Tim Harford, “What cities tell us about the economy“, Undercover Economist:

Jacobs is not the only person to argue that economic development may be profitably studied through a magnifying glass. A new research paper from three development economists, William Easterly, Laura Freschi and Steven Pennings, offers “A Long History of a Short Block” — a Shinohata-style tale of the economic development of a single 486ft block of Greene Street, between Houston and Prince Street in downtown Manhattan.

Easterly, a former World Bank researcher, is well known in development circles for his scepticism about how much development can ever be planned, and how much credit political leaders and their expert advisers deserve when things go well.

“Here’s a block where there is no leader; there’s no president or prime minister of this block,” he explained to me. Greene Street, he suggests, offers us a perspective on the more spontaneous, decentralised features of economic development.


The lessons of Greene Street? Getting the basic infrastructure right — streets, water, sanitation, policing — is a good idea. Aggressive planning, knocking down entire blocks in response to temporary weakness, is probably not. Predicting the process of economic development at a local level is a game for suckers. Most importantly, even a tremendous development success — the United States and, within it, New York City — is going to show some deep wrinkles to those who get in close.

Also from New York, here’s urban economist Ed Glaeser’s perspective on how that city could address its housing cost issues: “Build big, Bill“:

Mayor de Blasio’s focus on creating or preserving 200,000 “affordable units” over 10 years, many of which would come at the expense of market-rate units, offers the prospect of two permanently separate cities:

In one New York, the market-rate city, housing will be freely accessible to anyone willing to pay — but many will be priced out as increased affordability requirements raise building costs.

In the “affordable” New York, housing will be dribbled out unenthusiastically by lottery to lower-income renters by developers who have no incentive to treat them with respect but see them only as a cost of selling market-rate units that will be even more expensive to carry the subsidized ones.

The possibility of scoring a prized, below-market unit like the type the city creates through special carveout programs creates intense competition for these units; a flock of my graduate students once left class early to enter into the lottery for subsidized units in Boston’s new Mandarin Oriental.

Giving people a small shot of winning an affordable unit lottery is no substitute for encouraging the creation of a far more abundant supply of ordinary rental apartments. Radically upping that supply would leave far more New Yorkers able to rent something for a reasonable price on the free market, instead of allowing rents to continue to rise for most while a select few win access, via waitlists and lotteries, to a separate class of apartments at below-market rates.

Once upon a time, New York City understood this. In the early 1920s, we were building 100,000 new housing units, accessible to people across the economic spectrum, every year. As a result, the booming city remained affordable to working- and middle-class people like my grandparents.

How was this possible? For one, the city’s regulations back then were relatively modest, focusing primarily on safety and light. Thanks in no small part to that limited red tape, the cost of producing new units was relatively low, and — surprise, surprise — they got built.

In a similar vein, the San Francisco Bay Area Renters Federation asks: “How does adding expensive housing help the little guy?” They’ve put together a quite neat mathematical model to illustrate the dynamics of the market:

A central question of our housing debate is whether building new (expensive) housing protects existing low-cost housing, or destroys it. I have had the great fortune over the last year and a half to discuss this question with hundreds of people. What we all have in common is whatever side we take, we believe the relationship is perfectly obvious.

In an effort to clarify my own thinking, and double check that my mental model didn’t have any hidden internal inconsistencies, Davi Caetano, Phil Nova, Matt Lichti, Avi Flamholz, Jean-Ezra Yeung and I wrote a simulation of a housing market. Our initial conditions were 15,000 people and 10,000 housing units. Our intervention was adding 2,000 more “luxury” units.  What we found was that after adding luxury units, displacement decreased at every income level.


But who benefits from the new units? Is it only the higher income displaced people?

Let’s compare just the blue sections before and after the addition of the 2000 high-scoring units.


Displacement decreases in every income category, and completely disappears in the higher income categories. This is part of the result is particularly relevant to the conversation in San Francisco over whether to subsidize middle income housing, and whether to build “luxury” housing with abandon.

