Building our way to affordability?

It’s logical that when housing supply does not meet housing demand, prices will rise. Housing affordability is a huge issue in many cities around the world – with the blame often falling on planning rules and restrictions: both in the form of restrictions on sprawl and restrictions on the level of intensification. While there’s a logical connection between a lack of housing supply and higher housing costs, it is perhaps a little more complicated if we start to take this connection and apply it through saying that if we build a lot more dwellings we will start to make a positive difference to affordability.

For a start, there are two different ways in which we might try to improve affordability by constructing more housing supply: building more houses on the urban edge and building more houses through urban intensification. As many previous posts have pointed out on this blog there are likely to be a number of ‘false economies’ if you attempt to improve affordability by allowing urban expansion. Not only are many of the housing cost savings likely to simply be eaten up in transport cost increases, but there’s an enormous hidden cost in such an approach: all the additional infrastructure that’s required. An interesting Australian research paper suggests that the infrastructure costs of servicing urban expansion rather than urban intensification are huge:   There are really only two ways to pay for the additional infrastructure costs of urban expansion. The first option gets development in peripheral areas to properly ‘pay its way’ – adding huge development contributions to the cost of each dwelling and therefore significantly undermining the ability of this development to actually be affordable. The second option, which seems to be what happens in a lot of American cities that provide ‘affordable’ housing on their peripheries, is to hugely subsidise that development – largely through not requiring it to pay fully for the infrastructure necessary to service it. But then there’s a bit of a logic gap here – why is the rest of the city helping to subsidise those on the periphery who contribute most to congestion, the urbanisation of farmland, probably the greatest amount of CO2 emissions per capita and so on?

The other option is to provide a lot more housing through intensification. This is more logical in a number of ways:

  • You have lower infrastructure costs on a per capita basis and therefore the existing city either doesn’t need to subsidise the new development as much, or the development contributions don’t need to be so high.
  • Most demand seems to be for inner-city housing (that’s where prices are increasing so dramatically), so you provide housing where people actually want to live.
  • You avoid the ‘trade-off’ between housing affordability and transport affordability. More affordable housing in inner areas really will be more affordable for its inhabitants and they won’t see the gain eaten away at the fuel pump.

Harvard economist Edward Glaeser is a big proponent of the concept that you need to build your way out of affordability problems – criticising (for example) Jane Jacobs who wanted to maintain a mix of building ages in an area – even if that came at the cost of allowing additional development. The paragraphs below come from Glaeser’s fascinating article in The Atlantic, which is an excerpt from his book “Triumph of the City”:

But then, during the 1950s and ’60s, both public and private projects ran into growing resistance from grassroots organizers like Jane Jacobs, who were becoming adept at mounting opposition to large-scale development. In 1961, Jacobs published her masterpiece, The Death and Life of Great American Cities, which investigates and celebrates the pedestrian world of mid-20th-century New York. She argued that mixed-use zoning fostered street life, the essence of city living. But Jacobs liked protecting old buildings because of a confused piece of economic reasoning. She thought that preserving older, shorter structures would somehow keep prices affordable for budding entrepreneurs. That’s not how supply and demand works. Protecting an older one-story building instead of replacing it with a 40-story building does not preserve affordability. Indeed, opposing new building is the surest way to make a popular area unaffordable. An increase in the supply of houses, or anything else, almost always drives prices down, while restricting the supply of real estate keeps prices high.

The relationship between housing supply and affordability isn’t just a matter of economic theory. A great deal of evidence links the supply of space with the cost of real estate. Simply put, the places that are expensive don’t build a lot, and the places that build a lot aren’t expensive. Perhaps a new 40-story building won’t itself house any quirky, less profitable firms, but by providing new space, the building will ease pressure on the rest of the city. Price increases in gentrifying older areas will be muted because of new construction. Growth, not height restrictions and a fixed building stock, keeps space affordable and ensures that poorer people and less profitable firms can stay and help a thriving city remain successful and diverse. Height restrictions do increase light, and preservation does protect history, but we shouldn’t pretend that these benefits come without a cost.

This is an interesting debate, because of what Glaeser hints at halfway through his second paragraph: that while the new buildings themselves may not be affordable, they should contribute to an improvement in affordability by increasing general supply. The rationale seems to be that richer people currently living in older houses/apartments will shift to the shiny new houses/apartments, and their older houses will be less valuable and therefore more affordable. I’m not entirely sure whether I follow that logic. It makes sense for office space, as companies able to afford premium space generally lease it and therefore are keen to occasionally “trade up” to the shiny new buildings – leaving their previous space more affordable and now available for a second-tier of companies to shift into. But when it comes to housing, I’m not entirely sure whether building more inner-area apartments and terraced housing is going to make existing housing in that area too much less attractive for prospective buyers. In effect, you’ll have the choice of older lower-density housing or newer higher-density housing (which will probably be constructed to a fairly flash standard). Neither of those sounds particularly affordable to me.

This interesting Glaeser/Jacobs debate was picked up on in a post on the superb City Builder Book Club blog, which is going through Jane Jacobs’s masterpiece, The Death and Life of Great American Cities, chapter by chapter:

One of the most insightful observations that she makes about old buildings is that their capital costs have been written down and therefore the landlord does not need to charge a high rent. New construction is very expensive. It takes 20 or 30 years for a developer to pay off the mortgage. It is only then that there is less pressure on the owner to charge high rents.

This simple observation has recently been questioned. Edward Glaeser (Harvard economist and author of last year’s book, Triumph of the City), for example, has completely misunderstood this chapter. Glaeser asserts that keeping old buildings leads to nothing but high rents — that it’s a simple issue of supply and demand. He tells us that the only way to go is up, up, up, and if towers were built in the place of these older, smaller buildings, districts like Greenwich Village in cities all over the world would become far more affordable. That is, more density equals lower rents.

