The Determinants of Buyer Search Duration in “Hot” and “Cold” Residential Real Estate Markets

Presented by Ekaterina Chernobai page 1ERES 2011, Eindhoven 1

The Determinants of Buyer Search Duration in “Hot” and “Cold” Residential Real Estate Markets

Ekaterina Chernobai

• California State Polytechnic University, Pomona, U.S.A. College of Business Administration

Department of Finance, Real Estate, and Law• University of Nürtingen, Germany (Visiting Professor)

Department of Real Estate Management

Tarique Hossain

• California State Polytechnic University, Pomona, U.S.A.College of Business AdministrationDepartment of International Business and Marketing

Presented by Ekaterina Chernobai page 2

Idea

Housing liquidity

Time on the market Time to buy

Seller side Buyer side

We analyze:

• Determinants of search duration

• Differences in the effects in “hot” and “cold” housing markets

• Differences between private investors and non-investors

Also:

• Effects of “subprime crisis”

ERES 2011, Eindhoven


Idea

0

50

100

150

200

250

300Ja

n-95

Jan-

96

Jan-

97

Jan-

98

Jan-

99

Jan-

00

Jan-

01

Jan-

02

Jan-

03

Jan-

04

Jan-

05

Jan-

06

Jan-

07

Jan-

08

Jan-

09

Jan-

10

Cas

e-S

hil

ler

ho

use

pri

ce i

nd

ex

Los Angeles MSA

Composite 10

Composite 20

2004-2005 survey of recent house buyers

We run: 2007-2008 survey of recent house buyers

• Same geographic area in Southern California – most pronounced “bubble”• Same average house price in both time periods

January 2000 = 100



Idea

Geographical coverage of the surveys

San Luis Obispo county

Santa Barbara county

Ventura county

San Francisco

Los Angeles

San Diego


California


Idea

Some past studies of search duration:

• Case and Shiller (1988) “The Behavior of Home Buyers in Boom and Post-Boom Markets”

- House purchases during 1-year period

- Different geographic locations

- Some with rising some with declining prices

• Baryla, Zumpano, Elder (2000) “An Investigation of Buyer Search in the Residential Real Estate Market under Different Market Conditions”

- Houses purchased in different interest rate periods

- Different geographic locations (entire country)



Idea

• Krainer and LeRoy (2002) “Equilibrium Valuation of Illiquid Assets”

- Theoretical general equilibrium model of housing transactions

- Buyers’ valuations of houses are heterogeneous

- Weigh the search cost against the expected life-time utility less price

- Equilibrium: Longer expected stay in a house increases search duration

• E. Chernobai (working paper) “When Does Mobility Reduce Liquidity”

- Theoretical model that generalizes Krainer & LeRoy’s

- Short-term and long-term buyers

- Equilibrium: Long-term buyers search longer than short-term buyers



Research Questions

Questions of interest:

1.) Any effect of expected housing tenure on realized search duration?

2.) Are consumption-buyers different from investor-buyers?

3.) Any differences in 1.) and 2.) between “hot” and “cold” markets?

Our hypotheses:

Positive relationship if buy for consumption

Investors:- Time pressure to buy: Yes? No?- What does a typical consumption-

buyer want?

Short-term investors vs.Long-term investors

- Time to buy is shorter in “hot” market- Investors: Rel. proportion of short- &

long-term investors varies over time


Results


Survey 2004-2005 recent house-buyers

Mailed 6,000 questionnaires

Response rate: 11.3% (661)

Survey 2007-2008 recent house-buyers

Mailed 6,200 questionnaires

Response rate: 11.6% (719)

“Hot” market “Cold” market


Results


Variable Pooled “Hot” “Cold” Difference in means (p-value)Time To Buy 6.80 5.35 8.06 0.000***

