Introduction
Most of the questions presented in the two preceding posts don’t come up much when the Indianapolis Public Transportation Corporation (“IndyGo”) appears before the Indianapolis City-County Council. Empowered by a 2016 referendum, the Council imposed a transit income tax that greatly increased IndyGo’s subsidy. Candid answers to the first post’s questions would have revealed that IndyGo bus rides now cost taxpayers more in subsidy than they’re worth to the actual riders. Candid answers to the second post’s would have revealed that almost nothing IndyGo told referendum voters about bus rapid transit (“BRT”) has proved true of IndyGo’s first BRT route, the Red Line.
IndyGo is nonetheless forging ahead with plans to add two further BRT lines, to be known as the Purple and Blue Lines. So those lines will be the focus of this this post’s questions.
The Questions
How Much Purple Line Ridership Does IndyGo Expect?
The Purple Line will replace Routes 4 and 39. In the referendum year those routes’ ridership totaled 1,407,539. A couple of years ago IndyGo apparently told the Federal Transit Administration that it expected the Purple Line to carry 1,167,400 “linked trips” annually. But it’s not clear that this meant the Purple Line’s ridership was expected to be lower that its predecessors’; maybe that “linked trips” value was arrived at by subtracting from the expected ridership the portion likely to originate from a different bus route.
Anyway, IndyGo may want to revise its prediction now that (1) the pandemic has accelerated trends toward working from home and getting paid not to work, (2) increasing lawlessness has made taking the bus more unpleasant, and (3) the Red Line has not lived up to its ridership expectations. So it would be helpful for IndyGo to provide taxpayers an update.
Purple Line ridership is of interest because unless it greatly exceeds the predecessor routes’ referendum-year value the $188 million Purple Line build-out cost seems hard to justify, even if we ignore the $81 million in federal money included in that figure. Let’s say the long-term rate of return earned by taxpayers’ retirement accounts proves to be 7% (instead of the stock market’s long-term-average 10%). Then the ($188 million – $81 million ≈) $107 million local-tax share of the build-out cost could be earning taxpayers about $7.5 million annually if it weren’t being spent on the Purple Line. For taxpayers, in other words, the build-out’s opportunity cost could be some $7.5 million per year.
Now let’s weigh the benefit against that cost. IndyGo claims that the Purple Line will save riders four minutes on the 3.9-mile stretch between Post Road and Emerson Avenue. That seems plausible, although we note in passing that on the route the Purple Line will follow south of 38th Street the Red Line BRT takes about sixteen minutes to travel from 38th & Meridian to its Statehouse stop, whereas conventional Route 39 takes only fifteen minutes to travel a similar distance from 38th & Park to Market & Alabama, to which conventional Route 19’s trip from 38th & Central takes only thirteen minutes.
But for the sake of discussion let’s accept that on the Purple Line’s 9.9-mile non-Red Line portion, i.e., on the portion where the Purple Line build-out is actually taking place, riders will save time at the claimed rate. That is, we’ll assume that over an IndyGo-average 4.5-mile ride each Purple Line rider on that portion saves (4.5 mi ÷ 3.9 mi × 4.0 min. ≈) 4.6 minutes. Let’s also assume that the ridership on that portion will be (9.9 mi. ÷ 15.2 mi. ≈) 65% of the referendum year’s 1,407,539 rides. That would imply that in the aggregate riders would save (65% × 1,407,539 rides × 4.6 minutes/ride ÷ 60 minutes/hour ≈) 71,000 hours over the course of a year.
Dividing 71,000 hours into taxpayers’ $7.5 million annual opportunity cost tells us the build-out alone will cost taxpayers $105 in opportunity for each hour it saves bus riders. Even if we value the (typically low-income) bus rider’s time as equal to the $22/hour median wage, the build-out alone will cost taxpayers nearly five times what the saved time is worth to Purple Line riders—unless the Purple Line ridership exceeds what Routes 4 and 39 garnered in the referendum year.