Can you think of anywhere you’ve seen that happen? It certainly has not been my experience in the past decade of tall building construction in Toronto. Nor has it been the case anywhere in Manhattan that I am aware of. It probably isn’t the case in your city either. I cannot think of an example in an economically healthy city where an old building was torn down and replaced by a new taller/bigger structure and this new structure has cheaper rent than the building it replaced. 

It is certainly quite fascinating to compare these two argument – both of which seem to make logical sense, but at the same time find themselves almost diametrically opposed. What perhaps this highlights to us is that to improve affordability, we need to be a bit more specific than simply saying “expand housing supply”. What really needs to happen is a specific expansion of affordable housing supply. Perhaps the final word on this matter should go to this recent blog post by Cap’n Transit – looking at housing affordability and New York:

Suppose that tomorrow there’s a revolution in New York City. Zoning and rent control are abolished, and every member of the City Planning Commission and the community boards is sent off to the reeducation camps. Spreading out from the Empire State Building, developers cover the New York area in parking-free high-rises until there’s enough housing for everyone, at affordable prices. Sounds great, right?

Almost. But all housing is not created equal. Some apartments are bigger, some have better views, some have are more conveniently located. Some come with relatively superficial amenities like pools and package services. Some are dangerous or bad for your health, from crime, pollution, bad construction or neglect.

Some housing differences are a matter of taste, like neighbors who play loud salsa music or cook Indian food. Some people choose their housing out of racism, moving out if a Black family moves in. Some people want to live near people like them. Some people want to live where there is ethnic diversity, with no single group dominating.

All these factors affect the price of the housing. People who can’t afford higher rents will necessarily have to put up with some undesirable features, like bad views, loud music, crime or a long commute…

…There is a more market-oriented solution, though: build more cheap housing. And cheap housing is bad housing. The next question is: bad in what way?

If the only determinant of an apartment’s price is its distance from job centers, then the poor and the young will all wind up living on the outskirts of town, paying for their poverty with long inconvenient commutes. If the only determinant of price is proximity to a hazardous waste dump, or neglected housing stock, or gang activity, then the poor and the young will wind up in substandard housing, exposed to toxins and victimized by gangs. If the only determinant of price is proximity to “the right people,” then the poor will wind up clustered together, having little contact with other social classes.

To prevent segregating the poor into inconveniently located bad housing with crime and pollution, we need to make some safe, solid housing available closer in, integrated with the rich people’s housing. that is still affordable. In order to do that, we need to allow housing that’s cheap in the non-dangerous, non-segregated ways. That means housing that’s small or ugly, with crappy views and no doormen. Maybe housing that allows loud music if it doesn’t bother anyone else.

Ironically, a good example of this in the Auckland context are the much maligned ‘sausage flats’ built in the 1960s. While they’re pretty much universally disliked from an architectural point of view, they provide quite a lot of relatively affordable housing in places where people actually want to live (Mt Eden, Epsom, Three Kings, Herne Bay etc.) Their aesthetic unattractiveness, in a somewhat bizarre way, has ensured that they remain affordable and means that the supply of relatively affordable housing in inner Auckland is significantly greater than it would otherwise be.

I guess the key point is that just as building more 5 bedroom McMansions on the urban edge won’t make a blind bit of difference to housing affordability, building super-flash inner city apartments, townhouses and terraced houses is also unlikely to help. Clearly, constrained housing supply leads to housing becoming unaffordable, but to resolve that we need to not only build more houses generally, we need to build more affordable houses. How to do that in a way that still allows developers to make a sufficient level of profit for them to bother is perhaps one of the biggest questions facing Auckland in the next few years.

Manukau Timetable is out

The new timetable that includes Manukau is now available on MAXX. If you want to catch a train from there in the morning, arriving at Britomart between roughly 8am and 9am you have the following options:

  • 7:16 via Glen Innes
  • 7:37 via Newmarket
  • 7:57 via Glen Innes
  • 8:18 via Glen Innes

Each will take roughly 40 minutes to get to Britomart. To those that want to be on the very first revenue service to use the station you will want to catch the 7:26am from Manukau to Britomart on Sunday 15th April (I think I will most likely still be in bed) while the first revenue service to arrive at the station will leave Britomart at 7:06 and get there at 7:46. This does raise its own questions as it means train and it’s crew will sit around at the station for 40 minutes before making another run, hardly an efficient use of resources and something I hope gets fixed up quickly.

One thing I am very disappointed about though is that the only real timetable changes are to add in Manukau when last year Auckland Transport promised on a number of occasions that the new timetable would include improvements to the western line which simply haven’t happened.

Fixing Congestion: Carrot or Stick?

Congestion pricing seems likely to become a bit of a hot topic in Auckland over the next few months, as various ways to ‘plug the gap’ between the amount of money required for the transport projects wanted and the amount of funding actually available are explored. An article at the excellent Atlantic Cities blog points out some recent research into the effectiveness of congestion pricing at reducing peak time congestion compared to other measures – most specifically rewarding those who travel during off-peak times.

Congestion pricing is generally considered to have two aims: raising revenue and discouraging car travel either to certain areas or at certain times of the day. For most transport economists, the focus is on the second of these matters, as congestion charging is a pretty inefficient revenue generator – due to its significant administration costs (especially when compared to fuel tax, which is incredibly cheap to administer). The theory goes that roads are congested at peak times because peak time road-space is a scarce resource, and by under-charging for it we are creating too much demand for the relatively limited amount of available supply. In order to use this road-space more efficiently, we should charge higher rates at times (or in locations) where demand is highest. This will encourage people to travel outside peak times or to take public transport, walk or cycle, instead of driving.

