This is a guest post from Cantabrian reader Brendon Harre, taking a look at New Zealand’s slow productivity growth and whether better-functioning cities can improve it. It was originally posted on Brendon’s blog.

It is election season in New Zealand and recently the two main contenders squared off in a televised debate.

One contender, Labour’s new leader, Jacinda Ardern was concerned about New Zealand’s lack of productivity growth. Jacinda stated that better access to tertiary education from Labour’s proposed three years free tertiary education policy would help correct the country’s poor productivity record. While the current Prime Minister and National party leader Bill English denied that New Zealand was in a productivity recession and defended his government’s economic performance over the last nine years.

JB Were is an investment broker -they are advising its clients to reduce their exposure to NZ equities and property. The report and analysis can be read here and here

I will leave it up to reader to decide who is right about New Zealand’s general productivity situation. What I would like to consider is whether New Zealand needs to commit to an urbanisation project to increase productivity in its cities? Would improving some aspects of urbanisation -transport, housing, infrastructure…. boost incomes and productivity?

Frequently, it is argued that Auckland’s roads are congested, that this wastes a large of amount of time and is a productivity loss to Auckland and the country. A recent NZIER report which focused on this issue indicated;

The benefits of de-congestion to current capacity in Auckland would be between $0.9 billion and $1.3 billion (1% to 1.4% of Auckland’s GDP).

Note the $0.9 to $1.3 billion benefit is based on the realistic assumption of a lower speed that maximises road carrying capacity (D), not a higher speed (A) that is equal to the speed limit, which some previous studies assumed.

The NZIER report shows that Auckland’s time delay at peak times is the second worst in Australasia. Auckland’s time delays are similar to much larger cities, such as Melbourne and Sydney, and is significantly worse than comparable sized cities, such as, Perth and Brisbane.

There is some problems with the contention that congestion is a cost. In some ways congestion is a symptom of success. Congestion is really a tipping point problem. Having too few travelers is as much an economic problem as having too many.

Also while NZIER’s de-congestion figures include flow-on benefits from time savings to businesses and households, they do not include the agglomeration benefits of improved accessibility. Such as, allowing more choice for business locations, and better skill matching for workers.

An alternative and I would suggest better way to improve urbanisation related productivity is to look at improving access and mobility. Specifically, by using a Bertaud type analysis of measuring the total number of workplaces a worker can easily access in an acceptable timeframe.

Alain Bertaud explains in ‘Cities as Labor Markets’ that agglomeration results from cities generating scale economies by firms reducing costs per unit, knowledge spillovers, and lowered transaction costs due to more competing suppliers and consumers in close proximity. Bertaud’s paper demonstrates that workers’ mobility –their ability to reach a large number of potential jobs in as short a travel time as possible, is a key factor in increasing urban productivity. Importantly, he gives some empirical data on this effect.

In Korean cities, a 10% increase in the number of jobs accessible per worker corresponds to a 2.4% increase in workers’ productivity.

In the US, Melo et al. show that the productivity effect of accessibility, measured by an increase in wages, is correlated to the number of jobs per worker accessible within a 60 minute commuting range. The maximum impact on wages is obtained when the number of jobs accessible within 20 minutes increases; within this travel time, a doubling in the number of jobs results in an increase in real wages of 6.5%. Beyond 20 minutes of travel time, worker productivity still increases, but its rate decays and practically disappears beyond 60 minutes.

Both papers demonstrate that workers’ mobility –their ability to reach a large number of potential jobs in as short a travel time as possible, is a key factor in increasing the productivity of large cities and the welfare of their workers. Large agglomerations of workers do not insure a high productivity in the absence of worker mobility. The time spent commuting should, therefore, be a key indicator in assessing the way large cities are managed. (p. 24, 25).

So we can conclude the benefits from investing in better transport infrastructure are quite large. NZIER’s analysis shows that the excess costs of congestion are around $1 billion due to a combination of wasted time and productivity losses. If it was possible to transformationally reduce that, either by reducing congestion or reducing people’s exposure to it, that could bring large benefits.

Investing in better transport services for Auckland, depending on the specific cost/benefit business cases has the potential to have good payoffs. Also seriously worth considering is congestion road pricing, which would be very beneficial, for minimising travel delays and maximising the number of travelers able to use the road network.

There is another aspect to improving mobility outcomes. City residents may choose to permanently move house or job to avoid travel time delays. There is two ways this effect can exhibit itself -one negative, one positive.

Giving up on employment or not applying for specifically located work due to excessive time delays, inhibits agglomeration benefits, because it affects skill matching.

Being able to move residence to be closer to work (measured in time), will benefit agglomeration and therefore productivity and incomes. This ability will somewhat depend on the general housing affordability of a city. But most importantly it depends on the specifics of being able build more city residences of the desired size, price and location, such that supply elastically responds to demand.

