A N.H. commuter rail question: Should Paul Revere have taken the T to Lexington (or Nashua)?


When considering commuter rail in New Hampshire, here’s thought experiment that offers a great place to start. Should the Massachusetts Bay Colony have built commuter rail in Revolutionary-era Boston?

Assuming the technology had been available, would this have made any sense?

We can get to an answer by looking at the primary obstacle to building a successful commuter rail line in New Hampshire today. 

Any debate about commuter rail has to begin with basic demographic data because population density is the key to successful commuter rail service.

The general rule for light rail is that cities need a population density of about 10,000 people per square mile to generate enough riders to make rail a viable alternative to automobiles. (See here and here.)

“The performance of a rail or BRT [bus rapid transit] line is directly related to the surrounding densities,” writes author Christof Spieler. “For example, the most successful light-rail systems in the United States—San Francisco, Boston, Philadelphia, Seattle, Newark, Jersey City, Buffalo, and Houston—serve large areas of over 10,000 people per square mile.

“Put transit in densely populated places. [emphasis in original] The fundamental math of density leads to an obvious rule: put transit where the people are. Successful transit needs to go where population densities are highest.”

Some cities outside of the United States have commuter rail and overall densities well below 10,000 people per square mile. But they tend to be large metropolitan areas with compact, walkable, dense downtown areas. 

For example, looking at the city as a whole, Calgary is roughly as dense as Manchester. Calgary’s density is 3,442 people per square mile, while Manchester’s is 3,496 per square mile. Calgary has light rail and a street railway. But Calgary is also a city of more than 1 million with a dense downtown of about 7,778 people per square mile. It’s the fifth densest downtown in Canada. 

Manchester doesn’t have a single zip code with a density of even 4,000 people per square mile. 

Nashua’s population density is only 2,961.7 people per square mile.

Commuter rail boosters seem to believe that Nashua, Manchester and Lowell, Mass., can all support rail because they’re fairly close in population. Nashua has about 91,000 people, while Manchester and Lowell have about 115,000 people each.

But these totals mask huge variations in density.

Lowell’s population density is 8,489.8 people per square mile, even denser than downtown Calgary. That makes it much more hospitable to light rail than Manchester or Nashua. 

Boston’s population density, by comparison, is 13,967.7 people per square mile.

And that gets us back to our question about colonial Boston. 

In The Revolutionary: Samuel Adams, author Stacy Schiff notes that Adams lived in a Boston that spanned four square miles, with a population of 16,000. (Boston had about 16,000 people from the 1740s through the 1780s.) That’s 4,000 people per square mile, which makes colonial Boston denser than present-day Manchester. 

Few would argue that commuter rail would make sense in colonial Boston. But lots of people argue that commuter rail makes sense in Nashua and Manchester, which are much less densely populated than colonial Boston was. 

The hard reality is that no municipality in New Hampshire is close to boasting a population density anywhere near the level needed to support successful commuter rail. And that will be true for a very long time. 

U.S. Census data put Manchester’s population density at 3,310 per square mile in 2010. In the decade from 2010-2020, the city’s density grew by just 186 people per square mile. 

At that rate, Manchester will reach 10,000 people per square mile in 350 years.

If the city’s growth rate somehow doubled, it would still take 125 years to get to 10,000 people per square mile.

Building, or even preparing to build, a commuter rail line now makes about as much sense as telling Paul Revere to take the T to Lexington. 

(By the way, it took Revere about an hour to get to Lexington from Charlestown by horse. Today it takes about half an hour by car. By bus? About two hours.)