Renters who have very low and intermittent, or zero income – the elderly, people on disability, the temporarily unemployed – will need subsidies in order to afford to live in SF, or anywhere. That is uncontroversial. What’s phenomenal about the past few years in San Francisco is how many middle and higher income people are finding themselves displaced. Famous example: Kelly Dwyer. Dwyer works for the City of SF, and her husband is a firefighter. Nonetheless, they find they cannot afford the type of housing they want in San Francisco. Ironically, throughout her political career Dwyer has been an opponent of the rapid and substantial increases in residential housing capacity that would have enabled her to stay in SF.

Since we’re now in San Francisco, it’s worth reading Scott Weiner’s argument that “San Francisco should always have a subway under construction” in Medium. Weiner is a member of the SF Board of Supervisors, so his view carries some weight. Incidentally, while San Francisco is at the centre of an urban region of over 5 million people, the city itself has less than a million people. And it’s growing less rapidly than Auckland.

San Francisco is experiencing unprecedented growth. The city has 200,000 more people than in the early 1980s and 100,000 more than in the early 2000s. We are growing by about 10,000 people a year and are projected to add another 150,000 residents by 2040. We see the results of this growth on our streets every day, with more and more auto congestion and a harder time for our extensive bus network navigating the streets and meeting schedules. Indeed, Muni buses travel at the slowest average speed of any urban bus system in the country, at just over eight miles per hour on average.

We are working hard to make our surface transit system run more efficiently — reducing the number of cars on the road, increasing the number of buses and light rail vehicles, creating transit-only lanes and bus rapid transit lines. Yet, as important as this work is for our city, it isn’t enough. We need to move more transit underground, meaning we need more subways.

Subway construction is a critical part of the future of dense urban centers all over our country. New York, which is already served by an extensive subway system, is building the Second Avenue Subway to serve the east side of Manhattan. Los Angeles is expanding the Purple Line Subway, which will extend the city’s growing metro system from Downtown all the way to Westwood. Internationally, we have seen aggressive investments in subway systems pay off in places like Santiago, Chile, which has the second largest network in Latin America, after Mexico City…

If you read on, Weiner’s got some good ideas about institutional and funding changes needed to keep transport investment going.

Over at Washington Post’s Wonkblog, Emily Badger and Christopher Ingraham take a look at the most popular housing typologies in big US cities. A few interesting things pop out to me. Philadelphia and Baltimore are rowhouse cities (or terraced houses as they’re called in the UK and NZ) – pretty anomalous in the US situation. Los Angeles and Seattle have got a surprisingly large amount of large apartment blocks, but few rowhouses or small apartment blocks. And even some cities we think of as pretty sprawled out and subdivided – Atlanta, Houston, Denver – have a decent mix of apartments:

wonkblog most popular housing typology by city chart

However, all households aren’t created equal. A reader pointed me towards this disturbing article on rental tenancies in Auckland. (I rent, but I’m lucky enough to have a good landlady and a dry apartment. You shouldn’t have to be lucky to have a decent place to live!) Jess McAllen, “Rental nightmare: Are our tenants second-class citizens?” Stuff:

Damp spongy floors, peeling wallpaper and the smell of “mud” and “pond water” are recognisable descriptors to anyone who has lived with mould – which, it turns out, seems to be most Auckland renters.

Andrew King from the Property Investors Federation, whose group represents about 6000 landlords owning 22,000 properties, says that some of the power to stop it is in their hands. He says tenants often do things that encourage mould, such as not heating homes and drying clothes on clothes racks.

“A lot of tenants actually keep their curtains closed during the day. Both parties need to take responsibility for mould.”

Similarly, Housing Minister Nick Smith is proposing a $1.5 million education programme as part of his Minimum Standards bill to educate tenants and landlords about mould.

“The Green Party says mould is solely the responsibility of landlords,” says Smith in response to their alternative draft amendment to the Residential Tenancies Act of 1986. “But professional advice I have received is that mould largely depends on the way tenants are ventilating their homes.”

But the renters Sunday Star-Times interviewed, at least, have gone above and beyond their duties in ventilation – often spending lots of money on dehumidifiers and heat pumps – and find Smith’s comment patronising.

The dire conditions are making them sick, they say. In the case of one 20-year-old, mould destroyed the possessions of her dead mother.

