Bham1982,
1. Frank's metric is jobs created per dollar, not the intensity of a use. And let's see the numbers on what type of spinoff you think the Book-Cadillac has. Further, hotel occupancy is actually counted as a spinoff of other things like conventions - it's not an independent economic generator. Hotels in downtown Detroit largely have business due to Cobo [[oft-cited as a subsidy hog).
2. The southern states went through years of getting virtually nothing for what they paid to incentivize auto plants - in fact, they were mostly building knockdown kits for years [[the same way that Western companies built in China for years). We went through these business cases in grad school. I don't recall whether it was a "payoff" on a straight line or not, but I do recall that from local government's [[city, county, and state) standpoint, it was taking about 10 years to break even, if not more.
3. Numbers? Ok. Let me give you some concrete examples.
Take $60,000 per job [[round up to $4.5 million divided by 75). Assume that the city is directly paying 1/3 of that, or about $20,000 in cash, on a one-time basis, as an incentive. There is no indication this is actually happening, but since you and other seem to suggest that the city has better things to do with its money, let's assume it.
A. The Whole Foods site owner would pay property taxes again as an improved property [[and on a very high denominator). The city gets 81 mills of that. If the improved property ends up being worth even as little as $2 million [[a worst case), it's $162,000 per year or $2,160 per worker per year.
B. The workers themselves, assuming they live in Detroit and make $25K [[just picking a low number), end up paying $625 a year apiece.
So right here, going no further and assuming zero spinoff, this pays off for the city in 7 years [[assuming the "city" money really came from the city - and discounting the cost of capital, which may make it slightly longer). But go on to the indirect [[but still unit-quantifiable) impacts:
C. Adding one work/live Midtown median family income [[from a suburban location) provides more than $2,800 in city taxes/year.
D. Getting one midtown house repaired and back on the tax rolls can mean $3,000-4,000/year in property taxes.
E. Every added commuter job paying median income adds $ 500 per year in city taxes.
F. The parking deck behind the Ellington would be sucking up a lot of additional cars. The city gets 1% of the increased net revenue. A hundred more cars a day paying 5 bucks apiece means $1,825 a year in city taxes [[this computation is easy, since all costs in parking decks are essentially fixed costs). That works out to $24.33 per worker per year.
And then:
G. Local sourcing [[if it happens) creates jobs at local businesses.
H. The studies do show that supermarket availability affects property values [[as if you didn't know this from everyone who asks Detroiters whether they have grocery stores).
I. How much money would the city put on being able to say they have a Whole Foods? You could dump $1.5 million into saying how great Detroit is, but the flag of a mainstream retail chain could be a lot more effective.
And on to your point regarding social consciousness for the region?
A. Unless Whole Foods is announcing that it is closing its suburban stores or reducing their workforces, that this is a zero-sum jobs game is just as much speculation as anything else is.
B. If it is actually poaching from the region, the suburbs have for years called it "job growth" and "job creation," often fueled by highly subsidized road construction and subsidized real estate loans. Sauce for the goose.
C. And even restricting the "region" to Midtown, it's pretty well accepted that local competition actually boosts sales in a given area. This is, after all, the idea behind multiple gas stations at a given intersection or putting multiple middle-class department stores into the same mall.