Off the Kuff Rotating Header Image

May 17th, 2022:

DMN/UT-Tyler: Abbott 46, Beto 39

Here’s the story, which I currently can’t access. A very brief summary of it is in this Current article. The data is here and I’m going to riff on that, with references to the February version of this poll, for which the data can be found here. I will note that there are some primary runoff results in this sample, and I am ignoring all of them – that kind of polling is too tricky to be worth worrying about.

“In a race for Governor would you vote for Governor Abbott, Beto O’Rourke, or someone else?” I’ll generally be quoting the poll questions, which thankfully are the same in each sample. In May, as noted in the post title, it’s 46-39 for Abbott, basically identical to the 45-38 Abbott result from February. The shape of those numbers are a bit different. In February, possibly because both Beto and Abbott were in contested primaries, there was a considerable amount of crossover support for each, Dems were only 76-16 for Beto, while Rs were just 76-11 for Abbott. In May, those numbers were 82-9 among Dems for Beto and 85-7 for Abbott among Rs. Independents were 36-29 for Abbott in February and show as 16-6 for Abbott now, with 29% going to the Libertarian (there is a Green candidate named as well, who also gets 6%) and an astonishing 38% for “someone else”. This has to be a mangling of the data – among other things, given the size of the Indy subsample, it would have put the Libertarian candidate at nearly 10% overall, but the topline result gives him just 3%. Most likely, the 38 is for Abbott and the 29 is for Beto, or possibly all of these numbers are just wrong. I will shrug and move on at this point.

For approval numbers, President Biden checks in with 39-58 approval, which is obviously not good. Greg Abbott is also underwater at 46-50, while Beto has a 42-44 approval rating, which is the only one of the three to improve since last time. It was 39-57 for Biden, 50-46 for Abbott, and 40-46 for Beto in February.

Weirdly, Dan Patrick has 50-41 approval, and Ken Paxton has 42-41. Usually, Abbott does better in approvals than any other Republican, in part because fewer people have opinions about the rest of them. A separate question about Paxton asks “do you agree or disagree that he (Paxton) has the integrity to serve as attorney general?”, and it’s 30 for agree, 37 disagree, and 33 unsure. He was at 34-33-33 in February, so a bit of a dip there.

For some other questions of interest, the numbers are not bad for the Dems, and usually a little better than they were in February.

“If the general election was today, would you vote for a Republican candidate or Democratic candidate for the Texas House?” That was 49-48 for Republicans in May, 52-45 for Republicans in February.

“On orders from Governor Abbott, Texas Child Protective Services recently began investigating families who provide gender-affirming care to transgender children. Was this action” needed or unnecessary, with various reasons for each? There were three sub-options for each of those choices, and if you add them up it comes to 52-48 combined for “unnecessary”. Honestly, that’s better than I expected. There was no February comparison for this one, as that order had not yet been given at that time.

“Should the Supreme Court overturn its Roe v. Wade decision and allow states to decide abortion policy?” This was 53-46 for “no it should not be overturned” in May, and 50-47 in February. Again, a little better than I might have thought, and a tick up from before, which is to say before the draft opinion got leaked. Put those numbers in your back pocket for the next time someone claims that Texas is a “pro-life” state.

“Do you agree or disagree that K-12 teachers should be permitted to discuss how historical examples of discrimination in our laws apply to inequalities today?” Here, 61-24 strongly or somewhat agreed in May, and it was 59-22 for Agree in February. That means that for abortion, trans kids, and book banning, the Republican position is the minority one. Obviously, one poll and all that, but there’s nothing to suggest Dems should be running scared on any of this. Quite the reverse, in fact.

Now as we’ve said a zillion times, it’s one poll, opinions on issues often don’t drive voting behavior, and we’re still months away from an election where many other factors will affect the outcome. I’m quite scared of another COVID wave, especially if Congress doesn’t get some more funding for vaccines and treatments and whatever else passed in the very near future. But for now, and bearing in mind that it’s still a 7-point lead for Abbott, the numbers ain’t that bad. We’ll see what other polls have to say.

