How is the integrity crisis in business reporting like the integrity crisis in science?

We’ve talked a lot over the years about the replication crisis in science, but maybe we should be calling it the integrity crisis.

Lack of integrity isn’t quite the same thing as fraud. You can be doing bad science because you’re in a hurry, or because you have some point you want to make, or just because you’re using bad methods, and none of this is necessarily fraud or even “scientific misconduct,” but it still reflects a deficit of scientific integrity. Indeed, you might well be a person of high moral integrity in other aspects of life–you might pay your bills on time, be nice to your family and coworkers, you might even be doing your science with a goal of helping the world–and you might, to the best of your knowledge, be working with accurate data and be using legitimate research methods–but you could be having problems with scientific integrity (dictionary definition: “firm adherence to a code of especially moral or artistic values : incorruptibility”) in that you’re studiously avoiding potential problems with your work. This relates to Clarke’s Law: “Any sufficiently crappy research is indistinguishable from fraud.” When that guy was writing those beauty-and-sex-ratio papers, I didn’t notice any fraud, but I did notice that he avoided confronting the clear problems that had been pointed out with his work. I see this as a lack of integrity. Similarly, when those Nudge authors memory-holed their earlier effusive praise of a now-disgraced food-behavior researcher, that was a lost opportunity for them to learn, also a lack of integrity in that they were avoiding a chance to see how they were on the wrong track. Not fraud, but a deficit of scientific integrity.

I was thinking about this after reading this Columbia Journalism Review article by Sara Silver subtitled, “Understanding money is key to grappling with power. Business journalism isn’t set up for that,” and which begins:

When stock markets closed on January 19, 2021, Netflix posted a press release with its end-of-year financial results. Business is booming, the message went. About a year into the coronavirus pandemic, the company had provided “escape, connection, and joy” to more than two hundred million subscribers. A New York Times journalist had an hour to review Netflix’s earnings report: eleven pages, half corporate-speak, half dense tables filled with accounting metrics–some of them standard, some invented. During after-hours trading, Netflix stock began to rise.

The pressure was on to explain the numbers while standing firm against salesmanship. In its release, Netflix had highlighted operating income. That excluded hefty interest on its debt (the company had borrowed sixteen billion dollars to fund a spree of shows) and the expense of taxes (a year before, Netflix had quadrupled its profit by emptying a tax reserve that it had only recently stuffed). But Netflix’s finance whizzes had turned straw into gold. The Times story duly noted the company’s rising operating income, not the fact that profit was down 8 percent for the quarter–and the drop looked even steeper when compared with the burgeoning sales. . . .

The Times noted that Netflix would still have ten billion to fifteen billion dollars in debt, but said that the company “made enough revenue to pay back those loans while maintaining its immense content budget.” Netflix now makes enough to repay its debt, the story went; “the gambit seems to have worked.”

Silver summarizes:

The article was written, edited, and posted in seventy-eight minutes. Alas, it was wrong.

Ulp.

The business reporting that Silver is criticizing is an example of what I would call a deficit of integrity in business reporting. Again, the claim is not that the reporters are committing fraud or even knowingly making an error–as Silver puts it, “No one is dumb in this story. Nor is the situation unique . . . Consistently, and across the press, the rush to cover corporate earnings sets journalists up for failure.”

The rest of Silver’s article discusses the conditions under which these reporting problems arise:

Earnings coverage did not always work this way. In the nineties, when I [Silver] started out as a business reporter, journalists could lean on sell-side analysts, as they were known, who worked for banks that issue stocks and bonds for companies. It was their job to help us understand corporate strategy and to warn of red flags. They toured factories and took meetings with executives; they connected the numbers on an earnings report to the actions of CEOs, supply chain snafus, demographics, and emerging technologies. . . .

Seasoned reporters found ways to sidestep obvious conflicts of interest, knowing that analysts would recommend that clients buy stocks their banks were selling. In effect, Wall Street was policing itself. . . . [But] banks began axing their research departments, just as publishers were shrinking newsrooms. The financial crisis of 2008 and 2009 led to more bloodletting and consolidation, with fewer companies listed on public stock markets, and fewer people buying individual stocks than a slice of the market through index funds. . .

It’s an assembly line: One person listens to quarterly earnings calls with executives. Another checks news that could affect the company’s customers, suppliers, or competitors. One plugs earnings numbers into a spreadsheet. Only the head of the team makes calls–and the calls are narrow, predicting short-term earnings. These judgments are made within minutes. Earnings often beat expectations–mostly in alignment with companies’ guidance–which enables analysts to recommend that clients buy shares. That, in turn, sends stock prices up. For journalists, these recommendations are mostly useless.

And yet few business reporters, even at elite news organizations, understand the implications of all this change. Market news sites still churn out stories on individual stocks based on sell-side research. When a company’s numbers beat Wall Street estimates (set within a range the company provided), that typically becomes a story’s lede.

This is all interesting in itself, but also it reminds me so much of . . . the process of scientific publication and review: a push toward binary judgments (positive or negative), a trust in intermediate authorities, a hollowing-out of expertise, conflicts of interest.

And the integrity crisis, which is deeper than mere fraud and corruption.