Is it healthy to assume correlation means causation?

Association of obesity trend in the US with vehicles miles travelled (left) and with Justin Wolfer's age (right)

The link between the physical environment and health outcomes like obesity is fraught. The Victorian Legislative Council’s Environment and Planning References Committee should bear this in mind as it goes about its new inquiry into the contribution of environmental design to public health.

The Committee might want to start with the first chart in the accompanying exhibit, which comes from a recent issue of The Economist and purports to show that obesity has increased in the US in line with the increase in miles driven over the last 15 years. The chart is based on work done by researchers at the University of Illinois who found “a striking correlation between these two variables – but with a large time lag……This near-perfect correlation (99.6%) permits predictions about obesity rates”.

Expatriate Australian economist and Wharton business school Assoc Professor, Justin Wolfers, points out the folly of this claim. It is, in his words, a “nonsensical correlation”:

When you see a variable that follows a simple trend, almost any other trending variable will fit it: miles driven, my age, the Canadian population, total deaths, food prices, cumulative rainfall, whatever.

To demonstrate his point, Professor Wolfers prepared the second chart showing an even better correlation between changes in obesity over the period and changes in his age – he didn’t even need to resort to a time lag to get such a good fit! He acknowledges The Economist offered the customary caveat that correlation does not equal causation but this chart, he says, is so completely unconvincing as to warrant a different warning: “Not persuasive enough that you should bother reading this article” (in the interests of balance, here’s The Economist’s subsequent response to Professor Wolfer’s charge).

This exchange highlights a problem with much of the research that purports to show the physical environment — particularly density and/or public transport access —  has a strong effect on health-related variables like obesity. There’s plenty of evidence of correlation but not much evidence of causation. There’s no doubt obesity is inversely related to both density and access to public transport, but if it turns out these aren’t the underlying drivers of obesity then the economic cost of misdirected policies could potentially be significant.

There are special reasons why it’s hard to establish causation when dealing with real life infrastructure projects and transport/land use programs. These British epidemiologists reviewed 77 international studies examining the effectiveness of policy interventions to reduce car use. They concluded the evidence base is weak, finding only 12 were methodologically strong – and they mainly involved relatively small-scale initiatives like providing better information about travel options or direct financial incentives to reduce driving (incidentally, only half of those 12 actually worked i.e. reduced car use). Read the rest of this entry »


Is exempting petrol from the carbon tax such a big problem?

Is transport the main game? Source: Productivity Commission: Carbon emission policies in key economies

Given Australia already has a large excise tax on petrol, exempting automotive fuel bought by “families, tradies and small businesses” from the Gillard Government’s carbon tax is not the disaster some would have us believe.

Australia has a minority government so compromise was inevitable – two of the independents, Tony Windsor and Rob Oakeshott, wouldn’t be party to any increase in the price of fuel for their country constituents. It was either put a price on most but not all sources of greenhouse gas, or have the whole idea shot down yet again.

The exemption is expected to apply to petrol, diesel and LPG. Were the tax to apply to petrol, the impact would be modest – a $25/tonne tax is generally estimated to increase the price at the pump by around $0.06 per litre. The CSIRO calculates that even a $40/tonne tax would only raise the price of petrol by about ten cents per litre.

These amounts are much less than motorists already pay via the $0.38 per litre excise tax on petrol and diesel (there’s no excise on LPG). While it might have a “sin tax” dimension in relation to cigarettes and alcohol, in the case of automotive fuel the excise is not aimed at making motorists pay for roads or for the external costs of petrol – it’s just a convenient way of raising revenue (although it’s not as good as it used to be since John Howard abolished automatic indexation of the price in 2001).

Nevertheless the excise tax is a serious deterrent to driving. The Productivity Commission’s recent report, Carbon emission policies in key economies, calculates that “in 2009-10, fuel taxes reduced emissions from road transport by 8 to 23 percent in Australia at an average cost of $57-$59 per tonne of CO2-e”. Although not put in place with the purpose of abating emissions, the excise already has a much more significant effect on driving than any level of carbon price that’s been seriously touted in the political debate. Based on the CSIRO’s estimates, it could be argued its effect is equivalent to a carbon tax of over $100 per tonne (the relationship isn’t linear – there’re diminishing returns from a marginal increase as the fuel tax gets bigger).

