Where the Job Growth Is

Lately I've been seeing a lot of claims about job growth in states that pursue different types of policies. For example, Ben Wolcott of the Center for Economic and Policy Research recently had this to say:

Of the 13 states that increased their minimum wage [on January 1 of] 2014, all but one (New Jersey) are seeing employment gains. Furthermore, nine of the remaining 12 states are above the median for this period. The average change in employment for the 13 states that increased their minimum wage is +0.99% while the remaining states have an average employment change of +0.68%. 

Mark J. Perry of the American Enterprise Institute, meanwhile, notes that Texas has performed much better economically than California in recent years -- a classic red-state/blue-state comparison.

I took a slightly different approach to these questions -- rather than focusing on statistics in individual states, I grouped states together based on the relevant factors (states that raised their minimum wage, states that lean Republican) and looked at job growth in each territory as a whole. Not only does this condense a ton of data into a single trendline, but, unlike a simple average of state-level numbers, it gives more populous states a stronger influence on the results. You can see my spreadsheets, which use Bureau of Labor Statistics data, here.

Interestingly, I confirmed both Wolcott's and Perry's arguments: If anything, states that hiked their minimum wage this year saw strong job growth -- but red states have indeed been recovering from the recession especially well.

Here are my results for the states that did and didn't raise the minimum wage:

Sure, the gap is small -- in January, employment fell about 0.1 percent in non-wage-hike states while rising 0.1 percent in wage-hike states, creating a 0.2-point difference that might have narrowed to 0.1. (The very last data point is based on preliminary numbers.) But the timing is really quite incredible, with non-wage-hike states experiencing a hiccup at the exact moment that the other states changed their policies. If the situation were reversed, many people would be inclined to see that as proof that minimum-wage hikes kill jobs.

As I've said before, the effect of the minimum wage on total employment is almost certainly tiny, so we wouldn't necessarily expect to see it in an analysis like this. I'd prefer to look exclusively at the 16-24 age group, to focus on the demographic most likely to be affected, but the Bureau of Labor Statistics informs me that, for sample-size reasons, it doesn't keep monthly numbers by state and age group. (It does keep annual numbers, so in early 2015 a closer look will be possible.)

[Update: As Jed Graham of Investor's Business Daily pointed out to me on Twitter, we should also bear in mind that many of the minimum-wage hikes in question were small adjustments for inflation. You can see a chart here with inflation-adjustment states in a separate group, but the results don't change much. Also of interest is this new post from Ben Gitis at the American Action Forum; he suggests that the January hiccup in no-wage-hike states might be attributable to especially harsh winter temperatures, and analyzes the numbers for the retail and restaurant industries.]

What about the partisan divide? Using these statistics from the 2012 presidential election, I sorted the country into red, blue, and purple -- "purple" defined as supporting a candidate by a margin of less than 10 points. (I chose the presidential election because it presents the exact same two candidates to voters in every state. You can make a case for using control of the state government instead, considering we're looking for effects of state-level policies, but a government in one state might be very different from a government run by the same party in another.)

Here's how those numbers shake out:

Republican states had more job growth in the year before the crash, suffered less from the crash, and even, it appears, had slightly better job growth during the recovery. Obviously, there are factors that could throw off the analysis -- red states are different from blue states in ways that go beyond politics, and perhaps we should be looking at some type of ratio between jobs and population instead of just raw jobs numbers. (If job growth merely keeps pace with population, maybe that's not impressive -- but maybe the population is growing because people are coming for the plentiful jobs.) Further, it's purple states, not blue states, that are doing the worst, which complicates the narrative that red policies are working while blue policies aren't.

Complicated narratives are the norm when it comes to job growth, though.

Robert VerBruggen is editor of RealClearPolicy. Twitter: @RAVerBruggen