[1/11 update at bottom]
It looks as though takeup of Cost Sharing Reduction (CSR) subsidies by eligible buyers on HealthCare.gov is up slightly this open season, so far. CSR reduces deductibles, copays and maximum out-of-pocket costs for buyers with incomes below 250% of the Federal Poverty Level (FPL) -- but only if they select silver plans, the second-cheapest of four metal levels.
As of the end of open season for 2015 enrollment last February, 76.8% of CSR-eligible HealthCare.gov customers had selected silver plans and so accessed the benefit. For 2016 enrollees through 12/26/15, according to HHS's latest enrollment report, CSR takeup is up to 78.4% on the federal exchange.
Here's the basic calculation for CSR takeup among 2016 enrollees:
Total hc.gov enrollments through 12/26 8,524,935
Total enrollments with income known 7,844,313
Total CSR eligible (79% of income known) 6,197,007*
Total with CSR (57% of all customers) 4,859,212*
Percentage of CSR-eligibles with CSR 78.4%
The uptick in CSR takeup comes in the face of a fair headwind, because HealthCare.gov customers are a bit wealthier on average this year, probably because four states on the platform have belatedly expanded Medicaid eligibility, removing the lowest tier of private plan customers from the market in those states. In states that have refused the expansion, eligibility for subsidized private plans begins at 100% FPL, compared to 138% FPL in expansion states. The lower the income, the higher the CSR takeup, because a) the benefit is strongest for those with incomes below 150% FPL, and b) silver plans are much cheaper for customers in the lowest income tier -- capped at 2% of income for those up to 138% FPL, compared to 8.15% at 250% FPL
HHS statistics don't tell us exactly how many buyers are in the should-have-been-in-Medicaid income range (100-138% FPL).They do report the percentage in the 100-150% FPL range, though, and this year it's down from 40% as of last February to 37% through 12/26. The percentage of all buyers who are CSR-eligible (those with incomes under 250% FPL) is down this year from 83% to 79%. (Those percentages exclude those for whom income is unknown, which has risen from 6% to 8%. Buyers who don't report income are by definition ineligible for subsidies.)
Thus, the percentage of CSR-eligible buyers on hc.gov who access the benefit is higher this year than last even though the percentage of all buyers who accessed CSR is a bit lower, dropping from 60% at the end of open season last year to 57% among 2016 enrollees.
While CSR provides a major affordability boost to buyers with incomes up to 200% FPL, it weakens sharply beyond that point. CSR-enhanced silver plans provide an actuarial value of 94% to those with incomes up to 150% FPL, 87% for those in the 150-200% FPL range, and 73% for those from 200-250% FPL -- just 3 points higher than silver unenhanced by CSR. Accordingly, takeup falls off a cliff at 201% FPL, as some state-issued enrollment reports show. For buyers up to 200% FPL, takeup is probably at least 5 percentage points higher than the overall CSR takeup rate. In some states, it's more like ten points higher.
It is good news that higher premiums are not driving more enrollees into bronze plans, which have sky-high deductibles and an actuarial value of just 60% (compared, again to 94%, 87% or 73% for CSR-eligible silver buyers). Then again, overall premium hikes should only drive enrollees into bronze plans if bronze plans go up less than silver. Subsidized buyers -- 87% of all hc.gov customers -- are insulated from price hikes and not really helped by price drops, at least with regard to premium: the benchmark silver plan will cost every subsidized buyer the same percentage of income as every other buyer at her income level. Only the spread between the benchmark (second cheapest) silver plan and other plans the shopper may want to buy should affect decision-making. For example, if the cheapest silver plan in a given area is much cheaper than the benchmark, buyers in that area can get CSR at a discount.
One more sidelight: 87.4% of hc.gov enrollees who are eligible for premium subsidies are also eligible for CSR. That's worth thinking about from a couple of perspectives. First, if CSR were an automatic benefit and not dependent on buyers' metal level choice, almost all subsidized buyers would get it -- and underinsurance might be less of a problem among the current enrollee population. On the other hand, to date takeup of subsidized private plans has been much higher among lower-income buyers: the percentage of buyers in the 200-300% FPL range (who are eligible for either weak CSR or no CSR) should be higher, as the census shows a high concentration of uninsured in that income range. States that want to make the ACA work and provide affordable care to their all their residents should consider sweetening the pot on their own dime in this modestly higher income range.
A final footnote. The CSR takeup rate may be somewhat higher among those actually eligible for subsidies,as a small number of enrollees with income in subsidy range get no subsidy. California, in its June 2015 member profile (available here), lists the number subsidized and unsubsidized at each income level. About 3% of enrollees under 250% FPL in the state are unsubsidized. That may be because they have an offer of insurance from an employer that's technically affordable, or perhaps because they can't verify immigration status. If 3% of those under 250% FPL nationwide are subsidy-ineligible -- and we certainly don't know that they are -- that would bring CSR takeup to 80.7%.
