A modified spatial house price to income ratio and housing affordability drivers study: using the post-LASSO approach

• Date: 20 May 2024
• Pages: 1443-1460
• DOI: https://doi.org/10.1108/IJHMA-12-2023-0169
• Published date: 20 May 2024
• Subject Matter: Property management & built environment,Real estate & property,Housing markets
• Author: Qifeng Wang,Bofan Lin,Consilz Tan

A modif‌ied spatial house price to

income ratio and housing

af‌fordability drivers study: using

the post-LASSO approach

Qifeng Wang and Bofan Lin

School of Economics and Management, Xiamen University –Malaysia Campus,

Sepang, Malaysia, and

Consilz Tan

School of Economics and Management, Xiamen University –Malaysia,

Sepang, Malaysia

Abstract

Purpose –The purpose of this paper is to develop an index for measuring urban house price affordability that

integrates spatial considerations and to explore the drivers of housing affordability using the post-least absolute

shrinkage and selection operator (LASSO) approach and the ordinary least squares method of regression analysis.

Design/methodology/approach –The studyis based on time-seriesdata collected from2005 to 2021 for

256 prefectural-level city districtsin China. The new urban spatial house-to-priceratio introduced in thisstudy

adds the consideration of commuting costs due to spatial endowment compared to the traditional house-to-

price ratio. And compared with the use of ordinaryeconomic modelling methods, thisstudy adopts the post-

LASSO variable selection approach combinedwith the k-fold cross-test model to identifythe most important

driversof housing affordability,thus better solving the problemsof multicollinearity andoverf‌itting.

Findings –Urban macroeconomics environment and government regulations have varying degrees of inf‌luence

on housing affordability in cities. Among them, gross domestic product is the most important inf‌luence.

Research limitations/implications –The paper provides important implications for policymakers,

real estate professionalsand researchers. For example, policymakerswill be able to design policies that target

the most inf‌luentialfactors of housing affordability in their region.

Originality/value –This study introduces a new urban spatial house price-to-income ratio, and it examines how

macroeconomic indicators, government regulation, real estate market supply and urban infrastructure level have a

signif‌icant impact on housing affordability. The problem of having too many variables in the decision-making process

is minimized through the post-LASSO methodology, which varies the parameters of the model to allow for the ranking

of the importance of the variables. As a result, this approach allows policymakers and stakeholders in the real estate

market more f‌lexibilityin determining policy interve ntions. In addition, through the k-foldcross-validation methodology,

the study ensures a high degree of accuracy and credibility when using drivers to predict housing affordability.

Keywords Housing affordability, Machine learning, Post-LASSO variable selection,

K-fold cross-check, Price-to-income ratio, ArcGIS

Paper type Research paper

1. Introduction

Housing is a necessity in the personal lives of residents and is the foundation of family

stability and well-being. Due to the intricate intertwining of global real estate marketswith

This paper forms part of a special section “Spatial analysis and housing markets”, guest edited by

Koech Cheruiyot.

Housing

af‌fordability

drivers study

1443

Received5 December 2023

Revised11 March 2024

4April 2024

Accepted17 April 2024

InternationalJournal of Housing

Marketsand Analysis

Vol.17 No. 6, 2024

pp. 1443-1460

© Emerald Publishing Limited

1753-8270

DOI 10.1108/IJHMA-12-2023-0169

The current issue and full text archive of this journal is available on Emerald Insight at:

https://www.emerald.com/insight/1753-8270.htm

f‌inancial and capital systems, the substantial surge in housing prices has given rise to

signif‌icant challenges in housing affordability for numerous residents (Han et al., 2018).

Despite the f‌lourishing of the real estate sector and the growth of national economies

worldwide, the global communitycontinues to grapple with housing affordability problems,

notably affecting individuals with medium to low incomes and migrants (Saiz, 2023;

Yiu et al., 2023). The annual meeting of the United Nations Commission for Social

Development noted that the COVID-19 pandemic has further exacerbated the problem of

housing affordability(Aliyu, 2022).

In response to the urgency highlighted by the World Human Rights Organization, certain

countries, such as the prominent player China, are actively engaged in adopting suitable housing

policies to ensure the realization of human rights related to housing (United Nations, 2023). In

China, the government has implemented various housing security systems, including low-cost

housing, an affordable housing system and a housing provident fund system. However, the

existing policy system is not f‌lawless and has not completely succeeded in effectively mitigating

the increasing housing diff‌iculties and problems faced by the majority of Chinese residents. Due

to the mismatch between sustained high growth in urban housing prices and low growth

incomes in recent years, the persistent issue of low affordability among Chinese residents

remains a signif‌icant concern for both Chinese society and policymakers (Xie et al.,2013).

Therefore, in-depth understanding and testing of the resident’s ability to pay for housing and

clarifying its driving factors are of great signif‌icance in helping to formulate real estate

regulation and protection policies and improve the residents’ability to pay for housing, which is

what some scholars are currently delving into. Li et al. (2020), who examined 275 prefecture-level

cities in China from 2014 to 2018, found that some of China’s“superstar cities”, including

Beijing, Shanghai and Xiamen, had serious housing affordability problems, while the

affordability levels of other cities remained generally stable.

Housing affordability refers to the abilityof a household or individual to afford a decent

and comfortable home after meeting their minimum living condition and expenditure

(Galster and Lee, 2021). Housing affordability is measured in a variety of ways, including

price-to-income ratios, housing expenditure-to-income ratiosand residual income measures.

Among them, the price-to-income index is currently the most widely used method and the

computation of the ratio usesreadily available macroeconomic data to compare house prices

with individual or household income to determine whether a region’s housing market is

exceeding affordability levels.However, the computation of this housing price index lacks a

theoretical basis and it is diff‌icult to f‌ind a comprehensive model of the causes of the

housing price index in existing studies (Gan and Hill, 2009). Besides, existing research on

house price affordability indicators over the world is more concerned with theconstruction

and optimization of indicators, relativelylittle research has been conducted on the causes of

housing affordability, as well as research regarding the accuracy of prediction and

interpretabilityof the resulting statistical model.

Decomposition analysis is the main existingresearch approach to the drivers of housing

affordability. Several studies in different contexts have investigated, macroeconomic

indicators that are considered to be a priority and important consideration inf‌luencing the

housing price to affordability index (HPAI), such as Malaysia (Latif et al.,2020;Selvaraja

et al.,2018;Zarul Azhar bin Nasir, 2022), Kenya (Kieti and K’Akumu, 2017), New Zealand

(Robinson et al., 2006), the USA (Bogdon and Can, 1997;Calder, 2017;Hulchanski, 1995;

Kutty, 2005), the UK (Mulliner et al.,2013), Hong Kong (Hui, 2001) and Shanghai (Zhou et al.,

2010). Supply is also often a consideration for researchers. From a demand perspective, the

ability of a property to meet consumers’basic needs for food, education and transport is

considered to be one of the factors inf‌luencing affordability.From a supply perspective, the

IJHMA

17,6

1444