Thus there’s a good argument that automotive fuel is one of the few areas where consumers already pay a high level of tax over and above the GST. Indeed, if it were so minded, the Government could’ve imposed the new carbon tax on petrol and diesel and simply provided an equal offsetting reduction in the level of the existing fuel excise tax. There wouldn’t be a lot of political or economic sense in that, but it illustrates the principle. Read the rest of this entry »


Are these curves moving for the same reasons?

Annual per capita passenger kilometres by mode, Melbourne (data from BITRE)

Back in May I compared the historic level of passenger travel by car in Australia since 1970 against rail and bus, showing the significant flattening in car use from circa 2004-05 and the upturn in travel by public transport. This sort of long term perspective is useful for understanding the relative importance of the changes in each mode — something which isn’t as evident if only the last five or six years is examined.

The accompanying exhibit shows the change in passenger travel by mode just within Melbourne, using data from the Bureau of Infrastructure, Transport and Regional Economics (BITRE). Importantly, it also allows for the increase in population and hence shows the change in per capita passenger travel. The period is the 33 years between 1976/77 and 2008/09.

It can be seen that private – or individual – travel (i.e. car, van, motor cycle) has fallen sharply since 2004/05, by 1,236 km. Conversely, public – or shared – transport travel (i.e. train, tram, bus) increased by 301 km. While the curves are still a long way apart, it’s notable that the gap is closing primarily because Melburnians are driving less.

I haven’t seen anything which shows confidently and unambiguously where the fall-off in driving is happening. For example, is it outer or inner urban driving? Is it certain trip purposes only? Is it fewer trips? Is it shorter trips? Is it confined to particular demographics? Or is it something else entirely? As with most things, the outcome we see most likely results from the interplay of a number of factors, rather than from a single dominant force.

The usual suspects called on to explain these trends include increases in the price of petrol, in traffic congestion, in parking costs, and in the level and quality of public transport. Other explanations include the theory that baby boomers are getting older (and hence driving less) and the conjecture that the long distance drive is increasingly being replaced by cheap air fares (although this relates more to non-urban travel).

Then there’s Gen Y’s declining interest in driving, the impact of new communication technologies and growing interest in health & fitness and environmental issues. There’s also the theory we’ve hit saturation level with driving – we can drive to enough opportunities already, we don’t need more. Perhaps another reason is the increase in women’s workforce participation leaves them with less time and need for driving. Read the rest of this entry »


Are SUVs killers?

Railway timetables - Monty Python does Agatha Christie

My hate-hate attitude towards SUVs hasn’t improved after reading a new US research paper, The pounds that kill, by two University of California (Berkeley) researchers. They show being in a vehicle struck by a 1,000 pound heavier one results in a 47% increase in the probability of a fatality in the smaller vehicle (the paper might be gated for some – here’s an ungated version that looks very similar).

The authors note that from 1975 to 1980, the average weight of US cars dropped from 4,060 pounds to 3,228 pounds in response to higher petrol prices. However average vehicle weight began to rise as petrol prices fell in the late 1980s and by 2005 it was back to the 1975 level. In 2008 the average car was about 530 pounds heavier than it was in 1988.

A key reason for increasing weight is the safety “arms race”. Drivers seek large vehicles because they’re thought to be much safer for occupants than smaller ones, however they are more hazardous for other travellers in smaller vehicles. As the average size of the fleet rises, there’s an incentive for all drivers concerned about safety to trade-up. The authors note the “safety benefits of vehicle weight are therefore internal, while the safety costs of vehicle weight are external”.

Safety is a key public policy issue because road accidents are as dangerous to life as lung cancer. Moreover, traffic accidents kill more Americans aged under 40 years of age than any other cause. While only as quarter as many people die on roads each year as die from lung cancer, the average age of a road accident victim is 39 years compared to 71 years for lung cancer. The aggregate years of life lost from both causes is similar.

