Urban policies, economists, and anchoring bias

(Warning: this is a bit meta.)

As an economist, my job often boils down to helping people think clearly about issues. For example, the essence of good cost benefit analysis is being able to categorise stuff correctly – i.e. separating costs from benefits – and making sure that things aren’t being counted twice.

It’s surprising how often people get these things wrong. For example, it’s common to see politicians argue that megaprojects are a good idea because the government will employ people to build them. But if you think about it, government spending isn’t a benefit – it’s actually a cost that we bear in order to obtain other benefits! Unless the economy’s in a recession, those people probably would have been doing something else useful instead, like building houses.

As a result of stuff like this, I spend a fair amount of time thinking about cognitive biases, which can make it difficult for people to clearly think through issues. (Economists aren’t really any better or worse than any other group of people when it comes to cognitive bias. Training in statistical methods is a bit better, which helps in some areas but not others.) I encounter several cognitive biases on a regular basis: framing bias often affects (afflicts?) project scope; confirmation bias influences many literature reviews and data analyses; and fundamental attribution error is probably one of the largest causes of interpersonal conflict.

A project I was recently working on got me thinking about another type of cognitive bias: anchoring bias. Here’s Wikipedia’s summary:

Anchoring or focalism is a cognitive bias that describes the common human tendency to rely too heavily on the first piece of information offered (the “anchor”) when making decisions. During decision making, anchoring occurs when individuals use an initial piece of information to make subsequent judgments. Once an anchor is set, other judgments are made by adjusting away from that anchor, and there is a bias toward interpreting other information around the anchor. For example, the initial price offered for a used car sets the standard for the rest of the negotiations, so that prices lower than the initial price seem more reasonable even if they are still higher than what the car is really worth.

As this example suggests, anchoring plays an important role in negotiations. The initial proposal becomes the benchmark for subsequent suggestions. As a result, it makes sense for people to spend time and energy trying to “anchor” a discussion at a point that’s beneficial to them.


Some people are more anchored than others (source)

But what happens when two people are “anchored” at different points?

Rather than talk about my project – my policy is not to blog about things that I’m working directly on – I’ll use an example from Transportblog. Last month, I wrote a couple posts on publicly owned golf courses (1, 2, 3). In them, I made an argument that would be convincing to an economist: that the benefit of converting some golf courses to a mix of housing and public parks would exceed the benefit of retaining them in their current use, and hence we should consider doing so.

The posts led to a pretty lively and engaging discussion. Of course, not everyone was convinced and some people were angry about the idea. Fair enough. The interesting thing, in a way, was that different people’s “anchors” were so far apart.

As an economist looking at the topic, my “anchor” was the principle that resources are scarce and should therefore be used efficiently, taking externalities into account. (In this respect, being an economist is very congruent with being an environmentalist.) When an analysis showed that the golf course land could be used much more efficiently if it were used for something else, I was comfortable with the result.

An analogy could be drawn with congestion pricing, another policy that feels right to economists but is rarely put into practice due to fear of public opposition.

However, others had a different anchor point: a feeling that the status quo is basically good and shouldn’t change (too much). Several people expressed the view that open space in cities was valuable and, where it exists, should be preserved in perpetuity. Here’s a representative comment from one of the posts:

Never destroy public green spaces for housing. Plain stupid. Once gone, never get back. You can play games with fancy layouts and beautification till the cows come home – but once the core foot print is gone, its gone.

CAA NW Cycleway Chamberlain Park

Chamberlain Park from the outside. Source: CAA

These commenters were willing to contemplate some changes to publicly owned golf courses. A number of people suggested that Auckland’s golf courses could be converted into public parks or spaces for other outdoor activities, either now or in the future. Ben Ross, for example, wrote up a proposal for converting Chamberlain Park to an “urban forest”. However, actually developing some of the golf course land seemed to be too far away from the anchor point.

So what can we make of this?

First and foremost, anchors are important – people are attached to them and don’t like to be dragged too far away from them. Once anchored, it’s really easy to reject ideas that seem too far out – even if they would objectively be better.

