Measuring the Success Rate

Or “How to Lie With Statistics”. As I read James Taranto’s column today on the early reports of successful enrollments at Healthcare.gov, I realized that many people might not understand how sensitive error reporting can be to just how you discern and report errors. Take optical character recognition (OCR) software, for example. When reading the advertising blurbs for it you frequently encounter extremely high recognition rates, 99% or higher. 99% of what?

What they don’t generally tell you is that the recognition rate is characters. Not pages or, worse yet, documents. When you’re talking about a recognition rate of 99% of characters that means that on average the program will correctly recognize 99 out of 100 characters. That in turn means that on average for every page with more than a tweet-full of characters the page recognition rate is zero. The document recognition rate is zero. That can make the difference between a particular project making financial sense or not.

In the real world as opposed to in a lab that’s an important distinction. In dealing with a large project of thousands or even millions of documents it determines how many proofreaders you’ll need. It can determine how long the project will take and how much it will cost.

Twenty years ago the company of which I was a principle at the time had a project for the Federal Reserve that required us to scan, store, process, recognize, and index about a million pages worth of documents in a very tight timeframe. It was a monumental task, as you might imagine. The niggling little details matter.

That’s what I think of when I read reports of 50,000 or 500,000 enrollments. What’s the enrollment rate and how is it measured?

Let me also remind you that the open enrollment period for the healthcare exchanges ends on March 31, 2014 and that time is passing. What was 200 days is now more like 150. That affects the peak load requirements of the system. If anybody tells you that a system that can accept and process 100 applications a day can definitely accept and process 100,000 applications per day, fire him.

Can Healthcare.gov meet the challenge? Sure. Will it? Who knows? Pointing to the tremendously different processing requirements of the Massachusetts system as a model is fatuous. They are not comparable. We’re in unknown territory.

12 comments

No How, No Way

And speaking of worst case scenarios and predictions, Steve Forbes predicts that the PPACA will be repealed before the 2014 elections:

Prediction: even if HealthCare.gov is fixed by the end of the month (unlikely), Obamacare is going to be repealed well in advance of next year’s election. And if the website continues to fail, the push for repeal—from endangered Democrats—will occur very rapidly. The website is a sideshow: the real action is the number of people and businesses who are losing their health plans or having to pay a lot more. Fixing the website will only delay the inevitable.

My reaction to that is “No how, no way”. There are enough Democrats in safe seats in the House and Senate that a presidential veto, all but certain in the event of a vote to repeal, cannot be overridden.

35 comments

Will Obama Bounce Back?

Speaking of movie references, I’m guessing that Democrats’ main argument for the foreseeable future will be Han Solo’s “never tell me the odds”. Over at National Journal Alex Roarty suggests that it’s unlikely that President Obama’s approval ratings will bounce back:

In fact, no president in the last 60 years has watched his approval ratings bounce back during their second term. Either they didn’t make it to another stint in office (Ford, Carter, and George H.W. Bush), never dipped in the first place (Eisenhower and Clinton) or were removed from office at the nadir of their popularity (Nixon). Lyndon Johnson recovered somewhat, but only after announcing he would not seek another term. Ronald Reagan dropped from the low 60s to the high 40s amid the Iran-Contra scandal, and his popularity never recovered entirely until his last months in office. But it also never fell to lows experienced by Truman or Bush.

Or Obama, for that matter.

I’m in the analysis business not the prediction business but if I were to venture a guess my WAG as of this writing is that in 2014 the Republicans’ hold on the House will narrow a bit while the Democrats’ hold on the Senate narrows a bit. Things look different in the states holding elections for the Senate this time around than they do in the country at large. But, as I say, that’s no more than a guess.

1 comment

Worst Case Scenario for the PPACA?

Peter Suderman (Mr. Megan McArdle) reviews the worst case, hard landing scenario for the PPACA:

Over the weekend, several reports suggested that, despite continued assurances that Healthcare.gov, the problem-plagued online insurance enrollment portal run by the federal government, would be running smoothly for most users by the end of the month, it increasingly looks likely that the deadline will be missed.

Insurance industry consultant Robert Laszewski, who, thanks to his contacts with his insurers, has been a critical and frequently prophetic source of information about the law’s rollout, opened a blog post this weekend with the following assessment: “It is now becoming clear that the Obama administration will not have Health.care.gov fixed by December 1 so hundreds of thousands, or perhaps millions, of people will be able to smoothly enroll by January 1.” Laszewski says that months, not weeks, of work remain.

