Relevant and even prescient commentary on news, politics and the economy.

Michelle Malkin and Larry Katz

Robert Waldmann

is very very late with this one. Appears that long long ago (in blogtime) Michelle Malkin said that Larry Katz once wrote that unemployment insurance causes unemployment.

I know Larry Katz, Larry Katz was my PhD co-supervisor, Larry Katz is a friend of mine and Ms Malkin, you’re no Larry Katz.

Malkin was referring to “The Impact of the Potential Duration of Unemployment Benefits on the Duration of Unemployment” Katz and Meyer 1990 which did indeed show that a lot of people start working again exactly when their UI benefits run out. It also showed something even more remarkable. A lot of people start working again after exactly 6 months of unemployment even if, while they were unemployed, benefits were extended. The paper is a spin off from his dissertation chapter on how, if one wants to understand unemployment duration one has to distinguish between temporary layoff unemployment and other unemployment.

That’s not so true anymore. The sad fact is that the fraction of the unemployed who are on temporary layoff is much lower now (about 10% I think). This implies much slower employment expansion in a recovery. It also means that unemployment spells are not designed in advance, and, in particular, not planned with UI rules in mind. Katz’s story about the 6 month spike even when benefits had been extended was that firms scheduled a layoff of 6 months before benefits were extended. It is true that in addition to the recall from temporary layoff spike Katz and Meyer found a smaller spike in the hazard of exiting unemployment to a new job when UI actually stopped.

The economy has changed and the effect described by Katz and Meyer is probably smaller now than it was in their data.

Tags: , , Comments (0) | |

July Auto Sales

By Spencer,

July auto and light truck sales are being reported at 11.2 million annual rate.

That is a very significant increase from the second quarter average of 9.5 million and implies that consumption in the third quarter GDP data will be up significantly.

I do not have detailed data of how much of this is due to “Cash for Clunkers”.

But the point that people are willing to take on a new set of car payments is a good sign for consumer confidence and spending.

Tags: Comments (0) | |

Energy & Commerce version of HR3200: a treat for the Read the Billers

by Bruce Webb

How many people in the “Read the Bill” movement will in fact read the bill? Well here is their chance.
HR3200: Energy and Commerce passed version

I would be interested in knowing how this bill saves $100 billion over the version reported out by Education and Labor House Tri-Committee Health Care Bill. First thing of note is that the new Energy and Commerce version is the exact same page length as the older version suggesting that the changes were mostly in the formulas and not so much the structure.

CBO has not released a score, when it does I will update.

(Update: in comments Movie Guy suggest using some file path information that this is not in fact the final form of the bill. Well maybe, but it is the only one linked from the Energy and Commerce page that shows the activity undertaken on Friday the 31st. Which leaves the question of why all the amendments and the results of their votes are listed and not the bill as approved.
Anyone who can shed light on this please put something in comments.)

Tags: , Comments (0) | |

Will catastrophic only health insurance be rescinded in the end?


Taunter explains how the .5% rescission rate figure discussed in the hearings on health insurance premiums and coverage is used to dismiss Congressional concerns as inconsequential and actually is a good business practice that will not change. It is worth going to the post to accurately follow the train of thought. Taunter concludes that the chances for rescission for a serious illness is:

If the top 5% is the absolute largest population for whom rescission would make sense, the probability of having your policy cancelled given that you have filed a claim is fully 10% (0.5% rescission/5.0% of the population). If you take the LA Times estimate that $300mm was saved by abrogating 20,000 policies in California ($15,000/policy), you are somewhere in the 15% zone, depending on the convexity of the top section of population. If, as I suspect, rescission is targeted toward the truly bankrupting cases – the top 1%, the folks with over $35,000 of annual claims who could never be profitable for the carrier – then the probability of having your policy torn up given a massively expensive condition is pushing 50%. One in two.

Rescission definition practice can be found here and unfair examples here. Here is why door number two is NOT 50/50.

It is suggested that application forms are formed to allow for inaccuracies and that some criterion used to cancel policies is easily avoidable at the time of the application, the implication being future use for the inaccuracy makes it worthwhile to allow ambiguity and not vet information, mainly for those who wind up claiming over $35,000 and who have severe, chronic conditions that make premiums from that person irrelevant to profit.

This may have not been the intent long long ago, and certainly fraud occurs, but rescission having a basis purely monetary and not legal (from a common consumer point of view on what fraud is), is the policy today if testimony is to be believed.

The first notion, that it is a small problem except for the person involved, is discussed in the first post. The key to amended %’s lies in the fact that Medicare takes care of the 65/over group of chronically ill and elderly clients, so the privately covered population is smaller than the per centages indicate as well, a major oversight. The second and third part of the problem will be in Part 2 and 3 (clever, huh?)

