I was disappointed by what I thought were bonehead conclusions in a recent article about the effectiveness of education in changing life outcomes. I thought the data cited in the article provided clear evidence that more education is associated with a better future, and this applied to both rich and poor.
In my analysis, I described another chart that appeared in a source paper mentioned in the article. The chart shows outcomes for sons who were born to fathers who were in the poorest 10% of the United States and Canada.
What was missing in the Washington Post article was the "before" chart of the before-and-after comparison. I have simulated the "before" picture for the fathers-and-sons chart as well.
Naturally, all the sons of fathers in the bottom decile are themselves in the bottom decile at the start. The question is, what does the "after" chart look like? How many of the sons remain in the bottom 10% as adults? How many make it up to the 2nd decile, and beyond? Maximum credit for those who venture to provide estimates for all ten deciles for at least one country.
EDIT: the solution
I have a thought: how unlikely is it, a priori, for a bottom-tenth-son to be in the bottom decile in the "after" simply based on the intertia of new immigrants (to the US) replacing the bottom decile every generation? In other words, would they not have to actively go "backwards" in order to remain where they are and not be at least somewhat supplanted (in the bottom decile) by new immigrants? I haven't thought that out very far, but I will use it as the basis for my guess. Bottom: 5%
2nd: 15%
3rd: 20%
4th: 20%
5th: 10%
6th: 10%
7th: 10%
8th: 5%
9th: 5%
10th: trivial In other words, I'm guessing (with considerably less than 5 minutes of thought sadly) that it's reasonably easy to get somewhere above the bottom decile, but quite difficult to get past about the 7th. I'm probably comically wrong.
I like immigration, but it is a disaster for econometrics. People, would you please stand still while we measure you? You are probably right that most immigrants arrive into the lower deciles. A hundred or so wealthy EB-5 newcomers won't skew this measure much. But to be consistent, we should then assume that many of the fathers who started in the bottom 10% were themselves immigrants. I don't see any reason to think that future immigrants would be poorer than past immigrants, so if bottom-percenters of any background find themselves moving up the ladder, it is likely for the same reasons that natives advance, and not simply an artifact of poorer and poorer new arrivals. -- Hmm, it occurs to me that my argument is not so good in light of population growth. If immigrants are constantly increasing the number of people in the bottom deciles, it will tend to shift the goalposts so that most people's "place in line" improves even if their income number does not change. (On second second thought, this applies even if all the immigrants are equally poor. Also, a higher rate of reproduction among poorer deciles would contribute to this distortion.) So you are probably wise to account for some "decile inflation" given that U.S. and Canadian policy kinda sorta welcomes poor foreigners.
Below is some context about the data provided in the paper. The result is a longitudinal dataset in which synthetic individuals, part actual CNLSY data and part imputed data, pass through five life stages from birth to adulthood. This includes 5,783 children from the CNLSY, born between 1971 and 2009. The paper was published by The Federal Reserve Bank of Boston.The SGM [Social Genome Model] is constructed using two data sets from the Bureau of Labor Statistics' National Longitudinal Surveys. Our primary data set is the ‘Children of the NLSY79’ (CNLSY). It represents children born mainly in the 1980s and 90s, and is the source of our data for the birth, early and middle childhood, and adolescent stages. No respondent in the CNLSY is yet old enough to track through adulthood, so we impute their adult values with help from a second dataset: the ‘National Longitudinal Survey of Youth 1979’ (NLSY79).