Clay Shirky writes,

Failure is always an option. Engineers work as hard as they do because they understand the risk of failure. And for anything it might have meant in its screenplay version, here that sentiment means the opposite; the unnamed executives were saying “Addressing the possibility of failure is not an option.”

He talks about what I call the suits vs. geeks divide. Recall that I originally coined this to describe the unwillingness of financial executives to listen to geeks who worried about the risk in mortgages.

When a large organization, such as government or a legacy media organization, undertakes a new initiative, they are in effect starting a new business. Most start-ups fail, so that failure is in fact the most likely outcome. But if you have spent your whole life playing office politics (or real politics) to get to the top of an established organization, you may not have had any training in dealing with something as fragile as a new enterprise.

If you start a new enterprise, it is a good idea to start small and build incrementally. In fact, that is how President Kennedy’s initiative to put a man on the moon was carried out.

But the health care law was designed to be big and complex from the beginning, and nobody wanted to allow for the possibility of failure. Even now, many progressives deny that it will fail. My sense is that the law is in fact a total failure, with the web site perhaps the least of its problems.

Among people along all three axes.

1. The progressives are much less forgiving of the Obama Administration’s management failures than are the rest of us. Some of us saw the problem as baked into the law. It was pointed out that the law mentions the word “web site” over 130 times, which is an indication of how complex the requirements could be. I made my point that Amazon and Kayak emerged out of a tournament involving thousands of companies. I said that if Obamacare had been a private-sector start-up, its odds of success would have been less than one in a thousand. Others pointed out that in the private sector you usually start with a small, minimally-functional prototype, not with a full-blown, full-featured system. Still others pointed out that the features of Obamacare are so tightly interconnected that it required perfect execution, which was extremely unlikely.

The progressives (especially those over age 40) wanted instead to emphasize the fundamental management flaws, such as not having a strong executive in charge of the project. They insist that Obamacare could have worked. Clearly, to suggest otherwise was to cast some doubts about the progressive approach.

2. The suggestion that some of us had of creating a “data extract” for the web site, so that it did not have to access other databases in real time, could not have been implemented. There is a law against the IRS giving out copies of its data.

3. The story was told that when Medicare Part D debuted out (in 2003, I believe), the launch was delayed three weeks because the system was not considered ready. Someone said that the Bush Administration put out the story that they did not want to launch the site on the Jewish Holiday of Yom Kippur. Of course, that is only a one-day holiday. Someone (not Jewish) joked that “But then you have sukkot” (a week-long holiday that comes a week after Yom Kippur). The best hope for Democrats is that the web site works well quickly

4. Nobody of any persuasion was buying David Cutler’s line that the slowdown in health care spending since 2008 reflects Obamacare.

5. Some of the state exchanges may be working better because states have been given an exemption from verifying eligibility for subsidies (at least, that is what I heard people say, but I may be wrong)

6. One journalist insisted that when President Obama announced his recent “fix” that would allow insurers to renew old plans, everyone knew that this could not possibly be implemented. The journalist said that the whole point was to keep the Senate from voting on a proposal to implement the fix. I guess the choice is between believing that Obama was an idiot (for thinking that state insurance commissioners and insurance companies could pull off a renewal so quickly) or a liar, and this journalist, who tends to be sympathetic to the President, implied that Obama (once again) was being cynically deceptive.

7. It was pointed out that non-libertarian Republicans should be sad about the failure of healthcare.gov, because many Republican health-care reform proposals have involved some sort of health exchanges.

8. People said that it is hard to make big policy moves in today’s environment. Trying Obamacare was like trying to privatize Social Security. President Bush wisely backed off of the latter (although he was not trying to privatize the entire program), and perhaps the Democrats would have been wise to back off the former.

9. President Obama’s reputation with young people as being “cool” has taken a tremendous hit. This could have long-term implications for how young people look at politics. My thought (which I kept to myself) is that at the moment President Obama seems to be adapting Atlas Shrugged for the 21st century.

10. The scenario for Obamacare’s death is that the insurance companies “defect” by leaving the exchanges or raising rates to unacceptable levels. As Obama’s personal political strength weakens, the “defect” scenario becomes more plausible.

The Washington Post is troubled by President Obama’s “keep your plan” fix, but for the wrong reason.

Insurers who decided to act would try to hold onto their healthiest customers — those who don’t make many claims, if any. But the Obamacare marketplaces need those relatively healthy customers.

The threat of “adverse selection” in health insurance is over-rated and always has been.

