Yet another distributional analysis

The Hill has “details” on the latest “plan”-like scribblings of the Republican policy “wonks” on healthcare. There is one thing I want to look at before I start my morning coffee:

The core of the plan is a $2,500 tax credit that any citizen would be eligible for and use to purchase health insurance. The lawmakers say this gives flexibility to people, whether they get employer-based insurance or not, to more directly control their healthcare spending, for example by using a health savings account.

I’m looking at one of the sponsor’s web pages and I get very few more “details”

every American citizen is eligible to claim a $2,500 tax benefit as well as a $1,500 tax benefit per dependent minor. This benefit can be assigned to an employer, transferred to a Roth Health Savings Account, or advanced for annual distribution. With this benefit, individuals and families now have the freedom to use pretax dollars to plan and save for their health care futures.

Let’s look at the distributional consequences of this type of policy.

For people who make under 200% of FPL, pre-tax dollars aren’t too valuable as most of their dollars are minimally taxed.  For people making six figures and only have a kid or two at most, pre-tax dollars are fairly valuable as they are facing a much higher explicit marginal rate.  Worrying about pre-tax dollars is overwhelmingly an upper middle class to affluent problem.

More importantly it is the flat subsidy.

$208 a month is a decent subsidy.  In some regions that will buy the equivalent of a Silver plan with absolutely no out of pocket monthly premium.  That is fine for a healthy and young individual (as underwriting is back with a vengeance).  There are Silver plans for 40 year old non-smokers that cost under $200.  However, that same $200 a month Silver plan with a $3,500 deductible will cost a 63 year old $450 a month.  And odds are that 63 year old will need to use their policy a lot more than the 40 year old.

Furthermore, a flat subsidy is great for people who don’t need help.  I get my insurance through my employer and the visible premium payment is roughly two hours of pay per month for a Platinum like coverage for my family.  I don’t need help.  My family does not need help.  We already have access to good, high actuarial value, affordable coverage.

Families and individuals that are not mid-career professionals and are making under median income will see a far higher percentage of their income go to post-subsidy premiums.  The poorer you are, the higher the premium percentage is for a given level of individual risk.  And that is a major problem as the people who should bear the least risk are the one’s with the fewest available resources to mobilize in an oh-shit scenario.

TLDR: Comfort the comfortable



Trump-proofing the Republican nomination process in the future

This post is speculation. It assumes that Trump will lose and lose big in November and that the Republican establishment as defined by a variety of rules committees has the power and the will to institute changes to the Republican primary process to Trump-proof the process.

The easiest way for the Republican Party to Trump-proof itself is to stop lying to its supporters. The Republican Party elite is fundamentally not trustworthy to its base voters. The core example is the promise that a Republican House and a Republican Senate could force President Obama to unwind PPACA while he sat in the White House. That was not going to happen. Trustworthy elites won’t happen as there is too much money to be made from fleecing the rubes. Once we take policy honesty off the table, rule changes are the next step.

Trump is the delegate leader (and presumptive delegate majority holder once the process plays out) with a low proportion of the total vote.

He benefited from a split field and a rules system that allowed factional plurality leaders to amass delegate strength out of proportion to their actual vote counts. Winner take all elections with more than two candidates have this common failure. There were two sets of winner take all elections in this current Republican primary. The first was state level delegates where the winner of a state received a significant bonus number of delegates and then winner take all at the Congressional District level. The Republicans assigned three delegates to each Congressional District without regard to how many Republicans actually lived or voted in that district.

538 has a good example of how this flat allocation of winner take all delegates by district helped Trump:

If Ted Cruz wins by a huge margin in Milwaukee’s suburbs, as expected tonight, he’ll get all three delegates from Wisconsin’s 5th Congressional District, which cast 257,017 votes for Mitt Romney in the 2012 general election. But in two weeks, Donald Trump could capture just as many delegates by winning a majority of the vote in New York’s heavily Latino, Bronx-based 15th Congressional District, which cast only 5,315 votes for Romney four years ago.

Three weeks ago, Trump won three times as many delegates — nine — at the Northern Mariana Islands convention, which drew just 471 participants.

