Ever since reading through the cost-benefit analysis of the Waterview Connection (the old full tunnel version, we don’t even know what the cost-benefit analysis of the newer version is) I have become somewhat skeptical of the ways in which we analyse the costs and benefits of various transportation projects. To refresh our memories, here is the basic CBA of the Waterview Connection (the full tunnel option):
Now as I said above, this analysis is just for the full tunnel option – so the costs will be different to what has been chosen by the government as the preferred alignment. The construction costs will be a lot less, and I imagine the social and environmental costs will be higher. But that’s not really the issue here.
What I am much more interested in is how the benefits are calculated. How do we come up with $2.62 billion in time savings benefits? How do we work out $607 million of agglomeration benefits? How are the $690 million in congestion cost benefits different to the time-savings benefits? How can building a motorway actually create a CO2 benefit?
I have started digging into a few international transport policy journals to try and find out a bit more about international best-practice for making decisions on what transport projects should be built first, and more specifically how the estimated transport benefits of a particular project are calculated. One method is the New Approach to Appraisal (NATA) that is “a framework used to appraise transport projects and proposals in the United Kingdom”. This framework appears much more complex than the simple cost-benefit approach that is used in New Zealand, and includes a wide variety of effects (both positive and negative) that should be looked at when deciding whether to proceed with a transport project.
If we have a look into a bit more detail of what the NATA is, we can see that complexity:
Within the NATA framework, the impacts of transport projects are categorised in terms of five high level criteria (economy, safety, environment, accessibility and integration), reflecting the Government’s objectives for transport. Each of these criteria is divided into a number of sub-criteria and it is against each of these sub-criteria that the impacts of a proposal are assessed and presented in a 1 page Appraisal Summary Table (AST).The division of the five criteria is shown below:
1) Economy (Public Accounts, Transport Economic Efficiency: Business Users & Transport Providers, Transport Economic Efficiency: Consumers, Reliability, Wider Economic Impacts)
2) Safety (Accidents, Security)
3) Environment (Noise, Local Air Quality, Greenhouse Gases, Landscape, Townscape, Heritage of Historic Resources, Biodiversity, Water Environment, Physical Fitness, Journey Ambience)
4) Accessibility (Option values, Severance, Access to the Transport System)
5) Integration (Transport Interchange, Land-Use Policy, Other Government Policies)
Now I really am no expert on the details of how things are done in New Zealand (hence my numerous questions above) but it certainly seems as though we look into the first three criteria, but kind of get lost when it comes to accessibility and integration. Perhaps this is because our system is so focused on the road system, rather than public transport projects, so these issues aren’t seen as significant, but I am guessing that leaving out effects on accessibility and integration potentially plays a major role in how the analysis of transport projects in New Zealand seems to end up favouring expensive roading projects.
Further interesting analysis is available in a journal article in “Transport Policy” by A. Cundric, T. Kern and V. Rajkovic (can’t link for copyright reasons unfortunately) entitled: ‘A qualitative model for road investment appraisal.’ This article critiques the simplicity of the quantitative models that are currently used to analyse transport projects and their cost-effectiveness, and is particularly critical of the simple cost benefit analysis system that we see in New Zealand:
In the paper, we focus on road investments and firstly present an international overview of road appraisal methodologies which have been recently moving away from the traditional cost benefit analysis (CBA) estimating direct effects of road investment. These methodologies are thus evolving towards CBA that aims at monetising also some indirect effects and towards multi-criteria analysis (MCA) that aims at evaluating road projects by both quantitative and qualitative criteria.
After trying to make sense out of the “journal-speak”, I think what is importantly mentioned here is how the general trend of assessing the cost-effectiveness of a transport project (the appraisal process) is moving away from the cost-benefit system and its focus on direct effects.
I certainly have a lot more reading to do with regards to these issues, but it certainly seems to me that something’s broken in the way we look at transport projects in New Zealand – where it is somehow justifiable to spend the vast bulk of our transport funds on roading projects that are unsustainable in the medium to longer term due to oil price increases, and are unnecessary as demand for them is falling. It seems like the way we analyse the value for money from transport spending is as outdated as the thinking behind our general transport policies.
Processing...
I do not believe that CBA is the best way to assess anything which has environmental, social and cultural effects. How can a dollar value be placed on a human life (i.e. a doctor)? Or how can a river be restored back after years of increased pollution? Sometimes the costs far out weight the benefits, but most of these benefits are made around time saving (I guess if they build that motorway people will get home in time to sit on the couch and see the end of ‘Home and Away’) at the cost of local community environments.
If a CBA properly assessed a road development, we would not have many roads, however this is not the case, as CBA allows us to make that data say what we want it to say. This often favors road projects and their outcomes. In saying that, CBA has a place to play in the economics of a project, but other models such as multi criteria analysis need to be used in helping decide on how fix transport problems as well as community involvement.
There’s a great saying that if time savings benefits were so important to the economy, the government would buy everyone a dish-washer.
Air quality: $0.
Oh great. I can has third-world priorities?
That was the full tunnel option. I imagine that air quality would now be a reasonable cost given that parts of the route will be ‘at-grade’.
NATA has a major problem, it treats reductions in tax revenue to the Treasury as a “cost” and increases in tax revenue as a “benefit” when they are transfers.
Brent C: How can a dollar value be placed on a human life? Easy, everyone does it. You wont spend all the money you have to save many lives through charitable donations and the like. Stopping the last fatality on the road network would cost billions, which could save many many more lives if spent elsewhere (or left in the hands of the people taxed to pay for it). THAT is what this is about. It is how do you best spend money to achieve safety benefits. The environment is the same, if it is private property it should be about compensation, if not then what do people value the river for. It isn’t “priceless” because nobody would spend everything they have to save it, so it has to be valued against alternatives. So environment generally has to be evaluation of the cost of pollution on humans.
Time savings do matter, because people change behaviour to save time. People generally do not use public transport over driving because of time/convenience – that is by far the biggest factor. It is why cheap air fares killed off half of the long distance passenger rail in NZ in recent years.
CBA is a useful tool for two reasons:
– It allows different options to be compared for overall net value, so that the most economically efficient option can be chosen. Using similar criteria allows that appraisal to be made reasonably robustly and rationally;
– It can allow a floor – a threshold below which ANY option isn’t worth it. If the benefits don’t significantly exceed costs then it shouldn’t proceed.
However, it always is a proxy for preferences, for what those paying would prefer to pay for. It is about spending taxes for the greatest benefit, and whether what they are spent on are more valuable than what people would otherwise spend them on if they got the taxes back. For example, I doubt the average taxpayer thinks that subsidising high income public servants to go by rail from Khandallah to Wellington every day is a good use of their taxes
The problem with time savings benefits is that they tend to disappear over time. The classic example is that for all the transport investment in the world the time spent commuting is fairly similar now as to what it was at most points in the past. Whilst a lot of transport investment has been “running to stand still”, in that if it had not been made things would have got worse, projects are generally not “sold” on that basis: they actually promise to make commuting faster.
What actually happens is that people travel further. It’s still a benefit (improved accessibility) but it’s not really a time savings benefit.