The beauty of congestion pricing is that it gets around, to an extent, the problem of induced demand. Historically, we have tried to solve congestion by simply building more and more supply – and then for some reason being surprised when that supply is taken up quickly, leading to our roads being just as congested as before the widening/new building took place. Congestion pricing means that we don’t have to go and spend a huge amount of money on additional capacity, only to see it eaten up close to immediately. Rather, it means that trips which can be taken in another way often area, as many people with options will choose to avoid paying the charge. Of course induced demand doesn’t disappear completely – freer flowing roads encourage further traffic which can mean that congestion charges need to keep on being hiked up to ensure the roads stay free-flowing: the same ‘chicken-and-egg’ spiral you get with road widening, although more helpfully this time you’re making money rather than spending increasingly eye-watering amount on roading projects.

Of course an alternative to charging people for driving at peak time is to instead offer rewards for travelling off-peak. Fundamentally, either option could have the effect of helping to ‘spread the load’ and reduce peak time congestion. In a sense we do this at the moment, through subsidising public transport services for their ‘congestion relief’ benefits: which are particularly applicable at peak times. It’s in the best interests of motorists to have a lot of people on public transport. Therefore it’s sensible for drivers to help keep PT fares lower than they would otherwise be, through contributing to subsidies. The question is, what if we extend this theory to directly encouraging people to travel outside peak times?

That’s what this study in the Netherlands looked at, and here’s its abstract:

In a 13-week field study conducted in The Netherlands, participants were provided with daily rewards – monetary and in-kind, in order to encourage them to avoid driving during the morning rush-hour. Participants could earn a reward (money or credits to keep a Smartphone handset), by driving to work earlier or later, by switching to another mode or by teleworking. The collected data, complemented with pre and post measurement surveys, were analyzed using longitudinal techniques and mixed logistic regression. The results assert that the reward is the main extrinsic motivation for discouraging rush-hour driving. The monetary reward exhibits diminishing sensitivity, whereas the Smartphone has endowment qualities. Although the reward influences the motivation to avoid the rush-hour, the choice how to change behavior is influenced by additional factors including education, scheduling, habitual behavior, attitudes, and travel information availability.

A further academic article looks at making a comparison between the effectiveness of congestion pricing schemes and the kind of reward scheme the Netherlands study trialled. The results of the study are quite interesting – starting with the differences in the changes people made when either a cost or a reward was provided to encourage them to drive outside peak times:  I’m not entirely sure what’s with the bike result, but it seems that the major difference is that the reward scheme generally encouraged people to travel outside peak time or to shift to public transport at a higher rate than road pricing did. Road pricing seemed more successful at shifting people to riding bikes, other modes (presumably largely walking) or staying at home.

But this is only a part of the story. What really matters is the issue about how effective the two schemes are at achieving their primary purpose – which is to reduce peak time car travel. And the results of that are quite interesting: There are some variations in the methodologies for the different studies that went into creating these results, but the distinction seems pretty sharp at an overall level of analysis. It seems that people do respond better to rewards than to ‘punishment’ when it comes to shifting away from travelling at peak times by car.

Of course reward schemes are potentially quite challenging to fairly implement, and of course they cost money rather than generate it – as road pricing does. However, obviously they’re likely to be quite a lot more politically acceptable and if they are more effective at getting the results we want, then they might be a legitimate form of public intervention. Put in a slightly different way, spending money on reward schemes to reduce peak time congestion might be quite a lot cheaper, quicker and more effective than spending money on increasing roading capacity with the same goal.

Car vs Train

This morning my wife and I needed to drive to work so I thought I would do a little experiment and time the trip. Those who regularly read the blog will probably know that I catch the train from the Sturges Rd station our west.

7:14 – We cross the Sturges Road bridge and we were actually driving across the over bridge at the same time as the train we catch passes under it so it turned out to be a pretty fair comparison. If we are driving to the motorway, even off peak, we will avoid Lincoln Rd for as long as possible due to the amount of traffic and lights along there, going via Rathgar Rd it usually takes about 5 minutes to get to the intersection of Universal Dr and Lincoln Rd and this is where the problem starts.

7:42 – Green Marker- After sitting in bumper to bumper traffic we finally get to the intersection of Triangle Rd, Central Park Dr and Lincoln Rd.  Moving on to the interchange it self, as we had two people in the car we are allowed to use the T2 lane which saves us a few minutes waiting at the onramp lights.

One extra thing I will point out is I noticed a cyclist who stood out from some others due to what he was wearing, his bike and they type of helmet he was wearing, he crossed the intersection and was heading down Central Park Dr towards the city just before we crossed the intersection but more on this later.

7:42 – If we had of been catching the train and it was on time then it would have been at about Kingsland so would have travelled much more in the same time frame.

Back on the motorway we had just travelled past the Te Atatu interchange when I noticed the rider mentioned earlier about to cross the bridge over the Whau River however he disappeared again as we had to slow down in traffic. This was pretty good considering he had to negotiate a few back roads and get across Te Atatu Rd.

7:55 – Light Blue Marker – We were travelling along the causeway just past the works going on, traffic is still heavy but is moving.

7:55 – If the train is still running on time it would probably be somewhere between Newmarket and Parnell, this is after already completely passing the CBD and allowing for a 3 minute stop at Newmarket for the driver to change ends.

8:00 – Yellow Marker – After the Waterview interchange the motorway started to speed up quite a bit and we were making up for lost time, we were just passing Western Springs. At the Waterview interchange was where we finally passed the cyclist mentioned earlier, quite some distance from when we first saw him.

