- Количество слайдов: 23
Winning Bidders: Is there a strategy? Louise Brown, Stanley Mc. Greal and Alastair Adair
Research questions § Is there a distinct pattern emerging from the transaction histories of the winning bidders? § Does this change during different stages of the house price cycle? Under investigation § Duration of the bidding process § Observed behaviour of winning bidder through incremental time and amount of bids.
TOM § TOM and listing strategies (Bjorklund, Dadzie, Wilhemmson, 2003; Allen and Dare, 2004; Knight, Sirmans and Turnbull, 1994; Asabere, Huffman and Mehidian, 200; Mc. Greal et al, 2007) § Bid-Ask Spread (Glosten and Milgrom, 1985; Judd, Winkler and Kissling, 1995) § Buyer search under differing economic conditions (Baryla, Zumpano and Elder, 2000; Novy-Marx, 2009)
Bidding § Evaluation of methodologies (Leung, Leong and Chan, 2002; Pryce and Gibb, 2006; Black et al, 2003) § Extreme Bids (Levin and Pryce, 2007) § Jump bidding (Raviv, 2007) Aggressive bidders (De Silva et al, 2002) § Individual transaction histories, UK data (Merlo and Ortalo. Magne, 2004) § Bid acceptance strategies (Green and Vandell, 1998) § Bidding model of perfect competition (Robert Wilson, 1977) § Game theory and decision analysis (Howard Raiffa, various)
Lessons from other literature § Auction data and bidding at auctions (Some fundamental differences: 1)sequential open market transactions can be simultaneous 2) agreed finish time for an auction compared to unfixed duration of the bidding process. § Internet auctions reported in economic journals : cross bidders/observers (Anwar, 2006; Housers and Wooders, 2006) § Motivation: emotional bidders, rational/irrational exuberance (Schiller; Neo, Ong and Tu, 2008) particularly relevant in a rapid market (2006/7) § Entry point, opportunity cost of an early bid (Brosig and Reib, 2007)
Data § 2001 qtr 3 -2009 qtr 1 (31 quarters) § For a metropolitan area § Sample size = 991 complete cases § Source: an estate agency practice -high %sales for Belfast Metropolitan Area -spatially diverse -mass market not niche -representative of other agents -uniformity of approach across branches -long established network
Time period under investigation Source: University of Ulster
Characterization of the relationship between viewing and bidding periods
Data Variables Individual property identifier Date of listing Date of agreement Date of completion Listing Price Sales Price Full Postal Address House Type Age of Property Nominal Floor Area Number of Bedrooms Number of Reception Rooms Bathroom Garage Full Central Heating In need of modernization 00001 dd/mm/yyyy £ 300, 000 £ 315, 000 BT 37 0 QB Detached Bungalow 1960 -1980 1200 sqft 3 2 Yes Yes
Bidding Information Example Bidding Transaction Data Individual Property Identifier 00001 Date of Listing Date of 1 st Bid Date of 2 nd Bid Date of 3 rd Bid Date of 4 th Bid Date of 5 th Bid Date of 6 th Bid Date of 7 th Bid Date of 8 th Bid 26/3/7 2/4/7 4/4/7 5/4/7 13/4/7 1/5/7 12/5/7 Number of viewings: 44 Number of bids: 8 Number of bidders: 7 £ 315, 000 (LP) £ 315, 000 £ 315, 500 £ 316, 000 £ 317, 500 £ 318, 000 £ 320, 000 £ 322, 000 £ 330, 000 Bidder A Bidder B Bidder C Bidder B Bidder D Bidder E Bidder G *sale agreed
Descriptive statistics TOM Range = 0 to 1023 (2009 qtr 1), mean =96, mode=14 No of bids Range = 1 to 11 (2006 qtr 3), 90% less than or equal to 10 bids No of bidders Range = 1 to 13 (2004 qtr 1), 90% less than or equal to 4 bidders No of viewers Range = 0 -111 (2004 qtr 2), 90% properties had less than or equla to 29 viewers
Time on the market (listing to agreement)
House price cycle and viewers, bidders and bids
Incremental timing by qtr
Incremental amount by yr
Incremental time by yr
Correlations between bidding variables Winners Incremental timing of winning bid Losers Incremental timing of losing bids Winners incremental bid as a % of AP (normalised) Losers incremental bids as a % of AP (normalised) Sales Price Asking Price -. 227** -. 184** . 299** -. 178** Time on the market . 366** . 475** -. 225** . 417**
Findings 1 Volume of Viewers and Bidders § Peak of viewers and bidders dropped before the decrease was realised in prices therefore it could be a lead indicator.
Findings 2 Incremental amounts § Initial findings suggest that in a time of rapidly increasing prices winning bidders are prepared to go higher thereby securing the property. § Conversely in a time of rapidly decreasing prices winning bidders have >- SP-AP and still secure the property (possibly indicating better market knowledge).
Findings 3 Timing § Winning bids appear to wait for a shorter time than losers indicated by a lower incremental time between the penultimate bid and the last bid compared with the incremental time of all previous losing bids. § This may help explain the earlier finding on price. In thin markets with few sales the winning bidder bids acts more quickly but with a lower price taking advantage of market conditions. § Analysis shows that the difference between the amount of bids for winners and losers is more significant than the timing.
Bidder typologies for residential open market transactions Data suggests three typologies: § The sole bidder § The patient bidder (responsive to other bidders) § The aggressive bidder § Some evidence that the sole bidder (35% cases) and the aggressive bidder are winning bidders (bids in higher increments and with shorter incremental time).
Thank you for listening. Any questions?