The problems with implementation are under-rated and always have been. The Obama Administration has spent 3 years bulldozing the individual market in health insurance. Now, they expect the health insurance companies to rebuild it in 30 days.

The folks who have to implement the new Presidential edict are not pleased. From the state health insurance regulators:

The National Association of Insurance Commissioners (NAIC) suggested Thursday that the plan to allow insurance companies to offer non-compliant health plans into 2014 is not logistically feasible.

From the health insurance companies:

They have warned legislation in both chambers of Congress would provoke an administrative fiasco and could have serious implications for the new marketplace’s risk pool, premium prices and the cost of the law to the federal government.

From a WaPo blog story:

Here’s how Robert Laszewksi, an insurance consultant, put it in a note to clients earlier this morning:
“This means that the insurance companies have 32 days to reprogram their computer systems for policies, rates, and eligibility, send notices to the policyholders via US Mail, send a very complex letter that describes just what the differences are between specific policies and Obamacare compliant plans, ask the consumer for their decision — and give them a reasonable time to make that decision — and then enter those decisions back into their systems without creating massive billing, claim payment, and provider eligibility list mistakes.”

If I were the CEO of a health insurance company, I would engage in civil disobedience. I would design policies that are attractive to consumers, regardless of whether I offered those policies before and regardless of whether they satisfy regulatory mandates. I would then sell those policies to any customers who want them and defy the regulators to come in and take them away. I think in today’s environment, I could get away with that.

Consider two tasks:

1) Get the healthcare.gov web site to work by November 30.

2) Use legislation and regulations to enable people to keep their plans and keep their doctors.

I think that (1) is more likely to be achievable. Not that I am optimistic about (1), and today’s WaPo reports,

The insurance exchange is balking when more than 20,000 to 30,000 people attempt to use it at the same time — about half its intended capacity, said the official, who spoke on the condition of anonymity to disclose internal information. And CGI Federal, the main contractor that built the site, has succeeded in repairing only about six of every 10 of the defects it has addressed so far.

But there may be workarounds. A reader sent me this link to a story about web developers who created a site in two weeks that enables you to browse for plans to find the best one.

You can’t actually enroll on the HealthSherpa site, but they do provide contact information for companies offering the plans. Users who find a plan they like can go directly to the insurance companies without ever using HealthCare.gov.

On the other hand, “keep your plan, keep your doctor” seems to me a pretty hopeless case.

1. Even if Obamacare had never been enacted, it would have been difficult to back this promise. What is to stop an insurance company from deciding to tweak a plan or to get rid of it altogether? What is to stop a doctor from refusing to accept insurance from the given company, or from any company at all? These sorts of changes used to take place all the time, with or without Obamacare.

2. But now, the situation is FUBAR. The old plans no longer exist. In order to revive them, the insurance companies would have to have to put together brochures, mail them to customers, and give customers time to look them over and decide whether or not to renew. But before they can do that, the insurance companies have to go through the process of signing up health care providers, including negotiating agreements concerning compensation. But before they can do that, the insurance companies have to run their plans past regulators to make sure that they comply with whatever legislation/regulation the bureaucrats in Washington and the states come up with as they try to enforce “keep your doctor, keep your plan.” And before they can do that, regulations have to be written, go through a comment period, and published.

I would be surprised if this can be accomplished before health insurance cancellations take effect.

CNN reports,

On the first day, there was a big problem that we hadn’t heard about before. “There was a fix regarding residency for Medicaid and CHIP that was not fixed correctly and is denying … 90% of people based on residency,” the war room notes from the day read.

As the documents move on and issues are tackled one by one, sweeping problems continue to appear.

October 9: “A new problem in the system has been identified: for about 30% of the 70,000 applicants, the system has skipped applicants through ‘events’ that are required to complete the application.” In other words, nearly a third of applicants couldn’t fill out the form, because the website was skipping “events” or entries they needed to make.

Imagine what would have happened had there been no issues with speed on the site, and tens of thousands of people had been able to use it at once. The results would have been so catastrophic that they would have had to shut down the site. Of course, if this were a private-sector project, it would not still be live. I cannot imagine any private-sector company willing to take the sort of risks that the government is taking right now.

As an aside, here is columnist Suzanne Fields picking up on the suits vs. geeks divide.

As another aside, in another article from several years ago, the geek, Robert N. Charette writes,

If there’s a theme that runs through the tortured history of bad software, it’s a failure to confront reality.

That might serve as an epitaph for the healthcare.gov rollout. The entire article is worth reading.

CBS News headline, October 31.

Unlike HealthCare.gov, Washington state’s health insurance exchange appears to be working well

Seattle Times headline, October 25.