8:00 – By now the train is due to be arriving at Britomart with a probably about 300 hundred people on board having already dropped of a similar number of people at Grafton and Newmarket (the train is usually 5 or 6 carriages long and gets often gets completely jam packed).

8:07 – Red Marker – We finally arrive in the city and park the car, my work is abut the same distance walk from either the train station or the car park so it didn’t even have that benefit.

All up the trip via the road trip comes in at 18.3km while the train is 23.2km. The other thing is that things are much more stressful when driving, you constantly have to be on alert to what vehicles around you are doing and there is nothing more frustrating that sitting in traffic not going anywhere. On the train use my phone to surf the web, play games or listen to music or podcasts, much more relaxing.

Of course the trains that we use today are pulled by 40 year old freight locomotives that were never designed for the type of use they currently have, once we get brand new purpose build EMUs we can expect the rail trip to be at least 5-10 minutes faster than it is today. A little bit further in the future the CRL would drop this time by a similar margin meaning a trip to the middle of the CBD from Sturges Rd would be only 30-35 minutes, just imagine how popular that would be.

7 year-old newsflash: Kiwis driving less and loving it; MoT/NZTA curse freedom of choice

Hindsight may show that the transport sector is in the midst of unprecedented change.

NZTA’s latest report on state highway traffic shows that volumes were down 1.2% in 2011.  And if you look further back you can see that total vehicle volumes on the state highway network have been static since around 2004.  In the same period, New Zealand’s population has increased by approximately 7%, while real GDP grew by around 11% (Sources: Statistics NZ and Reserve Bank of NZ).  Taken together, this data suggests that per capita demand for vehicle travel has declined by around 1% per annum in the last seven years.

The plateau in vehicle volumes since 2004 is an unprecedented change from the last century, when we experienced consistent growth in vehicle travel.

In the 1950s and 60s this growth was largely driven by a combination of increasing vehicle ownership and sustained low fuel prices.  And then in the 80s and 90s increased workforce participation (i.e. baby boomers and two-income households) caused a second spurt in the demand for vehicle travel.   Throughout this whole time drivers were aided and abetted (albeit unintentionally) by poor planning/pricing practises that subsidized vehicles over other modes of travel.  Homogenous land use zoning and minimum parking requirements being among the most notable (despicable?).

Since 2004, however, this momentum has clearly dissipated: Vehicle volumes on New Zealand’s state highway network have been static or even falling.  And as this earlier post notes very similar trends are emerging overseas, even in countries that are considered to be ‘culturally remote’ from NZ.   This in turn suggests that the emerging trends in state highway traffic volumes in New Zealand are not a localised abberation, but in fact reflect broader forces that are changing people’s travel choices.

NB: The graph uses data up to 2007, which was the peak in NZ’s highway traffic volumes.  More recent data would show a longer plateau or even the beginning of a downward trend.  Nonetheless, a marked slowdown is already apparent in most of these countries.

So what gives?  Why are kiwis (and people overseas) driving less?  I think the following five factors are likely to lie behind the static/declining trends in per-capita vehicle travel that has been observed in New Zealand and overseas:

  1. Travel saturation – People already drive about as much as they can, given other constraints on their daily lives.  I suspect that for many people’s their lives are “saturated” by vehicle travel – hence they are unwilling to travel further.  This was not the case a few decades ago …
  2. Demographic shifts – The baby boomers are getting older and simply don’t need to travel so much, especially at peak times.  Meanwhile, young people simply aren’t as attached to their cars as previous generations; smart phones are the new i-sexy item.  Thus the need/desire for vehicle travel being undermined in both the younger and older age groups.
  3. Ongoing urbanisation – While we like to tell stories about our rural heritage, the reality is that NZ is a highly urbanized population, and increasingly so.  One of the key reasons people are attracted to urban areas is because they can reach many more activities without having to travel as far, especially by car.  My parents are in this category – not only are they semi-retired baby boomers, but they have also recently moved into the “city” from a lifestyle block on the urban periphery.  They now consequently drive far, far less than they did previously.  And they love it!
  4. Transport costs/policies – High fuel prices has slowly but surely driven small changes in travel and land use choices and increased demand for substitutes to vehicle travel, such as public transport and home delivery.  In downtown Auckland and Wellington, the removal of minimum parking requirements has allowed the value of parking to rise to more reasonable levels, which has in turn driven considerable mode shift and allowed for more intensive development patterns (NB: Minimum parking requirements should be abolished).  Meanwhile, the price of domestic airfares have reduced faster in real terms compared to car travel, which has subsequently chipped away at the demand for long distance vehicle travel.
  5. Technological change – the internet, which powers this very blog, is having a profound impact on travel demands.  Booking flights and internet banking are just two examples of tasks that are now routinely undertaken from home, without needing to travel anywhere.  More interesting, however, is the impact on global/domestic supply chains.  Whereas previously you would drive to the store to buy a book that had been delivered by truck from a warehouse, the same book is now delivered straight from the warehouse to your door, potentially replacing several intermediary trips.  Meanwhile, online journey planners and smart phones have helped to demystify public transport networks.  And finally telecommuting is slowly coming of age: I work from home at least 1-2 days per week and travel to work outside of peak times when I do, mainly because technology now makes it easy to do so.

These five factors are, I believe, causing much of the stagnation and subsequent decline in per capita demand for vehicle travel (let me know if I have missed something that you think is important!).  And the key point is that (aside from fuel prices) they are all the result of people’s choices.  And for this reason you would have to assume that the people making  these choices are doing so because it makes them better off.  Yes that’s right – Kiwis are driving less and they are doing so because it makes them better off.

