisite for efficient allocation of resources,
whereas flexible and fluid land tenure equivalent to insecure tenure will lead to
market inefficiencies. Thus they conclude a privatized property of land is the
sole most efficient and secure institutional arrangement for an economy with
markets restructuring. On the contrary, empirical studies show the evidence of
negative effects of land privatizing reform as in many African countries (Ho
and Spoor, 2006). As important supplementation and clarification, Ho and
Spoor (2006) address two accounts which are often equivocal when

2
There are three basic categories of institutions: constitutional order, institutional
arrangements, and normative behavioral codes. The constitutional order refers to the
fundamental rules about how society is organized—the rules for making rules. Institutional
arrangements are created within the rules specified by the constitutional order. These
arrangements include laws, regulations, associations, contracts and property rights. The third
category, normative behavioral codes, refers to the cultural values which legitimize the
arrangements and constrain behavior. The constitutional order and normative behavioral
codes evolve slowly; institutional arrangements may be more readily modified (Feder and
Feeny, 1991).

6
contesting. First, increased tenure security can be accomplished through
formal, as well as informal institutions. Second, in contrast to the assumptions
of Western economic advisors to national governments in developing and
transition countries, secure tenure should not be equalized with private
property per se. They conclude further, in a low economic development level
with large proportion of small-scale farmers, the state will risk creating new
institution nothing more than a paper agreement with little or even a negative
effect on the actions of social actors, if land titling is implemented.

Based on many historical cases, Deininger and Feder (1998) draw three
conclusions in favor of more secure and fuller property rights arrangements.
First, increased tenure security (not necessarily equivalent to formal title) has
an important impact on increased investment, with the increasing of
population. Second, higher degree of transfer rights provides additional
incentives for investments and for more efficient use of family labor. Finally,
the ability to use land as collateral to increase access to medium and long-
term formal credit markets is of importance if foreclosure is feasible.

2.3 Land rights and economic efficiency

As an institutional arrangement, land tenure and related property rights will

naturally be connected with the efficiency of agricultural production, through
the introduction of the concept of institutional efficiency. For transition
economies, we adopt Leibenstein’s theory of X-efficiency as our theoretical
foundation to explain the existing efficiency gap between perfectly competitive
market and real market. Together with the addition of institutional efficiency,
the traditional Farrell’s measures of economic efficiency can be applied to
estimate efficiency score and its components. The theoretical framework of
the study is illustrated in Figure 1.

For the case of China’s agricultural production, as already discussed in

the introduction session, the remaining ambiguity over land tenure rights
seem to show a robust explanation power for the source of unexpected
efficiency performance. Considering the hot dispute about institutional reform
related to land issues currently in China, whether it is the suitable choice and

7
right timing for the creation of new institutions will also be an interesting topic
to explore.

Perfectly Efficiency Real

competitive gap market
market

Institutional economics
Neo-classical economics

Farrell’s Leibenstein’s Institu-

measures of theory of tional
economic X-efficiency efficiency
efficiency

Land tenure Other

and property institutional
rights environment

Figure 1: economic efficiency and land rights in transition economies

Source: own figure

3. Land rights in rural China

3.1 The history of land issues

Land ownership was feudal in China before the 1949 Revolution, with only
10% of landlords controlling 70-80% of agricultural land. Since 1949, the
government implemented large-scale land reform confiscating those
landlords’ land without compensation and redistributing to peasant farmers
(Fan et al., 2002). Under the background of adoption of the heavy industry-
oriented development strategy, where a large amount of grain and other
agricultural products are demanded to support the urban industrialization, the
government began to organize large collective production system since 1953.
The land was collective owned and the production cooperative organization

8
evolved to greater scale of People’s Commune in the Great Leap Forward
(1957-1960) period. But afterward the production was decentralized into
smaller units of so called production teams due to the Great Famine (1959-
1961) disaster. This collective-based system of agricultural production
implicating collective ownership of land remained nearly 3 decades until 1979.

