A couple of my favorite people in the whole world have a new policy brief available for us to pay attention to. Kathy Pettit and Tom Kingsley looked at urban neighborhoods and identified which had improved and which had worsened, then tried to see if any predictive indicators existed to be able to tell ahead of time which neighborhoods were going to struggle. The article is interesting, and you ought to take a look if you're interested in neighborhood-level indicators.

Here's how Tom Kingsley at The Urban Institute described the article:

Concentrated Poverty: Dynamics of Change, by G. Thomas Kingsley and Kathryn L.S. Pettit.
Neighborhood Change in Urban America Series, Brief 4. Washington DC: The Urban Institute.

America’s urban neighborhoods generally fared better in the 1990s than they did over the preceding decade, but this brief shows patterns of change were far from uniform. It contrasts census tracts in the 100 largest metropolitan areas that improved over a decade (poverty rate decreased by 5 percentage points or more) with those that worsened (poverty rate increased by 5 points or more). Indeed, a larger share improved in the 1990s (11 percent) than in the 1980s (8 percent). But even though the numbers were declining, the shares that worsened were actually larger in both decades: 15 percent in the 1990s down from 19 percent in the 1980s. The share of neighborhoods that improved in the 1990s was much higher where markets were strong than where they were weak, but the results were always a mix; some neighborhoods worsened even in the strongest markets and vice versa. Neighborhoods that worsened most often saw sizeable increases in minority populations, but racial composition did not change as much in improving tracts, suggesting that gentrification was not the dominant explanation. While there were many exceptions, tracts that improved were most often found in the inner portions of the central city and the outer rings of the suburbs, while tracts that worsened were more prevalent in the outer portions of the cities and, in particular, the inner ring of the suburbs. Beyond that, we found no simple set of indicators as of 1990 that reliably differentiated how tracts would change over the subsequent decade. Local officials cannot be complacent about the good news that has been reported about urban trends of late. Clearly, they should make better use of local data to get early warnings of worsening and improvement and to learn more effective ways to address the challenges that both imply.

The abstract, some highlights, and a link to the full PDF article is available here: http://www.urban.org/url.cfm?ID=411527