




































In ternationa l
Scholars
Journa ls

 

African Journal of Agricultural Marketing ISSN 2375-1061 Vol. 4 (5), pp. 001-005, May, 2016. Available online at 
www.internationalscholarsjournals.org © International Scholars Journals 

 

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

Full Length Research Paper 

 

Land marketing and hedonic price model in Turkish 

markets: Case study of Karacabey district of Bursa 

province 
 

Hasan Vural1 and Halil Fidan2* 
 

1
University of Uluda , Faculty of Agriculture, Department of Agricultural Economics, Bursa, Turkey. 

2
University of Ankara, Faculty of Agriculture, Department of Agricultural Economics, Ankara, Turkey. 

 
Accepted 20 January, 2016 

 
The aim of this study is to identify the variables affecting land value. Examined land was selected from farms in 
Karacabey district of Bursa province. Data was collected from 54 farmers by using survey methods and from 
second sources on farms. A hedonic price analysis was conducted to determine the marginal return to different 
land characteristics using an econometric model corrected for correlation. Parcel characteristics such as distance 
to farm, organic matter, potassium, saturated water, pH, phosphate, salinity, size of land were used in easement 
values. As expected, the characteristics did not completely explain the easement values in selling. In this analysis, 
seven factors affected well to land value were used. The problem definition was worked on the basis the analysis 
of the regulation reasons of price of land on the land markets. The land prices were determined by real sale 
factors. In the land areas, the agricultural price of land was municipality-specifically affected by factors such as 
the salinity, soil pH and the land use. The land purchases were coined noticeably less by speculation 
considerations. At the same time, one observes a shrinking commercial probability for plots, which lie in 
municipalities with high salty density in the last years. The reason for this could be the top price regulation for 
agricultural properties. The results of this research can help formulate sale decisions and nationalization. 
 

Key words: Land market, price of land, hedonic price model, property rights. 
 
 
INTRODUCTION 

 
In Turkey, there are some important convenient agricul-
ture areas. One of which is Karacabey plain. Farmers in 
the plain have grown various animal and vegetable pro-
ducts. While some farmers are growing only a product, 
most farmers do variety on production. Through the se-
ven months of a year, climate conditions are appropriate 
for vegetal production.  

Turkish arable land amounts to 27 million hectares (in 
2006), about 30% of the total land area. The rest of the 
land is covered with forests, mountains, marshlands and 
lakes. Seventeen percent of the land in Bursa region is 
cover with plain. In Bursa, arable land occupies 40% of 
total culture area (429 599 ha) (Anonymous, 2005). 

Land valuation is a complex operation (Bible and Hsieh, 

1999). Accuracy in estimation of the value of land 
 
 
 
*Corresponding author. E-mail: fidan@agriankara.edu.tr.  

 
 
 
 

 
may be compromised by a whole range of factors of 
methodological, technical and legal nature. The main 
problem faced by analysts is related to methodological 
objectivity, and the difficulties involved in selecting a 
comprehensive set of variables to reflect the value and 
heterogeneity of the plots to be valued. 

Systematic analyses over the regulation reasons of the 
price formation on the agricultural land market are, how-
ever, hardly present in Turkey. In view of the high adjust-
ment density within the range of the area planning and 
land politics, it is expressed by the participants on land 
market, interested with the land. This considerations are 
revealed on the agreed upon prices for determination of 
data of an agricultural plot in dependence of their charac-
teristics to lead. Decision of market analysts is taken in to 
considerations for the land politics, strictly speaking, for 
the regional planning policy and agricultural policy. For 
example, if the height of the price of land is determined 
predominantly by agricultural factors, agrarian reforms 

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


 
 
 

 

lead to a change of the price parameters. 
Agricultural product can show scientific results over the 

price formation on the agricultural market. We tried the 
regulation factors of the price of land to identify in three 
case examples. With the analysis of the agricultural land 
market of the Karacabey, the strong characteristic 
interested us, which from the proximity to the economical 
center of dense development are Bursa results. In illus-
tration, development of the price of land is shown in 
connection with the land characteristics.  

