The Calcbench research department is back to non-GAAP financial metrics again, this time with a report analyzing how firms in the S&P 500 reported their adjusted earnings — and which adjustments accounted for the largest gaps between GAAP and non-GAAP.

You can download the complete report from the Calcbench Research Page. Meanwhile, here are some of the highlights.

What We Did

First, we examined the 2020 earnings releases of all S&P 500 firms that reported both GAAP and non-GAAP net income numbers. We measured the difference (in dollar terms) for the 60 firms with the largest differences between GAAP and non-GAAP net income.

We then counted the number of adjustments those 60 firms made, and grouped all those adjustments into 11 broad categories. Then we studied the size of each non-GAAP adjustment (in dollar terms) for those 11 categories.

The three big conclusions we found:

• Non-GAAP net income reported by our sample group exceeded GAAP net income by $132.3 billion. This is more than doubling total reported GAAP net income of $130.7 billion.
• The sample group made more than 240 adjustments to GAAP net income.
• Amortization of Intangibles was the single biggest category of specified adjustments, accounting for 30 percent of all adjustments.

The breakdown of adjustments across all 11 categories is in Figure 1, below.

A good example of the adjustments comes from Bristol Myers Squibb ($BMY). The company reported a $9 billion loss for 2020, and then added back another $23.78 billion in various adjustments — for a non-GAAP adjusted net income of $14.77 billion. See Figure 2, below. The column with the adjustments is boxed in red.

And an Other Thing...

We’d be remiss if we didn’t note that 22 percent of the non-GAAP adjustments we tracked fell into the ever-popular “Other” category, accounting for $29.47 billion. What are those Other adjustments? Who is making them?

We looked deeper. Turns out, 75 percent of that amount can be explained by just four firms:

• AmerisourceBergen ($ABC) and AbbVie ($ABBV) both had “contingent considerations” of more than $5 billion each.
• UPS ($UPS) had an almost $5 billion mark-to-market loss on its defined benefit plan.
• Hess ($HES) has $2.1 billion in exploration costs.

Our report has numerous other examples of firms and their non-GAAP adjustments, including examples of revised non-GAAP numbers that change when a firm’s reconciling items change. So download the full report, and let us know what you think!

The other day we had a blog post about CEO pay ratios, which are one example of the compensation data that Calcbench tracks. Today we’re going to stick with that theme, but explore something a bit larger — how to pull executive compensation data for a large number of executives quickly and simply (professional subscribers only).

Begin on our XBRL Raw Data Query page, and look near the top. You’ll see a link to another page called Non-XBRL Earnings Release data. See Figure 1, below.

That page lets you find various pieces of non-GAAP data that firms include in their financial report, such as adjusted earnings, free cash flow, and similar items. But Calcbench also lets users pull certain predetermined non-GAAP reports — and executive compensation is one of them.

Say you want a quick download of 2020 executive compensation data for all firms in the Dow Jones Industrial Average. You would start on this page, and set your “Choose Companies” selection to the DJIA. In the “Fiscal Year” field, top right, you’d enter 2020.

Then at the bottom of the page, you’ll see the option for an “Executive Compensation” report. See Figure 2, below; and note the red arrow pointing to the report you want.

Click that choice, and you’ll get a spreadsheet with all firms in the DJIA that have already filed their 2020 proxy statements. Each executive gets his or her own line, with a neat breakdown of compensation data by type: base salary, bonus, share grants, option grants, pension payments, and the ever-popular “other.” The far right column will report total compensation.

See Figure 3, below, for an example. It’s not the best shot because, well, there’s just too much data for us to show in one image. (#HumbleBrag.) But you get the idea.

You can also use this page to conduct more expansive searches of data. For example, you could search a single company for several years’ of data at once; or even multiple companies for multiple years of data, although you’ll probably get to a large pile of data quite quickly. And as you can see from Figure 3, above, every line of data we return also includes a link back to the original source document filed with the SEC.

Happy searching!

Calcbench recently had the pleasure of (virtually) chatting with Rani Hoitash, the John E. Rhodes Professor of Accounting at Bentley University. Hoitash is one of the early adopters of XBRL, having developed an interest in the data-tagging technology even before the SEC began requiring companies to submit their filings tagged in XBRL. 

Q. What drove your interest in XBRL?

A: In the early 2000s I became interested in XBRL, but at that point companies were not required or allowed to use this technology to file their financial reports. As a researcher, one of my primary interests is in metadata. I focus on the tags that enable me to extract every aspect of accounting data. Prior to XBRL, no one could collect this level of richness.  

Q: You were one of the first academics to use our platform. How did you come across Calcbench? 

A: I heard from a friend that Calcbench had been extracting and collecting XBRL data and I jumped on the opportunity to work with them. My initial exposure to Calcbench was in the form of a flat data file. It was a huge file, with every piece of data they had. I used that data in my Accounting Review paper on Measuring Accounting Reporting Complexity with XBRL. What excited me about the data from Calcbench was that it wasn't normalized, the way it is with Capital IQ and others.  

Following the Accounting Review piece, I recently published another paper in Auditing: A Journal of Practice and Theory on eXtensible Business Reporting Language (XBRL): A Review and Implications for Future Research. This paper summarizes past research and presents many opportunities for researchers. The paper also discusses practical implications and challenges in using XBRL, for example, XBRL data is currently not audited. We write in the paper that Calcbench is one of the major sources for those interested in working with XBRL data.

Q: How does Bentley University use Calcbench? 

A: I can’t speak to all the use cases, but some of my doctoral students are using Calcbench to extract text data from the filings. With Calcbench it’s easy to search for keywords. The platform is made for researchers who go beyond the numbers. There’s a lot of opportunity to analyze text data from the filings. With Calcbench, researchers can analyze text more efficiently and accurately. 

Q: When do you think XBRL will become mainstream?

A: Not everyone is aware that you can extract financial data without technical knowledge. Undoubtedly the awareness will come. We are living in an era of big data; investors, researchers and other stakeholders want more data, not less. The more people write about the power of XBRL, the more researchers and practitioners will take advantage of this great data source. 

Q: What do you want others to know about Calcbench? 

