A "key bit of data" sometimes

We say PolitiFact uses inconsistent methods in fact-checking political claims. An April 27, 2016 fact check of the "New Day for America" Super PAC serves as a recent example, paired with a comparable case from the 2012 presidential election.

The 2016 claim from "New Day for America" charged Democratic presidential candidate Hillary Rodham Clinton with promising to raise taxes by $1 trillion. PolitiFact's conclusion said the number was about right but charged the Super PAC with ignoring the 10-year time frame over which the tax increase would produce its revenue. PolitiFact  added that Clinton would not even be in office for the entire 10 years. PolitiFact also claimed it was important who paid the taxes, calling that "another key bit of data." "New Day for America" ignored two key bits of data, so "Half True":

[Clinton's] plan does, in fact, call for raising a trillion dollars, but it would do so over 10 years — longer than she could serve as president, even if she were re-elected. So if she brought in roughly $100 billion per year, even a two-term Clinton administration couldn't fulfill a promise to bring the total to $1 trillion.

Also, the statement ignores another key bit of data — that the money would be raised by tax changes targeted to the richest Americans, a group that has seen its top tax rate drop dramatically since the 1950s and early 1960s, when the marginal tax rate was over 90 percent.

Because the statement is partially accurate but leaves out important details or takes things out of context, we rate it Half True.

We couldn't remember PolitiFact using this approach to a tax proposal before, so we looked for a comparable example from PolitiFact's past. It turned out President Obama's re-election campaign back in 2012 said challenger Mitt Romney's tax plan would "add trillions to the deficit." PolitiFact examined that claim along with the associated claim that President Obama's tax plan would cut deficits by $4 trillion. In both cases PolitiFact found that the numbers were inflated (cherry-picking high estimates). The details of Romney's tax plan were unknown, so the accusations about his tax plan had little foundation in fact.

More importantly, the effects of the tax plans were estimated over a 10-year period in both cases. Neither Romney or Obama would serve as president over the full 10 years. The Obama ad said Romney wold cut taxes for millionaires,, but ignored other tax cuts and tax increases in the plan. Good enough for PolitiFact? The Obama ad received a "Half True" rating, the same as "New Day For America" even though the Obama ad committed the same errors and more.

We encourage readers to look at PolitiFact's rating of the Obama campaign, especially the summary paragraphs, to see how PolitiFact largely forgave the fundamental inaccuracies in the ad. Compare it to the summary PolitiFact offered for the ad coming from "New Day For America."

PolitiFact simply does not use the same standards for the two ads.

In Fact-Checking This Is a Big Deal

Why do we nitpick PolitiFact over this kind of thing? Both "Half True" ratings are justified on reasonable grounds, aren't they?

No. Full stop. Good fact-checking requires a consistent approach to the issues. PolitiFact repeatedly fails to achieve consistency.

PolitiFact's rating system has always been a sham, because PolitiFact follows no rigid definition for its ratings. Sure, PolitiFact usually attempts justifications, but they are all over the map. For example, PolitiFact once gave Mitt Romney a "Half True" rating because the problems with his claim matched PolitiFact's definition of "Mostly True." Seriously, that's how PolitiFact justified its rating of Romney. PolitiFact has let the error stand for years.

Critics left and right have panned PolitiFact's rating system. Defenders often claim that the ratings aren't important. The important thing, they say, is the detailed information we get in the fact check. But if PolitiFact varies in its approach, finding a problem in one instance and overlooking that same problem in another instance, it gives its readers poor fact-checking.

Of course the problem is worse when the inconsistencies favor one political leaning over another.


Afters:

In defending Clinton from the "New Day For America" ad, PolitiFact pointed out that her tax plan tries to increase taxes on the rich, ostensibly to help restore the balance in effect in the 1950s:

(T)he statement ignores another key bit of data — that the money would be raised by tax changes targeted to the richest Americans, a group that has seen its top tax rate drop dramatically since the 1950s and early 1960s, when the marginal tax rate was over 90 percent.

PolitiFact evidently did not bother to check its facts on this point. PolitiFact writer C. Eugene Emery Jr. supported his statement by providing links to raw data showing top marginal income tax rates. But changes to tax law affecting deductions have kept the effective tax rates on rich people from changing much. It was rare for anybody to pay the highest marginal income tax rate from the 1950s. Ari Gupta, in a paper for the Manhattan Institute, pointed out that popular liberal economists Thomas Piketty and Emanuel Saez admitted as much:

The reduction in top marginal individual income tax rates has contributed only marginally to the decline of progressivity of the federal tax system, because with various deductions and exemptions, along with favored treatment for capital gains, the average tax rate paid by those with very high income levels has changed much less over time than the top marginal rates.

Why is it okay to call out others for leaving out key information while at the same time omitting key information in one's own reporting? Ask PolitiFact. But don't expect an answer.