The House Price Index and its development from mean price thru Repeat Sales to SPAR Paul de Vries, Delft University of Technology & Erna van der Wal, Statistics Netherlands Department of Lands and Survey of Cyprus Kadaster International November 29 December 3, 2009 Delft University of Technology OTB Research Institute for the Built Environment CBS, Statistics Netherlands Challenge the future
Starting1628 300 250 reële prijs (euro 2009) x1000 Industrial Revolution 1850 Credit Crunch 2009 200 150 Napoleon in Amsterdam 1795 Oil crisis 1979 100 50 3 th English War 1672-1674 0 Crash Tulipmarket, 1637 1630 1650 1670 1690 1710 1730 1750 1770 1790 1810 1830 1850 1870 1890 1910 1930 1950 1970 1990 2010 Challenge the future
300 250 200 150 100 50 0 1965 1967 1969 1971 1973 1975 1977 1979 1981 1983 1985 1987 1989 1991 1993 1995 1997 1999 2001 2003 2005 2007 2009 Nominal House Price euro X 1000 Starting 1995 we use the nominal house price for estimating the House Price Index Challenge the future
House Price Index 300 index Jan 1995 = 100 250 200 150 100 50 Depends on registration Monthly ~ 16.500 transactions repeat sales index hedonic index spar index mean price 0 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 Challenge the future
Outline Aim 2005: Mean Price to Repeat Sales Index (RSI) 2008: RSI to Sales Price Appraisal Index (SPAR) Comparison Hedonics RS SPAR The House Price Index and its development 5 28
Aim The House Price Index and its development 6 28
Aim of the index: the actual house price Dutch Central Bank: No trust in common summary methods; Valuation portfolios: Haircut 25% by Price Rise; Thus: actual house price to low and higher reserve for mortgage risk; Solution: House Price Index for the housing market. The House Price Index and its development 7 28
Mean price and the Index 14 12 10 8 6 4 2 0-2 -4-6 -8 % Index Prijsindex [Kadaster/CBS] Gemiddelde [Kadaster/CBS] Mean 1995 1997 1999 2001 2003 2005 2007 2009 The House Price Index and its development 8 28
Hedonic regression The House Price Index and its development 9 28
Hedonic Valuation Process Observed House Quality House Price Sold or On-the-Market Hedonic regression - some characteristics - House Price Estimated House Price All properties The House Price Index and its development 10 28
2005 Repeat Sales Index The House Price Index and its development 11 28
Data Dutch Land Registry Office (Kadaster) Actual (delay: two weeks) Complete (all transaction) Less characteristics (only type) Only existing homes (no new dwellings) 2005: Given the data, we chose the Repeat Sales Model for calculation the index The House Price Index and its development 12 28
Why Repeat Sales Based on Price for single property at different points in time Literature Bailey et al. (1963); Case and Shiller (1987, 1989, 1990) Practice OFHEO House Price Index (USA) Problem revision The House Price Index and its development 13 28
Housing stock and transactions Owners occupied stock Sold 44% 1.500.000 Sold twice or more 46% 720.000 560.000 repeat sales 3.500.000 Sold once 54% Not sold 56% The House Price Index and its development 14 28
Repeat Sales The House Price Index and its development 15 28
Impression Data Month 2 between two sales 1,8 1,6 1,4 1,2 1 0,8 0,6 0,4 0,2 0 apartments terraced corner semi-detached detached % Repeat Sales Transaction 100,0 90,0 80,0 0 10 20 30 40 50 60 70 80 90 100 110 120 130 140 number of month 70,0 60,0 50,0 40,0 30,0 20,0 10,0 apartments terraced corner semi-detached detached 0,0 1993The 1994 House 1995 Price 1996 1997 Index 1998 and 1999 its development 2000 2001 2002 16 2003 282004 2005
Results 300 index under upper The Netherlands Total 250 200 150 100 50 280 270 260 200404 200405 200406 200407 200408 200409 200410 200411 200412 200501 200502 200503 Revision 0 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 250 240 230 220 210 200 2001 2002 2003 2004 2005 The House Price Index and its development 17 28
Results 350 300 250 index under upper South apartements 200 150 100 50 280 270 260 250 200404 200405 200406 200407 200408 200409 200410 200411 200412 200501 200502 200503 Revision 0 230 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 240 220 210 200 2001 2002 2003 2004 2005 The House Price Index and its development 18 28
2008 Sales Price Appraisal Ratio The House Price Index and its development 19 28
Background Owner-occupied housing currently excluded from HICP Eurostat pilot study The House Price Index and its development 20 28
Sale Price Appraisal Ratio method Bourassa et al. (Journal of Housing Economics, 2006). the advantages and the relatively limited drawbacks of the SPAR method make it an ideal candidate for use by government agencies in developing house price indexes. Used in new Zealand since early 1960s; also in Sweden and Denmark Promising results in Australia (Rossini and Kershaw, 2006) Based on (land registry s) sale prices p and official government appraisals a Model-based approach using appraisals The House Price Index and its development 21 28
Reliability of Appraisals R2 1995: 0,90 1999: 0,94 2003: 0,95 Appraisal reasonably useful for SPAR The House Price Index and its development 22 28
Distribution: Ajt / Sjt ~ 1 1995 1999 2003 Black bars: 0,95 1,00 1,00 1,05 The House Price Index and its development 23 28
Housing stock and transactions Owners occupied stock Sold 44% 1.500.000 Sold twice or more 46% 720.000 560.000 repeat sales 3.500.000 Sold once 54% Not sold 56% 1.500.000 SPAR The House Price Index and its development 24 28
Sale Price Appraisal Patio (SPAR) principle Ratio of Sum Sale Prices (S jt ) in current period and sum Appraisals (A j0 ) in base period divide by ratio of Sum Sale Prices (S j0 ) and sum Appraisals (A j0 ) in base period The House Price Index and its development 25 28
Sale Price Appraisal Ratio (SPAR) SPAR index The House Price Index and its development 26 28
SPAR <> Repeat Sales (development) 350,0 300,0 250,0 200,0 150,0 100,0 50,0 0,0 199501 199601 199701 199801 199901 200001 200101 200201 200301 200401 200501 200601 200701 SPAR RS The House Price Index and its development 27 28
SPAR <>Repeat Sales: precision 350,0 0,10 0,09 300,0 0,08 250,0 0,07 0,06 200,0 0,05 150,0 0,04 0,03 100,0 0,02 50,0 0,01 0,00 0,0 199501 199601 199701 199801 199901 200001 200101 200201 200301 200401 200501 200601 200701 Precision SPAR - Overijssel SPAR Precision RS RS - Overijssel The House Price Index and its development 28 28
Conclusion The Repeat Sales Index is a efficient model Drawback RSI: revision SPAR Method most precise index Appraisals (in the Netherlands) are reliable We uses the SPAR Model The House Price Index and its development 29 28
Finish De bocht van de Herengracht (1671-72) Gerrit Berckheyde The House Price Index and its development 30 28