Apparently Not Much More Than a Penny per Employee-Hour

The term mathematics suggests exactitude, as in the phrase mathematical precision.  But in public policy mathematics quite often ends up causing confusion.  We propose below to reduce such confusion in the statutory formula for funding IndyGo, the Indianapolis city-bus company. 

The statute that empowered the Indianapolis City-County Council to impose a transit tax (a dedicated increment to local income tax) would seem to require that IndyGo’s fixed-route system be funded at least 25% by fares and 10% by other non-tax sources.  But the statute relies for its execution on IndyGo and the Council, both of which are hostile to that result.  As a consequence, the intended limitation of taxpayer subsidies to no more than (100% – 25% – 10% =) 65% has been so perverted as probably to leave taxpayers saddled with about 90% of the system’s cost. 

We therefore propose below that the statute be made less subject to distortion and that its execution be placed in the hands of the Indiana Department of Revenue.  Specifically, the statute should be amended to prohibit the Department of Revenue from collecting transit tax at any rate greater than a limit L that equals half the ratio that a base year’s fare and donation revenue bears to the income implied by that year’s Marion County income-tax receipts:

where R/r is the Marion County income implied by Marion County’s income-tax rate r and resultant revenue R in the most recent year for which the latter quantity is determinable, F is IndyGo’s fixed-route fare revenue for that year, and D is that year’s contributions to the foundation set up in accordance with the statute to receive voluntary contributions.    We will presently see that this formula’s result roughly equals what the current statue’s would have been if it had been faithfully executed.


But first a little background.   Throughout most of the last century Indianapolis streetcar and bus operations were paid for only by fares, not by taxes.  This was consistent with the principles of free enterprise: although mistakes and externalities occur, voluntary transactions are more likely to result in a net societal benefit than involuntary ones are; almost by definition a voluntary transaction doesn’t proceed unless the buyer and seller both believe it will make them better off.

Unfortunately for bus companies, people can on average save more time and reach more places if they travel by car instead of by bus.  Bus ridership therefore fell as more people could afford cars, and by the mid-1970s the Indianapolis bus company could no longer attract enough riders to fund its operations through voluntary transactions.  So it became a municipal corporation subsidized by taxpayers.  Taxpayers’ share of its cost increased as the years went by, with local taxes eventually being supplemented by state and federal taxes. 

To illustrate the extent of that increase, we’ll use numbers adapted from the year 2016.  Specifically, IndyGo’s 2016 financial report says its operating expenses were $74,537,862.  That includes expenses not only of its normal, fixed-route system, but also of its van system, whose destinations are chosen by riders.  According to a preliminary report, that year’s van service cost $9,659,240.  This leaves $64,878,622 as the fixed-route system’s cost. 

To apportion most of that year’s tax revenue to the fixed-route system, we’ll subtract the report’s entire $717,970 value for advertising revenue and $9,308,211 of its $10,387,232 value for total fares from our $64,878,622 fixed-route-expense value.  That allocates $54,852,441 in taxes to the fixed-route system.  (Instead of $10,387,232, IndyGo apparently reported $11,051,807 to the FTA as that year’s total fare revenue, of which it said $9,903,750 came from its fixed-route operations.  We calculated our fixed-route-fare value by applying that FTA ratio to the annual-report number.)

In broad strokes our adopted numbers set forth what the approximate shares of the burden were before the transit tax: 85% from local, state, and federal taxes, 14% from fares, and a small amount from miscellaneous sources such as advertising. 

Obviously, the transactions are no longer purely voluntary; taxpayers are forced under penalty of law to buy bus rides they don’t take for people they don’t know.  And involuntary transactions are more likely to reduce aggregate welfare than increase it.  True, the rider is better off or at least thinks he is; otherwise he wouldn’t take the bus.  But over 95% of the taxpayers who foot (most of) the bill are worse off, because they’re deprived of the use of that money.  

IndyGo Is a Net Detriment

Moreover, the following thought experiment will demonstrate that the benefit to riders does not outweigh this detriment to taxpayers.  The experiment involves re-running the year 2016: we run 2016, Take Two. 