Unfortunately, the government seems to be cursing kiwis for making these choices, or at the very least are in denial about the fact that is occurring.  To provide a somewhat banal example, to access data on traffic volumes through NZTA’s website you actually have to click on a green box titled “Traffic volumes are growing – see how much“, as illustrated below.

Errr ….  no, they’re not growing actually.  And in the entire time that NZTA has existed as an organisation they have actually been falling.  But, honestly, what do facts matter?

More concerning, however, are comments made by the General Manager of MOT.  When he was recently questioned by parliamentary select committee on recent trends in traffic volumes, he suggested that we should not base future forecasts on recent trends.  Errr …. yes, you should actually.  The GM is mistaken for several reasons.  Firstly, these trends are not that recent; they have been evident for approximately 7 years.  Standard traffic engineering analyses will base their calculations of future growth on the last 5-10 years of data.  This means that very soon what the GM has dismissed as a “recent trend” will actually be “the only trend in town” (unless of course transport engineers sneakily start extending their time horizons back in time).  Second, as mentioned previously the very same trends are evident in other countries overseas, i.e. they seem to be a global phenomenon, and in some countries they have been evident for some time.

So what gives?  If kiwis are driving less and loving it, why is NZTA and the MoT so keen to dismiss their behaviour as an irritating aberration?  Do they begrudge the fact that people are freely and willingly choosing to curtail their driving?   I’m not a conspiracy theorist but I can’t help but suspect that the attitude  MoT/NZTA is quite deliberate and has nothing to with either a) reality or b) what the public wants.  It’s all about them.

Very soon static/declining traffic data will have catastrophic implications for the business cases of many of their projects.  The Puhoi-Wellsford business case, for example, assumed annual traffic growth of 4% per year over several decades, before being “trimmed back” to only 1.5% thereafter.  So if the MoT/NZTA actually acknowledged that traffic volumes were not actually growing that much, then it removes much of the justification for their pet projects.  It’s really just pure, unadulterated bureaucratic self-interest (please, can someone who lives in Wellington track Bill English down and let him know that the MoT/NZTA are out of control and are wasting billions of dollars?  As Minister of Finance he really needs to know).

Given that the views of MoT/NZTA are increasingly irrelevant then we will just have to think through the wider implications of the decline in per capita travel demands for ourselves.  Here are just some impacts that spring to my mind (I’m sure there are others):

  • Major road capacity expansions should be deferred: Overall vehicle volumes are not increasing – not unless the background population growth is sufficiently large to offset the decline in per capita demand for vehicle travel (approximately 1% per annum).  I would suggest that all major highway capacity expansions are deferred for the foreseeable future.
  • Declining fuel excise revenue – NZTA and the MOT will simply have to make do with less money.  The revenue implications may even be amplified by a shift towards more fuel efficient cars, which pay less excise tax per kilometre.  With less money coming into the NLTF it’s important that public transport becomes more efficient and reduces its dependence on operating subsidies.  Otherwise we risk being in a situation where we have increase demand for public transport, but less revenue available to fund it.
  • Flatter peak travel demands – the factors listed above are also likely to result in flatter peak travel demands.  I expect that (semi-)retired baby boomers will simply elect to arrange their lives to avoid travelling during the peak period, and why wouldn’t you?  Also, technological changes may continue to cause more people to work from home in the morning before heading into the office later on.  Taken together, I would expect the peak/base ratio to reduce over time.
  • Public transport networks need to focus on all-day travel – This is a natural extension of the previous comment.  Our public transport services typically focus services on meeting demands during peak periods, rather than all-day travel.  As travel patterns increasingly disperse (in a temporal sense) they will need to focus on all-day demand patterns.

None of this is to suggest that roads will suddenly be unnecessary or that targeted capacity expansions are not needed.  Indeed in rapidly growing areas some capacity improvements may well be warranted.   For the average New Zealander cars will still be “king”, and will continue to be so for the forseable future.

But the key point is this: Decisions on major new investment should be driven by the marginal user; it is their behaviour (rather than the average users) that should be the primary determinant of future investment.  And it seems patently clear (based on the available evidence) that, at the margin, New Zealanders are choosing to reduce their demand for vehicle travel.  To put it bluntly MoT/NZTA need to wake up from their self-induced coma  so that they can revise their planned transport investments in light of this “new information” (which I guess in Government is defined as anything that happens within the last decade or so, or if you’re a conservative then anything within the the last millennium is considered “recent”) .

It’s quite simple really: Per capita demand for vehicle travel has been declining steadily for around seven years and for many people, like my parents, has left them better off.  We’d be even better off, however, if the MoT/NZTA recognised that these changes were occurring and responded by revisiting their proposed transport investment mix.  How about it?

Final draft of the Auckland Plan

Tomorrow could be quite an important day for the city. Up before the councils governing body is the final version of the 30 year Auckland plan and while it isn’t a done deal, we hopefully won’t see to many changes. You can read the final draft here and it starts about half way down in Chapter 13. If you haven’t been reading much about the plan here is a high level summary of what it is about:

The Auckland Plan is the strategy to make Auckland an even better place that it is now and create the world’s most liveable city. It is also the plan through which we prepare for the additional one million people and around 400,000 new homes we may have to accommodate by 2040. Population, housing and business growth over the next 30 years will provide Auckland with important opportunities and resources to be even more liveable.

We have covered the plan, along with the changes to it quite a bit over the last few months as councillors have debated and tweaked it following the various rounds of consultation. I and my fellow bloggers haven’t always agreed with every part of the plan or the changes being made to it but overall I think the council have done a fairly good job and it help to make the city a much better place. The plan is pretty big at almost 300 pages so with this post I am just going to look at some bits of the final transport section. This section seems to have been one that has swung around like a pendulum a bit, first it came out talking about a lot of PT improvements (although was really a bit of PT wash). The revised drafts that came out contained a lot more focus on road projects which led to a bit of publicity and now things seem to have swung a little back towards PT.

