We have been pretty critical of many of the business cases for the roads of nation significance that have been produced because like as has been seen in Australia with their PPPs, many use incredibly over inflated traffic projections and even with those some still don’t make economic sense. Yet at the same time there are some fundamental issues with how we perform economic analysis that can skew the results, especially for big projects like the RoNS or the CRL. This was excellently highlighted last week by the councils chief economist.

Traditionally, projects are measured using cost-benefit analysis, – infrastructure benefits are given monetary values and compared with costs.

Unfortunately, many of the prescribed techniques used in a cost-benefit assessment understate the full benefits of infrastructure by some margin, particularly for large-scale transformative projects. It is somewhat puzzling that many of Auckland’s major transport priorities have been assessed as having very low benefit-to-cost ratios. The second harbour crossing has a ratio of 0.6 (meaning a return of 60 cents for every dollar invested); the Puhoi to Wellsford highway is not much better than one; and the City Rail Link gets a ratio of either 0.4 or 1.1 (depending on who you are talking to). On this basis alone, none of these proposed projects ought to inspire a great deal of public confidence.

There are two main reasons why these (and other) projects have been assessed as having such poor economic return. The first is that standardised government guidelines for appraisal miss many of the economic benefits. Critical infrastructure underpins economic growth by providing the foundations required for innovation to occur and jobs to be created.

For really big projects, infrastructure can fundamentally change the drivers of economic growth by creating changes in land use, enticing population inflows, and generally increasing the amount of economic activity (think of what the Auckland harbour bridge has done for population growth on the North Shore). Yet none of these benefits are considered in the economic appraisal of modern transport projects.

The second reason why long-lasting infrastructure projects attract such poor benefit-to-cost ratios is the way we value future generations. New Zealand Treasury policy necessitates the exclusion of all project benefits that accrue after 30 years – an astonishing requirement, not least because of the lifespans of the projects themselves. For instance, Auckland’s harbour bridge was built over 50 years ago, our storm and waste-water networks have been around for 100 years, the southern and western rail lines for well over 100 years and the port longer still.

All these remain among our most important pieces of infrastructure. We are extremely fortunate that previous generations did not use modern day cost benefit guidelines to decide on these projects – otherwise we might be aspiring to catch up with countries far poorer than Australia.

In fact, New Zealand policymakers have settled on some of the most short-sighted appraisal methods in the developed world. By comparison, Britain measures infrastructure over 60 years, while Australia uses a 50-year period. The difference is significant in terms of valuing infrastructure. If we applied the British framework to the City Rail Link, for example, the project benefits would be around six times higher. The resulting benefit-to-cost ratio would likely stoke a great deal of public confidence. Exactly the same is true of a number of other key infrastructure projects. The hurdle for long-lasting infrastructure in New Zealand is far too high, which erodes public confidence. This means that short-lived, piecemeal projects are usually given priority.

The result is an under-investment in quality infrastructure, which puts further strain on our global competitiveness.

What isn’t mentioned is the discount rate that is used which effectively discounts both the costs and benefits that occur in the future into present value terms so we can compare it better. Well it seems that the NZTA is now actively investigating making some changes to how it accesses projects which will address both the discount ratio and and the assessment term. A presentation released to me from an OIA request suggests that the NZTA are looking at changing the discount rate from 8% to 4% while also changing the assessment period from 30 years to 60 years.

As noted these changes would have potentially massive implications on the business cases of projects like the RoNS however the NZTA board seem to have taken a very cautious approach to any change with the minutes from the meeting stating

The Board expressed its in-principle support to a staged introduction of lower (than 8%) investment evaluation discount rates, and to extending project evaluation timeframes. The Board preferred that such changes to Investment strategy and procedures will not apply to this NLTP i.e. will apply to the 2015/18 NLTP and beyond. The Chair requested that any proposed new discount rate be externally peer reviewed, and that the logic behind any such proposed rate be clearly established and then set out.

It would be easy to be cynical to say that these changes were primarily aimed at improving the business cases of the RoNS but perhaps a project that benefits more than them will be the CRL. The original business case suggests that even with just looking at the conventional benefits we normally assess the BCR changes from 1.1 to 2.4. Interestingly the council also looked at how the project would fare under the UK system which uses a 3.5% discount rate and a 60 year term and the change is simply massive. As Geoff noted in his piece, the benefits received are around 6 times higher at ~$11 billion over the assessment period giving a BCR of about 5.2

Hopefully before any changes are made we will see the NZTA line up each project with this new methodology so we can get a better comparison but it would certainly be useful for projects like the CRL.

Share this


  1. I was taught that the cost benefit approach NZTA used was originally born out of the idea that it was used to compare small roading projects with each other which is why there are things like travel time savings, vehicle running costs factored into the benfits.
    I think in the past ten years its just been politically driven and used for a purpose that it was never intended for.
    NZTA needs to rethink there approach like you suggest in the article quite drastically before publishing such irrelevant numbers.

    1. I agree, the system got turned from a tool used to compare similar small projects to one use to justify large projects because a number like a BCR is a convenient political tool.

  2. I agree with Tim, it is good to see changes that should help us make longer term decisions around infrastructure investment but the whole basis of how we evaluate different projects needs to be reviewed to be sure we have as comprehensive an understanding of the likely benefits as well as he costs and externalities as possible.

  3. I think the only use of the BCR is comparing like with like, once start playing around with discount rate and return period it just becomes a joke.
    As Jarbury regularly blogged many of the benefits of roading projects are illusory and poorly defined anyway, so I don’t think the raw numbers mean much at all.
    And I think we are being a bit naive if you think this is going to make any difference to the governments and NZTA’s view of the CBDRL.
    We should concentrating on finding better ways of measuring costs and benefits of PT and large motorway projects. I hoping this will reduce benefits of motorways by shredding the time saving benefits, and on the other hand better capture the efficiency gains of Public Transport.

  4. Clearly reducing discount rate doesn’t magically create more money to fund all these projects that “now make sense”, so we’re likely to end up seeing a higher cut off point for projects to have to surpass in order to receive funding. Perhaps nothing under 2 will now be seriously considered?

    In general I like the changes as I think they would recognise the long-term impacts of transport investment far more and hopefully shift us away from the “low CAPEX, high OPEX” mindset that we’re often in.

  5. Does anyone know where 8% comes from?
    Maybe reasonable twenty-odd years ago, but it seems extraordinarily high when even first home buyers are able to borrow money at 5.__ per cent. (Admittedly long term rates are higher)…..
    Am I missing something?

    1. I think that is part of the problem. The rate was set some time ago by treasury to be conservative and hasn’t been an issue but now that it is being used to assess big long term projects the issues with the rate are being highlighted much more.

  6. I have just finished reading quite a bit of the documentation on the NZTA website about the RoNS and have come to the conclusion that even if Labour can get into power at the next general election, 90% of the roads will have been commited to construction, with the exception maybe of Puhoi to Wellsford, so that is $10b that Labour will not have for any other projects. These will, politically be, unstoppable.

Leave a Reply