In a debate almost totally infected with myth, perhaps the most tenacious myth about the Affordable Care Act involves the tsunami of old insurance policies that were supposedly canceled by insurers because they didn't comply with the ACA.
How many policies? The figures are all over the place -- some say 17 million, some say 4.7 million. The implication is also murky: However many cancellations happened, were all these people left without insurance?
Two experts at the Urban Institute have crunched the best numbers we have, and their conclusion is that 2.6 million policies were canceled because of noncompliance with the ACA -- but that more than half the policyholders were eligible for subsidized, low-cost replacement insurance.
In other words, slightly more than 1 million individuals or families had their insurance canceled and had to pay the full premium to replace it.
But that's not the same as saying they were cut off from insurance. Most of them almost certainly did replace it, often with more comprehensive benefits.
Is 1 million a lot? It depends on where you stand, and perhaps more if you're one of the 1 million. But it must be measured against the 30 million Americans whose access to health insurance will be eased by the ACA.
It's important to seek out firm numbers because the picture of a huge tide of people deprived of their insurance by cancellations continues to drive health insurance policymaking in Washington and state capitals.
It's the No. 1 reproach to President Obama, thanks to his "if you like your plan you can keep it" sound bite -- though as we observed in October the notion that all these plans were absolutely beloved by their owners, like the family dog, is also certainly mythical.
Just last week, the Obama administration extended through October 2017 the period in which people can keep their noncompliant policies (with the agreement of their state regulators and their insurers).
A political design
This move is plainly designed to serve political ends, a sure sign that it's not based on empirical conditions. Indeed, Adrianna McIntyre at theincidentaleconomist.com figures that the number of policyholders affected will be so small that the change will have little practical impact on the insurance market.
The myth of the tide of cancellations has opened the door to mischief at the state level. In Washington state, Republicans are offering a bill that would address the so-called cancellation crisis by allowing policies from other states to be sold in Washington without oversight by the state insurance commissioner.
That's just a formula for a race to the bottom and abuse of the consumer, says the commissioner, Mike Kreidler, who notes that it's based on a sizable overestimate of the number and the cause of cancellations in the state.
The individual policy market has been shrinking for years, even before the ACA, Kreidler says, with most coverage being dropped because customers found it too expensive for the skimpy benefits.
So what are the numbers? The Urban Institute economists, Lisa Clemans-Cope and Nathaniel Anderson, start with an estimate from the Centers for Disease Control and Prevention that 14 million people had individual health insurance in 2013.
The Urban Institute's Health Reform Monitoring Survey polled people in this category to determine whether their policies no longer would be available this year. About 18.6 percent said their policies had been ruled incompatible with the ACA. An additional 6 percent said their policies had been or would be canceled for other reasons. More than 75 percent didn't get a cancellation.
Of the 18.6 percent -- that's 2.6 million people -- more than half will be eligible for Medicaid or premium subsidies, according to Clemans-Cope and Anderson.
A few points need to be kept in mind when trying to analyze activity in the individual market. Most important, it is and always has been extremely volatile, with fewer than 17 percent maintaining individual coverage for two years or more.
For the vast majority of people in the market, individual coverage is a bridge or stopgap to employer-sponsored insurance coverage or to public coverage such as Medicare or Medicaid. That's one reason the size of the canceled population has been so hard to estimate.
Michael Hiltzik ,is a columnist for the Los Angeles Times.