I suspect that anchoring goes a long way towards explaining the shape of urban policies in general. For example, revisions to urban planning rules seem to happen incrementally, even in cases where very strong and comprehensive evidence for changing rules has emerged. Slow changes to bad policies can’t be explained by analysis – surely more rapid change would be better! But it starts to make sense once you account for the fact that people are anchored on the current plans and the current places.

Second, sometimes it’s really important to cut loose from the anchor of the status quo. If you’re an urbanist in a city like Auckland, you can’t be too attached to the way things currently are. That’s because you’re constantly going to be looking at places like this:

Lower Hobson St

Or this:

Albany beauty

And wondering whether this is really the best we could do, or whether we could live somewhere a bit more like this:


San Francisco. A good example of high-amenity midrise development in a hilly city. (Wikipedia)

Or this:


Amsterdam. A fine example of a humanised waterfront. (Wikipedia)

Which leads to a third point, which is that debates between people “anchored” to the status quo and people suggesting changes are always hard. The starting points are just so very different. But these debates are necessary. They’re a key part of building a case and a constituency for change.

People tell me that one way of overcoming such divides is to give people a picture – or better yet, an experience – of an alternative anchoring point. That’s essentially what Janette Sadik-Khan did as New York City’s transportation commissioner – put down cones, paint, planter boxes, and chairs to give people a chance to play with a different vision of the city:

I’m not a very visual person, so I tend to make quantitative rather than visual appeals for change. But despite that, I suspect that our ability to change in a better direction depends upon our ability to visualise, depict, and trial a better alternative. A pretty picture is a more appealing anchor than a large number!

36 comments to Urban policies, economists, and anchoring bias

  • Thanks Peter; wonderfully lucid.

  • Peter H

    We need to ask what a good anchor is for a city centre.
    Cities in English speaking countries tend to anchor there city centres around business.
    Medium size European cities centres tend to anchor on high residential areas, I’m thinking most don’t have a CBD.

    • Peter Nunns

      Not quite what I meant by “anchoring”, but…

      I’m not really that familiar with European cities, but older medium-density mixed-use areas of new world cities might offer a useful model. They typically combine residential and business use in close proximity – multi-storey offices and shops along a main road, and medium-density residential areas interspersed with the odd cafe, bar, or gym along the parallel streets. Sometimes you can also find the leftovers / heirs of old industries – joinery factories, panelbeaters, craft brewers – in pockets.

      One issue we have in NZ is that we’re probably too insistent upon separating residential and commercial uses. I can see why separating out heavy industry is a good idea from an externality perspective though.

      • Yes the need to separate Heavy Industry from people has absurdly led to a whole machinery of zoning which is deeply counterproductive. Our best suburbs are the ones where we have inherited a jumble of uses from the Victorians who first made them before zoning. I’m afraid this is one area where the planning profession have got into over reach. Other than the placement of aluminium smelters etc, almost everywhere should be considered mixed-use as a default. Homogeneity is the enemy of propinquity, and propinquity is urban value; is agglomeration.

        This is not the only area where the planning profession is in overreach. It is telling that their response in the PUAP process to the clear need to remove the absurdity that is minimum parking regulation is to come up with a whole array of complex and controlling mins and maxes…. just ditch the whole thing already; let the wild west sort it out. Or would lead to too much red tape disappearing, and too little control and work for planners?

        • Frank McRae

          Great article Peter. This is the problem – the planning profession in New Zealand is way too focussed on protecting the status quo. Many in the profession see themselves as gatekeepers, protecting the city from change without questioning what is so worthwhile about the current state of things or without thinking of how things could be fundamentally different. Often the only way developments get through the planning system is by proving that they won’t alter the current state of things too much. This is why a major barrier to development is proving that it won’t alter the current level of access for cars. Little thought is given to whether that level of car access is something worth protecting at all costs.

      • Yes, it’s that way in Europe too. Mixed use is what you’d expect in cities.

        Talking about Europe, density is a relative thing. If you look at those tramway era suburbs, they typically have 500 to 600m² sections. That’s very low density compared to suburbs this close to the city in Europe. There you’d usually have maybe 200m² sections.