A: You can be a data expert without knowing the structure behind it. XBRL code can be very confusing. Calcbench is able to do the complex stuff for you. In addition, the level of richness is unparalleled. When I refer to the richness of XBRL, it is analogous to a kid going into a candy shop. For anyone who loves data, Calcbench gives you shelves full of data candy.

-30-

Corporate America won’t start filing Q1 2021 earnings reports for a few more weeks yet, so the crack Calcbench research team decided to spend some time lately looking over corporate proxy statements for 2020 — which are arriving at a brisk pace these days.

That brings us to one of our favorite pieces of esoteric financial data: the CEO pay ratio!

For those unaware, the CEO pay ratio is a required disclosure for most U.S. public companies. It’s the ratio of total compensation for the CEO compared to the median total compensation of all other employees. The Securities and Exchange Commission began requiring CEO pay ratio disclosure for 2017 fiscal years, so this is now the fourth year of CEO pay ratio data we have.

Figure 1, below, is an example from Flowers Foods Inc. ($FLO), which filed its proxy statement on April 13.  

The calculations are fairly straightforward. Total compensation for CEO Ryals McMullian was $5.11 million, and median employee total compensation was $80,400. So the CEO pay ratio was 63.6 to 1.

The more interesting details, as always, are the footnote disclosures underneath those numbers. As part of the disclosure, a firm must describe how it defined and calculated that median employee compensation number. For example, did the firm include contracted labor? (Not in Flowers’ case, apparently.) Did it include overseas employees, whose compensation might be much lower than that for U.S. employees, and therefore could skew the ratio to some unnatural extent? (Flowers has no overseas employees, so that’s not an issue here.)

The exact presentation of the CEO pay ratio can also vary. Look at the disclosure for Southern Co. ($SO), which filed its proxy statement on April 12, in Figure 2. All the necessary numbers are included, but you have to read the words to understand what’s what; no easy-to-read table presentation like Flowers gave in Figure 1.

If you can’t sufficiently squint, CEO Thomas Fanning had total compensation of $22.38 million in 2020, and median employee total compensation was $122,760 — for a CEO pay ratio of 134 to 1.

How Do You Find This Data?

Calcbench can help you find CEO pay ratio data in several ways. If you’re looking for data from a specific firm, start at our Interactive Disclosure page. Select “proxy” from our Choose Disclosure Type pull-down menu on the left side of the screen. Then do a text search for “ratio” and you should find the disclosure pretty quickly.

You can also search for CEO pay ratio disclosures in bulk on our Multi-Company page. Set the group of companies you want to search, and then enter “CEO Pay Ratio” in the standardized metrics field (because,yes, Calcbench tracks the pay ratio piece of data).

Figure 3, below, shows the CEO pay ratio for S&P 500 firms that have filed their 2020 proxy statements so far. As you can see, Starbucks ($SBUX) tops the list with a ratio of 1,657 to 1; followed by Coca-Cola ($KO) at 1,621 to 1.

Yes, you can also track the CEO pay ratio numbers for a firm over time — but beware the details! For example, changes in interest rate assumptions might push the costs of a CEO pension plan upward; or a decision to redefine a median employee (say, to include newly acquired employees in a low-cost overseas region) might push the median employee number downward.

The devil really is in the details for this particular number. Thankfully, Calcbench has all the details you want.

You may have seen news last week of yet another research report declaring that a significant swath of Corporate America paid no federal income taxes last year. 

This time around the report came from the Institute on Taxation and Economic Policy, which found that 55 large businesses paid zero to Uncle Sam in 2020 even while they reported a total of $40.5 billion in pretax income. ITEP’s report hit the interwebs on April 2, and was promptly picked up by the New York Times and other big media outlets.

Our only question: Everyone knows this data is readily available, right? Calcbench has been providing data on corporate federal tax payments — or the lack thereof.

That is, publicly traded firms need to disclose their federal tax expense or benefit in their quarterly reports. Those numbers are tagged as XBRL data, which means Calcbench can find them with just a few keystrokes.

For example, we pulled up our Multi-Company database page, and entered “CurrentFederalTaxExpenseBenefit” in the search field that page has for XBRL tags. Within moments, we found what all firms the S&P 500 paid in federal taxes for 2020. Those numbers are the same data ITEP used to compile its report. See Figure 1, below.

The third column denotes federal taxes paid. Notice that those numbers are all in red. ITEP is correct that scads of large firms paid no federal taxes in 2020 even though they generated gobs of pretax income.

For example, DTE Energy ($DTE) reported a negative federal tax expense of $247 million — meaning, the company actually received that amount back from the U.S. Treasury, rather than paid anything into it. That’s the number ITEP included for DTE in its report, and it’s also what we found with a moment’s search on Calcbench.

ITEP’s report flags 55 firms that received $3.5 billion back from the U.S. government in 2020 while also reporting $40.5 billion in pretax income. Not all of those firms are in the S&P 500, but they are all publicly traded, so they’re in the Calcbench data archives somewhere.

On our first (and rather naive) try, we found a total of 126 firms that had assets of more than $100 million, and that had reported domestic profits, and that had negative federal tax expenses in their fiscal 2020 year end filings.  

Meanwhile, we can say that among the 451 S&P 500 firms that have filed their 2020 reports so far, those firms paid a total of $95.6 billion in federal taxes, against $1.076 trillion in operating income.

Of course, much more goes into corporate tax analysis than a firm’s federal tax payment. Many firms that didn’t pay federal taxes did pay state, local, or international taxes. And tax management, where a firm claims various deductions, credits, and other maneuvers to lower its cash payment, is a time-honored tradition among corporate giants.

You can find details about a firm’s tax payments via our Interactive Disclosure page. Call up the firm in question, select its tax disclosure from the drop-down menu on the left, and start digging.

Two big retailers filed their 10-K reports for 2020 this week: Macy’s ($M) and Abercrombie & Fitch ($ANF).  

As one might expect from retailers, their 2020 numbers weren’t terribly good because of the pandemic. So both firms also reported adjusted net income figures, to account for various one-time expenses related to COVID-19 disruptions.