This, however, is one of the risks involved in using sta-
tistical or econometric modeling and has given rise to 
some criticism (Des Rosiers et al., 2000). Statistical and 
econometric methods are not causal; they simply mea-
sure the relationships that exist between variables pre-
viously selected by the analyst. Though statistics based 
on dependence models provide a quantitative tool to 
ensure the objectivity of results when the model is con-
structed, they do not generally simplify the task of select-
ing the causal or independent variables. In other words, 
the objectivity of the method is no guarantee against a 
biased choice of explanatory variables (Cavailhes and 
Wavresky, 2003).  

The heterogeneity of farmland plots in Karacabey plain 
is such that over a distance of just a few kilometers, 
features such as soil composition, land relief, distance to 
farm centre or distance to road can vary widely. In more 
homogeneous areas, there are fewer factors determining 
the value of land. When this is the case, the construct is 
used to model the valuation needs to include a wide 
range of variables, but by doing so, the econometrician 
clashes headlong with whole purpose of modeling, which 
is to find simple, stable constructs to represent real situa-
tions and link land value to a known, limited and perma-
nent set of price-determining factors (King and Sinden, 
1994).  

Methods were commonly used to implement this 
approach, which include the hedonic technique pioneered 
by Griliches (1971) and formalized by Rosen (1974). The 
relationship between land prices and affected well factors 
have been studied in the hedonic framework by Miranow-
ski and Hammes (1984), Gardner and Barrows (1985), 
Ervin and Mill (1985), King and Sinden (1988) and Cas-
well and Zilberman (1986).  

The analyses represent the work over the agricultural 
lands market. It comes in the end that predominantly agri-
cultural factors affect the price of land formation.  

The purpose of this study, therefore, is to offer a 
proposal for an econometric model to be used in valuing 
farmland. These plots provided the variables that will be 
described below. We investigate empirically how this 
selectivity problem affects the hedonic valuation of the 
effect of land value on agricultural land. The outcome of 
this empirical investigation is that hedonic valuation tech-
niques might give rise to misleading conclusions about 
the effect of land value.  

In this study, some factors, which affected land value, 

 
 
 
 

 

were examined in Karacabey district. Land consolidation 
was done on twelve villages of Karacabey plain. After 
consolidation, irrigation cooperatives were founded and 
agricultural production was supported by government.  

The structure of the paper is as follows; Land markets in 

Turkey, material and methods, results and conclusions 

 

Land markets in Turkey 
 
The market of urban land-use right in Turkey can be 
divided approximately into primary and secondary five 
markets based on different subjects and levels of market 
transaction. The model of land transaction includes grant, 
transfer, leasing, pooling and authorized operation etc. 
 
1.) Primary urban land market, it is formed through grant 
of land-use right within certain periods from urban land-
owner to urban land user. In Turkey, the urban land is 
owned by the private sector. The government is the sup-
plier of urban land in primary market. At present, there 
are three key modes of transactions in urban primary land 
market: grant, leasing, pooling and authorized operation. 
Among them, grant is the most common one including 
pro- cesses by agreement, public bidding and auction. 
The leasing is also a common model of transaction. The 
pooling and authorized operation is the model usually 
implemented in land assets transaction of stock com-
pany.  
2.) Secondary urban land market, it is formed through 
grant of land-use right in residual term of years by gran-
tee holding the land-use right in primary urban land 
market to other land users who utilize the land according 
to specified terms and conditions after paying the grant 
charge. The land user can put this land-use right into 
market based on market rule. Accordingly, the market for 
land-use right transfer (namely secondary land market) 
has been established. The purpose and term of years 
specified in the grant contract cannot be violated while 
transferring the land-use right. Currently, there are four 
key transaction modes in this market: transfer, sub-
leasing, mortgage and pooling. 
 
The relationship between both markets (markets of land 
grant, transfer and leasing) is unified as well as equal 
transfer and competition. Owing to participation of the 
government, the land grant market plays a significant role 
in macro readjustment and control of the market, parti-
cularly in supply-demand relations of land market and 
land price equilibrium. The land transfer market is an 
important part and an energetic point of land markets. Its 
development gives direct presentation of sound land mar-
ket, and plays an accelerative role in land grant market 
(Plantinga et al., 2002).  