Take Two will differ from Take One in that instead of Take One’s $1.75 bus fare Take Two’s will be the ride’s true cost, $7.08.  (That value results from dividing our $64,878,622 fixed-route budget by 2016’s ridership, 9,163,327.)  But we’ll make sure that riders in Take Two can afford that true cost: each morning in Take Two we’ll magically place in each Take One rider’s pocket the difference between the bus fare he paid that day in Take One and what he’ll have to pay in Take Two if he takes the same rides. 

Will Take One riders take the same rides in Take Two?   Occasionally they may.  But most Take One riders will probably decide that making other arrangements is the better choice in Take Two.  Either they’ll find a less-expensive way to make the trip or they’ll dispense with the trip altogether because they want to spend the $7.08 on something they value more than the trip. 

Since the bus ride’s cost to society is $7.08, what this thought experiment demonstrates is that in most cases the ride’s value to the rider is less than its cost to society; IndyGo works to society’s net detriment.

Naturally, transit-subsidy partisans would argue that the conclusion would be different if externalities were taken into account.  For example, they may argue that the public would benefit from reduced traffic congestion and carbon-dioxide emissions.  We won’t revisit those arguments here; we’ve explained elsewhere that IndyGo actually emits more carbon dioxide than the cars that would replace it, that its overall effect on congestion is minuscule, and that actually its expansion will probably make congestion worse.

Employers’ Argument

But a further argument, which various business leaders made in lobbying for the transit tax, is more relevant in the present context: that expanding IndyGo would “attract talent and take people to work.”  Actually, there was little evidence that those benefits’ value outweighed the degree to which the added tax would impair residents’ ability to, say, buy hearing-aid batteries or pay off their credit-card balances.  That argument therefore encountered some skepticism.  The General Assembly told employers that if they really believed buses are so important they should put their money where their mouths were.  Specifically, the bill it passed told employers they could get the tax if they would voluntarily donate 10% of the fixed-route system’s expenses. 

To get enough tax to increase IndyGo’s budget to, say, $120 million, the donations wouldn’t need to be more than $12 million, or only 0.042% of Marion County employers’ $28.6 billion aggregate payroll: just over a penny increment to a $26-per-hour wage.  If you think a significant difference in attracting talent and getting them to work on time would be worth at least a penny an hour, you might have been inclined to believe Mark Fisher, then vice-president of government relations for the Indy Chamber, when he said in 2014, “I do believe we’ll get support, not just from corporations but also small businesses, individuals and the philanthropic community.”

Ignoring the Conditions

Additionally, the bill required that fares cover 25%.  So, while none of us is a mind reader, it seems entirely possible that most General Assembly members envisioned that the transit-tax bill they passed would result in something like the following:

That is, the transit tax would supplement the existing tax to make up no more than 65% of the fixed-route system’s cost, while actual riders would pay at least 25%, and voluntary contributions, mainly from employers, would supplement miscellaneous revenues like advertising to provide at least 10%.

But that’s not how it worked out.  When the time came to show their sincerity, local employers apparently concluded that adding more bus service isn’t as valuable as they had contended.   According to IndyGo’s 2019 financial report, the top ten Indianapolis employers accounted for 20.79% of Indianapolis employment.  If they account for a similar percentage of local payrolls, they could have met the entire requirement all by themselves for a donation equal to less than 0.2% of their payrolls.  Yet Mr. Fisher, who now is the Indy Chamber’s chief policy officer as well as an IndyGo board member, has been unable to obtain even that small a donation from Indy Chamber members. 

And the City-County Council has flouted the law by imposing the transit tax without requiring that IndyGo comply with the 25% and 10% conditions.  The expansion is not yet complete, so we don’t know what the overall result of that behavior will be.  However, the administration of Indianapolis mayor Joe Hogsett has predicted that ridership may increase by as little as 15%.  If we’re more optimistic and instead assume a 25% increase, we may speculate that revenue will be something like the following when the expansion is complete:

That is, the percentage of the burden borne by taxpayers may climb from 85% to 90% while riders’ share falls from 14% to 10%. 

