Australian Housing & Transport Affordability

Funded by the Lord Mayor's Charitable Foundation


Housing affordability is declining in Australia. According to the Reserve Bank, over the past 30 years, the ratio of housing prices to income has increased substantially. This project aims to develop a new tool to measure housing and transport affordability in Australia, with Melbourne metropolitan selected as the pilot study area. Housing affordability is traditionally measured using the percentage of income spent on housing costs. As a common rule, households who spend more than 30% of their income on housing costs while earning in the bottom 40% of the income range are considered to be under housing stress. An important cost that is usually overlooked in measuring affordability is the transport or accessibility costs.


Housing costs are defined as the regular expenses made by a household for accommodation. These costs refer to all housing expenses such as mortgage repayments, rent payments, maintenance costs, taxes and other expenses. A review of previous studies suggests two main approaches for estimating housing costs. The first method uses either mortgage repayment only or rent payment only, whereas the second method relies on both mortgage repayment and rent payment. Here, the housing cost is estimated using the weighted average of median weekly mortgage repayment and median weekly rent payment for each statistical area. Median rather than the mean values are used to avoid skewing the results by outliers. The model also distinguishes the costs of owning or renting a unit versus a house. The mortgage cost data was computed based on (1) average 2016 median property price at each SA2 zone, (2) a weighted average loan to value ratio of 71.25% from APRA 2016 data (September quarter), and (3) a 5-years fixed interest rate of 4.39% from major lenders such as Westpac Bank at the end of 2016. Weekly rental cost was based on 2011 Census data, however, it was adjusted for inflation to 2016 value. Other variables included in the housing costs equation were computed from the 2011 Census data.


The transportation costs mainly consist of the private vehicle cost as well as public transportation cost. The private vehicle cost consists of a fixed cost of owning a vehicle and its varying operational costs including fuel, tyres and car servicing, using an average of all car makes and models listed in the RACV report 2014 in total cost per kilometre. Recognizing the fact that travel patterns of households vary based on the trip purpose and time, we distinguish between the costs associated with journey to work, weekday non-work trips and weekend trips.

We compute work-related travel cost using two main datasets from ABS Census including the place of work and journey to work data. A 281 by 281 Origin-Destination (O-D) demand matrix is developed. The O-D matrix provides information on the number of people traveling between zone pairs using different travel modes. We calculate the costs associated to each of the frequently used modes of travel including private car, public transport, bicycle, and walking. We obtain the average travel distance between zone pairs from Google API represented in a 281 by 281 matrix. The average travel cost per kilometre by a private vehicle is calculated by taking the average cost over all vehicle makes and models included in the RACV's car ownership cost report. We calculate the cost of public transportation based on the figures obtained from Public Transport Victoria (PTV).

We extract information for weekday non-work trips from VISTA data. An individual person makes on average 1.2 non-work trips on a weekday in Melbourne metropolitan area. Using the household travel survey data and working population from the 2011 ABS Census, the average number of non-work trips per household can be calculated for each SA2 area. The transportation mode share is assumed to be the same for both work and non-work trips and thus, we can estimate the average percentage of private car users for each SA2. The travel cost associated to non-driving (non-auto) only includes the cost of public transportation.

We extract information for weekend travel data from VISTA to compute the household weekend travel cost. An individual person makes on average 2.76 trips on a weekend in Melbourne. We assume that the transportation mode share for weekend is the same as work trip or weekday transportation mode share. The percentage of population who are using a private car is also assumed to be constant. For private car trips, it is assumed that the family is traveling in the same vehicle. Non-driving travel cost is also assumed to include public transportation cost only.

The total transportation cost per household for each SA2 is simply the sum of weekday work and non-work travel cost, weekend travel cost, and average weekly vehicle ownership cost.


The traditional way of measuring housing affordability is inadequate as it overlooks transportation costs. We simply argue that living in the outer areas away from the CBD does not necessarily reduce the cost of living in Melbourne. Once the cost of transportation is taken into account, most of the outer suburbs become less affordable while some of the inner areas become more affordable.

A comprehensive analysis of location affordability must not be solely based on the housing and transportation cost distribution maps provided here. The comparison of affordability of two locations should also recognize differences in lifestyle, neighborhood characteristics, demography, etc. Overall, this project provides an improved way of measuring affordability of a location by taking into consideration both combined housing and transportation costs including public transportation expenses. This provides a more comprehensive way of thinking about the true affordability of place.



Meead Saberi

Chief Investigator

Lecturer in the Institute of Transport Studies at Monash University, PhD in Transportation Systems Analysis and Planning.


Dharma Arunachalam

Chief Investigator

Associate Professor in the School of Social Sciences at Monash University, PhD in Demography.


Jonathan Smith

Chief Investigator

Research Fellow in the School of Social Sciences at Monash University, PhD in Sociology.


Dr. Meead Saberi
Department of Civil Engineering
Monash University

+61 3 9905 0236
Room 101, Building 60, Clayton campus