First up the strategic direction.

The first thing I noticed here is the the PT trip targets has increased even further that what it was only a few weeks ago with the aim now being 140m trips per year on PT by 2022. That means in the space of a few short weeks we have gone from not having a patronage target at all to having quite a decent and challenging one (based on population projections that would put our trips per person per year close to 80).

Here are some maps of the strategic transport network by 2040 and the key projects will take place to reach it. One thing you will notice is that a NW busway has now been added which excellent.

And here is the final list of major projects by the decade they happen in:

First Decade: 2011 to 2020

  • Public transport service improvements, including bus services to the airport
  • Integrated transport ticketing and fares
  • Rail network electrification and increased train frequencies to 10 minutes
  • Western Ring Route, Newmarket Viaduct and Victoria Park Tunnel completion
  • Removal of pinch-points in the strategic road network to improve throughput (such as widening from Hill Road to Takinini on the Southern Motorway and others)
  • City Rail Link completion
  • City Centre transport improvements (as described in the Auckland City Centre Master Plan)
  • Completion of the Auckland Manukau Eastern Transport Initiative and the East-West Link between State Highway 20 Onehunga and State Highway 1
  • Arterial road network improvements (with a focus on the movement of Walking and cycling infrastructure improvements (includes completing 70% of the regional cycle network)
  • Ferry network extension to Hobsonville and Beach Haven
  • Route protection:
    • Dedicated rail corridor to the Auckland Airport
    • Additional Waitematā Harbour Crossing (road and public transport) Rail to the North Shore
    • Avondale-Southdown rail corridor
  • Construct the Pūhoi-Wellsford Motorway Phase 1; Pūhoi to Warkworth
  • Complete electrification of rail to Pukekohe

Second Decade: 2021 to 2030

  • Construct rail to Auckland Airport
  • Continue improvements to the arterial road network (with a focus on the movement of public transport and regional freight)
  • Complete construction of an additional harbour crossing (road and PT)
  • Construct improved rail and road access to the port
  • Extend the Northern Busway from Constellation to Silverdale and from Onewa to the City Centre with bus lanes on the Auckland Harbour Bridge
  • Triple-track the North Island Main Trunk rail line (the Port to Westfield to Papakura) for rail freight
  • Construct the Warkworth-Wellsford motorway
  • Improve airport road access – SH20A and 20B
  • Complete the regional cycle network
  • Continue removal of pinch-points in the strategic roading network to improve throughput
  • Continue City Centre transport improvements (as described in the Auckland City Centre Master Plan)
  • Route protection: o Constellation-Westgate-Extend the busway from Botany to Flat Bush to Manukau
  • Investigate extending the ferry services along both the north- eastern coastline and the southern coastline especially for recreational and leisure opportunities

Third Decade: 2031 to 2040

  • Renew optimisation initiatives to take advantage of technology developments
  • Construct busway along SH16 between Lincoln interchange and Waterview interchange
  • Construct the Avondale-Southdown rail connection
  • Transport infrastructure and services in place in new greenfield areas to support their development.

Lastly there is also a bit of information on the top three projects, most of it isn’t new so I won’t paste it here but I think that one important bit of information is this line from the CRL part:

Eighty per cent of submitters on the Draft Auckland Plan who referred to the CRL supported its construction.

80% support for such a large and costly infrastructure project is pretty impressive. I wonder what level of support the various RoNS would get if subjected to the same process?


Spot the Change

On my home tonight I noticed these new signs around the posts at Britomart, can you spot the new addition? I do like the use of the pillars to display these maps and combined with the recent improvements to how destinations are shown it should hopefully make things much easier for people unfamiliar with the system.

(sorry for the quality, it was taken with my phone while on the way to my train).


And the transport prize goes to …

This is the first of what I hope will be a series of short posts on good/bad transport behaviour.  The idea is that by “naming” people/companies that deliver good transport outcomes, and “shaming” those that perform poorly, we can create a semblance of public pressure for businesses to pull up their transport socks.

From the outset I should say that these posts are necessarily subjective and somewhat banal: They revolve around me and my mixed experiences in Auckland.  But if you have a particularly good/bad transport experience of your own then please email me at and I’ll consider your experience for inclusion in future award ceremonies.

Without further ado, my prize for best transport experience of the week goes to:

Big Brands Online – this website sells home appliances and provides free delivery with all purchases.  No need to leave home, just click your way to a new fridge and sit back as their delivery trucks bring home the bacon within 1-2 days.  Big Brands Online wins the Auckland transport blog prize for recognizing home delivery as an opportunity to cut huge costs (retail space, car-parking etc) from their business model – savings which are passed onto consumers in the form of lower prices and free delivery.  May the capitalist planets align to ensure the success of their business.



But at the other end of the spectrum my award for worst transport experience of the week goes to …

The Localist – 2-3 weeks ago I was crossing the intersection of Symonds/Alten Street and I was almost nailed by a red light runner.  As the car flashed past in front of me I saw “The Localist” sign emblazoned on the door.  I promptly lodged a complaint via the Localist website; three weeks later I have not yet received a response (let alone an explanation) from the company.   Upon perusing their website I noted that The Localist is a subsidiary of NZ Post.  In their “corporate sustainability” strategy NZ Post state the following mission (which they suggest applies to all their subsidiaries):

“enhance the sustainable growth of the New Zealand Post Group through acting to protect and build the environmental, human and social resources needed for the future.”

