In 2015, Houston Metro opened its Purple Line (aka Southeast Line). The Purple Line is a light rail facility that snakes 6.5 miles from downtown by Texas Southern and the University of Houston, ultimately terminating at Palm Center on Griggs Road.
When Metro first proposed this line in 2008, it estimated that the cost would be $591 million, and the line would carry 28,750 weekday riders. As of September 2020, the project had cost taxpayers $787 million and is projected to ultimately cost $822 million, 39% higher than projected. That equates to a cost of just over $126 million per mile. The line’s actual weekday ridership last month was 5,230, less than 20% of what Metro originally projected. As a result, Metro spent about $150,000 per daily rider to build the Purple Line.
How did Metro get it so wrong?
We happen to know a lot about this particular project because it was partially funded by the Federal Transit Administration (FTA), which required Metro to file an Environmental Impact Statement (EIS) with the Houston Galveston Area Council (HGAC). Also, the FTA included the Purple Line in an analysis it did of about thirty transit projects, known as a Projected versus Actual Study, in 2020. Here is what those records show.
This project was originally conceived as part of the Metro Solutions program that Metro put before the voters in 2003. By 2008, Metro had completed its original analysis and came up with a proposed route, which in transit vernacular is called the “Local Preferred Alternative” (LPA).
The EIS filed with HGAC included a chapter on how the Purple Line would impact transportation in the area where it would be built, including a projection of ridership. Below is the table from that chapter which showed Metro’s average daily ridership projection for the Purple Line at 28,750.
By the time Metro got around to applying for FTA funding for the Purple Line in 2011, it had quietly dropped its daily ridership projection to 17,200.
While transit ridership projections are notorious for their exaggeration, Metro’s ridership projection on the Purple Line was so egregious that it drew special attention from the FTA in its 2020 study. The report notes that Metro’s Purple Line projection was the worst of the thirty projects it reviewed. This is a chart that the FTA included in its final report showing what an outlier Metro’s projection was.
Because of this, the FTA asked Metro for an explanation of how the projection could possibly be so far off. Specifically, the FTA asked to review the models that Houston Metro used to make the projection.
However, by that time Metro had lost access to the software and the data used to make the original projection. So, instead of providing the FTA with the detailed modeling data it asked for, Metro sent an email to the FTA in which it speculated that the earlier projection models had relied on the assumption that all of the light rail lines included in the Metro Solution program (including the Richmond/Westpark and to each airport) would have been completed and feeding passengers onto the Purple Line.
The FTA skeptically responded, “It leaves us still pondering how changes in the transit network – by themselves – could triple the forecast of Purple Line ridership compared to the . . . the actual outcome.” Its skepticism was well founded because there was nothing in the 2008 EIS referencing the other light rail lines, other than the Main Street (Red) Line which had been in place since 2003. I think it is fair to conclude that the original projection was badly flawed, and Metro created this explanation after the fact as an excuse .1
Metro also admitted in the correspondence with the FTA, that it had run a new projection in 2013 that showed average daily ridership would only be about 6,500. Metro pressed the FTA that the 6,500 projection should be used in its study and not the earlier projections of 17,000 or 28,000. The FTA responded, “While METRO and the FTA no doubt wish that those revised assumptions had been used in the forecast in play at the time of the FFGA decision, we cannot substitute the revised forecast for the forecast actually provided to FTA at the time of the FFGA.”
The significance here is that the FFGA is the point in time when the FTA had committed to financially supporting the project. If Metro submitted a projection of 6,500 daily riders, the FTA would almost certainly have never approved funding the line. Of course, Houston taxpayers were never told that the original projection of 28,500 riders per day had been reduced by nearly 80%.
Last year, the total trips on the Purple Line were just under 1.2 million. In 2016, Metro estimated that the operating cost of the Purple Line was just under $13 million. Of course, that is undoubtedly higher today. So, the Purple is costing Houston taxpayers something like $12 per ride in operating costs.
However, that does not include the capital cost of building the line. If you assume that the capital cost of $822 million is, say 5%; that would add another $41 million of annual cost, or an additional $32 per ride. So, the actual ongoing cost of the operations and the opportunity cost of the capital is about $44 per ride. And that is before you even think about trying to recover the capital investment.
Keep in mind that any trip on this line is, at a maximum, 6.5 miles. A ride sharing trip would cost a small fraction of what taxpayers are paying to provide this service to Metro’s riders. By the way, according to a Metro study, the average speed is 14 miles per hour, and it takes about 30 minutes to travel from one end of the 6.5-mile line to the other.
In 2019, Houston voters approved the Metro Next plan, in which Metro incredibly included an additional 16 miles of light rail. Of course, Metro did not share with the public its abysmal record on the Purple Line; or the Green Line which is equally horrible, and I will be writing more about soon.
Based on inflation since the Purple Line was completed, that additional 16 miles of light rail would cost something close to $200 million per mile. If dropping another $3.2 billion on light rail in Houston is not the definition of insanity, I cannot image what is.
Note 1 - This “if you let us build it they will come” is a common mantra among transit advocates and is completely unsupported by any data. I will be writing more on this soon.