Table of contents
List of Tables
List of Abbreviations
1 Title and Abstract
4 Literature Review
5 Development of hypothesis
6 Research Design
7 Data and Sources
8 Expected outcome and contingent plan
List of Tables
Table 1: Variables of the hedonic regression model
Table 2: Testing Criteria Hypotheses 1a
Table 3: Testing Criteria Hypotheses 1b
Table 4: Testing Criteria Hypotheses 2a
Table 5: Testing Criteria Hypotheses 2b
Table 6: Expected Outcome of regression coefficients
List of Abbreviations
BNP Bank Nationale de Paris
CBD Central Business District
GDP Gross Domestic Product
GDR German Democratic Republic
GIS Geographical Information System
IVD Immobilienverband Deutschland
OLS Ordinary Least Squares
PIS Public Information System
RIWIS Regionales Immobilien-Wirtschaftliches Informations System
ROI Return on Investment
US United States
1 Title and Abstract
“How societal aging impacts apartment prices in Germany – the case of Leipzig”
This paper illustrates the effect of societal aging on apartment prices in Leipzig (Germany), comparing a sample of city-level demographic and apartment price index data between 1990 and 2018. The impact of small households, as well as the old age dependency ratio, will be inspected. A regression model (OLS) is constructed to examine both occurrences. Statistical significance of the variables is proofed by relevant research on the topic. Results show that the number of small households is positively, and old age dependency ratio negatively correlated to the average apartment price (€/m2) in Leipzig. Summing up, this paper recommends to critically examine societal aging factors in order to determine political, investment, and property developing strategies.
Current research contributions have forecasted a huge aging problem for Germany´s society. There should be significant impacts on the property market, too (Hiller and Lerbs, 2016). The society faced aging problems already during the second world war, but luckily the “baby boomer generation” averted this trend quickly.
Since 1989, Germany reunited, and vast differences between East and West have been tried to adjust. In terms of economy, wages, property values, demography, et cetera, significant gaps remained, especially in rural areas. Looking at Leipzig (former Eastern Germany), one can see an incredible contrasting development of the concerns mentioned above. Population size has increased by 18.7% to almost 600,000 over the last ten years (PIS Stadt Leipzig). German news (i.e., MDR, Leipziger Volkszeitung) recently published results of a study by the Berlin-Institut für Bevölkerung und Entwicklung, forecasting Leipzig to be the fastest growing city in Germany until 2035 with roughly 16% population growth (Munich 11%).
In order to highlight the city´s property market position, prominent market players are using synonyms like “boom town” or “hidden champion” (BNP Paribas, Wohnreport Leipzig 2018). Apartment prices have skyrocketed on average by 9,79% p.a. between 2010 and 2018 (BulwienGesa).
Nevertheless, Leipzig is also faced with common societal problems like upcoming old-age poverty (i.e., newspaper article: Altersarmut: Alter!) and had a strong upward trend in aging between 2000 and 2010 (PIS Stadt Leipzig). Existing empirical studies on the relationship between aging and apartment prices are not examining the distinctive implications for an emerging regional market like Leipzig. For these cities, studies on upcoming aging issues are very relevant for further market evaluation and risk analysis, because they become more and more a substitute for too expensive top cities. Therefore, it is to be questioned if and how Leipzig`s property market is affected by societal aging.
Demography in Leipzig has shifted over the last decades. The elder cohort tended to increase more rapidly compared to the working-age cohort. People seemed to be elder on average, but at a certain point, the phenomena changed. The old age dependency ratio (people aged 65+ over people aged 20 to 65) started to decrease from 2010 after it stagnated a couple of years earlier (Federal Statistical Office Saxonia). Leipzig kept many historical residential buildings that have not been destroyed during the war. From 2000, a big wave of refurbishments started to make the city attractive to young people. Due to a continuous economic upwards trend, many well-known corporations settled in Leipzig, while universities kept increasing student numbers. Looking at recent data, one can see that the ratio has stagnated again since 2016 and remains constant. Furthermore, the number of single and two-person households is increasing due to a declining household size. One reason may be an aging society.
Knowing that, it seems to be highly relevant to examine if an increasing number of small households affects the apartment prices per square meter significantly positive or not. On the other hand, testing the real risk of aging (by old age dependency) is going to be conducted and will be empirically tested through a regression model. These effects are going to be analyzed for the residential apartment submarket in Leipzig.
The results should support investors to evaluate the risk of aging in order to determine an investment strategy. Indicated correlations should help to assess ongoing aging shifts for capital allocation.
4 Literature Review
Starting in 1989, Mankiw and Weil found that changes in birth rates (affecting population size and age structure) have a considerable impact on housing demand and, therefore, substantially causing housing price changes. They forecasted a price decline in the US real estate market in the 1990s, due to a serious downwards shift in birth rates (“baby-bust”) during the 1970s. However, it was proved wrong, because the use of cross-sectional micro-level data ignoring housing consumption among the elderly is low in a cross-section.
For England/Wales and Scotland, Levin et al. (2009) found in their difference-in-difference methodology (panel data estimation) that a population decline and an aging society put downward pressure on property prices. Notably, the younger working-age cohorts aged 20-29 and 30-34 years as first-time property buyers were statistically significant and had a positive impact on housing prices.
