The latest Census numbers roiled Chicago, when people found out that the Chicago metropolitan statistical area (not the city, we don't have those numbers yet) and Cook County had lost population. There are a lot of reasons, and beneath those, further reasons analysts will be processing for a while.

The new figures happened to cross paths in the news with another theme—housing affordability, driven in some part by the work of sociologist Matthew Desmond and his new book Evicted. "We’re at point in the U.S nationally where one in five of all renters are spending more than half of their income on housing," Desmond told Belt. The generally accepted cutoff for whether a renter is "rent burdened" is 30 percent of income.

It's not just Desmond. Former Trulia chief economist Jed Kolko recently found that the urban revival is being driven by the top 30 percent of households by income, Gen Xers, college graduates, people with at least four years of college, and people with no children or ones under five (which, obviously, is about when children start school). NPR recently had a story titled "Low-Income Renters Squeezed Between Too-High Rents and Subpar Housing," which makes the valuable point that the floor for rents—for bad apartments in bad neighborhoods—is high.

On that last point, consider Chicago. The median household income in Englewood in 2010 was $21,000, and almost 60 percent of all households made less than $25,000. The median rent was $797. In North Center, one of the city's most sought-after neighborhoods, the median household income was $83,000, and 13 percent of households were under $25,000 (down from 2000). The median rent? $1,131.

So I was curious to see how rent burdens broke down across Chicago. Using data from the Chicago Rehab Network, I mapped out the changes from 2000-2010. It's pretty dramatic.

Here's a map for 2000. Green means that fewer than 40 percent of renters in the community area are rent-burdened; red means that 40 percent or more are.

Now here's 2010.

Not a great pattern. The only areas that escape the trend are, interestingly, near north side neighborhoods—perhaps due to the influx of people who are young, college educated, and without older children, who tend to have higher wages.

But as Daniel Kay Hertz has pointed out, the measure we use to say whether someone is rent-burdened—at least 30 percent of income goes to rent—is not necessarily a great measure. If you make $150,000 and spend $50,000 on housing ($4,200 a month!), you still have $100,000 left over. If you make $30,000 and spend $10,000 ($833 a month) on housing, $20,000 isn't a lot for everything else.

So, alternately, I mapped the change in the rent-burdened percentage of the population just among those with household incomes between $25,000 and $50,000. And that's where the dramatic change really is. Here's 2000:

And 2010:

What changed? Generally speaking, rents increased and household incomes stagnated, though the mix of those causes varies by neighborhood. In some of Chicago's poorest community areas, rent dramatically increased while income fell. In West Garfield Park, for instance, the rent jumped from $680 to $904 (in 2010 dollars) and the median household income fell from $29,750 to $24,000. In Englewood, the median rent rose from $600 to $800, but the median income fell from $24,000 to $21,000.

I wanted to double-check this general idea based on the most recent data, so I ran the American Community Survey five-year estimate for median rents at the census-tract level. As Hertz describes, it's arguably the best way to get a sense of what people have actually been paying in recent years, as opposed to the red-letter estimates that come from some listings companies. There's a time lag, because it comes from five years of estimates, but it's a useful tradeoff.

Below, green signifies median rents below $1,250 a month, red shows rents above $1,250.

And the spectrum of median apartment rents in Chicago is, relatively speaking, pretty narrow. There aren't a lot of census tracts with rents below $600 or above $1,700. The median rents in the tracts surrounding my last apartment—a small three-bedroom that was $1,130 with heat in a great neighborhood—run from about $900-$1,100. In the heart of Englewood, it's about $670-$750.

In short, the floor for rents in Chicago is pretty high; it's hard to find a genuinely cheap apartment if you don't make much money. (The ACS estimates don't capture subsidies, but there are a lot of people on that waiting list in Chicago.) Conversely, the ceiling is pretty low; you can pay a lot of money, but you don't have to, so if you're upper-middle-class, it's easy to find a place that gives you a comfortable cushion for everything else.

Chicago is hardly alone in this. The Joint Center for Housing Studies at Harvard reports that the number of severely rent-burdened households—paying more than 50 percent of income for housing—has basically doubled since 1990, because of modest increases in rents since 2000 and a precipitous decline in incomes for renters. Some of the worst problems are in struggling cities; Flint has the highest percentage of housing-burdened renters, at 72 percent; Detroit is fourth at 69 percent. Desmond found that in 2010, "the average monthly rent in Milwaukee’s poorest neighborhoods was only $50 less than the citywide median." Further, he found that "only 260 bucks separated the 90th percentile from the 10th percentile." The distribution seems to be wider in Chicago, where there is much more wealth, but it's still not that wide.

Desmond's proposed solution is vouchers: "each year, we spend three times what a universal housing voucher program is estimated to cost (in total) on homeowner benefits, like the mortgage-interest deduction and the capital-gains exclusion." Once upon a time, this might have been a program with bipartisan appeal, as analysts from the conservative Heritage Foundation have supported the use of vouchers in order to reduce the size of government.

If there's a bright side to this morass, perhaps it's this: According to the Center for Tax and Budget Accountability, the Chicago Housing Authority's surplus gives it leeway to issue more vouchers than it has over the past few years. That's because from 2008-2012, the CHA issued tens of thousands fewer vouchers than it received federal funding for. If the city's looking to stem out-migration, it could be one factor, and one of the most basic—one that literally determines where people live.