Over on the other end of town, housing is a challenge for different reasons. In his NZ Herald column, Bernard Hickey reminds us that location is an important, if costly, “status good” for families: “The lure of the ‘Double Grammar zone’“:

Is there any more powerful phrase in the minds and spreadsheets of Auckland’s real estate chatterati than “double Grammar zone”?

Just the thought of those three words plastered across a real estate billboard is enough to make any red-blooded agent salivate. It’s also enough to make any property developer in the zones see stars and dollar signs. That’s why as many as 1800 apartments are expected to be built in the Auckland Grammar zone over the next three years.

But those three words and the thought of all those apartments is also enough to make the principals of Epsom Girls Grammar School and Auckland Grammar School go green at the gills.

Hickey looks at a number of ways that the desirable schools could accommodate growth. His last option seems quite elegant – in fact, Act actually proposed something similar back in 2003. Perhaps it’s time for the local MP to stand up for his alleged libertarian principles:

The final option is to simply turn the schools private. There is certainly plenty of demand for such assets. Private equity group Pacific Equity Partners is reported to be in talks to buy Academic Colleges Group, which includes schools in Auckland, for $500 million.

Perhaps the free market Act party could get behind such a public asset sale? That would be a double-riot option with a sideshow lynching of the local MP.

All the options are ugly, but are the inevitable result of creating a resource with public money that is highly desirable and virtually free to all those able to buy into the zones.

On a completely different note, here’s a fantastic TED talk by Swedish transportation researcher Jonas Eliasson. He’s looking at the success of Stockholm’s congestion pricing scheme:

Eliasson draws a good analogy between road networks and other complex social systems:

Planning a complex social system is a very hard thing to do, and let me tell you a story. Back in 1989, when the Berlin Wall fell, an urban planner in London got a phone call from a colleague in Moscow saying, basically, “Hi, this is Vladimir. I’d like to know, who’s in charge of London’s bread supply?”

 And the urban planner in London goes, “What do you mean, who’s in charge of London’s — I mean, no one is in charge.” “Oh, but surely someone must be in charge. I mean, it’s a very complicated system. Someone must control all of this.”

 “No. No. No one is in charge. I mean, it basically — I haven’t really thought of it. It basically organizes itself.”

It organizes itself. That’s an example of a complex social system which has the ability of self-organizing, and this is a very deep insight. When you try to solve really complex social problems, the right thing to do is most of the time to create the incentives. You don’t plan the details, and people will figure out what to do, how to adapt to this new framework.

According to Eliasson, even rather small congestion charges can induce a big change in behaviour and hence congestion. Incidentally, the fact that a 1-2 euro charge got 20% of traffic off the road during rush hours means that those rush hour trips weren’t very valuable to the people taking them – an insight that transport agencies should take into account when planning to expand peak road capacity at a high cost:

Stockholm is a medium-sized city, roughly two million people, but Stockholm also has lots of water and lots of water means lots of bridges — narrow bridges, old bridges — which means lots of road congestion. And these red dots show the most congested parts, which are the bridges that lead into the inner city. And then someone came up with the idea that, apart from good public transport, apart from spending money on roads, let’s try to charge drivers one or two euros at these bottlenecks.

Now, one or two euros, that isn’t really a lot of money, I mean compared to parking charges and running costs, etc., so you would probably expect that car drivers wouldn’t really react to this fairly small charge. You would be wrong. One or two euros was enough to make 20 percent of cars disappear from rush hours. Now, 20 percent, well, that’s a fairly huge figure, you might think, but you’ve still got 80 percent left of the problem, right? Because you still have 80 percent of the traffic. Now, that’s also wrong, because traffic happens to be a nonlinear phenomenon, meaning that once you reach above a certain capacity threshold then congestion starts to increase really, really rapidly. But fortunately, it also works the other way around. If you can reduce traffic even somewhat, then congestion will go down much faster than you might think.