StatsGuy wrote in comments to the Taunter Media post:

The same light bulb went off when I read the 0.5%, but I could not have explained it _nearly_ as well. Very nice post.

I still wonder, though, whether it might be slightly worse than even this picture.

1) I believe your data is for the US population as a whole. (If I’m wrong, then this comment is meaningless – apologies.) But, in fact, much of the sickest part of the population receives health care via Medicare because older people are (to use your technical term) sicker.

So the % of people in the top tier AMONG PEOPLE NOT ON MEDICARE is much lower, which means that the conditional probability of suffering rescission given that you need treatment is much higher. Roughly, if the % of people among under-65 (and not on Medicaid) in the top bracket was half of what it is for the entire population, then the probability of suffering rescission given that you have a large claim is double even your current estimates.

2) The probability of losing the policy given that you really need it may be X% in any given year. But there’s a cumulative effect – over time, you build up a reservoir of uninsurable who lost insurance due to rescission, and now cannot get it back because they have a chronic condition.

on July 29, 2009 at 10:03 am | Reply Taunter
You are absolutely correct about #1, and this is a huge error factor. 10% of Medicare costs take place in the last month of life alone, and Medicare is roughly 45% of the national health care spend. So all of those patients are clogging up the top end of national distribution and not on private insurance in the first place. Unfortunately, I can’t find a private-only, or individual-pay-only distribution, and of course if I did find an individual-pay-only distribution it would be skewed on the top with denied claims (some people should be spending a lot, but actually spend much less, because their policy was pulled). The Reuters article says Medicare spends 30% of its outlay on the top 5% of its population, which means it has a flatter curve than non-Medicare (I would assume, without evidence, that fewer Medicare beneficiaries have negligible health expenses). This implies non-Medicare spending is even more highly concentrated with a few very high spenders.

On #2, I’m a little less confident, and it was one of the reasons I may have misunderstood James’ original post. There is a cumulative effect, but that effect is blunted to some degree by the fact that the people who account for the very high medical expenditures do not necessarily change much from year-to-year (with the obvious exception of the end-of-life expenses typically borne by Medicare). In fact, one of the reasons I suspect rescission became such a powerful phenomenon is that if Sally has breast cancer at a young age, she is going to be in the 99th percentile several times; the carrier is weighing years of such expenses against her premium. So it might not be the case that in a forty year career an average person has a 33% chance of ending up at some point in the top percentile (1-(.99^40)); it is probably the case that most people have a tiny chance of ever getting an expensive chronic condition (or at least an expensive, chronic condition before turning 65), and some people have a large chance of repeatedly being in the top percent.

Keep the discussion going. It is clear a profit motive has serious impact for some people…how would you bet even catastropohic only insurance premium money?

Tags: , , Comments (0) | |

Models for health care organizations?

Tom aka Rusty

Models for health care organizations?

In a recent post I opined the Mayo Clinic and Cleveland Clinic, both favorites of President Obama, were not models for wide application to the rest of the nation.

In response to a spirited debate (thanks Coberly) I wrote up these comments.

Starting in the 90s a common model has been the Medical Services Organization, or MSO.

(I seem to remember the MSO evolving from failed staff model HMOs.)

In the model, the hospital or integrated network buys and/or starts physician practices, manages those practices, and the physicians are employees of the MSO corporation.

Many of the MSOs were/are an attempt to provide adequate primary care medicine in an era when primary care practices are tough to manage. The MSO model allowed a central organization to hire, deploy and re-deploy physicians in accordance with the needs of the community.

Many of the early MSOs failed, for a variety of reasons, but often bad management. Hospital executives thought “I can manage a hospital, therefore I can manage physician groups,” and that was very misguided thinking. Some still use that line of thinking, producing poor results.


The MSO model can be used in both large cities and small communities.

The model allows (or forces) better coordination between physicians and hospitals.

The model can be combined with an insurance product, or not.

The model allows better coordinated negotiations with private insurers.

Patients can often keep their doc, just under a different nameplate.

There is a potential in integrate information systems.

Hospitals often do a poor job of management.

Hospitals often botch physician contracting.

Most hospitals lose money on MSOs, theoretically making up the difference by capturing ancillaries and through management efficiencies.

Many physicians are lousy employees. Physicians rarely trust hospital executives.

Physician motivation and work ethics change, structuring physician contracts is very difficult.

Specialists, by and large, have not bought into the MSO model, which remains a province of primary care (often including internal medicine and ob-gyn).

