Expressions of Discontent

Of all the urban transit systems that became mainstream in the 19th or early 20th century (cars, underground or elevated trains, buses, trams and trolleybuses), all have thrived continuously, except trams and trolleys. They died out almost everywhere in the world between the 1930s and 1950s. Why? Simple: of all the transit systems that ever became mainstream, they are the worst. They cannot compete with buses for cost-effectiveness or flexibility, and they cannot compete with cars or underground and elevated trains for speed or capacity.

Yet, strangely, a trend has emerged recently for cities to reintroduce the tram. This article concerns itself with understanding why this trend is occurring, and sets out to explain why as a councillor or voter you should reject any proposal to introduce trams or trolleybuses in your area.

The one plus-point of trams is this: they are powered by electricity, so they pollute less (in the locality) than conventional buses or cars powered by internal combustion engines. Everything else about trams is negative.

Compared to buses, trams are much more expensive. A bus capable of carrying 80 people costs about £150,000 new, while a trolleybus that carries up to 125 people costs £400,000, and a tram that carries a maximum of 250 people costs £1,000,000. If we leave aside the cost of laying a track or guideway for a moment, we find that the cost per passenger of a bus is £1,875, while that of a trolleybus is £3,200, and that of a tram is £4,000. (For perspective, a Ford Focus Zetec, having a £12,550 list price in 2004 and 5 seat capacity, costs £2,510 per passenger.) In other words, a trolleybus costs nearly twice as much to buy per passenger that can be carried, compared to a conventional bus, and a tram costs more than twice as much, and that doesn't take into account the cost of any track or guideway.

I'm taking these numbers from a pro-tram source, so they are unlikely to be biased against trams in favour of buses. Here's the source: it's a spreadsheet, and it resides on this website. (Incidentally, if you go down to the footnotes at the bottom of the document, you'll find that it admits that the buses it's talking about are actually capable of accommodating 96 people, but we'll ignore that. We're happy to show trams in their best light, and buses in their worst, because even then it is clear that trams are worse than buses.)

So, the vehicles are expensive. Perhaps there's some compensatory benefit to make up for that extra expense? Certainly, advocates of trams and trolleybuses say there are, but their claims are dubious, to say the least.

Using the same data source referred to above, we discover that an existing bus system that carries 1,600 passengers per hour at peak time can be replaced by a tram system that will be able to carry 1,840 passengers per hour, if predicted increased usage occurs. That's an increase of 240 passengers per hour, or 15%. The increase in capacity afforded by a trolleybus system would be slightly less. What's not mentioned is the reduction of capacity of the road itself, caused by the presence of the tram. If the tram takes a reserved right of way, it could cut the capacity of a four lane road by half or more, which would not just cancel out the benefit of switching from bus to tram, but cancel out the benefit (in terms of capacity) of having any public transport system at all. If cars travel along a lane maintaining a two-second second gap between each vehicle (as advised by the highway code), and there is an average of 1.3 persons per vehicle (this is a conservative figure, since according to UK Office of National Statistics, the average occupancy of cars in 2000 was 1.56), the capacity of the lane is 2,340 persons per hour. Reserving the lane for the benefit of trams means taking away a car-borne capacity of 2,340 persons per hour. If our tram is to travel two ways, we need to take away two lanes, which is to say, we need to take away a car-borne capacity of 4,680 persons per hour, in order to provide 1,840 persons-per-hour capacity by tram. Let's be nice to the tram, and assume that it takes a circular route, and steals only one lane from the car. In that case, we're still reducing the capacity of the system by 500 persons per hour. How does that help solve the problem of congestion? It doesn't, that's how.

Most proposals for tram and trolleybus systems in Britain include proposals to reserve lanes of roads for exclusive use by the trolleybuses or trams. The spreadsheet from which my numbers come makes the assumption that the trolleys have a "clear run between stops", but that buses do not. This assumption makes it possible to claim higher average trip speeds for the trams or trolleybuses than for ordinary buses, and, as a corollary, a need for fewer vehicles in order to achieve the required capacity. That way, it can be claimed that although the trolleybus or tram vehicles cost a lot more each, they don't cost much more in total. The spreadsheet we referred to earlier claims that if a lane is reserved for the tram, it will cost £9,000,000 in the purchase of nine vehicles to provide the capacity that would require 51 vehicles, at a total cost of £7,650,000, if we stuck to buses. If trams really are a lot better than buses, then investing a mere £1,350,000 extra for trams might be a good idea. However, if a lane is reserved for the bus, the bus can go much faster, and fewer buses are required (just 21, in fact). Then, the total cost of buses to required to proved the desired capacity drops to £5,250,000. So reserving lanes for buses looks like a much better option than buying trams. The extra capacity for public transport comes from the act of reserving the lanes, not from the purchase of trams.