Twisting the Language

Beyond just flouting the law, IndyGo has seized upon statutory language to avoid the interpretation that Fig. 2 suggests.  Consider, for example, the “operating expenses” and “only those expenses incurred in the operation of fixed route services” language in Indiana Code ¶ 36-9-4-58(b), which imposes the  25%-fare-recovery language:

If a public transportation corporation providing public transportation services in Marion County expands its service through a public transportation project authorized and funded under IC 8-25, the public transportation corporation shall establish fares and charges that cover at least twenty-five percent (25%) of the operating expenses of the urban mass transportation system operated by the public transportation corporation.  For purposes of this subsection, operating expenses include only those expenses incurred in the operation of fixed route services that are established or expanded as a result of a public transportation project authorized and funded under IC 8-25.

To the extent that the intention was a 25-10-65 split among riders, employers, and taxpayers, the choice of the “operating expenses” language was unfortunate, because IndyGo can squirrel away taxes such as the transit tax into, say, capital accounts.  This means that, at least until such time (if ever) as depreciation catches up with capital expenditures, the tax burden will be more than 65% of the operating expenses upon which the 25% and 10% conditions are based.

Another problem arose from the nature of the expansion that the transit tax would fund.  The expansion would consist of (1) increasing revenue hours 70% by extending hours of operation and having buses run more frequently, (2) converting IndyGo’s route layout from a hub-and-spoke arrangement to a grid, and (3) making three of the resultant routes “bus rapid transit” (“BRT”) lines, which would travel mostly on dedicated lanes.  What this means is that the “fixed route services that are established or expanded as a result of a public transportation project” are those of the entire fixed-route system; it’s the entire system that’s being converted from hub-and-spoke to grid and having its hours extended and its frequency increased.  So it’s the expenses of that entire system that are to be at least 25% covered by fares.

If a recent Indianapolis Star article is any guide, however, IndyGo professes to see it differently.  The first of IndyGo’s BRTs is called the Red Line, and the Star reported, “The Red Line's operating costs were $9.3 million last year, according to a fiscal note.  It succeeded in raising 25% of that through fares.”  That is, IndyGo seems to be contending that the statutory percentages apply only to the BRT lines. 

Furthermore, 25% of the purported $9.3 million in Red Line expense is $2.3 million.  But an FTA report puts the Red Line’s year-2020 ridership at only 916,441.  Since various discounts and giveaways reduce IndyGo’s nominal $1.75 fare to less than $1.10, Red Line fare revenue probably didn’t much exceed $1.0 million.  If IndyGo is contending that it has “succeeded in raising 25% of that through fares,” therefore, it seems on the one hand to be basing operating expenses on just a single route but on the other hand taking credit for the entire fixed-route system’s fares in satisfying the 25% requirement. 

Obviously, IndyGo is playing fast and loose with the numbers.  The statute needs to be changed to reduce opportunities for such mischief. 

This is not to say that interpretations differing from Fig. 2 couldn’t have been made in good faith.  As Fig. 4 illustrates, for example, some members may have viewed the 25% and 10% to apply only to the increase:

Fig. 4’s lighter segments represent the pre-expansion values we’ve assumed, while its darker segments represent expansion amounts that exhibit the statute’s 25-10-65 relationship.  This would have made the existing-plus-new total consist of 19% fares, 5% other non-tax revenue, and 75% taxes. 

Now, that view doesn’t fit the facts very well; again, it’s the entire fixed-route system that’s to be “established or expanded.”  But it does show that as a matter of human nature the statute is vulnerable to misinterpretation, intentional or otherwise.    

The Proposal

To reduce that problem, the proposal we made at the outset bases the statutory criterion only on quantities that leave little room for interpretation:  (1) a previous year’s Marion County income-tax rate r, (2) the resultant income-tax revenue R, (3) that year’s fixed-route fare revenue F, and (4) that year’s donations D to the foundation.  Again, the Department of Revenue would be prohibited from collecting a transit tax if the rate set by the Council exceeds a limit L calculated in accordance with:

In a more-perfect world the fourth quantity would be defined more broadly to include not only donations to the foundation but also other non-tax revenue, like advertising.  But IndyGo has displayed bad faith in interpreting the statute; it contended, for example, that revenue “other than taxes and fares” includes funds from the federal government.  So a more-specific definition is necessary to avoid further IndyGo shenanigans.  The statute revision should probably also prohibit the Department of Revenue from collecting the tax if IndyGo fails to make timely certifications of the fare and donation amounts.