The dates he lists are important, and not only because of the administration’s self-imposed deadline of November 30. Anyone who wants to purchase insurance that kicks in at the beginning of next year must complete enrollment by December 15. If the system isn’t working smoothly at least a couple weeks prior to that rapidly approaching date, then large numbers of people simply won’t have a chance to sign up.

The quick summary is that millions of people would lose their healthcare insurance and suffer a discontinuity of insurance of unknown duration. The Obama Administration would have caused it and there would be nothing they could do about it.

He does propose one interesting stop-gap solution:

So the insurers have suggested a temporary measure: Let the insurers estimate the subsidies on their own. Any estimates that are too low would be reimbursable, and any estimates that are too high, the insurers would get to keep. In other words, the federal government, backed by taxpayers, would be on the hook for their bad estimates.

He thinks that’s pretty unlikely and I do, too.

His worst case scenario reminds me of the scene in Crocodile Dundee in which Mick, visiting New York, is confronted by a mugger armed with a switchblade knife:

Sue Charlton: [guardedly] Mick, give him your wallet.

Michael J. “Crocodile” Dundee: [amused] What for?

Sue Charlton: [cautiously] He’s got a knife.

Michael J. “Crocodile” Dundee: [chuckles] That’s not a knife.

[he pulls out a large bowie knife]

Michael J. “Crocodile” Dundee: THAT’s a knife.

What Mr. Suderman describes is no worst case scenario. IMO the worst case scenario is the above plus the contingency that most of the hardy few successfully enrolling for healthcare insurance under the exchanges are either a) Medicaid enrollees who qualified under the pre-PPACA rules or b) old and sick with pre-existing conditions.

The former condition would impose an intolerable fiscal strain on state governments. Medicaid enrollees who qualified for Medicaid under the pre-PPACA rules but for some reason weren’t enrolled in Medicaid don’t qualify for reimbursement by the federal government. Their entire tabs would be picked up by the states, many of which, like Illinois, are already teetering on the brink of fiscal disaster with no clear way to save themselves.

The latter condition would induce what’s been known as the “death spiral” in which insurers raise their rates and the PPACA’s costs soar in deadly embrace.

And that doesn’t even take what could happen in the group insurance market into account. We could be seeing millions of people formerly covered by their employers thrown into the exchanges.

I’m not predicting any of these scenarios. We’ll know in due course. All could be roses and lollipops.

6 comments

Declaring Victory

Econ prof David Cutler declares victory over rising healthcare costs:

Even as coverage efforts are sputtering, success on the cost front is becoming more noticeable. Since 2010, the average rate of health-care cost increases has been less than half the average in the prior 40 years. The first wave of the cost slowdown emerged just after the recession and was attributed to the economic hangover. Three years later, the economy is growing, and costs show no sign of rising. Something deeper is at work.

The Affordable Care Act is a key to the underlying change. Starting in 2010, the ACA lowered the annual increases that Medicare pays to hospitals, home health agencies and private insurance plans. Together, these account for 5 percent of the post-2010 cost slowdown. Medicare payment changes always provoke fears — in this case, that private plans would flee the program and that the quality of care in hospitals would suffer. Neither of these fears has materialized, however. Enrollment in private plans is up since the ACA changes.

If the PPACA is the cause in the slowing of healthcare cost increases, well, good. If the relationship between the PPACA and healthcare costs is one of false causality, we’ll know in due course. It continues to be too early to tell and, frankly, I’d have more confidence in the pronouncement if an opponent of the PPACA were to proclaim it was lowering costs than when the Obama Administration’s senior healthcare advisor does so.

There are two things we should also bear in mind. First, healthcare costs continue to rise at a multiple of the rate that other costs are. Until healthcare costs are constrained to something affordable, i.e. rising no faster than income, rising no faster than revenues, etc., they’re still increasing too fast. Second, comparing “healthcare inflation” to “general inflation” is specious since healthcare is a major component of personal consumption expenditures. If healthcare comprised 100% of the economy, healthcare inflation (in the sense of rising costs) would by definition be the same as the general rate of inflation. If healthcare were 50% of the economy, its costs were increasing and that increase was twice as fast as the overall increases, all of the increases would be attributable to healthcare.

Healthcare is now a bit over a sixth of the economy. If healthcare costs increase at 6% per year, that’s nearly all of the increases in the entire economy.