We decided to analyze and compare those non-GAAP disclosures, to see how one item labeled “adjusted net income” can include very different things, even between two firms that work in the same industry.

Example 1: Macy’s

Macy’s reported $17.34 billion in revenue for 2020, down 29.4 percent from $24.56 billion in 2019. The company also swung from $564 million net income in 2019 to a net loss of $3.94 billion in 2020. Most of that 2020 loss came from a $3.58 billion charge for restructuring, store closing, and other costs.

In the earnings release Macy’s filed on Feb. 23, we can see that the firm reported an adjusted net loss of only $688 million. The reconciliation is in Figure 1, below.

OK — so what does that $3.58 billion restructuring and impairment charge actually entail?

To decipher that number, we had to jump to the Interactive Disclosure tool and read the restructuring note Macy’s included in its 10-K, filed on March 29. There, we can see that most of the charge stemmed from a $3.28 billion impairment to goodwill; plus another $154 million in a restructuring charge that was mostly severance pay; plus assorted other charges. See Figure 2, below.

The other significant portion of Macy’s non-GAAP net income was that $412 million adjustment for taxes. Again, using the Interactive Disclosure viewer to find the details, we can see that Macy’s arrived at the $412 million number through nine separate credits and charges, including a $205 million carryback benefit allowed under the CARES Act.

Example 2: Abercrombie & Fitch

Abercrombie & Fitch, meanwhile, reported $3.12 billion in revenue for 2020, down 13.7 percent from $3.62 billion in 2019. The company swung from $39.36 million net income in 2019 to a net loss of $114 million in 2020.

The adjusted net income numbers are a bit more tricky here. In its earnings release filed on March 3, A&F includes an adjusted operating income in dollars, but the adjusted net income is only reported in earnings per share. See Figure 3, below.

The most important number on this entire table, however, is that Footnote 4 next to the phrase “excluded items.”

What does that mean? In the notes of the earnings release, A&F says: “Excluded items this year consist of pre-tax asset impairment charges which are principally the result of the impact of COVID-19 on store cash flows. Excluded items last year consist of pre-tax asset impairment charges related to certain of the company's flagship stores.”

So Abercrombie had an operating loss of $20.47 million in 2020, and then made an upward adjustment of $72.94 million for “excluded items” that resulted in a non-GAAP, adjusted operating profit of $52.47 million.

To decipher that $72.94 million, we first had to look at Abercrombie’s income statement on the Company-in-Detail page. The $72.94 million was there plain as day, listed as an asset impairment charge. Then we had to open up the Interactive Disclosure database again, and found the Asset Impairment footnote explaining exactly what assets were impaired.

The answer is Figure 4, below. It’s all impairments to operating leases and property, plant and equipment.

Abercrombie also has another footnote explaining the impact of COVID-19, and in that disclosure the company coughs up a $14.8 million write-down of inventory that apparently went out of style during the store closures of early 2020. But that inventory charge is separate from the charges above, and not reflected in the non-GAAP adjusted numbers in Figure 3, either.

So there you have it: two retailers, reporting two non-GAAP numbers to reflect the costs of COVID-19. The details are always in there somewhere, if you know where to look.

Non-GAAP financial disclosures are a fact of life in modern corporate reporting, and Calcbench is fine with that. We also like to see neat, thorough reconciliation of non-GAAP financial metrics back to their GAAP counterparts.

How should those reconciliations work? For your consideration, we submit the outstanding non-GAAP reconciliation that Ulta Beauty ($ULTA) included in its March 11 earnings release.

Ulta operates retail stores across the United States to sell cosmetics and other beauty products. So as one might imagine, a global pandemic that closed retail stores for weeks on end and left millions of people with no need to put on makeup for office jobs and nights out — well, that hit Ulta rather hard.

Annual revenue fell 16.8 percent, from $7.4 billion in 2019 to $6.15 billion in 2020. Net income fell 75.1 percent, from $705.9 million to $175.8 million, although that was partly driven by a $114.3 million charge for goodwill impairment and restructuring costs.

That brings us to Ulta’s non-GAAP financial metrics.

As part of its earnings release, Ulta reported 11 adjustments to that $175.8 million net income number. When all was tallied up, the firm’s adjusted net income was $264 million.

Most impressive, however, was Ulta’s step-by-step reconciliation between those two numbers. As you can see in Figure 1, below, the company walks analysts through the math of what adjustments add to net income and which ones subtract from it, and then presents the final non-GAAP number.

All firms must reconcile their non-GAAP metrics back to GAAP in any earnings releases that include the non-GAAP numbers. It’s just a welcome sight to see the arithmetic laid out so plainly.

We now have a critical mass of fourth-quarter 2020 filings for Calcbench to do another of our quarterly “quick analysis” guides, where we do year-over-year comparisons across a range of financial disclosures, to see how Corporate America has been doing lately.

You can download and read the full analysis if you like. Meanwhile, the headline for Q4 2020, compared to Q4 2019: net income down, revenue down, capital and operating expenses down — but cash is way, way up.

Our quarterly quick analysis is a straightforward exercise. We aggregated the disclosures of more than 2,100 firms, excluding financial firms (whose unusual balance sheet structure distorts the overall picture) and most retailers (whose quarterly periods don’t align with the calendar year). Then we do a year-over-year comparison across 12 major financial disclosures that firms make, to see how this quarter compared to the year-ago period.

Figure 1, below, tells the tale for Q4.  

For anyone who can’t quite make out the numbers, the major ones are:

• Revenue: down 4.25 percent
• Net income: down 16.7 percent
• Capex: down 13.9 percent
• Operating cash flow: down 7.1 percent
• Assets: up 5.4 percent
• Liabilities: up 8.5 percent
• Cash: up 38.4 percent
• Debt: up 9.4 percent

Those numbers shouldn’t surprise anyone. The economy was roaring along in those halcyon days of Q4 2019, so even as economic fortunes recovered over the course of 2020 (just look at our restaurant and airline industry posts from a few days ago), comparing any quarter of 2020 to 2019 was going to look grim. Revenue and net income dropped, lots of firms cut expenses, and lots of firms stocked up on debt so they’d have a sufficient pile of cash to carry them into more rosy times in 2021.