However, the strategy of reserving lanes for public transport, whether buses or trams, is only reasonable if there is a surfeit of road capacity, so that when a lane is taken away from cars and given to trams, the total capacity of the system is increased. Unfortunately, in most British cities (and in most cities in developed countries around the world where trams are being proposed), there is no such surfeit of road capacity, so taking lanes away from cars to give them to either trams or buses reduces the capacity of the system as a whole, and makes congestion worse. This is, of course, exactly what the people selling trams (and buses) want to see happen. The worse congestion is, the easier it is for them to convince people to buy buses and trams, because everyone tends to think that road congestion is caused by cars, whereas in this particular situation, it is caused by public transport.

So far, we've worked out that the vehicles are more expensive if we switch from buses to trolleybuses or trams, and we've worked out that the vehicles don't give us extra capacity for public transport, but rather, it is the act of reserving lanes that gives us that extra capacity. Furthermore, we've worked out that reserving lanes reduces the capacity of the system as a whole, and makes congestion worse. Why, then, are we supposed to switch from buses to trams? Are there some benefits being promised? Yes, there are two alleged benefits.

The first is that trams and trolleybuses are more energy efficient and cleaner. By drawing their power from the electricity grid, they avoid polluting the area where they travel, and they use a cheaper, and supposedly more efficient energy source. According to our spreadsheet, a fleet of diesel-powered buses will use £136,000 worth of fuel to run, while a trolleybus fleet will use £117,000 worth of electricity to carry the same number of passengers, and a tram fleet will use £87,000 worth. There's no mention of how much it would cost to run a hybrid diesel-electric bus, but manufacturer claims are that such buses use 60% less fuel than their standard diesel equivalents. If we assume that the manufacturers are being optimistic, and that the real gain is just 40%, the cost of running buses drops from £136,000 to £81,600 -- less than the energy cost of trams. A sixty percent reduction would bring the fuel cost of buses down to £54,400. What about the green credentials of such a system? Well, manufacturers claim a 90% reduction in harmful emissions for hybrid buses relative to conventional diesel. Finally, diesel engines can use biofuels, and if they do, they are liberated from dependency on fossil fuels, and no longer contribute to the increase in atmospheric CO₂ that is alleged to be causing global warming. The total cleanness of a vehicle powered from the electricity grid depends on how the electricity is generated. In Britain, most of that power comes from natural gas, which is cleaner than petrol, but is still a fossil fuel, and still not totally clean. In France, most of the power would come from a nuclear source, and in Germany, the source would be coal, also a fossil fuel, and dirtier than petrol. The green credentials of trams are therefore open to challenge. A hybrid bus is likely to be at least as good.

The second benefit claimed for trams is lower operating costs. Because they don't have engines to go wrong, they are maintained more cheaply and last longer. According to the claims in our spreadsheet, a fleet of trams costs £1,601,000 per annum to maintain and operate, while an equivalent fleet of buses £4,088,000 to run. So trams look a lot cheaper per annum. However, this figure excludes the cost of maintaining the tramway, as well as all capital costs. When the cost of maintaining the tramway is brought into the equation, the balance tips the other way, and buses become cheaper (£5,363,000 versus £6,267,000). In fact, the claimed lower operating costs for the vehicle are much more than offset by the capital costs. Not counting the cost of disruption to traffic during the several years while work is in progress, the cost of laying a tramway is, according to our pro-tram source, a mere snip at £4,000,000 per km, or £1,000,000 per km for a trolleybus way. The cost of track for buses is, of course, zero. When the capital cost of the tramway is factored in, according to our pro-tram source, a trolleybus service costs £123,000 more than buses per annum, and trams cost £693,000 per annum. A hidden assumption is that an entire fleet of new buses is being bought, at a cost of £19,125,000. Without that assumption, the cost of running a bus service suddenly gets a lot lower. There are very few towns cities currently contemplating getting trams that do not already have a substantial fleet of buses.

The two claimed benefits of trams relative to buses are empty: they don't cost less, and they aren't greener. The implied claim that switching to trams increases capacity is also false. The increase in capacity comes from reserving rights of way for exclusive use by public transport, not from switching to trams, and that increase in capacity for public transport is offset by a larger decrease in capacity of the transport system as a whole. If increased congestion in the system as a whole compels some car drivers to abandon their cars and adopt the tram, that will be hailed by tram advocates as a success, but the total percentage of drivers projected to make such a switch is small. Our spreadsheet promises a 15 percent increase in ridership if trams are adopted. That's an extra 240 riders per hour, as a result of removing more 2,340 persons per hour capacity from the road. The rest of car drivers will presumably still be driving in the more congested roadway remaining available to them, but more slowly than before, and so less fuel efficiently, and more pollutingly.

There are other problems, too. Proposed tram and trolleybus systems run fewer, larger vehicles. What this entails is that fewer vehicles travel on a route each hour, meaning in turn that passengers will have to wait longer for their ride. The spreadsheet tells us that passengers will have to wait three times as long at peak times for a tram than they would for a conventional bus service. It also tells us that the stops also have to be further apart, meaning that passengers have to walk 33% further to catch their ride. That amounts to a significantly worse service.