So the various proposed housing market reforms being discussed during this election season do have the potential to improve productivity in New Zealand’s cities.

Making it easier for cities to grow organically, both up and out has been my motivation for many of my articles -in particular my idea about reciprocal intensification. An idea which was first published on Interest.co.nz in August 2016 (Part two: of this article). Greater Auckland recently reviewed legislative changes that will make reciprocal intensification easier in an article they published in August 2017.

Reciprocal intensification is my proposal for an acceptable way for New Zealand to grow its cities organically. Organic city growth has been very successful elsewhere, for instance Tokyo, which I about write here. There may be a better way to achieve organic growth than reciprocal intensification and certainly a broader package of urbanisation reforms will be needed.

I think it is an exciting time for urbanisation in New Zealand. A real conversation is being had on the nature of our built environment. Perhaps for the first time ever in New Zealand, certainly in my lifetime, issues of housing, transport and urbanisation in general are rising to the top of the public debate.

Over 80% of New Zealanders live in urban environments and 33% live in Auckland. Making those environments as productive as possible should be a priority for New Zealand.


I have tried in my own way to contribute to New Zealand’s housing and transport conversation with a collection of writings on a blogsite called –New Zealand needs an urbanisation project. Hopefully this article is a good contribution to that effort.

Share this

41 comments

  1. My reading of NZIER’s report wrt to this statement “the excess costs of congestion are around $1 billion due to a combination of wasted time and productivity losses. If it was possible to transformationally reduce that, either by reducing congestion or reducing people’s exposure to it….” is they do not measure all the productivity loss -only that which is directly related to the wastage of time. The NZIER did not measure the wider agglomeration benefits -the sort of benefits which Bertaud’s paper discusses. So the potential benefits will be larger than $1 billion but that currently most of that benefit is unmeasured.

  2. Not fully measuring agglomeration productivity effects is an important issue and I should have made that clearer in my original paper. If something is not measured then it is not valued. For instance, the government has not had an agreed suicide target or childhood poverty target (although Bill English announced one in the recent leaders debates) and these lack of targets/measures has hindered analysis of what works/doesn’t and ultimately how to fix the problem.

    Not fully measuring agglomeration productivity gains means the value of the built environments is not fully appreciated. It means that the status quo perception that value is only created on the farm or the natural environment (tourism) is not challenged.

    Not measuring agglomeration is ignoring the elephant in the economy. Over 80% of kiwis are attracted to urban environments -they perceive an advantage for doing this -but as a country New Zealand doesn’t value this, we are not fully measuring it and we are not systematically enhancing it.

    1. Regarding Kiwis urban attraction vs national viewpoint – It’s frustrating that so many would prefer a short and simple answer, either ignorant of (or not caring for) the fact that simplicity demands a reduction of accuracy. The result is that people will often cite telecommuting and flexible hours as magic pills that will solve congestion issues (don’t get me started on how driverless cars will magically “solve” spatial efficiency issues).

      It is as if people are feeling a profound lack of control or influence in the development of our society, so go for soundbites of ideas rather than a well thought out plan – That way it’s easier to have a sense of belonging, with the apparent strength in numbers that neo-tribalism comes with. Not measuring agglomeration is an example of this tendency. We need to start investigating how to measure this now, regardless of how uncomfortable it may make us. The alternative will see Auckland doomed to poor fiscal performance, at best.

      1. YES!

        “…It is as if people are feeling a profound lack of control or influence in the development of our society, so go for soundbites of ideas rather than a well thought out plan…”

        One of my favourite quotes is H L Mencken; “For every complex problem there is a solution that is simple, popular, and wrong”. I strongly believe that the world is full of complex problems where politics is using the simple, popular, wrong solutions. Congestion pricing properly understood, is complex and so consummately beneficial, it is a no-brainer.

        But your insight is that the people feel they “aren’t in control” if they have to take the word of some expert. But I still don’t see that they couldn’t think a bit harder and longer, and intelligently support win-win-win concepts. Then you have the mainstream media not really helping; explanations of complex problems don’t really “sell”, and people probably haven’t got the patience to read or listen or concentrate hard enough.

        People probably were more willing to use their faculties back when our civilisation was lifting itself up into the “developed” condition, otherwise we might never have got to where we are now! I am impressed, reading things from the past by non-economists like Frank Lloyd Wright, Henry Ford, Charles Booth, and Raymond Unwin, just how instinctively they understood the complex problem of the relationship between transport systems and “urban land rent”, which so vexed Karl Marx and Henry George and their contemporaries.

