As housing development and housing market policies involve many long-term decisions, improving house price predictions could benefit the functioning of the housing market. Therefore, in this paper, we investigate how house price predictions can be improved. In particular, the merits of Bayesian estimation techniques in enhancing house price predictions are examined in this study. We compare the pseudo out-of-sample forecasting power of three Bayesian models—a Bayesian vector autoregression in levels (BVAR-l), a Bayesian vector autoregression in differences (BVAR-d), and a Bayesian vector error correction model (BVECM)—and their non-Bayesian counterparts. These techniques are compared using a theoretical model that predicts the borrowing capacity of credit-constrained and unconstrained households to affect house prices. The findings indicate that the Bayesian models outperform their non-Bayesian counterparts, and within the class of Bayesian models, the BVAR-d is found to be more accurate than the BVAR-l. For the two winning Bayesian models, i.e., the BVECM and the BVAR-d, the difference in forecasting power is more ambiguous; which model prevails depends on the desired forecasting horizon and the state of the economy. Hence, both Bayesian models may be considered when conducting research on house prices.

@en

Waar voorgaande crises gepaard gingen met dalingen in de huizenprijzen, blijft een dergelijke prijscorrectie bij de huidige coronacrisis vooralsnog uit en neemt het groeitempo zelfs toe. Wat zou hieraan ten grondslag liggen — is er soms sprake van een zeepbel?@en

This paper compares two model-based multilateral price indexes: the time-product dummy (TPD) index and the time dummy hedonic (TDH) index, both estimated by expenditure-share weighted least squares regression. The TPD model can be viewed as the saturated version of the underlying TDH model, and we argue that the regression residuals are “distorted toward zero” due to overfitting. We decompose the ratio of the two indexes in terms of average regression residuals of the new and disappearing items. The decomposition aims to explain the conditions under which the TPD index suffers from quality-change bias or, more generally, lack-of-matching bias. An example using scanner data on packaged men's T-shirts illustrates our framework.

@en

Location is capitalized into the price of the land the structure of a property is built on, and land prices can be expected to vary significantly across space. We account for spatial variation of land prices in hedonic house price models using geospatial data and a semi-parametric method known as mixed geographically weighted regression. To measure the impact on aggregate price change, quality-adjusted (hedonic imputation) house price indices are constructed for a small city in the Netherlands and compared to price indices based on more restrictive models, using postcode dummy variables, or no location information at all. We find that, while taking spatial variation of land prices into account improves the model performance, the Fisher house price indices based on the different hedonic models are almost identical. The land and structures price indices, on the other hand, are sensitive to the treatment of location.

@en

One of the main approaches to constructing quality-adjusted price indexes is the time dummy hedonic method. An alternative but rather unconventional method is the estimation of quality-adjusted unit value indexes. An advantage of the latter method is the interpretation of the implicit quantity index as the simple ratio of quality-adjusted or standardized quantities. In this paper we compare the two methods. We show that the expenditure-share weighted time dummy price index and the quality-adjusted unit value index can be written as ratios of weighted geometric and harmonic means, respectively, of quality-adjusted prices. Next, we argue that the two indexes will have similar trends and volatility if the quality-adjusted prices in the quality-adjusted unit value index are based on the estimated time dummy model. Our theoretical findings are illustrated on New Zealand scanner data for seven consumer electronics products.@en

Purpose Risk in home ownership from mortgage providers’ perspectives has received tremendous attention than individual home owner’s perspectives in existing literature following the financial crisis in 2007/2008 within the euro zone. This paper explored the risks factors in homeowners from the individual household’s perspectives within the owner-occupied housingsector in Dutch housing market. Design/methodology/approach The paper adopted a broader review of extent literature on the different concepts and views on risk in homeownership. These concepts are unified into a framework that enhances our understanding of the perceived sophisticated risk in owner-occupier with focus on the Netherlands. Findings: From the perspective of the homeowner, two main types of risks were identified: default payment and property price risk. The paper has unearthed a quantum number of factors which underline the above risks. These factors relate to the initial amount of mortgage loan taken out, the future housing expenses and the income development of the owner-occupier. Family disintegration is identified as one of the main causes of mortgage default and that of property price risk are mainly influenced by income levels, interest rates and conditions in the social and private rental sectors. Research limitations/implications Findings of the paper are based on review of the extant literature in the context of the Dutch housing market. Possible rigorous situational analysis using other tools are recommended for further research Originality/value This paper contributes to the much needed body of knowledge in the owner-occupied sector and provides a better understanding of risk in home ownership from the individual perspectives. @en

