Content uploaded by Izabela Rącka
Author content
All content in this area was uploaded by Izabela Rącka on Nov 29, 2016
Content may be subject to copyright.
172
CHARACTERISTICS AFFECTING HOUSING PRICES IN KALISZ
Senior Lecturer, Ph.D. Izabela Rącka
The President Stanisław Wojciechowski University School of Applied
Sciences in Kalisz
Abstract
The paper presents the characteristics of flats and the way they affected housing
prices in Kalisz in the years 2006-2014. The author pays attention to the
differentiation of residential properties, which causes further segmentation of the
real estate market. The characteristics and prices of over 2,000 sold flats have
been analysed and – using econometric modelling – the impact of the
characteristics of the flats on their prices has been estimated.
Keywords: housing market, prices, flats, econometric modelling.
1. Introduction
The factors influencing the housing market situation and consequently the
housing prices, include macroeconomic, political, legal, social, demographic,
technological and environmental elements, as well as the situation in other seg-
ments of the real estate market and finally – the characteristics of the real estate.
The research into the changing levels of prices is crucial due to the linkages of the
real estate market, including the residential segment, with the economy. The real
estate market is affected by the economy, influencing it at the same time. For that
reason, market changes in the countries with a well-developed real estate market
are being examined and the results of the research are of interest to investors,
banks, entities shaping housing policy and many others.
The objective of the article is to present the changes in the prices of flats in
Kalisz in the years 2006-2014 and to demonstrate the influence of one group of
factors (characteristics of the real estate) on the prices of flats. The choice of the
173
objective determined the research methods, which were: selection of literature on
noise and its influence on the residential real estate market, price changes in the
residential real estate market, complete assessment of market transactions in flats
and econometric modelling. Finally, econometric models were constructed show-
ing the impact of noise and other factors on the prices of flats in the analysed
period.
The article incorporates information on prices and characteristics of residen-
tial dwellings sourced from notary deeds and real estate price and value registers,
data from real estate cadaster), title deeds, geoportals (information regarding util-
ity infrastructure, intended use in the development plan, noise levels, etc.) as well
as site visits. For the research purposes, information about all transactions in Ka-
lisz in the years 2006-2014 was collected. Out of those, the author eliminated
transactions with incomplete data (e.g. no address), non-market transactions (con-
tracts of donation, division of the estate, partition of joint property etc.) or con-
cerning non-residential real estate (commercial, industrial, etc.) and those with
specific conditions of sale (discount, delayed payment) and extreme ones, sug-
gesting a heightened need for the transaction (e.g. forced sale). The number of
analysed records meets the requirement of a representative sample.
2. Prices of flats
Beside demand and supply, flat prices are one of the elements of the housing
market. The prices are governed by market rules, undergo changes under the
influence of internal, real estate market factors, as well as external ones, on a
macro level. The prices inform the market participants whether their asset
allocation enables them to reach optimal results. The prices are therefore the best
indication of the direction of the asset flow, as well as an encouragement to
commence or suspend activity (Zaremba i Będzik 2010, p. 42). Real estate is the
only asset, which is clearly dependent on the character of the surroundings, i.e.
the location (Siemińska 2012, Fanning 2014).
The price in the real estate market depends on many factors, e.g. location, the
174
quality of the surroundings, the distance from the city centre, access to
communication network (including public transport), the age and materials the
building is built of etc. The price variability of residential real estate causes further
segmentation of the market (Ranci, Brandsen and Sabatinelli 2014, p. 167).
E. Kucharska-Stasiak, M. Załęczna and K. Żelazowski (2012, p. 23-25)
note that there has been no recognised distribution of prices for particular
characteristics of the real estate. Furthermore, the characteristics are poorly
reflected in the real estate prices. Due to the low frequency of transactions, rather
than the exact price, the market sets a range, within which the price falls. The level
of transaction price, understood as an average unit price (PLN/ m2) within a given
period, is influenced (except for the characteristics and sales forms and strategies)
by negotiating skills of the market participants, linkages between the entities,
motivations, emotions etc.
