The rise of poverty on the outer fringes is not unique to America. Relative poverty exists in the suburbs of Melbourne as well. How do we know? Some of the evidence is readily accessible.
It seems intuitive to think that house prices decrease the further you live from the City. The Department of Health’s Local Government Area Profile for 2012 provides a readily accessible dataset for both median house prices for 2012 and distance from Melbourne.
Figure 1 Median house prices indicates that distance from the CBD is a relatively good proxy for relative median house price. Exceptions include established inner [Bayside, Boroondara and Stonnington] and middle ring suburbs [Glen Eira, Manningham, Monash and Whitehorse] as well as beachside localities [the Borough of Queenscliffe, the Surf Coast and Mornington Peninsula Shires]. In the main however, the closer you live to the CBD, the more expensive your house purchase price is likely to be.
In an equal world we might assume that housing stress and housing poverty fall the further out from the CBD you go, given the fall in house prices over distance. This proposition is worth exploration. It assumes people who live further out do so because either can’t afford to pay higher mortgages, can’t attract more finance for larger mortgages, or don’t feel comfortable with higher mortgages required to purchase a house closer to the CBD.
Figure 2 illustrates Mortgage Payment rates by LGA from the 2011 Census. Not only are households further from the CBD more likely to pay mortgages rather than rent but the proportion of households further from town paying higher mortgages is higher than for those living closer to town. The proportion paying lower mortgage levels is also relatively higher. This makes the proposition both true and false at the same time. True for the small proportion paying between $600 and $2199 per month, and false for those paying on average more than $500 per week, of which there are more the further the distance from the CBD.
To recap, it is more likely that the further out you live the bigger mortgage you will pay, rather than lower repayment rates.
Rental housing, visible as part of the gap between reported mortgage payment rates and 100%, is more likely to be available, and available in increasing proportions, closer to the CBD. Moving to rent at the urban fringe is unlikely to deliver ready access to rental or cheaper rental.
Paying a higher mortgage and living further from town is no indication of capacity to pay. People in the CBD earn more relative to those living further from the CBD. The further you move from the centre of town the more you can expect relative median wages to fall.
Figure 3 Median Household Income illustrates median income levels for all Victorian Local Government Areas in the 2011 Census, by distance from the CBD. The exponential trend line illustrates the relative performance against the trendline, where some local government areas perform better or worse than relative distance might otherwise suggest, but generally the trend holds across the State.
Figure 4 Mortgage Stress illustrates Department of Health derived data for mortgage and rental stress of low income households only by Local Government Area from the Census in 2011. The proportion of low income residents in mortgage and rental stress rises slightly with distance from town, suggesting any relative benefit distance brings in lower house prices is undermined by relatively lower income levels. The pattern does not appear consistent with distance, however the use of only low income resident data to estimate mortgage and rental stress will have influenced these results. In most localities in Victoria an estimated 30%+ of low income households suffer from mortgage or rental stress by paying more than 30% of income in housing costs.
To the issue of access to work, the tyranny of the daily commute and access to public transport. As you move into the outer suburbs people have comparatively lower access to public transport and are more likely to be reliant on driving most places and particularly commuting to work. As a result of this equation [more limited access + lower incomes + higher mortgage stress] we might expect those more remote from the CBD to be maintaining more cars per household, possibly older and less efficient cars for driving longer distances. These ‘choices’ are relatively more expensive. This burden of ‘choice’ falls disproportionately on lower income households.
Again, thanks to the Department of Health there is readily accessible dataset to illustrate the state of play.
Figure 5 Car Ownership Rates illustrates car ownership per 1,000 persons, and Figure 6 Vehicle Age and Commuting by Distance illustrates vehicles over 10 years old as well as those with a 2+ hour commute.
Residents at the core of the urban centre [Melbourne, Yarra and Port Phillip] own a higher proportion of cars per 1,000 perhaps reflecting the larger number of single households in these localities. These localities also have a higher proportion of households with no cars. This ‘no car’ household halo effect of appears to be limited to households within 15km of the CBD except for those households more than 200km from the CBD. Possible explanations for the more remote carless households are that these represent older, shared or student households, in which householders are proximate to public transport, likely to poorer and economising.
