President Biden is Telling the Truth About Taxes, so Why Is a Fact-Checker Calling him a Liar?

It was sensible to welcome the emergence of fact-checkers at major newspapers and as independent journalistic operations.  With the internet speeding up the spread of falsehoods, and with politicians like Donald Trump and his cultists willing to pile lie upon lie, it seemed necessary to have people act on the adage attributed to Daniel Patrick Moynihan that "everyone is entitled to their own opinion but not their own facts."  Someone needed to call out the liars, right?

There are three problems with this idea -- none of them fatal to the basic value of true fact-checking, but all of them serious enough to make it necessary to adjust our expectations regarding what all of this can accomplish.  The people doing the fact-checking are mostly journalists, after all, which means that they will bring with them all of the biases and professional norms that have prevented plain old journalism itself from performing a reliable fact-checking function.

This leads to the first problem with fact-checking, which is that journalists hew strongly to the conventional wisdom on almost every topic, which causes them to believe fervently that, say, fearmongering about federal borrowing is an objective statement of fact rather than a tendentious policy position.  (See my most recent cri de coeur on this particular policy question here.)  The second problem is bothsidesism, which causes journalists to want to prove their even-handedness by showing that they will call out "lies" by people who disagree with Trump and other Republicans just as readily as they will call out right-wingers' lies.  The third problem is that it simply might not matter whether any fact is ever checked, because there is plenty of reason to believe that no one's mind is changed by factual corrections -- and indeed that repeating the lies in order to debunk them reinforces rather then undermines their influence on public discussoin.

The first two problems are strongly interrelated because, as noted, they both are rooted in a form-over-substance version of journalistic "fairness and balance."  Even there, however, some of the examples of fact-checking gone sideways are extremely frustrating.  One of the most infamous in its time was PolitiFact's 2011 "Lie of the Year," which was not only not the worst lie of 2011 but not a lie at all.  I will not go into the details here (which had to do with Democrats and Medicare), but interested readers can find a short contemporaneous criticism here.

Similarly, I have occasionally mentioned a 2013 fact-check regarding a Republican claim that the US tax litigation system has a presumption that the government's position is correct -- that is, that Americans are supposedly treated as guilty until proven innocent, just like in France!!  (See two of my columns on that topic here and here.). That was especially annoying because the legal presumption in favor of the taxpayer is black-letter law, but that did not stop the fact-checker from finding a superseded (and quite obscure) Treasury regulation that could possibly have misled some people to think that the legal burden was on the taxpayer.  Yet the fact-checker did not say, for example, that the claim was "false but possibly understandable" but instead that it was "mostly true."  Arghhhh.

Note that both of those examples preceded the emergence of Trump.  And indeed, because Republicans discovered that they could work the refs and pass off outright lies as either mere opinions or as facts (especially in the 2012 Romney/Ryan campaign operation), we have been relying on at least ad hoc fact-checking (no matter whether we formally call them "fact-checkers") for many years.  Again, and to be very clear, the inherent problems with fact-checking (at least as it has been done in the US) are reason for concern, but not having fact checkers would almost surely be worse overall.  When they go off the rails, however, it is truly a sight to behold.

Why bring all of this up now?  In my column earlier this week discussing why societiess regulate businesses, I gingerly raised the issue of fact-checking in the context of an August 2023 claim by Robert F. Kennedy Jr., which Washington Post fact-checker Glenn Kessler had ably debunked: "Being who he is, however, [Kennedy] could not keep his story straight, so much so that even Glenn Kessler, the fact-checker for The Post who by no means bats 1.000, had fun taking Kennedy down.  Kennedy does makes it easy, presenting everything in his usual conspiratorial terms and getting basic facts wrong."  The low-hanging fruit includes Kennedy's mixing up of the companies Blackstone and BlackRock and, more seriously, his risible claim that only three companies own 88 percent of the corporations that make up the S&P 500.

As I was writing that column earlier this week, I went out of my way to highlight Kessler's highly suspect track record in part because I had in mind the example that I will discuss here.  Specifically, he recently issued a fact-check in which he decided what counts as "income" for tax purposes and then gives President Biden "two Pinocchios" for lying about how regressive the US tax system is.  The problem is that Kessler does not know what he is talking about or, worse, he is deliberately mischaracterizing Biden's argument for ideological reasons.

Biden has included in recent speeches the fact that high-income Americans pay about 8.5 percent of their incomes in federal taxes, which is a much lower rate than less well-off Americans pay (by a lot).  This idea gained widespread notice because of a 2021 study by Pro Publica, which measured income not by the arbitrary definitions that Congress has created in the tax code (gross income, adjusted gross income, taxable income) but by a true economic measure of income.  Pro Publica thus measured what they call "true tax rates," finding that the some of the wealthiest Americans pay close to zero tax rates (0.10 percent, in Warren Buffet's case).