3. Factors determining the prices of residential real estate
The influence of various factors on the prices (and value) of residential real
estate is well researched. Authors focus mostly on the features of the real estate
influencing its price. Other significant factors (Issac 2002, pp. 31-32) include:
international situation and the level of interest rates,
investment climate and the condition of national economy,
fiscal and monetary policy,
local economy,
geographical environment and location,
trends and local demand (demand for real estate in a given location),
individual features of the real estate (architecture, style, interior, size and
features of the backyard etc.),
tenants (the value of tenanted and vacant real estate may differ),
condition, frequency of refurbishments, quality of the fittings,
possibility to extend and renovate,
175
ease of sale (the amount of time necessary for finalising the transaction),
accessibility of information regarding the real estate.
The research carried out across various real estate markets shows the influence
of numerous factors on the prices of flats. S. F. Fanning (2014, p. 51) underlines
the impact of location on the price of the real estate. K. Szopińska, M. Krajewska
and M. Bełej (2012) analysed the effect that noise plays in shaping the real estate
price. Noise is an example of a negative effect of the environment and translates
into lower prices. S. Huderek-Glapska and R. Trojanek (2013) measured the in-
tensity of the impact an airport has on the real estate prices in Warsaw in the years
2007-2011. The authors established that the price of the real estate depended on
its location with reference to the airport, the distance, whether it was located in
the landing zone, as well as on the degree of influence (including noise generation
or pollution caused by the airport). The flats located within the area of limited use
were on average 1% cheaper than identical flats located outside of the area.
Based on transaction prices, Ł. Mach (2011) points to the following variables
influencing the value of residential properties: the number of storeys in the build-
ing, the storey the flat is located on, the surface area of the flat, the year of building
completion and the standard of interior finishing. The analysis of the offer prices
showed that some additional significant factors included: the number of rooms,
the type of kitchen, the location of the flat within the building, the location of the
building, technology and the year of completion. M.F. Dziauddin, K. Ismail and
Z. Othman (2015) demonstrated that in Malaysia, in the years 2010-2011 the price
of flats was influenced by their surface area, as well as some aspects of the neigh-
bourhood, e.g. distance to high schools and the CBD. These dependencies were
noted in all the local markets analysed, however, the degree of influence of the
particular factors was different depending on the location.
The way that the distance to the CBD, employment centres, communication
hubs and public spaces affect the real estate prices is also an area of focus for
176
T. K. Koramaz and V. Dokmeci (2012). The authors analysed the effect of envi-
ronmental amenities on the housing prices in Istambul, by using the semi-hedonic
pricing model and analysing the spatial data with the Kriging method.
The research showed that the housing prices depend on the following environ-
mental amenities: distance from the city centre, arterial roads and the seashore.
R. Trojanek (2015) examined the influence of the quality of life in a given Warsaw
suburb on the value of the real estate. The suburbs on the extreme ends on the
quality of life spectrum, also featured extreme values of housing (the suburb with
the highest quality of life and the most expensive housing is Śródmieście (Down-
town) and the one with the lowest quality of life and cheapest housing is
Białołęka).
When researching the prices of flats in Slovenia, M. Romih and Š. Bojnec
(2008) established that they differ, depending on the region: the most expensive
flats can be found in Ljubljana, in its suburbs and in the area of Slovene Littoral
(Primorska), medium range prices dominate in Upper Carniola (Gorenjska),
while the cheapest flats can be purchased in the north-eastern part of Slovenia:
Carinthia (Koroška), Styria (Štajerska) and Prekmurje (Prekmurje). The analysis
of hedonic prices showed various factors influencing the prices in the above men-
tioned regions. In the region with the dearest flats, the size of the flat has a signif-
icant effect on the unit price, which is not the case in the other areas of the country.
The age of the flat and level of location within the building influenced the price
within the dearest and the cheapest regions.
The next research area in terms of the influence of the real estate characteristics
on its price is the analysis of the buyers’ preferences. It is based on the information
regarding the desired flats’ characteristics, as provided by the potential buyers.
