Debates over major transport investments often get caught up in arguments over benefit-cost ratios, or BCRs. In recent years, projects such as the Transmission Gully and Puhoi to Warkworth motorways and the City Rail Link have been criticised for their low BCRs. These debates have often raised more questions than they resolve. So it’s necessary to ask: What is a BCR, how is it calculated, and what does it mean?

The good news is that there is a manual that explains it – New Zealand Transport Agency’s Economic Evaluation Manual (EEM). The bad news is that it’s tediously long and not written for a general audience. This series of posts aims to provide a guide for the perplexed:

In part two of this series we examine a tricky topic – the benefits of transport projects. As in the first post, I’m going to focus on explaining the conventional evaluation procedures, rather than presenting challenges to said procedures.

Transport infrastructure projects are often expensive. The Waterview Connection will cost an estimated $1.4 billion; the proposed Puhoi to Warkworth motorway an additional $800m or so; the City Rail Link an estimated $2.8 billion, although this figure includes inflation and costs to buy and run new trains over a 30-year period; and so on and so forth. For these costly projects to be worthwhile, the benefits of these projects – the “B” side of a BCR – need to be of a similar magnitude.

What does that mean in practice? If we say that the CRL will have several billion dollars in benefits for Auckland, what sort of benefits are we talking about? There are three key things to understand about the economic benefits estimated in transport evaluation.

First, the benefits of transport projects do not translate directly into increases in GDP. Benefits of transport projects are estimated by assigning monetary values to a range of outcomes. But just because the EEM assigns monetary values to benefits does not mean the benefits actually manifest as more “money”. For example, a project that saves people a small amount of time on their morning commute might mean that those people work longer hours. But it’s more likely that they will sleep in a bit longer or read the morning TransportBlog post instead. Naturally, this makes people better off – and hence is ascribed a value – but it doesn’t increase GDP.

Second, there are a number of different categories of benefits, and it’s a bit misleading to combine them all into a single measure. Broadly speaking, there are three main categories of benefits that are quantified in transport evaluations:

  • Transport user cost/time savings tend to be quantified for all transport projects and make up the central component of most transport evaluations
  • Health and environmental externalities are often included to some degree; however, some types of benefits that are harder to quantify tend to be excluded from many evaluations
  • Wider economic impacts such as agglomeration and increased labour supply are typically only calculated for major projects.

This post will cover the first two categories of benefits and, returning to our Ruritanian case study, present a worked example of how one might go about calculating these categories of benefits. I’ll leave the wider economic impacts to the next post, as they require a fuller explanation.

The third important fact about transport benefits is that travel time savings make up the majority of measured benefits. Under conventional evaluation procedures, the main benefit of new transport projects is almost always that they save time for travellers. Other benefits, including vehicle operating cost savings and emissions reductions, are minor by comparison. Moreover, as alluded to above, travel time savings cannot be equated to increased economic activity.

I’ll illustrate this using the Ruritanian case study, which shows that the principal benefit of introducing a new bus lane is likely to be travel time savings for users. But before doing so, I’ll briefly run through several categories of benefits.

Transport user benefits

The central component of most evaluations is an estimate of the effect that new transport infrastructure or services will have on the time and monetary cost of travelling. Over in boffin-land, all of these factors are combined together into a figure called “generalised cost” (GC).

GC is a composite measure that covers all of the monetary and non-monetary costs of travel. It can be thought of as the “utility cost” associated with a trip:

GC = Travel time + Vehicle operating costs + Tolls + Parking Costs + PT fares + user amenity.

This measure is that it allows monetary and non-monetary costs to be converted to equivalent measures and compared. So, for example, the EEM provides conversion factors that allow you to place a dollar figure on an hour spent travelling. These values are in the range of $15-25 per hour for most trip purposes. While the EEM used to ascribe a higher value of time for car users than PT users, NZTA decided to equalise the value of time for different modes.

“User amenity” is a quite broad category that attempts to cover all of the subjective factors that people take into account when using transport. So, for example, the EEM provides values that allow you to estimate the value that people place on (say) each minute spent contributing to a traffic jam, or having a real-time board at a bus stop, etc.

Panmure Station 1
Panmure Station was built to provide a better transport experience for passengers

With that in mind, the following table summarises the components of GC and describes whether or not they are monetary costs. It distinguishes between household travel and business travel, as workers’ time does have a monetary cost when they’re travelling on employer business. However, business travel only accounts for a small share of overall trips.

Generalised cost componentHousehold travel (e.g. commutes, retail trips)Business travel (e.g. freight)
Travel timeNon-monetaryMonetary
Vehicle operating costMonetaryMonetary
Tolls, PT fares, and parking costsMonetaryMonetary
Trip amenityNon-monetaryNon-monetary

Health and environmental externalities

Contrary to popular belief, NZTA doesn’t simply ignore environmental and health outcomes in its evaluation procedures. Indeed, the agency has progressively been attempting to build additional health/environmental elements into its evaluation.