28 comments

The Eleventh Hour of the Eleventh Day of the Eleventh Month

Today is Armistice Day, commemorating the end of World War I, the “war to end all wars”. The Great War ended 96 years ago today.

Like Confucius, I believe that the essence of truly human life consists in observing rituals and some of those rituals mark actual, individual events or persons. Like Armistice Day. I don’t find that trying to commemorate abstractions or generalities like “veteran-ness” or “presidents” has the same hold on our emotions as remembering actual events or people. Would “Civil Rights Leaders Day” have the same impact as “Martin Luther King’s Birthday”? I don’t think so and I don’t recall a push for it as an alternative when the day was set aside as a holiday.

So, today I’ll continue to reflect on the end of World War I, why it was fought, and the many errors in the peace that followed that caused it to fail to end all wars.

4 comments

The Damage Done

At the New Yorker John Cassidy writes on the potentially permanent damage done to the U. S. economy by the Great Recession:

Trying to sort the signals from the noise in the jobs report is a tough task in the best of times: the margin of error for the payroll figures is plus or minus ninety thousand jobs. So instead I’ll say a few things about a new research paper by three economists at the Federal Reserve, which is getting a lot of attention because it suggests that the recession and its aftermath have not only done terrible things to the U.S. economy in the immediate sense—high rates of joblessness, tepid gross-domestic-product growth, falling incomes—but also seriously undermined the economy’s capacity for future growth. Gavyn Davies, of the Financial Times, drew attention to the study earlier this week; Reuters published a story about it; and, in today’s New York Times, Paul Krugman devoted an entire Op-Ed to it, which was perfectly justified. It deserves to be discussed widely.

The authors of the paper are David Wilcox, the head of economic research at the Fed, and two of his colleagues, William Wascher and Dave Reifschneider. Although the study uses some sophisticated statistical methods, its basic point is straightforward: in the long term, economic output (G.D.P.) is constrained by the quantity and the quality of economic inputs (labor, capital, and technology). If the growth rate and quality of these inputs decline, the potential growth rate of G.D.P. will fall, too—it’s just a matter of arithmetic.

Since the financial crisis of 2007, the authors argue, that’s precisely what has happened. With hiring rates down, many workers have given up searching for jobs and have dropped out of the labor force. (According to today’s employment report, the labor-force participation rate hit yet another low in October: 62.8 per cent.) With budgets tight, corporations and government departments have cut back on investments in new plants and machinery, computer hardware and software, and research and development. And, with investments in innovation depressed, the rate of over-all productivity growth has slowed down.

If you put all of these things together, they seem to suggest that the economy’s capacity for growth is lower than it was before the financial crisis. How much lower? The authors come up with a variety of numbers, including one that has received a lot of attention: potential G.D.P.—broadly speaking, the level of G.D.P. consistent with stable inflation—“is currently about 7 percent below the trajectory it appeared to be on prior to 2007.” According to the latest figures from the Commerce Department, the G.D.P. is now close to seventeen trillion dollars, and seven per cent of that figure is $1.2 trillion. This is a lot of money to have gone missing, especially if it will never be recovered.

He reproduces the familiar “potential GDP” graph, showing a shortfall of $2 trillion in GDP from where we might be if the growth trend of 2007-2008 had continued. I’m going to propose an alternative scenario.

Above I’ve reproduced his graph but I’ve introduced a solid green line that illustrates where we would be if the growth trend after the end of the dot-com boom had continued. It’s hauntingly familiar.

I agree with Mr. Cassidy that serious damage has been done to the U. S. economy. I disagree about when. Here’s my alternative story.

The dot-com boom was not caused by the policies of the 1990s. It was the consequence of more than a decade worth of capital investment in information and telecommunications technology that finally paid off. Our economy is now a mature economy like France or the United Kingdom rather than a developing one like China. We’ve adopted policies including fiscal, trade, immigration, education, healthcare, etc. that taken together have resulted in the growth we’ve seen for much of our history to slow. The dot-com boom offset the effects of those policies until it ended. When it did the Powers-That-Be were desperate to keep the party going and so adopted policies which resulted in the housing bubble of the Aughts.

The gains produced by that bubble were illusory. There’s no use wondering where all that potential growth went. There was nothing there. It never really existed.