M. Trojanek (2009) published the results of the analysis of the buyers preferences
within the primary market of flats in Poznań in 2007. Of the 22 characteristics
presented, the owners of flats bought off the plans, considered as the most im-
portant the following: location, price of the flat, balcony/terrace, accessibility of
177
transport, maintenance costs, distance to green spaces (parks, green areas). An-
other research into the buyers preferences within the market of new flats in the
years 2010-2011 (Trojanek M. 2013) was carried out based on the analysis of
notary deeds. The results presented the direction of change in the structure of the
buyers in socio-demographic (life style, family model) and economic (wealth)
terms. Another researcher of buyers preferences in the Poznań market was
H. Gawron (2012). According to his conclusions, the variability of buyers prefer-
ences depended on their socio-professional and age group.
According to the English Housing Survey1 the owners of flats stated that the
most important reason for a move was the desire for a larger flat (25%) and the
desire to improve their surroundings (16%) (Department for Communities and
Local Government 2016). R. Burrows (1998) claims that the characteristics of the
property neighbourhood and the size of the flat are of equal importance. The qual-
ity of life is also emphasized by T. Champion et al. (1998).
The impact of property characteristics on its value is an issue which has been
raised several times by S. Malpezzi. Already in 1980 (Malpezzi, Ozanne and
Thibodeau 1980; Malapezzi 2003) Malpezzi pointed to the discrepancies in as-
sessing the individual characteristics of a flat by different buyers and difficulties
in measuring the impact of real estate characteristics on its price.
In most segments, the real estate market, as well as real estate characteristics,
are of a local rather than universal nature (Prystupa 2014). Due to that, the re-
search regarding buyers preference is not carried out in the national market. There
is no need and no merit in generalizing the results of small sample analyses. The
local nature of the real estate market justifies quality analysis, which, contrary to
quantity analysis, is not carried out on large representative samples and does not
allow for conclusions on the concentration of attitudes within the population (Mai-
son 2010, p. 31).
4. Turnover of dwellings in Kalisz in the years 2006-2014
Taking into account the basic determinants influencing the prices, for the data
178
analysis, the author used the real estate characteristics which had been grouped
into sub-categories. The characteristics used in the analysis are divided into
measurable, i.e. quantitative (e.g. the floor space or price of the real estate) and
non-measurable, i.e. qualitative (e.g. the suburb or type of land ownership).
The number of transactions and the basic characteristics of the flats sold in a
given year are shown in Table 1.
Table 1. Basic descriptive statistics for transactions in Kalisz in the years 2006-
2014
Market segment/
year
2006
2007
2008
2009
2010
2011
2012
2013
2014
Total
Number of transactions
primary market
16
3
39
87
120
112
134
151
77
739
secondary market
120
143
80
108
142
174
151
224
216
1358
Total
136
146
119
195
262
286
285
375
293
2097
average floor space
49
46
51
53
55
52
51
52
50
51
standard deviation
15
17
15
18
17
17
15
16
18
17
variation coefficient
31%
38%
29%
34%
32%
33%
30%
31%
36%
33%
Source: Own research.
Transaction prices within the years analysed are presented in Figure 1.
Figure 1. Unit prices (PLN/m2) for flats sold in the secondary market in Kalisz
on a quarterly basis in the years 2006-2014
Source: Own research
179
Transaction prices in the secondary market had been growing rapidly in the
period from the third quarter of 2006 to approximately mid 2008, after that there
was a gentle correction in unit prices.
The prices of flats varied depending on the floor space of the dwelling. The
lowest unit prices in the secondary market were noted for flats with the floor space
from 55 m2 to 65 m2, which were mostly two-bedroom flats. In the primary
market, there was no transactions within all of the floor space groups in each year.
However, the following trend could be noted: in 2007 the dearest flats were small
(from 25 m2 to 45 m2), while in the years that followed, the highest unit prices
belonged to medium-size flats, similarly to the secondary market.
