So there have been proposals submitted suggesting a massive increase in capacity for Birmingham Airport might be a viable approach to easing the constraints on flights.
It is an interesting, if very difficult, question to ask what impact this will have on the UK and specifically on the west Midlands. The scale is difficult to judge, certainly from my quick reading of the report, the comparisons with Heathrow are not well founded. Heathrow carried 69.4 million passengers in 2011; the master-plan is forecasting 27 million passengers per annum by 2030.
The first part of the answer is probably the most simple: Direct jobs. More flights means more people employed at the airport, something that can only help the unemployment problems in the area. Of course there is a multiplier effect as each job spends their extra income, so the shops and businesses in the area benefit. From a local perspective the multiplier will be small as a lot may be bought in from outside the area but from a national perspective there will be somewhat of an increase. Unfortunately from a national perspective these jobs are jobs that would exist wherever the extra capacity would fall and so it is only the impact of different levels of crowding out between areas of high unemployment and low unemployment that would matter. Not that these would be insignificant, but the impact at a local level would be more than at the national level.
These factors shouldn’t be overstated. Many of the most highly paid jobs would not necessarily go to local people. Pilots, for example might relocate if they flew from Birmingham regularly but there are likely to be few unemployed pilots in the region who would suddenly find themselves with a job. This said even economic migrants moving into the region would still spend their money in the local economy.
Secondly there is the direct transport impact. Travellers from the region will take less time to get to their airport and will have less journey time variability on a shorter trip. This means less time set aside for travel and this is time that can be put to productive use. This is fairly standard and uninteresting
Where judging the impacts gets really interesting is when one looks at the bigger picture. What would an expanded airport do to local productivity and wealth? What would it do to prosperity in the region through its effects?
There has been a lot of work in this field, although little of it is entirely convincing in a quantitative sense. There are benefits to connectivity, usually approached under the heading of agglomeration. Agglomeration is usually addressed in the context of expanding cities and moving workers and businesses closer together. Workers benefit from a greater choice of job opportunities, businesses have a greater choice of supplier and workforce and transport costs for moving people and goods fall as the distance between them also drops. Cities allow specialisation; clustering of industry allows knowledge to be shared giving rise to areas like Silicon Valley, Hollywood and the City of London.
At best, the studies that have been done disagree on both the origins and the magnitude of the productivity benefits of agglomeration. Whilst all of the above impacts (and more) can have an effect different study techniques in different cities unsurprisingly give different answers as to the importance of this. One thing we are confident of however is that agglomeration benefits do exist and they can be big. A transport scheme that supports an increase in population density that allows an extra 0.25% increase in productivity may not sound impressive but when you consider that this is applied across a large population the total value can be huge.
Obviously an expansion of flights into Birmingham International will not work to increase agglomeration in the sense described above; it wont allow an increase in population density and so most of the measured statistical relationships (insofar as they ever agreed in the first place) would not be applicable. What does remain is the rationale for these benefits.
Firstly consider how workers benefit from agglomeration. They get a wider choice of jobs, they get to share knowledge and they get to specialise. An expansion of Birmingham International’s traffic would make the City much more international. In doing so it would make it much more feasible for workers based in Birmingham to work on international projects, workers from overseas could work in the city without such a long trip home to visit family and friends. Companies also could expand into the area, particularly specialist companies, in the knowledge that they could recruit more readily from an international pool of talent rather than just the local pool in a less well connected city.
All of these benefits are similar to the more “classic” agglomeration benefits in nature and I believe it makes sense to try and consider them in the same light. There are however a couple of problems.
Firstly agglomeration benefits skilled labour more than unskilled labour. Sharing of skills and knowledge only has an impact on the economy when the workers in question’s skills and knowledge is important. In unskilled jobs this is not the case. Likewise the ‘right person for the right job’ effect, where there are productivity gains from a deeper local labour pool and a deeper jobs pool is only relevant to the extent that workers are not homogeneous, i.e. only if there can be productivity gains by a worker swapping roles with another is this relevant. Given that workers develop comparative advantage through learning this will raise the productivity of skilled labour more than unskilled labour. Finally, the development of specialisms requires specialist knowledge.
