For example, the Philadelphia metro area has over 6 million residents, while the city itself has only about 1.6. Problem is, the metro area includes extremely rural and small towns within about a 1-2 hour drive of the center of the city [1]. I'm not sure I'd consider someone living in exurban/rural Chester or Montgomery counties, 50 miles from the city, to be living in Philly.
>rural and small towns within about a 1-2 hour drive of the center of the city
According to this, the vast majority of the metro area is within 1 hour.
https://app.traveltimeplatform.com/#/search/0_lng=-75.16843&...
>I'm not sure I'd consider someone living in exurban/rural Chester or Montgomery counties, 50 miles from the city, to be living in Philly.
The number of people living in rural areas is small compared to the people living in closer more densely populated areas. Subtract 10-20% from the population numbers to account for them and you still have a number more useful and less arbitrary than population within city limits.
Other measurements like job opportunities also tends to favor small/mid cities. "U.S. Census Bureau metropolitan area population estimates for 2017, which show a significant increase in domestic migration away from the 53 major metropolitan areas with populations over a million and toward the 54 middle-sized metro areas in the 500,000 to 1 million range." [2]
As to your point about subsidies...many of those subsidies are for the production of food that the cities obviously rely on. I think it would be interesting to see how livable large cities -- dozens or hundreds of miles away from food sources, waste disposal sites, and water sources -- would be if the true, un-subsidized costs were passed through.
[1] http://worldpopulationreview.com/us-cities/
[2] https://www.forbes.com/sites/joelkotkin/2018/05/15/growth-in...