Due to the main problem of incentive structure of collectivization policy,

the government started to address a series of reform in the late 1970s. The
reforms dismantled the commune system, and implemented household
responsibility system (HRS) attempting to introduce private property
incentives while leaving formal ownership of the land in state hands or
collectively owned. Land use rights and residual income rights were granted
to individual farm households between 1979 and 1983 (see Brandt et al.,
2002). According to Lin and Zhang (1998), the duration of the initial land
allocation was 15 years and extended to another 30 years after existing
contracts expired in 1993.

As the growth rate of grain output slowed down in late 1980s, there is a
need to inspect the ongoing reform policy. Lin and Zhang (1998) point out that
the current household responsibility system for land is actually a village-based
communal land tenure system, and the individual farm households do not
have the legal titles to land. They further argue that the restrictions on land
markets, frequent land redistribution according to population changes and
small scale landholdings rooted in this village-based land tenure system have
become the major hindrance to the efficiency of resource allocation and the
improvement of agricultural productivity.

3.2 Empirical results related to land rights

Although land issue topics are currently often contested, there is relatively
little empirical research studying the impact of land rights on economic
efficiency. While some scholars have presented results related to land issues,
which might be helpful for the processing of our study. Using satellite images
to examine the changes of the area of cultivated land and its potential
agricultural productivity in China, Deng et al. (2006) find that between 1986
and 2000 China recorded a net increase of cultivated land (1.9%), which

9
almost offset the decrease in average potential productivity, or bioproductivity

(-2.2%), which means that conversion of cultivated land has not hurt China’s

national food security. They also argue that more recent change in cultivated
area likely has had little adverse effect on food security. What is interesting to
us is, in an international perspective, they state that the loss of cultivated land
is not definitely due to the weak property rights, while losses to development,
abandonment due to low profitability and conservation set aside.

Through their survey work, Brandt et al. (2002) suggests enormous

heterogeneity of land rights at the village level. In some villages farmers have
the right typically associated with a private property regime, albeit short of
being able to buy or sell the land. In other villages, tenure is shorter and the
use of the land is constrained in various ways. For the reason of reallocation
decisions, they state a more important explanation as the incentives of local
leaders, which are directly tied to leaders’ responsibilities for fulfilling state
policy, and possibly also to opportunities for rent seeking. They conclude that
in the long run China needs a land management system that provides long-
term security of tenure and promotes the efficient use of land. Secure use
rights and the expansion of rental markets are ways to facilitate the
reorganization that is required in the farm sector, however, this policy will only
be effective if the incentives of local leaders are aligned in this direction.

The results of Li et al. (1998) show that land tenure and associated
property rights in rural China affect the farmers’ production behavior, and that
the right to use land for long periods of time encourages the use of land-
saving investments. Whereas the results clearly show there are gains of
reforming China’s land system, considering the lack of land courts, land
registration system, or good credit markets for farmers, they propose the
question for policy makers whether or not at this stage of China’s
development or during this point of time in the economic transition the gains
are worth the costs or risks.

10
4. Methodological framework
4.1 Productivity change and efficiency

The productivity of a production unit is defined as the ratio of the output(s) that
it produces to the input(s) that it uses. Productivity change occurs when an
index of outputs changes at a different rate than an index of inputs does
(Kumbhakar and Lovell, 2000). A change in productivity can be caused not
only by a change in efficiency but also by a change in the production
technology and the environment in which the production unit operates (Lovell,
1993).

In early econometric studies, productivity change was allocated

exclusively to shifts in production technology (the magnitude of neutral
technical change); eventually roles were also assigned to the biases of
technical change and the structure of the technology (scale economies)
(Kumbhakar and Lovell, 2000). However, in practice not all production units
are equally efficient provided the same production technology and
environment. So efficiency change should also contribute to productivity
change. Farrell (1957) proposed that the efficiency of a production unit
consists of two components: technical efficiency, which reflects the ability of a
firm to obtain maximal output from a given set of inputs, and allocative
efficiency, which reflects the ability of a firm to use the inputs in optimal
proportions, given their respective prices and the production technology.
These two measures are then combined to provide a measure of total
economic efficiency.