5. Analysis of correlation and multiple regression in Kalisz in the years
2006-2014
In the first instance the author analysed the correlation between the prices of
flats in Kalisz and the characteristics of the real estate. For the assessment of the
correlation of quantitative characteristics, the author used the Pearson correlation
coefficient. For establishing whether there is a correlation between the
quantitative variable and the qualitative variables, the test of independence 2 was
used. The assessment of the power of the influence was executed using the
corrected contingency coefficient. Explanatory variables with little internal
diversity (V≤0,10) were eliminated, as well as those, which were not statistically
significantly correlated with the dependent variable. The information obtained is
presented in Table 2.
Table 2. Correlation between the characteristics of the real estate and its price in
Kalisz in the years 2006-2014
Indicators
Flats
secondary market
Flats
primary market
Nominal price
(PLN/m2)
(PLN)
(PLN /m2)
(PLN)
Year
0.2809
0.1936
independ.
independ.
Number of the quarter
0.2817
0.1906
independ.
independ.
180
Number of the month
0.2819
0.1905
independ.
independ.
Transaction date
0.2821
0.1906
independ.
independ.
Area
independ.
independ.
0.6836
independ.
Usable area
independ.
0.7040
-0.1726
0,8566
Size of belonging premises
independ.
0.1466
0.1253
0.2162
Basement
independ.
independ.
independ.
independ.
Garage
0.0904
0.1058
independ.
independ.
Number of rooms
0.0684
0.5200
independ.
0.6095
Size of the plot
0.2080
0.1473
0.1982
0.0787
Title to land
independ.
0.1754
independ.
independ.
Number of above-ground storeys
independ.
independ.
independ.
independ.
Number of below-ground storeys
-0.1631
-0.1709
0.3215
0.1039
Height of the building: low/high
independ.
independ.
independ.
independ.
Built period
independ.
0.2437
-
-
Distance from swimming pool
-0.0694
-0.0954
-0.2144
independ.
Distance from city centre
0.2732
0.1699
-0.1531
-0.1246
Distance from shopping centre
0.1342
0.1264
-0.233
-0.0728
Distance from retail
0.2651
0.2871
-0.1843
independ.
Distance from park
0.1898
0.1146
-0.1179
independ.
Distance from kindergarten
independ.
independ.
-0.1181
independ.
Distance from bus stop
0.1306
0.1020
-0.1114
independ.
Distance from school
0.1349
0.1095
-0.1262
independ.
Distance from social housing
0.1520
independ.
0.1783
-0.0921
Distance from unemployment zone
0.3040
0.2586
-0.0726
independ.
Distance from poverty zone
0.2859
0.2609
independ.
independ.
level of noise
-0.2531
-0.2677
0.1560
0.1611
number of buildings
-
-
-
-
type of building
independ.
independ.
-
-
Legend: The table shows values of statistically significant correlations (=0.05) only; bold font– clear
correlation, ‘-‘ – did not appear or no differentiating impact; independ. – no grounds for rejection H0
with the independence of the analysed characteristics (=0.05).
Source: Own elaboration.
The analysis was carried out on the base of econometric modelling with the
use of multiple regression equations. When building the econometric models
using multiple regression techniques, the author assumed that there is a linear
relationship between the average unit price and the explanatory variables. In the
chosen procedure, the dependent variable was the price of flat in 2006-2014 in
Kalisz. From the 30 initially chosen independent variables, after carrying out the
consecutive steps of regressive elimination, in the final version of the model there
181
were 14 remaining (x1 - primary/secondary market, x2 – transaction date, x3 – title
to land, x4 – noise level, x5 – distance from shopping centres, x6 – distance from
social housing zone, x7 – distance from primary school, x8 – garage, x9 - number
of above-ground storeys, x10 – distance from poverty zone, x11 – number of below-
ground storeys, x12 – distance from kindergarten, x13 – distance from park, x14 –
distance from bus stop), and the estimated equation of relations between the
variables could be written as:
P = – 436.22 x1 + 0.23 x2 + 127.31 x3 – 16.05 x4 – 0.14 x5 + 1.53 x6 – 0.36 x7 + 752.94 x8 +
34.53 x9 + 0.66 x10 – 108.43 x11 – 0.53 x12 + 0.13 x13 + 0.52 x14
Table 3. OLS using observations 1-18, dependent variable: unit price 2006-2014
Standard error of regression s=631.87
Determination indicator R-squared R2=0.279
Adjusted R-squared R2adj=0.274
