Urban Planners need to understand land rent like rocket scientists need to understand gravity. Why don’t they?
This essay is an edited version of a posting on a specialist forum called “The Transportation Modelling Improvement Program”. I have introduced a line of argument many times to that forum, regarding the commonly used computer models in urban planning and transport planning; that they fail to include major invisible “forces” in urban evolution, particularly economic rent in land. I argue this is akin to trying to do rocket science without taking gravity into account.
The “science of forecasting”, as explained by Scott Armstrong, Kesten Green and others, holds that as long as the forecasters are constantly having to readjust their computer model to account for wrongly predicted outcomes to date, any predictions of the future should not be regarded as “scientific”. Unfortunately, the modelling of urban policy and transport policy outcomes very much falls into this “unscientific” category. But besides the need to show a track record of successful predictions, if a “computer model” is to be of practical use, it should also be possible to apply it to any city. Known current statistics can be inputted and known current outcomes replicated, within essentially the same mathematical model. Any city planning department should be able to purchase the same model, and put it to use.
This is certainly not the case. Every computer model so far, has been subjected to massive “tweaking” to harmonize inputs and outputs uniquely to the city that is being studied. And this tweaking is always ongoing; and correct predictions of outcomes of changes in this or that policy; zoning, new transit routes, new highways, congestion pricing schemes, etc remain elusive. Armstrong and Green and colleagues would feel confirmed — the future cannot be modelled when the “causes and effects” of the past are not understood anyway. In reality, urban economies and their societies are too complex to be able to “model” with computers. But the following are some suggestions of obvious “big” factors in urban-system evolution that strangely, are little understood by policy makers and their “expert” advisors. Shallow assumptions and existing entrenched “worldviews” that are straight-out wrong, tend to predominate.
Cities, as opposed to rural areas, enable more activities, particularly economic interactions. Whatever planners and policy makers are trying to do, they should understand this first and foremost. The McKinsey Institute has identified exceptions to the “law” that bigger cities are more productive than smaller ones. It looks to be an important factor that a lack of “mobility” can erode the anticipated productivity benefits. On this point, Alain Bertaud’s recently-published book contains some powerful insights. For example:
“…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 ensure high productivity in the absence of mobility…”
Once an urban area grows beyond a certain size, its patterns of economic interaction tend to become more fragmented. It is unrealistic to expect either a “monocentric” form with everyone converging on a single centre in which to interact as workers or consumers; or a “contained poly-centres” form where all the interaction is within each geographic segment of the whole. The reality is a chaotic web of interactions in all directions throughout the entire urban area. Each “agglomeration” is overlapping and in fact many “agglomerations” are near-invisible because of the chaos and the overlapping. Nevertheless there are obviously limits to the distance over which each individual contribution to that chaotic interaction can occur.
Bertaud’s central insight is that the greater the distance over which individuals are able to interact within accepted travel time budgets, the more productive the urban area; there are simply more invisible agglomerations in action. This holds true regardless of density; it can be argued that density can substitute for “distance able to be travelled”, but even when there is high density, productivity is foregone if distance able to be travelled is sub-optimal. Tory Gattis actualy calculated years ago that Houstonites are able to interact with the same number of people as Manhattanites within a given time frame.
Last week I tried taking Jane Jacobs' four tenants of vibrancy and applying them to the car-based city, describing the…
Furthermore, the extensive analysis of “urban scaling” done by West, Bettencourt et al, indicates a relationship between productivity and “size”, period; density can be present or absent without affecting the outcomes in the big data.
Density is Not the Issue: The Urban Scaling Research
The "urban scaling" research of Geoffrey West, Luis Bettencourt, Jose Lobo, Deborah Strumsky, Dirk Helbing and…
A core principle that planners and modelers need to understand is that “clustering” is not a “monocentric” phenomenon (and agglomeration economies include a lot more things than just physical clustering — communications, for example, has increasingly substituted for proximity. But the focus of this essay is the physical “cluster”). Clusters are of multiple types, with multiple differing inputs depending on the range of activities that happen to be “complementary” to the evolving “cluster”. Furthermore, the factors that affect the location of each type of cluster, vary — particularly such things as land cost (when high land consumption is required, eg for production lines), proximity to rail-heads, highways, ports, airports, raw resources, etc. Each evolving cluster has a kind of “magnetic attraction” to potential participants.
Unfortunately, there are also “magnetic repulsion” factors that hamper the evolution of clusters, which are multiple in number.
1) lack of availability of space for new participants, at practically affordable rents. This is especially an issue for “grassroots” new “industries” — for example, it is doubtful that anything like Silicon Valley could have evolved without lightly regulated, low-cost greenfields. “High Tech” sectors in high-cost cities all involve “old money” high tech, not new grassroots industry growth.
2) lack of housing at practically affordable rents or prices, for potential workforces, within effective transport distance
3) lack of mobility restricting the potential catchment area for participants and workforces
One of the most common shallow assumptions today is that “induced traffic” is bad, and is a reason to stop building road space. But surely “induced traffic” should be viewed as “evolving clustering effects”! Why would more people than before be going somewhere if there wasn’t some “magnetic” attraction at work? But planners and modelers tend to think too narrowly when it comes to “clustering” and the enabling of it, which is really what their primary duty, “enabling economic interaction” is all about.
First, they often overlook the reality of “what is required” for some cluster types. Urban Planning as it has been done in the UK since 1947, absolutely certainly has foregone a lot of clustering efficiencies and no-one even knows that anything has been foregone. Planners look at “what is” and think they have done “a good job”, perhaps because urban densities are high and transit mode shares are high.