        Measuring agglomeration efficiencies is in its infancy; what we really have is a bunch of concepts to explain why larger populated cities have higher incomes than smaller ones, all else being equal. In urban planning at least, there has been far too much uncautious use of “agglomeration efficiencies” as justification for policies that could well be having the opposite effect. For example, the UK’s productivity gap has not been convincingly explained as yet by anything other than its Urban Planning system (Evans et al, Cheshire et al). But a crude assumption among contemporary urban planners is that “preventing sprawl” creates agglomeration efficiencies rather than foregoes them. But Evans et al and Cheshire et al convincingly argue that the opposite is the case. The correlation between density and agglomeration efficiency has been observed of course, BUT in all such cases the density and the agglomeration evolved simultaneously under market forces, not planning fiat!

        One obvious and massively significant agglomeration you would forego by “preventing sprawl”, is “Silicon Valley”. All those whiz-kids with bright ideas and little capital, got their start in rudimentary buildings in near-Wild-West regulatory conditions on low-cost exurban land.

  3. I wish people would stop using that stylised curve of speed vs volume, or at least explain that the bottom part only exists upstream of a bottleneck meaning that others entering the system at the bottleneck are using the capacity instead.

    1. The stylised speed vs. volume curve comes from real data obtained by Lupton and Wallis of actual observations from a NZ motorway study.

      Congestion will generally always occur centrally and then radiate outwards because growing cities direct increasing volumes of traffic to a small and fixed amount of central road space.

      Read about congestion and other spatial economic considerations here https://medium.com/land-buildings-identity-and-values/successful-cities-understand-spatial-economics-95c272ac04c9

      1. Fascinating you should think it is so new. That curve was published by a guy named Greenshields in 1935 and it dominated for 50 years. There are some problems. 1/ It didn’t fit any motorway data. 2/ He used holiday data for a two lane road, 3/ The bottom bit was assumed to be symmetrical to the top curve. The best thinking now comes from a guy called Boris Kerner who has focused on what actually happens from observation, particularly as stable flow breaks down.
        My biggest complaint is that that bottom part is not applicable unless there is a down-stream bottleneck to cause it. If you measure down stream of a bottle neck say after an on ramp you can only see points on the top part of that graph. If you measure at a bottleneck, say mid interchange you can measure only points on the top but right through to the capacity point but nothing on the bottom. If you measure upstream of a bottleneck you will see the left part of the top and some points on the bottom if flow has broken down, but you cant get the capacity point. Why does this matter? Because any policy decision you make based on that bottom bit means you are proposing something not specifically related to the bottleneck capacity.

        1. You’re absolutely correct that bottlenecks matter. That doesn’t change the general principle. Either we attend to the bottlenecks with added capacity, or price the road at the bottleneck. Of course attending to the bottlenecks is far more cost-effective that the assumption that dozens of kms of roads need widening. Both things are no-brainers – the congestion pricing and fixing bottlenecks. Both maximise the return on past public investment in the network as a whole.

          Another point is that the more dispersed travel patterns are, the less the whole network is affected by a few bottlenecks. It is interesting that Auckland still has such congestion problems even with such dispersion of employment. Part of the problem, is lack of alternative “ways around” the bottlenecks.

          Auckland is extremely under-provided with road network and highway lane-miles, relative to European cities and even more so relative to US ones, and we wouldn’t be going mad on car-dependence if we played a bit of catch-up for a while. But we can be smart with our budget. Besides fixing bottlenecks, alternative routes round the outside of the city on low-cost greenfields, would be the best bang for the public buck. Dispersion of travel patterns still has benefits to give, Auckland is still only averagely dispersed. The most ideal hypothetical model would be jobs evenly dispersed and workforces living next door to them; central concentration is dead as an overarching principle. Not to say that central cities can’t evolve, but they will evolve quicker in conditions of spatial competition. New York evolved like it did even as it was spreading out at low density at the same time. If prescriptive planning tries to force urban economic activity into the centre, the centre becomes a rentiers and speculators paradise at the expense of productive grassroots economic activity.

        2. Wow of course. We just keep building motorways until the congestion is fixed. It’s so obvious now. If only someone had thought of that 60 years ago and just kept going with that motorway policy, ignoring any other alternatives like public transport or cycling, we could be congestion free.

          Thanks Phil!

        3. “alternative routes round the outside of the city on low-cost greenfields”

          Would those be the greenfields in the harbours, or the ones in the mountain ranges?

          Perhaps you should try to drive “round Auckland outside the city” yourself sometime. (Hint, you’ll need a boat and rock climbing hear).

        4. You guys slightly misrepresent my position. But it is true that congestion delays are minimised by keeping up with increasing road network capacity. The evidence is clear on that; it is absurd to believe that induced traffic is kind of infinite and exponentially elastic in its response to provision of capacity. It is true that “you can’t build your way out of congestion” but this relates to roughly constant, moderate congestion. If you don’t bother adding capacity decade after decade, you will end up with congestion delays of, say, 45 minutes per hour of driving at peak instead of 15. It is true that you can’t build your way out of 15-minute congestion, but you can avoid ending up with 45-minute congestion.