Health care reform will drive providers to start new MSOs and to expand current MSOs. There will be a honeymoon period, but the marriages may not live happily ever after.

Tags: , Comments (0) | |

The ‘Read the Bill’ Movement

by Bruce Webb

Republican obstructionists to Obama’s agenda have developed a new tactic. Seemingly forgetful of how Republican majorties jammed through huge bills without letting Dems actually read them, they now insist that every Dem has to read every bit of every bill and more specifically the Health Care Bill. Well in reality that is not that much of a challenge. Behold a typical page of legislation:

If you click on the image to expand it you will see that is is is quite large print, is double spaced, and much indented. Moreover large parts of it are clear boiler plate, necessary to get the exact language in the subsequent U.S. Code correct, but mostly not necessary to get the sense correct. I don’t mean that you don’t need to read bill language carefully, but it is not like reading a 1000 page bill is the equivalent of reading War and Peace.

This sample is from the Senate HELP bill which is 175 pages long. Why is it so much shorter than the House version? Because it only covers the policy components of establishing the benefits package and administering the Exchange, the bulk of the bill is taken up with various tax provisions, important yes but not so much so that you cannot evaluate the policy options without them. Lets look at the text of HR3200 as introduced.
On page 2 we see that the bill is divided into Division A – Affordable Health Care Choices and Division B – Medicare and Medicaid Improments. Division A takes up the first 215 pages while Division B takes up the remaining 803.

I am not saying that Division B is not important or that it doesn’t have some controversial pieces, but mostly it is tinkering with programs that already exist, the real policy innovations are in Division A. And if we look at that we see that it is divided into four Titles. Title 1 – Protections and Standards for Qualified Benefit Plans, Title 2 – Health Insurance Exchange and Related Provisions, Title 3 – Shared Responsibility, and Title 4 -Amendments to the Internal Revenue Code. Of the 215 pages Title 1 takes up pg. 5-71, Title 2 pg. 72-143, Title 3 pg. 143-167. Even at that much of the material is administrative or spelling out specific language changes to current existing code.

The Republicans are going to try to make hay out of this ‘Have you read the bill’ message and to laugh at Obama’s offer to take people through the bill “line by line” but there is nothing that laughable about it. Because most of the meat is concentrated in just a small number of sections in Division A, a couple of hours of briefing or browsing is probably enough to get the average Congressman up to speed on this legislation. In fact the biggest frustration is that while the pages are numbered there are no page keys to translated Section numbers to pages, but except for that this is not that difficult a read.

When people make the claim that no one can really understand a 1018 page bill realize that mostly you don’t need to deeply engage with more than maybe seventy or eighty double spaced, highly indented pages to get a pretty solid grasp of what the bill proposes.

Tags: , Comments (0) | |

Declining customer base for insurers means declining companies?


Slate has an article that is interesting. Are current health insurance companies a doomed model anyway?

The declining numbers aren’t simply a function of job loss. A Bureau of Labor Statistics study released this week found that in March 2009, only about 70 percent of private-sector workers had access to employer-provided medical care benefits, “and only 25 percent of the lowest wage earners—those with average hourly wages in the lowest 10 percent of all private industry wages—had such access.” Note the difference in the data between the public and private sectors. For government workers, 88 percent have access, and the participation rate is high. For private-sector workers, 71 percent have access, and the participation rate is lower. What accounts for the difference? It’s unclear. But at least for single employees, the government picks up more of the tab (90 percent, compared with 80 percent for private sector jobs). For family coverage, the split is the same, 70-30, in both the public and private sectors.

In fact, there’s pretty good evidence that government spending is all that stands between the struggling insurers and complete disaster. Look through the insurers’ earnings reports, and you’ll see that a portion of the loss in commercial business has been offset by growth in Medicare and Medicaid programs. At UnitedHealth in the past year, for example, enrollment in its public programs rose from 6.185 million to 7.115 million.

The system of employer-provided health care coverage is crumbling before our eyes, and for more Americans—and for more American insurance companies—government-funded health care is all that separates them from financial disaster. A Gallup poll found that the percentage of Americans who say they get their health insurance from an employer has fallen from 58.9 percent in January 2008 to 56.5 percent in May 2009, while the percentage who get it from the government (Medicare, Medicaid, VA benefits) has jumped from 26.5 percent to 29 percent. (The rest purchase it on their own.) But this poll understates the case. About 17 percent of payroll jobs today are government jobs. Crunch the numbers, and it’s more like 39 percent getting insurance from government sources (public programs and public-sector jobs) and about 47 percent from private-sector jobs.

Simply by doing nothing, we’re slowly nationalizing health care.

Comments (0) | |