In October last year I did what was obviously a very rough cost-benefit analysis of the Puhoi to Wellsford “holiday highway”, and I came up with a cost-benefit ratio of 0.19. Of course my analysis didn’t have all the traffic models that various consultancies that do these things for real use, and I must admit my understanding of NZTA’s economic evaluation manual is fairly limited, but there was a bit of method in my process, as you will understand if you read the post.

So I must say I was a little bit surprised to find out, in January this year, that the cost-benefit ratio for the project had been calculated at 0.8: and that was just the worst case scenario. Add in the ephemeral “wider economic benefits”, plus a lower discount rate and the BCR got as high as 2.0, as can be seen in the table below: I could understand if I had under-estimated the cost-benefit ratio by half, but four times out on the standard analysis and out by a factor of 10 if you reduce the discount rate and add in the wider-economic-benefits. Something seemed a bit odd.

A couple of weeks ago I finally got my hands on a full copy of the December 2009 “Puhoi to Wellsford Four Laning: Business Case Statement” which I have fully scanned in so that anyone can have a look through it (though at a relatively low quality to ensure the file size wasn’t massive). While having a good look through that study explained a reasonable number of things I had missed in my admittedly unofficial first analysis of the project last year, it appears to me as though the benefits of the project have been highly over-estimated by this Business Case Statement, which means that the real “cost effectiveness” of the project probably lies somewhere between 0.8 and 0.2.

Let’s take a look at the Business Case Statement in a bit more detail: in particular the last few pages of it where the cost-effectiveness of the project really gets analysed. There are four main types of conventional “benefit” that the road is anticipated to generate: journey time reliability, travel time savings, vehicle operating cost savings and accident cost savings. Added to these conventional benefits are the “wider economic benefits” that I will explain in more detail later on. All up, the analysis of the project came up with the following totals for the amount of “Net Present Value” benefits: The total amount of benefit is $689 million, including the wider economic benefits. Comparing this to an estimated project cost of around $1.4 billion in 2008 dollars gives us a “return on investment” that’s a lot lower than the 1.1 in the first table of this post (perhaps that’s taken account of by the benefits being net present value, I’d be interested to hear more about that.)

Going through each of the stated benefits, if we look first at the improvements to journey time reliability, the amount generated is relatively low: at $8.4 million for the project. Journey time reliability benefits are calculated separately from time savings benefits, I guess to recognise the gain from knowing more clearly when you’re going to arrive at where you’re going: which is probably particularly useful for freight. While the issues I have with the way time savings benefits have been calculated may also apply to the reliability benefits, in general I think the amount is low enough to not make much of a difference.

Shifting on to time-savings benefits, as indicated in the table above these are a huge part of the economic justification for the project, with two-thirds of the conventional benefits arising from “travel time savings”. Putting aside my general scepticism of time savings benefits, it is interesting to analyse how this figure of $352 million was worked out. This is detailed more in the table below: Wow, it’s difficult to know where to start it outlining the problems with this. Let’s start at the top: as I note that the Warkworth to Wellsford section has been estimated with an opening date of 2029, not the 2022 that NZTA are talking about. As I am sure is obvious, the longer you leave an upgrade the more necessary it becomes and the more benefit you get from doing it – which means that shifting the completion date of Warkworth-Wellsford forwards to 2022 means lower vehicle numbers, and a lower amount of time-savings benefits.

The next issue I have is the level of assumed traffic growth. What the data actually shows is that traffic around Wellsford has been decreasing (although as a pretty slow rate) over the past few years. Further south, yes there have been increases in traffic flows, but with rising fuel prices I think it’s incredibly optimistic to be suggesting 4% increases in traffic over the next 16 years. Lower levels of traffic growth will mean fewer vehicles benefitting from the upgrade, and once again – lower time savings benefits.

The next issue I have, which is probably the most significant, is the incredibly low average speed of 60 kilometres per hour that was used as the “current situation” to compare the improvements to. While I definitely agree that the current road gets congested at times, and that there are slow parts: particularly between Warkworth and Wellsford – it seems utterly wrong to suggest that the current average speed is as low as 60 kilometres per hour on this road. Wises suggests that your average speed would be about 83 kph between Puhoi and Wellsford: If the current speeds are higher, then the benefits gained from the upgrade would be a lot less, and as a result the time-savings benefits would be dramatically lower.