Among all the western economies, the UK should receive major attention from urban economists and modelers. The Planning system has successfully perpetuated the highest urban densities in the western world. But the McKinsey Institute; Alan W Evans and colleagues; Paul Cheshire and colleagues; all have blamed the UK’s “productivity gap” on “what has been foregone” due to the urban planning system. Taking the three factors above: there is absolutely minimal “spare land” anywhere; land rent is higher than western Europe by a factor of around 30 and higher than southern USA by a factor of around 100; no housing options are “affordable” by the metric of “3 times household income”, no matter how crammed and minimalist the housing has been permitted to be; and a combination of high congestion and high taxes on fuels reduces the “radius” over which workers and economic participants can interact.
Worse, the higher that land rents are “everywhere” in an urban area, the more powerful the “pricing out” effect. That is, “efficient” or high-demand locations are affordable to far fewer potential participants than equivalent locations in “low overall land rent” cities like those in southern and heartland USA. Upzoning merely feeds into higher site rents, and perversely, the smaller that average housing-unit sizes can be shrunk by prescriptive planning, the higher the per-unit price (the opposite of common assumptions).
Therefore, the “benefits of density” as a “substitute for speed” are over-rated. “Pricing out” effects swamp the alleged benefit of “proximity” — a disproportionate amount of the workforces in cities like Hong Kong and London have monster commutes of more than an hour in each direction.
High transit mode shares are seldom correlated with short-duration commuting. Furthermore, they are seldom a “congestion reducing” device, contrary to the bad-faith ways in which transit “investment” is often sold to the public. The relationship between transit mode share and congestion, is that higher congestion incentivizes a higher transit mode share. If transit reduced congestion, it would be a victim of its own success (as Anthony Downs pointed out as part of his “triple convergence” principle). If providing more road space “induces more drivers to take that road space” then any drivers “taken off the road by transit” also will be replaced on the road by “induced drivers”.
It should also be noted that transit itself does not “induce riders” to the same extent as roads “induce driving”, at least certainly not in the presence of practical automobility alternatives. This is probably an inevitability of the respective “supply of land” enabled by each mode; the lesser supply enabled by transit will lead to more rent capture and extraction, whereas property-rentiers power tends to be reduced by automobility’s dispersion-enabling feature.
“What happens” in response to unaddressed congestion, and “behavioral nudges” towards transit, depends on a wide range of factors in the urban area as a whole. The most powerful response, if it is not disabled by prescriptive planning, will be relocation of economic interaction to elsewhere that economic rent is lower and to which travel speeds are higher. But the extent to which this happens depends very much on the “magnetic attraction” of the centrally located cluster that transit is a practical option for. Can this magnetic attraction overcome the magnetic repulsion of the congestion, the long trip times from “fringe” affordable-housing locations, and the tendency of economic rent to increase? In the case of “global city” or similarly unique centres of finance and bureaucracy and media and so on, the answer is “yes” for large numbers of people. But in most other urban areas most of the time, the answer will be “no”. The loss of manufacturing from central cities is a perfectly rational phenomenon. Likewise the attraction of suburbs for big-box retailing.
But what if planning is deliberately circumscribing this “escape” for economic participants to elsewhere than the central area, because the “objective” of “enabling economic activity” has been confused with “enabling economic activity in a single central area served by transit”? The outcomes of this kind of thinking institutionalized as the guiding principle, differ only from the oppression manifested under central planning in the former USSR, in “magnitude”; they are morally similar. The employment that many people might have had in suburban type-specific industry; the decent housing that many people might have had at affordable cost; the manageable burdens of daily travel time that more people could have had given greater choices of co-location of home and employment; foregone along with “multi” clustering efficiencies, productivity and increased incomes.
Planners who look at a city like London and credit themselves with “a job well done”, and look at a city like Houston with disdain, are intellectually trapped in nothing less than a combination of cargo-cultism and elitism. Cargo-cultism because of the implied assumption that planning like London’s “produces more Londons” (rather than humbly recognizing centuries of path-dependent historical accidents and the uniqueness of near-weightless primary income sectors) and elitism because who cares about the far superior conditions of housing and location options and career pathways and startup opportunities for the “bottom 90%” under Houston conditions?
Why say all this political stuff this on a forum dedicated to modelling? Because the modelling approaches we have so far, are not providing the connections between policy choices and real-life outcomes, that would justify policy choices other than those enjoying fad status at the current time.
I have recently published on Medium.com, a literature summary on the history of urban land rent, with particular attention to its relationship to macroeconomic cyclical volatility. But urban-economy modelers could read it and see if their approaches to modelling reflect any understanding of the historical evidence about the relationships between transport systems, “land supply for urban economies”, economic rent, and evolution of the modern highly-productive urban economy.
The History of Urban Land Rent and Cyclical Volatility: The Elephant in the Macro-Economic Room
The recent episodes of increased cyclical volatility in some property markets, and chronic worsening of housing…
The counterfactual for the ubiquitous “sprawling” and productive modern urban economy of most developed nations today (typified by the Ruhr Valley) is not “Manhattan”, it is the developing-nation cities identified by the McKinsey Institute as suffering a productivity penalty from lack of mobility. Along with this, the perpetuation of the “informal housing problem” in many cities is only to be expected if planning is centred on rationing land supply “to support slow transport modes” (and avoid the “mistakes” of the first world)! Bertaud ironically notes that the motor scooter (becoming afforded by most people now in developing nations) is at least enabling “informal housing” to spread out and hence become more spacious — even as the authorities officially “increase housing supply” in the form of apartment blocks on transit routes that nevertheless perversely remain unaffordable to most.