Let’s Get Granular

We also did a deeper dive into the 20 largest industry groups according to SIC number. Those 20 industry slices contained roughly 860 firms, 40 percent of our total sample population. Then we did a year-over-year analysis for each financial disclosure.

For example, Figure 2, below, shows the industry breakdown for revenue. As you can see, the two sectors that took the biggest hits were hotels (down 51.5 percent) and oil & gas (down 26.6 percent).

Again, that makes sense, since most of humanity parked the car in the garage and took a staycation for Q4, while a second wave of COVID-19 washed over the Western world.

Figure 3, below, shows the change in net income by industry. Note the jump in computer processing (up roughly 200 percent) and smaller but appreciable jumps in other computer sectors. Such is life when everyone is on Zoom calls, watching Netflix, while placing Amazon orders.

We’ll be back in several more months with our first quarterly quick analysis for 2021, comparing this year’s early numbers to Q1 2020. That should be quite a show. Stay tuned!

Today we have an interesting use case from Indiana University in Bloomington. Our guest interviewee, Terry Campbell, is a Clinical Professor at the Kelley School of Business.

How did Indiana University come in contact with Calcbench?

The first time I came in contact with Calcbench was when I was shown a Strategic Finance example. Then I saw Calcbench displayed at an American Accounting Association meeting and scheduled a demo. The rest is history.

What Calcbench tools do you use?

I like the comparative disclosures, the ability to run side-by-side comparisons,  previous period to previous period. It amazes me and my students that all the manual labor that we used to perform, doesn’t have to be done. 

You do a lot with experiential learning. It’s a big focus in academia now. Tell us Indiana’s program.

At the Kelley School of Business, we have an accelerated MBA program for accountants whereby students work with a company for several weeks during the second semester of their first year. The students perform comparative analysis, valuation, strategic analysis, whatever the company requests.  We dedicate several hundred hours of student time to these projects. The hallmark of the program is that students are actually working for the client during this time. 

While we have been doing this for over twenty years, over the past few years, we’ve been able to intensify the analysis and the comparative work. Calcbench has been an integral part of this. 

How does Calcbench factor into experiential learning?

We’re currently working with a startup in Vietnam in the cinnamon business. The company’s goal is to get into the food chain around the world. To get their ingredient distributed globally they need to connect into very large food manufacturing trade shows and very large multinational brokerages. We try to understand how a tiny minutia business, such as cinnamon,  might be reported. We use tools such as Google Scholar,  Statista to find the world’s largest consumers of cinnamon. Turns out Starbucks in one of them. We use Calcbench to understand the financial data of the industry sector or the companies we’re trying to target. In the case of this cinnamon company, we were able to use Calcbench to find a publicly listed brokerage out of Chicago.

What other courses have you seen Calcbench integrated into?

We demonstrate Calcbench, PowerBI, Tableau and Python in a career success skills course. It’s the beginning of the master’s program just to jumpstart everyone and let them know what they will run into and how to be prepared. The two features that we have almost 100% hit rate on are Calcbench and PowerBI. PowerBI to build planning models and Calcbench to get the data off in a comparative form.

How would you like to use Calcbench in the future?

IFRS is a great addition to Calcbench, but we often need data for European and Asian companies. For that we use Orbis to get revenue and profitability by country. In the luxury good space, where I have a lot of connections, there are a lot of private companies. We use Orbis for that as well and it would be great if the two platforms talked or connected with each other. 

-30-

We’re going to continue our one-year anniversary look at the COVID-19 pandemic today, this time looking at the airline industry.

In April 2020 we noted that United Airlines ($UAL) booked a $50 million impairment charge on the value of its flight routes from the United States to China. In May 2020, we examined the first federal bailout of U.S. airlines.

So today we go back to see what happened to six major airlines:

• Alaska Air Group ($ALK)
• American Airlines ($AAL)
• Delta ($DAL)
• JetBlue ($JBLU)
• Southwest ($LUV)
• United Airlines ($UAL)

To state the obvious, 2020 was not a good year for any of these firms. Year-over-year revenues tumbled by roughly 60 percent; net income turned into note losses for all six. See Table 1, below.

Comparing Q3 and Q4 of 2020, however, we see a bit of lift. Revenues increased by a range of 12 percent for Southwest to 37 percent for United. With the exception of Alaska Airlines, all others saw a decrease in the net loss quarter over quarter.

On the other hand, the pandemic’s effect may be long-lasting. All six airlines took on significant amounts of debt. See Table 2, below.

Moreover, when we examine the footnotes for more information about those debt issuances, it’s clear that some of the debt is quite expensive. For example, Delta Airlines issued $1.3 billion in debt last June with an interest rate of 7.375 percent. JetBlue issued $115 million in debt last August at 8 percent.

At least some of that debt seems to have been salted away as cash, perhaps to be prepared for future challenges. See Table 3, below.

Check out the Excel spreadsheet that was used for this analysis and see what other insights you can find!

Ten months ago, Calcbench introduced Restaurant Week — a series of posts exploring how firms in the restaurant business were faring, given the brutal disruption of the pandemic.

Now that most of the world has weathered one full year of COVID-19, we decided to revisit the restaurant industry again and see how those firms are doing.

To start, we looked up all firms identified as “Retail Eating Places,” otherwise known as the SIC 5812 filing category. We found 113 firms, and 30 of them have already filed their 2020 annual reports. Then we compared both yearly results (2019 against 2020) and recent quarterly numbers (Q3 2020 against Q4 2020).

At the annual level, the results are not good. Most firms saw both revenue and net income for 2020 well below 2019 levels. For example, Aramark ($ARMK) saw annual revenue drop 20.9 percent, while net income went from $448.5 million in 2019 to a $461.4 loss for 2020. McDonald’s ($MCD) saw its revenue drop by 8.9 percent, net income by 21.5 percent.

Not all the news is grim, however. Many restaurant firms seem to have found their footing in the second half of 2020 as we all grew accustomed to new eating habits under the pandemic.