There are many different methods used to estimate efficiency if the

assumption that all production units are fully efficient is relaxed. Lovell (1993)
provides an excellent introduction to this literature. The two principal
approaches that have been used are data envelopment analysis (DEA) and
stochastic frontier analysis (SFA), which involve mathematical programming
and econometric methods, respectively. Both approaches have their pros and
cons. Unlike the DEA approach which is deterministic in the sense that it
assumes that all deviations from the frontier are only due to inefficiency, the
stochastic frontier approach allows for random disturbances, such as weather

11
conditions, the effects of pest and diseases, and measurement errors in the
output variables. As Coelli et al. (1998, p219) observe SFA approach is likely
to be more appropriate in agricultural applications, especially in developing
countries, because of these reasons.

4.2 The SFA distance functions

Aigner, Lovell and Schmidt (1977) and Meeusen and Van den Broeck (1977)
independently proposed the SFA models. A number of comprehensive
reviews of literature on stochastic frontier estimation are available, including
Førsund et al. (1980), Schmidt (1986), Bauer (1990), Greene (1993) and
Murillo-Zamorano (2004). The basic idea is that, the production frontier has an
error term with two components, one for random effects beyond the control of
the producer (weather, etc.) and another for technical inefficiency, which is
under the firm’s control.

When multiple inputs are used to produce multiple outputs, which are
usually the case in agricultural production, Shephard’s (1953, 1970) distance
functions provide a functional characterization of the structure of production
technology. Input distance functions characterize input sets, and output
distance functions characterize output sets (Kumbhakar and Lovell, 2000,
p28). Distance functions are particularly useful for analyzing agriculture in
transition for the reasons of data availability and behavioral assumptions:
output and input data are often more readily available and often of better
quality, and agriculture in transition is neither competitive nor are all farm
decision makers profit-maximisers (Bruemmer, ?). An output distance function
is defined as

DO ( x, y ) = min{µ : y µ ∈ P( x)} (1)

an output distance function takes an output-expanding approach to the

measurement of the distance from a producer to the boundary of production
possibilities. It gives the minimum amount by which an output vector can be
deflated and still remain producible with a given input vector (Kumbhakar and
Lovell, 2000, p30).

For multiple outputs case, Kumbhakar and Lovell (2000, p49) give an
output-oriented measure of technical efficiency by the function

12
 TEO ( x, y ) = [max{φ : DO ( x, φ y ) ≤ 1}]−1 (2)

this provides us a measurement of technical efficiency with

TEO ( x, y ) = DO ( x, y ) . A stochastic distance function model is given as

1 = DO ( xi , yi ; β ) exp{ui − vi } (3)

to convert equation (3) into an estimable regression model, linear

homogeneity property of output distance functions is exploited, which state
that DO ( xi , λ yi ; β ) = λ DO ( xi , yi ; β ), λ > 0 . So equation (3) can be transformed to

yi
τ −1 = DO ( xi , ; β ) exp{ui − vi } (4)
τ

where τ , the normalization variable, could be one of the outputs, or the

Euclidean norm of the output vector. There are many discussions about the
choice of normalization and related function quality of estimation, including
Kumbhakar and Lovell (2000, p94), and Bruemmer (?).

4.3 Measurement of productivity change and its decomposition

The traditional total factor productivity growth Divisia index is defined as
M K
T F P = ∑ Rm q m − ∑ S k xk (5)
m =1 k =1

where a dot over a variable indicates its rate of change; Rm is the revenue
share of output qm, and Sk is the cost share of input xk. Under the assumption
of profit maximization, after some transformations of equation (4) and (5)
(Bruemmer, ?), the TFP growth index can be rewritten as
M K K
∂ ln DO ∂u
T F P = ∑ ( Rm − µ m ) qm + ∑ (λk − S k ) xk + ( RTS − 1)∑ λk xk − − (6)
m =1 k =1 k =1 ∂t ∂t

K
∂ ln DO ( ) ∂ ln DO ( ) ∂ ln DO ( )
where µm = ; − RTS λk = ; RTS = −∑ : returns to
∂ ln qm ∂ ln xk k =1 ∂ ln xk
scale. The relationship in equation (6) decomposes TFP change into 5
∂ ln DO
components: technological change [− ] , technical efficiency change
∂t
K
∂u
[− ] and scale component [( RTS − 1)∑ λk xk ] , which are connected to
∂t k =1

