Thanks for joining me!
The link to a National Public Radio article below introduces an exceptionally interesting and extremely significant GIS-exercise. Harvard economist Raj Chetty with a team created a tool that marries individual level U.S. Census Bureau data of ca. 20 million US residents born between 1978 and 1983 with data from the Internal Revenue Service plus county & smaller level data from additional national databases and superimposes the result on a high detail interactive US map. [https://www.npr.org/2018/10/01/649701669/the-american-dream-is-harder-to-find-in-some-neighborhoods]
By clicking the map in the article, you can zoom it in and out for your wanted areal detail. It looks at how the children born in the different US counties, cities and finer subareas into homes representing three household income percentiles (25th, 50th and 75th], five races [Black, White, Hispanic, Asian and American Indian] and the two genders are succeeding in life, measured as their current [at age 35…40] household income level.
The Opportunity Atlas is so fascinating and intellectually challenging that I have spent several days playing with it, and I am certainly not over yet. The first and obvious finding is that all else [childhood home income level, race and gender] being equal the state, county, city, suburb, i.e., geographic location where you are born still matters a great deal.
There are a few details, however, which need to be considering when making comparisons within the database, as they may have significant impact on the value of each message and its interpretations. I recommend reading at least the METHODS which you can find clicking in the upper left field in the Opportunity Atlas [https://www.opportunityatlas.org/]:
The county on the map is the county where the child was born (1978-83), i.e., the current income [and other] data given for, e.g., Orange County, NC, does not represent the current population of that county but instead the current status of everyone who was born there in 1978-83, regardless of where they currently live. [my daughter, presently living in Northern Italy, was born there but missed this cohort by 2 years.]
The childhood home income percentiles are national, not specific for the location of the home or race of the child. This could blur the interpretation for some exceptionally poor, e.g., Native American reserve (?) areas, such as Sioux County, ND, where most households would be below the national 25th income percentile. For many outcome parameters the relative status in the regional/local income/social ladder, however, is more relevant than the absolute income level.
There is no universally applicable and comparable, available-from-census-statistics indicator of quality of life [QoL]. Relative social status would probably be the best. Available household income is a rather poor indicator of social status, years of education, for example, is better. But one has to live with the data that are available, and despite this limitation, I think this is a tremendous tool. Household income still remains a perfect measure of household income.
I began looking for two clearly different but non-extreme counties, one providing a clear advantage for the children to grow up [BLUE] relative to one low advantage [RED], (colors refer to the opportunity Atlas, not party voting preferences): Combining all races and both genders, the children born in Lincoln [County, NE], to parents at the 25th income percentile now have an average household income of 39 000 $/year, while the respective children born in Washington [County, MS], now earn 23 000 $/year. The respective numbers at the 75th percentile are 57 000 and 48 000 $/year. Not only do the households of the Lincoln born now earn in average 9 000 to 16 000 $ more per year, but also their income distribution is significantly narrower.
Particularly the children born into low income families in the low advantage states and counties are much more likely to be doomed to lifetime poverty than the children born to low income families in the high advantage states and counties. For the black children the current income difference in favor of Lincoln is bigger still. Those born black in Lincoln county now earn 83% of those born white, while those born black in Washington county earn only 40% of those born white. Being born into a black family is still a both absolute and relative economic disadvantage in America, but it is a much smaller disadvantage in certain states and counties than it is in others.
In some cases you find some surprising, even contradictory results, obviously due to the fact that some subgroups, e.g., Asian females born to low income families in Pasadena Ca are likely to be small – in fact their current household income turn out to be high, 84 k$/a, regardless of the income of the family they were born into. Such comparisons do not tell the whole story of success and happiness, but they do tell a lot.
One possible explanation that comes to mind is religion and ruling party. In the generally low advantage states [NC, SC, GA, FL, AL, TN, VA, KY, OH, MI, IN, MO, and LA – consistently republican in bold normal font, democrat in bold italics] 36.1% consider themselves evangelicals, 14.6% mainline protestants, 10.6% traditional black protestants, 13.4% catholics and 19.4% unaffiliated. In the generally high advantage states [MN, ND, SD, MT, ID, WY, NE and IA] there are respectively 24.4% evangelicals, 23.6% mainline protestants, 1.8% traditional black protestants, 19.0% catholics and 22.8% unaffiliated. The main differences appear to be more catholics and mainline protestants, less evangelical and traditional black protestants in the BLUE vs. RED states. Aside of the traditional black protestants, the differences are smaller than +/-50%. Religion does not appear decisive and party dominance seems to make little difference. Most of the democrat dominated states appear to fall between the low advantage [RED] and the high advantage [BLUE] providers.
The big picture is striking, though, and calls for more detailed analysis and, obviously, political action.
The Opportunity Atlas tool supports thousands of similar and different analyses. The Big Data behind Opportunity Atlas provides an absolutely fascinating playground, raises a lot more questions than answers and shoots down some presumed truths. Opportunity Atlas could and should become a broadly applied study-hypotheses generating tool for sociology, socioeconomics, and what not. To begin with, a rather easy to do what-if-analysis about how much more wealthy the US of A as a whole would be if all counties could provide their children as good foundations for life as 10% of the best do today would yield mind blowing results.
Finally. I do envy the American nationwide statistics and databases that cover with similar criteria the entire US. EU and UN organisations are struggling against nationalistic pride and other hidden political interests to create similar Europe-wide statistics, but we still remain light-years behind. I cannot even imagine a similar database and mapping tool first made possible, then be done and finally made publicly available here on our side of the Ocean.
Congratulations – now start using it.