Also, the same presentation that claimed each bus rider will save four minutes between Post Road and Emerson Avenue admitted that each motorist will lose three-and-a-half minutes on the same stretch. At that rate the 71,000 hours that bus riders save will cost motorists 620,000 hours in lost time if ten times as many motorists as bus riders travel the Purple Line route. Even assigning (typically not low-income) motorists’ time only the same, $22/hour value we assigned to bus riders’ would imply that the street reconfiguration will add $192 in motorists’ time to the $105 in taxpayers’ opportunity cost: almost $300 to save a single hour of Purple Line riders’ time.
Again, that’s based on the assumption that the Purple Line ridership will equal its predecessors’ referendum-year ridership. So it would be helpful to know whether IndyGo believes the ridership will be higher than that.
How Much Faster Will the Purple Line Be than Route 39?
IndyGo’s Transit Is Essential Pocket Guide (“the pocket guide”) claims the Purple Line will reduce transit travel time by more than 25%. A Route 39 bus takes about 50 minutes to travel between 38th & Post and the Transit Center. So a 25% reduction would mean that the Purple Line will make the trip in no more than 37.5 minutes.
But on its identical route from the Park Street Station to the Transit Center the Red Line takes 26 minutes, and IndyGo has said the Purple Line will take 10 minutes to travel from 38th & Post to 38th & Emerson. The 1.5 minutes that this would leave for the 2.6-mile stretch between Emerson and Park suggests that the pocket guide’s “greater than 25% reduction in transit travel time” claim may be open to interpretation.
Some clarification would be helpful.
How Many Buses Will the Purple Line Require?
The number of Purple Line buses in IndyGo’s plans doesn’t seem entirely in accord with that route’s FTA project profile.
Its profile’s stated bus count is fifteen, and with the promised 275-mile electric-bus range this would have made sense if IndyGo’s claim of “more than 25% reduction in travel time” is correct. On the Purple Line’s 15.2-mile route a round trip at IndyGo’s 13.5‑mph conventional-route average would take 135 minutes. A reduction of more than 25% would lower that duration to no more than about 101 minutes. Adding in a fifteen-minute break1 brings the total to 116 minutes, which the promised ten-minute bus interval translates to twelve buses.
Now, without en route charging the Purple Line’s 15.2-mile route length could make that many electric buses inadequate if their range were no greater than the promised 275 miles. If Purple Line buses depart in both directions on 15.2-mile route traversals every 10 minutes for 19.5 hours2 of the Purple Line’s nominal 20-hour day, they would travel at least (19.5 hours × 60 minutes/hour ÷ 10 minutes/traversal × 15.2 miles/traversal/direction × 2 directions ≈) 3557 miles per day in the aggregate, whereas twelve 275-mile-range buses would have an aggregate range of only 3300 miles. But fourteen buses would raise the aggregate range to 3850, so the profile’s bus count of 15 would have seemed reasonable.
However, it has been reported that instead of fifteen Purple Line buses IndyGo plans to buy twenty-eight for $37.8 million. Also, IndyGo told the Council’s Municipal Corporations Committee on October 27, 2021, that the Purple Line could need as many as thirty at a cost of $42 million, and the $52,074,400 given as their cost in IndyGo’s 2023 budget suggests as many as forty. Why?
Of course, we now know that the electric buses’ range could be as low as 152 miles, but even with that low a range and no en route charging the Purple Line would need only twenty-four buses (since 24 buses × 152 miles/bus exceeds that the Purple Line’s 3557-mile daily total). And IndyGo does have inductive chargers for en route charging, so the required bus count should be less.
Indeed, it could be as low as eighteen. If eighteen buses all start out fully charged and IndyGo recharges them one at a time to total 18 hours of charging at 200 kW over the course of the Purple Line’s 20-hour day, then at the 3.8 kWh/mile maximum consumption rate we inferred in the previous post the (18 hours × 200 kW =) 3.6 MWh from en route charging adds (3.6 MWh ÷ 3.8 kWh/mile ≈) 947 miles to the (18 buses × 152 miles/bus ≈) 2736 miles of range the buses started out with. The resultant 3683-mile sum exceeds the 3557 miles the Purple Line requires.
That’s cutting it close, of course, so as a practical matter eighteen would probably be too low. But thirty or forty still requires some explanation.
How Long Will Blue Line Buses Take to Traverse their Route?