In this area, Important of Karacabey was increased from 

the pressure of agricultural growing for the price of land, 
additionally. On the basis of the available investigation, 

the first evaluation of the land characteristic is possible. 



 
 
 

 

A comparison among the land which farmers one buys 
and which leaves flows on the hereditary way into the 
property, however, the probability of this goal to reach to 
land quality contains signs for the creation of right 
inequality, as only one group of prospective customers 
may buy agricultural plots. These statements lead to the 
following policies. They leave themselves in two main 
topics with different temporal realization possibility, sum-
marize to urgent, and faster in measures convertible, a 
flexible organization of the lease land market and a stric-
ter separation agricultural areas are in the agriculture 
zone. With it, a redefinition of the term "land purchase" is 
connected. The results of all case examples point out the 
fact that the market value of agricultural plots will approxi-
mate hardly for productive value. Over the land market, 
agriculture enterprises cannot secure themselves in the 
necessary areas at favorable prices, in order to be able to 
profit from the cost regression. 

 

MATERIALS AND METHODS 
 
The choice of the functional involution forms a relatively difficult task 
which can be used to represent results of the hedonic price model 
(Johnson and Wicherin, 1992). The methods used for decision (t-
test) lead to the choice of a linear involution form. The examination 
of the coefficient of the parameters takes place in connection with 
the introduction of the SPSS. It is assumed that individual para-
meters exhibit different values with the different legal basic con-
ditions. The model was estimated for 54 farms (Table 1). For this 
reason, the purchases of the land were not considered. All condi-
tions for the computation of the involution model (size of land, salty, 
phosphate, pH, saturated water, potassium, organic mater, and 
distance to farm) are fulfilled in the selected linear model.  

In this analysis eight factors that explained land value well were 
used. These factors are: land size, organic matter amount in soil, 
salt, phosphate, pH, saturated water, potassium and distance to the 
farm. Data such as pH, salt, potassium, and saturated water con-
tent were obtained from farm records. Farm records contain these 
data because farm soils have been analyzed due to land consoli-
dation. The reason for the importance of soil as a deter-minant of 
the value of farmland is that it is this feature of the natural environ-
ment that has the greatest impact on its potential uses. It may, 
therefore, be considered to be the main factor when valuing land, 
particularly, because it is a technical and therefore objective para-
meter. On consolidation area, since the land characteristics are the 
main factors affecting the land value were used in study.  

In this study, farm data such as land value, size of land, and 
distance to farm were obtained using survey method. Specifically 
land which sold really was selected. These farms’ lands have been 
sold recent years. For this reason, sampling and analysis was done 
base on real selling price.  

Land was classified into four different categories for organic 
matter criteria which coded as poor, medium, good and very good. 
When examining the effects of distance to farm criterion, metric 
distance scale (km) was used. Data were obtained through a survey 
of the district in question. Professionals specialized in land valuation 
were consulted in order to confirm these values, providing the 
variable expert value.  

Land selling in the region is generally done by property owners. 
However, land is also partly marketed by real estate agents. In this 
case, real estate agent takes commission fee and the amount of the 
commission is established by bargaining method. On the other 
hand, real estates are generally marketed by real estate agents in 
urban areas, contrary to rural areas. 

  
  

 
 

 
In this research, data of land which were sold recently were ana-

lyzed (Aakre et al., 1997). Therefore, with this purpose, data were 
obtained from farmer chamber, the union of irrigation cooperatives 
and county headship of agriculture ministry in this region.  