Noting that no detailed attempt has been made to measure the external costs of vehicle weight, the authors sought to:

quantify the external costs of vehicle weight using a large micro data set on police-reported crashes for a set of 8 heterogeneous states…..The data set includes both fatal and nonfatal accidents…..The rich set of vehicle, person, and accident observables in the data set allow us to minimize concerns about omitted variables bias.

They estimate a 1,000 pound increase in striking vehicle weight raises the probability of a fatality in the struck vehicle by 47%. Moreover, they find that light trucks like SUVs, pickups and minivans “raise the probability of a fatality in the struck car – in addition to the effect of their already higher vehicle weight”. The authors suggest this additional effect could be due to the stiffer chassis and higher ride height of light trucks, or possibly to the behaviour of light truck drivers. Read the rest of this entry »


Does business oppose congestion charging?

Projected avoidable costs of traffic congestion (BITRE)

I’m very disappointed with the line one of the State’s largest employer associations, VECCI, is taking on road congestion charging. This issue was raised in a report prepared by consultants Acil Tasman for the Competition Commission’s (VCEC) inquiry into a State-based reform agenda.

Congestion imposes such a high cost on business – whether freight or personal business travel – that I’d expect the great majority of VECCI’s members would be better off with pricing. The Bureau of Infrastructure, Transport and Regional Economics puts the current cost of congestion in Melbourne at around $3 billion per year, rising to $6.1 billion by 2020 with unchanged policies.

VECCI says it is opposes the idea of congestion charging for three reasons. First, motorists already pay both the CBD Congestion Levy and the fuel excise. Second, there’s no spare capacity in the public transport system to take displaced motorists. Third, the economy can’t handle another tax on top of the forthcoming mining tax and the carbon price.

In regard to VECCI’s first objection, the obvious point is that these existing imposts on motorists aren’t working – many Melbourne roads are still congested at peak times. Despite the lofty title, the CBD Congestion Levy is actually a tax on parking. It’s only had a very minor impact on traffic because most drivers don’t pay it personally – their employers do. It isn’t in any event peak-loaded and of course it does nothing to manage the level of through traffic. And as per the name, the levy only applies to the CBD. The $0.38 fuel excise tax (no longer indexed) does reduce the overall level of travel by motorists, but has no appreciable effect on congestion because it doesn’t vary with the level of traffic or time of day.

So far as the “no spare capacity on public transport” objection is concerned, a congestion price only has to remove a relatively small number of vehicles in order to get traffic moving at an acceptable speed. The vast majority of motorists won’t come seeking a seat on the train – they will continue to drive but, rather than pay with time as they do at present under congested conditions, they’ll pay in cash. Their numbers may increase over time, but so will public transport capacity. Also, revenue from congestion charging should be applied to improving public transport and increasing capacity.

VECCI’s argument that the economy can’t handle another tax is about as nakedly political as you can get. Australia is one of the world’s most vibrant economies and has been growing for 20 years. Motorists, both business and private, can afford to pay their way. In any event, a congestion charge isn’t a tax — as the name says, it’s actually a charge. Motorists pay directly for the quantity of road space they use and get a direct and immediate benefit – faster travel. That means motorists who travel more pay more. It also benefits drivers from all income strata by enabling them to reduce delays when they make high value trips. Read the rest of this entry »


Is parking the best use of CBD streets?

Pop-up cafe, Pearl St, Lower Manhattan

Here’s an interesting nugget of information from Melbourne City Council’s new Transport Strategy – there are 4,190 parking spaces in Melbourne’s CBD, of which 3,077 are metered. There are however more than 60,000 off-street parking spaces “in the centre of the city”. That means on-street spaces account for just 6% or so of all city centre parking.

This suggests that in the CBD at least, on-street parking is not that important in the overall scheme of things. It’s currently used solely for short-term parking, but commercial parking stations can also perform that function. Indeed, they’d probably prefer the higher income that comes with rapid turnover. Car storage is a remarkably low value use for such premium land. Even in the case of the metered spaces, the price charged is well below the value the land could theoretically fetch in some alternative use.

