How can we plan transport when politics and budgets get in the way

When Alexander the Great consulted the Delphi Oracle, the duty Pythia, or priestess, told him that she didn’t want to see him and to come back another day. The story goes on that he dragged her by her hair out of the Oracle’s chamber into the daylight and only released her when she screamed (perhaps less politely than these words) “leave me alone, you beast, you are invincible”. He wasn’t, of course, so he died just like all those billions who don’t have “The Great” as a surname. At the time, though, he didn’t know that. He was happy with what he took as the Oracle’s prophecy rather than the words of a frightened woman being assaulted by a bully seeking affirmation.

From this story, I’m making my usual leap of analogy, likening strategic transport models to oracles. What’s more, Alexandrian treatment of our modern oracle-models is common. Since the 1990s decision-makers have dragged our models by their digital hair from the chamber of planning where they belong, into the harsh light of economics where they turn to ash in a terrible mixed metaphor. There the decision-makers force a validation of projects that are all too often little more than follies and cost a lot more than they’re worth.

Strategic transport models are nothing more than tools that planners use to understand how transport demands change in response to changes in urban structure and socio-economic conditions over decades. Their outputs are the result of the planner’s inputs and assumptions rolled around in some mathematics and sprayed out in a vast array of numbers. These utterings of our oracle-models need to be studied and interpreted carefully to formulate plans or to assess their value.???????? ?

The models range from back-of-the-envelope calculations to massively complex four-step or activity-based models, which are all, let’s face it, digital back-of-the-envelope calculations. But man, you should see the size of the envelope.

Antipathy towards strategic models has grown, perhaps because the role of models in the ill-judged toll road forecasting or stamping a benefit-cost ratio onto a piece of infrastructure in business cases. I always feel that their most strenuous critics don’t understand the role of models in transport planning. Strategic transport models are neither evil nor wicked, they don’t use “broken algorithms” and they don’t construct a fantasy world. They’re neither good nor bad; they’re ethics agnostic, to use modern weasel words. They’re not accurate or inaccurate, as this suggests that there is a fixed target in the future that we should aim at. They’re not wrong or right; there is no tangible,? absolute ethical standard against which to judge their forecasts.

Strategic transport models need to be used in good faith and for the purposes for which they are designed. That is, they need to be used for transport planning. The four-step model was developed to assist the planning transport for cities and regions, not for assessing the economic worth of individual items of transport infrastructure or for estimating how rich transport infrastructure may make their operators. Business cases are not planning exercises. Different models should be used for them, or the same models should be used differently.

To mark 50 years since the last comprehensive transport plan for Melbourne, Eric Keys , De Gruyter, Chris and Graham Currie FTSE - a trio of heavy hitters in Melbourne’s transport planning scene - wrote an article for The Conversation. They argued (correctly in my opinion, or I wouldn’t quote them) that transport models are not the problem in planning; political abuse of models is. They also point out that there has been no planning of transport in Melbourne for the past 50 years, which may be true of all the cities in Australia. This is easily demonstrated because the freeway part of the 1969 plan is still being implemented. Much of the proposed freeway network has been constructed, despite communal opposition to the construction of freeways. A new 10-km freeway link recommended in the plan is currently being constructed. At a price of $16b it is the most expensive transport project ever built in the state of Victoria. On the other hand, a few rail projects were recommended in the 1969 plan, but only one – the rail loop around the Melbourne CBD - was implemented. Mercifully, the plan recommended the retention of Melbourne’s tram system, unlike many earlier studies in other cities where trams were replaced by cheaper, but less popular buses.

Comprehensive transport planning was common in the 1960s and 1970s. Auckland, Wellington, Adelaide, Melbourne and Sydney, amongst others with apologies to those not in the South Pacific, forecast their transport futures and developed transport plans. They were not the first of transport plans in those cities, but they were the first in the dawning age of computers and transport models.

In each case, apart from Wellington, an extensive network of freeways or motorways was recommended. Public transport was recommended to various extents. In Melbourne only around 15% of the estimated transport budget was allocated to public transport, with roads and parking snaffling the remaining 85%. In Sydney, the split between roads and public transport was a little more balanced, with 60% of the proposed budget for roads, and 40% for public transport. And in Auckland in 1968, the focus was on freeways and other roads, but a “rail rapid system” was proposed. The consultant, De Leuw Cather, compared an all-bus network to the proposed rail rapid system. They wrote a separate memo to the Auckland authorities, warning that a bus only solution would not cut the mustard and would lead to an increasing flight of patronage from public transport and to increasing congestion on Auckland’s roads. The memo provided financial and economic arguments to show that although the rail rapid system would be expensive, it would pay for itself in benefits. In the end, Auckland built some of the proposed freeways but not the proposed rail. Residents of Auckland can judge whether their leaders were visionary enough in their decisions.?

In Wellington (1963), where four-step modelling was not used, a single motorway was recommended with changes to the town centre’s traffic flow.? The study suggested that the rail service was already better that Wellington needed and that the motorway should be built before an extension of the rail to the city centre could be considered. Most of the recommendations were implemented, after some angst about the cost of the motorway. The railway extension was never built.

In Adelaide, Bob Nairn's paper to the Australian Transport Research Forum (ATRF ) shows that none of the proposed plans of a 1968 Metropolitan Adelaide Transport Study (MATS) were implemented. Apart from freeways, the study recommended replacing the only operating tram with an expressway and conversion of the heavy rail network to a “rail rapid” transit network, with an extension of the rail line across the CBD via an underground tunnel. Although the sitting government supported the plan, the public reacted to the freeway proposals with some vigour. Political shenanigans followed, including catching the minister of transport trying to bias the output of a poll. After a change of government, the plan was abandoned. It seems to me, though, that one of the main recommendations of the Adelaide study, a north-south motorway spine, is currently being implemented.