UPDATE, 1/11: Jed Graham points out that late enrollees in last year's open season skewed toward bronze selection, and that overall bronze selection on hc.gov is ahead of the pace from the comparable point last year:
UPDATE 2, 1/13:A look at 2016 premium tables from the Robert Wood Johnson Foundation shows that bronze plan premiums rose more this year on a percentage basis than silver premiums did. The average bronze premium rose 13.2%, vs. 11.3% for average silver, while the cost of the cheapest plan at each metal level in each area rose an average of 13.6% for bronze and 12.4% for silver. Perhaps that convergence has pushed a few more enrollees toward silver. For subsidized buyers, price spreads from the benchmark matter more than absolute prices. On the other hand, in dollar terms, both spreads widened slightly.
----
* Since HHS provides only rounded percentages for those in each income level and those who have accessed CSR, the precise-looking totals calculated below for "CSR eligible" and "CSR accessed" are actually estimates. But the takeup percentage should be on target.
Related:
ACA enrollment 2016: pool somewhat wealthier
The ACA's uncertain shield against underinsurance: A CSR compendium
It looks as though takeup of Cost Sharing Reduction (CSR) subsidies by eligible buyers on HealthCare.gov is up slightly this open season, so far. CSR reduces deductibles, copays and maximum out-of-pocket costs for buyers with incomes below 250% of the Federal Poverty Level (FPL) -- but only if they select silver plans, the second-cheapest of four metal levels.
As of the end of open season for 2015 enrollment last February, 76.8% of CSR-eligible HealthCare.gov customers had selected silver plans and so accessed the benefit. For 2016 enrollees through 12/26/15, according to HHS's latest enrollment report, CSR takeup is up to 78.4% on the federal exchange.
Here's the basic calculation for CSR takeup among 2016 enrollees:
Total hc.gov enrollments through 12/26 8,524,935
Total enrollments with income known 7,844,313
Total CSR eligible (79% of income known) 6,197,007*
Total with CSR (57% of all customers) 4,859,212*
Percentage of CSR-eligibles with CSR 78.4%
CSR swims upstream
The uptick in CSR takeup comes in the face of a fair headwind, because HealthCare.gov customers are a bit wealthier on average this year, probably because four states on the platform have belatedly expanded Medicaid eligibility, removing the lowest tier of private plan customers from the market in those states. In states that have refused the expansion, eligibility for subsidized private plans begins at 100% FPL, compared to 138% FPL in expansion states. The lower the income, the higher the CSR takeup, because a) the benefit is strongest for those with incomes below 150% FPL, and b) silver plans are much cheaper for customers in the lowest income tier -- capped at 2% of income for those up to 138% FPL, compared to 8.15% at 250% FPL
HHS statistics don't tell us exactly how many buyers are in the should-have-been-in-Medicaid income range (100-138% FPL).They do report the percentage in the 100-150% FPL range, though, and this year it's down from 40% as of last February to 37% through 12/26. The percentage of all buyers who are CSR-eligible (those with incomes under 250% FPL) is down this year from 83% to 79%. (Those percentages exclude those for whom income is unknown, which has risen from 6% to 8%. Buyers who don't report income are by definition ineligible for subsidies.)
Thus, the percentage of CSR-eligible buyers on hc.gov who access the benefit is higher this year than last even though the percentage of all buyers who accessed CSR is a bit lower, dropping from 60% at the end of open season last year to 57% among 2016 enrollees.
While CSR provides a major affordability boost to buyers with incomes up to 200% FPL, it weakens sharply beyond that point. CSR-enhanced silver plans provide an actuarial value of 94% to those with incomes up to 150% FPL, 87% for those in the 150-200% FPL range, and 73% for those from 200-250% FPL -- just 3 points higher than silver unenhanced by CSR. Accordingly, takeup falls off a cliff at 201% FPL, as some state-issued enrollment reports show. For buyers up to 200% FPL, takeup is probably at least 5 percentage points higher than the overall CSR takeup rate. In some states, it's more like ten points higher.
It is good news that higher premiums are not driving more enrollees into bronze plans, which have sky-high deductibles and an actuarial value of just 60% (compared, again to 94%, 87% or 73% for CSR-eligible silver buyers). Then again, overall premium hikes should only drive enrollees into bronze plans if bronze plans go up less than silver. Subsidized buyers -- 87% of all hc.gov customers -- are insulated from price hikes and not really helped by price drops, at least with regard to premium: the benchmark silver plan will cost every subsidized buyer the same percentage of income as every other buyer at her income level. Only the spread between the benchmark (second cheapest) silver plan and other plans the shopper may want to buy should affect decision-making. For example, if the cheapest silver plan in a given area is much cheaper than the benchmark, buyers in that area can get CSR at a discount.