But what about all of those unemployed people and unutilized assets? (I hear someone ask.) My explanation is somewhat along the lines of Tyler Cowen’s “zero marginal product” one. Businesses continue to invest but they’re investing overseas and what domestic investment they’re doing depresses employment. Despite that we continue to import large numbers of unskilled and semi-skilled workers and skilled workers in areas where employment isn’t growing. Businesses are convinced, with a confidence based in experience, that they’ll be able to import the workers they need and continue pushing wages down even as the number of unemployed or discouraged burgeons.

19 comments

What Do You Think About the Deal With Iran?

The negotations with Iran over its nuclear development policy are teetering:

World powers failed to reach an interim deal with Iran over its disputed nuclear program after lengthy talks in Geneva despite days of encouraging signs from the White House that a deal was imminent.

Catherine Ashton, the European Union’s top diplomat, said talks will resume on Nov. 20 in Geneva.

Ashton said there had been, “concrete progress but some differences remain,” BBC reports.

The news came as Secretary of State John Kerry was joined by foreign ministers from the United Kingdom, France, Russia, China and Germany to hold a series of meetings with each other and with Iran’s delegation, headed by Zarif.

France in particular objected to the proposed deal, questioning whether it would go far enough to limit Tehran’s nuclear ambitions, AFP reported. French Foreign Minister Laurent Fabius said Iran’s continued operations at its Arak nuclear reactor and its enriched uranium stockpiles should be addressed to remove Iran from the path of developing the capability to build a nuclear weapon without detection

I think it’s refreshing that we’re not the bad guy in this. The French have taken up that role. I think that John Kerry is an idiot and anything with his fingerprints on it is automatically suspect. I also think that our policies with respect to foreign policy, nuclear proliferation, and Iran’s nuclear development program in particular have been wrong-headed—simultaneously too bellicose and unconvincing.

I’m struggling to form an opinion on the emerging deal with Iran. What do you think?

13 comments

Is the PPACA a Good Policy?

The editors of the Pittsburgh Post-Gazette to remonstrate with those who complain that the president lied about the PPACA. So what? It’s still a good policy:

Plans are being canceled because they don’t meet the coverage set out in the Affordable Care Act and leave many people poorly insured. Those Americans who have been canceled can obtain other policies that may be more or less costly but should offer them better coverage than before.

In short, the furor of the supposed great lie is an embarrassment to Mr. Obama, but it obscures the larger and more important truth that the Affordable Care Act remains good policy.

Is it? I’d like to see somebody make the case. What I’ve seen is people defending the idea of the PPACA or its possible future evolution. If you’re defending the PPACA, here’s what you need to defend. The estimate of people without healthcare insurance is between 30 million and 50 million people. According to the CBO, no more than a third of those would be insured under the terms of the PPACA. Consequently, you need to defend the PPACA not on the basis that it insures everybody or that someday everybody might be insured but that leaving between 20 million and 40 million people uninsured is just fine.

You need to defend the PPACA’s expansion of the Medicaid paid by the states beyond their ability to pay. You need to defend the cancellation of hundreds of thousands or millions of people’s insurance not on the basis of their potentially getting better insurance someday but on the basis that they shouldn’t have had even the insurance they had.

I wish them the best of luck. The key point is not merely that the PPACA is bad but that it is inadequate and doesn’t provide the path to future improvement its advocates think it will.

The other day in comments over at OTB a frequent commenter made what I think is the best and most succinct summary of the PPACA I’ve ever read: it was the least disruptive meaningful change the Democrats in Congress thought they could enact. IMO with respect to our healthcare system disruption is directly proportional to meaning.

Here’s my less than fifty word synopsis of our healthcare system. It provides pretty fair healthcare for those who can afford it. It’s mostly subsidized and the subsidies are determined based on political expediency rather than need or merit. Most of the benefits of the subsidies go to providers, stockholders, and the wealthy.

I think that’s a system that needs more disruption rather than less.

5 comments

Hell, No. We Won’t Go

Showing enviable prudence, most of the uninsured are ignoring Healthcare.gov:

PRINCETON, NJ — In the midst of widespread news coverage of problems with the federal health exchange website, relatively few uninsured Americans (18%) — the primary target population for the exchanges — have so far attempted to visit an exchange website. The percentage is slightly higher, 22%, among uninsured Americans who say they plan to get insurance through the exchanges.

the Gallup organization has found. They also find that nearly half don’t plan to purchase insurance, period. As my wife sometimes say, you can lead a horse to water but that won’t make him into a duck.

33 comments