Monday, March 16, 2009

Auckland rail business case fisked

So let's look at the truth behind ARTA's rail business case, what it does to cook up a positive case for pouring hundreds of millions of your money into a transport system that will always need subsidies. For starters I already said that evaluation "exiting rail" as an option was dismissed because it would cost a lot of money ($1 billion when rail is actually $1.4 billion) and would be inconsistent with the ARC and central government's strategies (strategies shouldn't be that specific anyway).

So how about the evaluation of the business case?

Discount rate

"the Rail Development Plan compares favourably as an investment if all the benefits are
considered more broadly and over a longer period reflecting the life of the assets and the ongoing generation of benefits"

Favourably? For starters, does this "investment" include the operating cost subsidies growing? However, more importantly, by taking a different discount rate for this project compared to others, you are no longer comparing like for like. ALL other projects need to be re-evaluated, which may STILL mean, it's a bad "investment" compared to building roads. More importantly, is it better than letting people spend their own money?

Value of time

"There is a strong case for using the same value of time for car users and PT users, particularly for rail users in peak times" Is there? Make it. Oh you don't. The reason PT users have a lower value of time is related to their average incomes, so earning capacity. You can play around with this of course, but it really only should apply to PT users who switched from cars. Oh you don't really talk about that do you?

Benefit-Cost ratio: Core Network upgrade

Here the business case shows it up for how lousy it is. We get ten benefit/cost ratios. Five based on NZTA evaluation procedures, five on ARTA's own evaluation procedure. Three under NZTA's procedures have costs higher than benefits. None of ARTA's do. ARTA's base upgrade one is a BCR of 1.5, not spectacular (and with no confidence range around it. Is this the top end or the mid range (it wont be bottom)). Then we have three of the scenarios are fictional "what if fuel went in price a lot", "increase value of time" and some new economic snakeoil called "Other economic benefits relate to CBD (agglomeration) benefits (increased productivity per employee) – residential intensification – wider effects of car use – environmental benefits of new rolling stock"

Now hold right here. Increased productivity per employee because of agglomeration in the CBD? That's a pretty bold assumption. It implies rental prices increase too, of course. It implies Auckland's CBD is more competitive (than where?). $0.5 billion over 40 years? Who believes this?

Residential intensification? Why is THAT a benefit? A benefit that people live together with less space?

Wider effects of car use? What about the BENEFITS of car use? No, that's right. You'll just count some unrecognised "costs".

Oh and environmental benefits of new rolling stock are already part of the evaluation, but let's count them again hey?

Other factors

Each train trip costs in subsidy today around $7. That is what each rail commuter should be paying extra, but doesn't. That is estimated to drop to $5 by 2016 after spending over a billion on upgrading the system - for 15 million trips a year. Yep you can figure that one out yourself.

"Abandoning the rail passenger system in Auckland would require construction of new busways running parallel to the existing rail lines" In many cases it wouldn't as you could rip up lines except the main trunk to the Port. That gives you busways from Britomart to West Auckland and south to Southdown, beyond that buses can operate on local streets or join motorways and use the hard shoulder as a lane.

Conclusion

ARTA/ARC have dressed up the rail business case to suit the answer they wanted, on grounds that the government's own funding agency would question. It will continue to cost taxpayers $5 per trip when electrified, it will generate very modest benefits, and most of those who benefit will be those who get their trip subsidised. It will make diddly squat difference to those using the road network, at best it might increase property values for those living nearby a station and work nearby one on the same line, or businesses who may have a catchment from those able to use the train.

At best, it needs independently appraised - not by anyone in Auckland local government - to determine if the appraisal itself is robust, the levels of confidence and optimism bias around costs and benefits, and whether a thorough appraisal of alternatives has been included.

Sadly, National has been taken for a ride, and you're being asked to pay.

2 comments:

john-ston said...

"At best, it needs independently appraised - not by anyone in Auckland local government - to determine if the appraisal itself is robust, the levels of confidence and optimism bias around costs and benefits, and whether a thorough appraisal of alternatives has been included."

I would suspect that, if anything, the figures provided are far too pessimistic. If we look at the experience of Perth and Brisbane post electrification, we have seen that rail patronage on the legacy routes have increased by in excess of 50% after electrification - what is known as the "sparks effect". We have also seen massive hikes in overall public transport patronage, particularly in Perth which has followed an electrification scheme. Brisbane had a bumpy ride, but since the late 1980s, overall public transport patronage has also increased

Also, you need to remember that ripping up the rail lines and replacing them with bus routes will, if anything, cause a decrease in overall public transport patronage. When Perth's Fremantle Line was closed in 1979 and replaced with buses, patronage dropped by 30%. When it was re-opened four years later, patronage went back to the level that it was.

I must also ask that while you have commented on minimal impact on congestion on the New World systems, that is only part of the picture. Has peak hour shrunk at all in those centres? Has the rail system absorbed what would have been increases in congestion brought about by natural factors?

Tony said...

Interesting, where do you get a copy of the Auckland Rail Business Case ?

My work in Wellington shows the same story. While the original Wellington Passenger Rail Business Case has never been released in full, I did get hold of a copy and it too has a benefit cost calculation that is shonky to the core. For example, it assumed rail was replaced by buses that would run on roads without any investment in busways. Its model also assumed the buses would run on roads parallel to the rail line on the same schedule !. That's right, at 7:04 9 x buses will leave Upper Hutt; at 7:18 another 9 buses will leave ... The model also included a commuter preference for rail by reducing the rail travel times by 10% ! There were many other suspect calculations (including assuming a bus fuel consumption as 3.4 litre/kilometre). The Business Case could therefore state that 25% of commuters would switch to cars. Rail was never compared to a real bus service. Even so, the Benefit Cost Ratio was only 1.48.

Now the updated Wellington Regional Rail Plan is being discussed (here as Item 5 Attachment 2). Two key points:

Firstly, the already Committed Base Case 25-year Cost is $2,590M (i.e. 2 1/2 times Transmission Gully) (Page 41)

Secondly, the GWRC wants still more trains (called "RS1") and a further investment of $440M has come from nowhere to be 3rd priority on the Wellington RLTS. This will only get another 1,000 commuters but the benefits of doing this somehow outweigh building either the Mt Vic or Terrace road tunnels that failed to show benefits (Details on RS1 in pages 42 - 48). I have the background spreadsheets and can confirm the RS1 benefit ratio of 1.5 is just as shonky as the original business case !

Everyone here (including you) keep debating the merits of funding Transmission Gully while those in power are pulling off a rail investment three times larger. And we wonder why this country is going down the toilet :(