This is problem #1. The GOP primary delegation process favors plurality winners and it favors candidates who can win in very low turnout environments. There is a massive variance between the minimum number of votes needed per delegate and the maximum number of votes needed per delegate. Some districts are extremely efficient and some are extremely inefficient places to win. The Republicans treat districts like the Senate treats states. The first rule change would be to scale the delegate award to some measure of Republican vote strength.

Read more



GOP Ceilings

Just grabbing a bit of a Chris Cizilla post on Trump’s ceiling:

There’s no debate that Trump’s electoral college floor is lower than that of any other possible Republican nominee in recent memory. You could also argue —as I have — that his electoral college ceiling is higher than (or at least different from) any of the party’s past nominees. For what it’s worth, here’s what a Trump best-case-scenario map might look like….

That’s Trump at 285 electoral votes, 15 more than he would need to become president….WaPo Max Trump 20160405

 

Cilizza is arguing that Trump is the highest variance Republican candidate. I can see that argument, and rationally speaking, if the 2016 fundamentals are neutral to slightly lean Democratic with a combination of a decent economy, general peace and a Presidential year electorate that is looking more and more liberal as the conservative base is dying off much faster than the liberal base and the liberal base is getting refreshed much faster than the conservative base, a high variance strategy is not bat shit insane. It guarantees more blow-out losses but it will get a few more wins.

But assuming no colossal cock-ups, the ceiling for Republicans under the current political alignment will continue to shrink so the incentive for the GOP to push high variance national candidacies will grow, not shrink, in 2020 and beyond.



Damn it, Janet!

I’ve been a little disappointed in the low level of vitriol n’ derp so far in the Democratic campaign, so I was glad to see Susan Sarandon say maybe Trump would be better than Hillary because revolution.

Of course, the reality is that almost no Bernie supporters will vote for a Republican in November if Hillary’s the nominee and almost no Hillary supporters will vote Republican in November if Bernie’s the nominee.

On the other hand, I’m not so sure about the #NeverTrump people. I originally thought they’d all come around to Trump if he’s the nominee (I don’t think it’s a given he will be) but now I’m not so sure. Federal Republican office holders and national GOP party operatives have to back him because it’s the smart move (Tessio was always smarter), but I wonder if the conservative punditocracy can use the Trumpocalypse to prove to the world that they truly are independent-minded Burkeans of no clique or party.

Is there any chance that we could see a post-Both Sides Do It campaign? I’m not talking about the real dead enders — I know Ron Fournier will be in a bunker tweeting death to the duopoly for years after our current political system, as well as twitter, have ceased to exist. But now that Cokie herself is committing journalism against Trump….



Three ways in the House

Reading the Huffington Post, I saw this political bodice ripper and I still can’t figure out how to make the mechanics of the piece actually work in our shared reality:

Suddenly they realize, “holy shit, what if we could stop Donald Trump and keep Hillary Clinton out of the White House?”

So they run a moderate establishment Republican as a third-party candidate — 100 percent as a spoiler candidate. Worst case scenario oh, they prevent Donald Trump from winning the White House. Best case scenario they pull enough votes away from Hillary Clinton to prevent her from securing the necessary majority of 270 electoral votes.

Then the election goes to a House of Representatives ballot presided over Speaker Paul Ryan, Mitt Romney’s former running mate in 2012.

If neither candidate gets 270 electoral college votes, Congress picks the president. And he will be called President Mitt, the one who is laying the groundwork for this doomsday electoral scenario.

The basic theory is that a third party candidate who is Generic Republican Establishment (no not Pawlenty) would be able to do three things at the same time:

  • Insure that Trump does not get 270 electoral votes
  • win at least one electoral vote
  • Insure that Hillary Clinton does not get 270 electoral votes

In an alternative universe, that could work, but in this universe, I am having a hard time seeing how to actually make it work with a generic Republican running as a non-Trump alternative.

I think the first part is achievable.  However, the third party Republican spoiler is not needed.  Continual video playback of Trump’s speeches to non-Trump fans will isnure that.  If the Republican establishment decided it needed at least one electoral vote, it’s sock pocket could probably win Utah or a Congressional district in Nebraska.  Worse comes to worse, an elector could be a faithless elector.  I’ll concede the mechanics on this one.

The problem with this pre-emptive pants shitting is the third part.

Read more



The cost of spite

Utah recently passed an expansion of Legacy Medicaid.  This is an improvement for the lives of the 16,000 people who are projected to receive coverage.  But it is also an illustration of the cost of spite.

