Astute IBM (NYSE:IBM) shareholders probably saw the headlines. Jim Chanos, president and founder of Kynikos Associates, recently claimed in an interview with CNBC that the aging tech giant isn't nearly as profitable as it appears to be on the surface. Chanos suggested the company's effective bottom line for the past year is only a little better than half of the profits being reported. He described the disparity as "financial engineering," and though he didn't explicitly levy the accusation against IBM, the IBM-centric conversation included the phrase "sometimes the greatest scams are hiding right in plain sight through the use of pro forma accounting."

As is so often the case, though, there's more to the story.

An investor reviewing her portfolio at a desk.

Image source: Getty Images.

What Chanos said

The interview in question took place during CNBC's "Halftime Report" show on Monday, Oct. 18. Chanos explained:

IBM is supposed to earn almost $11 [per share] this year. For the trailing 12 months they've earned less than $9. But ... if you look at IBM's operating earnings, and add their IP [intellectual property] royalty stream and tax it at a normal 21%, the actual earnings are $6. So you have this almost $5 spread between what IBM is really earning and what they claim they hope to earn on an 'adjusted' basis this year.

Chanos landed a parting shot with, "This is just more financial engineering that this company is doing."

And, he's right ... sort of.

He's absolutely right in the sense that IBM's actual profits and the profit figures being featured within its quarterly reports aren't the same. Namely, operating profits have been and continue to be much bigger figures than actual earnings once all the bills are paid and all the write-downs are booked.

To call it financial engineering, however, is a bit disingenuous.

GAAP vs. non-GAAP

Financial engineering isn't an official accounting or legal term. Rather, it's a vague, unofficial description of how a company may push the limits of accounting standards -- and even laws -- to make its financial condition look stronger than it truly may be. That's not exactly what IBM is doing.

But first things first.

Chanos' beef ultimately points to the difference between GAAP (generally accepted accounting principles) and non-GAAP reporting. GAAP numbers are considered the more accurate of the two in that they factor in every penny coming in or going out in any given quarter, including unusual one-time costs. Non-GAAP figures, or operating figures, reflect how well an organization would have fiscally performed without those unusual expenses being factored in.

Both help investors gauge the strength of a company, though it's actually non-GAAP results that help determine how well that company may do in the future. The "pro forma" accounting Chanos described is the rectification of the two sets of numbers, for the purpose of painting a clearer picture of an organization's current financial situation.

Lately, IBM's GAAP and non-GAAP earnings admittedly haven't been as closely aligned as we'd generally like. Chanos' point that IBM's trailing-12-month earnings (on a GAAP basis) of around $6 per share versus analysts' non-GAAP/operating estimates near $11 is on target. Through the first six months of 2021, IBM reported per-share operating earnings of $4.10, yet that number was whittled down to only $2.52 per share after accounting for all the costs incurred during the first two quarters of the year. However, $0.79 of that reduction is the result of costs linked to acquisitions. Another $0.60 reduction reflects expenses related to the company's retirement plan. Yet another $0.20 reflects costs linked to the upcoming spinoff of IBM's Kyndryl unit, which should improve cash flow, by the way. Add those back in and operating earnings ramp back up to $4.10.

It's also worth noting each of those adjustments was plainly disclosed and described in the company's recent reporting.

That's not to suggest investors should simply ignore these costs. Everything matters, and to be fair, these GAAP-to-non-GAAP adjustments appear to be increasingly the norm for IBM.

The thing is, they're the new norm for IBM because IBM is in the midst of a major overhaul. Aside from the upcoming sale of its managed infrastructure business Kyndryl, the company completed the acquisition of Red Hat in 2019. In the meantime, it has acquired Turbonomic, The Weather Company, and more. It's also sold several marketing, commerce, and software assets since 2019, all in an effort to better focus on hybrid cloud computing. The difference between operating profits and actual profits should be wide, as it's the result of a much bigger plan.

Consider more than reported (or operating) income figures

So, how does one measure the true financial health of IBM, or for that matter, use the past to make informed decisions regarding the future?

It's still tricky to be sure. But smart investors are looking at both GAAP and non-GAAP profit trends, while also considering cash flow trends; cash flow isn't a figure that's as easy to adjust as operating income is.

With this in mind, the image below speaks volumes. While reported net income has been less than operating income for some time, if you can look past last quarter's income and cash flow setbacks that were at least somewhat impacted by the chip shortage and IBM's effort to reshape itself, you'll actually see the company's operating cash flow has been firming up for a few years now. It crawled back above operating and actual income back in 2016. That's a big clue that IBM is gaining the traction it's been aiming for.

IBM Operating Income (Quarterly) Chart

IBM Operating Income (Quarterly) data by YCharts

And this can't be stressed enough: Judging any company on their reported income since the COVID-19 pandemic took hold early last year isn't exactly a fair assessment of that company. We're in very unusual circumstances.

Keep it in perspective

Again, none of this is to suggest IBM's GAAP/non-GAAP disparity should be dismissed. The company is spending a ton of time and money to transform itself. That's not easy, or cheap. Patient investors understandably have high expectations of IBM's new hybrid cloud computing focus.

Still, it's shortsighted to jump to conclusions based solely on a set of recent numbers without understanding the "why" behind those numbers. Now priced at only 13 times this year's projected operating/non-GAAP earnings and 12 times next year's estimated earnings, the wide difference between them and the company's probable GAAP results is arguably already built into the stock's price.

And here's another important detail: Jim Chanos' Kynikos Associates is shorting IBM shares. If the stock loses value, Kynikos makes money. That makes the voicing of his assessment at least a bit biased.

IBM reports its third-quarter results after Wednesday's close.

This article represents the opinion of the writer, who may disagree with the “official” recommendation position of a Motley Fool premium advisory service. We’re motley! Questioning an investing thesis -- even one of our own -- helps us all think critically about investing and make decisions that help us become smarter, happier, and richer.