Malmberg (2010) regressed aggregate housing demand on age distribution for Sweden (1981-2006), using regional panel data. He figured out that population aging, as well as low birth rates, have a slowing impact on property prices growth. Similar to Levin et al. (2009), age cohorts 15-29 and 30-49 have a significant positive, whereas people aged 75+, a significant adverse effect on housing prices. Looking at another angle, Hui et al. (2012) examined the relationship between elderly dependency, housing prices, and the fertility rate in Hong Kong. If housing prices and old age dependency ratio increased by 1%, birth rates would decrease by 0.23% and 1.23%, using an ARDL to co-integration approach.
Takáts (2012) used an extensive panel regression dataset of 22 advanced economies from 1970-2009 (OECD countries) to observe the relationship between demography and real estate. He found that demography boosted house prices over the last four decades by around 30 basis points, whereas property prices would decrease in the upcoming 40 years by around 80 basis points. Accordingly, he explains upwards price movements are mostly caused by population growth, whereas the old age dependency ratio will put intense downward pressure on housing prices, ceteris paribus.
Saita et al. (2016) observed the impact of societal aging on real estate prices in Japan and the US. They figured out that the old age dependency ratio and housing prices are inversely correlated, whereas population size impacts housing prices positively. Furthermore, they forecasted the impact of aging on housing prices in Japan, and suggested a 2.4% percent decline p.a., compared to a decrease of 3.7% p.a. between 1976 and 2010, as the old age dependency ratio is not expected to increase as much as before. Since there was no entire quality-adjusted housing price index for Japan, they constructed one using a hedonic regression approach, which is considered critically.
A major contribution to the German property market has been published by Hiller and Lerbs (2016), who analyzed 87 German cities on demographic and housing price data between 1995-2012. They determined two major demand characteristics for the real estate market: the size and the age effect. They figured out similar elasticities of population growth and old age dependency with property prices as Takáts (2012) and Saita et al. (2016) showing that Germanys elasticity (old age dependency) has the highest impact on property prices (-1.83) compared to Japan (-1.73), the US (-0.54) and OECD Countries in general (-0.68).
5 Development of hypothesis
The above literature illustrates the importance of societal aging as one of the major macroeconomic determinants that impact the housing market. It is proved that Germany will suffer a lot more from low fertility and a high aging population than other nations (i.e., Hiller and Lerbs, 2016; Just and Maennig, 2017). To outline the influence of aging in Leipzig, two hypotheses which seem to reveal the most relevant factors impacting Leipzig’s property market are developed.
The population size is one of the most significant drivers for demand and so property prices, but recent research has shown the number of households to be the real direct driver (Just and Maennig, 2017). They found that even during the population size decline of 800,000 inhabitants (between 2003 and 2010) in Germany, the number of households increased. Shifts in the age structure strongly influence this number. In general, smaller apartments by size, are asking a higher per square meter price than larger condos. Therefore, it is necessary to look at the number of small households as well as the unit size and their impact on the price (€/m2). Hereafter, it is examined if this phenomenon applies to emerging Leipzig. For this reason, the first hypothesis (part a and b) denotes as follows:
A rise in the number of small households (single and two-person households) impacts apartment prices (€/m 2 ) positively, ceteris paribus.
When the number of small households rises, the size of an apartment declines, which leads to an aparment price (€/m 2 ) increase of smaller units, ceteris paribus.
As demand seems to be determined by the number of households, different age cohorts should have different effects on the price. Elderly are demanding other housing attributes than younger people. Besides, they have a different income level and purchasing behavior. Therefore, it is justified to question the impact of a growing number of old aged people on apartment prices in detail. The old age dependency ratio is defined by people aged 65 and above years to people being at working age (20 to 65). Many contributions to recent research have shown a significantly negative correlation among old age dependency and housing prices for several countries, i.e., Germany, Japan, the US (Levin et al., 2009; Takáts, 2012; Saita et al., 2016; Hiller and Lerbs, 2016). This impact is expected to depend on the apartment size as well, as elderly tend to live in smaller households, which might increase demand for smaller units. This correlation may impact Leipzig’s apartment prices too. For finding an appropriate answer, the second hypothesis (part a and b) is analyzing the effect as follows:
The old age dependency ratio (age group 65+ over 20-65) affects apartment prices (€/m 2 ) negatively, ceteris paribus.
A decline in apartment sizes due to a rising old age dependency ratio (age group 65+ over 20-65) leads to an aparment price (€/m 2 ) increase of smaller units, ceteris paribus.
6 Research Design
Properties are heterogeneous assets with many different variables impacting the price, knowing that a regression model has to be established. It assesses the significance of one attribute while keeping all other variables constant. With the model, one can examine the buyer`s marginal willingness to pay for the looked-after attribute. When evaluating the willingness to pay, the price function can be derived by the concerned character. For examination purposes, a model is chosen, where the dependent variable is equal to the sum of all independent variables (Rosen, 1974).
For the paper´s regression model, a mixture of cross-sectional (i), property-based variables, and time series data (t), that is observed over time (one year), is used and defined as follows:
where P is the apartment price per square meters (dependent variable); SIZE shows the living area (square meters), ROOMS reveals the total number of rooms, DCBD measures the distance to the CBD (meters), QUALITY rates the quality and AGE marks the actual age of the apartment (years). The age is calculated as the difference between date of sale and completion. In reality, properties are renovated and improved over time, which is not accounted for in this model (due to lack of data) and can slightly decrease the accuracy of the estimate. QUALITY is a categorical variable with three levels. Level one means simple, level two standard, and level three luxury quality.
- Quote paper
- Benjamin Liers (Author), 2019, How societal aging impacts apartment prices in Leipzig, Germany, Munich, GRIN Verlag, https://www.grin.com/document/542906