New Study Analyzes Light-Rail Proposals for Kansas City

Today, the Show-Me Institute released a new policy study, "Review of Kansas City Transit Plans," by Randal O’Toole. The study takes a look at recent light-rail proposals in Kansas City proposals and weighs their costs and benefits, looking at statistics and case studies from light-rail projects in other municipalities throughout the United States.

O’Toole documents:

• the typical high cost overruns of other light-rail implementations;
• the low ridership capacity of light-rail and streetcars;
• the increased operating costs over more flexible forms of transit;
• the years of advance planning required, and the accompanying uncertainty about future transit needs or preferences;
• the decrease in per-capita transit ridership that most other regions have experienced after building light-rail lines;
• the ways in which light rail increases traffic congestion;
• the unsuitability of light rail for municipalities without high concentrations of downtown jobs;
• the increased crime statistics correlated with light rail;
• the fact that most light-rail lines consume more energy and emit more greenhouse gases per passenger mile than the average passenger car;
• and the fact that light rail has not spurred urban development in other cities without huge additional subsidies from the government.

In short:

This information suggests that light rail and streetcars would not be a good fit for Kansas City. Instead, this study recommends that the Kansas City Area Transportation Authority contract out bus operations to private companies, which is likely to save 30 to 40 percent of costs. This, in turn, will allow a 50- to 60-percent increase in bus services, including several new bus?rapid transit routes. These improvements should result in far more new riders using public transit than would be gained from light rail — without increasing the cost to taxpayers.

We’ve already received coverage from the Kansas City Star about this study. The Star‘s article is generally fair, allowing O’Toole and Show-Me Institute Chairman Crosby Kemper III to make brief arguments, but the piece is misleading in places. For instance, in this section:

The author is Randal O’Toole, a senior fellow at the libertarian Cato Institute in Washington D.C. He has written several papers and books denouncing light rail, and national pro-rail groups such as the Center for Transportation Excellence have accused him of presenting flawed and inaccurate arguments about this increasingly popular form of mass transit. Some of those arguments — such as associating light rail with crime — appear in the Kansas City study.

“I think it’s important to look at this study in the context of its author,” said Mark Huffer, executive director of the Kansas City Area Transportation Authority, which is leading Kansas City’s light-rail planning. “I think the conclusions ? were a foregone conclusion before he (O’Toole) started.”

O’Toole doesn’t disagree. In an interview, he said: “Generally, light rail is not a good fit anywhere. … It doesn’t carry as many people as buses but it costs a lot more.”

The only argument of O’Toole’s that the Star piece makes any effort to discredit is the study’s use of crime statistics, which is mentioned in the first paragraph quoted above. But this is juxtaposed with the second paragraph, where a critic suggests that O’Toole’s conclusions were "foregone," followed by a third paragraph with O’Toole agreeing — but only about the economic cost/benefit portions of his argument. The way these paragraphs flow, it makes it appear as though O’Toole is agreeing that his interpretation of crime statistics was also "foregone," which is not the case. O’Toole was certainly familiar with these statistics before he began the Show-Me Institute study, but his interpretation of them is accurate.

The Star piece elaborates on its statistical quibble later:

However, O’Toole’s Kansas City study includes some methodologies and arguments he’s been criticized for elsewhere, such as drawing sweeping conclusions from small samples of statistics.

Take crime. O’Toole wrote that light rail has “by far the worst crime record in the transit industry” and cited a decade worth of federal statistics for crimes such as homicide. However, those statistics show all light-rail lines across the country were associated with just two homicides in the last seven years of federal data.

It’s true that the statistics show a marked decline in homicides in recent years — for all forms of transit alike. While this is good news — and some might consider those two deaths an outlier — O’Toole’s study uses statistics "between 1996 and 2005," which goes back a little further than the years considered by the Star. Although light-rail homicides weren’t too much higher in raw numbers in the late 1990s, they still outstripped buses by a huge margin, when considered in terms of per-capita passenger miles. For instance, in 1998, buses had 40 homicides — the highest raw number by far for buses in any of the years O’Toole’s study considers. The homicide count for light rail saw an uptick to 4 for that same year, which seems insignificant until you consider that U.S. transit riders only logged 1,128 passenger miles for light rail in 1998, compared to 20,360 for bus. If we compare the homicide rate per passenger mile, it’s as though there were about 72 homicides on light rail in comparison to those 40 on buses.

Even if we were to consider the homicide rate for light rail to be too low for a meaningful statistical comparison, it’s much easier to see the crime trend when looking at robbery, theft, and burglary. While homicides have fallen off on light rail since the ’90s, other types of crime (including rapes) have increased on light rail — especially in these categories. Just looking at the robbery statistics for 2005, there were 535 incidents reported on buses and 377 on light rail. When considered in terms of robberies per passenger mile, it’s as though there were nearly 4,840 robberies on light rail in comparison to those 535 on buses. (Or, alternatively, it’s like there were about 42 robberies on buses in comparison to those 377 on light rail.) I could show similar comparisons for almost any of those measurements of transit crime.

In other words, the rates of violent crime, in general, only seem lower for light rail because there are far fewer miles of light rail track than there are miles of bus routes. But if both are considered in comparable terms, light rail is far riskier.