For example, Ruth’s Hospitality Group ($RUTH) saw its annual revenue plunge 40.7 percent compared to 2019 — but revenues in Q4 2020 were 22 percent higher than those in Q3. Net income also went from a $5.3 million loss in Q3 to a $1.4 million profit. Yes, that’s still only a fraction of the $14.5 million net income Ruth’s enjoyed in fourth-quarter 2019, but small profit is better than no profit.

Likewise, Jack in the Box ($JACK) net income fell by 4.95 percent from 2019 to 2020; but it rose 34.4 percent from Q3 2020 to Q4 2020.

And not all firms had a bad 2020. Apparently we ate many pizzas during lockdown; annual revenue for Papa Johns ($PZZA) actually increased in 2020 by 12 percent, and net income soared from $4.8 million in 2019 to $57.9 million last year. Chipotle Mexican Grille ($CMG) also saw annual revenue increase by 7.13 percent, net income by 1.6 percent.

Attached is the spreadsheet we used to conduct the analysis. Download and use it to find your own slice of analysis.

Great news for everyone addicted to auditing, accounting, and data analytics — our webinar on exactly that subject is now available for your viewing pleasure on YouTube!

We had an excellent discussion, exploring everything from the practical steps necessary to use data analytics in a project; to the skills necessary to put analytics to best use in the corporate world; to several examples from our speakers of how they use analytics in their work today.

Our speakers, by the way, were two leading thinkers in the field:

• Brian Wolohan, partner in charge of the audit innovation team at Grant Thornton; and a member of the data analytics task force at the Public Company Accounting Oversight Board.
• Vern Richardson, professor of accounting at the University of Arkansas, who has been working in data analytics for years and literally wrote the book the subject — aptly named Data Analytics for Accounting.

Anyway, you can watch the full webinar (roughly one hour long) on the Calcbench YouTube channel, or just press play for the embedded video below.

As always, if you have a suggestion for our next webinar or other analytics projects we should undertake, drop us a line at info@calcbench.com any time.

For the past few years, we have published a list of firms and their sales to China.  With 10-K season winding down, we thought that we would publish the list again.  This year we are doing it slightly differently.  We are publishing the entire file.  Please note that it does have embedded Calcbench formulas in it, so  for it to work, you will need a Calcbench subscription.  

Here’s a picture of the top firms that sell to China as a percentage of total revenue.  

Cybersecurity firm SolarWinds Corp. ($SWI) filed its Form 10-K this week, giving analysts their latest glimpse into how much financial harm SolarWinds might suffer from its role in that massive cyber attack Russia launched on the U.S. government last year.

As you might recall, the feds discovered late last year that Russian operatives placed malware into the Orion Software Platform that SolarWinds sells to corporate and government customers. That maneuver allowed Russia to penetrate the cybersecurity defenses of a vast swath of the U.S. government and Corporate America; the president of Microsoft ($MSFT) called it “the largest and most sophisticated attack the world has ever seen.”

Calcbench previously looked at what SolarWinds had to say about the event in a blog post from December, when the business announced the attack on Dec. 14 in a Form 8-K filing. Now we have SolarWinds’ first quarterly report since everything went sideways, so let’s review the latest.

The company offered 590 words of disclosure related to the cyber attack, right at the top of the Commitments & Contingencies section in the 10-K. Most notable: in the 17 days from Dec. 14 (when SolarWinds first announced the attack) through Dec. 17 (the end of the filing period), SolarWinds racked up $3.5 million in expenses related to the attack.

For the record, SolarWinds also reported $1.02 billion in revenue for 2020 — so the attack expenses were only 0.34 percent of that number. Not a material amount per se, but the $3.5 million was incurred in only 17 days. Presumably the ongoing costs will be much higher as 2021 unfolds.

As SolarWinds proceeds to note, “As a result of the cyber Incident, we are subject to numerous lawsuits and investigations … In addition, there are underway numerous investigations and inquiries by domestic and foreign law enforcement and other governmental authorities related to the cyber Incident, including from the Department of Justice, the Securities and Exchange Commission, and various state Attorneys General.”

Translation: a whole lotta lawyers are coming after SolarWinds. Whatever the final costs of that litigation might be — and SolarWinds does not provide any estimate of that number — the process will not be cheap.

What Happens Next

SolarWinds did disclose that it maintains a $15 million insurance policy for cybersecurity attacks. Obviously that amount will help, but it’s unlikely to cover the full cost of things. (Nevermind the potential loss of business from customers too spooked to keep doing business with the firm.)

In that case, SolarWinds will likely need to set aside cash over time, placing it in a contingency account as the true cost of the cyber attack becomes more clear.

The good news is that the company does seem to have that money. For example, gross profit was $744.6 million in 2020; and operating profit after sales and marketing, R&D, and other routine expenses was still $107 million. See Figure 1, below.

On the balance sheet, we see that SolarWinds had $370.5 million in cash at the end of 2020, roughly 6.5 percent of $5.71 billion in total assets.

On the other hand, SolarWinds was also carrying $4.25 billion in goodwill. So if the firm suffers a loss of reputation in the market, which then translates into lower revenue or share price over the long-term — could that lead to a goodwill impairment? Which could, then, slash into stockholder equity since so much of the company’s total assets are tied up specifically in goodwill?

That’s only speculation on our part. We do know, however, that any hit to customer loyalty or revenue isn’t yet reflected in SolarWinds’ financial statements, because the attack happend so late into the company’s fiscal year. And while the company has seen a plunge in share price since December, when you look at the prior 12 months altogether (Figure 2, below), the plunge isn’t so bad.

What’s next? Check back with us in three months, or set SolarWinds as one of the firms you track with our email alerting function. Whenever there’s new data to study, rest assured that Calcbench has it.

Some firms disclose only a little about their goodwill testing and impairments; other firms disclose a lot.

And then, once in a blue moon, you get a disclosure like what Kraft Heinz Co. ($KHC) included in its annual filing on Feb. 17 — an opus nearly 4,000 words long, and a fascinating glimpse into how shifting corporate strategies translate into shifting numbers on the balance sheet.