        And the streets over here are really wide. All of them. I lived in a couple of places in Belgium, and it would be very unusual to see a street wider than 6 to 7 metres, except for the arterials. Most are too narrow to fit 3 cars side by side.

  • Max

    Nice. As a fresh graduate, I helped design the “dumbbell” or “not-quite-double-roundabout” you show up there. It’s not what I want for my city though! So colour me un-anchored to that stuff.

    Yet when working on large new developments, even those where I actively work to reduce car traffic, I come up against the old paradigmas (even reduced car use is still a lot of car use with a new development), and doing halfway things that I wish I could avoid. Growing a city is hard. Growing one away from the car even more so. But we gotta do it.

    • john smith

      I assume you you mean you were in the relevant organisation and watched it happen. I’d welcome any more comment you can make on how it did end up that way.

      Watching urban planning from the outside, I’ve always been mystified by how a well-established profession, full of no doubt honest, well-meaning, appropriately credentialled people, still manages to produce such relentlessly mediocre results.

      What type of people designed the dumbbell? What were they thinking? What were the approval processes that let it through? I really want to know. I try to imagine what it would be like to be a fly on the wall at their planning meetings, and I come up blank..

  • Warren S

    Peter – Thank you so much for the aerial view of the Albany roadworks. The overhead view is the most intriguing view I have ever seen and I can imagine how the original conceiver would have been so proud of his work, as he sat admiringly at his desk.
    The reality however, is that the shapes don’t translate or work properly at ground level and the whole plan is a disaster of monumental proportions – a commercial success – maybe, but a visual and practical disaster especially when you want to move from one area to the other using one’s own legs.

    • Peter Nunns

      Apparently you can blame Max for those roundabouts! See the next comment up.

      • Max

        Only for the development part (i.e. in the mall). I didn’t design the roundabouts of Albany itself – those had been around for a while already…

        Similarly, the car-centric nature of Auckland, and the transport model culture (with nary a pedestrian aspect in the traffic models) – where Councils and NZTA require developers to show their extra traffic can be accommodated, or ask for larger intersections to be built, at pain of refusing consent otherwise. These days, Councils require more attention to pedestrian and cyclist designs than 8-10 years ago when this was done, but models still reign.

    • mfwic

      “I can imagine how the original conceiver would have been so proud of his work, as he sat admiringly at his desk.” -or maybe proud of her work… at her desk.

  • Ross

    Nicely written article. Are you sure though about the example used for anchoring? You suggest that a negative response to your previous blog was the result of anchoring, yet there is nothing to suggest that ‘maintain the status quo’ was the first piece of opinion the respondents had come across. Isn’t there a difference between an opinion influenced by the first piece of data seen and an opinion influenced by a strongly held, and possibly well debated, opinion?

    • Peter Nunns

      I had a much clearer example, but I can’t really write about it as it’s something I’m doing at work.

      For what it’s worth, I didn’t write this post to complain about a negative response to something I’d previously written. I think people are entitled to their views, even if they don’t agree with me. (But they are _not_ entitled to reject empirical evidence just because it contradicts their views.)

      My aim in writing was to _explore_ the dynamics of how discussions about change might proceed, using the response to the post as a case study. I think there’s a lot to learn.

      • Ari

        Great post.

        I think the only way NY got so much change is that they had lots of nice one way roads to narrow. Not so easy in Auckland with few one way roads. Of course someone has to still push for the radical change in the first place.

        I am entitled to reject empirical evidence. It is called democracy. 🙂

        • john smith

          Yes, you’re entitled to reject evidence in favour of strongly held prior beliefs, and that’s what most people do most of the time because it’s cognitively easier.

          The problem is that doing so often leads to negative outcomes, because nature can’t be fooled. You’re entitled to believe that garlic cures cancer. But then you die.

          So it would be good if folks who are a bit ahead of the pack could gently encourage others to be more open to evidence and less driven by strongly held priors. The post discusses how to do that.