Any benefits of this type from air travel will, even more than from most transport schemes, benefit those already wealthy (whilst it is now accepted that trickle down effects are weak, there will nevertheless be a benefit to the wider economy from raising aggregate wealth). When we are reaping benefits from accessing a global pool of Labour we are obviously talking about the most specialist jobs and the most talented people, people that could not be found locally. It is also worth noting here that this needn’t be restricted to standard labour contracts, contractors brought in to work on projects or internally moving the most appropriate staff around in a company to best fulfil certain roles all fall into this category. This obviously is of a benefit both ways, local problems can be solved by the most appropriate people in the world and local problem solvers can solve the problems they are best suited to.
So whilst we might be familiar with the benefits of agglomeration, in this case they would be very much focussed on a narrow section of the community.
The second problem is a matter of assessment. As has already been mentioned forecasting benefits of agglomeration is difficult at the best of times. No result is likely to be reliable and pulling together an appropriate reference class would be controversial to say the least. For this type of benefit it is even more difficult. Firstly you would want to look at productivity growth due to airport expansion. This is not a problem by itself; there are roughly 1,000 international airports and there have been many expansions.
The problem lies with what is comparable. Is a new airport the same as an expansion? Does the impact of an expansion depend on connections from existing connections at the airport in question or other accessible airports? Is the impact of expanding an airport in the 1950’s the same as the impact of an expansion now after globalisation? Do the effects depend on the size of the associated city? Is it important which additional cities are connected to – i.e. would an expansion of a Spanish airport increasing connectivity to the cities of Europe be the same as an expansion in Vietnam increasing connectivity to the cities of east Asia? These are just a few of the questions that could (and indeed should) be asked. The result of which is that what looks like a promising data set of impacts of changes (even discounting the inevitable unreliable local statistics in some regions) becomes insufficient to establish a strong enough relationship to forecast the impacts of expansion, especially given that effects are likely to have long and variable lags.
There may be solutions. There is always the temptation to “go structural” and adopt a reductionist approach, producing a model based on a smaller number of parameters and more assumptions about behaviour. This might work and sometimes it is the only way but such approaches usually give poor results or uninformative results. I imagine that there are few transport modellers that have been surprised by and have accepted the outputs of a model and have later been impressed by how accurate it has been. Calibration often incorporates a somewhat subjective view of what is “reasonable” and therefore is almost like a Bayesian approach with a strong prior.
It is possibly worth a quick mention of leisure flights at this point as much of the above context is business. Leisure flights do give value to the population but much of this wont show up in GDP as expansion basically means people get to pick from a wider range of holiday destinations. Possibly more suitable holidays but little positive impact. Insofar as this may make it easier/more attractive for potential holidaymakers to spent money outside the country it may have a negative effect. As Birmingham is not such a major tourist destination it is unlikely that incoming tourists would offset this loss.
Certainly this is an interesting problem and a major challenge to policy makers.
No cost benefit analysis would be complete without an assessment of the costs, although here I will be brief.
There is some concern expressed in the Independent that a Midlands hub would not be as attractive as a London one: Virgin has also implied that it is not convinced by Birmingham as an alternative, saying Heathrow and Gatwick are “full at peak times because passengers want to fly from those airports”.
There may be some truth to this but it is also probably overstated. The information suggests that at Heathrow 34% of passengers are transfer passengers. With the limited number of . From a commercial viability perspective there should not be a great deal of trouble here. Landing fees will be adjusted to get good use from the airport and this may mean lower fees and a shift of transfer flights away from Heathrow if Birmingham as an origin/destination proves unpopular. Obviously these flights do not benefit Birmingham greatly but from a national perspective they would free up spaces on planes from Heathrow and deliver benefits there. On the other hand if demand at Birmingham is high (and the current forecasts anticipate capturing 57% of regional traffic with expansion up from 34%) there could be substantial benefits to the region. A safe plan may be to anticipate that in the short term charges will be low making it attractive as an interchange/transfer airport and anticipating that the market will grow as people and companies who value this connectivity migrate there.
Incoming flights have an environmental cost through particulates and noise pollution that must be borne by local residents. In addition there are costs to local and regional residents from additional traffic on the roads leading to the airport.
Judging the balance of all this is difficult. There are a lot of unknowns. Traffic problems arise if the expansion is successful for the local economy but not if the airport doesn’t work well for the region. If the airport doesn’t attract/generate large numbers of local users it doesn’t much benefit the local economy, but if that frees more space for transfer flights it benefits air traffic at Heathrow. Air transport is environmentally concerning but shorter, less congested road trips may begin to offset this somewhat, but less so if the scheme is economically successful in development terms.
Assesing this will require skill, judgement and a lot of very difficult work.