One more sidelight: 87.4% of hc.gov enrollees who are eligible for premium subsidies are also eligible for CSR. That's worth thinking about from a couple of perspectives. First, if CSR were an automatic benefit and not dependent on buyers' metal level choice, almost all subsidized buyers would get it -- and underinsurance might be less of a problem among the current enrollee population. On the other hand, to date takeup of subsidized private plans has been much higher among lower-income buyers: the percentage of buyers in the 200-300% FPL range (who are eligible for either weak CSR or no CSR) should be higher, as the census shows a high concentration of uninsured in that income range. States that want to make the ACA work and provide affordable care to their all their residents should consider sweetening the pot on their own dime in this modestly higher income range.
A final footnote. The CSR takeup rate may be somewhat higher among those actually eligible for subsidies,as a small number of enrollees with income in subsidy range get no subsidy. California, in its June 2015 member profile (available here), lists the number subsidized and unsubsidized at each income level. About 3% of enrollees under 250% FPL in the state are unsubsidized. That may be because they have an offer of insurance from an employer that's technically affordable, or perhaps because they can't verify immigration status. If 3% of those under 250% FPL nationwide are subsidy-ineligible -- and we certainly don't know that they are -- that would bring CSR takeup to 80.7%.
UPDATE, 1/11: Jed Graham points out that late enrollees in last year's open season skewed toward bronze selection, and that overall bronze selection on hc.gov is ahead of the pace from the comparable point last year:
One thing that's clear from the latest enrollment period is that those who sign up close to the wire are much more likely to choose bronze plans — 29% of HealthCare.gov signups did in last year's final month vs. 19% in the first two months. That makes sense, because people who drag their feet but sign up to avoid paying a fine are more likely to opt for the cheapest plan available.I did not know that there was a late 'bronze surge' last year, and that does compromise my implied conclusion here. However, while bronze selection is up slightly from the same point last year, so is silver selection, 71% to 70% -- a difference perhaps too small to be meaningful, but silver at least is not falling. The difference comes out of platinum selection, which is down to 1% from 3% last year. Also, for January 2015, we do not have CSR takeup numbers. Whether CSR increases, drops or stays the same remains to be seen.
Thus, the bronze share of plan selections is sure to rise over the next month, yet bronze is already more popular among HealthCare.gov signups this year than it was in the first two months of enrollment a year ago: 21% have opted for bronze vs. 19% at the same point.
UPDATE 2, 1/13:A look at 2016 premium tables from the Robert Wood Johnson Foundation shows that bronze plan premiums rose more this year on a percentage basis than silver premiums did. The average bronze premium rose 13.2%, vs. 11.3% for average silver, while the cost of the cheapest plan at each metal level in each area rose an average of 13.6% for bronze and 12.4% for silver. Perhaps that convergence has pushed a few more enrollees toward silver. For subsidized buyers, price spreads from the benchmark matter more than absolute prices. On the other hand, in dollar terms, both spreads widened slightly.
----
* Since HHS provides only rounded percentages for those in each income level and those who have accessed CSR, the precise-looking totals calculated below for "CSR eligible" and "CSR accessed" are actually estimates. But the takeup percentage should be on target.
Related:
ACA enrollment 2016: pool somewhat wealthier
The ACA's uncertain shield against underinsurance: A CSR compendium
Thanks for all your great coverage of health care and insurance policy! What are some good sources to go to on the issue of underinsurance? It seems like it's the policy and political issue on the horizon, but I'm eager for a treatment of the subject that really deals with the assumptions, definition, and empirical reality behind the topic/concept. At a most basic level, is the spectre of underinsurance mostly about controlling health care provider costs--making "insurance" and "sufficient" actuarial value broadly affordable? Or is underinsurance about changing the preferences of consumers or at least making sure they are crystal clear on what type of insurance they are buying (something closer to catastrophic, not comprehensive/low deductible). I've long been dissatisfied with the way the term "insurance" is used in the matter of health care, because it's confusing, unlike almost all other insurance markets, in which people are generally trying to mitigate against exceptional events, not find something offering substantial coverage of everyday out of pocket costs. Layering on top of that the judgment of under vs over insurance seems to make things even less technocratically clear, since it seems to me what's really being hashed out here is an ideological political dispute over to what extent all health care expenses should be, at base, socialized. And I'm speaking here as a strongly left liberal. Have I confused myself here or is the terminology itself sometimes confusing?
ReplyDelete