Deseret News has the details:

Gov. Gary Herbert told reporters that he would sign HB437 (sponsored by Rep. Jim Dunnigan, R-Taylorsville), which would extend Medicaid coverage to Utahns who have the greatest need

Bill details are in this story:

A bill backed by Republican House leaders extending Medicaid coverage to Utahns in the “greatest need” because of homelessness, run-ins with the law, substance abuse or mental health issues won committee approval Monday.

“We have the opportunity to improve the lives of 16,000…

Gov. Gary Herbert saw his Healthy Utah plan to use the hundreds of millions of dollars available under Obamacare to cover all of the estimated 110,000 Utahns eligible for Medicaid expansion pass the Senate but fail in the House last session….

The cost to the state is about $30 million, he said, with the federal government paying about $70 million. He said Utah hospitals will pick up 45 percent of the state’s share, some $13.5 million.

Spite is expensive.

There are approximately 110,000 people in Utah who qualify based on income for Medicaid expansion (thank you very much Chief Justice Roberts).  In the current calendar year, the Federal government will pay 100% of the cost of covering those individuals via Medicaid Expansion.  In CY 2017, the feds will pay 95% of the cost, then 94% in 2018, 93% in 2019 and 90% in 2020 and over the long term.  A Legacy Medicaid expansion which this is means the Federal government is picking up 70% of the cost from the first day of the program start to the last day of the program.  Utah will need to come up with the other 30%.

In 2016, complete Expansion means the state is paying on net, nothing to cover 110,000 people.  Under this legacy expansion, the state is paying $30 million to cover 15% of the eligible population.  In 2017, the state will actually have to kick some money in.  However, it will be less than then the  $30 million obligated as the current program is skimming individuals who are highly likely to be high cost individuals.  The uncovered population will tend to be healthy and cheap.  By 2020, the state may be spending slightly more to cover the entire population that is eligible rather than the small subset of the population that they want to cover.

However, expansion at that point will be bringing in massive net federal cash flows that show result in higher state tax revenue so a dynamic analysis will probably show that expanding a subset program under Legacy Medicaid match rates versus a full expansion is still close to a net wash from the state finance perspective.  At most, covering the entire population at the enhanced federal reimbursement rate is a pittance more than covering an expensive subset at the regular Legacy Medicaid federal match rate.

Spite is expensive.

 



CAP and the Republican primary

I just want to highlight two tweets I saw since Cole dropped his truth bomb:

And a question as to why Rubio is not dropping out:

The anti-Trump agenda is ensnared in a massive collective action problem. The anti-Trump movement is better off if Rubio drops out. However, the problem is simple for a party that really does not believe in collective action problems solved through societal actions and instead believes or at least publicly spouts off that everything can be modeled on the basis of individual rational behavior to get optimal societal results. There is a stable equilibrium that is extraordinarily negative for the anti-Trumpers where everyone is asking the other individuals to impale themselves on the barbwire so that they can use the body as a bridge to get into Trump’s trenches.

Let’s just look at Rubio for his incentive structure. Right now, he has shitty chances. The betting market has him at 8% chance of nomination and probably an implied 4% to 5% chance of the White House. Those odds suck, especially compared to his odds in December. However they are better than his 2020 odds. He has gone 1 and done in the Senate. He has indicated he actually hates the process of governing so a run for Governor in 2018 and then a summer long camp-out in South Carolina in 2019 is unlikely. If he loses now he becomes 2012 Rick Santorum without a natural base of dedicated supporters and a similar humiliating loss.

8% odds suck. They are much better than his 2020 or his 2024 odds.

So why would he get out?

Applying that same logic to all to Kasich and Cruz, and their odds suck now, but they are better than they would be in 2020.

And given that the promises that are made in March of 2016 are highly contingent promises that Trump can first be beaten and then Clinton can be beaten plus the promiser has few strong constraints in his actions after Election Day, the promises made to move someone out are not particularly valuable nor credible.

The traditional solution to a collective action problem is to have an external entity be able to move people off of stable but negative equilibriums and compensate losers from the much larger net social gains. The RNC is not a strong governing entity and Republicans don’t do collective action problems well anyways…

So pass the popcorn.