Cutting Edge Tech at Absolutely No Cost to You

A closer look at the R&D spending of the security software companies

Once upon a time, the price-to-book ratio was a widely used tool in investing. An investor could take the accounting value of a company’s assets, subtract its liabilities, and be left with a number that was at least a decent starting place from which to determine what was left for the equity holders to split amongst themselves. Comparing this amount to the market value of the equity could be an informative first stop in determining if the stock was currently a bargain or not.

However, things change. As technology advanced, less of the typical company’s value was tied up in hard assets like assembly lines and delivery trucks. Price-to-book ratios were no longer informative in these new industries where a company’s valuation in the marketplace was based on imagining how a new software solution or hardware platform could disrupt an existing market-or perhaps even create an entirely new one of its own.

Given the key role that research and development plays in such a market, we are baffled by the fact that accounting conventions and current reporting practices allow companies to effectively erase the bulk of R&D costs from their earnings. We will explore the two main ways companies do this by taking a close look at the R&D spending of the major security software players including Okta (OKTA), Cloudflare (NET), Palo Alto Networks (PANW), CrowdStrike (CRWD), ZScaler (ZS), and Fortinet (FTNT).

---

Erasing R&D Expense Method #1- Pay the Developers in Stock and Ask the Market to Ignore the Cost

It’s no secret that most tech companies rely on stock-based compensation to pay their employees and there is nothing wrong with this. However, readers know our feelings on the matter of companies asking investors to accept the practice of adding back stock-based compensation to non-GAAP earnings as if the expense was imaginary. While issuing options is a non-cash transaction, buying back shares to limit dilution isn’t. Neither is the cash a company would have to pay its employees if it didn’t pay them with options. These are real expenses that shouldn’t be ignored.

What’s more, many of the options are used to compensate employees in the R&D department. The following table shows the percentage of R&D expense that each company pays in stock for the last seven quarters. As you look through the table, remember that the ability of these companies to utilize their stock depends on their employees’ belief that the stock is valuable and likely to go up. This has become less certain in recent months.

We see that OKTA pays the largest share of its R&D budget in stock. In addition, this percentage is increasing on a year-over-year and sequential basis. ZS is also heavily dependent on stock to pay for research as is NET, which also like OKTA, has seen this percentage trend upward in recent quarters. Adding back stock compensation pretends that more than one-third of these companies’ R&D budgets doesn’t exist.  

Let’s take a look a more detailed look at each company below.

• OKTA uses stock for the largest share of its R&D spending, closing in on half. This percentage is increasing.

• Not surprisingly, OKTA’s R&D paid in stock as a percentage of sales is also the highest of the group making it the largest beneficiary of adding back stock compensation to non-GAAP results.

• Both total R&D and cash-only R&D appear to have flattened out in the last three quarters.

• Adding 100bp as a percentage of sales to R&D would cost OKTA just under 3 cents in non-GAAP EPS per quarter.

• Despite stock compensation rising from 30% of R&D to 45% in just a few quarters, OKTA is still reporting negative non-GAAP EPS.

• Also note that R&D in total dollars fell sequentially last quarter, as the R&D percentage went up. It is likely there was a timing issue that saw the R&D rise faster than expected two quarters ago and the last quarter corrected that. However, this is worth watching, especially as OKTA guided to poor sales forecasts.

---

Zscaler (ZS)

• Almost 43% of ZS’s R&D budget is paid in stock- the second highest in the group.

• ZS was seeing its percentage of R&D expense paid in stock decline which was diluting the benefit of adding back stock compensation to non-GAAP results. However, this jumped again in the latest quarter as cash R&D declined while Stock R&D increased.

• he increase in stock compensation overall only accounted for 0.4 cents of non-GAAP EPS in the quarter but this may be more of a headwind going forward.

• ZS is posting positive non-GAAP earnings and guided to strong growth in the coming quarter. It still relies heavily on adding back stock compensation, but it has not varied the percentage very much.

---

Cloudflare (NET)

• The cash spending on R&D has stayed very close to 20% of total sales.

• Total R&D spending has grown as NET has added more stock compensation. R&D-related stock compensation expense has risen from 7.2% of sales to almost 12% of sales. The shift to stock compensation has effectively shielded non-GAAP profits from the increase in total R&D spending.

• NET is not even earning positive non-GAAP EPS. Like OKTA, it has recently seen the amount of R&D paid for with stock jump significantly. Even with that boost to non-GAAP EPS, they still cannot earn a profit.

---

Palo Alto Networks (PANW)

• R&D as a percentage of revenue has been declining. Strong guidance for revenue growth and even faster growth in deferred revenues should help PANW continue to leverage in this area.

• The 60bp drop in cash R&D in the 7/22 quarter helped EPS by 7 cents.

• Total dollars spent and cash-only R&D continue to rise in dollar terms.

• PANW could report positive earnings at this point even if it DID NOT add back stock compensation.

---

CrowdStrike (CRWD)

• CRWD pays for roughly 30% of its R&D with stock. However, this percentage has remained fairly constant over the last few quarters.

• It is likely a positive that the company continues to grow its R&D spending, but it could become a larger headwind.

---

Fortinet (FTNT)

• ·FTNT’s stock compensation as a percentage of its R&D budget is the lowest in the group, as is its total R&D spend as a percentage of sales. R&D spending in total dollars and in cash-only appears to have stalled while stock compensation used for R&D has increased slightly.

• If FTNT needed to grow R&D by 100bp as a percentage of sales, it would only cost quarterly EPS about 1 cent.