Moving along to vehicle-operating cost benefits and safety benefits, I don’t have too many problems with the way these have been calculated: except to note that most of the safety benefits could probably be achieved by far far less expensive means. I would also note that on current trends, around 50 people will die on this road before the proposed upgrades are completed.

That brings us to these wider economic benefits. Now from the outset I think it’s important to note that I am not against counting the wider economic benefits of transport projects, as in fact I think that they are probably the most real benefits (along with safety) that upgrading transport infrastructure will bring. It is likely that, for example, the CBD Rail Tunnel project will generate enormous wider economic benefits – and that they would form a critical part of the business case for that project. The issue is about the veracity of the wider economic benefits, or put more simply: will they happen or not?

Turning back to the holiday highway, I am sure that it will generate some wider economic benefits. It should bring Northland a bit economically closer to Auckland, even if mainly in an imagined rather than a real way (I can’t see a 10 minute time-saving making the world of difference) which would have some benefit. But the problem I have with the wider economic benefits for this project (and they make a big difference to its cost-benefit ratio remember) is the “flip-flopping” that studies into them have done.

We have the 2008 SKM study into an upgrade of this road saying the following about wider economic benefits: In short, the wider economic benefits are limited, modest, small in scale and unlikely to make a significant contribution to the viability of the project (although obviously there would be some benefit).

However, the 2009 SKM Study (the business case that this post is about) says this about wider economic benefits:

I must say I find it highly strange that wider economic benefits that were limited, modest, small in scale and unlikely to make a significant contribution to the viability of the project in 2008 can somehow magically transform into being significant, substantial and even a guess that this assessment is “conservative”. What magically happened between 2008 and 2009 to so significantly change the economic impact that this road upgrade would have? (aside from who the Minister of Transport is).

Overall, reading through this Business Case Assessment confirms many of my fears about the poor quality of analysis and assessment that has been undertaken into determining the cost-effectiveness of the holiday highway project. While it may be a bit of a laugh to call this road a “holiday highway” and fun to poke a bit of fun at the Minister of Transport by saying it’s a monument to him – this is actually a deadly serious situation. By pursuing this project not only will people unnecessarily die on this stretch of SH1 as its safety upgrades are delayed, but also in my opinion it is clear that even the relatively low cost-benefit ratio this project has is vastly over-estimating the benefits of the current project. While the Puhoi-Wellsford road will undoubtedly have its benefits, it is extremely clear to me that they are not anywhere near worth the cost, and we should be looking at far cheaper alternatives.

Remember, the $1.5 billion dedicated to this could be the funding so desperately needed for the CBD rail tunnel. I very much look forward to seeing the CBD rail tunnel justified by a proper, high-quality and robust Business Case. Not the rubbish that is being used to justify the holiday highway.

Share this

17 comments

  1. Nice summary here, just further cements my view of these BCRs as a complete and utter waste of time and how they can be more or less manipulated and adjusted to say whatever you want…..

  2. I hate the holiday motorway. It can never be justified. We are spending the thick end of $2b for what? 12,000 motorist and truckies? Come on people. Waterview’s pricetag barely justifies spending that much and that has some 100,000 vehicles passing through by 2026.

  3. I think everyone except Joyce Banks and Key agree with you Marwan. Oh and the trucking and road building industries.

  4. Great post. The thing about using a lower discount rate is a bit specious – if a lower discount rate is used, all potential projects will look better. This just means that the cut-off BCR at which a project gets the go ahead would have to be higher. It also favours high initial capital cost and lower maintenance costs.

    I would be interested to see if concrete motorway pavements had a higher BCR over ashphalt if these lower BCR’s were used (concrete pavements are a classic example of a high capital cost and low maintenance cost investment). If concrete did win out with say a 6% or 4% discount rate – well I don’t see any concrete pavements on the motorway network…

    Also, whenever I do a sensitivity analysis in my job, I will use the current/average value for the variable (in this case a discount rate of 8%) and then look at the sensitivity of changing the variable both upwards and downwards. Why did the report only go one way from the NZTA specified discount rate, and not also look at 10% or 12%? 10% used to be the specified discount rate until recently anyway. Sensitivity analyses are about looking at whether changing a variable dramatically changes the result, not about getting a better answer.