          But my main argument is that whatever capacity you provide, needs to be maximised by pricing, and maintaining flow.

          It is also not supported by evidence, that spending public money on “other modes” instead of roads, minimises congestion as effectively. The win-win position is to have roads and urban design that allows buses, bicycles and pedestrians to co-exist with cars. But it is decades too late to retrospectively do this to the first world’s most street-space-starved city (according to the authoritative UN database). I would be one of the most experienced cycle commuters in NZ, and I heartily concur with the great cycling guru John Forester, that all we want are nice wide roads with wide shoulders. Slapping green paint on inadequate, narrow roads, to “exclude the motorist”, doesn’t make things any safer for anyone.

          There is already talk of another harbour bridge or tunnel, and good foresight would suggest it should go across Manukau Harbour near the Heads. Duplicating the current one would be a stupid waste of money, and a rent-seekers ploy associated with the primacy of CBD property. The money should be spent where it ties into “New Cities” on greenfields, not where it dumps yet more cars into the existing choke point. There is massive potential in the land mass between Waimauku and Otaua. Did our forefathers throw up their hands in horror and decide that topography meant that no city should be built where Auckland is now, and where Wellington is now?

        5. Hi
          Sorry to come in so late in the argument, it was drawn to my attention. Brendon wasnt saying that Ian and i discovered the curve, just that the curve he used as illustration was taken from our paper. I am of course familiar with Greenshields et al, The observed data fits more closely to a BPR curve with power 4 but it can also be fitted with an Akçelik function. As you will know the former is based on link density and the latter based on queueing (ie a bottleneck). As for the observations themselves, they are from old (pre ramp metering) Transit NZ data and are 15 minute observations over a month. If the dominant impact was the downstream bottleneck, you would expect the observations to be clustered (as you describe) – as indead is observed at some other Auckland sites and in the data i have for the Hutt road. (The clusters nevertheless sit where you would predict based on the BPR curve). These data cover the whole range, which suggests to me that we are seeing link congestion. Does it matter? If you are looking for engineering solutions, maybe it does – although i would have thought any good engineer would look for the pinch point – be it a section of road, a merge or whatever – and address that rather than worry about the shape of the curve, I am more interested in pricing solutions in which case it doesnt matter whether the congestion is caused by a bottleneck or a restricted link – the prescription is the same (as per Vickrey 1962 etc) – to price so that demand equals capacity.

      2. Oh and Greenshields only had one point on that bottom part of the curve but he managed to sketch that pretty curve through it.

        1. Which means it was measured down stream of a bottleneck. In this case caused by the Greenlane on ramp, Penrose on and SEART on. That part of the line is showing a relationship between speed and density when queuing occurs. It actually tells you nothing else and most certainly is not telling you the capacity of that piece of road. People used to think (most of us were trained to think) that as density increased speed dropped. Now we know it is simply the opposite, as speed reduces due to what is happening ahead, the following cars bunch up and increase density. If you want to make a difference you have to either increase the capacity at the actual bottleneck or price off people from travelling through the bottleneck and that will allow more of these people to pass through it instead. But enough with the bullshit about magically getting more through at higher speeds.
          Actually I have looked at that again. I am at a loss as to how you would get that curve off one site, every other data set I have seen shows part of the classical relationship, never the whole thing like that.

        2. I really wasn’t expecting this to be the disputable point in my article. NZIER and GA all seem to be ok with the graph. Does anyone else dispute its validity?

        3. Other way round Sailorboy. I have no problem at all with the article. I just really don’t like that symmetrical graph as it represents an idea I thought we had moved away from over the last 20 years. The graph doesn’t really help the article so if it were me I would just leave it out. To be really pedantic the top part of the stylised graph is too steep- the research behind COBA in the UK had a flat top for most of it. They showed it might as well have a slope of zero for most of the range with a dip at the far right. The bottom part of the graph is usually either left out (like the Highway Capacity Manual Chicago data) or shows only a few points like the HCM New York Parkway data did (in that case it actually showed the effect on inbound flow of traffic lights and had nothing to do with the road link itself). I thought most people were now on-board with the Three phase theory where traffic flow breaks down as a wide moving jam (J) and synchronized flow (S).

        4. Sorry, should have been more clear, my understanding is that you were challenging the interpretation that increasing congestion leads to flow breakdown and that you prefer the interpretation that flow breakdown downstream results in low speed and low throughput.

    2. The implications of the demand / flow curve are dramatic and it is a typical example of a complex problem where the inability of most people to understand it, is an obstruction to good policy making.