First 0.24 0.25 0.23 0.538

Offers 1.78 1.83 1.74 0.279

New 0.09 0.07 0.10 0.076*

Miles 201.40 178.49 221.33 0.251

Price 612,625 602,267 621,900 0.284

Moving Up 0.32 0.32 0.32 0.987

Coast 0.38 0.40 0.36 0.336

Fixed 0.53 0.41 0.63 0.000***

HighLTV 0.34 0.33 0.35 0.504

1-5 years 0.34 0.41 0.27 0.000***

6-10 years 0.22 0.20 0.24 0.158

10+ years 0.37 0.31 0.42 0.000***

1-5 years *Investor 0.04 0.07 0.02 0.001***

6-10 years *Investor 0.01 0.01 0.02 0.281

10+ years*Investor 0.01 0.01 0.01 0.674

Investor 0.09 0.11 0.07 0.074*

Winter 0.16 0.06 0.25 0.000***

Spring 0.22 0.23 0.22 0.523

Summer 0.29 0.32 0.26 0.060*

Fall 0.32 0.38 0.27 0.001***

(months)

Correlation matrix:

no issues

DESCRIPTIVE STATISTICS


Results


Variable Pooled Hot market Cold marketIntercept 0.899 (<0.0001) 0.305 (0.1277) 0.291 (0.1440) -0.434 (0.1259) 0.887 (0.0008)

First 0.056 (0.5945) 0.051 (0.6236) 0.057 (0.5770) 0.373 (0.0089) -0.165 (0.2216)

Offers 0.218 (<0.0001) 0.212 (<0.0001) 0.221 (<0.0001) 0.195 (<0.0001) 0.280 (<0.0001)

New 0.476 (0.0009) 0.437 (0.0017) 0.445 (0.0010) 0.987 (<0.0001) 0.124 (0.4447)

Miles 0.000 (0.9190) 0.000 (0.7421) 0.000 (0.6030) 0.000 (0.0523) 0.000 (0.6554)

Price 0.000 (0.0054) 0.000 (0.0116) 0.000 (0.0120) 0.000 (0.0029) 0.000 (0.4476)

MovingUp -0.113 (0.2220) -0.150 (0.0991) -0.123 (0.1764) -0.284 (0.0306) 0.010 (0.9276)

Coast 0.236 (0.0040) 0.224 (0.0055) 0.227 (0.0046) 0.385 (0.0007) 0.227 (0.0323)

Fixed 0.027 (0.7399) -0.007 (0.9335) 0.012 (0.8768) 0.218 (0.0508) -0.200 (0.0595)

HighLTV -0.186 (0.0418) -0.143 (0.1163) -0.143 (0.1115) -0.186 (0.1783) -0.062 (0.5831)

1-5 years 0.621 (<0.0001) 0.528 (0.0009) 0.718 (0.0006) 0.405 (0.0614)

6-10 years 0.557 (0.0007) 0.560 (0.0006) 0.659 (0.0038) 0.487 (0.0247)

10+ years 0.866 (<0.0001) 0.877 (<0.0001) 0.803 (0.0002) 0.787 (0.0001)

1-5 years *Investor 0.540 (0.0060) 0.336 (0.1299) 1.028 (0.0034)

6-10 years *Investor -0.160 (0.6475) 0.617 (0.3038) -0.465 (0.2406)

10+ years*Investor -0.545 (0.1145) -0.197 (0.7039) -0.712 (0.0872)

Winter 0.351 (0.0031) 0.336 (0041) 0.315 (0.0065) 0.782 (0.0007) -0.030 (0.8263)

Spring 0.180 (0.0965) 0.172 (0.1027) 0.148 (0.1589) 0.274 (0.0519) 0.099 (0.4874)

Summer 0.259 (0.0102) 0.254 (0.0097) 0.247 (0.0110) 0.340 (0.0078) 0.105 (0.4400)

N obs. 806 806 806 375 431

Dependent variable:

logTimeToBuy

Weibull distribution

REGRESSIONRESULTS

consumption incentive


Results


We also looked at Pooled regression model with:

• “Hot” dummy interaction terms for all variables significant for many Splitting into “hot” and “cold” sub-samples is justified

• “Investor” dummy interaction terms for all variables significant for Tenure variables, Coast, and New

To identify consumption- and investment-driven submarkets, split “Hot” and “Cold” samples into

- Coastal zip code areas (43% of zip codes, 38% of obs.)