SCOTUS asked to again block that stupid social media censorship law

Please save us from the lawless Fifth Circuit. Having to make such an ask of this SCOTUS sure is a jaw-grinding experience.

Lobbying groups representing Facebook, Twitter, Google and other tech companies filed an emergency request with the U.S. Supreme Court on Friday, seeking to block a Texas law that prohibits large social media platforms from banning users based on their political views.

The Texas law went into effect on Wednesday when the 5th U.S. Circuit Court of Appeals granted the state’s request for a stay of a district judge’s injunction blocking the law.

The law forbids social media companies with more than 50 million active users per month from banning members based on their political views and requires them to publicly disclose how they moderate content.

[…]

Internet lobbying groups NetChoice and the Computer & Communications Industry Association filed a lawsuit against the measure, and U.S. District Judge Robert Pitman in Austin, Texas, issued a preliminary injunction in December.

Pitman had found that the law would harm social media companies’ free speech rights under the First Amendment of the U.S. Constitution.

The tech groups, in their emergency request, asked the Supreme Court to “allow the District Court’s careful reasoning to remain in effect while an orderly appellate process plays out.”

See here for the previous update, and here for a more detailed analysis of why the Fifth Circuit’s no-words ruling was so bad. You know how much faith I have in this court to ever do the right thing, but maybe this was a bridge too far. Maybe. Ars Technica and The Verge have more.

“Silver loading”

I did not know this.

Sen. Nathan Johnson

The latest state to take strong action to improve affordability in its ACA marketplace is … Texas?

[…]

But let me explain some background details first. Below 200 percent of the federal poverty level (that is, $25,760 for an individual or $53,000 for a family of four in 2022), benchmark silver plans on the exchanges are both very cheap—costing nothing up to 150 percent of FPL, and just 0 to 2 percent of income between 150 and 200 percent of FPL—and further enhanced with cost-sharing reduction (CSR) subsidies, which make them roughly equivalent to platinum plans.

At incomes over 200 percent of FPL, however, the marketplace’s offerings will be greatly improved by Texas’s new state-enforced pricing regime.

More than two-thirds of Texas enrollees in ACA plans (69 percent) have incomes below 200 percent of FPL. But thanks to the more generous subsidies created last March through 2022 by the American Rescue Plan—which removed the prior income cap on subsidy eligibility—2022 enrollment in Texas grew fastest at higher incomes. Enrollment at incomes over 200 percent of FPL rose 57 percent, to 566,000. Those enrollees stand to gain most by the change in law. At higher incomes, silver plan deductibles average over $4,500 (though many services are not subject to the deductible). Gold plan deductibles, by contrast, average $1,600.

So if gold plans have lower deductibles, why should they be cheaper than silver plans? As mentioned above, for low-income enrollees, CSR raises the value of silver plans (by reducing out-of-pocket costs) to a roughly platinum level, with average deductibles under $200 for the lowest-income enrollees and $800 at the next level (150 to 200 percent of FPL). In Texas, 89 percent of silver plan enrollees have incomes below 200 percent of FPL, and so the average silver plan sold in the state really does match a platinum-level “actuarial value,” or the percentage of the average enrollee’s costs the plan is designed to cover.

During the Obama administration, the federal government reimbursed insurers directly for providing CSR, and silver plans were priced as if no CSR were attached. When Trump abruptly cut off those direct CSR payments in October 2017, however, almost all state regulators responded by allowing or encouraging insurers to price the value of CSR directly into silver plans—a process that came to be known as “silver loading.”

Here’s where the quirk comes in. Because ACA premium subsidies, designed so that the enrollee pays a fixed percentage of income, are set to a silver plan benchmark (specifically, the second-cheapest silver plan), higher silver premiums mean higher premium subsidies across all plans—and discounts for subsidized buyers in bronze and gold plans.

But silver loading has stopped halfway. As the author of the “focused rate review” bill, Texas state Sen. Nathan Johnson, points out in the bill analysis, “insurers have not approached silver loading in a uniform manner. The resulting misalignment of premiums has caused Texans to lose out on hundreds of millions of dollars in federal marketplace subsidies, making coverage less affordable.”