  5. http://www.nzta.govt.nz/resources/research/reports/392/docs/392.pdf

    This report on discount rates carried out for the NZTA may be of interest. I am just looking through it now. Some bits on PT

    New public transport initiatives generally have ongoing operating and maintenance costs that occur over
    future periods that are large relative to the initial investment. While a lower discount rate serves to make
    such initiatives more economic at first, future operating costs are more heavily weighted, which offsets the
    extent to which BCRs increase (whereas this factor works in the favour of maintenance-oriented initiatives).
    Thus the relative priority of public transport initiatives is expected to be lower, all else being equal,
    following a material decrease in the discount rate in the absence of measures to access additional sources
    of funding…

    …A material lowering of the discount rate may lead to investments that do not naturally align themselves
    with achieving the New Zealand Transport Strategy 2008’s targets for 20401, particularly those relating to
    public transport and active modes.

    Also this – talking about major capital works (for roading, eg new motorways):

    Whether relatively high BCRs for such works can be sustained indefinitely is uncertain. If realistic travel
    options to motorists can also be provided when major urban corridors are developed, particularly public
    transport services, then travel demand management initiatives such as congestion charging may be a
    viable alternative to further major capital works. This would reduce the case for sustained capacity
    expansion. Lower BCRs would result if these alternatives were included in CBA methodologies.

    In other words, the authors dont think ongoing development of new motorways is likely to be economically justified in the future – once the motorway network in Auckland is complete for example.

  6. To now justify Roads of National Significance economically it seems as though all the different roads are getting rolled into one, and then justified that as a whole the idea makes sense: http://www.nzta.govt.nz/resources/rons-economic-assessment-2010-05/docs/full-report.pdf

    Never mind that obviously some parts (Vic Park Tunnel) make significant sense while other parts (holiday-highway) make about as much sense as flushing hundreds of millions of dollars down the toilet.

    1. Yes, it looks like that – quite bizarre.

      In the report (section 5) there are a few graphs showing benefits of the various RoNS separately. The holiday highway looks to have benefits substantially less than all the other projects, including Tauranga Eastern Link, and Christchurch motorways.

      NZTA has the cost of Tauranga Eastern Link at $435m, the Christchurch motorways project at $730m, compared with $1.38b for the holiday highway.

  7. What oil price in 2029 is factored in? Higher fuel prices mean less money available for the rest of the economy and higher foreign debt. Thats even before considering the equity argument of building roads for wealthy people to get to work at the expense of other projects.

  8. Maybe I don’t understand this too well, but I thought that the discount rate was related to the opportunity cost of the capital to indicate the net present value.
    I was under the impression that it should be equal to the same as the interest on capital, plus an additional weighting to factor in the relative risk of the project. Considering the costs of borrowing money (or the interest to be gain by having money in the bank) currently hover around 6% or more without the risk weighting, how could they justify anything like a 4% discount rate?

  9. Yeah you’re right Nick, it is linked to interest rates – to answer the “are we better off investing money in this project or chucking it in the bank” question I imagine. Discount rates also reflect that $1000 this year is more valuable than $1000 next year, and in particular, potentially much more valuable than $1000 in 20 years time.

  10. I didn’t even have to read you ripping apart the flimsy sugar-coating of this project to dismiss their case – all it took was to get to read their talk about the discount rates. Anything which starts with a “if only our current rules were changed to fit our project better” STINKS.

    The fact that changing these rules would screw public transport is just icing on their roads cake.

  11. I find it hilarious that this road is scheduled to be completed the year the IEA says oil will peak… A giant waste of money to be completed just in time for it to be empty…

  12. Jeremy, the positive side to it is that I think this is never going to get built. As fuel prices start to become more erratic, and rising over the coming decade, even the current government, if it is voted back in, will have to reassess and will quietly drop a project like this that has become an embarassment and a money hemorrhage.

    Then, we will “only” have wasted a couple hundred mil on design and preparatpry works, rather than 2 billion.

  13. Our Steven is likely to be standing in the Rodney electorate, this is his big vote buyer, so I wish I could share your optimism for reason to break out over this outrageous diversion of public money… Also these guys believe in this with their very souls; progress is getting the SUV to Omaha a little quicker and a little smoother than now to all the people he mixes with. You’ve seen how ‘facts’ like the BCR can change with a little determination.

  14. I dont think Stephen needs to buy votes in Rodney.
    More about the appealing to the middle classes of Auckland, and making the govt look like is busy doing something.

  15. On a less cynical vein, I believe Steven Joyce considers him as doing right by Auckland and New Zealand. I’d call this less a cycnical or electioneering move rather than a combination of a “think big” attitude (which I actually like) and a (sadly outdated) concept of how transport should be prioritised.

Leave a Reply

Your email address will not be published. Required fields are marked *