      The main reason for the political unsaleability of congestion pricing is in fact a wrong assumption, but even a lot of “experts” don’t understand – ask David Lupton about the struggles he has had with “experts” denying the reality of the U-shaped curve.

      This is the graph curve of the relationship between demand and speed/flow – David has an actual data-based illustration taken from Auckland motorway.

      The main reason to public resistance to congestion pricing, is the perception that a significant proportion of the people currently travelling, will be “priced off” in favour of wealthier drivers, who will represent a lucky reduced proportion of the current traffic, to be granted “access” at the crucial times.

      In fact it should be viewed as a modest payment by an increased number of drivers relative to what are achieving access under the status quo, in return for which faster access is provided. This is because the current flow rates under stop-start conditions (breakdown congestion) are around 1000 vehicles per lane per hour, and 1600+ is easily possible. In fact 2300 is the figure they work to in some countries! (Slow drivers obstructing outside lanes is probably the reason for NZ underperformance).

      But ironically, this at least partly solves the problem that mass transit is supposed to solve. Potential ridership under the status quo of breakdown congestion, reduced numbers of vehicles getting through, is potentially lost. Hypercongestion is probably a bigger inducement to use mass transit, than a modest charge to get quicker road access.

      Of course you can set up congestion pricing so it doesn’t increase vehicle throughput, either by setting the fee too low and perpetuating hypercongestion, or by setting it too high so that you do “price off” some of the 1000 vehicles per lane per hour currently getting through, and the few hundred better-off payers of the charge whiz through on what appears to be half-empty motorways. Either way, you are doomed to defeat by public backlash.

      1. I like the idea of congestion charging and have no fear of it. However I’d like to point out another concern that people have – How will it be implemented and will it be easy to understand?

        That was a big concern for the Northern Gateway. Of course, once the tolls were in place and people realised that ALPR works well enough the dissent died down. However there will be the same concerns when road pricing is first introduced into Auckland. Most likely be a lot of FUD too from vested interests.

        I wonder if the council will be empowered to bring forward road pricing plans, allow it to be in place sooner than a decade from now. I’d hope that whatever government in power allows the fees to be collected by NZTA (per Nth GW Toll Rd). It _is_ a state HW after all.

        1. Northern Gateway is an excellent example of an exceptional simple case. That is, there will be a massive time saving relative to the status-quo circuitous route. Cesare Marchetti is the go-to author on this. The Harbour Bridge was another classic case. Marchetti argues that governments are forever underestimating the potential for paying off viaducts, tunnels etc with tolls, when massive time savings are provided.

          Congestion pricing an existing network is highly complex, while the above is simple.

        2. A valid criticism 🙂

          The point that I was alluding to was that society tends to have an innate fear of change, per the cycle of acceptance. To be fair, I should say that I’m referring to a grossly simplified version of the cycle. Should probably also accept that I may be coloured in my thinking by my own experiences in corporate IT, where change (on so many levels) is the norm and knowledge of the cycle effectively mandatory.

        3. +1, the toll should have been set far higher. High enough to actually encourage some people to use the old route, so I’d guess $12-15?

  4. Comparing Auckland and Wellington’s congestion with Australian cities is benchmarking ourselves a bit low. They and us are both outlier-bad globally, for cities of a given size. Cities like Nashville and Indianapolis absolutely slaughter Auckland and Aussie cities of comparable population; and it is even more ridiculous when you compare piddly little 500,000-population Wellington, with comparators like Boise and Wichita. GPS-measured delays per hour of driving at peak, in urban areas of 500,000, are generally between 5 minutes and 10 minutes. Somehow Wellington manages to match cities like Los Angeles, pop 14 million; at a delay of around 45 minutes.

    Part of the problem is a kind of Walter Mitty syndrome on the part of Kiwis, dreaming that our cities “are” equivalents to London, Paris, New York, Hong Kong, etc. It is only slightly less absurd for, say, Ashburton and Levin to think this way too.

    1. Auckland and Wellington both have difficult geographies -which limit travel to a few narrow corridors. This means understanding congestion is an important factor in making those metropolitan areas work more productively.

      Even flat Christchurch has limitations -to the south is a rugged set of hills with only a couple of passes, one road and one rail tunnel (freight only), to the east is sea, to the north is a large river with only two road bridges (total 3 lanes each direction) and one rail bridge (freight only), to the west is a large airport, with an even larger no build noise zone.

      A correctly priced (as Phil Haywood describes) road congestion charge would send the right signals for transport users and transport providers.

      1. Congestion charge is a part of a solution.

        But without alternative modes, people will then just price the congestion costs into their lifestyles.

        I’d say that developing alternative transport is more important in the short term before things like congestion pricing are developed. Otherwise 1) any congestion costing model you develop will change if alternative modes introduce new equilibrium dynamics, 2) you hold back other forms of productivity development – non-car lifestyles, housing without parking, etc.