I’d suggest that by failing to follow up on my complaint about their driver’s dangerous red light running  “The Localist” has in turn failed to meet their self-proclaimed corporate responsibility.  It is for this reason that they win my award for worst transport experience of the week.

P.s. Don’t forget to email me with your own transport experiences, good or bad!

Fairer and Smarter Fares

In public transport circles there is always going to be a lot of debate over the levels of fares. Should they be higher to reduce the amount of subsidy required? Should they be lower to encourage greater ridership and/or for social equity reasons? Should it be free to eliminate the costs of collection and to truly incentivise PT use?  Through the PT benchmark study undertaken last year, we learned that Auckland’s public transport fares are relatively high – while NZTA’s farebox recovery policy seems likely to raise fares even higher in the longer term as it seeks to reach a 50% farebox recovery ratio (the proportion of operating costs paid for through fares).

The table below is from the benchmark study and highlights Auckland’s high fares on a per passenger kilometre basis: 

The extent to which we generally raise fares in the future is largely a political decision around how much we wish to subsidise public transport because of its external benefits like congestion reduction, environmental gains and support for desirable land-use outcomes. Lower fares may also raise patronage so much that the revenue loss from each individual ride is made up for through an increase in the number of riders overall – although that will largely depend on when and where those riders are added to the system and whether they require expensive additional services and/or infrastructure.

What I’m more interested in is exploring ways in which we might fiddle around with the level of our fares on a cost-neutral basis so that its structure contributes more to where we want public transport heading to. There are a few relevant matters here:

  • The extent to which longer trips should be more expensive than shorter trips
  • The level of discount we give to unlimited trip passes (daily, weekly & monthly)
  • Whether (and to what extent) we discount travel that takes place outside peak times


Charging more for longer trips

Starting first with comparing fares for long trips with those for shorter trips, it certainly seems as though Auckland’s fare structure starts very low and increases quite dramatically when compared to other fare structures around the world. Going by the cash fares for buses, we have a single stage costing $1.80 heading right up to 8 stages costing $10.30. In contrast, Vancouver’s base fare is $2.50 for travel with Zone 1 (including free transfers) and increases to a maximum fare of $5.00 for travel within all three zones (with a single intermediary step of $3.75 for travel within 2 zones).

In some respects, charging more for longer trips in a proportionate ways makes sense – as longer trips require more service kilometres, which comes at a cost. The rider is also gaining a greater benefit from the longer trip themselves, so there’s also likely to be a willingness to pay a higher fare. But on the other hand, there are actually some compelling reasons why we might want to reward longer trips by not charging such a high fare – after all, longer trips generate greater congestion reduction benefits while buses and trains have to complete the whole run anyway, so the extra passenger riding further doesn’t add a marginal cost (although obviously longer routes are much more expensive to run, but that might be a network design issue more than anything else).

My feeling is that, out of the three issues with our current fare structure, this is the one least likely to need significant change – but in the long run as we shift to a zone based fare system, reducing the number of zones and flattening the cost difference between shorter and longer trips, might be a smart thing to do – to help encourage those longer trips onto public transport and get them off the road.


The second matter I think our fares need to be smarter around relates to the level of discount we give to unlimited trip passes, be they daily, weekly or monthly passes. When looking at this issue I tend to think that a fundamental decision should be made around trying to attract as many of our public transport customers as is possible to be using weekly or monthly tickets. People with such tickets will obviously try to get the best value from them, so will be regular and frequent users of PT – and they should be rewarded for doing so. Not only do they pay for their travel ‘up front’, but they are also likely to become regular users of off-peak, evening and weekend services – as those trips will effectively be ‘free’ to them (as generally people budget for a monthly pass on their weekday travel requirements).

In some parts of the city monthly unlimited passes work well but in other parts they aren’t effective at all. If we take the price of monthly rail passes in Auckland, there are two available.

  • A ‘City Monthly’ which is unlimited travel on the network between New Lynn, Onehunga and Otahuhu. This is equivalent to a 3 stage journey.
  • An ‘All Zones Monthly’ which extends that travel out to Waitakere and Papakura. This is equivalent to a 6 stage journey

The table bellows shows what the costs for each of the passes cost and the number of trips you would need to take break even for the passes vs the cash or 10 trip fare. As you can see the number of trips you would need to take if you were within a 1 stage area just wouldn’t make a monthly pass viable. By comparison if you live in a 5 stage zone like I do you need to take just 28 trips (or 14 workdays) to have been better off.

Part of the problem though is these passes are only useful for people going through town, people going from say Morningside to Henderson on a regular basis would only really be able to use cash or 10 trip fares so miss out on the benefits of monthly passes. This suggests that we need to rethink the cost and conditions of our monthly passes so that they are more useful for a greater proportion of the population. Switching to a zone-based system could mean a wider variety of monthly passes being available (a pass for 1 zone, 2 zones etc.) while generally if we are to look at anywhere in our fare system where we want to decrease the cost of fares, I think monthly passes would be the place to start. Let’s reward our best customers with a good deal.

Peak vs Off peak Pricing

Finally, we get to that vexed question of whether there should be a price differential between peak and off-peak travel. On the negative side first, it obviously adds some complexity to our fare system – at the very time we’re trying to simplify it and make public transport easier to understand. There’s also the issue of matching up this shift with a greater focus on monthly passes – as it seems unlikely most people would want their monthly pass restricted to off-peak travel only. The ability to just jump on anything, anytime, is a great attribute of having an unlimited travel pass. The other possible downside of peak pricing is that it is our peak time PT users who create the greatest congestion relief benefits for road-users, so it could come across as a bit counter-productive to charge highest fares for those people who are easing congestion the most.

