14: Learning from history

I was recruited to a post that was the start of my urban modelling career in the Autumn of 1964 by Christopher Foster (now Sir Christopher) to work on the cost-benefit analysis of major transport projects. My job was to do the computing and mathematics and at the same time to learn some economics. Of course, the project needed good transport models and at the time, all the experience was in the United States. Christopher had worked with Michael Beesley (LSE) on the pioneering cost-benefit analysis of the Victoria Line. To move forward on modelling, in 1965, Christopher, Michael and I embarked on a tour of the US. As I remember, in about ten days, we visited Santa Monica and Berkeley, Philadelphia, Boston and Washington DC. Having set the context, this finally gets me to the point of this piece: we met a good proportion of the founding fathers – they were all men – of urban modelling. A number of them influenced my thinking in ways that have been part of my intellectual make-up ever since. These threads can easily be traced in my work over the years. An interesting question then: for those recruited in subsequent decades, what are the equivalents? It would be an interesting way of writing a history of the field.

Jack (I. S.) Lowry was working for the RAND Corporation in Santa Monica where he developed the model of Pittsburgh that now bears his name. I recall an excellent dinner in his house overlooking the Bay. His model has become iconic because it revealed the bare bones of a comprehensive model in the simplest way possible. Those of us involved in building comprehensive models have been elaborating it ever since. The conversation for me reinforced something that was already becoming clear: the transport model needed to be embedded in a more comprehensive model so that the transport impact on land use – and vice-versa – could be incorporated.

The second key proponent of the comprehensive model was Britton Harris a Professor of City Planning at the University of Pennsylvania but, particularly important in the context of that visit, he was the Director of the Penn-Jersey Land-Use Transportation Study. The title indicated its ambitions. This again reinforced the ‘comprehensive’ argument and became the basis of a life-long friendship and collaboration. I spent many happy hours in Wissahiken Avenue. The Penn-Jersey study used a variety of modelling techniques, not least mathematical programming which was new element of my intellectual tool kit. More of Brit later. At Penn – was it on that trip or later – I met Walter Isard, a giant figure in the creation of regional science and who contributed to my roots in regional input-output modelling. Walter was probably the first person to recognise that von Thunen’s theory of rent could be applied to cities – see his 1956 book ‘Location and the space-economy’. Bill Alonso was one of his graduate students and he fully developed the theory of bid-rent. We visited Bill in Berkeley and I recall a letter from him three years later, in the heady days of 1968, starting with ‘As I write, military helicopters hover overhead ….’! Then back to Penn. For me it was Ben Stevens who operationalised the Alonso model in his 1960 paper with John Herbert – as a mathematical programming model. This fed directly into work I did in the 1970s with Martyn Senior to produce an entropy-maximising version of that – making me realise that one of the unheralded advantages of that method was that it made optimising economic models – like the Alonso-Herbert-Stevens model – ‘optimally blurred’, to recognise sub-optimal real life.

In Harvard, we met John Kain, very much the economist, very concerned with housing models – territory I have failed to follow up on since. A new objective! He was at the Harvard-MIT Joint Centre for Urban Studies whose existence was a sign that these kinds of interdisciplinary centres were fashionable at the time – and have been in an out of fashion ever since – now fortunately fashionable again! An alumnus was Martin Meyerson who by this time was Chancellor of the University of California at Berkeley (and we dined with him in his rather austere but grand official dining room – why does one remember these things rather than the conversation?!). There also was Daniel (Pat) Moynihan who had just left the Centre to work for the President in Washington – another sign of the importance of the urban agenda. I was urged to meet him and that led to my only ever visit to the White House – to a small office in the basement. Of course he later became very grand as a long-serving Senator for New York.

The Washington part of our visit established some other important contacts and building bricks. We engaged directly with the transport modelling industry through Alan Voorhees – already running quite a large company that still bears his name. It was valuable to see the ideas of transport modelling put to work and I think that reinforced my commitment that modelling was a contribution to achieving things – the use of the science. We met Walter Hansen who was working for Voorhees who was probably the inventor of the concept of ‘accessibility’ in modelling through his paper ‘How accessibility shapes land use’ and T. R. (‘Laksh’) Lakshmanan of the ‘Lakshmanan and Hansen’ retail modelling paper – other critical and ever-present parts of the tool kit. From a different part of the agenda, there was Clopper Almon who was working for the Government (as I remember) on regional input-output models.

Much of what I learned on that trip has remained as part of my intellectual tool kit. Much of it led to long-standing exchanges – particularly through regional science conferences. Some led to close working collaboration. Brit and Walter between them, ten years later, recruited me to a position of Adjunct Professor in Regional Science at Penn where I spent a few weeks every summer in the late 70s. I worked closely with Brit and those visits must have been the basis for my work with him on urban dynamics that was published in 1978 – and is still a feature of my ongoing work plan. I could chart a whole set of contacts and collaborations for subsequent decades. Maybe the starting points are always influential for any of us but I was very lucky in one particular respect: it was the start of the modern period of urban modelling and there was everything to play for.

Alan Wilson, May 2015

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s