First, let’s look at the headline numbers Kraft reported. As you can see in Figure 1, below, Kraft started 2020 with $35.55 billion in goodwill and ended it with $33.09 billion. That’s a drop of $2.46 billion, or 6.9 percent.

Those numbers, however, barely begin to tell what Kraft was doing last year. As always, you need to read the details in the footnotes!

First, Kraft reorganized both its internal operations and its reportable operations in early 2020. For example, it moved its Puerto Rico operations from its Latin America division to its U.S. division, and consolidated all its European operations into one, new International division. Ultimately Kraft landed on three reportable segments divided by geography: the United States, Canada, and International.

That’s important because a change so significant prompted Kraft to do two impairment tests of its goodwill assets — one immediately before the reorganization, the other immediately after. (Like, the two tests were performed on the same day, since the “reorg” was really just management moving items around from one P&L statement to another.)

The impairment test done after the reorganization resulted in two impairment charges totaling $226 million: one charge of $83 million for the Australia, New Zealand, and Japan unit; and another for $143 million in the Latin America unit. (Both units now included in one new International segment for reporting purposes.)

Impairment Tests, Part II

After those reorganization-related impairment tests at the beginning of 2020 came Kraft’s standard goodwill impairment test, which the company always conducts on the first day of its second quarter — March 29, 2020.

That test resulted in another impairment of $1.8 billion across four internal operating divisions:

• $815 million charge in Kraft’s Canada Retail unit
• $205 million charge in its Canada Foodservice unit
• $655 million charge in its U.S. Foodservice unit
• $142 million charge in its EMEA East reporting unit

Add those numbers together, plus the $226 million impairment charge from the reorganization test, and you get $2.043 billion — the amount reflected in Figure 1, above.

What drove those impairments, you ask? The company had this to say:

These impairments were primarily due to the completion of our enterprise strategy and five-year operating plan in the second quarter of 2020. Management, in completing the five-year operating plan, developed updated expectations regarding revenue growth and profitability opportunities associated with our reporting units and, as a result, has recalibrated our future investments to align with the opportunities for which we see greater potential for a return on those investments.

Translation: Kraft is tempering its expectations for future growth, as consumer tastes drift away from mass-produced food in a box toward fresh produce and more organic food. This isn’t news per se; recall that several years ago Kraft took an impairment charge of $15.4 billion in 2019. That impairment was an earlier, much larger example of shifting consumer trends hammering Kraft’s goodwill valuation.

But Wait, There’s More

Kraft then reorganized the operations within its U.S. reporting structure — “to align to the management of our new platforms, which were established to support the execution of our new enterprise strategy and five-year operating plan” — which meant the company had to perform a third impairment test, this time in June 2020.

That test didn’t result in any new impairment charges. But also that quarter, Kraft announced “the Cheese Transaction,” where the company agreed to sell its global cheese business to Groupe Lactalis for $3.3 billion. The Cheese Transaction was still pending at the time Kraft filed its 10-K (the deal should close sometime this spring), so to account for everything properly, Kraft had to shift $580 million from goodwill to assets held for sale. That number is also reflected in Figure 1, above.

We’re going to stop here because we have lives to lead, but analysts could delve into Kraft’s goodwill and impairment details all day long. Heck, right after the goodwill impairment analysis, Kraft launches into an equally long discussion of impairments to its intangible assets.

And don’t forget, you can also skim our Calcbench Research Guide for Goodwill & Intangible Assets. As Kraft demonstrates, there’s a lot of stuff to consider here.

“Human Capital” is a new disclosure item required by the Securities and Exchange Commission as of Nov. 9, 2020. All filers are supposed to include details about their workforce such as:

• Total number of employees, and employees per business segment when those segments are material to the business;
• Any “human capital metrics or objectives,” such as employee turnover, tenure on the job, or percentage of employees in unions; and
• Any information the business deems material about diversity and inclusion, either among employees as a whole or management specifically.

In our recent blog, Human Capital Disclosures, we share some recent examples of what’s included in Human Capital disclosures, including Sirius XM $SIRI and homebuilder MDC Holdings $MDC. 

Given that these Human Capital disclosures have a lot of room to share information about their people, we thought we would do a more comprehensive look and put Human Capital disclosures from the S&P 500 into a visual for you. 

Enjoy!

We want to follow up today on our previous post about critical audit matters (CAMs), to take a closer look at one particular set of CAMs: uncertain tax items. 

You might have noticed that tax-related items accounted for a significant portion of the CAMs we’ve seen so far in 2020 annual reports. We found 13 CAMs related to uncertain tax positions, plus another three related to unrecognized tax benefits. Taken together, that’s nearly 20 percent of the 85 CAMs we identified in total.

Well, exactly what are those tax positions? What’s the nature of the uncertainty, and what makes these disclosures qualify as critical audit matters?

Let’s first remember what a critical audit matter is. As dictated by accounting regulators, all CAMs have two parts:

• They relate to items or disclosures that are material to the financial statements; and
• They involve “especially challenging, subjective, or complex auditor judgment.”

In that case, you can see how various corporate tax issues might qualify as critical audit matters. Plenty of tax disclosures can be quite large and therefore material to the financial statements. And given the complexity of modern tax law in the United States and around the world, uncertain tax disclosures will almost always meet the second criteria, too: especially challenging, subjective judgment on the part of the auditor.

One example of this is Pepsico ($PEP), which reported $1.6 billion in reserves the company is salting away for unrecognized future tax benefits. That is, Pepsico might get that $1.6 billion sometime in the future, if certain disputes with tax regulators go the company’s way — but if not, Pepsico will have the cash to cover taxes due.

Pepsico’s auditor, KPMG, still flagged the issue as a CAM. In its auditor’s report (where audit firms disclose CAMs), KPMG had this to say:

The Company establishes reserves if it believes that certain positions taken in its tax returns are subject to challenge and the Company likely will not succeed, even though the Company believes the tax return position is supportable under the tax law. The Company adjusts these reserves, as well as the related interest, in light of new information, such as the progress of a tax examination, new tax law, relevant court rulings or tax authority settlements.

