Every discipline has its challenges. In this post I briefly list some of the challenges to economists that arise in transport planning; these are those things that make it just a bit more complicated than it might be.
1) The wide scope of the field
Transport planning involves any areas of economics at very different scales. Transport models deal with the behaviour of individual consumers – specific utilities and discrete choices. It deals with strategies – route choice on a transport network is a Nash equilibrium it deals with development economics, labour economics, households, inequality and energy economics. It deals with macroeconomics (although usually in a stripped down fashion) and it deals with economic geography, with contracts and conflicts, competition and monopoly.
2) Normal assumptions often do not hold
There are a lot of convenient assumptions that are generally good enough for most applications in economics, be they perfect competition, perfect information, frictionless markets or rational behaviour. Not all of these hold all the time in any area but transportation suffers more than most. Here are a few that don’t always work out:
Convexity vs increasing returns to scale
Transport has a lot of natural monopolies. Rarely are the two motorways competing for traffic between two cities. Only the largest cities will have competing airports. infrastructure is expensive and duplication is often hideously inefficient. Understandably much of the transport infrastructure is managed by the government.
Simply put, the minimum efficient scale for a lot of operations in transportation is very, very large. Projects costing hundreds of millions of pounds are financed by banks and governments.
At the other end of the scale there is often scope for positive feedback. If you increase the frequency of a bus service you increase demand, which when met will give higher capacity through more buses, increasing frequency and so on. The capacity for there to be multiple equilibria in these markets sometimes means that common analytical tools work poorly. Even where there are not strictly speaking multiple equilibria these feedbacks often mean that it is very difficult to make accurate quantitative forecasts.
A lot of transport economics can be unconcerned by consumers having imperfect information. If a commuter does not know the cost of wear and tear on their car accurately how many applications does this matter in? There are issues for welfare economics and some impacts for consumption models but generally speaking motorists behaviour is not significantly changed by modest changes in costs – if you need to drive, you drive and make the cuts to consumption elsewhere.
On the other hand sometimes costs do matter much more when there are meaningful choices to be made. Traffic models for determining trips (origins, destinations, modes, routes) have progressed immensely in the past 15 years but are still not reliable. Too often they are either overly simplistic top down models that cannot be realistically calibrated to match the changes that planners are interested in or they are models with micro-foundations of behaviour that rely on perfect rationality and perfect information to work. If a motorist wants to consider taking a bus to work they need to know the cost of the bus, the cost of car travel in terms of both money and time. Practically, this would usually be found by making the trip. Given variability in travel times it would take many trips, all with time set aside for delays to gain enough information. Firstly this is a significant cost, secondly this cost limits the number of different options that can even be considered.
Positive and negative externalities abound in transportation. Usually this is not too complex for an economist to deal with in a practical sense but the ubiquity of these externalities does mean that care must be taken.
There are the familiar externalities of pollution – particles, noise, carbon dioxide etc.. Then there are the specifics of things such as congestion. Travelling a route causes delays for other motorists (hence the advantages of congestion charging).
There are also the positive externalities, which can be difficult to quantify sometimes: improved transport links can help break monopolies, improve labour supply, lower costs and improve social welfare. People consume transport as it generates consumer surplus for them – each additional purchase of “transport” will in general, mean a benefit to each side of the transaction that is made.
It is also worth noting that these challenges interact. Consider a commuting motorist looking to change mode of transport. The cost of overcoming their lack of information must be borne by the motorist – they must investigate the travel alternatives themselves. If they do change mode and move to bus or train then the benefits of lessened congestion are shared amongst all the other road users.
3) It is a very interdisciplinary Area of Work
This isn’t so much an argument for good economists as for good communicators. That said a through understanding of a field often helps communicate it without excessive use of jargon and in a way that others understand. In consultancy this may mean client organisations that you are advising, but it can also be other members of the team. You need a good dialogue between transport modellers, financiers, economists and engineers. On small projects some of these may be the same people but on larger works this becomes crucial. Of course, just as it is helpful for people other than the economists to know enough economics to communicate well with economists, it is also important that economists in the field are open enough to ideas and broad enough in their outlook that they can appreciate the work and outlook of other parts of the project team.