I wrote most of this post back in November, when the latest CBD-specific employment data came out. We didn’t publish it at the time, since we felt we had given the topic enough coverage already, but it seems appropriate to revisit it now that the Ministry of Transport has released some new documents on progress towards the government’s targets for starting the City Rail Link early.

In a way, this can be thought of as an unofficial peer review of one of the MoT documents – a report by Richard Paling Consulting (abbreviated as RPC below). At least, it’s my own attempt to tackle the same questions as that report did, and come up with specific, measurable targets from what was actually a fairly vague government statement. Here goes, with the embedded quotes being parts I wrote in November:

Following on from our earlier posts here and here, let’s take a closer look at National’s employment target for starting to build the City Rail Link early (i.e. around 2015, not 2020). When John Key announced that the government would consider making an early start, he gave two conditions. Auckland doesn’t have to meet these conditions immediately, but simply show that it is on track to meeting them.

The employment target is for “Auckland city centre employment [to increase] by 25 per cent over current levels – that is half the increase predicted in the Future Access Study”

As far as we’re aware, National hasn’t actually explained how the employment condition works. There doesn’t seem to be a formal policy relating to it: just John Key’s initial announcement. There are several questions which need to be answered (and should have been already) if National sticks with its current position:

  1. What data source is used?
  2. How is the “city centre” defined?
  3. What base year is used?
  4. What final year is used?

What data source is used?

The census is probably the most accurate data source on employment at the local level. The employment results from the 2006 census are out, but the 2013 ones aren’t. The next census is still years away – a year for it hasn’t even been finalised – and the data would come out too late to make a difference to the CRL start date. For a 2016 census, we’d be looking at the data coming out in 2018. Overall, I don’t think the census will be much help in showing whether the city centre is on track to reach National’s target.

The other data source is business demographics. These aren’t as accurate, but they are released each year. This is the data source that National will probably need to base their condition on, and it’s what our analysis has focussed on so far.

RPC reached the same conclusion – that business demographics data are the most suitable for measuring progress against the target.

How is the “city centre” defined?

The definition of the city centre is something we’ve looked at several times on the blog. The “CBD” is usually defined as the area bounded by the motorways. However, when you’re dealing with data, the most common definition is based on Statistics New Zealand boundaries, and those are slightly different. It cuts out the southern side of Karangahape Rd, including Ironbank, and some of Auckland’s more colourful nightlife spots. It cuts out the eastern part of Symonds St south of Wellesley – plenty of university buildings there.

If you limit the CBD definition to the area inside the motorways, you cut out everything west of the Northern Motorway, including Beaumont Quarter,  Freemans Bay, the new GHD building, Victoria Park New World, and so on. You cut out everything east of Stanley St, including the Carlaw Park precinct, the creative quarter on the southern side of The Strand, and so on.

As far as I can tell, the City Centre Future Access Study which John Key refers to uses the traditional CBD definition, whether it’s based on the Statistics New Zealand boundaries or something slightly different.

In fact, Nick and Matt have argued that the “central city” covers a larger area than just the traditional CBD. Given that central areas like Newmarket, Parnell, Grafton and Newton will benefit significantly from the CRL, you can make a strong argument for their inclusion.

In short, it’s hard to find a perfect definition for the city centre – but the choice of definition can be pretty important. This is especially true given that the Freemans Bay area unit (immediately west of the CBD and including all those developments west of the Northern Motorway) saw major employment growth in the last year.

RPC included parts of Grafton and Newton in their definition, in order to ensure that all the area inside the motorways was covered – as shown in their map below:

CBD definition by RPC

This is a valid definition, one of several that might have been chosen. Again, there’s the difficulty of trying to choose a specific meaning from a vague target. However, I think it’s more of an issue that the target itself reinforces a misleading view – that the CRL is mainly about the CBD. It’s not; the project has many benefits, including dramatically increasing the capacity of the rail network throughout Auckland, reducing congestion for vehicle drivers, and removing an impediment to further employment growth in the CBD. It is a regionally (indeed, nationally) significant piece of infrastructure.

What base year is used?

As for what base year is used, your guess is as good as mine. The City Centre Future Access Study uses a 2006 base year, and the most logical approach National could take would be to do the same. Otherwise, they might try to argue for a 2012 base year, although there isn’t really a clear rationale for doing so, except that they announced their ‘support’ in 2013, and the most recent data available then was 2012.

If  2006 is the base year, then we’ve already come a way since then – using business demographics data, employment in the CBD (based on Statistics New Zealand boundaries) has grown from 80,390 in February 2006 to 91,130 in February 2013. This is 13.3% growth in seven years, much faster than the Auckland or New Zealand average in that time, and despite a very flat economy.

RPC adopts a 2012 base year, although they don’t seem to give a reason for this choice. Their interpretation and mine are both valid, although I think mine is more consistent with the reference to the CCFAS in National’s targets. You could take the view that the 25% growth starts from the time of National’s announcement, or that 2012 data – and indeed, the announcement does say growth from “current levels” – but removing six years from the modelling period isn’t really fair. Following that logic, if National had made the announcement in 2017 rather than 2013, the CBD would only have a couple of years to grow by 25% – which clearly wouldn’t be possible.

What final year is used?