        In a sense congestion charging is really just “adding a lane” to the existing network. Pushes the problem down “the road”.

        1. Part of my point was that it is even worse than you say – it is not that people “just price the congestion charge into their lifestyle”, they actually get value for money for it in the form of quicker access. The whole point is not to “price low income drivers off”, it is to maintain traffic flow and hence provide access at the crucial time for up to twice as many drivers as the status quo. All the congestion charge has to to, is dissuade the small number of drivers who tip “demand” into the territory where it exceeds the roads capacity and causes flow to break down. For example, around 1000 vehicles per lane per hour are currently getting through at rush hour, which is around 17 per minute. It could be 27 per minute if the flow was maintained, but it is breaking down because at one crucial minute, as few as 5 extra vehicles over and above the 27, have attempted to join the flow. This need only happen once, and flow is collapsed for the next two hours!

          The number of drivers who need to be dissuaded from driving so as to preserve flow, is actually incredibly low. Because there is currently only 1000 per hour and everyone is ultimately making their plans around that, if flow was maintained instead of breaking down, the massive queue we currently would simply get through quicker. Many people could quite happily leave later, probably more people than what need to be dissuaded by the charge in the first place! When you get this, it is so frustrating that it is politically impossible. People just don’t understand it, that’s all. One of my favourite quotes is H L Mencken; “For every complex problem there is a solution that is simple, popular, and wrong”. I strongly believe that the world is full of complex problems where politics is using the simple, popular, wrong solutions. Congestion pricing properly understood, is complex and so consummately beneficial, it is a no-brainer.

        2. Sure that sounds good, but it’s not really an answer to my question/statement.

          It just pushes the problem further down “the road”. In this case the road in question can also be a function of time. Even if we increase the congestion fee as congestion increase, that is still just pushes the problem down the road.

          Further without alternatives your solution further isolates sectors of society that have trouble at the moment.

          Then it still comes back to my question – why do congestion pricing if changes in the network will have a major impact on equilibrium?

          There are many aspects of the world which are counter-intuitive. For example Braess’s Paradox. Or adding fees can actually increase rather than decrease usage.

          Complexity either arises due to lack of information or they arise from the system.

          Just because small hammers are wrong, doesn’t been a big (complex) hammer is the right solution.

          If your solution set is restricted on howto more cars. Then ok.

          But really we should be focused on the big problem/solution set – how to move people.

        3. Thanks, Nicholas. “Braess’s Paradox. Or adding fees can actually increase rather than decrease usage.”

          In the case I am describing, congestion pricing, the increased usage is “no-cost”, it is merely utilising the existing infrastructure better. The revenue can be put to some use, but it is not the same as a “tax”, it is a payment for which the payee directly receives something of value in return.

          I agree that the fee might have to be increased and increased and increased over time if there is no extra capacity provided, and if alternative “public” modes are not cost-effective – however, it is still preferable to the status quo where addition of capacity is problematic anyway, and congestion is probably costing everyone more than the fee that would solve it! And we don’t have a “happy” source of revenue to put towards longer-term solutions! It is all “just more taxes” to everybody at the moment.

          The isolated sectors of society are trapped in a paradigm where they need the flexibility of automobility more than everyone else. The problem with investment in public transport routes, is that the amenity value ends up getting “priced” into local rents and property prices anyway. So lower-income people end up priced out to remoter locations, and under-represented in PT ridership, and PT investment becomes a kind of lifestyle subsidy for well-off people with CBD jobs.

          As with a lot of things, poorer people would be far better off with cash assistance, than being used as a political pretext for big-project public expenditure that might not reach them with its benefits anyway.

          In the biggest picture, nothing reduces “isolation” more than “decentralisation” of amenities. For example, if you spend all your public budget on one massive centrally-located medical facility, you will end up maximising “isolation” relative to having a medical centre in every suburb. We already do schools in a decentralised way, so that no-one needs to be “isolated” from a single huge centrally-located education facility. Jobs are already decentralised anyway. Far too much planning is perversely attempting to centre urban travel patterns on the urban “centre”, and this is 180 degrees wrong for improving “isolation”.

          The kind of money being spent getting CBD employees to work from their toney “transit oriented” housing nodes, is at least as much as the cost of providing poor isolated people with a leased used small car and petrol vouchers, and I know which “solution” the poor isolated people would prefer the money to be spent on. Suburban bus services also could be the focus of PT subsidies instead of being socked with the highest per-km ticket costs by a wide margin. I see Tony Randle in the DomPost this very morning pointing out that people who ride trains 20-60km to the CBD are getting most of the subsidy expenditure at the expense of suburban-network bus services.

          The once-famous econometrist and land economist Colin Clark made the following argument in one of his last books, “Regional and Urban Location”, 1982.

