On Wednesday afternoon, WNYC-FM’s Leonard Lopate Show tackled the topic of “Why Cities Fund Professional Sports Stadiums,” a subject of more than passing interest around here. Guests were investigative tax reporter (and my editor on the inequality anthology Divided, still available from finer internet trading conglomerates near you) David Cay Johnston and Grantland staff editor Andrew Sharp, who wrote a long article last month calling on Congress to address stadium subsidies because local officials afraid of losing their teams sure won’t. (Or mostly won’t, anyway.)

Now, one of the problems of talk radio (okay, talk anything) is that hosts feel obligated to pit guests against each other, so here Sharp ended up cast as the pro-stadium side, or at least Mr. Glass Half Full. After Johnston led off by outlining the billions of dollars in public cash that goes to stadiums as “just a drop in a very large bucket” of ways that the public end up subsidizing billionaires, Lopate turned to Sharp for any silver lining, and got this response:

“One of the reasons that cities sell themselves on these investments is that every now and then, particularly when you invest in a stadium in an urban area, it can help stimulate growth around that area, and it can turn into a win-win situation where the owners obviously get their subsidies, but then also the surrounding businesses around those stadiums can prove pretty beneficial to the city at large.”

As an example of one of these wins, Sharp cited the new Washington Nationals stadium, which he said “helped revitalize the whole waterfront area” — though he immediately added that there are far more examples of failures than successes.

I got dragged into this last night when someone asked about it on Twitter, which led to me questioning why anyone would consider the Nats stadium an economic success, and eventually to one of those Twitter conversations where nobody is quite arguing about the same thing and everyone just feels icky and misunderstood. So let me try presenting my side here, in a bit more detail.

First off, as Johnston immediately noted on the air, sports venues are “human surge tanks” — crowds sweep in and sweep out on game days, but most of the year the place is dark, which isn’t a great anchor for neighborhood development. Sharp countered that there are now more bars and restaurants around the stadium, and “you can’t deny that now and then these things work.”

There are a few problems here. First off, you can absolutely deny that now and then these things work, especially given that economic study after study has found no measurable economic benefit for cities that build new stadium, or get new teams, or get teams back in action after strikes and lockouts. If there really are outliers that are win-wins, they’re awfully well-hidden in the data.

Secondly, have you been to D.C. lately? You can’t go anywhere without seeing construction cranes — it’s one of the hottest real estate markets in the U.S., and that’s true of virtually every neighborhood, with or without a stadium in it. It’s impossible to say what would have happened to the Navy Yard area if the Nats were still playing at RFK Stadium (or in Montreal, for that matter) — and even if that area wouldn’t have been developed to the same degree, might developers and residents and restaurateurs have gone elsewhere in the city instead? It’s a huge “but-for” problem, albeit one that stadium boosters love to overlook, especially when they just built a stadium in a neighborhood that was already starting to take off.

But fine: Let’s grant that the arrival of Nats Park at least prompted a handful of sports bars and the like to locate in the immediate neighborhood. (I wouldn’t dispute that.) The question here is whether the stadium project is “beneficial to the city at large,” and you can’t determine that without taking into account the price tag. As I’ve noted many times before, there’s a price point where subsidizing stadiums makes sense: In most cases I’d be fine with spending $1 in public money towards a new sports venue, and even the $20 million or so that San Francisco put up for the Giants‘ stadium is arguably reasonable, even if the SoMa neighborhood was already going gangbusters before Pac Bell Park was built.

Nationals Park, though, cost D.C. taxpayers something on the order of $600 million. That’s a crazy-high figure to justify with a few sports bars, but on Twitter at least, Sharp said that the cost isn’t the point:

@neildemause @FechjoCesarano Well sure, but that's just moving the goalposts, and it assumes city's make smart investments elsewhere.

— Andrew Sharp (@andrewsharp) September 11, 2015

I think I get Sharp’s point: We shouldn’t criticize spending $600 million on a stadium just because there are even better investments a city could be making with the money. But that’s not what I was saying at all — rather, the point is that since doing just about anything with $600 million, including sitting on it or throwing it from a helicopter, would be better for the local economy, handing it over to the owners of the Nationals for a new stadium is a massive waste of taxpayer funds.

Let’s start with the simplest example: What if D.C. simply hadn’t collected the money in the first place? About two-thirds of the money came from a tax on large D.C. businesses, and while I’m not about to start defending them as efficient economic engines, they would have done something else with that cash, whether it was hiring more entry-level staff, giving more perks to corporate bigwigs, or (hahaha) cutting prices for local consumers. Sure, probably only a small share of it would have been spent in D.C. — but it’s still a non-zero cost to the local economy. And if that cost is more than the benefit of those handful of sports bars, suddenly the Nats stadium is a net loss for D.C.

The other third (roughly) of the public cost, meanwhile, came from kicking back sales taxes on money spent at the ballpark — not a sales-tax surcharge, mind you, but refunding to the Nats sales taxes that otherwise would have gone to the district. So if the Nats had been playing at RFK, this would have been money that would have gone into the public treasury — and if the Nats had never come to town, at least some of those sales taxes would have been collected when locals spent at other entertainment options in D.C. Again, it’s not 100% — but you can make an excellent case that even doing nothing would have been more economically beneficial to the D.C. economy than building a baseball stadium.

What we’re left with as a pro-Nationals Park argument, then, is that if a city is going to blow a few hundred million dollars on something, at least putting it in a promising neighborhood downtown might shift a little bit of development to that locale. That’s certainly true — Tim Chapin at Florida State has done some good work in this area — but using it as an argument that some stadiums are good public investments is like saying, “Sure, the Pentagon budget may be bloated beyond belief, but aren’t these some cool hammers?”

I don’t want to get on Sharp’s case too much — he was asked to present a counterexample of a stadium deal that’s worked out better, and he threw out one that seems to have been a relative success, at least on a “look around and see if the surrounding streets are blowing with tumbleweeds” basis. But that’s the problem: Just looking at what is there misses what would have been there — and elsewhere in the city — if the project hadn’t been done. Pointing to a full sports bar is easy; pointing to the bar across town that closed, or was never built, because public or consumer spending was diverted away from there is hard without a time machine. And stay away from those things, man, they’re dangerous.