13
 M
technology; allocative effects for outputs [∑ ( Rm − µm ) qm ] and allocative
m =1

K
effects for inputs [∑ (λk − S k ) xk ] , which are connected to market (Bruemmer,?).
k =1

4.4 Empirical specification of SFA

Now an appropriate functional form is needed to characterize the features of

production technology, which is crucial to the estimates of efficiency, as well
as policy analysis and implications. Appelbaum (1979) and Berndt and Khaled
(1979) generalized the application of the Box-Cox transformation function,
which integrates a variety of functional forms and provides comparison by
parametric tests. The generalized quadratic Box-Cox model, assuming input-
biased technical change, can be written as
J
1 J J
1 J
Yit(δ ) = α 0 + ∑ α j X (jitλ ) + ∑ ∑ α jk X (jitλ ) X kit( λ ) + β1t + β2t 2 + ∑ γ j X (jitλ )t + ε it
j =1 2 j =1 k =1 2 j =1

i = 1,2,…,N, t = 1,2,…,T (7)

where the variables Yit(δ ) and X it( λ ) are the Box-Cox transformations of output

and inputs, respectively, defined as (Box and Cox, 1964)

Yit2δ − 1 X λjit − 1
Yit(δ ) = and X (jitλ ) = (8)
2δ λ

where δ and λ are the transformation parameters to be estimated. Under

appropriate parametric restrictions for the values of δ and λ, the generalized
quadratic Box-Cox transformation yields the four locally flexible functional
forms (i.e., translog, generalized Leotief, normalized quadratic, squared-root
quadratic) as well as the non-homothetic CES and Cobb-Douglas
specifications (Giannakas, Tran and Tzouvelekas, 2003).

For multiple outputs case, applying linear homogeneity property of output

distance functions, our specified SFA distance function can be written as

14
 P Ypit(δ ) J
(λ ) 1 P P Ypit(δ ) Yqit(δ )
τ −1
= (α 0 + ∑ ω p + ∑α j X jit + ∑ ∑ω pq
p =1 τ j =1 2 p =1 q =1 τ τ
1 J J
(λ ) (λ )
P J Ypit(δ ) 1
+ ∑ ∑α jk X jit X kit +∑ ∑ϕ pj X (jitλ ) + β1t + β 2t 2 (9)
2 j =1 k =1 p =1 j =1 τ 2
P Ypit(δ ) J
+∑ κ p t + ∑ γ j X (jitλ )t ) g (l ;θ ) exp {uit − vit }
p =1 τ j =1

where τ , the normalization variable, could be one of the outputs, or the

Euclidean norm of the output vector; Y it(δ ) (p, q = 1,…, P) are the Box-Cox

transformation of multiple outputs; the other definitions are the same as in

equation (7) and (8).

g (l ;θ ) measures the institutional effects of land tenure and related

property rights. l is a vector of variables expected to explain the situation of
land rights with respect to individual households. The proxies of l could be
alternatives reflecting the extent of land tenure and the expanded tenure
security, as the survey results of enormous heterogeneity at village level
suggested by Brandt et al. (2002). Their work shows some farmers have more
freedom to dispose farm land, which is closer to a private property regime.
While in other villages, land use is strictly restrained.

The vits are random errors assumed to be i.i.d. N(0, σv2), and
independent of the uits. uits are non-negative random variables that account
for technical inefficiency in production and satisfy the scaling property (see
Wang and Schmidt, 2002), which means u equals a function of z times a one-
sided error u* whose distribution does not depend on z. Here u* is assumed to
be i.i.d. truncated (at zero from below) normal distribution and u is written as

u = h( z , δ )u * (10)

where z is a vector of exogenous variables (firm characteristics) used to

explain variation in technical efficiency. δ is a vector of unknown parameters
to be estimated.

15
References

Brandt, Loren; Huang, Jikun; Li, Guo and Rozelle, Scott. (2002). Land Rights
in Rural China: Facts, fictions and Issues. The China Journal, No. 47 pp.67-97.