It is to be accepted that a land- characteristics, different compo-
sition of the affecting attributes or their change, e.g. by economic 
changes, has a large influence on the price of land. Measured 
variables, which determine the price of land development on the 
basis of the capitalization theory over the time (growth of incomes), 
loose under the circumstances at force of expression, described 
above. A hedonic price model is suitable for the empirical analysis 
of the agricultural land market. The hedonic beginning represents 
an indirect method of the preference measurement. It is used in 
order to evaluate individual characteristics of a property, whose 
value consists of a bundle of characteristics (heterogeneous pro-
perty). For heterogeneous goods a direct comparison of market 
prices is impossible, because these are comparable only with 
difficulty with one another (Shonkwiler and Reynolds, 1986). The 
theoretical development of the model is explained on work of 
Griliches (1971). Roses (1974) and Freeman and Myrick (1993) 
independently converted the theoretical bases of Griliches in 
models, which serve as basis for the estimation of the marginal 
values of goods characteristics.  

Today, hedonic price models are not only used for the deter-
mination of the price of land and the landscape values, but also for 
the evaluation of many other heterogeneous goods as for example, 
the quality of consumer goods (Carew, 2000; Fidan, 2004), for the 
evaluation of (not-agricultural) real estates for the economic model, 
Subordinate the water quality (D'arge and Shogren, 1989). The 
hedonic beginning assumes that the prices are characterized by the 
function P(Zi) . In the present work, the property characteristics are 
arranged into three components; the prices are, therefore, charac-
terized by the following function: Equation 1: formula characteristics 
of an agriculture land. In other words: The market price of lands 
shows the value of its agricultural use. In the hypothesis test, it is 
examined whether the characteristic is significant for an additional 
unit of a characteristic other than 0. Hedonic method is used and 
missing prices are calculated by means of regression analysis. In 
equation 1 below, regression is shown as land price. Land price has 
been sold in the base period with a quality of Z. The quality Z is 
here meant as quality-vector, which include the values of several 
quality features.  
 

P(Zi)=P(Zland(d ) Zsatty Zphosp., Zsat.water ZpH, Zpotass. 

Zorg.cmater Zdist.tofarm) 
 
 
This is expressed in a significant t-value. In the investigation addressed 

here one is interested in many cases v.a. in the statement whether a 

characteristic has at all a significant influence. 

 

RESULTS 
 
Descriptive statistic (mean, standard deviation and the 
number data) on Table 1, regression analysis for the he-
donic price model is also shown on Table 2.  

Independent variable: Value of land the relationship 
among factors which affect land value has been given in 
Table 3. There is high positive correlation between orga-
nic matter and land size factors. Changing variables with 
changing value of land level:  

In particular for distance to farm, it is stated that with 

increasing distance to farm, farm land value level 

increases (Pope et al., 1979), it has increased the price of 

land over on the average 6.577 with high increase rate of 



 
 
 

 
Table 1. Result of descriptive statistics of factors.  

 
Factors Mean Std. Deviation N 

Value of land (TL/da) 8.203 2.237 54 

Size of land (da) 63.592 76.794 54 

Salty (%) 60.180 10.320 54 

Phosphate (kg/da) 7.262 0.293 54 

pH 9.380 5.840 54 

Saturated water (%) 59.180 24.260 54 

Potassium (kg/da) 1.610 1.080 54 

Organic matter (%) 0.070 0.500 54 

Distance to farm (km) 2.090 2,060 54 
 

 
Table 2. Regression analysis for hedonic model  

 
   Linear   Logarithmic  

 Variables B Std. Error Sig. B Std. Error Sig. 

 Constant 13.355 7.763 0.092 51.656 17.680 0.005 

 Size of land (da) 0.009 0.005 0.067 -6.313 2.843 0.031 

 Salt (%) -0.067 0.039 0.090 -10.252 8.145 0.215 

 Phosphate (kg/da) -0.603 1.115 0.591 0.416 0.566 0.466 

 pH 0.155 0.068 0.028 1.294 1.029 0.215 

 Saturated water 0.018 0.016 0.269 0.449 0.685 0.515 

 Potassium 0.216 0.309 0.488 -0.112 0.308 0.718 

 Organic matter 0.001 0.001 0.163 1.080 1.029 0.300 

 Distance to farm 6.577 11.183 0.559 0.005 0.005 0.282 

 R2 0.613  0.004 0.542  0.034 
 

 
Table 3. Correlation matrix.  
 