But on the plus side, there is a lot of sense in having a price difference between peak and off-peak travel for a number of reasons. The most obvious is so that we can use our system more efficiently, flattening out the ‘peaks’ of demand and therefore being able to carry a lot more passengers without the need for extra buses, trains or PT infrastructure (or, more realistically, the need for proportionately less additional buses, trains or PT infrastructure). Getting an extra bus on the road at peak times, for example, is always going to be pretty expensive – because you need to be buying more buses, hiring more drivers and probably running more empty service kilometres (repositioning after that peak run).

Auckland’s PT situation, where we want to grow patronage significantly in the near future, but we don’t have the money to spend a huge amount on PT services and infrastructure (largely because of Central Government’s squeeze on funding) means that ‘flattening’ demand and getting a lot more out of our existing system is likely to be necessary. There are other ways we can do that to just ‘peak pricing’, such as providing a better quality of service throughout the day, but I think that creating a price discount for off-peak travel to reflect the lower cost of providing extra off-peak service, is something that Auckland will need to look long and hard at.

So I think we could be much smarter about public transport fares. Hopefully these issues will be looked at as Auckland Transport focuses on implementing integrated fares to go with the integrated ticketing we’ll have finished by the end of this year. Also one positive is that we haven’t actually had a fare increase for about 2 years, hopefully the huge increases in patronage in the last few years are helping to reduce the need for them.

Land Hungry Cars

Seems like the best way to become a regular blogger on this site is to continually bombard the authors with Guest Posts – I’m excited about joining the team!

One of the fundamental geometric advantages of public transport is that it uses less space to shift a certain number of people, than your typical car does. This is most often talked about in terms of corridor space, with such metrics as “this railway line carries the equivalent of ten lanes of traffic” being commonly discussed. This is very true of course, but misses another element of the comparison – which is storage space. While rail and bus depots look fairly large, on a per person carried basis, they are a fraction of the space that gets dedicated to car storage – more commonly known as parking.

While these matters are generally quite logical, being able to quantify the space impact of different transport options is pretty useful when we’re looking at how to integrate our land-use aspirations with the transport projects we prioritise. A useful article by Todd Littman of the Victoria Transport Policy Institute helps by doing some of this work.

Newman and Kenworthy (1999, Table 3.9) found that automobile dependent cities average about 7 meters of road length per capita, while less automobile-dependent cities average about 2.5 meters, and parking supply follows a similar pattern. This indicates that automobile-oriented transportation increases facility land requirements 3 to 5 times. Put differently, 66% to 80% of the land devoted to roads and parking facilities in modern cities results from the greater space requirements of automobile transport.

In addition, motor vehicle traffic tends to reduce development density indirectly by increasing the need for sidewalk and building setbacks to avoid traffic noise and dust, so larger boulevards, highways shoulders and front lawns can be considered, in part, a land use cost of motor vehicle transport.

The key point here is to think about land as the obvious scarce resource it is – especially in our cities. If land is being used for wider and wider roads and larger and larger parking lots, that is land which cannot be used for buildings, parks and other – arguably more productive and desirable – uses. Littman goes on to detail this on a per capita basis:

An urban arterial traffic lane can typically accommodate about 1,000 peak-period vehicles. If the average urban automobile commuter drives 10 kilometers each way on a 3-meter wide lane, each requires 60 square meters of additional road space (3m width x 10,000m length x 2 daily commutes ÷ 1,000), plus two to four parking spaces (one at home, one at work, and a share at other destinations) that average 10 square meters for curb parking or 20 square meters for off-street parking. Each additional urban motorist therefore requires 80 to 140 square meters of land for additional road and parking space to avoid increasing traffic and parking congestion.

Compare this with other urban land uses. A typical urban resident uses about 100 square meters of land for a small-lot (400 sq. m.) single-family home with four residents, and less for multi-family housing (townhouses, condominiums and apartments). A typical employee needs about 10 square meters of office space or about 30 square meters for retail. This indicates that an automobile requires more land than a typical urban resident uses for housing, jobs and commercial activities. Automobiles more than double the amount of land required per capita.

That final sentence is critical, that automobiles double the amount of space required on a per capita basis. In other words, providing for cars halves the amount of space available for more productive activities.

What becomes particularly relevant is how this varies across different land uses – which is illustrated in the graph below:

You can really see this domination of commercial areas by roads and parking when you look at recently developed employment centres in suburban USA – with Tysons Corner near Washington DC probably being the most classic example: The ironic thing about having so much land dedicated to roads and parking in commercial centres like this is that this is most probably extremely valuable land. Obviously the zoning is fairly permissive for higher-intensity developments, obviously there’s already a number of businesses located here (the 12th largest employment centre in the USA), and it has good access to both downtown Washington DC and Dulles Airport. Yet so much of this precious land has been eaten up by roads and parking – because the auto-oriented model the places was developed around is just so extremely inefficient when it comes to utilising space.

(By the way, there are some pretty grand plans to completely change Tysons Corner over the next few decades, once the Silver Line Metro System is threaded through it).

What Littman’s work, as well as observations of areas like Tysons Corner, tells us is that if we want to be successful at utilising our urban space more efficiently, then we really need to think very hard about the contribution of transport to this goal, or how transport can undermine this goal if we continue to rely on a primary method of transportation that’s so extremely space inefficient. Furthermore, from an economic perspective if we are to utilise this huge amount of available land that’s often located in logical areas for growth and development, then we need to be providing an attractive alternative to the auto-centric planning paradigm we often get so stuck in. Quite simply, you can’t turn a Manukau or a Botany or a Westgate into the next Newmarket unless a massively lower proportion of people drive – there simply isn’t the space.