Brümmer, B.; T. Glauben and W. Lu. (2006). Policy Reform and Productivity
Change in Chinese Agriculture: A Distance Function Approach. Journal of
Development Economics, Volume 81, Issue 1, pages 61-79.

Button, Kenneth J. and Weyman-Jones Thomas G. (1992). Ownership

Structure, Institutional Organization and Measured X-Efficiency. The
American Economic Review, Vol. 82, No. 2, pp.439-445.

Coelli, Tim; D. S. Prasada Rao and George E. Battese. (1998). An

Introduction to Efficiency and Productivity Analysis. Kluwer Academic
Publishers, London.

Curtiss, Jarmila. (2002). Efficiency and Structural Changes in Transition: A

Stochastic Frontier Analysis of Czech Crop Production. Aachen: Shaker, 2002.

Deininger, Klaus and Feder, Gershon. (2001). Land Institutions and Land
Markets. Chapter 6 in Handbook of Agricultural Economics, edited by Rausser,
Gordon, and Bruce Gardner, Amsterdam: Elsevier, 2001.

Deng, Xiangzheng; Huang, Jikun; Rozelle, Scott and Uchida, Emi. (2006).
Cultivated Land Conversion and Potential Agricultural Productivity in China.
Land Use Policy 23, 372-384.

Fan, Shenggen. (1997). How Fast Have China’s Agricultural Production and
Productivity Really been Growing? New Measurement and Evidence. EPTD
discussion paper No. 30.

Fan, Shenggen. (1999). Technological Change, Technical and Allocative

Efficiency in Chinese Agriculture: the Case of Rice Production in Jiangsu.
EPTD discussion paper No. 39.

16
Fan, Shenggen; Zhang Linxiu and Zhang Xiaobo. (2002). Growth, Inequality,
and Poverty in Rural China: The Role of Public Investments. IFPRI.

Farrell, M. J.. (1957). The Measurement of Productive Efficiency. Journal of

the Royal Statistical Society Series A 120(3), 253-290.

Feder, Gershon and Feeny, David. (1991). Land Tenure and Property Rights:
Theory and Implications for Development Policy. The World Bank Economic
Review, Vol. 5, No. 1: 135-153.

Giannakas, Konstantinos; Tran, Kien C. and Tzouvelekas, Vangelis. (2003).

On the Choice of Functional Form in Stochastic Frontier Modeling. Empirical
Economics 28: 75-100.

Ho, Peter and Spoor, Max. (2006). Whose Land? The Political Economy of
Land Titling in Transitional Economies. Land Use Policy 23, 580-587.

Kumbhakar, S. C. and C. A. K. Lovell. (2000). Stochastic Frontier Analysis,

Cambridge University Press.

Lambert, D.K. and E. Parker. (1998). Productivity in Chinese Provincial

Agriculture. Journal of Agricultural Economics 49(3): 378-392.

Leibenstein, H. (1966). Allocative Efficiency vs. X-efficiency. American

Economic Review 56: 392-415.

Li, Guo; Rozelle, Scott and Brandt, Loren. (1998). Tenure, Land Rights, and
Farmer Investment Incentives in China. Agricultural Economics 19, 63-71.

Lin, Justin, Yifu. (1992). Rural Reforms and Agricultural Growth in China. The
American Economic Review, Volume 82, Issue 1, 34-51.

Lin, Justin, Yifu and Zhang, Fan. (1998). The effects of China’s Rural Policies
on the Sustainability of Agriculture in China. The 11th Biannual Workshop on
Economy and Environment in SE Asia, Singapore, November 10-13. 1998.

Lovell, C. A. Knox. (1993). Production Frontiers and Productive Efficiency. in

The Measurement of Productivity Efficiency - Techniques and Applications

17
edited by Harold O. Fried, C. A. Knox Lovell, and Shelton S. Schmidt, Oxford
University Press, New York.

Wang, Hung-Jen and Schmidt, Peter. (2002). One-Step and Two-Step

Estimation of the Effects of Exogenous Variables on Technical Efficiency
Levels. Journal of Productivity Analysis, 18,129-144.

18