The income elasticity of mortgage loan demand

• Author: Manthos D. Delis,Chris Tsoumas,Iftekhar Hasan
• Published date: 01 May 2019
• Date: 01 May 2019
• DOI: http://doi.org/10.1111/fmii.12108

DOI: 10.1111/fmii.12108

ORIGINAL ARTICLE

The income elasticity of mortgage loan demand

Manthos D. Delis1Iftekhar Hasan2Chris Tsoumas3

1Montpellier Business School, 2300 Avenuedes

Moulins, 34080 Montpellier, France

2FordhamUniversity and Bank of Finland,

5 Columbus Circle, New York,NY 10019

3School of Social Sciences, Hellenic Open

University, 18 ParodosAristotelous, 26335

Patra,Greece

Correspondence

ChrisTsoumas, School of Social Sciences, Hellenic

OpenUniversity, 18 Parodos Aristotelous, 26335

Patra,Greece.

Email:ctsoum@eap.gr

Abstract

One explanation for the emergence of the housing market bubble

and the subprime crisis is that increases in individuals’ income led to

higher increases in the amount of mortgage loans demanded, espe-

cially for the middle class. This hypothesis translates to an increase

in the income elasticity of mortgage loan demand before 2007.

Using applicant-level data, we test this hypothesis and find that the

income elasticity of mortgage loan demand in fact declines in the

years before 2007, especially for the mid- and lower-middle income

groups. Our finding implies that increases in house prices were not

matched by increases in loan applicants’ income.

KEYWORDS

distribution of income, income elasticity, mortgage loan demand,

subprime crisis

JEL CLASSIFICATION

G21, G01, D12, E44, R21

1INTRODUCTION

Is there a role for the income elasticity of mortgage loan demand in the shaping of the bubble that led to the subprime

crisis? Is this elasticity stable overtime or was it different in the pre-crisis period? The answers to these questions have

fundamental implications for the way we think about the origination of banking and financial crises, as well as the way

policy should respond to fluctuations in loan demand. The main demand-side explanation of the crisis is that credit

expansion in the market for mortgages came, inter alia, from political incentivesin the form of lax fiscal and monetary

policy, credit-promotion policies, and deregulation of the financial system, that increasingly induced the middle class

to own their own houses and raising loan demand (e.g., Rajan,2010).

This explanation translates to three main, non-mutually exclusive mechanisms, reinforcing the formation of the

housing bubble from the demand side. The first is that mortgage prices decreased (holding income and income elas-

ticity constant) and this yielded a surge in the quantity of loans demanded. A simple look at the yearly number of loan

applications (data are obtained from the Home Mortgage Disclosure Act-HMDA), especially between 2002 and 2006,

confirms these developments (see Panel A of Table 1). The second mechanism comes from an increase in individuals’

c

2019 New YorkUniversity Salomon Center and Wiley Periodicals, Inc.

Financial Markets,Inst. &Inst. 2019;28:115–139. wileyonlinelibrary.com/journal/fmii 115

116 DELIS ET AL.

TABL E 1 Summary statistics for the mortgage applications

Panel A: Number of applications byyear Panel B: Summary statistics for income and loan amount byyear

log(income) log(loan amount)

All

Conventionalloans for

home purchase

Conventionalloans for

home purchase with no

missing data

%oforiginated

conventionalloans for

home purchase with no

missing data Obs. Mean Median St. dev. Mean Median St. dev.

1992 12,006,230 3,347,873 1,215,554 70.83 1,215,554 3.94 3.95 0.68 4.39 4.51 0.86

1993 15,380,827 3,773,883 1,244,150 72.33 1,244,150 3.97 3.97 0.71 4.50 4.61 0.80

1994 12,194,357 4,686,721 2,080,713 68.73 2,080,713 3.89 3.89 0.69 4.35 4.47 0.85

1995 11,234,064 5,007,230 2,811,449 66.27 2,811,449 3.90 3.89 0.67 4.32 4.44 0.86

1996 8,571,193 3,561,029 1,611,305 64.89 1,611,305 3.88 3.87 0.68 4.27 4.38 0.85

1997 16,155,418 6,211,933 1,695,618 68.16 1,695,618 4.11 4.09 0.68 4.60 4.72 0.83

1998 24,661,738 7,746,152 3,812,334 61.97 3,812,334 4.00 4.01 0.69 4.45 4.56 0.85

1999 22,911,779 8,181,955 4,073,249 60.55 4,073,249 4.05 4.04 0.70 4.50 4.61 0.87

2000 19,233,174 8,079,906 2,184,819 60.79 2,184,819 4.19 4.17 0.69 4.61 4.74 0.92

2001 27,578,487 4,250,118 1,359,592 63.29 1,359,592 4.31 4.29 0.66 4.81 4.97 0.92

2002 31,236,040 4,455,971 1,560,352 63.30 1,560,352 4.34 4.32 0.66 4.92 5.06 0.88

2003 41,556,864 8,831,184 1,886,185 60.93 1,886,185 4.36 4.33 0.64 5.01 5.14 0.87

2004 33,607,736 11,448,747 7,008,954 61.97 7,008,954 4.30 4.28 0.66 4.86 4.94 0.87

2005 36,439,157 14,247,608 9,000,639 58.05 9,000,639 4.37 4.34 0.67 4.85 4.94 0.92

2006 34,105,441 13,819,614 8,143,137 57.34 8,143,137 4.45 4.42 0.68 4.86 4.95 0.93

2007 26,276,463 9,394,399 5,795,093 53.89 5,795,093 4.45 4.42 0.71 5.01 5.10 0.88

2008 17,391,570 4,625,621 2,806,395 56.86 2,806,395 4.45 4.42 0.74 5.12 5.20 0.84

2009 19,493,491 3,244,697 1,809,530 60.26 1,809,530 4.43 4.39 0.75 5.08 5.15 0.83

2010 16,348,557 2,753,491 1,754,707 59.20 1,754,707 4.46 4.44 0.76 5.06 5.15 0.88

2011 14,873,415 2,830,345 1,773,290 59.48 1,773,290 4.47 4.45 0.76 5.03 5.11 0.91

2012 18,691,551 13,613,630 2,102,655 61.84 2,102,655 4.49 4.48 0.75 5.09 5.18 0.87

Note: Panel A reports the number of mortgage applications byyear. Panel B reports summary statistics for the income and loan amount by year.