Purpose: This paper aims to examine the existence of the ripple effect from Amsterdam to the housing markets of other regions in The Netherlands. It identifies which regional housing markets are influenced by house price movements in Amsterdam. Design/methodology/approach: The paper considers the ripple effect as a lead-lag effect and a long-run convergence between the Amsterdam and regional house prices. Using the real house prices for second-hand owner-occupied dwellings from 1995q1 to 2016q2, the paper adopts the Toda–Yamamoto Granger Causality approach to study the lead-lag effects. It uses the autoregressive distributed lags (ARDL)-Bounds cointegration techniques to examine the long-run convergence between the regional and the Amsterdam house prices. The paper controls for house price fundamentals to eliminate possible confounding effects of common shocks. Findings: The cumulative evidence suggests that Amsterdam house prices have influence on (or ripple to) all the Dutch regions, except one. In particular, the Granger Causality test concludes that a lead-lag effect of house prices exists from Amsterdam to all the regions, apart from Zeeland. The cointegration test shows evidence of a long-convergence between Amsterdam house prices and six regions: Friesland, Groningen, Limburg, Overijssel, Utrecht and Zuid-Holland. Research limitations/implications: The paper adopts an econometric approach to examine the Amsterdam ripple effect. More sophisticated economic models that consider the asymmetric properties of house prices and the patterns of interregional socio-economic activities into the modelling approach are recommended for further investigation. Originality/value: This paper focuses on The Netherlands for which the ripple effect has not yet been researched to the authors’ knowledge. Given the substantial wealth effects associated with house price changes that may shape economic activity through consumption, evidence for ripples may be helpful to policy makers for uncovering trends that have implications for the entire economy. Moreover, the analysis controls for common house price fundamentals which most previous papers ignored.

@en

This paper establishes a simple affordability model that implicitly incorporates the major Dutch market features to elucidate long-run house prices under a regulatory environment. The results reveal a long-run relationship for house prices under strict regulations. The association among house prices, income, interest rates, and inflation is verified using an aggregated dataset. In the long-run, incomes and interest rates function as the two prime forces driving price dynamics, whereas the role of inflation is limited.

@en

House price modeling has been frequently used to investigate the dynamics of housing markets, especially competitive markets; yet less attention has been given to markets that have experienced considerable interventions. The aim of this study is to demonstrate a mismatch between conventional house price models and the case of the Netherlands and to provide reasons of such mismatch. We first describe and classify the conventional house price models into asset-pricing house price model, stock-flow model, multi-period utility model, and repayment model. These models are subsequently applied to the Netherlands, where considerable government interventions took place. As expected, the empirical results are unsatisfactory to explain the Dutch house price development. The degree of mismatch of the repayment model and the multi-period utility model, however, seems to be fairly limited.@en

The recent Global Financial Crisis has lent even greater urgency to the need for households to understand the risks and dynamics of the residential property market better. This paper uses a rich dataset on individual residential property transactions between 1995 and 2014 in Amsterdam to study the risks and the inter-dependency of house prices in the sub-district housing markets. The paper also examines the impact of house price growth in Amsterdam on the wider national trend. Simple summary statistics are adopted to characterise the dynamics and to compute the risks, while the inter-dependencies and the city-wide impact are analysed using Granger causality and cointegration techniques. The analysis establishes that house prices are generally higher, growing at faster and more volatile rates as we move from the peripheral to the districts into the central area. Furthermore, the appreciation rate of property prices in Amsterdam has a significant impact on the national trend, while there is limited systematic inter-dependency among the sub-markets themselves.@en

We find that unit value prices used for constructing the CPI should be for the same period as the index to be constructed, rather than for a sub-period. The latter approach can lead to an upward bias in the CPI.@en