F(14, 2082)= 57.476
p value (F) < 0,0000
Coefficient
Std. error
t-ratio
p-value
Const
3146.122
193.1008
16.2926
0.000000
primary=0 secondary=1 market
-436.220
36.5473
-11.9358
0.000000
Date – day number
0.227
0.0167
13.6444
0.000000
land ownership=1 perpetual usufruct=0
127.312
33.2142
3.8330
0.000130
noise level
-16.055
2.9400
-5.4607
0.000000
distance from shopping centres
-0.144
0.0345
-4.670
0.000032
distance from social housing
1.532
0.2384
6.4241
0.000000
distance from primary school
-0.362
0.0973
-3.7235
0.000202
garage
752.936
194.1345
3.8784
0.000108
building - above-ground storeys
34.525
9.6085
3.5932
0.000334
distance from poverty zone
0.656
0.2398
2.7340
0.006310
building - below-ground storeys
-108.429
31.4163
-3.4514
0.000569
distance from kindergarten
-0.535
0.1219
-4.3868
0.000012
distance from park
0.131
0.0485
2.6919
0.007163
distance from bus stop
0.516
0.2266
2.2788
0.022778
Source: Own elaboration
The calculated regression coefficients may be interpreted as the influence of
the given variables on the changes in the unit price: by how much will the unit
price change (increase or decrease) on average in a situation when the independent
variable relating to the given coefficient raises by a unit.
Verification of the models was made based on:
verification of statistical significance of the effect of independent variables
182
on the dependent variable,
determination coefficient,
adjusted value of the determination coefficient, resistant to increasing the
number of explanatory variables
Based on the model constructed, it is possible to conclude that the prices of
flats in Kalisz secondary market in the years 2006-2014 were influenced by:
positive variables: x2 – transaction date, x3 – title to land, x6 – distance from
social housing zone, x8 – garage, x9 - number of above-ground storeys, x10
– distance from poverty zone, x13 – distance from park, x14 – distance from
bus stop,
negative variables: x1 - primary/secondary market, x4 – noise level, x5 –
distance from shopping centres, x7 – distance from primary school, x11 –
number of below-ground storeys, x12 – distance from kindergarten.
The parameters of the regression function showed statistical significance (at
the level of p<0.05) and the determination indicator R2 was 0.279. This means
that the constructed model explained more than 27% of changes of the dependent
variable. The value of adjusted R2 (accounting for the number of independent
variables in the model) equalled 0.274 and did not deviate significantly from the
general coefficient (table 3).
6. Conclusions
The article presented the results of the research into the influence of real estate
characteristics on its price, based on real estate markets across the world. The
author demonstrated the level of prices of flats in Kalisz in the years 2006-2014.
Next, the factors determining the unit prices of flats were revealed.
An econometric model was built, to describe the prices of flats in the analysed
period. In the course of the research, it was established that the real estate
characteristics explain its price only to a certain extent, which means that there
must be other factors determining the level of prices. Consequently, further
research must take into account the effect of other factors e.g. prices in the real
183
estate markets of the largest cities in the country, macroeconomic, demographic,
social and other factors.
Notes:
1. English Housing Survey is a continuous national survey commissioned by the
Department for Communities and Local Government. It collects information
about people’s housing circumstances and the condition and energy efficiency of
housing in England. It has two component surveys: a household interview and a
physical inspection of a sub sample of the properties.
References
1. Burrows, R., 1998, The Dynamics of the Owner-Occupied Housing Market, CML Re-
search Report.
2. Champion, T., Atkins, D., Coombes, M., Fotheringham, S., 1998, Urban Exodus,
CPRE, London.
3. Department for Communities and Local Government, https://www.gov.uk/govern-
ment/collections/english-housing-survey [access: 5.09.2016].
4. Dziauddin, M.F., Ismail, K., Othman, Z., 2015, Analysing the local geography of the
relationship between residential property prices and its determinants, Bulletin of Ge-
ography. Socio-economic Series, Vol. 28, Iss. 28 (Jun 2015), pp. 21-35.
