best regards

Coupla points:

-I like the ‘delay buying the first car, and delay buying the second one’ which means thinking in human life stages and cohorts. Encourage those 16 years old RIGHT NOW to avoid needing a car by specifically improving PT to the universities, entertainment and part-time jobs such as retail. The avoid the second car after the first has been bought by tailoring the system to meet the needs of stay-at-home mums/dads, or looking at arrangements that suit couples that both work, maybe soft measures involving ticketing and so on. Also hard stuff by looking at urban planning around life stages.

A lot of the urban planning ideological stuff is rubbish – reality is people can BOTH want to live in high density or low, maybe at different life stages or within different housing types in the same suburb. For example, you might want to live with your parents in Box Hill/Chatswood, then live in a high rise also in Box Hill/Chatswood to save some money, but then later buy your own house, also near Box Hill/Chatswood. The di/tri/polychotomies of urban planning ideology are too constraining.

-I agree that this question of selection bias is really a question of how demand is constrained. People DO factor PT availability in all but the most hard-nosed auto-loving cases, as a contingency. They don’t make the decision to use PT every day, but maybe once only, on purchase of the property. Those areas with really bad PT have what I would call a chronic problem, in that not only is the primary breadwinner unable to use PT to get to work, but I expect every single journey, including kids going to sport/music practice etc becomes a mums taxi journey. They must get their licence as early as possible and a car the same day.

That said, an excellent analysis Phin. Do we have any comments on people in SY/Toorak/Armadale etc not using PT while it’s in abundance? Or is it? Have you actually tried to get on a train at Hawksburn lately? Is it possible?

http://blogs.theage.com.au/yoursay/archives/2009/04/going_up.html

it got more posts in reply than almost any other topic ever. people really get fired up by integrated land-use planning!

The discussion above excites a few ideas in me:

on selection bias:

If people were absolutely able to maximise their preferences in making decisions about where they live, then selection bias would be a big problem. However, I suspect the housing heuristic ends up being a ’satisficing’ one rather than a maximising one. What that means is that because people have preferences over a range of characteristics, they have to trade off something they value (e.g. proximity to station) for somethign else they value (four bedrooms, room for vegie patch, in the catchment zone for local high school, etc.)

i.e. I bet people who live far from train lines don’t live there cause they hate PT – just because they were trying to meet some other needs.

On the fixed costs of car ownership:

A few years ago I read an article about some rockstar buying a Prius. He was quoted as sayign that it wouldn’t stop him from still driving his Hummer sometimes, but he wanted to do his bit. The journalist swooned. I nearly tore my hair out.

The environmental costs of making a car are, I have read (somewhere??) , greater than the costs of its operation until about the ten-year mark. (steel, aluminium, plastic all very energy intensive). Incorporating the costs in disposal probably makes the trade-off point even later.

Perhaps people are quite good at understanding the environemtnal costs of flows, but crap at stocks? for example, people love to turf their old fridge for one with a higher-energy rating, or throw out perfectly good lightbulbs and replace them with fluoro ones. I doubt these are environemtnally sensible actions.

So one implication of this ill-referenced rant is that there is potentially an environemtnal case for a high fixed cost tax on car ownership.

A seond implication is the dynamic relationship between the relative costs and beenfits of car ownership and PT patronage. This is similar to my favourite theory about PT design, which i believe i have expounded on here before.

1. Probably the single biggest determinant of PT patronage is car-ownership.
2. Once a first car is purchased, the marginal cost of making all sorts of incidental trips on PT becomes relatively high. total PT use falls.

There’s my preferred way of forestalling the first car purchase – by improving services which meet users needs:
3. People are liable to buy a first car to satisfy travel needs that are low-frequency but highly valuable, e.g visting their girlfriend’s house cross-town at strange hours of the night, going down the coast on the weekend, getting to basketball training / volunteering.
4. Systems that are designed to meet these low-frequency, high-value needs, can add as much value to the punter as those that are designed to meet needs that are more predictable, and have more substitutes.
5. Building a system that is designed to forestall car ownership may mean some individual services run at a loss (assuming these high value trips can’t be charged for separately), but total patronage on the system is potentially higher.

Or we can forestall the car purchase decision by having a high rego fee. In practice, they would work well as a combo.

3. On the variable costs of car use.

Clearly, the private costs of driving at certain times are way too low. The image of traffic embodies the concept of negative externalities. Increasing the petrol price is a pretty blunt tool to solve this, as it hurts counter-peak flows, business, farmers, etc.

it is probable that the private costs of driving are too low at all times. Not only is there insufficient tax, but there is a whole lot of subtle, implicit subsidy. One thing that really gets my goat is the compulsory provision of parking a) on streets b) in major developments. this pushes the cost of car ownership off car users, and onto people who use shops and offices. hardly seems fair.

I was also playing around with some car rego and fuel cost figures in excel on the train this morning. I’m hoping to find some ratio of fixed to marginal costs and see where that takes me. If I find anything meaningful or interesting I’ll try to do a post on it in the next few days

cheers
Phin

you’re only talking about hte negative externalities of the operation of hte car. Most cars (given we import abuot 70% of cars to Aus) have very high costs in production and disposal that noone ever thinks about.

I have often wondered if someone did a life cycle analysis on cars if it would be better to run cars to absolute death (like we do in Australia) or to use the Japanese/Singapore model of increasing ergistration costs teh older the car adn therefore making a car older than about 8 years too expensive to bother keeping (which is why so many 18y/o Australian males drive imported used Nissan Skylines)

anyway. we can agree to disagree on the point as we are basically coming at it from different angles. you are suggesting that for people who will always have a car it is better to have up fixed costs cheap and use costs high to discourage car use. and I’m saying that for people who might choose not to have a car it is better to have fixed costs high and use costs (high also preferably).

perhaps someone needs to do market research to work out which one would get a better result for more people. i.e. are their more ‘weekenders’ or more who’d consider no car at all.

You summed up my point in one sentence

It should also be remembered that restricting parking at the destination (rather than origin) reduces use without removing the opportunity to own car.

Agree that the increase in PT use in Melbourne over the last few years supports the idea that people will respond to incentives. In the case of Melbourne’s increased PT use, I’d argue it’s exogenous to PT quality, but rather a reflection of increased car costs (and a few other things) – certainly shows that car use is far from price inelastic. The spatial data above, as well as evidence from Smartbus, shows that endogenous factors make a big difference too.

I have to admit that I’ve always liked old cars in much the same way as I like old trains (and planes and ships) – they’re a fun hobby to have. Peter, if I got my hands on a 4 car garage I’d fill it with old broken Renaults within 2 weeks (maybe like a car version of Dorrigo) – unless my girlfriend killed me first! But I’d absolutely hate to have to drive to work every day and wouldn’t choose a house not close to decent public transport. I’d suggest that none of us here are likely want to live away from public transport – perhaps that suggests that selection bias is apparent. Although I’d venture that contributors to this blog may not be a representative sample of transport users.

Regarding the question of whether PT is better encouraged by making the fixed or marginal cost of cars higher, I have admit I’m biased. Like Drwaddles, I own a car which I don’t use during the week but sometimes use on weekends. If rego cost me $5000 a year I might start to question whether it’s worth it, but you have to ask where the harm is from my car ownership. Someone who owns a car and drives very little is not doing very much harm at all – the negative externalities from cars are roughly proportional to use. Consequently I’d consider it much more sensible to place the tax at the point of use – that is on fuel. Although I admit I have a big economic incentive to say that