Thousands get wrong subsidies data from state health exchange

Why would some of the state sites, such was Washington state, be working more smoothly than healthcare.gov? Perhaps they do not rely as much on national government databases to verify identity. It seems that the Washington state site asks for motor vehicle information, suggesting that they use their motor vehicle database as a way to check identity.

Today’s WaPo story seems to vindicate my instinct to blame the project organization, not the technical people.

Recall what I wrote in my letter to the editor.

I doubt that presidential anger can do much to cure the dysfunctional organizational dynamics that I suspect are at the heart of the Web site problems.

Today’s story is about those dysfunctional organizational dynamics. Most devastating is a 3-year-old memo from health economist David Cutler. He barely even touches on the technical issues of implementation. He sees the Administration dropping the ball on preparing the health industry and the public for the new system.

Running exchanges is a collaborative process. As just one example, the person who ran the Commonwealth Connector in Massachusetts estimates that he had 500 town meetings to discuss reform, the equivalent of 17,000 meetings nationally – and this was in a state where two-thirds of people, along with insurance companies, supported reform. The person newly appointed to head the insurance oversight office has a reputation as an insurance bulldog, not a skilled facilitator. Remember that most people will get their information about reform from their doctor and their insurance agent. If you cannot find a way to work with hesitant states and insurers, reform will blow up.

An important implication is that even if the Web site were working perfectly, we would have plenty of implementation problems. Has there been enough consumer education to enable people to understand which plan to choose? Have doctors been coached about how to help with the process? etc.

Later, he writes,

Above the operational level, the process is also broken. The overall head of implementation inside HHS, Jeanne Lambrew, is known for her knowledge of Congress, her commitment to the poor, and her mistrust of insurance companies. She is not known for operational ability, knowledge of delivery systems, or facilitating widespread change. Thus, it is not surprising that delivery system reform, provider outreach, and exchange administration are receiving little attention. Further, the fact that Jeanne and people like her cannot get along with other people in the Administration means that the opportunities for collaborative engagement are limited, areas of great importance are not addressed, and valuable problem solving time is wasted on internal fights

The WaPo story makes it seem as if no one was really in charge.

The Medicaid center’s chief operating officer, a longtime career staffer named Michelle Snyder, nominally oversaw the various pieces, but, as one former administration official put it: “Implementing the exchange was one of 39 things she did. There wasn’t a person who said, ‘My job is the seamless implementation of the Affordable Care Act.’ ”

That last sentence gets to the heart of the matter. As I wrote here,

I suspect that the technical problems are mere symptoms. Probably what is fundamentally messed up in this health insurance brokerage business is the org chart.

Cutler’s memo includes a recommendation for a functional org chart. I think that anyone with business experience or common sense would think along similar lines. President Obama’s supporters can spin like Dervishes, and yet it will be difficult to escape the inevitable inference that the White House clique lacks business experience and common sense.

He emails me (but not for attribution)

The key to the whole thing is the back-end. The big problem here is likely having to query various legacy DBs in real time (e.g., have the website fire a query to see if your name and SSN match). This was a home-run approach by the original technical team, and in my view a huge blunder (though hindsight is 20/20, and I’ve made some very dumb technical decisions in my life). They already have an Oracle (basically, a modern relational database) instance up and running. The crucial change would be to grab a batch transfer from the legacy systems as of say October 25th and move ALL of it into the Oracle instance. You could then have every query fire against the Oracle database. This would radically simplify every technical problem, as you would now have end-to-end control over the DB – logic – website. You could structure highly simplified queries, and build the whole data model around exactly one job: make this one website work. You could then periodically batch update from those source legacy systems, say daily or weekly once you were in production.

This occurred to me also. You create extracts of the data on legacy systems, and then you run the web site against the extracts. Some issues.

1. You have to hope that the existing system has a clean separation of business logic from physical location of the data. That way, you just change the statements that access data to find it in the extract rather than in the legacy system. If there is business logic mixed in with the data-access statements, then there is more to re-code and test.

2. The job of building the data model for the extract database is not trivial.

3. The job of tuning the database to obtain good response times is not trivial.

4. In some sense, you have two databases. You have a “read-only” database of extracts from other systems. You also have a database that is written to as the user goes through the process. Tuning the performance of that latter database is a nontrivial problem–although they should already be working on it.

5. We start with a deficit in terms of testing, and that deficit gets much larger if we redo the data interchange process.

6. This creates new security problems, and it may create new logistical problems for the people managing the legacy databases. Not that those issues will be given priority at this point.