I am getting into the weeds a bit for this post. I hear more questions regarding an urban-rural split in the state, and it is not always easy to parse this out. For one thing, the definitions of rural and urban are not universally decided and are relative to some degree. I am originally from Chicago, so no city in North Dakota appears urban to me. As a result I will need to define some terms in order to advance the analysis.
One way to address this is to finesse the issue by avoiding defining the terms “rural” and “urban” to instead make use of definitions within the data. The Quarterly Census of Employment and Wages (from the BLS) gives county level data and I use that and which counties in North Dakota are part of metropolitan statistical areas and micropolitan statistical areas. For the record, counties in North Dakota that are part of metropolitan statistical areas include Cass, Burleigh, Morton, and Grand Forks. North Dakota counties that are part of micropolitan statistical areas include Billings, McHenry, Renville, Richland, Stark, Stutsman, Ward, and Williams.
The jump in employment shows up clearly in the North Dakota data. What is interesting is that the metropolitan area area employment takes a slight dip prior to 2010 and then continues in a roughly straight line. The jump in employment seems to come from the micropolitan area employment.
So lets consider two alternative calculations that deduct metro area employment from the total state employment and then both metro and micropolitan area employment from the total. These would be two possible representations of rural employment. There are obviously others possible, but these are the most readily available from the data.
Let’s start with the lower line, North Dakota employment less the metro area and micro area employment. There is some increase in this number from around 75,000 in 2010 to a peak around 100,00 in 2015. Since 2015 there was some retreat, but not a complete one. The green line adds back in the micropolitan area employment, so it is simply the North Dakota total less the metropolitan area employment. The graph shows a significant increase in the employment from 150,000 in 2010 to a peak of 225,000 around 2015 with a retreat to over 175,000 at the beginning of 2017. So what does this tell us?
If we consider micropolitan areas as “rural” then rural employment increased significantly. However, if we classify the micropolitan areas as “urban” then the rural areas were responsible for some increase, but the bulk came in the urban areas. The fact is most of the change is coming from a select area, the Bakken region of North Dakota, as the map below confirms. The counties with the higher percentage changes here are Dunn, McKenzie, Mountrail, and Williams, the core counties in the Bakken oil region.
The discussion of urban growth and economic dynamics compared to rural is an important one. While they are gross classifications they are common ones that resonate in regions around the country and with policy makers. In addition, when you consider forecasts of economic performance in the various regions it is even more important that we have meaningful classifications of these areas. The data seem to confirm that the majority of growth is due to changes in the micropolitan areas. What remains is to decide how these counties should be classified.