First of all, this is not an exact science. Estimating the cost is the hardest thing a company has to deal with. It is forecasting the future. Sometimes, numbers are so big that the public should never know them or the mayor risks his seat in office. But after all, the people have the right to know. Any student armed with enough time, paper and pencils should be able to dig out the real numbers. The mayor might not agree, but for the sake of the city’s finances, it should be done objectively.
It is a fact. Subway is the fastest way to travel in a city. With top speeds ranging from 60-90 km/h, the trains are a safe bet during rush hours. Yes, sometimes they slow down due to incidents and yes, if something happens, the scarcity of lines limits the number of underground alternatives. But, still, in most of the cases, the gains are impressive. A quarter of hour ride below street level is better than a half hour car drive. With cities growing larger than ever, crossing them is becoming problematic. With traffic lights every couple of hundred meters and with an average waiting time of almost a minute, a good chunk of the time spent in a car or bus is lost stupidly. Self-driving cars will solve this problem, but we are not there yet. Today, the subway is the solution and every solution has a cost. The cost of a subway line is expressed in billions of dollars. Is it possible to cut one zero from that cost ? Maybe.
There are three type of metro lines: underground rail, light rail and elevated rail. Each has its advantages and drawbacks. Speed, cost, noise are but a few of the limitations. For the municipality, the cost is the most important factor, but not the only one.
The best subway is, of course, underground. Silent for the neighborhood, cool during hot summers, relatively warm during cold winters, time conscious almost always, the underground trains are the trademark of every large modern city. The access to the trains is optimized for all the categories of people: stairs, escalators, elevators. When there is a crossing and more than one subway line stops at a given station, the transit between lines is relatively short and safe comparing to the street crossing. It is good to separate the cars on the streets from the trains on rail, much better to separate the people taking the subway from the cars. Perfect choice as it is, it comes at a cost. At $250million/km, a 12km line costs about $3billion (stations included). A huge investment that is there for tens of years.
Going up, we have the light rail or street level subway. This is a fast tram running on tracks that are physically separated from the car lanes, so that the trains can run undisturbed by the other vehicles. Laying down tracks on a street is much cheaper than digging up a tunnel. We can cut a zero from the price of the infrastructure. Modern trains are less noisy than 30 years ago, yet their passage cannot go unnoticed. Cheaper, yet more disturbing. Also, it is easier to have line crossings, thanks to the traffic management systems (those big stacks of screens at the central dispatch). Cost of a line : $30/km. A typical 18 km line with stations every half a kilometer costs about $700 million. Much cheaper than the underground subway.
In some cases, the geography of the place does not leave place for light rail. Modern cities are crowded and streets are already narrow. With a width of 3m for a one way track, which is more than the width of a car lane, the consequences are important. There is a third way: the elevated rail. Having a reduced ground footprint (one 2x2m concrete pillar or two 1×1 pillars every 20m or so of track, this leaves much space for the traffic or walking. The cost is between that of the underground and that of the street level variants. On the average, the elevated rail costs around $120 million/km. The cost of the stations is bigger than the one of their ground level equivalents.
The numbers I give here are just rough estimates. There is a good article about the costs of subway lines here. However, estimating the costs is not easy. The municipalities like to play down the figures, so that people don’t criticize them too much. After all, the dream of every mayor is to stay in office at least two terms. Generally, the final cost of the subway is 1.5-2 times the initial bidding cost. In most cases, the increase comes from delays. From my experience, the best way to estimate the costs of a subway line is to go to the real figures. Most of the subway companies are public enterprises backed by the state. As so, they are supposed to publish the real figures in their annual reports. The bad news is that those reports are not very accurate or they are incomplete. Sometimes one has to read hundreds of pages to put together the numbers. In some cases, the reports are not available to the public. But it is the horse’s mouth.