Drop in low income uninsured is partly due to drop in low income population

In 2015, I noted that Democrats had paid in political blood for devising a health reform scheme that primarily benefited people in the lower third of the national income distribution  (those with incomes below 200% of the Federal Poverty level), where more than half of the country's uninsured were concentrated pre-ACA.

Chance led me this morning to take a bird's eye view of how the income distribution of the uninsured changed from 2013 (pre-ACA) to 2017. This snapshot is based on the Census Bureau's annual September report, Health Insurance Coverage in the United States, for 2014 and 2017.

The percentage of the uninsured in the lowest income brackets has shrunk considerably. But that's in part because the percentage of the total population in those brackets has also shrunk.  In 2013, 39% of the population and 59% of the nation's uninsured were in households with incomes below $50k. In 2017, 49% of the uninsured and 34% of the total population were in households with incomes below that threshold.

Here is the breakout by income group.

Uninsured Population by Household Income, 2013 vs. 2017

Numbers in thousands

Year	Total pop	Total uninsured	Uninsured  less than $25,000	Uninsured 25-50$k	Uninsured $50-75k	Uninsured $75-100k	Uninsured more than $100k
2013	313,443	41,462	11,611	12,759	7,752	4,023	5,717
2017	323,156	28,543	6,482	7,618	5,570	3,546	5,328
Decrease		12,919	5,129	5,141	1,782	477	380
Percent decrease		31%	44%	40%	24%	12%	7%

Source: Census Bureau, Current Population Survey, Annual Social and Economic Supplements (CPS ASEC), 2014, 2015, 2017, 2018

The shift is exaggerated by apparent growth in household income. In 2013,  there were 94 million households with income over $100k. By 2017, there were 116 million. The total number in each cohort  up to $50-75k shrank in this period, while cohorts over $75k grew. The percentage of the total population in households with incomes under $75k shrank from 56.7% to 50.3%.

Here's the extent to which the uninsured population dropped in each income bracket in percentage terms. The gap between the higher and lower brackets in reduction of the uninsured population  is not as pronounced as in the raw change by bracket noted above.

Percent Uninsured by Household Income, 2013 vs. 2017

Year	All income  groups	Less than $25,000	$25-50k	$50-75k	$75-100k	More than $100k
2013	13.2%	20.9%	19.0%	13.3%	9.7%	6.1%
2017	8.8%	13.9%	12.3%	10.4%	7.9%	4.6%
% decrease	33.3%	33.5%	35.2%	21.8%	18.6%	24.6%

While the report on 2017 adjusts income brackets for inflation vis-a-vis the comparison year, 2016, I'm not sure how inflation affects the 2013-2017 comparison, especially since the CPS's income questions were redesigned for 2014. The report's breakout by income as a percentage of the Federal Poverty Level (FPL) shows a more even reduction in uninsured rates across income brackets -- in part, again, because population increased in wealthier cohorts and decreased in lower income ones.

In 2017, the percentage of the population with incomes under 200% FPL was 30.5%, down from 33.5% in 2013. The percentage with incomes above 300% FPL, conversely, was up from 50.6% in 2013 to 54.3% in 2017. Here is the extent to which the uninsured population decreased in each income FPL bracket, according to the CPS ASEC.

Percent uninsured by FPL, 2013 vs. 2017

Year	Total pop	0-100% FPL	100-200%	200-300%	300-400%	Over 400%
2013	13.2%	23.5%	20.4%	15.8%	10.3%	5.6%
2017	8.8%	17.0%	12.8%	10.9%	7.7%	4.3%
% decrease	33.3%	27.6%	37.2%	31.0%	25.2%	23.2%

I am frankly not entirely sure what to make of all this. The relatively small drop in the uninsured rate at incomes below the poverty level is in part affected by a spike upwards from 16.3% uninsured in 2016 to 17.0% in 2017. The rather strong drops at high FPLs are something of a surprise, and $50-100k looks weaker than the roughly corresponding 200-400% FPL. Perhaps the biggest surprise to me, if not to people who follow income stats closely, is an overall population movement  from lower to higher income brackets, in FPL as well as in household income. That at least is good news.