Book Review: Order without Design

Several years ago I read a short article written by urban planner Alain Bertaud. It explains how, one way to help understand how cities function, is to think of them in terms of labour markets. It’s a simplification, but I found it helped focus the mind and was quite a neat way to boil down something extremely complex into an idea that is simple enough to be comprehensible.

I only found out a few years later that the article was an excerpt from his book, Order without Design. This year, almost a decade after reading the article, I finally got around to reading the rest of the book.

The first few chapters were great, talking quite generically about cities and the way they function. But I also started spotting problems in his thinking which I hadn’t noticed ten years ago, mostly related to his thoughts on transport.

The overriding theme of the book is that cities would operate better if planners just let them grow organically rather than trying to impose their own regulations on them. He argues that planning tools like height limits , zoning, and urban boundaries interfere with the free market. If everyone was given more freedom to build whatever they want on their land, live however they want, work however they want, then they’ll make decisions that are best for them and overall society will be better off.

The mayor, with his team of municipal mangers, is not the city’s ruler, nor is he the city’s designer. A city is entirely created by its citizens’ initiatives.

To me this is a solid place to begin, and we are starting to see some of this market-first philosophy manifesting itself in NZ planning reform. For example the recent NPS-UD and MDRS aim to remove unnecessary regulation and allow housing markets to operate much more unrestricted than previously.

My hunch is that this is probably not a bad step, but you don’t have to look at cities for too long before it becomes obvious that there are market distortions all over the place, which will need to be addressed for it to work properly. Chapter 5 is all about mobility in cities, and in my mind that’s where the logic really starts to disconnect.

Transport has especially large unpriced externalities, which prevent the transport market from functioning properly. Bertaud acknowledges this.

Urban transport is different from other consumer products because its users pay only a part of its cost. Car users pay a market price for their car and the gasoline they consume (in most countries), but they are usually not paying for the public road space they use, or for the pollution, congestion, and other costs they have imposed on others.

He talks a bit about air quality, and how there’s not really any pricing mechanism that could be used to fix that distortion. Therefore a second-best option involving government regulation needs to be used instead (e.g. clean car standards).

He talks similarly about carbon emissions. Ideally some sort of pricing could be implemented, but in the absence of that then government regulation is the next best option (the book was written a few years ago before widespread carbon pricing).

And he talks about congestion. Again pricing is the ideal option, but it’s very difficult and therefore not widespread. So again other government policies and regulation has to be used instead.

But he doesn’t translate this back into the earlier chapters and realise a lot of the interventions he derides (like urban boundaries) are in fact the “second-best” regulation that is made necessary by the lack of pricing mechanisms.

He uses this graph lots to explain his points. It shows land values by distance from city centre in a hypothetical uniformly circular city (monocentric city model). The edge of the city is the point where land values drop to the same level as the surrounding rural land.

He acknowledges (but never graphs) that if transport is subsidised (which he states that it is) then the curve will be artifically stretched horizontally. If you corrected that by, say, implementing congestion pricing, then the curve would contract horizontally, like my crude pencil mark above.

You can see that the edge of the city would be several kilometres closer in than it is with subsidised transport. To me that’s the justification for an artificial urban boundary: it’s acknowledging we have a market failure (unpriced road space), noting that hopefully one day we’ll fix that properly (with congestion pricing), and in the meantime it doesn’t make any sense for councils to invest in costly infrastructure out on the margins where we know development isn’t economic, it’s only being pursued because of our temporary failure to price roadspace. This is also the justification for lots of other planning tools: transit oriented development, upzoning around public transport routes, etc. One day we won’t need any of these because we’ll price transport properly, but that day is not today.

All up though, I’d still recommend the book if you work in this field (or just find cities fascinating things). It’s pretty long though (400 pages) so if you don’t have time just read the first few chapters (I wouldn’t rush to read the chapter on mobility). At a minimum I’d highly recommend spending 30 minutes reading the chapter linked to above on cities as labour markets.

As a sidenote I recently attended a short course on urban economics, taken by Drs Shane Martin and Stuart Donovan. The content of this course aligned pretty strongly with most of the book. One area they were were more nuanced in was the way they discussed transport and the shortcomings of the current set up. I would also highly recommend that course to anyone interested in this stuff.