5. Fanning, S. F., 2014, Market Analysis for Real Estate, Second edition, Appraisal Insti-
tute, Chicago, USA.
6. Gawron, H., 2012, Ewolucja funkcji mieszkania i preferencji klientów na rynku
mieszkaniowym, w: Trojanek M., Strączkowski Ł. (red.), Z prac Katedry Inwestycji i
Nieruchomości. Aktualne problemy rynku nieruchomości w Polsce, Zeszyt Naukowy
Nr 231, Wydawnictwo Uniwersytetu Ekonomicznego w Poznaniu, Poznań, ss. 7-20.
7. Huderek-Glapska, S., Trojanek, R., 2013, The Impact of Aircraft Noise on House
Prices, International Journal of Academic Research, May2013, Vol. 5, Issue 3, pp.
397-408.
8. Isaac, D., 2002, Property Valuation Principles, Palgrave, London.
9. Koramaz, T.K., Dokmeci, V., 2012, Spatial Determinants of Housing Price Values in
Istanbul, European Planning Studies, Vol. 20, No. 7 (Jul 2012), pp. 1221-1237.
184
10. Kucharska-Stasiak, E., Załęczna, M., Żelazowski, K., 2012, Wpływ procesu integracji
Polski z Unią Europejską na rozwój rynków nieruchomości, Wydawnictwo Uniwersy-
tetu Łódzkiego, Łódź.
11. Mach, Ł., 2011, Budowa praktycznego modelu regresji opisującego zależności wystę-
pujące na rynku nieruchomości mieszkaniowych, w: Zeszyty Naukowe Wyższej
Szkoły Bankowej we Wrocławiu nr 20/2011 Forlicz S. (red.), Wydawnictwo WSB,
Poznań.
12. Maison, D., 2010, Jakościowe metody badań marketingowych. Jak zrozumieć konsu-
menta, Wydawnictwo Naukowe PWN, Warszawa.
13. Malpezzi, S., 2003, Hedonic pricing models: A selective and applied review, w: T. O.
Sullivan & K. Gibbs (Eds.), Housing economics and public policy: Essays in honor of
Duncan Maclennan, Blackwell, Oxford, UK.
14. Malpezzi, S., Ozanne, L., Thibodeau, T., 1980, Characteristic prices of housing in
fifty-nine metropolitan areas, Research Report, December, The Urban Institute, Wash-
ington, DC.
15. Prystupa, M., 2014, Wycena nieruchomości i przedsiębiorstw w podejściu porównaw-
czym, Wydawnictow Replika, Zakrzewo.
16. Ranci, C., Brandsen, T., Sabatinelli, S. (red.), Social Vulnerability in European Cities:
The Role of Local Welfare in Times of Crisis, Palgrave Macmillan, London.
17. Romih, M., Bojnec, Š., 2008, Višina in oblikovanje cen rabljenih stanovanj v Sloveniji
(Level and Determinants of Housing Prices in Slovenia at Micro Level), Management
(18544223). Vol. 3, Iss. 2, pp. 165-184.
18. Szopińska, K., Krajewska, M., Bełej, M., 2012, Apartment Market Analysis Consider-
ing Environmental Noise Level in Poland, European Real Estate Society 19th Annual
Conference, Edinburgh, Scotland.
http://web.sbe.hw.ac.uk/eres2012/Book%20of%20Abstracts_Main.pdf
19. Trojanek, M., 2009, Preferencje nabywców na pierwotnym rynku mieszkaniowym w
Poznaniu, 2009, Acta Scientiarum Polonorum. Administratio Locorum 8 (1), ss. 5-19.
20. Trojanek, M., 2013, Buyer behaviour patterns on the primary housing market in Poz-
nan in the period 2010-2011, Real Estate Management and Valuation, Vol. 21, No. 4,
s. 47-54.
21. Trojanek, R., 2015, The Relation Between the Attractiveness and Value of Districts in
Warsaw, Montenegrin Journal of Economics, Vol. 11, No. 2 (December 2015), pp.
137-146.