The difference between tax rates that are measured against statutory measures of income and "true" tax rates is a result of arbitrary exclusions of some types of income from official measures of income.  Congress is allowed to say (as it often does) that something can be excluded from gross income or deducted when computing taxable income, but that does not mean that the exclusion or deduction was not income in the first place.  If I receive something worth $100,000 with no offsetting liability, I am $100,000 richer, and I thus have received $100,000 in income, no matter what Congress says I should call it.

The most important statutory game that distorts the picture is the difference between "realized" income and "unrealized" income.  Both are income, the difference being only that the former kind of income has been converted into cash while the latter has not.  If I receive, say, a million dollars of income but am able to exclude $900,000 of it because Congress says so, and if I then pay $25,000 in taxes on the remaining $100,000, my tax rate as measure by standard methods is 25 percent (25k divided by 100k), whereas my true tax rate is 2.5 percent (25k divided by 1000k).

There is an entirely separate set of questions about whether unrealized income should be taxed and even whether it is possible to tax it as a practical matter.  This is also the underlying constitutional question in the Supreme Court's Moore v. US case this term, which I discussed two months ago after oral argument.  I will write more about that case soon enough, but Kessler's foray into this area is a different matter, because he wants to say that Biden is lying about the underlying facts.

After quoting Biden's oft-repeated claim that the one thousand US billionaires pay only 8.5 percent on average in federal taxes, Kessler offers this riposte:

But if you check Treasury Department calculations for what the richest Americans already pay in taxes, you would see that the top 1 percent pay in excess of 20 percent in income taxes and more than 30 percent in all federal taxes. Even if you drill down to the top 400 wealthiest taxpayers — data that was publicly available on an annual basis until President Donald Trump killed the report — they paid an effective tax rate of 23.1 percent in 2014.

At this point, we can already see what is happening.  Treasury's method of computing tax liability would come up with, in my example above, the 25 percent rate, because they leave out unrealized income.  Biden's numbers include all of that income, so his estimate is the equivalent of the 2.5 percent computation in my example.  At worst, one would think that this is a judgment call -- one might even say a political argument -- about which not-false numbers to use, not a fact-checkable moment.  Kessler disagrees.

Other than quoting several conservative economists' rants to the effect that unrealized income should not be taxed (perhaps not even discussed), however, Kessler comes up with nothing.  (I should add, however, that to his great credit the conservative economist Alan Viard does admit bluntly that "unrealized income is income," which Kessler quotes and then effectively waves away.)  Kessler's pithy summary: "The president is describing a tax system that he wishes existed — not the system in place."

No, the President is describing the tax rate as measured against the true total income that these billionaires receive.  I should also add that these billionaires do not receive their income in nontaxable form by accident, because they lobby heavily to be able to continue to exclude unrealized income from their statutory income and they are then able to make sure that they receive their incomes in that tax-favored form.

Kessler tries to undermine the notion of true income by saying that two economists long ago "proposed a broad definition of income," which is like saying that two physicists "proposed" an accurate definition of velocity.  He then runs some numbers and says cheekily: "So, just by fiddling with the income figures, we have cut the tax rate in half."  You see, we are merely messing around with some definitions, which is apparently supposed to seem fishy.  He also tries to undermine Biden's case by saying repeatedly that his source was a "blog post" -- as if that is a bad thing? -- and thus not a matter of rigorous analysis.

While I agree with Kessler's very minor point that Biden should update the number that he uses in his speeches to the most recent estimate (which is slightly higher), that is most definitely not the drum that Kessler is beating.  He concludes:

There’s some value in highlighting how not all income obtained by the wealthy is currently subject to taxation. But he shouldn’t suggest that billionaires already are paying such a low percentage of their income in taxes when numerous reports by the Treasury Department, the Congressional Budget Office and other agencies show they pay a much higher effective tax rate under the current tax system.

That is worse than fatuous.  Biden is not "suggest[ing]" that billionaires are paying a lower rate than the "numerous reports" by Treasury and CBO but rather that billionaires are in fact paying that lower rate.  Now.  Because they are.  There is no factual dispute here.  Kessler simply seems determined to say that Biden is lying, no matter the facts.

Why would Kessler do such a thing?  When I wrote earlier this week that he "by no means bats 1.000," I linked to a column from 2018 in which I discussed a Kessler fact-check about Bernie Sanders's Medicare-for-All plan.  Kessler insisted that Sanders was a three-Pinocchios liar for having relied on a particular assumption about how much money Sanders's system might save.  Kessler said that Democrats were "cherry-picking," which is far short of a lie; and even that was dishonest, because in fact Democrats were pointing to an analysis by a conservative economist that showed that their analysis was accurate under a specified set of assumptions.

What do these two examples have in common?  One sees Kessler faulting Democrats for favoring a progressive change to the health care system, and the other sees Kessler faulting Democrats for favoring a progressive change to the tax system.  In neither example can he show that they lied, only that they presented facts in a way that he dislikes.  So he calls them liars.  When does fact-checking stop being about checking facts?  When mainstream pundits panic and feel the need to defend the status quo by shouting down unwelcome facts.