As a result of their work in this area, which has included the development of new modelling approaches and commissioning of reports on the benefits of walking and cycling activity, the EEM now contains recommendations on how to value:

  • The benefits of reduced CO2 emissions from transport behaviours
  • The benefits of reduced emissions, and reduced road noise, which have an effect on amenity and health within affected areas
  • The health benefits of walking and cycling travel.

In practice, it’s not yet standard to value all of these benefits for all projects. Conventional evaluations usually include benefits from reduced CO2 emissions, as they tend to be closely related to vehicle operating costs. NZTA recommends valuing carbon emissions at $40/tonne – a value that seems high relative to current Emissions Trading Scheme prices, but which is low relative to other transport benefits.

Other emissions and noise impacts do not tend to be valued for many road and PT projects, as they’re hard to robustly estimate. Their effects depend upon a whole range of factors, such as population density around roads, topography, weather patterns, and so on and so forth.

Most people aren't that keen on vehicle emissions. Emphasis on "most".
Most people aren’t that keen on vehicle emissions. Emphasis on “most”.

Likewise, health benefits of active transport only tend to be considered for walking and cycling projects. These benefits aren’t necessarily small in value – the EEM states that each kilometre spent walking generates $2.70 in social benefits, which can add up quite quickly. This is potentially a problematic exclusion for evaluation of public transport, as people tend to walk to access bus and train routes. For example, given the size of the average walk-up catchment, each bus user could be walking an additional kilometre or more each day.

Back to Ruritania: Calculating the benefits of a new bus route

In order to give a sense of how these values come together in an evaluation, we return to our Ruritanian example. If you recall, transport planners in Streslau, the capital city, are trying to evaluate a new bus line between two suburbs (A and B). The road between the suburbs is getting increasingly congested, as it’s limited in size and lacking in public transport alternatives. At this point, they’re seeking to determine whether putting in a dedicated bus lane would be a good idea.

While this is a hypothetical exercise, I’ve tried to make it as consistent as possible with New Zealand evaluation practices to give a sense of what an evaluation might look like.

The first step for Streslau’s transport planners is to determine which categories of benefits to measure. After a bit of discussion, they’ve decided to hew to the conventional evaluation procedures, and focus on quantifying reductions in generalised costs of travel and carbon emission reductions. They’re also going to consider whether there are likely to be any health benefits associated with walking to bus stops.

The table below summarises the values that Streslau’s transport planners are planning on using. As Ruritania’s also a developed country, its valuation parameters are pretty similar to those in the EEM.

Valuation parametersValue ($)
Value of time ($/hr)$15
Vehicle operating cost ($/km)$0.40
Greenhouse gas emissions ($/km)$0.02
Health benefits of walking ($/km)$3

The next step in the evaluation is to determine the level of demand for the new PT service. There are a number of approaches to doing so, including integrated transport modelling, surveys of potential users, or desktop analysis based on known factors and historical growth rates. I’m not going to cover demand forecasting right now – that’s a knotty topic for a future post!

For now, all you need to know is that Streslau’s transport planners have reached into their black box and estimated that roughly one-tenth of the existing trips between the two suburbs will switch modes after the introduction of a new bus line. Changes between the Do-Minimum (the current state) and the Option (the new bus line) are summarised in the following table.

Daily travel demand between A and B (in Y1)Do-MinimumOption
Car trips50,00045,000
PT tripsN/A5,000
Average walking distance to PT stop (m)N/A500

Likewise, it’s necessary to forecast the effects of the change on transport speeds. The road between the two suburbs is roughly 5 kilometres long. At present (under the Do-Minimum) it’s quite congested – traffic flows at an average rate of 25 km/hr.

After further rummaging around in the black box, Streslau’s transport planners have determined that the introduction of the new bus line will result in travel time savings for all users. Drivers will have a bit less road space, but congestion will drop due to reduced vehicle traffic. The net effect is that car speeds are forecast to increase to 30 km/hr after the introduction of the new bus line – saving the average driver roughly two minutes per trip.

Buses will also be faster, but some of the effects will be offset by the need to stop and pick up passengers. As a result, average speeds for buses will increase to 27 km/hr, saving the average PT user almost one minute per trip. (In this simple analysis, I have ignored PT fares and PT user amenities such as real-time message boards, assuming that they approximately offset each other.)

Estimated travel distance and timeDo-MinimumOption
Distance (km)55
Average car speed (km/hr)2530
Average bus speed (km/hr)N/A27
Change in TT for car users (min/trip)2.0
Change in TT for new PT users (min/trip)0.9

In short, the new bus line is expected to remove 5,000 cars from the road every day, while improving travel times for remaining users. This is expected to result in:

  • Cumulative daily travel time savings of roughly 1,500 driver hours and almost 100 bus user hours (remember, these travel time savings are valued at $15/hr)
  • A cumulative daily reduction of 25,000 vehicle kilometres, which is expected to reduce vehicle operating costs by $10,000 every day and reduce the social costs of carbon emissions by $500 each day

The total estimated benefits of the project are reported in the following table. For the sake of simplicity, we have assumed that benefits are experienced only during working days. As there are approximately 250 working days in a year, the total annual benefits of the new bus line are expected to be approximately $8.5 million.