  • Excellent. I really enjoyed reading this. I imagine that so much of the inconsistency in decision making in areas of planning (where there is ambiguity) can be boiled down to this.

    Also, when I read ‘congestion pricing’ my immediate thought was ‘congestion charge’ which I fear nobody is willing to do due to the short term voting impact of public opposition, rather than the direct opposition itself. If done well, the initial outcry may just be outweighed by the impact, but someone would need to move their anchor away from ‘staying at the top’.

    • Peter Nunns

      Thanks! For what it’s worth, I suspect that congestion pricing would prove to be more popular than people think – as long as people feel like they have alternatives, such as travelling at off-peak times or taking other modes.

  • Jackson

    Quite disingenuous to compare Pacific heights in San Francisco to a shopping mall in Albany. A more accurate comparison would be Rockridge in Oakland but why let the truth stand in the way of another anti car anti suburban story.

    • Dave B (Wellington)

      How is this article “anti-car”?
      What is it about “pro-car” people that they feel so threatened by any suggestion there may be better ways of doing things?
      Being anti-car and being anti car-overdependency are not the same thing.

    • Peter Nunns

      You’ve illustrated my point about anchoring bias quite nicely. Your response to the question “Why can’t we be like this?” is, basically, “It’s too different and unfamiliar.”

  • Lord Maths

    The problems emerge when specific decisions that are individually beneficial, however, result in an overall loss (as an economist Pete I’m sure you’re familiar with this)\\

    Converting public space to private space may be right for a specific case, but as an overall principle is leads to increased capitalism and so on, causing harm

    Same as paying terrorist ransoms. Of course $1m to get 10 people back makes sense, but the overall effect is to actually lead to more people being taken captive (and in every case, paying $1m to get them back continues to make sense)

    • Peter Nunns

      I don’t think your comparison between transactions between participation in a market economy and “paying terrorist ransoms” is apposite. Market transactions generally do not involve coercion or use of force.

      • Lord Maths

        The analogy is that a decision taken in isolation can be beneficial, but each decision creates a broader environment that actually creates a sub-optimal result. For example, while each transaction in a market economy might individually be beneficial, the end result of those interactions is a libertarian fantasy that is very far from beneficial. In this case, nx where both n and x are positive can actually result in a negative.

        In this case *any* decision that strengthens capitalism over/above public ownership may have immediate benefits but longer term harms in further strengthening the capitalist deathgrip.

        • Peter Nunns

          A negative social outcome from private decision-making will only arise if there are negative externalities associated with those decisions. For example, if I want to turn my house into a dance club it would benefit me (and my customers) while imposing negative externalities on my neighbours. In this situation it would be appropriate to prevent me from running a dance club, or require me to pay compensation to neighbours.

          I’d observe that there can also be _positive_ externalities. For example, people’s private decisions to locate businesses in a dense commercial area can foster agglomeration economies through, e.g., knowledge spillovers between firms. In this case, private decisions can result in broader social benefits, not costs.

          There is a large economic literature on externalities – causes, consequences, appropriate remedies – that I would recommend you investigate if you are interested in this subject.

  • Cognitive biases are behind most human activity. Food, water and shelter are birth rights, not something to be earned, but most people are conditioned to believe they must “earn a living”, and so devote their lives to justifying and qualifying their obtaining of what was always their right in the first place.

    Cognitive bias can indeed make it difficult for people to think through an issue. But it can also, first and foremost, cause people to overlook whether the issue is even real in the first place.

  • unaha-closp

    Second, sometimes it’s really important to cut loose from the anchor of the status quo. If you’re an urbanist in a city like Auckland, you can’t be too attached to the way things currently are. That’s because you’re constantly going to be looking at places like this:

    Your anchors for Auckland are the bottom side of an underpass and the entrance to a car park. I think those are very strange anchors to choose.

    And wondering whether this is really the best we could do, or whether we could live somewhere a bit more like this:

    The Coit Tower picture is very nice, but I don’t think any prettier than the views of Mt Eden.

  • Ted F

    Thank you for another interesting and thoughtful blog