The latest US state- and city-level unemployment data was out today; here are 3 points to consider:
#1: Although the national unemployment rate declined in September to 7.9 percent from 8.4 percent in August, the improvement was not spread evenly around the country. Here are the month-to-month differences for the 15 most populous US states (where two thirds of the American population live).
Large states that saw increases in unemployment from August to September 2020:
- Texas: 6.8 percent (August) to 8.3 percent (September)
- Florida: 7.4 pct to 7.6 pct
- Georgia: 5.6 pct to 6.4 pct
- North Carolina: 6.5 pct to 7.3 pct
- Virginia: 6.1 pct to 6.2 pct
- Arizona: 5.9 pct to 6.7 pct
- Average unemployment, August and September: 6.4 percent and 7.1 percent
- Average increase in unemployment: 0.7 points
Large states that saw decreases from August to September 2020:
- California: 11.4 percent (August) to 11.0 percent (September)
- New York: 12.5 pct to 9.6 pct
- Pennsylvania: 10.3 pct to 8.1 pct
- Illinois: 11.0 pct to 10.2 pct
- Ohio: 8.9 pct to 8.4 pct
- Michigan: 8.7 pct to 8.5 pct
- New Jersey: 10.9 pct to 6.7 pct
- Washington: 8.5 pct to 7.8 pct
- Massachusetts: 11.3 pct to 9.6 pct
- Average unemployment, August and September: 10.4 percent and 8.9 percent
- Average decrease in unemployment: 1.5 points
Takeaway: September’s lower national unemployment rate was due to a decline in joblessness in populous states that were most hard-hit by this spring/summer’s shutdowns. Conversely, the headwinds came from states which reopened more quickly but could not sustain quick improvements in their labor markets. From this short-term data set it does seem like the US has a structural unemployment rate of 7 percent on the low end (the first cluster of states mentioned) and 9 percent (the second cluster). Future months may shade that to 6 – 8 percent if labor markets can improve, but we’re a long way from 2019’s 3 – 4 percent.
#2: Urban centers are still seeing very high levels of unemployment, even if these are slowly improving. The data here is still coming in, but we do have information for the 3 largest US cities:
New York City: 14.1 percent in September, down from 16.0 percent in August and a June peak of 20.3 percent.
Los Angeles: 15.1 percent in September, down from 16.8 percent in August and a May peak of 20.8 percent.
Chicago: 11.3 percent in September, down from 12.7 percent in August and a June peak of 16.3 percent.
Worth noting: there are 1.7 million unemployed workers in just these 3 cities. That is 13.5 percent of all jobless Americans, even though the populations of New York City, Los Angeles and Chicago only make up 4.6 percent of the US population.
Takeaway: urban labor markets face the toughest slog until there is a vaccine, and since 14 percent of all unemployed Americans live in NYC, LA and Chicago we can see joblessness in these cities remaining well above national levels into mid-2021.
#3: For readers with an interest in US politics, here are the recent unemployment trends for some commonly cited Presidential election swing states:
As mentioned above, these swing states saw an increase in unemployment last month:
- North Carolina
These swing states, however, had better employment trends in September:
- Iowa (6.0 percent in August, 4.7 percent in September)
- Maine (6.9 percent, 6.1 percent)
- Minnesota (7.4 pct, 6.0 pct)
- Nevada (13.2 pct, 12.6)
- New Hampshire (6.5 pct, 6.0 pct)
- Wisconsin (6.2 pct, 5.4 pct)
Takeaway: in any other Presidential election year we would say marginal changes in unemployment would weigh heavily on state-level outcomes, and perhaps it is wrong to think it is any different in 2020. We keep coming back to the fact that online prediction/gambling odds for Trump vs. Biden remain fairly close (40/60 or so) and the setup we’ve outlined in this point shows why the race really is too close to call definitively.