Almost two-thirds of these benefits actually arise from travel time savings for car users. This is actually fairly common for public transport projects, as the removal of some cars from the road gives everyone else a much easier ride. For example, some of the biggest beneficiaries of the City Rail Link will be people commuting by car to the city centre or through Spaghetti Junction.

Estimated benefits from a new bus lineDaily benefits ($)Annual benefits ($m)
Time savings for car users$22,500$5.6
Time savings for new PT users$1,111$0.3
Reduction in VOC$10,000$2.5
Reduction in greenhouse gas emissions$500$0.1
Total$34,111$8.5

Finally, Streslau’s transport planners want to understand whether there are likely to be any significant health benefits. We’ve assumed that bus users walk an average of 500 metres to their stop. As we are expecting an estimated 5,000 bus trips per day, this means that bus users are walking a cumulative 2,500 kilometres every day.

Walking catchments to stations. (Source: Human Transit)
Walking catchments are larger when street grids are well-connected (Source: Human Transit)

That’s a surprisingly large number! At a value of $3 in health benefits per kilometre, it adds up to an additional $1.9 million in annual benefits. In other words, including these benefits raises our estimate of the benefits of Streslau’s newest bus line by 20%. That’s potentially a big category of benefits that’s being ignored in many PT evaluations.

Next time: But wait! What about these “agglomeration benefits” I keep hearing about?

Share this

11 comments

  1. Aren’t time savings benefits often overstated due to induced demand or because people aren’t really willing to pay as much for a saving as we might think they are.

    For example all those PPP roading projects in Australia that went bust because people didn’t want to pay the toll the modelling had suggested they’d be happy to pay.

  2. Still, a cost-benefit analysis is only as good as the underlying assumptions. Since many of those assumptions really aren’t easy to monetize, there is a great potential for over or understatement. Is 1 km walking worth $3 or $1 or $2.25? Or nothing at all? What is the value, negative or positive, of whatever that person’s alternative mode of travel? And on and on. Cost-benefit is a great tool but is easily perverted to argue any case. Furthermore, many public benefits cannot be measured and trying to do so just distorts the actual benefit of a service.

    And another furthermore, I believe – I know – CBA is a trap that is set by the likes of highway proponents to present what appear to be robust and rational arguments that support their projects and disadvantage PT. I also believe that we (the sort of folks who believe in the value of PT and environmental protection) have bent over backwards to play this game, and it’s now time to start stating the moral and ethical reasons for public policies again.

  3. Um, you’re missing the other big transport project benefit: accident savings. In many projects, this is the biggest benefit; in fact, travel times may sometimes increase (e.g. install a right-turn arrow for safety reasons, implement traffic calming).

    1. I think you will find that for major new roading projects such as the RoNS, motorists’ time savings are far and away the largest component of the justification. Accident-cost savings are minor by comparison.

      1. Economic evaluation is used for every project, not just the major (big $) ones. For most rural projects, safety is the biggest driver, even when it’s a realignment that saves a few minutes of travel too. Projects like Puhoi-Wellsford and Waikato Expressway are also aiming to tackle the major safety issues north of Auckland and in the Waikato. And nationally, the total social cost of crashes is more per year than the cost of congestion, so I know what I’d be focusing more on.

  4. The biggest weakness I see with B/C analysis is that costs and benefits fall on different people. That matters to the people involved particularly if you are the loser. At an efficiency level it doesn’t matter so long as everyone values a dollar of cost or benefit the same way, but we don’t. A person with limited money gains a much higher level of utility out of each dollar than a person with a big pile of dosh. $20 to someone with nothing can feed their family while $20 to Bill Gates has now real utility value as it doesn’t alter his options in life one bit. Transport construction projects often involve taking land in poorer areas to build roads or rail lines that are then used by wealthy people to travel through. In effect there is a transfer in wealth from poor to rich people which means a decrease in overall utility. Now I like economic analysis as it makes us rigorous and avoids the pork barrel, but we need to keep in mind who are the winners and who are the losers of each project.

      1. I hadn’t considered Newton, but yes. In an economic assessment a benefit is the area under a Hicksian or compensated demand curve which assumes that if people make a substitution then there is no benefit or cost as it is only the wealth effect that matters. The problem is the impact of wealth effects depend on how much you have to start with. There is no easy way around that but we can’t just assume it away.

  5. In theory NZTA considers health and environment costs and benefits. But with significant roading projects, the impression given is that expert consultants are involved to demolish objections to applications on health and environment issues. Reading expert testimony supporting NZTA applications to EPA shows the same experts repeating dismissive comments to a range of projects, particularly on noise, public health and stress. Boards even say that climate change is outside the scope of their decision making so it doesn’t feature in CBR analysis.

Leave a Reply