When comparing the price of apples, make sure you aren’t including oranges.
That’s a basic rule in grocery shopping and data analysis. But with the help of a Census report, the Daily Yonder and other media outlets got our fruits mixed up earlier this week.
The Daily Yonder reported September 15 that rural America had seen a decline in median household incomes and an increase in poverty from 2014 to 2015. But nonmetro counties actually saw improvement in both these measurements, according to a report released September 16, 2016, by the Center for Budget and Policy Priorities.
It turns out, median household income in nonmetropolitan America actually increased from 2014 to 2015, from $42,698 to $44,212, according to figures from the American Community Service and reported by the Center for Budget and Policy Priorities. That’s an increase of 3.4 percent, and is roughly on par with the income increase in metropolitan areas during the same period of 3.6 percent.
Similarly, the nonmetro poverty rate improved by 0.9 percentage points, to 17.2 percent in 2015.
This is not to say that the report indicates a rosy picture for nonmetro America versus metropolitan American. The gap between nonmetro and metro median household and poverty remained flat. That means that rural American continues to lag significantly behind the rest of the nation by these economic measurements. (See more about the interpretation of the poverty rate below.) But the problem didn’t get worse from 2014 to 2015.
So why did the Census Bureau’s report “Income and Poverty in the United States: 2015” state that nonmetropolitan America median household income had dropped by 2 percent and poverty had increased?
The discrepancy involves two factors: polling methodology and the definition of “rural.” (I know this reference to methodology is likely to cause the eyes of some readers to glaze over. I understand. I had the same response when sitting through my research methods class in 1985.)
First, between the release of the 2014 and 2015 numbers, the federal Office of Management and Budget revised its classification of counties as metropolitan or nonmetropolitan. This is the definition that the Census Bureau used in this income report. The OMB makes this reclassification every 10 years, as the population and commuting patterns of U.S. counties change.
When the new classifications were released in 2013, 113 counties that were previously nonmetropolitan became metropolitan. Thirty-six counties went the other way, from metropolitan to nonmetropolitan. Here’s the story we ran reporting the changes.
The problem in the Census report is that it used the old metropolitan classification list for the 2014 data and the new one for the 2015 data. The result: apples and oranges.
So, this negates the report that nonmetropolitan (what we commonly refer to as “rural”) fell from 2014 to 2015.
That leaves us with the second part of the correction. If the Census report didn’t tell us that rural incomes decreased, how can we report that they actually grew?
The Center for Budget and Policy Priorities dug to find a separate Census report that gives a more accurate picture of how things changed from 2014 to 2015. That report, based on the American Community Survey (a more thorough representation of the U.S. population) compared 2014 and 2015 using the same list of counties (apples to apples). That report shows that median household incomes grew in nonmetro counties at about the same rate as they did in metropolitan counties. And it showed that nonmetro poverty fell at about the same rate as metro poverty over the same period.
One more note on the nonmetropolitan poverty rate: The Center for Budget and Policy Priority also notes that if we take into account the difference in cost of living between rural and urban areas, and we count government support programs, rural areas actually have a lower poverty rate:
Under the official measure, poverty is much higher in non-metro areas than in metro areas. But under the Census Bureau’s Supplemental Poverty Measure (SPM), which accounts for most government benefits and adjusts for local cost-of-living differences, the poverty rate is actually lower in non-metro areas (13.2 percent) than in metro areas (14.5 percent). Although some analysts do not fully agree with the SPM’s approach to geographic adjustment, many analysts agree that either those adjustments or another, similar approach makes sense.