Why? Since a majority of marketplace enrollees have incomes below 200 percent of FPL, silver is still the most popular metal level, and so when insurers are in a competitive market, the price of a silver plan gets competed down (though when there is just one insurer, or a dominant insurer, they can and do take advantage of the rules, often to create huge discounts).* In most of the country, gold plans are still priced well above silver, and bronze plan discounts are not as big as analysts (including the Congressional Budget Office) expected them to be when anticipating Trump’s cutoff of direct payments for CSR. A handful of states, however, have effectively ordered insurers to fully price the value of CSR into silver plans.

With Johnson’s bill, Texas joined them. S.B. 1296 directed the Department of Insurance to, as the bill analysis phrases it, “focus its rate review in a manner that uniformly maximizes the benefits of silver loading, making coverage more affordable.”

On March 28, the Texas Department of Insurance issued a proposed rule to flesh out that legal directive. The rule directs insurers to use a “CSR pricing factor” of 1.35—that is, to price silver plans at 1.35 times what they would charge if there were no CSR. That rule effectively prices silver plans close to a platinum level. Gold plans are generally priced at about 1.2 times the cost of silver with no CSR.

The CSR pricing factor is slightly below the level implemented in 2022 in New Mexico, 1.44, which led to the lowest-cost gold plan in each rating area being priced an average of 11 percent below the benchmark silver plan. In New Mexico markets, several gold plans are priced below benchmark. Accordingly, in 2022, 69.5 percent of New Mexico enrollees with incomes above 200 percent of the federal poverty level chose gold plans, compared to 24.4 percent of Texas enrollees above the same threshold.

The rest of the story is about how momentum for this idea came from a couple of conservative actuaries plus a former aide to Greg Abbott now working at the non-partisan think tank Texas 2036, who were able to get Republican buy-in for the idea, and the House sponsor for Sen. Johnson’s bill, Rep. Tom Oliverson. The details are wonky and I definitely don’t understand all of them, but the bottom line is that in Texas another 200,000 people now have access to affordable heal insurance. And the guy that wrote the bill to make it happen is the guy that escorted the vile Don Huffines out of the Senate in 2018. Not too shabby. Charles Gaba, who brings even more wonkiness, has more.

2022 primary runoff Day One EV report: Lots of mail ballots

No news story yet as I write this, so let’s just jump right in. Here is the Day One EV report for the primary runoffs. Note that there are only five days of early voting in the runoff – as of this morning, there are now four days left – so I won’t be doing any comparisons with March, and since every runoff is its own little universe I won’t compare with previous years. You can see the final EV report for March here, though do note that several thousand more mail ballots arrived between the Friday and the following Tuesday – in total, there were about 29K total mail ballots returned as of the final results. Just over 50K mail ballots were sent out to the primary voters – we know what happened to a bunch of them, but however you want to think about it a bit less than sixty percent of all mail ballots were successfully returned.

Here are the totals so far after the first day of early voting for the runoffs:


Party    Mail   Early    Total
==============================
Dem    20,357   3,050   23,407
GOP    20,733   8,049   28,782

That’s 41K mail ballots returned, with just under 55K ballots being sent out, for a successful return rate close to 80% so far, and that will go up as more ballots come in. Maybe, just maybe, that’s a sign that the problems of March have been at least somewhat ameliorated. To be sure, these are people who almost certainly voted in March and thus have learned their own lessons from that experience. This is why I was so keen to see numbers from the May election, because that had to include a lot more first-timers. This is still an encouraging sign, even if it’s for a smaller population.

This also means that the main thing to watch for going forward is the in person voting population, as there aren’t that many mail ballots left to return and there won’t be any more sent out. I don’t feel like trawling through the past to see what the pattern for these five-day EV periods looks like, but I’d bet a dollar that Friday will be the busiest day. It’s probably not too busy now, so take advantage of the shorter lines while you can.