- Spotify is hammered for failing to report profits, but that’s irrelevant.
- Major expense items in its P&L, like R&D and certain finance costs, aren’t really operating expenses.
- What really matters is the business model, and Spotify’s is positive for customer growth, churn, and gross margin. (Less so for customer duration and ARPU.).
The media is abuzz about Spotify (SPOT) being deep in the red: “If Spotify is so huge, why is it losing money?” ponders ABC News; “Spotify intended to lose more money.” states the Motley Fool, and Fortune buries the music streaming company for good: “Why Spotify will never make money.” (May 3, 2018). Spotify seems to have reached its end.
And yet, despite this gloom, Spotify has a capitalization of almost $25 billion, and its current (December 10, 2018) share price around $130, isn’t far from its initial listing price of $132. (Spotify’s NYSE listing in early April 2018 was not a traditional IPO, raising money from investors. Instead, the company just listed its shares for trade.) And no fewer than 459 institutions hold 52% of the company’s outstanding shares. Haven’t these investors heard that Spotify “will never make money”? What’s going on here?To be sure, a superficial reading of Spotify’s recent income statements conveys a bleak picture: Net losses for the years, 2015-2017, were €230 million, €539 million, and €1,235 million, respectively. Thus, as monthly active users grow (91, 123, and 159 million, respectively, for 2015-2017), and paying subscribers multiply (28, 48, 71 million, respectively, for 2015-2017), Spotify’s losses mount. No wonder Fortune predicted that the company “will never make money.”
Yet, a closer examination of Spotify’s income statements reveals a different situation. Two, so called accounting “expenses” — R&D and finance costs — stand out. The 2017 R&D cost was €396 million. Readers of my recent book (The End of Accounting, Wiley 2016) and my research on intangibles are well aware of my position on R&D, which is widely shared by economists: R&D is not an expense. (R&D has been treated as an investment in the national income accounts for years.) An accounting expense is defined as a cash outlay without future benefits, like wages, rent, and insurance, which are payment for past services. R&D, in contrast, is conducted with the sole aim of generating future benefits from products and services. It is thus an investment and shouldn’t count against earnings. This is particularly true for early-stage enterprises that invest heavily in R&D to generate future growth, and in the process record accounting “losses.”
Spotify’s R&D is aimed “… to design products and features that create and enhance user experiences, and new technologies are at the core of many of these opportunities …. Expenses primarily comprise costs incurred for development of products related to our platform and service, as well as new advertising products and improvements to our mobile application…” (Form F-1, 2018). Do these activities sound to you as expenditures without future benefits?
If the expensing of Spotify’s R&D doesn’t make much economic sense, its finance expense will floor you: €855 million of finance costs for 2017. You think: interest paid to banks or bondholders. Far from it. This finance cost is a stunt that only accounting standard-setters could concoct. A footnote on page 74 of Spotify’s F-1, 2018 report states that the finance costs are: “… due primarily to the issuance of the Convertible Notes in April 2016, which are accounted for at fair value with any changes in fair value recorded in the statement of operations [income]. Due to the implicit interest rate and an increase in the value of ordinary shares, the expense recorded for Convertible Notes increased by €279 million… The expense recorded for outstanding warrants increased by €255 million due the increase in the value of ordinary shares…”
What’s going on here? An increase in Spotify’s share price raises financing costs by over €500 million? Yes, by the following accounting “logic”: As Spotify’s share price increases, the likelihood of convertible notes and warrants to be converted to shares increases and their value accordingly rises. But this is not an increase in Spotify’s liability (debt) to note holders. In case of no conversion, the liability, of course, remains the same. Furthermore, there isn’t any cash outflow associated with this “financing expense.” All it indicates is a possible dilution of shareholders’ value upon the conversion of notes. This is far from an operating expense, as it is presented on the income statement. Ironically, as Spotify was battered recently by analysts for being “profitless,” its share price decreased, and so were the financing costs (€76 million only for third quarter 2018), creating an accounting miracle: share price decreases, and voilà, earnings rise. A new value driver.
The upshot: Much of Spotify’s 2017 loss of €1,235 million isn’t really an operating loss, and the company isn’t that far from breaking even. This, indeed, is corroborated by Spotify’s 2017 positive cash flow from operations (€179 million), and positive free cash flow (€109 million). Moreover, for the third quarter of 2018 Spotify reported a small profit (€43 million).
Harassing Spotify’s managers to report positive earnings by the deeply flawed accounting measurement of this indicator is obviously unproductive, putting pressure on management to reduce R&D and other growth-driving investments. These accounting earnings cannot be used to evaluate the progress and viability of a business enterprise. Nokia reported for years tons of earnings while it lost its mojo to Apple. Amazon was for years a chronic “loser,” masking its huge strides toward dominance.
If not by earnings, how else to evaluate Spotify’s performance? Here, as elsewhere, you should examine carefully the company’s business model and its execution. Spotify’s monthly active users increased from 91 million at end-of-2015 to 159 million at end-of-2017, and further to 191 million on September 30, 2018. Paying users increased from 28 million in 2015 to 71 million in 2017, and to 87 million on September 30, 2018. A healthy customer growth, to be sure. Gross margin increased in the third quarter of 2018 to 25%, from 22% in the same quarter a year earlier. Another plus for Spotify. The key indicator of monthly churn (% subscribers leaving) decreased from 7.5% in 2015 to 5.5% in 2017. Good, but a 5.5% monthly churn rate means that subscribers stay with Spotify only 18 months, on average (1 over churn). Not a particularly long period. Another cloud on Spotify’s horizon is the constant decrease of ARPU (average revenue per user): from €7.06 in 2015 to €5.24 in 2017, and further to €4.73 in September 2018. ARPU decreases generally reflect competitive pressures. Spotify explains the decrease as: “… due principally to growth of the Family Plan and movement in foreign exchange rates.” More details would have been helpful.
Obviously, Spotify is not out of the woods yet, but not because of the absence of earnings. Key components of its business model―customer retention and ARPU, in particular―must be shored up.
Comparisons with Netflix, often made, are misguided. While Netflix’s marginal costs (the costs of servicing an additional subscriber) are very low, given that the investment in content was completed, Spotify’s marginal costs — payments to music right-holders for subscriber downloads — are high. So, unlike many other Internet-services companies, Spotify has to reduce royalties to record labels; a challenging task, given the dominant position of the three major label companies (Sony, Universal, Warner). The key to Spotify’s success is to grow and cash-in on alternative revenue sources, such as advertising and providing use of its subscribers’ “big data” to artists and labels. So, Spotify is still a work-in-process, but, I believe, it’s doing the right things.
Disclosure: I/we have no positions in any stocks mentioned, and no plans to initiate any positions within the next 72 hours.
I wrote this article myself, and it expresses my own opinions. I am not receiving compensation for it. I have no business relationship with any company whose stock is mentioned in this article.
This article first appeared here at Seekingalpha.com on Dec 11, 2018.
Edit: This article was not completely posted earlier on the blog, and subsequently have been updated to be complete.