According to Greville Pabst, chief executive of valuers WBP Property Group, a “car space in a typical city apartment can add from $40,000 to the purchase price and, in some instances, for upmarket apartments in good locations, it can add more than $100,000 to the price tag”. Even in inner city residential areas, the Mayor of the City of Yarra estimates a parking space adds about $50,000 to the value of an inner-city property. In Sydney’s CBD a garage costs as much as $120,000 to $150,000.

There is an opportunity here to do away with all or most on-street parking in the CBD and instead use the space for something more valuable. It could be used for high capacity vehicular modes like buses, trams and motorcycles; for highly valued sustainable modes like cycling, walking or shared car schemes; or for amenity-enhancing uses that could take advantage of ground level proximity to pedestrian traffic.

Parking spaces could be dedicated permanently to new uses – for example a cafe. Given an unrestricted brief, businesses would come up with innovative ways to use these narrow spaces for other purposes. Manhattan’s “pop-up” restaurants provide an interesting take on possible alternative uses.

Of course Council could simply start charging parking fees that reflect the real value of the land, hopefully with a demand-responsive tariff. Prices would presumably be relatively similar to what commercial parking operators charge – somewhat less because they’re not protected from the weather or supervised, but somewhat more for those that are a bit closer to the action. It would need to be examined closely but my view is the social value of alternative uses would still be higher. Read the rest of this entry »


Should motorcycles have a bigger role?

Scooters and Motorcycles (SAMs) in Hanoi

Public transport, cycling and hybrid cars get a lot of attention as responses to climate change and peak oil, but the potential of scooters and small motorcycles seems to pass largely unnoticed. That’s a pity because powered two wheelers are the mode of choice in places like Hanoi where fuel prices are very high relative to incomes. They offer the key advantages of cars – on-demand response, a direct route to the traveller’s destination and speed – but at much lower cost. Indeed, if fuel prices go stratospheric, I expect a very large number of urban Australians will choose this mode of travel ahead of public transport.

A lot of policy attention is quite properly given to improving the safety of riders, but very little is given to developing scooters and motorcycles (hereinafter SAMs) as part of a comprehensive transport strategy. That’s unfortunate because there would potentially be many benefits for the wider society if they had a much larger share of all travel within our capital cities.

SAMs (and I include power-assisted bicycles in this category) have many attractions for riders. They are remarkably inexpensive to purchase and cost little to run. They can be ‘threaded’ through congested traffic, are cheap to park, and in some places they can be parked legally on footpaths. Most machines can carry a passenger and small items like groceries (or much more in Hanoi!). Most importantly, riders don’t have to wait for them, or transfer to another one mid-journey, or stop frequently, or take a circuitous route to get where they’re going. In most places they’re so easy to park that riders don’t have to walk far to them either.

Compared to the large proportion of car journeys that involve only the driver, SAMs also have many social benefits. They require only a fraction of the road space, consume much less fuel and emit considerably less greenhouse gas. They need little space for parking and don’t require new dedicated infrastructure (although some improvements to roads would help). Most importantly, they are potentially attractive to new classes of travellers because, like bicycles, they’re private and hence flexible.

The share of trips taken by SAMs may increase organically in the future in response to higher petrol prices and increasing traffic congestion. But given their social benefits, it would be good policy to actively encourage greater uptake of SAMs in lieu of driving. Safety is the key obstacle, so that’s where any strategy has to start. Current approaches to safety stress factors like rider skills and greater awareness of SAMs on the part of drivers. These are important, however within large urban areas the best strategy is safety in numbers. Achieving a critical mass of riders could be facilitated by initiatives such as lowering registration charges for small capacity SAMs, relaxing restrictions on lane-splitting, giving riders access to transit lanes, and increasing the supply of dedicated parking spaces (with locking points). There might also be scope for small works – for example, access lanes could be reserved to enable riders to move directly to the head of traffic at intersections.

But there are other important issues that need to be addressed too. Engines in SAMs are not generally as sophisticated in dealing with pollutants on a per kilometre basis as car engines. Nor are they generally as quiet under acceleration. These drawbacks are partly technical – it’s harder, for example, to incorporate catalytic converters in SAMs – but they can also be addressed through better regulations and more pro-active enforcement. In the medium term, electrically powered SAMs may mitigate some of these problems. Read the rest of this entry »