Share this


  1. I too now have a very good landlord. House is warm and dry and he takes care of us as we take care of his house (note he is not a property investor, this is his old family home that he kept after moving to the regions). But I had bad experiences with mould before. In a previuos home a bought a 500$ dehumidifier, left it on 24h, opened the windows, heated up the rooms but the problem was the house was badly built. We had to sleep in the lounge for a month because the mould had completely enveloped the bedroom before the owner finally fixed the building. That place should have never been allowed to be rented for profit!!!! We don’t sell mouldy biscuits, why can we rent mouldy houses????

  2. “An insight transport agencies should take into account when planning to expand peak road capacity at high cost”. Or indeed peak transit capacity, particularly when it is being subsidised on the basis of positive externalties to road users.

    1. Yes and we already have the power to price the peak differently for Transit users as Transit use is already priced. Half price off peak, and something in between for the shoulders, would likely be very helpful in both spreading our peaky Transit load and incentivising more mode shift to that part of the network with plenty of capacity. And at very low to zero marginal cost.

    2. hmm … genuine question: Does low marginal WTP for vehicle travel at peak times necessarily imply low marginal value from increased PT travel?

      The former (low marginal WTP for vehicle travel at peak times) means that by definition there are a bunch of people (say 20%) driving at peak times that don’t value it that much. But this doesn’t tell us much about the WTP for the remaining 80%. So low marginal WTP for vehicle travel doesn’t necessarily imply low *average* WTP for reduced travel-times.

      Another way to look at it: Additional road capacity benefits the marginal road user, whereas increased PT usage benefits the average road user (assuming the congestion reduction benefits from the switch to PT are distributed equally between drivers).

      So I’m not sure this implication runs both ways, as you suggest.

      But correct me if I’m wrong!

      1. Ok my main point was that – even if building roads or transit yield positive net benefits based on the benefits to, amongst other things, road users, there is still potentially a better alternative – road pricing.

        However as to your interesting point. Yes I agree building transit does affect road users in a different way to building more road capacity. In an environment of no road pricing, there isn’t a functioning market, so we can’t talk about demand and supply in the conventional sense. But travel time is analogous to price insofar as you can imagine a demand curve sloping downward, with the y axis being travel time.

        Supply is a bit different. In the short term supply is fixed for roads, but there is a travel time supply curve to compare to the travel time demand curve. The supply curve is basically the throughput of vehicles vs travel time. You get increasing throughput with increasing travel time (more congestion), but it is highly non-linear. At low levels of throughput, the travel time doesn’t change much, so the supply curve is close to being horizontal, then as you start to get congestion travel time will start to rapidly increase with increasing throughput until the supply curve is vertical (and will actually start going backwards, but let’s leave that out for now).

        So, what effect on this “market” does building more road capacity have? It moves the supply curve to the right – more capacity for a given travel time. So what effect does this have on the total throughput and the “market travel time”? Well the demand curve slopes down so travel time should decrease and throughput increase. So we would expect, all else being equal, travel times to decrease when new roads are built and this will therefore benefit all road users. The amount by which travel times decrease will depend on the gradient of the demand curve.

        Now as for building new transit. This has the effect of improving alternatives to driving. So it should lead to a movement of the demand curve downwards. Again travel times will decrease and throughput will also decrease. So again all road users benefit, all else being equal.

        So I don’t think you can say that road building only benefits the marginal users while transit benefits all users – both types of investment should benefit all users. The question as to what is likely to be more beneficial will come down to the individual projects and the shape of the demand and supply curves, and I don’t know that much can be said about that generally.

        What this little exercise has brought to light in my mind is that when assessing transport benefits from a new project in a no-road pricing environment, you need to know something about the shape of these curves (in both the short and long term) to be able to make predictions about travel time savings. Is this done for the demand curve? I am not aware of it being done. I think demand is assumed to be exogenous which is intuitively wrong.

        1. that sounds reasonable.

          And yes from what I understand vehicle demand is usually assumed to be exogenous, except when dealing with larger projects that induce more significant land use changes. PT demand is more regularly assumed to be endogenous, especially where infrastructure investment is linked to more service (you wouldn’t run more service if you did not expect more demand). In my experience (not necessarily best practice) is that the endogeneous relationships between PT supply/demand is incorporated into analysis through the use of short (<12 months) and long run (<5 years) elasticities, where benefits to new (marginal) users are factored down due to the "rule of half".