Fig. 5 shows what a possible result would be if the fares and donations justify enough transit tax to fund the Fig. 3 expenditure level.

Although in theory this proposal differs markedly from the current statutory scheme, Fig. 5 shows that it approximates Fig. 2’s envisioned results; the taxpayer burden is only slightly less than what Fig. 2 depicts: 63% instead of 65%.

We say that Fig. 5 illustrates “one possible result” because our proposal imposes no condition on fares or donations individually; it specifies only the minimum to which they must add.  As a consequence, those quantities’ relationship could differ considerably from what Fig. 5 shows.  As we will now see, a possible reason for wanting it to do so is disappointing ridership.

In any given year, about 29 out of 30 residents didn’t take even a single bus before the expansion.  So there seemed to be a great potential for increased ridership, and IndyGo had encouraged the impression that the expansion would unleash a huge pent-up demand.  If ridership, say, tripled, then fares would require little or no adjustment in order to cover 25% of expenses.  If ridership were instead to increase only proportionately to the planned 70% increase in revenue hours, then at the expense level we’ve been assuming fares could be made to cover 25% of expenses by raising the fare from $1.75 to $2.00 and eliminating the current discounts and giveaways.    

However, the Hogsett administration has stated that the ridership increase may end up being as low as 15%, and we’ve been assuming 25% in our examples.  If ridership increases by that 25%, the fare would have to increase from the current $1.75 nominal fare to a $2.70 uniform fare to cover 25% of the expense.  We will presently see that the proposal provides a way to avoid so large a fare increase.

But first let’s consider what $2.70 would mean for, say, a high-school kid starting out at McDonald’s for $8 an hour.  If he pays $2.70 each way to commute to and from an eight-hour shift, that’s just over 8% of his pay.  But remember that the average American household spends 16% of its income on transportation.  And people who don’t ride the bus would be paying $8 on top of every $2.70 the rider pays.  Yes, the $10.70 that society will be paying IndyGo for a single bus ride is exorbitant.  But in the scheme of things the young employee’s share of the resultant commuting cost would not be unreasonable. 

As Fig. 6 illustrates, though, the employer community can spare riders such a fare increase if it wants to.  That figure’s first bar is the referendum-year funding depicted in Fig. 1, its second is the Fig. 3 projection of what IndyGo is on track to reach when the expansion is complete if IndyGo continues to flout the law, and the third and fourth columns represent the present proposal. 

Each bar’s bottom two segments represent fares and other non-tax revenue.  As the leftmost bar indicates, other non-tax revenue, such as that from advertising, was minuscule before the transit tax was imposed.  If IndyGo keeps flouting the law, it will remain so, as the next bar indicates.  Under our proposal, on the other hand, no transit tax would be imposed that exceeds half a previous year’s sum of fares and other non-tax revenue.  The bottom two segments’ sum in each proposal bar therefore equals twice its top, transit-tax segment if those quantities don’t vary much from year to year.

However, since the proposal imposes a requirement only on those quantities’ sum, not on their individual values, fare revenue would not have to be as great as the third bar’s bottom segment indicates.  The rightmost bar’s bottom segment, for example, represents the fare revenue that would result without a fare-structure change.  (Here we’ve ignored the possibility that fare revenue could increase faster than ridership because higher bus frequency lures a higher proportion of opportunistic, full-fare riders.) 

To justify such lower fares, the employer community would have to donate more than under the current scheme.  But the required donation would still be little more than a tenth of a percent of its payroll.  That’s still just a drop in a bucket if the expansion is as big a benefit as the Indy Chamber had implied to get the transit tax.


As we have explained elsewhere, IndyGo’s expansion could end up costing as much as $40 in transit tax for each added bus ride, and it will likely increase both carbon-dioxide emissions and congestion.  So it would actually be beneficial to society as a whole for us to dispense with the transit tax as a result of employers’ continued refusal to contribute.  If employers really think the expansion will confer as great a benefit as they contended, though, providing their share of its funding won’t cost them much.