          “…If rail and subway services to the center of large cities were charged at full cost….two consequences would follow. The employers of the lower paid workers in the city center would have to raise their wages, reduce the level of service offered, or move to suburban locations…Meanwhile the higher paid…would have an incentive to move their residences closer to the center…

          “…These movements would have their reflection in the price of land…….In net effect, the subsidies on rail and subway transport are subsidies to the owners of certain types of land – for which there is no social justification…”

      2. The geographic limitation factors are often more similar than we realise, from city to city. What we might regard as “a city on a flat plain with plenty of room to spread out at low cost of infrastructure”, often has managed to stay affordable and low-congestion in spite of rivers, swamps, unbuildable soft ground, etc. For example, a river like the Hutt River would in most cities have at least twice the intensity of bridge crossings that we do. Geography didn’t stop our ancestors 100 years ago from developing Kelburn, Northland, Karori etc with tunnels and a viaduct.

        It is a question whether undulating topography is any disadvantage given that “grade separation” of conflicting travel directions always needs costly bridges in a flat city. Undulating topography means half your engineering work is already done; this is why interchanges in some locations qualify so early under cost-benefit analysis (eg Haywards Hill). Tunnelling is getting cheaper and cheaper, we should be using it.

        1. Walkability is something that should be retained along with higher-speed high-flexibility transport modes, rather than lost or diminished. But what we need to understand about transport modes, is their relationship with economic land rent and hence housing justice.

          Cities that have never spread out with ubiquitous higher-speed high-flexibility transport modes, are always stuck with slums, tenements, and rental slavery in crowded, low-quality conditions. It really matters how much “land supply” is enabled for the urban economy, if it is only a small amount into which everyone is confined, the small land-owning class can gouge everyone for all eternity.

          The reason that high density, high quality conditions co-exist anywhere, like in Manhattan, is because the land-rentier class lost control long ago and New York, the urban area in which housing choice exists, is almost as big as the entire Netherlands. Alternatives to flattening urban land rent with automobiles and roads, include nationalisation as in the former Communist Bloc (Marx’s solution to the rent injustice issue); mixed “leasehold” with State ownership, as in Singapore; and highly competitive multiple integrated “public transport and property development” enterprises, as in Japan. Motor scooters are performing as a proxy for cars, in some developing countries, Bangkok is moving rapidly to improvements in affordability of housing of all kinds because of this.

        2. Brussels also should use congestion pricing if it doesn’t have it. It also has quite a lot of public transport, so it is interesting if this has not helped traffic congestion.

        3. Many part of Belgium have traffic issues.

          Even with heavy rail options between towns, my sister-in-law complains that it’s better to drive and deal with traffic. Since the train stations are outside and she’d need to call for a pickup anyway.

        4. “Cities that have never spread out with ubiquitous higher-speed high-flexibility transport modes”

          Can you give some examples.

          It’s funny as Auckland CBD get’s a 100% Walk Score.

        5. Nicholas, I mean literally every city in the developed world. The density of third world cities today, and Victorian-era western cities, is and was far higher than Manhattan today. Manhattan itself in 1900, was significantly denser than it is today. The densest part of Hong Kong in history was always Kowloon, which was never built “up”.

          People in these super-dense “crowding on the surface” conditions might be able to walk everywhere but they don’t see this as a net advantage. It is always accompanied by severe housing injustice, crowding and insanitary conditions. This was the norm everywhere in the west up to the end of the 1800’s. Every city in the west has spread out and reduced its overall density. The average factor is “16 fold”. At the same time, housing conditions improved dramatically and people got to own their own larger space.

          I specifically said that this spreading-out process did not necessarily supplant walkability in specific locations; merely that those walkable locations are now more spacious, sanitary and affordable than they would be without the spreading-out and comparative emptying-out. Exceptional built-“up” locations that have maintained density to a great extent, like Manhattan, are that way because of the way the local economy evolved. The vast majority of western cities are only a fraction as dense anywhere, even at their core, than they were in 1900 (of course in 1900 “the core” was the whole city).

          If you haven’t read Shlomo Angel et al “A Planet Full of Cities”, I strongly recommend it. There is some very interesting data on the density of Victorian-era cities and the trends since. Every significant European city and even the legacy US ones were once as dense as Dhaka and Lagos today (that is, denser than Manhattan today), and the proportion of home ownership, legal tenancy, and illegal slums, and the cost of rents in the legal market, would have been comparable to Dhaka and Lagos today. It was these conditions of housing injustice that led to the appeal and rise of Marxism. The west fixed it with the automobile. I am not saying there aren’t other ways, just that we need to understand what we are doing.

