Our study reviewed 20 LRT projects, five BRT projects, and one diesel multiple-unit (DMU) light railway, all opened since 2000. Major challenges we encountered, highlighted in the report, were that not only is there no standard, consistent process for evaluating how well these projects meet their goals, but in fact crucial data elements are either obscured or unavailable for convenient public access.
For example, we'd originally intended to compare operating & maintenance (O&M) costs among these different projects, but found that BRT's O&M costs are routinely blended in with systemwide bus data. O&M data for new-start extensions to pre-existing LRT systems were similarly blended with systemwide LRT costs.
The availability of any BRT data was by far the worst problem. In contrast with the relatively meticulous projections required for new LRT starts, BRT projects seem to be waved through for approval with the barest minimum of an available data trail.
Data for DMU light railway projects were also a problem. In the end, we had consistent, reliable data for only one such project—North County Transportation District's Sprinter line linking Oceanside and Escondido, California.
Eventually, our study had to be limited to comparing just three measures:
• Capital cost per mile (km)
• Ratio of final cost to original budget
• Achievement of ridership target
Incidentally, our study strictly defined BRT to refer to "bus services operating on exclusive or reserved alignments with in-line stations" — thus excluding so-called "BRT Lite" and similar operations in mixed traffic (that's "rapid transit?). For each BRT project accepted for this study, only route segments in exclusive or reserved alignment (i.e., in "guideway") were considered.
Because of the wide disparities between projects with little to no major civil works (bridges, tunnels, etc.) and those with them, capital cost was divided into two categories:
• Minimal installation—less than 5% of route length involving heavy civil works
• Substantial installation—5% or more of route length involving heavy civil works
So, once the numbers were crunched, what were the results?
In average cost per mile for the Minimal installation projects studied, BRT did appear to have some advantage:
• LRT—$51.8 million/mile
• BRT—$30.8 million/mile
However, planners should keep in mind that (as we emphasized in our conclusions) total annualized cost is a far better gauge of cost-effectiveness than merely raw capital cost.
However, in terms of Substantial installation projects, BRT was pricier per mile than LRT by more than 5 times:
• LRT — $79.8 million/mile
• BRT — $451.7 million/mile
As we concluded, "These results suggest that BRT projects do not have any particular advantage when very heavy installation (tunnels, elevated structure, etc.) is involved."
Incidentally, at $25.2 million/mile, the Sprinter DMU light railway (a Substantial installation project) compared favorably in this evaluation.
How well did LRT and BRT final costs compare with budget estimates? LRT again did better, exceeding budget by only 2% on average, while BRT averaged 35% above budget. The Sprinter DMU project, in comparison, exceeded budget by 38%.
In terms of ridership, we measured the rate at which each project was gaining ridership and compared this with the rate needed to achieve the planning forecast. Here again, while BRT averaged quite well (60% over target), LRT was the big over-achiever, exceeding the target ridership rate of increase by 230%. The Sprinter DMU line, unfortunately, was the big loser — coming short of the target by 20%.
So, what's the bottom line? Our results tended to rebuff the common myth and misconception that BRT is "just like light rail, but cheaper." While the Minimal BRT installations studied seemed to have modestly lower capital costs, BRT didn't seem to have any particular advantage when very heavy installation (tunnels, elevated structure, etc.) was involved. Furthermore, LRT projects on average actually performed much better in terms of staying close to budget.
In terms of ridership, our results suggested that BRT is nowhere even close to LRT, which on average met ridership targets twice as well as the BRT projects.
However, perhaps our biggest finding was the astonishing lack of any consistent, standardized evaluatory procedure for these major public works projects, most of which involve hundreds of millions of dollars in investment. The need for such a standardized evaluation seemed to strike a responsive chord among many participants at the conference.
In my next article, I'll focus on more details from our study — including more insight into the disparity between the relatively rigorous federal scrutiny of LRT projects vs. the comparatively facile approval of new BRT starts.