Are the suburbs dormitories?

"Centre of gravity" of jobs, 1981 and 2006

There are many misconceptions about the suburbs. A common one is that they are dormitories for workers who commute to the CBD. Another is that jobs in the suburbs are mostly low skill and low pay.

The reality is most economic activity in our capital cities takes place in the suburbs. In Melbourne, for example, 72% of jobs are more than 5 km from the CBD, 50% are more than 13 km away and 25% more than 22 km away.

Jobs have been moving away from the centre for a long time. The “centre of gravity” of jobs in Melbourne is now 7.9 km south east of the CBD, in the vicinity of Tooronga station, East Malvern. That’s up from 5.9 km in 1981. The “average”  job is 15.6 km from the CBD (12.4 km in 1981).

This decentralised pattern holds for most industry sectors. More than 70% of jobs in the Community sector and more than 80% of jobs in the Retail, Wholesale and Manufacturing sectors are in the suburbs (defined as more than 5km from the CBD). Even in the Commercial Services sector, which is the inner city’s great strength, 49% of metropolitan jobs are in the suburbs.

Over 90% of Melburnites live in the suburbs and the great bulk work there too. Less than 10% of workers who live in outer suburbs like Casey, Cardinia, Dandenong, Knox, Maroondah, Mornington work in the centre (City of Melbourne). Even in older suburbs like Hobsons Bay, Brimbank, Maribyrnong and Moonee Valley, less than 25% of the workforce works in the centre.

The view that the suburbs only have low-skill jobs is inaccurate. In fact, the majority (58%) of jobs in Melbourne occupied by a graduate are located in the suburbs. Around 80% of all graduate jobs in the largest sectors are located in the suburbs. The exception is the Commercial Services sector, where the suburbs account for a still-respectable 33%.

Most suburban jobs are spread at relatively low densities. As I’ve pointed out before, even on a relatively generous definition of what constitutes concentration, only 20% of suburban jobs are located in activity centres.

Those jobs are in 31 centres, most of which are small and all of which are very low density compared to the CBD. Many cover a substantially larger area than the 400 metre walk distances implied in Melbourne 2030.

Nine of the centres account for half of the jobs in suburban centres. All centres except one have a pronounced specialisation in at least one of sixteen industries. Nearly all also have a clear specialisation in graduate jobs in at least one sector.

Thus the suburbs are not a mere dormitory but a major job arena. This can easily go unnoticed because of the low density of activities compared to the inner city and particularly the CBD.

The Community Sector – which includes jobs in the Health, Education, Culture and Government industries – is a key reason why the suburbs have so many high human capital jobs. It also partly explains why jobs tend to be dispersed away from activity centres – schools and many smaller health facilities, for example, tend to be located closer to population.

The thing about Melbourne’s suburbs is that apart from Manufacturing (which has declined dramatically in importance), most jobs are in consumer-oriented sectors. They are there to serve the suburban population which, over time, has out-sourced more and more activities from the home.

Unlike many US cities and even Sydney, Melbourne does not seem to have attracted non-consumer jobs away from the CBD to the suburbs in large numbers. Where this is occurring it appears to be mainly in the inner city and some parts of the inner eastern suburbs.

There’re some interesting issues implied by this analysis.

First, what sort of sustainable transport policy should be pursued when 90% of the population and 72% of jobs are at low densities in the suburbs? Four fifths of suburban jobs are spread out and even those that are in centres are at low density. There is a major legacy issue here.

Second, how successful will the Government’s policy of creating six large suburban “CBD-type” Central Activities Districts be when suburban firms largely eschew density?

Third, does the ability of the CBD to expand outwards into areas like Docklands limit the pressure on firms to locate in suburban centres?

4 Comments on “Are the suburbs dormitories?”

  1. Moss says:

    Good stuff Alan.
    However what about university students. For all intents and purposes they “commute” to Melbourne University, RMIT, VUT and the numerous “training colleges” in the city. I imagine this would have some impact on your numbers?

  2. Peter Parker says:

    >First, what sort of sustainable transport policy >should be pursued when 90% of the population and 72% >of jobs are at low densities in the suburbs?

    This can be broken down into creating an urban environment conducive to the more sustainable forms of transport, and, in the case of public transport structuring networks (primarily bus) to optimise service level, connectivity and legibility.