Surprisingly, the lowest proportion of residents with 2+ hour commuters live in regional Victoria, leading local lives, making local trips to local workplaces. The most self contained Rural City was Greater Shepparton with only 2.4% of residents commuting for more than 2 hours followed by the Campaspe [2.5%], Wellington and Swan Hill [3.2%], Northern Grampians [3.3%], Moira [3.4%] and South Gippsland [3.9%] Shires.
Perhaps more predictably the highest percentage with 2+ plus hour commutes live in Wyndham [26.2% of all commutes] followed by Frankston [22%], Melton [21.9%], Macedon Ranges [21%], Banyule [26%], Monash [19.1%], Manningham [18.8%], Moorabool [18.7%] Whittlesea [17.8%] and Yarra Ranges [17.5%]. Six of 10 of these municipalities are located on the outer fringe, in growth areas and with relatively limited public transport.
Comparatively small proportions of residents in the inner city are undertaking 2+ hour commutes; Yarra and Stonnington have 3.4% of all residents each, and only one in 20 residents living in Port Phillip commutes for this long each day. However more than 10% people from as close to the CBD as Bayside [12 km] and as far away as Murrindindi [129 km] commute for more than 2 hours daily.
More than 50% of those living within 45km of the CBD [as far as Casey to the South East, Melton to the West and Hume to the North] have relatively better access to accessible public transport than others living further out. Figure 7 Proximity to Public Transport illustrates the percentage of those within 400m of a tram or bus stop and within 800m of a train station1. The data indicates better than the exponential trendline access for suburbs within 40km and significantly lower than predicted access for suburbs outside this ring. The exceptions are regional city centres.
Long time/distance commuting has health consequences on community vitality, local activity centre viability, and obesity rates to name a few. Just yesterday I was reading the relative impact on health of a longer commute and found this fabulous poster courtesy of Melanie Penola. The consequences of the long commute on individual and community health should not be underestimated.
To Other Impacts
We all know that that energy prices have been rising. As prices rise, those with lower incomes experience the relative impact more severely, and have less capacity to reallocate income from other budget areas. This impacts those on fixed incomes most severely [pensioners, unemployed, disabled and retirees], and those on lower incomes as well. Practically what happens, given limited resources people choose between necessities and prioritise. They pay for heating or for food, or budget for a little heating and a little food, or forego heating altogether for the sake of eating better. Social workers have been witnessing these choices happening in outer and regional areas in Victoria, counting the increasing number of blankets on couches in people’s homes. Professor Marmot refers to this as fuel poverty.
It is likely this is the lived experience of lower income households on Melbourne’s fringe this winter and the welfare agencies that support them. The Salvation Army document common poverty issues related in the recently released National Economic and Social Impact Survey May 2013
But it is not only the choice between heating and eating we should worry about.
As Elizabeth Kneebone and Alan Berube say… ‘The changing map of American poverty matters because place matters. It starts with the metropolitan areas, the regional economies that cut across city and suburban lines and drive the national economy. Place intersects with core policy issues central to the long-term health and stability of metropolitan areas and to the economic success of individuals and families—things like housing, transportation, economic and work- force development, and the provision of education, health, and other basic services.
Where people live influences the kinds of educational and economic opportunities and the range of public services available to them, as well as what barriers to accessing those opportunities may exist. The country’s deep history of localism means that, within the same metropolitan area, a resident of one community will not necessarily have the same access to good jobs and quality schools, or even basic health and safety services, as a person in another community, whether across the region or right next door.’ Elizabeth Kneebone and Alan Berube, Confronting Urban Poverty in America, Chapter 1, p4 2013.
Despite being recognised as one of the worlds most liveable cities, the same can be said of Melbourne.
We recommend this book as a salient read.
1 As this data is calculated on Department of Health GIS modelling and based on the Estimated Resident Population for 2008, it is unlikely to take into account recent additions to the public transport system including for instance the South Morang Station but gives a relatively current indication of the level of public transport route drop off by distance from the CBD. This data is relatively limited in that it only speaks to existing routes and gives not an indication of frequency of service or the number of destinations one can access from the point measured.