          I think road projects may be more sensitive to this issue because the bulk of the benefits are associated with travel-time benefits, which are impacted significantly by 1) shifts in demand curve and 2) differences in average/marginal users value of time. In essence, by assuming exogenous demand we are over-estimating the benefits to average users. Of course endogenous demand will result in more new (marginal road users, but these will have 1) lower VoT and 2) be subject to rule of half.

          Which brings me to back to the original point: Experience with road pricing suggests there's actually quite a lot of "low value" vehicle travel occuring at the margin, even at peak times. And perhaps of greater concern is the fact that as roads become increasingly congested, then I would expect travellers with high VoT to be effectively "priced out" of the market, i.e. driven to find alternatives (location changes perhaps?). Hence, ironically congestion itself may drive down the average value of travel. And possibly result in people with high VoT needing to live closer to the city centre than they would do otherwise (i.e. they spend money on proximity rather than travel).

          My gut feeling is that road pricing is likely to be even more beneficial than it appears on the surface, because it 1) reduces the need for us to analsyse imperfectly functioning markets and thereby second-guess people's actual preferences - ultimately leading to better investment decisions and 2) it is likely to have spillover benefits for land use and labour markets, ultimately leading to more efficient urban land use structures.

    3. “..because traffic happens to be a nonlinear phenomenon meaning that once you reach above a certain capacity threshold then congestion starts to increase really, really rapidly.
      But fortunately, it also works the other way around. If you can reduce traffic even somewhat, then congestion will go down much faster than you might think.”

      As anyone here who experiences the sudden drop in road congestion every school holiday period can attest.

      And that is why investing money for peak transit is much better than fixing congested roads with ever more expensive solutions which invariably require massive subsidies.

      Because those “positive externalities for road users” mean making peak PT attractive to all, means road congestion is improved at the same time.

      And while the thought of subsidies for peak Transit sticks in your craw Matthew, its a way better ever more subsidised investment in yet another lane on the motorway for a twice daily peak.

      Even better would be to do as Stockholm did – penalise peak road users a little, and use the money gathered to further improve peak PT – a win/win for everyone.
      Except maybe motorway and road builders, and probably also Matthew’s belief in a “free” market solution?

      But free markets have their limits, its why we live in society after all, with things like taxes and laws so that externalities can be accounted for, if the governments so choose.

      That double grammer zone problem referred to is what happens when you don’t price “free” goods appropriately [or at all], the market abuses the situation to the detriment of almost everyone [except the apartment builders and those who manage to get one in the double grammer zone].

      and what would you propose Matthew to fix the problem with the looming “grammer zone crisis” “widen” the schools? reduce the zones? put in a cut off date for age of home for eligibility to the grammer zone or privatise the lot?

      1. See my reply to Stu above. I don’t think you can generalise about whether transit projects by definition will be more beneficial than road projects. And the point I think Peter was trying to make about road projects and I was generalising to all transport projects, was that using price signals to decongest roads is potentially, if not probably, a more efficient solution than expensive infrastructure investment.

        1. Nah, you didn’t say that its only relevant to “expensive” transit projects you said basically all transit projects that are subsidised to assist road users are bad – to quote you:

          “… Or indeed peak transit capacity, particularly when it is being subsidised on the basis of positive externalties to road users.”

          And where do you draw the line on “expensiveness” and “subsidised”? $1 from the public purse and/or $1 more than some road project? And what about the subsidisation for the RoNS that goes on? Since they save far less money than they cost they are always “subsidised”.

          Investment in PT projects is generally better because it benefits both the marginal PT user and the marginal road user, maybe not 50:50 but a road only project seldom enhances just PT even when its simply adding a bus lane.

          Because, the road users in that case get more benefit from having the bus out of their lanes than the PT users do and when its adding buslane ieven if its a half-arsed “bus shoulder” its invariably called a PT project not a roading project and is lumped with all the costs and seldom subsidised while the adjacent road widening project is fully subsidised from NZTA. And thats when its actually built, too often its merely a “future proof” footnote on some road project.

