In the early 2000s, in one of my first attempts at angel investing, I was a partner in a British clone of an American DVD rental company, Netflix. We got off to a good start but needed more capital, which was harder to come by at the time. I persuaded Stelios Haji-Ioannou, the founder of low-cost airline easyJet, to inject additional capital, but the deal fell through. The company was eventually sold to LoveFilm before being in turn acquired by Amazon.
I should have bought Netflix stock. It would have been easier and, if I had held it, much more profitable. Today, I’m a Netflix customer and a big fan of Reed Hastings, the co-founder and president, but I’ve had some disagreements with the company — and one of its major backers — over its content compatibility. I originally looked at this two years ago and think it’s time to revisit now that content spending is on the rise again.
I think Netflix has become less conservative in its accounting for some time and its content amortization method could inflate its profits.
In this article, I take a look at how Netflix counts content. I just have to point out that at least until recently, no one cared too much about Netflix’s revenue. Thus, whether it adopts a more or less conservative policy could have a limited impact on its share price.
Netflix’s policy is to capitalize content costs and amortize them over their expected life. This amortization is accelerated because most viewing occurs when the content is new and the lifespan does not exceed 10 years, with 90% in the first four years.
For those like me who are pedantic on the subject, here is the account policy:
“For content produced, the company capitalizes costs associated with production, including development costs, direct costs and production overhead. Participations and balances are accounted for as expenses according to the amortization of production costs.
“Based on factors such as historical and estimated viewing patterns, the Company amortizes content assets (licensed and produced) in “Cost of Revenues” in the Consolidated Statements of Income over the shorter of the contractual window. availability of each title or the estimated period of use or ten years, from the month of first availability. Amortization is on an accelerated basis, as the company generally expects more initial viewings, and the Amortization of movies is more accelerated than amortization of TV series. On average, more than 90% of a produced content asset is expected to be amortized within four years of its month of first availability. The company continuously reviews factors affecting the amortization of content assets The Company’s estimates of these factors require considerable judgment by management.
“The Company’s business model is subscription-based, as opposed to a revenue-generating model at a specific title level. Content assets (licensed and produced) are primarily monetized as a group and are therefore reviewed globally at group level when an event or change in circumstances indicates a change in the expected usefulness of the content or the fair value may be less than the amortized cost To date, the Company has not identified any event or such change in circumstances. If such changes are identified in the future, these aggregated content assets will be stated at the lower of amortized cost or fair value. In addition, unamortized costs of assets that have been or should be abandoned are written off.
Conservative damping test
The first test I recommend when looking at capitalized intangible assets is to determine the average life by comparing the amortization expense to the average asset. My calculation adds the accumulated depreciation to the net figure shown in the balance sheet and then averages it out. Note that Netflix used to separate current content assets but no longer does so for technical reasons – with this more refined approach, the average lifespan was six years in 2019 compared to 6.8 years on our new baseline.
The trend for Netflix is illustrated in the graph below, which clearly shows Netflix continuing its trend of becoming less conservative in its content lifespan assumptions.
Netflix is open and helpful about its content count policy, even producing a video on the subject. In a useful way presentationPanorama of content accounting in January 2018 (since update ), it indicated that the above test should not be applied, and gave the following justification:
- The content library is presented net of depreciation, not on a gross basis
- Content is amortized on an accelerated basis
- Amortization over a given period is also affected by the mix of content, as different categories of content are amortized on different schedules (based on historical and projected viewing patterns)
In a previous BlogI explored these arguments:
- Since the contents are presented net, my calculation of the average life span adds back the accumulated depreciation to date. It is possible that some assets have been fully depreciated and should be eliminated from the calculation. Therefore, I may have included in my calculation the depreciation of an asset that was fully depreciated. This would extend the life of the chart, but wouldn’t be enough to create this trend, as Netflix has dramatically increased its spending on content – fully depreciated assets would almost certainly be a lower proportion of the total.
The graph shows that the average percentage delisted at Netflix increased from 60% to 66% during this period, a sign of some aging of the portfolio. Netflix’s spending has accelerated – in the past three years it has spent more than in the previous eight. The fact that the asset is aging indicates that the company is not super aggressive. But if 90% were amortized in four years, I would expect the profile of lives to be shorter and aging to be more advanced.
2) If the content is depreciated on an accelerated basis, there should be an obvious deviation from straight-line depreciation. Here is an excerpt from Netflix’s Accounting Policy in 10-K (2021, p44-5):
“Amortization is on an accelerated basis, as society generally expects more initial viewing, and amortization of movies is more accelerated than amortization of TV series. On average, more than 90% of a licensed content asset or product should be amortized within four years of its month of first availability”.
I tested this by amortizing STRAIGHT LINE content additions over four years to a residual value of 10%, then comparing it to Netflix’s depreciation expense. The results are displayed in the graph:
Obviously, I would expect the actual charge to Netflix’s financials to be considerably higher than my estimate, reflecting that content was being depreciated at an accelerating rate. I don’t understand why Netflix’s amortization charge isn’t higher.
3) The question of the mix is covered by the policy stipulating that on average 90% of the content must be amortized during the first four years.
My opinion is that Netflix’s content numbers are inconsistent with the company’s stated policies. This is rarely a positive point for investors. To evaluate the business, it would be better to use cash expenses as an indicator of the expense.