---

Erasing R&D Method #2- Acquire It and Ask the Market to Ignore the Cost of the Deal

In addition to developing intellectual property in-house, a company can acquire it via acquisition. Companies in research-intensive industries acquire other companies all the time and when they do, they are obtaining all the value of that company’s past research and development spending. In the case of more mature companies, the company is also receiving the value of what that company spent to build its brand name and establish customer relationships. A point that is often lost in the discussion is that the acquiring company would have had to spend money to develop that technology in-house if it hadn’t acquired it. However, accounting rules and current reporting practices allow companies to obscure this important point.

In practice, approximately half of the value of a tech acquisition is assigned to goodwill. In years past, goodwill was amortized over a period not to exceed 40 years. However, decades ago the FASB declared goodwill to have an infinite life by ending the practice of amortizing goodwill. (For a thorough discussion of goodwill accounting see our January piece on The Impact of Acquisitions on Earnings Quality- Part 1. )

The remainder of the value of the typical tech acquisition is assigned to intangible assets that are to be amortized over the estimated expected life of the asset. Analysts can examine the length of the amortization period used looking for signs of a company using an aggressively long period. However, in an investing world that now relies on non-GAAP results, the length of the estimated lives has become less relevant given that almost all tech and medical companies add the amortization of acquired intangibles back to their non-GAAP profits on the grounds they are non-cash. While the amortization expense may not technically be paid in cash, the acquisition leading to the intangible asset itself required cash and/or stock which much either be bought back or dilute the existing shareholders.

Also, note that many of the acquired companies have stock award obligations to current employees. During the acquisition, the acquirer frequently converts those stock awards into its own common stock for the workers coming onboard. That creates future dilution for existing shareholders unless the stock is repurchased which increases the distortion from adding back stock compensation to non-GAAP earnings as discussed in the first section above.

The following table shows an estimate of what goodwill amortization would be over 40 years and amortization expense relative to EPS for each company.

• If we look at just the sheer size of goodwill and intangibles – PANW and OKTA appear to be the companies benefiting most from acquisition accounting.

• OKTA posted a non-GAAP loss of 10 cents per share but gained 35 cents from ignoring expensing goodwill assets and adding back amortization.

• PANW saw 18 cents in non-GAAP items in these areas. At least it has non-GAAP EPS of $2.39 before looking at acquisitions, so this is 8% of the reported EPS.

• Don’t ignore ZS is getting over 2 cents of its reported 17 cents (14% of non-GAAP) EPS here.

• And NET reported 0 cents for its non-GAAP EPS but added 2.3 cents from adding back acquisition costs. If NET had to expense that 2.3 cents as R&D, it would be reporting about $8 million more in R&D per quarter.

In terms of settling stock obligations of the acquired companies, we will refer to the stock option discussions in this report and the first report from June. All of the new stock awards would not be expensed through R&D. Some will be charged to Sales and Marketing, General & Administrative, or Cost of Goods. The result should be rising stock compensation, which all these companies add back.  We can quantify what is coming from recent deals that are significant:

• OKTA has over $800 million of stock compensation expense coming to replace the awards held by Auth0 employees. That is more than a full year of stock compensation.

• PANW has $260 million of stock compensation going forward to replace awards held by multiple deals. That is roughly a 25% increase in annual stock compensation.

• NET has over $330 million of future stock compensation going to employees of acquired companies - $300 million to the Area 1 deal alone. This is more than double the annual stock compensation before the deal.

---

---

Contact behindthenumbers@btnresearch.com for questions regarding our institutional service.

---

Please share our column with any interested friends and colleagues…

---

You can follow us on Twitter here

---

Disclosure:

This article is intended for educational purposes and is not investment advice.

Behind the Numbers, LLC is an independent research firm structured to provide analytical research to the financial community. Behind the Numbers, LLC is not rendering investment advice based on investment portfolios and is not registered as an investment adviser in any jurisdiction.  All research is based on fundamental analysis using publicly available information including SEC filed documents, company presentations, annual reports, earnings call transcripts, as well as those of competitors, customers, and suppliers. Other information sources include mass market and industry news resources. These sources are believed to be reliable, but no representation is made that they are accurate or complete, or that errors, if discovered, will be corrected. Behind the Numbers, LLC does not use company sources beyond what they have publicly written or discussed in presentations or media interviews.  Behind the Numbers does not use or subscribe to expert networks.  All employees are aware of this policy and adhere to it.

The authors of this report have not audited the financial statements of the companies discussed and do not represent that they are serving as independent public accountants with respect to them. They have not audited the statements and therefore do not express an opinion on them. Other CPAs, unaffiliated with Mr. Middleswart, may or may not have audited the financial statements. The authors also have not conducted a thorough "review" of the financial statements as defined by standards established by the AICPA.

This report is not intended, and shall not constitute, and nothing contained herein shall be construed as, an offer to sell or a solicitation of an offer to buy any securities referred to in this report, or a "BUY" or "SELL" recommendation. Rather, this research is intended to identify issues that investors should be aware of for them to assess their own opinion of positive or negative potential.

Behind the Numbers, LLC, its employees, its affiliated entities, and the accounts managed by them may have a position in, and from time-to-time purchase or sell any of the securities mentioned in this report. Initial positions will not be taken by any of the aforementioned parties until after the report is distributed to clients, unless otherwise disclosed. It is possible that a position could be held by Behind the Numbers, LLC, its employees, its affiliated entities, and the accounts managed by them for stocks that are mentioned in an update, or a Peek Behind the Numbers article.