  Value of Size of Salty (%) Phosphorus pH Saturated Potassium Organic Distance  to 
  land land (da)    water  matter farm 

 Land value 1.00 -0.07 -0.21 0.31 0.24 0.36 0.11 0.15 0.14 

 Size of land(da) -0.07 1.00 0.36 0.49 0.68 0.14 0.43 0.80 0.25 

 Salty (%) -0.21 0.36 1.00 0.03 0.37 -0.13 0.06 0.21 0.10 

 Phosphate 0.31 0.49 0.03 1.00 0.59 0.64 0.49 0.65 0.09 

 pH 0.24 0.68 0.37 0.59 1.00 0.44 0.38 0.68 0.34 

 Saturated water 0.36 0.14 -0.13 0.64 0.44 1.00 0.29 0.29 -0.04 

 Potassium 0.11 0.43 0.06 0.49 0.38 0.29 1.00 0.40 0.45 

 Org. matter 0.15 0.80 0.21 0.65 0.68 0.29 0.40 1.00 0.13 
 Dist. to farm 0.14 0.25 0.10 0.09 0.34 -0.04 0.45 0.13 1.00 
 

 

the distance to farm. The faster the distance to farm in-
creases, the higher the prices for areas of arable land are 
paid. The main reason of this was consolidations 
according to farmers.  

Hedonic model analysis was used to investigate rela-
tionships between land value and factors which effect 
land values (I gın and Forster, 2006). Organic matter, pH, 
saturated water, potassium and distance to farm fac-tors 
have positive coefficients, so they affect land value 
positively. Other factors did not affect land value. Gene- 

 

 

rally, factors have small coefficients because land has 
similar soil characteristics and land values are not more 
different. It is seen that main factor which affect land 
value is distance to farm (I gın and Forster, 2005).  

In particular, one farm with a high phosphate density, 
with using highly fertilizer are increased the prices for 
agriculture land to be paid. The influence of the "distance 
to farm and phosphate” variables is statistically highly 
significant (56 and 59%, respectively).  

Logarithmic model also used in the present work serve 



 
 
 

 

hedonic price models primarily of the examination at the 
basis of the lying conditions of market equilibrium (Garett 
and Laird, 1995). If the land market with regularity was 
certain areas and the agricultural land market was not 
any longer in the equilibrium, in such a case, a basic con-
dition of the hedonic beginning would be hurt. For the 
illustration of the model estimations for the LN of the pH, 
organic mater and saturated water point prognoses of the 
periodic commercial probability are accomplished. The 
variables concerned are varied gradually from their mini-
mum to their maximum development. For all other varia-
bles the average values are used. 

Logarithmic models support the basic assumption that 
the land value is being increased with LN of the pH, orga-
nic mater and saturated water. If the connection between 
distance to farm, salt and farm size were other one, one 
would have had to reject the hedonic beginning. 

 

Conclusion 
 
Land purchasers who arose actively up to the intro-
duction of the rural and urban areas on the land market 
were not ready to pay high prices for agriculture areas in 
municipalities with a very strong increase of the pH, 
organic mater, phosphate, and distance to farm. In agri-
cultural land markets, the signs of purchases on specu-
lation abated and the average prices almost halved them-
selves. Nevertheless, the prices for agricultural plots are 
affected further by the same land factors.  

A sufficient surface mobility is to be secured over a 
flexible lease in land market. The productive value regula-
tion (assumption of a land purchase to the productive 
value) will remain necessary. Definition of the term "land 
purchase", thus it is to be ensured our judgment that the 
assumption of agriculture areas and agricultural infra-
structure is granted predominantly to full- time farmers. It 
is not a goal of the land policies to create favorable living 
conditions for people active in the agriculture. Signifi-
cantly higher price of land in municipalities with a high 
pH, distance to farm, phosphate, and organic matter 
shows that straight such advantages in the price of land 
has. It is a strict separation among agriculture areas 
which is not agriculturally used any longer or for except-
agricultural purposes to be used to be able to aim. The 
results of the empirical investigations confirm such expec-
tations. Regression shows that the top price regulation 
belongs to the market forces and land characteristics. 
This expresses itself in a very weak commercial activity 
for plots with high agricultural land potentials. 