    On the first, low density is less of an impediment than the way we use culs-de-sac and loop streets that increase walking and cycling distances or allow roundabouts that make roads uncrossable on foot.

    Buildings need to address the street to lessen walking distances, and mistakes such as setting facilities like Monash University or Chadstone Shopping Centre so back from their respective roads need to be avoided. Redevelopment or expansion proposals that move facilities towards the street (rather than behind fences or behind footpathless driveways) should be approved.

    There is significant scope for cycling for much local commuting. While pedestrian subways are out of fashion due to crime concerns (and overpasses are a 1960s freeway relic), suitable bicycle underpasses at major roads as part of an off-road bicycle network (more space to do this in these areas than closer in) could be practical and faster than driving (if no red lights).

    The Monash precinct is an example of poor walkabiltiy and where facilities are further from transport than they need be. Where a facility (such as Monash) requires an off-road interchange due to it not addressing the street, the cost in extra bus operating time and delays to through passengers is immense.

    On the bright side, if the industrial area is spread along a major road on the way to somewhere else (escpecially between two major rail-based centres), there is probably already a direct bus route, so there are no additional costs of running a service. An example are the light industrial and business park areas along Springvale Rd, Blackburn Rd, Stud Rd etc which now even boast SmartBus routes.

    A grid street layout where roads are at approximately 1km intervals (as in much of Melbourne’s north and east) should be fine.

    Making cars (average speed 50km/h) deviate by 5 or 10km km to reach a major road would be unthinkable by the average traffic engineer. However current road layouts, too few pedestrian signals, building setbacks and locating bus stops away from intersections all impose similar time penalties on pedestrians (average speed 5km/h). Higher densities is not necessarily a bad thing in light industrial areas or office parks, but the need for finer grained design is more urgent. My own view is that one of Melbourne 2030’s failings is that it’s basically a ‘coffee shop’ policy that ignored walkability and access in industrial areas and bulky goods strips (that its practitioners pretend don’t exist) despite their commercial and employment significance.

    Given that most of what we have now will still be there in 20-30 years, retrofitting existing commercial and industrial areas will be as big a priority as good design of new areas.

    An environment a bit less hostile to transit should help walking and cycling a bit, but isn’t sufficient in itself for public transport.

    Our bus network is basically a mix of what 1950s housewives used with increased outer suburban coverage (often indirect low-productivity routes, sometimes limited by street layout) and some orbital SmartBuses grafted on.

    As stated by the Minister a couple of weeks back, the No 1 design criterion for buses is that most people live within 400 metres of a route. Matters of frequency, effectiveness, connectivity or network legibility are lower concerns.

    Hence we have a legacy route structure that does quite well for access to shopping but quite poorly for access to major employment centres, particularly if ‘blue collar’ or light industrial. This is especially in the western suburbs (eg Laverton North and Tullamarine), which unlike in the east are not roads runninbg between big centres like Ringwood and Dandenong.

    Additional services are sometimes overlaid onto old routes without the latter being reviewed for necessity. Paul Mees recently mentioned the absense of service to Toyota, whereas other areas were much more intensively served. Mees may not agree, but I would consider the 7 buses per hour off-peak between Altona and Altona Gate (on three bus routes) an excessive service that is asking for pruning and better use elsewhere.

    Bus nework efficiency gains from removing duplications like these aren’t going to provide a super-frequent service to everyone. Nevertheless they should release sufficient resources to allow more widespread harmonisation of bus and train frequencies and extended transit coverage in high employment areas that buses had previously missed. Such connectivity is quite crucial since few residents will be within an industrial area’s bus route unless they can transfer from another bus or train.

    To conclude, ‘sustainable transport’ in suburban employment areas hasn’t been on many people’s radar so it’s appreciated that you’ve brought it up. I attribute this to the ‘coffee shop’ mindset of Melbourne 2030 and a reluctance to take service planning, especially for buses, seriously and hence miss out on developing a network suitable for today’s travel needs.

  3. […] Are the suburbs dormitories? There are many misconceptions about the suburbs. A common one is that they are dormitories for workers who commute to the CBD. Another is that jobs in the suburbs are mostly low skill and low pay. […]

  4. […] great majority of jobs are also now in the suburbs. In Melbourne, 72% of jobs are located more than five kilometres from the CBD and the median job is […]

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s