Has spare infrastructure capacity in the inner city disappeared?

Guess what this architectural gem in Stockholm is? Click to find out.

The received wisdom is it costs much less to provide infrastructure for an inner suburban dwelling than for one in the outer suburbs. However, as I noted last time, we don’t know how big the difference is or even, for that matter, if it’s positive or negative – we simply lack reliable evidence.

There are reasons, however, to suspect the savings in infrastructure outlays associated with urban consolidation might be much less than is widely thought. It’s plausible that the popular claim of an $85,000 per dwelling saving could be well off the mark (note I’m only talking in this post about the capital cost of infrastructure, not the economic costs and benefits of a fringe vs central location).

From the time urban consolidation was first seriously put on the table in Australia as a policy option, a key premise was the availability of ‘spare’ infrastructure capacity in the inner city. This part of the city had previously supported larger working class and migrant populations, so there was ‘free’ infrastructure to be had in support of a restoration of earlier population levels.

There’s not much sense in assuming any capacity is free (it all has to be paid for) but looking from the perspective of 2011, there are reasons to question if there actually is any spare physical capacity left, at least in relation to some types of infrastructure.

A key reason is a lot of whatever spare infrastructure capacity existed has already been used up by gentrification. At the 2006 Census, there were 36,488 more residents in the inner city of Melbourne than there were in 1976 (and 76,422 more than when the inner city was at its lowest ebb in 1991). In fact of the 31 municipalities in metropolitan Melbourne, only the City of Moreland and the adjacent City of Darebin had significantly fewer residents in 2006 than in 1976 – Moreland had 14,585 fewer and Darebin 17,137 fewer. That is not a lot in the context of projections Melbourne will grow from a current population of four million to seven million by circa 2049.

Even where there are fewer residents today than in the past, they might still have a much larger “infrastructure footprint” than their predecessors. Modern households have many more resource-intensive devices like flat panel TVs, air conditioners, heaters, computers, spas, and so on, than their predecessors. They have more cars than former residents, so there’s less room for parking. They also have higher standards – the primary school that used to accommodate 300 kids in six or seven classrooms now has to build twelve to handle the same enrolment.

Moreover, households today are smaller on average, so they have fewer ‘economies of scale’ in resource consumption than earlier generations. Two households of three persons each use more gas for heating than they would if the same six residents shared a single dwelling. Gentrifying households are also wealthier on average than the sorts of households who used to live in the inner city and inner suburbs 30 to 40 years ago. On a per capita basis, wealthier households consume more of just about everything worth having. Again, that will require more infrastructure capacity.

Thus it’s possible infrastructure in some locations could be at or above capacity even with a much lower population than those places housed in the 1970s. Read the rest of this entry »


Is living at density the same everywhere?

A reader, Ian Woodcock, took me to task yesterday about my post on whether outer suburban families would willingly choose densities of 25-30 dwellings/Ha so they could walk to local shops and services.

In particular, Ian reckons I invoked a straw man when I argued that “it can’t be assumed that merely increasing outer suburban housing density will create an environment with the ‘buzz’ and ambience of a St Kilda or Fitzroy”. He says the relevant comparison:

is not with inner-suburbs like St. Kilda that have the highest density in Australia (!), but with the streetcar (tram) suburbs of the inner-middle ring, like Kew, Camberwell, Malvern, Armadale, etc. where densities are higher than the outer suburbs in many places, public transport is good, there is housing diversity and access to a wide range of shops and services, and moreover, social capital is high.

Ian’s broadened the discussion beyond walkability to the merits of density more generally. I’m not sure there are substantial parts of suburbs like Camberwell with average densities as high as 25-30 dwellings/Ha, but nevertheless I’m happy to make the comparison with them. I’ve set out my argument about the attractiveness of higher densities to Growth Area residents below:

My contention is that living in a townhouse or apartment in the inner or middle suburbs – whether it be St Kilda, Kew, Fitzroy or Camberwell – delivers an entirely different set of benefits than it would in the outer suburbs. In essence, it’s worth choosing to live in a town house in the former but not at present in the latter.

Camberwell residents who can afford it generally choose to live in bigger dwellings, usually a detached house with a yard. The rest live in smaller dwellings like newish apartments and townhouses because that’s the only way they can afford to live there. They want to live in Camberwell for all the usual reasons – it’s close to the city centre, close to large centres like Glenferrie Rd and close to private schools. They also like it for the status it confers and because it’s ‘people like us’ i.e. wealthy or aspiring to be wealthy.  As Ian says, it also has good public transport and a wide range of services. But most of those who live in apartments and town houses do so because they have to – it’s primarily about location, not dwelling type.

Some will argue there are people who could easily afford a big dwelling but actively prefer a small one. Undoubtedly there are some, but there aren’t many and they’re not usually families (who’re the main household type in the Growth Areas). The fact that empty nesters tend to hang on to their large houses signals that, if they can afford it, people put a high value on space. People who can afford it don’t buy studios or one bedroom apartments in Docklands or Southbank – they buy three bedroom units and penthouses. All those renovated and extended terraces and houses in the inner city don’t say “small is better”. Location and dwelling size are what economists call superior goods. As incomes rise, consumers tend to buy more of them – they might move to a better suburb or a bigger dwelling (although only the very rich can usually do both).

Now compare Camberwell to the Growth Areas. In the latter a detached house with a garden costs much the same as a townhouse. Households don’t need to forego space in order to live there, so why would they choose to buy a townhouse? Read the rest of this entry »


Are lots smaller in the outer suburbs than in the inner suburbs?

The Financial Review (paywalled) reported on Saturday that the average house sold within 10 km of the CBD in Melbourne is located on a 511 m2 block, according to data collected by property information provider, RP Data.

Central Green, Highlands, Craigieburn - some lots are 213 sq m (click to enlarge)

Of course there is considerable variation in property types within that average. Houses in the inner city (0-5 km) are predominantly terraces on small lots whereas those in the inner suburbs (5-10 km) are predominantly detached houses on larger lots.

However what is especially interesting about this statistic is how it compares with the outer suburbs. The median size of lots sold in the Growth Areas is currently 513 m2, according to property consultants, Oliver Hume.

In other words, the average lot size is much the same in the two areas.

This is surprising, but a number of caveats are in order. First, some of the properties closer to the CBD would be destined for redevelopment whereas that seems unlikely in the Growth Areas. Read the rest of this entry »