COVID-19 Coronavirus

The US Department of Labor just released the new claims for unemployment insurance filed last week.  The number is worse than expected 5 million new claims:

Unemployment Insurance Weekly Claims

Initial claims were 6,606,000 for the week ending 4/4 (-261,000).

Insured unemployment was 7,455,000 for the week ending 3/28 (+4,396,000).https://t.co/ys7Eg5LKAW

— US Labor Department (@USDOL) April 9, 2020

Since the start of the economic coma, close to 15 million people have filed for unemployment. There are probably millions more who are either filing this week for the first time or are still trying to file as antiquated and underfunded systems are crashing.

One of the major concerns for folks who have been laid off is their health insurance:

6 million new unemployment claims again this weekPost + Comments (15)

One of the major provisions of the CARE ACT is a four month enhanced unemployment bump payment of $600 per unemployed individual per week. This is federally funded and it is in addition to regular unemployment insurance. It will have complexity for low income individuals who are currently insured through the ACA.

I want to run through a set of good faith scenarios with the following baseline: single individual working forty hours a week at $10 per hour for all of the first quarter. This person was laid off on April 1 and is trying to provide good faith estimates to Healthcare.gov on a timely basis of their changing circumstances. I am assuming this individual will qualify for regular unemployment insurance at 50% of average weekly wage ($200/week) in addition to the four month bump payment. I am assuming that they go back to work after four months of unemployment.

I am assuming this person started the year with a good faith estimate that they would have constant earnings and no major life events. Under these assumptions, they entered an income of 163% Federal Poverty Level (FPL) for a single individual and their net premium to buy the benchmark silver plan would be $86 per month.

Scenario 1: Update income on a monthly basis where a given month’s estimated income is also the year’s estimated income.

There is a lot happening here.  Three different actual net premiums will be collected over the course of the year.  The first is $86 per month for the first three months of the year.  However with the enhanced $600/week unemployment benefit that begins to pay out in April and ends on the last day of July, take home pay increases dramatically and therefore eligibility for APTC decreases.  The monthly expected premium for the benchmark plan goes to $334 a month.  However, the administrative apparatus tries to balance out annual APTC over the course of the year so that there is no overpayment or underpayment by the end of the year.  So the system thinks that recipient received way too much APTC in the first quarter, and tacks on an additional amount to the benchmark base so that the monthly premium for the benchmark plan now increases to $416 per month.

However, we’re also assuming that the individual in question goes back to their job on August 1 where their new expected benchmark net premium is back to $86 per month.  However Healthcare.gov thinks that they’ve been massively undersubsidized now so the calculation now tries to pay out more subsidy so that the IRS does not have to write a bigger refund check in the Spring of 2021 for 2020 APTC.  Benchmark premiums are now $0.

There is a lot of yo-yoing happening when a person assumes that their monthly income is annualized in their estimates.  Yo-Yos are bad.

Scenario 2:  Perfect foresight and re-annualization

Here is the perfect behavioral response with perfect foresight.  Annual income increases in April above the initial estimate and the individual perfectly forecasts that they go back to work on August 1.

The net premium yo-yo is far less. There is one jump from $86 to $195 per month. The four months of the enhanced $600/week payments have health insurance premiums cost 5.6% of monthly income. That is not too bad. However, since Q1 has overpayment of APTC to the individual, and the reconciliation is spread out over the remaining nine months, the last five months of the year have the benchmark net premium come out to be 11.3%. That is unaffordable.

Predicting anything, especially the future, is hard.

And that is true in normal times where the near past is usually somewhat similar to the present and the near future. I don’t think anyone can say that the near past is correlated with the near future in 2020 or that year over year comparisons are even facially plausible in a pandemic and an economic coma.

Right now, we are asking for the fundamentally impossible for people to predict their incomes when that prediction is fundamentally correlated with vaccine development interacting with aggressive and effective contact tracing to quarantine people who will be potentially exposed to new infections in the future.  The best epidemiologists in the world can’t make that prediction within a two or four week time frame.  It is absurd to expect someone who is making $10 an hour to be able to perfectly project when the economy will resume normal(ish) operations and if the recovery of the re-opening will be V or U or slightly drunk L shaped.

So what is the policy solution?

One approach could be for the federal government to either waive or significantly reduce repayment of APTC clawbacks for anyone who collected a single week of enhanced, $600/week federal unemployment payments.  Low income workers will catch a break financially and cognitively as the management load will decrease.  These workers are also the class of individuals who are most likely to spend any additional cash so this should be fairly effective stimulus when the economy needs stimulus instead of merely social insurance.

SEP, Estimating incomes and APTC clawbacksPost + Comments (11)

On Monday, a team of epidemiologists, infectious disease specialists, actuaries, data scientists and researchers in North Carolina released a brief on hospital capacity in the state**. They used an ensemble of three different modeling approaches to project hospital need and ICU bed need in the state through June 1. They model two scenarios. The first is maintaining current policies throughout the analysis period. The second lifts social distancing restrictions at the end of April and does not replace those policies with substitutes of equal effectiveness. The forecast is based on the probability that capacity is exceeded.

There is good news in the brief:

Under current policies, our models suggest that the volume of available acute care beds throughout the state will be sufficient to handle growing COVID-19-related case volume in the next few weeks. In the second half of April, estimates for the near-term suggest a substantial increase in confirmed state case volumes, to as many as 5,500-6,500 confirmed cases by April 15, up from approximately 2,402 confirmed cases on April 4.6 The probability that acute care bed demand will outstrip available supply will likely increase by mid-April but remains low (<10%) as the available capacity is likely sufficient.

Policy matters. Social and physical distancing is working to flatten the curve in the state. As long as distancing is maintained, capacity is likely to be sufficienct even as the state peaks. Avoiding current infections and thus avoiding future hospitalizations dominates any capacity response of adding new beds to the hospital systems in the state.

The difference in probabilities of demand exceeding bed supply across the surge levels is approximately 2 to 4 percentage points.

Those few points are very valuable points and are worth pursuing but another few hundred beds will be swamped by half a million avoidable infections.

To put the findings in context, the models’ composite estimates indicate that approximately 750,000 people in North Carolina may be infected with the virus by the end of our forecast period, June 1, 2020. This is if the social distancing policies are fully lifted at the end of April and are not replaced by other policies with equal assumed effectiveness to reduce transmission. On the other hand, if the same policies (or some other policies with similar effect) remain in place, the composite estimates indicate that an estimated 250,000 North Carolinians may be infected by June 1.

Until there is massive, rapid, no barrier access to testing that has few false negatives with accompanying public health contact tracing and mass quarantining in place, opening up a state just guarantees a massive shut down and overwhelmed hospitals in a few doubling cycles again. Even as we may be approaching local peaks, those peaks are only peaks because significant chunks of the nation started to shut down in early to mid-March to avoid millions of infections even as tremendous numbers of people are currently and soon to be infected. Opening back up again will just recreate the infection storm from a much higher base of currently infected individuals at the end of April compared to the base of people who were infected at the start of March.

** Disclosure: Most of the authors of the brief are some combination of co-workers, colleagues, co-authors, friends and beer drinking buddies of mine.

Wake me up when September comes: Physical Distancing in North Carolina editionPost + Comments (15)