          Of course back when western cities were super dense, most of the population was still rural. The rural-to-urban population shift was really a rural-to-suburban shift. Some developing countries today are just adding ex-rural migrants to the same dense, under-infrastructured morass with the same intractable problems. These cities are not reaping the standard benefits of “size agglomeration”. If anything, agglomeration efficiencies depend on expansion of infrastructure, not the addition of population per se. The assumption that it is about added population, missed the factor that at least in the history of the west, the added urban populations correlated with the addition of infrastructure.

          Because it is so prohibitively difficult to expand infrastructure right in the middle of teeming humanity, this addition always necessarily involves the breaking-in of new greenfields land. Developing nations who are failing to do this, are often trapped in a culture of corruption where the well-connected, including incumbent land owners, siphon all the public money off to their own benefit, or at the very least, rig the development game to ensure that “housing” remains tight and high-rent.

        6. Yes Brussels has quite unusually high congestion for a city of its size (slightly bigger in terms of population than Auckland). Why → it’s complicated.

          Brussels is on a flat plain. Didn’t save it from getting congested.

          The tunnels are in various state of disrepair, quite a few have been closed recently for long periods. There’s talk to close some of the tunnels since a city of that size doesn’t need that much tunnels.

          Congestion pricing, well, lots of blah blah, but nothing is happening.

          There’s a few other factors why congestion is so bad over there:

          Spatial planning in the hinterland is uniquely poor. You can easily spot this even on satellite images. If you look at the Dutch countryside you’ll see villages with paddocks in between. In Belgium, villages are barely recognisable, because all the roads between villages have also been built up with houses. This creates large terminally car-dependent areas, with a lot of people who do a 60+ minute commute each way, and little hope of ever doing so faster on PT.

          Finally there’s a lot of employees in Belgium who get a company car. Not because they need it, but as part of their remuneration package. That’s a consequence of a poorly balanced tax system (the value of that car is way undertaxed compared to actual euros). Until recently many of them also got unlimited fuel with their car, but that at least is less common now. But still, for a lot of people in those traffic jams there’s exactly $0 difference in cost between driving 1km this year, and driving 15,000 km this year.

        7. Population density in Manhattan started to increase again in 1980 having declined from 1910 as you say Phil. In Tokyo prefecture, population and therefore density has been increasing continuously since WW2.

          The real interesting fact though is people’s demand for personal residential built space is a normal good i.e. as incomes rise then they consume more personal space. So for instance, in Tokyo prefecture since the 1960’s population has increased from 10 to 13.3m, at the same time, as residential built space has increased from 15 to 32sqm per person. The graphs etc for these facts are in my spatial economics article
          https://medium.com/land-buildings-identity-and-values/successful-cities-understand-spatial-economics-95c272ac04c9

        8. Yes, Manhattan and Tokyo Prefecture have reversed the population decline that prevailed for decades, and so indeed have many other western cities cores. But not one of them has caught up back to the circa-1900 level of density, and the examples that are closest to doing so, are exceptional cities with exceptional local economic evolution rather than a norm that any city can aspire to by planning.

          Even in the case of Manhattan, Ed Glaeser published an essay a few years ago, “Wall Street is Not Enough”, warning that even a powerful global-finance agglomeration could one day go the way of Detroit’s motor industry, and greater economic diversity would be wise. And Manhattan has started to fall into the economic rent trap that bedevils every city where the absence of competitive land supply, results in urban sites becoming a speculative commodity like gold. Auckland should have quickly learned this lesson already; site owners are sitting on them expecting more capital gains, just like gold hoarders don’t sell the product to the Plating, Dentistry, Electrical, and Jewellry industries at any price lower than where the latest speculative bubble has taken it. So the resource, land, remains underutilised, and purchases for actual development of it becomes far too expensive and risky. Manhattan is suffering from a similar paradigm now that all the competitive-price rural land is dozens of miles away and locked up in “rural character preserving” municipalities. So I would be surprised if Manhattan maintains its trajectory.

          Tokyo has the incredible, pure-genius advantage of multiple, competitive Integrated Transit and Property Development enterprises, who compete fiercely with each other for tenants, trip attractors, and ridership. That means low rents around stations, as well as low fares. People who want to save the planet with public transport and density should wake up to the only “market” structure that is “delivering”. In western cities, private property owners gain wealth transfers commensurate with the level of subsidies to PT, charge higher rents and reap greater capital gains even if they don’t bother to redevelop and provide more floor space for more PT riders. In Hong Kong there is an Integrated Transit and Property Development enterprise but it is a government monopoly, and price-gouges on property rents like any good capitalist would. In HK, this has the upside that it is government revenue that substitutes for taxes that would be higher otherwise. Libertarians talking about “small government” and “low tax revenue” in HK should stop omitting the “size of government and its revenue” in property and transit.

        9. hmmm yes good point re using the geography to assist you with “grade separation”, we probably overstate the advantages of flat cities.

Leave a Reply

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