Geography of Recovery: Cumulative Job Gains/Losses since December 2007 by Metropolitan Statistical Area

October 1, 2015

By: TIP Staff

When we released the Geography of Jobs in spring 2008, our goal was to visualize the answer to a seemingly simple question: How did the impact of the recession play out across the country? The resulting animated map—which shows the 12-month rolling job change for all US metros from 1999 to the present—was a resounding success. It provides a vivid illustration of the magnitude of pre-recession job growth and the subsequent dramatic job losses. What our approach failed to capture, however, is the recession’s cumulative impact.

In the second quarter of 2014, it was widely reported that the US had “recovered” all the jobs lost since the start of the recession more than six years earlier. But as we traveled across the country, it didn’t take much to see that many areas were still suffering. With our latest map, the Geography of Recovery, we use the same data to explore this issue. As the name suggests, our new data visualization picks up on the question of recovery: How have individual metro areas fared since the start of the recession? Which metros felt the greatest job losses as a percentage of pre-recession employment? Which have yet to recover the jobs they lost? Which areas recovered faster? Were there any that saw minimal negative impact or even emerged unscathed?

How to Read the Map

Unlike the prior map, which illustrates the change in jobs relative to the same period 12 months earlier, the Geography of Recovery compares employment levels in each metro area to the number of jobs reported at the beginning of the economic downturn. To simplify the comparison, the map uses an index to illustrate this relationship. Each metro starts at 100 percent, which represents total employment in December 2007 (the recession’s official start). From that point forward, the size of each metro’s corresponding bubble grows or shrinks based on the percentage of jobs gained or lost relative to the baseline. A red bubble indicates a cumulative job loss; a blue bubble represents cumulative job gains.

Like the original Geography of Jobs, you can hover over each metro bubble and watch the actual percentage change over time. You’ll also notice two animated “dashboard” features on the left of the map that track with the animation’s timeline. The first is a simple percentage, titled “US share of 2007 employment,” which shows the nation’s job change relative to the baseline. The second indicator is a set of bars representing the number of metros above (in blue) or below (in red) December 2007 employment levels.

Revelations on Recovery

The most striking revelation from this visualization is the unevenness of the recovery. By the time the US returned to its December 2007 employment level in May 2014, the majority of metro areas had not recovered. As of July 2015—more than one year later—fully one-third (120) of the more than 300 metro areas analyzed had not yet recovered the number of jobs lost during the recession.

At TIP Strategies, we are always looking for ways to translate data into insights about economic development. We hope you will help us with this task by providing feedback and sharing your insights at the end of this blog post.

Footnote: We recognize the limitations to this approach:


  • It does not account for population change in each metro over time. Because jobs can grow faster or slower than population, the impact of employment change on a metro’s population may not be reflected.

  • We picked Dec 2007 as the starting point, since this was the date that the national recession officially began. But, some metros, such as Detroit, had already experienced significant job losses in the previous 2 years. Detroit was in a recession long before the official national recession began, therefore their bubble does not reflect losses from the time prior to December 2007.

  • Following the 2010 Census, the federal Office of Management and Budget revised the official definitions of a number of metropolitan statistical areas (MSAs). This once-a-decade overhaul (released in February 2013) resulted in the addition of a number of new metro areas, the change of metro boundaries, as well as the loss of the MSA designation for a number of existing areas. Some added counties, lost counties, or were combined with neighboring metros to form larger MSAs; others lost their designation due to population declines. In implementing these new standards, the US Bureau of Labor Statistics could not produce seasonally adjusted data for all the affected metro areas beginning with its March 2015 release of data from the Current Employment Statistics (CES) program, the data series used to create the Geography of Recovery. While 69 metro areas without seasonally adjusted data are not included in the animation, we have provided a table [PDF] showing the annual percent change in employment since December 2007 using unadjusted data.