The Sydney Area Transportation Study’s (SATS) documentation (links at the end of the article) is readily accessible on the web and includes detailed descriptions of the modelling. Frank Milthorpe compared the SATS forecasts for 2000 with the actual 2000 outcomes in a paper for the ATRF conference in 2005. To summarise, the outcome was a mixed bag, but a separate check shows that the model’s forecasts of traffic in corridors across the rivers in Sydney were within 10% of the counted volumes in 2000. Despite its flaws the modelling would have been useful enough to form a strategy for the future.

The trick to making plans and strategies more appropriate to the times is to review the modelling, forecasting and the strategy regularly. That was never done.? In 1975, the planners didn’t foresee the Sydney Olympics in 2000 and the surge of infrastructure investment it inspired. The Sydney Harbour Tunnel opened in 1992 and the Sydney Airport Rail Link opened in 2000, just before the Olympics. Neither were envisioned in the study. Employment forecasts did not foresee the collapse of the manufacturing industry in Sydney nor the rise of the IT industry. Regular reviews, maybe at 5-year intervals would have improved the planning outcomes.

Most interesting in the SATS documents, though, was not about the model. The study supported the implementation of tolls across the 500km of motorways recommended and provided financial evidence to show the revenue that could be collected. It argued that nobody should object to paying their own way across the road network. SATS was undoubtedly the seed of the current toll network in Sydney. It also notes the social discontent inspired by freeway proposals and construction, where there were never the same objections to the construction of railway lines. It then attributes the anti-freeway sentiment to misguided activists whose opinions don’t reflect the attitudes of the general public. A more logical reason may be that the community preferred railways to freeways.

These planning studies show that transport models are not the problem. There were problems with the “obvious” assumptions and inputs that went into the models. American companies did most of the work for these studies: Wilbur Smith Associates for Melbourne, De Leuw Cather for the rest. It seems to me that on of their implicit assumption was that these cities would be like American cities. Indeed, there is a statement in the SATS that the current dominant mode in Sydney was the car and would continue to be.? In the Adelaide study, the documents even included photographs of huge freeways in Los Angeles, to show what the proposed roads would look like. Residential land use was assumed to be along the lines of every family owning their own plot of land. Alternatives were barely examined. No vision of the future of the cities were developed, and there appears to be no sense that the future could be built. The forecasting seems to have been almost fatalistic: "This is what is going to happen to us in the future, so this is how we’ll plan for it. Demand will grow in these corridors, so this is where we’ll build freeways and provide some public transport services".

We know that these studies were influential because decades later, at least in Melbourne and Sydney, they are still being implemented. There may be some differences in the exact routes of the ironically named freeways, and in the type of mode (heavy rail compared to metro or light rail) but a glance at a map of the cities compared to the planned roads and rail show the similarities between what’s in the studies and what’s on the ground.

The transport studies talked about in this article were carried out in relative openness. They may have been the last thorough transport planning and modelling exercises in these cities. We wouldn’t know, because projects are often cloaked in cabinet or commercial confidentiality. Transport for New South Wales has developed a Future Transport Strategy, which is laudable, but it was largely developed in-house. The document espouses TfNSW’s vision for the future of Sydney’s transport network, which may not necessarily reflect the community’s vision. There is limited insight what Sydney will look like beyond transport in the future, how the city will operate, how we will live and work in the challenging future that we face. In the meantime, projects are proposed by corporations, run through a business case and a study into the impacts on the environment and then constructed so that the corporation can make a buck.? Our politicians make plans based on ideology, prioritising projects that are beneficial for their party rather than the community. The result is on-again-off-again infrastructure projects that waste hundreds of millions of dollars without benefit to society.

Our planning game is not good enough and has not been for decades. We need to improve urgently. Compare the unexpected in the Sydney study with what is likely to come in the next few decades. We need to take the impacts of global warming more seriously and plan for it with more urgency. We need to more holistic, less focused on maturation of investments and prepare plans for robust and resilient infrastructure for power, water, education and health as well as transport. Our cities need a vision for the future that is driven primarily by the young people who will live in them. Infrastructure Victoria’s Young People’s infrastructure Forum was a good start; programs like it need to be expanded everywhere.?

If we ever get to a point where we can plan maturely, I cannot conceive of an outcome that would not be heavily assisted and assessed by modelling of demand and supply, not just of transport but of population and employment, power and water supply, provision of housing, hospitals and schools. A review of the vintage transport studies shows that the models, flawed though they may have been, were not problems in the planning of cities. The forecasts that were made with them were not wrong, or inaccurate. It was the way they were used, the assumptions that drove them, the ideological solutions and the political wranglings that were the problem.

If we don’t get to that point, well, there’s not much to say apart from that I apologise wretchedly to my descendants. Our failure to plan will damage the quality of life of future generations. I hope my children never look back and think that I did absolutely nothing to improve their future.

Whatever we do, I am confident that our models will help to plan future infrastructure, testing resilience of the transport networks and the best combinations of infrastructure to help us live in whatever future we make. And I have no doubt that political interference and obstructionism from vested interests will continue.

Links to SATS documents

• SATS Volume 1: Base Year (1971) data report
• SATS Volume 2: Transport model development and forecasts
• SATS Volume 3: Passenger Transport Systems
• SATS Volume 4: Freight Transport Systems