Oakmark's Bill Nygren Is Looking At Amazon's Valuation Incorrectly

| About: Amazon.com, Inc. (AMZN)
This article is now exclusive for PRO subscribers.


Value investor Bill Nygren recently invested in Amazon.

Assumptions about profitability takes a leap of faith.

Nygren's valuation method incorporating an adjusted price-to-sales ratio is fundamentally flawed.

Since August 2012, I have been tactically short Amazon (NASDAQ:AMZN) through outright shorting and the use of options strategies. Even though the share price has appreciated from my original short price of $243.69 to $355.90 at the close Friday, an increase of 46%, I have eked out a small profit as I wait for the tide to turn to reveal that Amazon is swimming naked. One of the signs that the tide may be about to go out came from the CNBC headline earlier this week which read "Top value investor Bill Nygren buys into Amazon." If value investors start buying Amazon, that certainly has to exhaust the list of potential buyers for the shares, and could signal there is nowhere to go but down. My first thought was is this guy really a top value investor; if so, why does he like it and is there something I'm missing?

For those of you not familiar with Bill Nygren, he and Kevin Grant have managed the value oriented Oakmark Fund since March 2000. During their tenure, their annualized return has been 10.3% per year compared to just 3.6% for the SPDR S&P 500 ETF (NYSEARCA:SPY), an impressive outperformance of 6.7% per year. That's the difference between turning $100,000 into $404,000 rather than only $166,000 in an S&P 500 index fund. Oakmark's long-term track record indicates they are the real deal.

So, what is their investment thesis and why do they find value in Amazon? From the article:

"...for the shareholders, it's much better that they scale up and become two to three times the size they are now over the next five years that it would be to raise prices a little bit today, quit the R&D spending and then get a normal margin on the business they are today but then only grow like an average retailer..."

Nygren has bought into the fact that there is some sort of R&D spending that is hurting profitability that is able to go away at some point without hurting the business. However, I have yet to see anyone detail out what this spending is and why it will go away. Two large expense items that have contributed to the lack of profitability over the last few years have been their spending on video content to support the Prime product, a number that is not disclosed, as well as the massive increase in subsidized shipping costs. Shipping costs in excess of revenues have increased to $3.5 billion in 2013 from only $630 million in 2008. Plain and simple, these items are not investments; they are expenses. Growth in revenues is only valuable if it is profitable. From 2008 to 2013, Amazon has increased revenues by $55 billion while income from operations has decreased by $100 million to $724 million, or less than 1% of revenues. If they could eventually fully eliminate the subsidy to customers for shipping without hurting their business, they could get back to a net profit margin of about 3.5%, which would be in the range of leading retailer Wal-Mart (NYSE:WMT).

Being a value-oriented investor, he attempts to justify the current valuation as follows:

"We think when you look at Amazon, instead of looking at what they're currently earning, it's useful to look at what the scale of the company is and how does it look on a price-to-sales basis. And if you adjust for the fact that a lot of their sales now are third-party…and you gross those sales up to a full gross sales rather than net sales, Amazon's actually selling at a pretty big discount to the average retailer."

Unfortunately, this line of reasoning is fundamentally flawed. In an article I wrote in September 2012, I showed that the price-to-sales ratio an investor should pay for a company is directly related to the ultimate profitability those sales can generate. How much profitability would Amazon gain if you gross up its revenue number to account for the portion of the 3rd party sales that isn't currently recognized as revenues? The answer is zero. Profit Margin = Profit / Revenues. In determining their profit margin, the addition of third party revenues with no profitability would have the effect of increasing the denominator only: (Profit + Zero) / (Revenues + third party revenues) = lower profit margin. So while the price-to-sales ratio is lower if you include the third party revenues, your profitability is equally lower. Amazon should trade at a price-to-sales ratio equal to the average retailer only if those sales were as profitable as the average retailer. They currently aren't able to generate average retailer profitability without including a lot more revenue that has no additional profitability. Therefore, if you use an adjusted price-to-sales ratio to account for third party revenues, the ratio at which it is valued should be a discount to the average retailer by an amount equal to the impact of including the third party revenues. An exercise in futility.

Let me give you a real life example of the difference between using direct revenues and including third party sales numbers to determine valuation on a price-to-sales basis. Visa (NYSE:V) charges a small fee on every transaction that utilizes its payment system. Visa has an enterprise value of $118 billion, $12.3 billion of revenues over the last twelve months and $8.125 billion of EBITDA for an EBITDA margin of 66%. It trades at an EV-to-sales ratio of 9.6x and an EV/EBITDA multiple of 14.5x. But what if instead of the revenues they actually generate from those transactions, you included the total amount of the transactions that were processed (which is equivalent to Amazon's third party revenues) to determine the price-to-sales ratio? The total value of transactions processed over the last twelve months was $7.05 trillion. Using this number, the EV-to-sales ratio would be a mere 0.0167x revenues, but since all of the additional revenues had zero profitability, EBITDA margin would fall from 66% of revenues to only 0.115% of revenues. Not so magically, the EV/EBITDA ratio would still remain 14.5x, even though the EV-to-sales ratio was 99.83% lower.


Nygren takes a leap of faith to assume that Amazon will eventually be able to earn normal margins on their revenue base, even though they have not had any profitability on the last $55 billion of incremental revenues. Normal profitability could happen if they eliminated the shipping subsidy they provide to customers without hurting their business. But that begs the question, if it wouldn't hurt their business then why don't they eliminate the subsidies now?

However, he is way off-base on the topic of valuation using price-to-sales while incorporating incremental third party revenues that earns them no additional profit. Comparing price-to-sales ratios between companies is only relevant to the extent that the profitability of those sales is equivalent. If Amazon were valued based on price-to-sales, adjusting for revenue growth rates, and assuming they will ultimately reach average margins, the share price would be significantly lower than where it trades today.

Disclosure: The author is short AMZN. The author wrote this article themselves, and it expresses their own opinions. The author is not receiving compensation for it (other than from Seeking Alpha). The author has no business relationship with any company whose stock is mentioned in this article.