Route 8 currently takes about 83 minutes to travel from Cumberland to the airport. Taking I-70 instead of Washington Street west of Holt Road would probably cut about ten minutes off that time, leaving about 73 minutes as the time a conventional bus would take for the planned Purple Line route.
The Blue Line’s project profile said it would reduce travel time by as much as 38%, which would imply as little as 45 minutes, while the reduction reportedly expected a couple of years ago for selected stretches appeared to be about 30%.
What is IndyGo’s current estimate?
Why the Fixation on Electric Buses?
The Blue Line’s project profile envisioned twenty electric buses, but during the March meeting Ms. Evans seemed to say that the number had ballooned to fifty-three before IndyGo pulled the plug on the electric-bus approach:
We shifted for the first initial onset of the grant to shift from electric buses to hybrid buses which helped to bring back the number of buses from . . . the electric buses went from twenty-nine buses to fifty-three buses so going to hybrid buses brings us back down to around twenty buses.
Twenty hybrid buses should be adequate even if there’s no reduction from the 73-minute conventional-route traversal time we estimated above; ten-minute headways divided into a (2 × 73 =) 146-minute round trip implies fifteen buses if breaks aren’t included. Despite this dramatically lower bus count, however, IndyGo seems to view hybrid use as only temporary; the goal is an all-electric fleet. This is puzzling because at least as far as carbon-dioxide emissions go the all-electric approach wouldn’t seem to provide much benefit.
To demonstrate this let’s assume press-reported bus prices of $7.5 million for thirteen conventional diesel buses, $25.8 million for twenty-seven hybrid buses, and $6.5 million for five all-electric buses. We’ll also assume that diesel fuel costs $3.50 per gallon, that burning it emits 22.4 pounds of carbon-dioxide per gallon, that the so-called social cost of carbon is the Biden administration’s $51 per metric ton of carbon dioxide, that conventional- and hybrid-bus diesel mileages are respectively 3.6 mpg and 4.5 mpg, that an all-electric bus consumes 2.5 kWh/mile, that electricity costs 9.5¢/kWh, and that a bus travels 540,000 miles in its lifetime.
Now, those assumptions are open to question. In particular, the conventional-diesel and possibly the hybrid buses are forty-foot models, not the Red and Purple Lines’ sixty-foot articulated behemoths. Moreover, electric buses’ longevity could differ from their internal-combustion counterparts’. And we’ve chosen not to believe IndyGo’s claim that its hybrids emit 75% less carbon dioxide than conventional diesels do. So we shouldn’t read too much into the illustration below of those assumptions’ implications.
But the differences among the top segments of that plot’s bars make a valid point: even at the probably exaggerated $51/t “social cost of carbon” that Washington bureaucrats have conjured up, the value of electric buses’ carbon-dioxide-emissions avoidance is minuscule in comparison with other bus-cost components. And a factor that the plot omits is that IndyGo seems to require something like twice as great a bus count for all-electric routes as it does for diesel routes—and presumably twice as much bus-storage area, too.
We hasten to add that it’s also hard to see why IndyGo thought it needed 53 buses. Theoretically, if a Blue Line round trip is 48 miles and with dedicated lanes can be made in 144 minutes, then with 56 minutes of charging at the end of each round trip the Blue Line could theoretically be run with only 20 buses and six 200-kW en route chargers even if the buses consume 3.8 kWh/mile and their ranges are only 50 miles. That’s a simplistic analysis, of course, and in practice the required bus count would undoubtedly have to be higher than its calculated 20. Even though it’s merely theoretical, though, a value that low suggests that the case against electric buses isn’t as bad as might be implied by the 53-bus count that IndyGo apparently came up with.
So perhaps a reasonable case can be made for going all-electric. But the plot above shows that IndyGo’s electric-bus goal is questionable if it’s based on a desire to avoid carbon-dioxide emissions.
How Can IndyGo Prove That BRTs’ Net Effect on Development Is Positive?
One of the topics subsidy proponents prefer over discussions about ridership is so-called transit-oriented development. And the proposition that city buses and other subsidized transit spur development is easy to believe. Billions of tax dollars are spent in this country propping up transit agencies, which in turn expend a lot of effort on promoting that proposition. Moreover, companies like FedEx are always in favor of public expenditures that spare them having to pay their employees enough to afford transportation to work. So it’s always easy for city-bus companies to procure testimonials.