          As Patrick said, pricing off peak PT appropriately is important to ensure people see how expensive Peak PT usage is. The same should apply to roads. We have however a bunch of politicians too scared of that approach, so like the Grammer Zone we have a market free for all instead.

      2. As for the Grammar Zone crisis. If we are sticking with zoning, I would say that the schools should be allowed to expand, as any successful school should be able to if they can clearly demonstrate robust long term demand. A cut off date is fairly abhorrent. Privatising the schools seems arbitrary – it solves the problem by destroying it – ie the zone is overpopulated zone, so let’s remove the zone altogether. And why privatise some schools and not others?

        1. I agree, remove the zone is the simplest fix, but it will never happen, too many homeowners paid too much money already for an “in zone” property to ever allow that to happen.
          Why that would like allowing the intensification of Epsom.

          Think of that like the Thermo-Nuclear version of Skypath – you’ll have thousands of court cases for Africa clogging the courts for years if you try.

          So whats to be done?

          Expand, maybe, but who pays? Well that would be the Ministry of Education, but why should that be done at Tax payers expense to privilege those developers and the buyers of their apartments as the ultimate [and only] beneficiaries?
          The rest of us won’t be able to make use of that bigger school unless we pay the Grammer Zone tax to live in zone?
          Might as well put dedicated free Lexus lanes in to an adjacent Decile 10 schools instead, because at least off peak the rest of us could use them as freight lanes or parking or something and at least get something in return for our tax dollars.

          1. There is no problem to fix.

            Buying a house “in zone” is done by people who commodify their children or who are looking for status. They’re taking a risk. They know they take a risk by buying on a zone boundary. If it backfire? Well that’s what a risk is, isn’t it?

            Suggesting that AGS and EGGS should, when faced with a problem of roll growth and limited size to expand, resort to options that for any other state school would never be considered is sending a message. Considering any solution beyond “restrict the zone” is simply a case of saying, “Hey, these schools actually are better”. It’s stratifying. If you treat them like they’re not special (because, you know, they aren’t “special”) then you crack their mystique which has positive follow on effects for other schools (e.g. there’s a reason people consider “white flight” and other kinds of roll decline an issue).

            Zoning is a fundamentally good thing. Sure, you get strange results where some roads have one side go to one school and the other side go to There is no problem to fix.

            Buying a house “in zone” is done by people who commodify their children or who are looking for status. They’re taking a risk. They know they take a risk by buying on a zone boundary. If it backfires? Well that’s what a risk is, isn’t it?

            Suggesting that AGS and EGGS should, when faced with a problem of roll growth and limited size to expand, resort to options that for any other state school would never be considered is sending a message. Considering any solution beyond “restrict the zone” is simply a case of saying, “Hey, these schools actually are better”. It’s stratifying. If you treat them like they’re not special (because, you know, they aren’t “special”) then you crack their mystique which has positive follow on effects for other schools (e.g. there’s a reason people consider “white flight” and other kinds of roll decline an issue).

            Zoning is a fundamentally good thing. Sure, you get strange results where some roads have one side go to one school and the other side go to another school, but, fundamentally, you should not be allowing state schools to turn away someone who lives over the fence or within a ten minute walk because of some other qualities that don’t meet their ethos. If you let AGS have the opportunity, it would use its classification testing to simply admit everyone in order based on those tests: geography be damned. Your son could well end up spending an hour or more commuting to a school just because on one test/series of assessments he didn’t do well enough to get into a state school two minutes down the road. That’s insane and that’s what would happen.

            But, wait, it’d get worse. There are quite a few schools that are like AGS. They like to convince parents and people that they’re of a certain class or distinguishable by a certain something. All these schools, in the absence of zoning, would be like AGS and use entrance tests to exclude as well as classify entrants (despite what John Morris might try to convince the readers of the Herald, most schools stream their pupils). This means stuff because, as the “(double) grammar zone” phenomenon indicates parents do buy into this but also because test techniques actually do make a difference to how well you do on tests (and, on a related note, I guess so does exam doping).