 
REFERENCES 
 
Aakre DG, David M, Saxowsky M, Vreugdenhil HG (1997). North 

Dakota land valuation model, Agric Econ. AE 97001  
Anonymous (2005). The report of Agric Headship of Agr. Ministry in 

Bursa city. www.bursatarim.gov.tr  
Bible DS, Hsieh C (1999). Determinants of vacant land values and 

implications for appraisers. Appraisal J. 67(3): 264-269. 

  
  

 
 

 
Carew R (2000). A hedonic analysis of apple prices and product quality 

characteristics in British Colombia. Canadian J. Agric. Econ. 48: 241– 
57. 

Caswell M F, Zilberman D (1986). The effects of well depth and land 
quality on the choice of irrigation technology. Am. J. Agric. Econ. 68: 
798-811. 

Cavailhes J, Wavresky P (2003). Urban influences on periurban 
farmland prices, Eur. Rev. Agric. Econ. 30 (3): 333-357.  

D’Arge RC, Shogren J (1989). Okoboji experiment: comparing non-
market valuation techniques in an unusually well-defined market for 
water quality. Ecol. Econ. 1(1): 251-259. 

Des Rosiers FM, Theriault P Villeneuve Y, (2000). Sorting out access 
and Neighborhood factors in hedonic price modeling, J. Prop. Invest. 
Finan .18 (3): 291-315. 

Ervin DE, Mill J W (1985) Agricultural land markets and soil erosion: 
policy relevance and conceptual issues. Am.J. Agric Econ. 67: 938-
942. 

Fidan H (2004). Quality Level And Price In Turkish Durum Wheat 
Market, Quality & Quantity International J .Methodol. Kluwer Acad. 
Pub. 38: 319-329. 

Freeman A, Myrick III (1993) .The measurement of environmental and 
resource values: theory and methods, Resources for the Future,. 

Garett M, Laird F (1995). Regression models for bivariate discrete and 
continuous Outcome with clustering. J. Am. Stat. Assoc. (JASS); 845.  

Gardner K, Barrows F (1985). The impact of soil conservation 
investment on land prices. Am. J. Agric. Econ. 67: 943-947.  

Griliches Z (1971). Price Indices and Quality Change. Cambridge, 
Mass.: Harvard University Press. 

I gın T, Lynn Forster F (2005). Using real options theory to analyze the 
impacts of urban development on farm real estate markets, Turkish J. 
Agric. For. 29: 409-417. 

I gın TF, Forster L (2006). A hedonic price analysis of farmland option 
premiums under urban fluencies, Canadian J. Agric. Econ. 54: 327-
340. 

Johnson RA, Wicherin DW (1992). Applied Multivariate Statistical 
Analysis. 3rd edition. Prentice Hall Inc. Englewood Cliffs. N.J. 07632. 
London.UK. 

King DA, Sinden JA (1988). Influence of soil conservation on farm land 
values. Land Econ. 64: 242-255.  

King DA, Sinden JA (1994). Price formation in farmland markets, Land 
Econ. 70 (1): 38-52.  

Miranowski JA, Hammes DB (1984). Implicit prices of soil charac-
teristics for farmland of Iowa. Am J. Agric. Econ. 66: 645-649. 

Plantinga A.J, Lubowski .R.N, Stavins R.N (2002). The effects of 
potential land development on agricultural land prices, J. Urban Econ. 
52: 561-581. 

Pope RP, Kramer RA, Green RD, Gardner BD (1979). An evaluation of 
econometric models of U.S. farmland prices, Western J. Agric. Econ. 
4: 107-119.  

Rosen S (1974). Hedonic Prices And Implicit Markets - Product 
Differentiation In Pure Competition J. Poli. Econ. 82: 34 

Shonkwiler JS, Reynolds JE (1986). A note on the use of hedonic price 

models in the analysis of land prices at the urban fringe, Land Econ. 

62(1): 58-63. 


