Skip to main content
Fachgebiete
Automobil + Motoren Bauwesen + Immobilien Business IT + Informatik Elektrotechnik + Elektronik Energie + Nachhaltigkeit Finance + Banking Management + Führung Marketing + Vertrieb Maschinenbau + Werkstoffe Versicherung + Risiko
DE
EN
Bücher
Zeitschriften
Themenseiten
Organisationspsychologie Projektmanagement Marketing Smart Manufacturing
Weitere Formate
Podcasts Webinare Technik Webinare Wirtschaft Kongresse Veranstaltungskalender Awards
Jetzt Einzelzugang starten
Zugang für Unternehmen
Referenzkunden
SLX-Digitalkonferenz: Zukunftswerkstatt 2024

Mehr Umsatz mit Top-Kontakten

Am 7. Mai erfahren Sie, wie Vertriebsteams Leadgenerierung, Leadnurturing und Leadstrategien effizient steuern können.
Erweiterte Suche
Anmelden
nach oben
Review of Regional Research
Erschienen in: Review of Regional Research 2/2018

24.08.2018 | Original Paper

Regional heterogeneity in age-related housing depreciation rates

The case of German single-family houses

verfasst von: Sören Gröbel

Erschienen in: Review of Regional Research | Ausgabe 2/2018

Einloggen

Aktivieren Sie unsere intelligente Suche, um passende Fachinhalte oder Patente zu finden.

search-config
loading …

Abstract

This paper analyses the regional variation of age-related depreciation rates for housing: firstly, using observations of single-family house prices, our analysis estimates housing market-specific age-cohort depreciation rates, while taking into account the heterogeneity in attributes, location, and state of maintenance. Secondly, we examine whether the derived depreciation rates correlate with determinants of the regional supply- and demand-side. Considering the durability of the housing stock and substitution effects in housing demand, we find that a rise in excess supply will not only lead to declining house prices in the aggregate, but also to a widening of relative house price differentials among houses at different ages.
Vorheriger Artikel Evaluation of the Brazilian regional development funds
Nächster Artikel Physische und digitale Erreichbarkeit von Finanzdienstleistungen der Sparkassen und Genossenschaftsbanken

Sie haben noch keine Lizenz? Dann Informieren Sie sich jetzt über unsere Produkte:

Springer Professional "Wirtschaft+Technik"

Online-Abonnement

Mit Springer Professional "Wirtschaft+Technik" erhalten Sie Zugriff auf:

  • über 102.000 Bücher
  • über 537 Zeitschriften

aus folgenden Fachgebieten:

  • Automobil + Motoren
  • Bauwesen + Immobilien
  • Business IT + Informatik
  • Elektrotechnik + Elektronik
  • Energie + Nachhaltigkeit
  • Finance + Banking
  • Management + Führung
  • Marketing + Vertrieb
  • Maschinenbau + Werkstoffe
  • Versicherung + Risiko

Jetzt Wissensvorsprung sichern!

Jetzt informieren

Springer Professional "Wirtschaft"

Online-Abonnement

Mit Springer Professional "Wirtschaft" erhalten Sie Zugriff auf:

  • über 67.000 Bücher
  • über 340 Zeitschriften

aus folgenden Fachgebieten:

  • Bauwesen + Immobilien
  • Business IT + Informatik
  • Finance + Banking
  • Management + Führung
  • Marketing + Vertrieb
  • Versicherung + Risiko




Jetzt Wissensvorsprung sichern!

Jetzt informieren
Anhänge
Nur mit Berechtigung zugänglich
Fußnoten
1
Due to the prevailing incapability of measuring housing quality appropriately, building age usually serves as an approximation for all sources of depreciation instead. Thus, age-related housing depreciation does not distinguish between declining functionality, physical deterioration, and an increase in incidental cost for maintenance (e.g., Smith 2004; Francke and Minne 2017).
 
2
The age-related depreciation of housing is the key ingredient of the filtering theories in which old dwellings represent the housing supply for households with lower income.
 
3
This represents a consequence of an indispensably high building activity given the shortage of living space as well as of a tendency toward suburbanization caused by an increased degree of motorization.
 
4
This work is therefore related to a large strand of literature which estimates age-related depreciation rates on the basis of cross-sectional data and hedonic modelling. For a literature review see Malpezzi et al. (1987), Wilhelmsson (2008), and Francke and Minne (2017). For an empirical application of a repeat sales approach see Clapp and Giaccotto (1998) or Harding et al. (2007).
 
5
A higher proportion of land value decreases the absolute price differences of single properties and therewith the relative price differences in structural characteristics. Hence, assuming the proportion of structural building value to land value to differ, the regional variation in depreciation rates is also likely to be a phenomenon of the spatial diversity in the land-to-structural value ratio (e.g., Bostic et al. 2007; Bourassa et al. 2011; Bourassa, Hoesli, Scognamiglio, and Zhang 2011; Francke and Minne 2017).
 
