Wednesday, May 20, 2020

No light at the end of this tunnel - reflecting on failed infrastructure

The big risk and high cost of thinking big.
I flagged a concern in the last post about the fiscal and productivity impact of projects that don’t stack up economically. The risk is that the post-Covid recovery leads to indiscriminate infrastructure spending which would compound the already severe fiscal impact of essential deficit spending on public health, household incomes, and business support .

Given their dismal track record here and overseas, it is likely that large infrastructure and especially transport projects will dig the fiscal hole deeper without delivering the benefits that might help the country climb out of it. We know that the Think Big energy projects of the early 1980s precipitated a foreign exchange crisis.  Had they been economically sound the ten hard years of economic restructuring that followed may have been moderated. 

Looking back to go forward
There is no sense in trying to replicate the past.  But it does make sense to learn from it. 

In this case, it seems the lesson was not learned. I posted several critiques of Auckland’s Central Rail project back in 2011 and 2012. Its worth revisiting that project issue to see once again the risk that a new found enthusiasm for Thinking Big Projects entails.


Auckland’s Central Rail Link, 25c in the dollar?
How ever long it takes to finish and however much it gets used, Auckland's Central Rail Link looks to be an economic disaster.

The first cost estimate for the CRL tunnel was $2.3bn, released by Auckland Council in 2011. That was seen as unworthy of government support by Transport Minister Brownlee, with “a decidedly weak benefit:cost ratio of just 44 cents in the dollar”. 


And that doesn’t count the prior commitment of $500m to the electrification of Auckland’s rail in 2011, seen as a necessary step towards an underground link, or the expenditure on station development necessary to take advantage of increased capacity.


In any case, the estimated tunnel benefit:cost ratio of 0.44 turns out to have been on the high side.  That the project was under-specified is evident in the 2018 announcement that platforms were to be lengthened, adding around $250m to the costs. 


And it was under-costed.  By April 2019 the cost estimate was up to $4.4bn.  This covered construction cost increases of $327m, “non-direct costs” of $130m, and a new provision for escalation and contingencies of $310m.


Converting the original budget and additional costs to December 2019 dollars (using the price index for construction inputs) reveals an over-run of around $1.7bn, a 70% increase on the original budget in real terms.  Given that there has been no suggestion that the projected benefits will increase, the potential economic return now sits at around 25c in the dollar.


We can expect further cost escalation given that completion date (prior to Covid19) was pushed out from 2021 to 2024. This is likely to be extended further by pandemic-related constraints including disruption to contractors, labour, and supply chains, and increased competition from local and international “shovel-ready projects”. That's more bad news for those central city businesses that have seen revenues plummet in the face of ongoing disruption by the works, losses that do not appear to figure in the costs.


And there is no sign of increasing benefits to offset all these additional costs. 


Strong growth rates are misleading

Let's consider potential benefits in light of the past ten years' public transport performance.  The introduction of electric units in 2014 and station and service improvements across the network saw strong relative growth in rail patronage. It seems the benefits of improved service levels on the network are already being reaped without the $4.4+bn CRL.

However, this needs to be kept put in perspective. While rail patronage almost tripled over the ten years to February 2020, the real gains were in bus use (70% of the total[1]):




Significantly, 87% of gains in bus patronage were in “frequent, connector, local, targeted” services according to Auckland Transport.  This strengthens the argument for flexible bus services rather than high cost, fixed route rail. 


It is also likely that gains to rail included a transfer of some passengers from buses so that the impact on car use and the increase in public transport use will be less than indicated simply by the gross increase in PT trips. 


How important is rail to central city commuting?

According to the 2018 Census, a relatively low 55% of work trips by the 159,000 people working in the Waitemata Local Board Area (which contains most inner and outer central city employment) were by private or company vehicle. Of those, 6% of were made by passengers.  Company vehicles accounted for 11% of the total. As these vehicles are most likely required for work purposes their occupants are unlikely to transfer to PT. 

This means that the market for improved rail and bus services is just 46% of possible commuter trips .  Public transport already has a high penetration rate of 29% of commuters working in Waitemata.  However, less than a third of these were by rail, despite the relative growth in numbers. The prospects of getting many of the remaining private car users to shift to rail are low. Rail patronage may have to grow mainly through trips transferring from buses.


Narrowing the focus, there were 18,000 commuters to the inner city in 2018. Only 19% relied on a private or company vehicle (between 3,100 and 3,200 vehicles) in 2018. The likelihood of getting a significant reduction in this number is slim. 


A surprisingly high 50% said they walked to work, while 22% used public transport (only a fifth of those by rail).  The strategy of getting more inner city workers living there seems to be working. Ironically, it’s a success that raises questions over expectations that investment in the CRL will influence local travel in the inner city. 


Will CRL even deliver a significant mode shift? 

The Council wants people out of cars.  Whether or not that's achievable - or even reasonable  - was the CRL the way to achieve it?


Apart from the fact that the project is uneconomic and fiscally damaging, the fact is that over three quarters of Auckland’s labour force works outside Waitemata Local Board area, with 77% of them relying on private or company vehicles to get to work.  Even if the billions invested into the CRL were to effect a significant lift in public transport patronage, it is a spend that could have been much more effectively directed towards offering  more flexible bus-based transit serving the wider urban area.

And that was before Covid19.  
Today, the lack of flexibility of rail comes into even sharper focus in light of the potential changes in working practices, the diminished appeal of high density living, commuting, and working, possible land use changes, and the imposition of social distancing for the foreseeable future. These prospects, along with post-Covid19 delays in constriction, mean that the CRL is likely to fall even further short of helping to achieve “Government’s plans for higher economic productivity and the Auckland Plan vision of being the world’s most liveable city” (City Rail Link, Business Case 2015).

Spending $4.4bn (and increasing) on lifting the capacity of rail patronage by building the  CRL tunnel looks not only like an economic and and fiscal fail. It is also looking like a policy fail.


Which brings us to the even bigger white elephant in the room, Auckland's proposed MetroRail. This is the subject of my next post.



[1]    The growth of trip numbers will have been somewhat less given the adoption of electronic charging in 2011 and integration of charging across modes in the Hop Card in 2014 will have increased multi-modal trips.

No comments: