Background
In a September 16, 2021, meeting Indianapolis’s city-bus company proposed a 2022 budget to the Indianapolis City-County Council’s Municipal Corporations Committee. IndyGo, as the bus company is called, presented some top-level budget numbers, but most of its presentation was directed to huge capital projects, race- and sex-discrimination efforts, appearances with politicians, and similar activities.
What the presentation lacked was the type of context that would have shed light on whether the transportation that IndyGo will deliver justifies the $246.1 M ($660 per household) in capital and operating expenditures that IndyGo wants to make in 2022.
This post will provide some of the missing context.
Passenger-Miles
We start by stating the obvious. IndyGo is a bus company. Its business is moving people; its product is passenger-miles. So the most-basic measure of IndyGo performance is how many passenger-miles its huge budget delivers.
Now, the phrase passenger-mile didn’t come up in IndyGo’s presentation, and it can’t be found anywhere in IndyGo’s 93-page budget book. This is entirely understandable; collecting passenger-mile data is more complicated and error-prone than simply counting fares. But the number of passenger-miles traveled is a better metric than passenger trips for comparing city buses with other transportation modes.
Secular Decline
So, although we have reason to question how consistent IndyGo’s collection methods have been, we’ll start with the passenger-mile data it reported to the Federal Transit Administration (“FTA”):
We see that IndyGo’s fixed-route volume jumps around considerably but that even before the pandemic its long-term trend had been downward. (In this post we concentrate only on IndyGo’s regular, “fixed-route” bus system although for selected residents IndyGo also runs vans that instead of following a fixed route go where riders direct them.)
While IndyGo’s transportation volume has declined, its costs have soared:
Fig. 2 displays FTA data because the FTA provides a relatively long record, but the FTA apparently omits certain cost elements. In an attempt to provide a more-complete picture Fig. 2’s red curve therefore represents approximations based on the annual reports available at IndyGo’s Web site.
Unfortunately, the information that IndyGo shares with the public doesn’t often distinguish between bus and van costs. But van costs for the years 2013–2018 did appear in a published study, and the red curve’s approximations for those years are the differences between the study values and total costs taken from corresponding annual reports. Its approximations for other years are differences between the annual-report numbers and extrapolations from the van study.
Dividing Fig. 2 by Fig. 1 yields Fig. 3, which shows that IndyGo transportation has become hideously expensive:
Even before the pandemic IndyGo bus transportation’s cost had exceeded $2 per passenger-mile. And there’s little reason to believe that it will ever fall below that level again; IndyGo persists in expanding its operations, yet ridership languishes as the resultant expenses rise. Despite the planned service expansion IndyGo’s budget book admits that at least through 2026 its fare revenue will remain below even the anemic, decade-low level to which it fell when only eleven months’ fares were collected in the last pre-pandemic year.
Spin
Although the deterioration in IndyGo performance has recently accelerated, the trend should already have been apparent by 2016. Indianapolis residents nonetheless voted in that year to increase IndyGo’s subsidy by imposing a transit tax that has come to extract about $60 million per year from Indianapolis residents. As we have recounted elsewhere, a largely tax-funded campaign served to hide IndyGo’s dreary decline from the public. What the public saw instead was misleading propaganda like the following excerpt from a January 20, 2015, press release:
IndyGo’s ridership hit a 23-year peak in 2014, soaring to 10.29M passenger trips with seven individual months hitting highs that had never been seen before.
“Growing ridership shows that there’s a desire for more public transportation in Indianapolis,” said Mike Terry, President and CEO of IndyGo. “2014 was a huge year for us, and 2015 will be even bigger with the grand opening of our Downtown Transit Center.”
The “23-year peak” language also found its way into that year’s annual report. Had the press consulted previous reports and FTA records, though, the public might have formed a different impression:
Since IndyGo had struggled in the previous decade’s second half, its ridership did indeed enjoy a peak in 2014. But that peak wasn’t a “23-year peak,” and Fig. 4 shows that the Transit Center’s effect on 2015 ridership may not have been what Mr. Terry predicted.
In any event, ridership is only part of that year’s story. By displaying IndyGo buses’ average number of passenger boardings per hour, Fig. 5 shows that despite the ridership peak in 2014 IndyGo’s productivity had already resumed a quarter-century decline that has now erased half its early-’90s value.
Red Line Roll-Out
Such secular trends give us the big picture. But there’s a shorter-term result that deserves mention in light of IndyGo’s transit-tax campaign.
As the first of three bus-rapid-transit (“BRT”) lines that the transit-tax-funded expansion will eventually provide, the Red Line became a transit-tax-campaign synecdoche for the expansion as a whole. Later, IndyGo marked the Red Line’s September 2019 roll-out by making bus rides free system-wide for two weeks.1 On the Red Line they were also free during October and November.
By displaying monthly values Fig. 6 shows the resultant September productivity spike. But it also shows that even after seasonal adjustment for the normal winter ridership slump IndyGo’s productivity in the remaining five pre-pandemic months fell below the already-downward trend. While not much can reliably be inferred from only five months, that tepid response to the Red Line’s introduction provides scant evidence of the “desire for more public transportation in Indianapolis” that Mr. Terry claimed.
But it may illustrate a subsidy-caused pathology: that the bus company ends up responding to politics instead of economics. Specifically, residents who have little time for reflection are subjected to a barrage of misleading propaganda and accordingly vote for a fantasy. When they see the reality, though, they stay away in droves.
In particular, although during the transit-tax campaign a great many residents voiced their intentions to frequent the Red Line, its paid daily ridership has never averaged even half the 11,000 that IndyGo “conservatively” predicted. IndyGo’s new CEO Inez Evans has disputed the shortfall by claiming that the 11,000-rider prediction referred only to the level the Red Line will eventually reach, after the yet-to-be-implemented Purple and Blue Lines have begun operation. As we’ve pointed out, though, Ms. Evans’ argument was risible; even the Indianapolis Star, ordinarily an IndyGo cheerleader, was unable to accord IndyGo’s prediction any such construction.
When the Purple and Blue Lines do become operational, moreover, it’s unlikely that they will feed many more passengers to the Red Line than are already fed to it by the routes those lines will replace. Perusal of their maps reveals that the Blue Line will be little more than a dedicated-lane version of existing Route 8, while with minor differences the Purple Line will just be a dedicated-lane combination of existing Routes 4 and 39. Buses already come every fifteen minutes on Routes 8 and 39, so it’s unlikely that the Purple and Blue Lines’ ten-minute headways will attract much additional ridership.
Note also that IndyGo has apparently repented of its plan for ten-minute Red Line headways: the wait between Red Line buses has increased to fifteen minutes. So the Purple and Blue Lines may end up not even providing the planned headway reduction. Since the Purple Line will coincide with the Red Line more than Route 39 does, moreover, it may actually reduce Red Line ridership instead of adding to it.
True, the Purple and Blue Lines’ fancier stations will likely draw some ridership that their predecessors do not. But much of it will be probably be cannibalized from other routes. And if the Red Line is any indication their speed advantages will be slight.
“Access”
Maybe such dismal prospects are why Ms. Evans has started changing the subject to “access” when ridership questions arise:
“For our community, expanded transit service means access,” Evans said in a statement provided to IndyStar. “Are we helping students get to school? Can elderly riders get to the doctor? Are our essential workers able to get to work? Yes. Reliable, frequent transit service — like the Red Line — is that connection. And that is success in my book.”
True, access (or “mobility” as IndyGo’s mission statement and a previous Naptown Numbers piece called it) is a useful metric. Variously defined as the number of jobs or other destinations the average resident can reach within a given time period, it measures an important aspect of how well a city functions as a market. Also, it’s helpful in determining where a city-bus company should put its service.
Like city buses generally, though, IndyGo’s buses don’t really provide much access. According the University of Minnesota’s Accessibility Observatory, a car in Indianapolis affords 12 times as much access as a bus does if access is defined as the number of jobs reachable in ten minutes. If the criterion is 40 minutes instead, the multiplier increases to 48. And a car affords over 100 times the access a bus does if the criterion is the 20-minute duration of an average IndyGo bus ride.
Another consideration is that access has little value if few people use it. Note in this connection that Ms. Evans was talking about expanded transit. Buses had already been taking students, the elderly, and essential workers to their respective destinations long before IndyGo started expanding. So by her measure IndyGo had already been a success without the transit tax and the expansion it’s funding.
Completing the Expansion
And nothing we’ve seen so far suggests that the expansion will actually add many riders. IndyGo averaged 14% more buses on the streets in 2019 than it did in the referendum year, 2016. Yet despite over a month’s worth of free rides in 2019 the ridership that year was only 0.46% greater than 2016’s. At that rate the 70% greater bus presence the transit-tax-funded expansion’s completion is supposed to provide will increase actual IndyGo transportation by only 2.3%.
Of course, thus projecting from a single difference is a parlous enterprise, and according to the Hogsett Administration’s Thrive Indianapolis plan the increase will instead be at least 15%. But 15% doesn’t justify the expansion either. IndyGo estimates that its operating expense will already have reached $140 million in 2026, although we’ll guess about $12 million of that will be attributable to vans instead of buses. Let’s generously say ridership increases by 25% of 2016’s 9.2 million instead of the 15% the administration predicted. That would come to $11.18 per bus ride when the expansion is complete even if operating expenses don’t rise further when the Blue Line starts operation in 2027.
Cost Comparison
Let’s place that $11.18 figure into context. According to the FTA data, IndyGo bus rides average about 4.5 miles, making the after-expansion cost of IndyGo transportation about $2.50 per passenger-mile if ridership increases by 25%. If it increases by 15% instead then the cost will be about $2.70. Meanwhile, an Indianapolis cab ride of the average bus-ride length costs about $14 with tip. That would work out to about $2.25 per passenger-mile if 21%, 6%, and 3% of cab rides are respectively shared by two, three, and four passengers.
Those cab-occupancy rates are just guesses, but they suggest that IndyGo bus rides’ cost to society as a whole will likely average more than cab rides’—and certainly more than Uber or Lyft rides’.
Private-car transportation, of course, is much less expensive. As transit expert Randal O’Toole observes, the American Automobile Association’s 62¢-per-mile car-cost estimate is based on the assumption that people buy a new car, pay full finance charges for it, drive it 15,000 miles a year, and replace it as soon as it’s paid off. Instead of the 36¢ per passenger-mile that the national-average occupancy of 1.7 passengers per car combines with the AAA estimate to imply, Mr. O’Toole has detailed how government statistics for the last pre-pandemic year actually work out to only 25¢ out of pocket plus 1.1¢ in subsidies per passenger-mile.
Now, that includes both fixed and variable costs. Although buses no doubt spare a few residents the need for a second or even a first car, the more typical Indianapolis rider probably saves only variable costs such as fuel, oil, and wear and tear. Depending on the car, that might be in the neighborhood of, say, $0.15 per passenger-mile. And something between that and the total cost may be a more-valid value for comparison.
For discussion purposes we’ll nonetheless take Mr. O’Toole’s total-cost value, and we’ll increase it to reflect the fact that on urban streets car occupancy is less than the broader average. We’ll call car cost 36¢ per passenger-mile.
Fig. 7 summarizes the comparison, its first bar representing our car-cost estimate. Its second and third bars show that although IndyGo was already too expensive in the referendum year it probably still cost less than cabs (albeit possibly more than Uber or Lyft). As the third and fourth bars indicate, though, IndyGo’s post-expansion transportation cost will exceed cabs’ even under the generous assumption that ridership will increase by as much as 25% instead of the Hogsett Administration’s 15%.
Now, we’re assessing IndyGo transportation’s value to society as a whole, so we haven’t discussed funding much; we haven’t distinguished between taxes and fares or among local, state, and federal taxes. And partly because they’re only one-time events we’ve ignored that the expansion has attracted federal contributions to the BRT buildouts and that some of the contributions fund things like sidewalks that are benefits separate from transportation.
Not only is the buildout a one-time event, but assessing its net benefit is problematic. Some sidewalks will be added, to be sure, but experience with IndyGo suggests that its characterization of such projects may take credit for replacing infrastructure that the BRT buildouts impair. And, as we’ve discussed elsewhere, some of the BRT buildout’s effects are actually detrimental. So the assessment of those one-time expenditures is beset with imponderables.
But we’re entitled to speculate that the buildout’s actual value to residents won’t approach the cost in federal funds used for the project, which in any event comes to less than only a few years’ transit tax. So from a long-term perspective the quantity relevant to assessing the expansion is the cost of each bus ride the annual transit tax adds: the ratio that $60 million in transit tax bears to the 25% increase added by the expansion to 2016’s 9.2 million riders.
That cost, $26 in transit tax for a single added bus ride, implies the $5.85 per added passenger-mile that Fig. 7’s rightmost bar represents. If the ridership increase instead turns out to be only 15%, then that marginal cost is $9.74 per passenger-mile. And the cost balloons to $64 per passenger-mile if ridership increases by only the 2.3% that our projection from the last three pre-pandemic years implied.
The value that almost all taxpayers would place on a city-bus ride is less than whatever they’d have spent $26 on it if it had been left in their pockets. That should tell the Council that it did not choose wisely when it imposed the transit tax.
Other Effects
Now let’s turn from cost to one of the benefits commonly expected of the transit expansion: that it will ease traffic congestion. We’ve demonstrated previously that dedicating lanes to buses will impair Indianapolis mobility, but here we’ll ignore the dedicated lanes.
In the referendum year IndyGo had an average of 59 buses at a time on the streets (more during weekday peaks, none at night), and those buses averaged 5.8 passengers per bus. If the 70% expansion’s addition of 41 more buses attracts 25% more riders—and it may well attract less—its effect in terms of congestion and emissions will be as though all the added buses averaged only 2.1 passengers per bus.
Moreover, only around half of bus riders say they would otherwise have traveled by car, cab, or ride share. So the added buses will take up much more room on the streets than the cars they arguably displace. The expansion will therefore make traffic congestion worse, not better.
Another purported benefit is that by displacing cars the expansion would reduce carbon-dioxide emissions. But electric buses consume around 2‑3 kilowatt-hours of electricity per mile. This electricity must come mostly from the power company; as we’ve explained elsewhere, IndyGo’s solar array is largely a red herring in this regard. And in Indianapolis the power company has to emit about a pound of carbon dioxide for each kilowatt-hour of electricity it generates.
Even if all the added buses were electric, IndyGo’s expansion would therefore be causing about a pound or more of carbon-dioxide emissions per added passenger-mile. That’s if ridership increases 25%. If IndyGo’s expansion instead garners only a 15% ridership increase, IndyGo’s emissions could exceed 2 pounds per added passenger-mile. In contrast, gasoline combustion’s 19.6 pounds per gallon in carbon-dioxide emissions means that a 24-mpg car averaging 1.25 passengers would emit only 0.65 pound per passenger-mile.
Conclusion
So by the time IndyGo’s 2022 budget has been expended residents will already have paid about a quarter billion dollars in transit tax for an expansion that is increasing not only carbon-dioxide emissions but also traffic congestion. And there must be ways of helping our low-income neighbors that are better than buying $26 bus rides for the small fraction who are bus riders. (As a general proposition, moreover, subsidized transit may actually harm lower-income people, even the minority who do ride the bus.)
In a saner world the question when IndyGo comes before the full Council would therefore be how soon we could end such outlandish spending on a service that not one resident in twenty will use even once a year. It won’t be, of course. If past is prologue, in fact, no one will even be curious enough to ask that IndyGo project how much transportation the expansion will ultimately add.
In light of the projection IndyGo made for the Red Line, perhaps that’s just as well.
A previous version of this post erroneously said that rides were free system-wide for the whole month instead of for only two weeks.