Then there's the problem of the fixed guideway or tramway: with buses, if you decide to change or extend your route, all you have to do is tell the driver to take a different route. With trams, any change of route, however minor, involves laying new track, at a cost of millions per kilometre. Temporary changes of route are utterly unfeasible.

Finally, the pro-tram information that we have been using claims that an operating profit will be possible because of increased use of the system. Experience shows that tram systems in Britain lose money. Passenger numbers are lower than expected, and even when people do ride, fare dodging is easy; also costs of installation have been higher than expected. In 2003, the Croydon tram system lost £8.8m according to the National Audit Office. All the other existing tram systems are similarly losing money, the leader being Midland Metro, which lost £11.4m in the same year. None of the seven systems installed in Britain since 1980 cost less than £140 million to install. The only light rail system in Britain that operates at a surplus is the Docklands Light Railway, which is not a tram system, does not operate at grade (i.e., at street level), but is elevated, and entirely separated from the street, and is driverless. (These figures are from a National Audit Office report.)

Using information on data supplied by a pro-tram and trolleybus site, plus a few well-authenticated additional sources, we find conclusively that there is no benefit to be gained by switching from buses to trams.

So why do cities keep installing tram systems, or trying to install them? There are a few reasons: First, there are lots of people who find trams romantic, and those people get themselves involved in local government. Second, installing trams makes local governments look good, as they can be seen very conspicuously to be "doing something" about public transport. Third, there are lots of companies energetically selling trams, because they can make a lot more money from selling trams than from selling buses. Buses are a commodity product, and margins are not high, especially given that there are companies in China and Malaysia that will happily make a bus for you quite cheaply. Tram systems, by contrast, are proprietary, and often protected by patents, so if you buy a system from a company, you will have to go back to the same company any time you want change or extend your routes, or buy new vehicles. They are a boondoggle, nothing more, nothing less.

If local government officials are serious about providing good quality, good value-for-money public transport that increases total capacity rather than reduces it, they should steer clear of trams and trolleybuses. Either stick with good old, cheap old buses, or if you absolutely have to buy something that goes on a track, get an elevated railway (like DLR) -- preferrably monorail, since monorails save space relative to traditional steel-tracked railways.

Besides the fact that elevated railway systems actually increase, rather than decrease the capacity of transit systems, three other reasons for choosing them rather than trams are safety, safety, and safety, because they're separated from other traffic. They can also be made driverless, to reduce the risk of disruption of service as a results from strikes.

If you really want to be radical...

If you really want to be radical, innovative and green, and you really want to take road capacity away from cars, here's a suggestion. Take the route that you had marked out for a tram, and instead of installing a tram, do the following: on either side of the reserved lands install a traffic barrier along the entire length of the route (with gaps at junctions). At a cost of £80 per metre on each side, a crash barrier will cost £3,200,000 for a twenty-kilometre stretch. Then build a roof over the entire length. If you use clear prolypropelene panels, and the route is an even six or seven metres wide, it will cost no more than £160 per metre, or £6,400,000 for a twenty-kilometre stretch for the roof and the struts to support it. Total cost for twenty kilometres (in case your wondering, twenty kilometres is the length of a proposed tram route for West London), £9,600,000, or just under half a million per kilometre, being one eighth of the cost of a tramway.

Now all you have to do is announce that the covered, protected way that you have built is reserved for bicycles, tricycles and scooters, including electric bikes. Since the bicycle is the most energy-efficient ground transport system in existence, and the electric bike is very likely the most efficient powered vehicle, there really could be nothing greener than this proposal.

The cruising speed of bicycles is between 25 and 35 kph, depending on the fitness of the rider. Even a casual rider who is not very fit can ride a bicycle faster than a tram covers its route (24 kph according to our spreadsheet), and can beat a bus (15 kph) without breaking sweat. Six or seven metres of road width reserved for cylists is sufficient for cycling two abreast, or for overtaking, in both directions. If cyclists stick to the same safe headway of 2 seconds as cars are advised to do, then the total capacity in each direction of the reserved lane is 3,600 vehicles (and therefore 3,600 persons) per hour. The total capacity both ways is 7,200 persons per hour. That's hugely better than a tram does -- it actually amounts to a substantial increase in the total capacity of the system, even though capacity has been taken from the car, and the running costs to the operator are practically zero (occasional maintenance of the roof is all). The average cost of a new bicycle is less than the typical cost of a year's worth of tram tickets for a regular user. The freedom of a bicycle is that it is available at all hours of the day, with no waiting required, and journeys can be door-to-door, rather than from where the tram starts to where the tram stops. With a roof in place, commuting by bicycle becomes a valid all-year-round, all-weather proposition, and the crash barrier creates a level of safety that will encourage riders who would be put off by the danger of riding a bicycle in the midst of motor traffic. This suggestion is far greener, far more efficient, and far cheaper than a tram system. Why is something like this not being done?

Perhaps because a desire for efficient, environmentally-friendly travel is not the real reason for local governments wanting to install trams, but rather, fashion.