We identified the evaluation of the Company’s unrecognized tax benefits as a critical audit matter because the application of tax law and interpretation of a tax authority’s settlement history is complex and involves subjective judgment. Such judgments impact both the timing and amount of the reserves that are recognized, including judgments about re-measuring liabilities for positions taken in prior years’ tax returns in light of new information.

What does Pepsico itself have to say about unrecognized tax benefits? You can find that using the Interactive Disclosures tool and pulling up the firm’s tax disclosures. For example, in the 10-K Pepsico filed on Feb. 11, the company mentioned a $364 million gain in 2018 from a tax dispute with Russia that was resolved in Pepsico’s favor. The company also disclosed this table, below, showing how its tax reserves changed over the course of the year.

If you’re feeling ambitious, you can also use our Multi-Company Page and search for “unrecognized tax benefits” in the Standardized Metrics search field on the left side of the page. Then you could, say, identify all firms in the S&P 500 where unrecognized tax benefits were a material amount of money; and next search the auditor reports for those firms to see if any have unrecognized tax benefits as a CAM.

Uncertain Tax Positions

A close cousin of unrecognized tax benefits are uncertain tax positions. They’re conceptually similar — firms reporting a tax item as a potential payment or benefit — but uncertain tax positions encompass a wider range of tax items, including potential losses that might not materialize.

Amazon.com ($AMZN) is a good example of what we mean here. The company reported this table of tax contingencies in its annual report from Feb. 3:

Even for Amazon, $2.8 billion in uncertain tax positions is a material amount of money. And sure enough, we see that the company’s auditor, Ernst & Young, flagged this as a CAM. 

The Company is subject to income taxes in the U.S. and numerous foreign jurisdictions and, as discussed in Note 9 of the consolidated financial statements, during the ordinary course of business, there are many tax positions for which the ultimate tax determination is uncertain. As a result, significant judgment is required in evaluating the Company’s tax positions and determining its provision for income taxes. The Company uses significant judgment in (1) determining whether a tax position’s technical merits are more likely than not to be sustained and (2) measuring the amount of tax benefit that qualifies for recognition. As of December 31, 2020, the Company accrued liabilities of $2.8 billion for various tax contingencies.

Auditing the measurement of the Company’s tax contingencies was challenging because the evaluation of whether a tax position is more likely than not to be sustained and the measurement of the benefit of various tax positions can be complex, involves significant judgment, and is based on interpretations of tax laws and legal rulings.

Companies can be a bit more liberal in how they tag uncertain tax positions, so you might need to search the XBRL tag field on the Multi-Company page for this disclosure. Try “LiabilityForUncertainTaxPositionsCurrent” or “LiabilityForUncertainTaxPositionsNonCurrent,” and that should get you the results you’re looking for.

In 2019 we first wrote about how to access Critical Audit Matters (CAMs) from corporate financial statements using Calcbench.

Today, we will re-examine the information from the latest set of filings that have come in. Please note that we are still early in filing season, so don't draw too many conclusions from this just yet!

Some summary-level data first:

1. We examined 86 S&P 500 firms 10-Ks.
2. Identified a total of 124 CAMS in our sample (for example, Amazon reports one CAM while Becton Dickinson reports three, so that's a total of four CAMs).
3. Eight-five firms reported at least one CAM.
4. Seven firms reported three CAMs.

We compiled a summary of the most commonly reported CAMs in the table below.

Please note that this is our initial attempt to organize CAMs by topic. There is much more detail behind this summary. In addition, here is a link to the file that contains each firm we reviewed, its primary auditor, and the list of CAMs reported in its 10-K.

Lastly, in the list, we found 2 cases that had personal interest. These were from Disney (ticker:DIS CashTag:$DIS) and Netflix (ticker:NFLX CashTag:$NFLX). Both had to do with the Amortization of Content related costs. In Disney's case, this was a new CAM, whereas Netflix had also presented this CAM in last years 10-K.

Now that the 10-Ks for 2020 are rolling in, you may be noticing a new disclosure item in those reports: “human capital” factors.

What are those, exactly? They are new disclosures required by the Securities and Exchange Commission as of Nov. 9, 2020. All filers are supposed to include details about their workforce such as:

• Total number of employees, and employees per business segment when those segments are material to the business;
• Any “human capital metrics or objectives,” such as employee turnover, tenure on the job, or percentage of employees in unions; and
• Any information the business deems material about diversity and inclusion, either among employees as a whole or management specifically.

Human capital disclosures will typically be reported in the Business Description part of the 10-K. For now those disclosures will also vary quite a bit, because the SEC didn’t include a definition of “human capital” or any specific metrics that must be reported. Each firm can decide for itself what counts as a material item related to human capital that should be disclosed. (Compared to financial disclosures, for example, where firms have extensive and precise guidance on what goes into the filing.)

Anyway — yes, Calcbench can help you find these human capital disclosures! Just use our Interactive Disclosure tool, selecting “Business Description” from the pull-down menu on the left side. Skim down through the text, and you should see the material under a heading like “Human Capital.”

Firms only started making human capital disclosures in 10-Ks filed after Nov. 9, 2020; which means we don’t have many examples to study yet. But now that year-end filers are starting to submit their reports, Calcbench traipsed through a few examples. Here’s what we found.

MDC Holdings

Home-builder MDC Holdings ($MDC) is interesting because it provided a breakdown of employees by business unit in table format, in the 10-K it filed on Feb. 2. See Figure 1, below.

The rest of the disclosures are all narrative, which is fine although nothing terribly interesting. MDC only jumped out at us because on that same day, its rival Pultegroup ($PHM) also filed its 10-K, and didn’t use table format. Instead, Pulte described the same human capital information in written format:

On December 31, 2020, we employed 5,249 people, of which 945 were employed in our Financial Services operations. Of our homebuilding employees, 282 are involved in land acquisition and development functions, 1,752 are involved in construction and and post-closing customer care functions; 1,189 are involved in the sales function; and 1,081 are involved in procurement, corporate, and other functions. Our employees are not represented by any union. Contracted work, however, may be performed by union contractors. We consider our employee relations to be good.

So similar firms can take different approaches to the same disclosures. You’ll have to look carefully.