But there are a couple of reasons why a critical thinker should be slow to accept arguments that city-bus service’s effect on development is significant or even a net positive.
The first is that disinterested parties who have investigated the proposition find little hard evidence for it. Consider what transit expert Randal O’Toole found when he studied Portland, Oregon:
Because the demand for living in small apartments on noisy streets is limited, Portland and other cities have had to subsidize such developments. When Portland opened its first light‐rail line in 1986, it zoned everything near light‐rail stations for high‐density development. Ten years later, planners reported to the Portland city council that not a single such development had been built. To encourage such development, the city decided to use a variety of subsidies, the most important of which was tax‐increment financing, to dense developments along the transit lines. Overall, Portland has spent roughly $5 billion building its light‐rail system and close to $2 billion subsidizing developments near rail stations.
He similarly found in the case of Dallas that “many if not most of those developments were subsidized through below-market land sales, tax-increment financing, and other government assistance.” Such government pressure seems driven by a political-class desire for density that turns out not to fit what most people want when they actually get around to making their own life-style choices.
As Joel Kotkin put it in The Human City: Urbanism for the Rest of Us:
[T]he notion that development be “steered” into ever-denser pockets violates . . . the wishes of the vast majority. These attitudes reflect a remarkable degree of disrespect and even contempt toward the choices people make. If people move to the periphery, it is not because they are deluded or persuaded by advertising but because they perceive that is where their quality of life is higher.
A second reason for skepticism is that the business case for a net benefit is poor. Sure, employers like FedEx were willing to provide lip service in support of a transit tax. But where were they when the time came to contribute the statutory 10% of IndyGo operating expenses? Sure, businesses can be found on East 38th Street that will welcome the Purple Line. But remember that Route 39 buses already come every fifteen minutes along most of that street. Will ten-minute intervals really increase business that much? Moreover, most people who travel 38th Street are motorists, not bus riders, and the Purple Line’s dedicated lanes will drive many of them away.
Why Doesn’t IndyGo Raise Fares?
Infrastructure is something else IndyGo brings up to justify BRT cost. And it’s true that substantial project outlays have been made on things like sidewalks and drainage. It’s been reported that 18% of the Red and Purple Lines’ construction costs were drainage and that 42% of the Blue Line’s would be if current plans proceed. But we should be careful about weighing one-time expenditures against ongoing ones.
It helps in this connection to step back and consider why IndyGo doesn’t just raise the fare price in order to meet the statutory requirement that fares cover 25% of its operating cost. Politics is involved, of course. But a major part of the answer is that ridership’s price elasticity is probably too high for any fare price to recover that percentage of IndyGo’s current of expenditure level.
That’s because the bus rides IndyGo provides just aren’t worth much more than what riders are already paying. So for probably less than $10 million in value residents have for the last five years spent more than $100 million per year on bus-ride subsidies. That’s around $100 million per year in lost wealth, and it will continue indefinitely. So when IndyGo’s pocket guide touts “$126 million improving infrastructure” we should remember than IndyGo destroys that much wealth in less than two years.
Also, although a lot of that infrastructure does have value, there's also much that's detrimental. In all probability, for example, much of the Purple Line project's sidewalk work on 38th street will have a positive effect on safety. But weighed against that should be the safety issues that arise from so reconfiguring a major artery as to divert some of its former traffic onto neighborhood streets. We shouldn’t be surprised, for example, if the Purple Line affects school children’s safety as adversely as the Red Line may already have.3
Conclusion
A minimum criterion for a legislature’s imposing a tax should be that what the government will do with the proceeds is more beneficial than what taxpayers would have spent it on themselves. Candid answers to the questions in this and the previous two posts would have revealed that taxes used to subsidize IndyGo do not meet that criterion.
At least theoretically buses don’t really require breaks if riders switch off, but we’ve made this calculation conservative by marrying bus and driver.
This calculation assumes that the intervals between buses will be ten minutes all day, but IndyGo plans longer intervals in the evenings, so the resultant number of bus hours may be a little high.
See Councilor Barth’s remarks at 49:00 et seq. in this video.