6
The set of existing rules formulated in the DIN 4108 Waermeschutzbestimmung had no legal force.
 
7
Defined in the Waermeschutzverordnung WSchV’95.
 
8
The empirical strategy is related to the literature dealing with estimating the demand for housing attributes. See Bayer et al. (2016) for a current study and a literature review. Our primary issue under consideration is, however, to examine the regional variation in age-related depreciation rates. Therefore, the analysis does not go beyond the first stage of the Rosen (1974)-two-step approach.
 
9
See also Sect. 2. For comparison, we make use of a polynomial function of the second order, which is commonly used in housing literature, as well as unspecific decade depreciation rates, \(j\in[<1950\), 1950–60, 1970–80, 1990–00, 2000–10], which are not justified by changes in legal settlement.
 
10
There has been no consensus in the housing literature so far as to how one should control appropriately for locational effects. In the last two decades the traditional hedonic model has been augmented by methodologies from spatial statistics and econometrics (e.g., Basu and Thibodeau 1998; Dubin 1999). Bourassa et al. (2007) finds, however, that simple ordinary least squares models containing community dummies do well in terms of predictive ability.
 
11
We use a fixed number of neighbors, \(k=10\), for the spatial structure. For the time period, we use a fixed number of objects relying on the absolute dimension of offered objects. The general weighting matrix can be described as follows: \(W={\sum^{m_{s}}_{i=1}{\rho^{i}S_{i}}}/{\sum^{m_{s}}_{i=1}{\rho^{i}}}\), with \(S_{i}\) (\(i\in[1,m_{s}]\)) the spatial neighbor, \(m_{s}\) represents the predefined number of regarded neighbors; \(\rho\) (\(\in[0,1]\)) represents the decay parameter \(\rho\) (\(\in[0,1]\)) which measures the relative impact of the individual neighbors which is then set to \(0.7\) (results are however robust to changes in \(\rho\)). By using a lower triangular weighting scheme with a large temporal dimension, we avoid endogenous problems and can apply a simple least squares methods for estimation (Pace et al. 2000).
 
12
Clapp and Salavei (2010) show that their findings are robust when the ratio of interior size to mean interior size of new houses in proximity is used instead. Accordingly, the calculated intensity of a house j is given as follows: \(\textit{int}_{i}=\boldsymbol{X_{i,int}}/(\boldsymbol{X_{j,int}}\,\vline\,j\in\boldsymbol{S})/{n_{X}}\) whereby \(\boldsymbol{S}\) represents the vector of neighboring houses to object i, and \(\boldsymbol{X_{i}}\) contains \({n_{X}}=4\) variables: living space, living space to overall lot size, building age, elevated equipment, and modernized.
 
13
According to Clapp et al. (2012), there is a problem in the identification of significant positive option values with regard to measuring land value. Therefore, the positive option value has only been identified correctly if the marginal effect (the log price difference between a house with low intensity and the middle (omitted) category) for the lower quartiles is positive as well as if the difference between a low intensity and a high intensity is positive as well.
 
14
Though considering rental prices, Thomschke (2015) is the only German data-based housing study to consider quantile effects.
 
15
For a review of quantile regression methodology, with a detailed discussion of the truncation problem and estimation procedure by minimization of the weighted sums of absolute residuals, see e.g. Koenker and Hallock (2001).
 
16
Although the dependent variable represents estimates, we do not need to bootstrap standard errors, seeing as derived coefficients stem from separate estimations.
 
17
See Lerbs and Teske (2016) for a literature review on the house price-vacancy relation.
 
18
We use the ratio of land-to-structural value of new construction as this age-cohort represents the reference group for the relative depreciation scheme in the hedonic models. By use of structural and locational variables discussed in previous subsections, we control for intraregional differences in land value as well as structural characteristics, such as lot size and living space among age-cohorts.
 
19
Single-family houses are defined as detached and semi-detached houses. Special housing types, such as mansions, are not included. The analysis of relative price differences among various housing types may be part of future research.
 
20
Some regional districts (Emden, Hameln-Pyrmont, Luechow-Dannenberg, Wilhelmshaven) have been deleted from the overall sample due to a very low number of observations.
 
21
Potential buyers will have a stronger market position in housing markets with high excess supply, which in turn will increase the local bargaining bias (e.g., Harding et al. 2003).
 
22
This supports the hypothesis that excess supply will also lead to a selection and therewith to lower re-use probabilities of existing older houses, notably in housing market areas where the share of single-family houses is high.
 
23
Table 9 shows dynamics for selected regional variables from Lower Saxony together with dynamics for all German regions.
 