- Inland zip code areasERES 2011, Eindhoven

Results


Variable“Hot” (2004-2005) “Cold” (2007-2008)

Coastal Inland Coastal Inland

Intercept -0.378 (0.3663) -0.150 (0.6918) 1.099 (0.0107) 0.991 (0.0019)

First 0.508 (0.0663) 0.387 (0.0249) -0.612 (0.0118) -0.041 (0.8009)

Offers 0.176 (0.0097) 0.255 (<0.0001) 0.170 (0.0623) 0.323 (<0.0001)

New 0.297 (0.4324) 1.495 (<0.0001) 0.857 (0.0183) -0.009 (0.9608)

Miles 0.000 (0.0942) 0.000 (0.2333) 0.000 (0.2430) 0.000 (0.2254)

Price 0.000 (0.0984) 0.000 (0.1175) 0.000 (0.2863) 0.0000 (0.4855)

MovingUp -0.097 (0.6428) -0.348 (0.0331) -0.048 (0.7961) 0.188 (0.1797)

Fixed 0.281 (0.1264) 0.230 (0.0968) -0.074 (0.6780) -0.297 (0.0188)

HighLTV -0.644 (0.0116) 0.010 (0.9521) -0.189 (0.2999) -0.088 (0.5318)

1-5 years 0.986 (0.0028) 0.507 (0.0503) 0.192 (0.5651) 0.608 (0.0270)

6-10 years 1.059 (0.0027) 0.353 (0.2232) 0.229 (0.4749) 0.696 (0.0109)

10+ years 1.290 (<0.0001) 0.456 (0.0915) 0.965 (0.0014) 0.742 (0.0048)

1-5 years *Investor 0.419 (0.2046) 0.533 (0.0687) -0.930 (0.1218) 1.399 (0.0004)

6-10 years *Investor 0.189 (0.8648) 0.630 (0.3684) -0.962 (0.1176) -0.235 (0.6253)

10+ years *Investor 0.158 (0.8296) -1.258 (0.0740) -1.751 (0.0029) 0.039 (0.9429)

Winter 0.967 (0.0134) 0.554 (0.0689) 0.220 (0.3705) -0.239 (0.1256)

Spring 0.540 (0.0274) 0.106 (0.5366) 0.217 (0.3676) 0.093 (0.5740)

Summer 0.499 (0.0125) 0.104 (0.5301) 0.110 (0.6388) 0.119 (0.4519)

N obs. 149 226 157 274

Dependent variable:

logTimeToBuy

Weibull distribution

REGRESSIONRESULTS

consumption incentive

investment incentive - short term -

investment incentive

- long term -??? next housing cycle


Results

Presented by Ekaterina Chernobai page 13ERES 2011, Eindhoven

One possible explanation of differences between Coast & Inland results:

Hot Cold

Coast Inland Coast Inland

Variable Mean SD Mean SD Mean SD Mean SD

Search duration(months)

5.52 7.64 4.39 7.25 8.67 12.695 7.60 9.43

Purchase price($)

618,639 250,283 578,151 218,562 627,267 278,226 615,884 287,611

FRM financing (% of borrowers)

31.30 46.55 46.34 49.99 65.44 47.73 72.06 44.96

Conforming loans (% of borrowers)

12.35 32.90 15.15 35.86 36.73 48.21 33.98 47.36

Purchase price ($)

469,048 201,542 501,429 148,267 531,861 182,174 519,207 181,868


50.00 51.30 72.73 45.23 75.93 43.16 81.82 38.79

Non-conforming loans (% of borrowers)

87.65 32.90 84.85 35.86 63.27 48.21 66.02 47.36

Purchase price ($)

644,631 245,991 622,959 236,499 652,444 296,068 655,288 323,907


28.80 45.47 39.53 49.04 60.22 49.21 69.59 46.14

(means)

< <

<

<

<

>

Summarize


• Consumption-driven purchases: - “hot” market in coastal areas - “cold” market in inland areas

Long-horizon investment activity: - in just the opposite sub-markets

• Short-horizon investment activity: - in inland areas

Investment activity leads consumption activity


• The boom & burst of the “sub-prime” bubble affected search duration in Hot & Cold periods

The Determinants of Buyer Search Duration in “Hot” and “Cold” Residential Real Estate Markets

Documents

The Determinants of Buyer Search Duration in “Hot” and “Cold” Residential Real Estate Markets