A More Extensive Example

If you want to see what maximal disclosure might look like, bounce over to the 10-K that Sirius XM ($SIRI) also filed on Feb. 2. That disclosure — nearly 1,100 words! — ranged from number of employees, to philosophy on corporate culture, to diversity and inclusion programs, to support for local COVID-19 relief measures.

Sirius didn’t include too many numbers, but did give glimpses into its human capital programs. For example, the firm included this about its corporate culture:

We are focused on creating a corporate culture of integrity and respect, with the goal of working together to drive our business to be creative, innovative and competitive. To achieve these objectives, we have adopted and regularly communicate to our employees the following core values, which we call “AMPLIFY”:

• Applaud and encourage new thinking
• Move forward and be purposeful in our desire to win
• Prioritize honesty, integrity and respectful communication
• Lean on each other and learn from one another
• Invest in our actions and commit to the follow through
• Find ways to give back by focusing on community and feeding your individuality
• You Matter. We embrace our differences, empower each other and include everyone

We operate a performance-based environment where results matter and financial discipline is enforced. We have tried to create a highly collaborative culture in which employees feel a sense of pride that their input is sought after and valued. At the same time, we believe in holding individuals accountable and have tried to create a culture in which employees “do what they say they are going to do.” Still, we believe that our culture is a long-term competitive advantage for us, fuels our ability to execute and is a critical underpinning of our employee talent strategy.

Those are only a few of the disclosures that are pouring into the database these days. If researching that data is your bag — yep, Calcbench has it.

Today Calcbench has a guest post from Olga Usvyatsky, doctoral student in accounting at Boston College and a long-time whiz at financial disclosures.

Last summer we had a post exploring how companies adjust their EBITDA metrics to account for COVID-19 costs. We identified a sample of more than 40 companies that chose to present pandemic-related costs as a separate line in the reconciling non-GAAP tables, and noted that in many cases, the adjustments were related to direct pandemic-related expenses such as protective equipment.

That was then. What’s been happening lately? Lots.

In December 2020, the SEC’s Division of Corporate Finance clarified that it wouldn’t object to the presentation of COVID-19 adjustments, so long as those costs were directly attributable to the pandemic and certain other conditions were met. Notably, SEC staff warned that adjusting for lost revenue would not be appropriate.

Meanwhile, companies continued to present non-GAAP COVID-19 expenses in third-quarter 2020, but the magnitude of those adjustments appeared to decline. For example, T-Mobile US ($TMUS), a wireless network operator that we looked at in our June blog, disclosed $117 million in Covid-19 costs in first-quarter 2020, and another $341 million for the second quarter. By Q3, however, T-Mobile said that “employee payroll, third-party commissions, and cleaning-related COVID-19 costs were not significant for Q3 2020.”

Non-GAAP metrics need to be presented consistently across periods (see Question 100.02 in the SEC’s Compliance & Disclosure Interpretations). Yet, changes in economic conditions or certain company-specific developments may prompt companies to modify the presentation and introduce new metrics or adjustments. So, what were the non-GAAP modifications introduced in 2020?

The first adjustment we looked at was related to the U.S. Treasury Department’s extended Payroll Support Program (PSP) established in early 2020 to provide payroll support to airline carriers. The relief provided by the program was material to many (if not all) of the airlines, whose business was severely disrupted by the pandemic. Notably, although most non-GAAP adjustments improve net income, PSP adjustments cause non-GAAP income to be lower than the number reported under GAAP. (Most times, non-GAAP income is higher than GAAP income, a trend Calcbench noted in a post last week.)

To illustrate, let's look at the non-GAAP section of Alaska Air Group ($ALK). In third-quarter 2020, Alaska Air recorded a "Payroll Support Program wage offset" adjustment that decreased non-GAAP income by $398 million (and by an aggregate $760 million through the first nine months of 2020). Based on Alaska’s latest quarterly report, the company “expects to record an additional $10 million in wage offset in the fourth quarter.”

To better understand how other airlines reported PSP adjustments in the non-GAAP section, we looked at American Airlines Group’s ($AAL) disclosure. The company didn’t provide the PSP relief as a separate reconciling line. Still, American did clarify in a footnote that in Q3 of 2020, $1.9 billion and $228 million of PSP assistance were included in mainline and regional operating special items non-GAAP adjustments.

We can’t provide an opinion about whether any of the metrics are compliant with Regulation G (the SEC rule governing non-GAAP reporting), and we are not implying that any of these disclosures are incorrect or non-compliant. Yet, as we approach the filing season, it’s worth remembering that Question 100.03 of the SEC’s CD&I states that if a company excludes charges in the calculation of a non-GAAP measurement, it should also exclude gains.

Non-GAAP and Taxes

The second type of adjustment we looked at was related to changes in tax legislation in 2020. One such change in legislation, the U.K. Finance Act of 2020, was signed into law effective April 2020. The law abandons plans to reduce the corporate tax rate from 19 percent to 17 percent — which results in an increased tax liability estimate for some companies (because their expected tax cut didn’t arrive).

For example, in third-quarter 2020, Hasbro ($HAS) modified its GAAP income by a discrete "tax reform" adjustment of more than $13 million and stated that the adjustment was caused by “revaluation of Hasbro's U.K. tax attributes in accordance with the Finance Act of 2020 enacted by the United Kingdom on July 22, 2020.”

In another example, L Brands (LB) adjusted its Q3 net income by $23 million and disclosed that the adjustment was related to “foreign investments and recent changes in tax legislation.” Notably, in Q2 of 2020, the company also recorded a discrete tax benefit of $21 million related to “changes in tax legislation included in the CARES Act.”

Some of the adjustments introduced in 2020 appear to be company-specific. A recent blog by Calcbench discussed an unusual "civil disruption cost" adjustment recorded by a grocery chain Albertson ($ACI). Calcbench was unable to identify any other retailer that adjusted GAAP results for a similar item.

Finally, the pandemic put economic pressure on whole sectors such as retailers, forcing companies to restructure operations and re-evaluate their expansion plans. Unsurprisingly, these restructuring costs may show up as a “new” adjustment in the non-GAAP section.