            Removing zones is not going to lead to a more equitable world. It would, in fact, lead to the exact opposite (or, to the extent that the grammar zone means it exists today, merely preserve the current situation)] because the parents in a position to put their children into programmes designed simply to get their children into schools like AGS? Well, they’re not going to be the ones that can’t already afford to live in the grammar zone.

            If Grammar can expand, let it expand. If it can’t there should be only one option: restrict zones. A few people on the border aware that there are risks getting hurt is better than the alternatives.

          2. I think it’s probably important to just note the transport effects of removing zones. I don’t think that it is necessarily true that parents will try to get their children into schools based on what is closest to their places of work. Indeed, I think most parents would firstly consider what they largely do now: desirable schools. Second choice schools would likely be determined more by some sort of assessment of proximity to work. This could manifest itself as schools on the inhabited fringes experiencing falling rolls (this, generally, is a bad thing) and those closer to common workspaces with room to expand getting closer to their capacities (again, generally, bad). Or it could manifest more positively. Basically, transport could be better off without zones, worse of without them or, more than likely, not too different except the traffic flows have slightly different patterns. As such, it’s uncertainty is another reason not to do away with zones.

    1. It would be funny if it wasn’t so true.

      Belgium has the ultimate exponent of Motordom. A lot of employees get part of their salary in the form of a company car, because otherwise their employers would bankrupt themselves paying taxes. It’s estimated that over half of the cars in those traffic jams are company cars. And usually the company pays everything—maintenance, insurance, and unlimited fuel. It literally $0 to drive another 100km.

      (side note: despite that, it’s still a lot less hostile to pedestrians and bicyclists)

      And the other fatal problem is the lack of spatial planning. You have to see that for yourself to believe it. Just look at satellite images of the countryside. There are no towns, just long ribbons of houses along the roads. Thanks to the government accidentally subsidizing that practice, as they agreed to pull utilities to anywhere, no matter where you build.

      So the consequence, obviously, is that on said countryside it’s considered a rare luxury if your commute takes less than half an hour.

  3. I am uneasy about using a return period for death while cycling. Once every 8000 years sounds very safe and could lead us to think there is no point investing in improved safety. The point is not the return period it is that for those killed 100% of them died! I have only known one person killed cycling (hit by a milk tanker while going to school) and my great grandfather who I didn’t know who was hit by a truck when he was 72. So it must be quite rare. During my life I have met two people killed in car crashes so they must be rare as well. The point is it is low probability but bloody high cost. Return periods look like what we do to argue against spending. And as for any injury every 20 years surely is is higher than that? A cycle advocate once told me he was knocked off on average about once a year.

    1. What I liked about that tweet – without thinking too hard about the figures – was that it highlighted the significant difference between _perceived_ and _measured_ safety. My instinct is that crash reports are a quite poor source of information about the quality of street environments for people on bikes or on foot.

      For example, if an urban intersection hasn’t had any recorded car-pedestrian crashes, it might simply because it feels so unsafe that people detour to avoid crossing it, or simply avoid walking in the area. If I had the research funding for it, I’d quite like to send a bunch of volunteers out with heartrate or cortisol monitors to make a map of the most stressful intersections in the city. I suspect that it would yield some really valuable data for prioritising intersection upgrades.

      1. Sounds like an interesting idea. The problem with using severity of injury and fatalities with rare events is you are dealing with very small numbers. I got knocked of my bike by a bus in Newmarket at slow speed when I was 19. I didn’t report it and I was only a bit sore for a while so it would have been the lowest of there injury categories. But if the lady in the lane I fell in had been going faster it could have been any of the other categories. The severity was independent of the actual crash. To make sense of that you really need to aggregate and not get too detailed.

        1. For what it’s worth, NZTA’s Economic Evaluation Manual does recommend “scaling up” reported minor injury crashes and non-injury crashes to account for non-reporting issues. From memory, they estimate that there are 6 non-reported non-injury crashes for every reported crash. But you’re still dealing with a small and noisy population of reported crashes, which means that the results might be meaningless for any individual road.

          On the topic, you might find this amusing: https://www.youtube.com/watch?v=puK5CwThaq4

Leave a Reply