Just as important as the choice of base year is the choice of final year. Does employment need to grow by 25% over 10 years, or over 20? Because 14.0% of jobs in Auckland, or 4.7% of jobs in New Zealand, are in the CBD (using the tighter Statistics New Zealand boundaries). The number of jobs which need to be created is significant at the regional and the national level, and they won’t just come from nowhere, overnight.

The City Centre Future Access Study assumes that CBD employment grows by 54% over 35 years, from 2006 to 2041, or annual growth of about 1.24%. National should be giving us around 17 years, i.e. from 2006 to 2023 or so – anything less would be pretty unfair, especially with the recession in the early years of that period, and with the CRL not yet built. That’s a target Auckland could meet easily, and we’re well on track. I doubt National has this kind of timeframe in mind, though, or it wouldn’t be an aspirational target.

National do seem to be talking about a 2020 end date for the 25% growth, and RPC also use this as an end year. I’ve since heard that the CCFAS, using some slightly sketchy modelling, assumed very strong employment growth to 2020 when the CRL was assumed to open, of the order of 25%, and that this was part of the rationale for National choosing the target. With that said, I still stand by what I wrote in November!

Summing up

National’s targets are pretty arbitrary: they’re poorly defined, and after all, the CRL is about much more than just increasing CBD employment.

Realistically, the Council needs to stick with its current work programme, which is to make all the required preparations to be able to start building the CRL in 2015 or soon after. If National manage to stay in power after the next election, and stick to their guns over their two conditions (and choose their definitions to make them as hard to reach as possible), they could manage to delay the project.

I’d rather point to things such as the higher residential population in the CBD, or the network capacity which the CRL unlocks, or widespread support from both Aucklanders and the business community. There is still a big job to do in advocating for an earlier start to this project.

Of course, Len is now also pointing to several major projects which will be majorly benefited by an early start to the CRL.

Share this


  1. I posted this a while ago on twitter and our facebook page. If you look only at the blue area (as they did) then employment was up 2.1%. If instead you also included the red areas which would also be strongly positively impacted by the CRL then the increase was 4%

    Further even just sticking to the area they’ve used, if you compare the number of jobs in 2013 to what they were when they peaked in 2008 pre GFC then it’s up 8.2%

    1. Well clearly Parnell and Newton need to be include in this silly game as that is where two of the four new stations will be.

  2. Putting aside that trains seem to require targets to acquire funding that roads magically don’t, city centre employment is not an accurate statistic to use because it’s not just about journeys to the CBD. What does that L in CRL stand for again?

    For instance, if I’m a student living at GI and studying at Unitec (Mt Albert or Waitakere) campus, the CRL is going to reduce a train journey by 10-15 mins making me far more likely to take the train. CBD employment increase = zero, but many point to point journeys on the train network become far more attractive.

    Also K Road and Newton both become more attractive places to live with a different vibe to downtown, but now minutes away by train. Those existing CBD workers may move to dwellings near the new stations (Parnell too I guess). CBD employment increase = zero, but other modes of transport (bus, car) decrease -> less need for more roads.

  3. I agree that they should use an expanded definition of the CBD to include the areas that will directly benefit from the CRL but I think it is fair for the data start point for the base year to be 2012. The commitment was for growth based on current levels. That means 2012.
    Frankly I believe the project should be started earlier simply because it will only get more expensive to build. The dollar may weaken making the cost of imported components more expensive and labour costs are likely to increase as well. The same cost benefit argument can be applied also to the harbour tunnel project. Its going to be built – why not build it today.

    1. Agreed regarding Crl there are big ticket items we need to build we need to decide which are the highest priority and start Asap also considering future costs.

      I also agree that the same is true of the Harbour Crossing but you know that a road crossing would be well down my list (after everything in the cfn and more)

  4. My experience is that it is common to have funding or project start targets when you are planning for the future. Examples are signalising an intersection when land plan change areas are developed or waiting for traffic increases before building a new route. In the past few years we havent needed them as most road projects are built only after existing flows have increased to a point where major congestion already occurs daily. Examples are the SH18 motorway from SH1 to the Upper Harbour Bridge and the Hobsonville motorway through to Westgate. That method of waiting for a major problem was due to the 10% discount rate we used for years and even the 8% it dropped to which essentially said future issues can be ignored only current problems get funded. If CRL is intended to facilitate a major shift in land use policy then it is probably sensible to link it to evidence of strengthened CBD employment as opposed to the major residential focus of the last 20 years. But don’t imagine for a moment I am trying to defend Rons. They should be subject to the same sorts of assessment and funding controls.

  5. 25% is a lot of growth. Shouldn’t the system be in place to accommodate that growth? Or will government just continue to move the goalposts – 25% now, but 20% in 2017, maybe a little more in 2019, etc. The sooner the system is built the sooner those jobs will materialize. Not being critical of government, but setting targets like that can be an effective stalling tactic.

  6. Great to see Labour’s promise today that they would fund the CRL immediately if elected and rationalise the RoNS to fund it. They intend making it a point of difference in the election campaign. Have to say I think they’ve probably earned my vote!

  7. I’ve just been doing some work on PT usage for the Wellington CBD. What came out of that, was the need to distinguish the half-mile round the railway station (where PT use is very high), against that elsewhere in the CBD, where it is quite a bit lower. In Auckland’s case: all I would say is that you need to report the public transport use separately for the CBD (that in blue – it has about 80,000 jobs?) and then the central city (that in red); my hypothesis is that PT use in the latter would be quite a bit lower, because of fewer convenient bus routes and more parking.

Leave a Reply

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