24
The labor market of Wolfsburg is strongly marked by the head offices of Volkswagen.
 
25
The highest and lowest vacancy rates for all residential houses can be found in Salzgitter (\(10.43\%\)) and Wolfsburg (\(1.39\%\)) respectively.
 
26
Form of presentation is strongly inspired by Clapp et al. (2012). Coefficients are declared as statistically significant if their p-value is smaller than 0.1. Empirical results are robust against smaller \(\alpha\) values. Estimation results with \(\alpha=5\) will be provided upon request. Huber-White robust standard errors are calculated.
 
27
Although empirical results reveal that the model with a polynomial function (that is commonly used in housing literature) and the model with specific age-cohorts depreciation scheme do not differ significantly in measure-of-fit, the latter is more appropriate for the analysis of regional differences in marginal effects.
 
28
Due to a low spatial density of micro data, we concentrate mainly on indirect neighborhood effects. For a more differentiated analysis of price effects of houses in direct proximity see Zahirovich-Herbert and Gibler (2014).
 
29
Our results are therefore in contrast to the findings of Clapp et al. (2012). This may stem from the fact that we already use a large set of variables to control for locational values. However, for an interpretation of significant option value in sub-markets, the results must be identified by the strategy proposed by the former.
 
30
Significance is assessed by bootstrapped standard errors and significance levels of \(\alpha=10\) or better. Pseudo R‑squared statistics are reported as well. Results are robust against smaller \(\alpha\) values. Empirical results for a critical value of \(\alpha=5\%\) as well as for the specification without spatial effects are not reported but will be provided upon request.
 
31
For lack of describing the characteristics of the distribution-on-distribution of the estimators, we use t‑test statistics (two-sided) to test for coefficient homogeneity among the upper and lower quantile mean estimators in Fig. 4. Results are only reported for estimators that differ significantly.
 
32
Due to the relatively low number of observations, we present only selected estimation results with relevant accompanying variables. Empirical analyses on the basis of regional data often suffer from spatially correlated error terms due to omitted variables or spillover effects. To test for the presence of such phenomena, we use Moran’s I statistics, but do not find remaining autocorrelation in residuals.
 
33
There is however a strong correlation between vacancy rates and shares of age-cohorts on the total supply. We suggest that this stems from a long-term process: regions with steadily increasing population and low vacancy rates respectively are more likely to show a higher share of newly built houses. The endogeneity of new construction and deterioration is already discussed in the seminal paper of Sweeney (1974).
 
34
Offering prices, on the other hand, are likely to be biased due to bargaining effects. Hence, further research may deal with this problem in greater detail.
 
Literatur
Dubin R, Kelley PR, Thibodeau TG (1999) Spatial autoregression techniques for real estate data. J Real Estate Lit 7(1):79–95CrossRef
Kholodilin KA, Menz JO, Siliverstovs B (2010) What drives housing prices down?: evidence from an international panel. J Econ Stat 230(1):59–76 (Jahrbuecher fuer Nationaloekonomie und Statistik)
Koenker R, Hallock K (2001) Quantile regression: an introduction. J Econ Perspect 15(4):43–56CrossRef
Lerbs O, Teske M (2016) The house price-vacancy curve. Tech. rep., ZEW Discussion Paper No. 16-082
Munneke HJ, Womack KS (2015) Neighborhood renewal: the decision to renovate or tear down. Reg Sci Urban Econ 54:99–115CrossRef
Shilling JD, Sirmans C, Dombrow JF (1991) Measuring depreciation in single-family rental and owner-occupied housing. J Hous Econ 1:368–383CrossRef
Smith BA, Tesarek WP (1991) House prices and regional real estate cycles: market adjustments in houston. AREUEA J 19(3):396–416CrossRef
Metadaten
Titel
Regional heterogeneity in age-related housing depreciation rates
The case of German single-family houses
verfasst von
Sören Gröbel
Publikationsdatum
24.08.2018
Verlag
Springer Berlin Heidelberg
Erschienen in
Review of Regional Research / Ausgabe 2/2018
Print ISSN: 0173-7600
Elektronische ISSN: 1613-9836
DOI
https://doi.org/10.1007/s10037-018-0125-3

Weitere Artikel der Ausgabe 2/2018

Review of Regional Research 2/2018 Zur Ausgabe

Original Paper

Physische und digitale Erreichbarkeit von Finanzdienstleistungen der Sparkassen und Genossenschaftsbanken

Original Paper

The detection of natural cities in the Netherlands—Nocturnal satellite imagery and Zipf’s law

Original Paper

Evaluation of the Brazilian regional development funds

Original Paper

Gauging two sides of regional economic resilience in Western Germany—Why sensitivity and recovery should not be lumped together