When I started writing about Amazon.com (NASDAQ:AMZN), I had a divergent view from the market. It was late October 2011, and the consensus was that the earnings weakness was temporary due to an "investment phase". My own view was that the earnings weakness was structural, driven by several unfavorable trends working against Amazon.com, from the changing mix of sales, to the migration to digital OS-integrated stores, to the need to collect sales taxes. My own view was that earnings would get worse and the "investment thesis" was basically false. My own view was that Amazon.com would soon post earnings losses, instead of showing strongly growing earnings as expected.
And back then, I thought earnings mattered. Thus, with the stock already trading at 169 times 2011 estimates at the time ($1.28 per share), I thought that if I was right regarding what was going to happen to earnings, I'd be right regarding the stock.
I was wrong.
I was not wrong regarding the earnings, though. Amazon.com went on to post a quarterly loss (in Q3 2012). It even went on to post a yearly loss for the whole of 2012. And even more brutal, this is what happened to 2014 estimates over the last 3 years (Source:bwolf, Twitter stream. This is non-GAAP, so it includes compensation paid in stock)
In a way, Amazon.com did worse in the earnings front than even I could have ever imagined. I didn't expect it to post a yearly loss, and it did. Time and time again, Amazon.com failed to deliver the expected earnings and estimates were drawn lower and lower quarter after quarter. At some point, each passing quarter meant estimates for the next quarter got cut 80% or more. And it was not just short-term earnings estimates that were being cut, as we can see above with the 2014 non-GAAP estimates.
And yet, I was wrong.
The stock proceeded to march ever higher on ever lower earnings estimates. The stories started being more and more otherworldly, with analysts even pulling targets from 2019 earnings multiplied by 50, when they couldn't get next quarter's earnings right 3 months ahead.
But I was the loser in all of this.
Let there be no doubt, I was the loser. I would have been better served by complete ignorance of Amazon.com's fundamentals. I would have been better served by not knowing that investments are capitalized and depreciated over time - because time and time again we see analysts, commenters, authors, pulling the "earnings are not there because they are being reinvested" meme.
I would have been better served by not understanding Amazon.com's business, not seeing the flaws and structural problems. Not seeing shipping costs shooting up even with the expanded distribution network. Not seeing how 3P and AWS are not really 100% gross margin businesses, like Amazon.com accounts them for, and how that inflates gross margin as a percent of sales.
I simply knew too much, where ignorance would have been bliss.
I knew too much, because if I was going to go back to October 2011 with the knowledge I have today of what happened to Amazon.com except for the share price, I'd have done it all again. And I would have lost, again, just like I did.
Yes, I lost because I knew too much.
Perhaps this is simply what Bernanke wants from the markets. To drive compliance into people's skulls, or have them lose heavily no matter what happens. Thou Shalt Not Sell Short, no matter what. I had partially learned my lesson, in that I no longer sold short and held anything except Amazon.com. But it was not enough, the point had to be driven home.
And it was, I lost. No matter what.
Regarding this latest quarter
A quick comment regarding Q3 2013:
- Earnings met expectations due to an income tax benefit; they'd be somewhat lower without it;
- Amazon.com is again seeing CSOI margins dropping, unexpectedly;
- Amazon.com's revenues were probably helped by the ebook accounting change, driving revenues from 3P to 1P. Amazon.com won't say how much this effect was, I'd estimate it at around 3.9% of revenues ($666 million);
- Gross margin dropped from 28.62% (Q2 2013) to 27.65% (Q3 2013), an effect that was expected if the accounting change helped revenues. This drop is consistent with around $666 million in additional revenues from the accounting change;
- Gross shipping costs continued up;
- Net shipping costs also increased marginally, which is surprising.
All in all, a quarter that was like many which preceded it. Not becoming of a $170 billion market capitalization company, but it all doesn't matter.
Regarding the accounting change, the analyst from JPMorgan, Douglas Anmuth, did ask Amazon.com in the earnings conference call about the impact. This was the question and answer:
Douglas Anmuth - JPMorgan
Great, thanks for taking the question. Just want to ask two things. First, Tom, can you give us some color on where you are in the shift from third-party to first-party e-books and how much of the factor that's been in reaccelerating media revenue, you have seen reacceleration in media in the last three quarters, I think in North America? And then secondly, it looks like there is six fewer shopping days this holiday season between Black Friday and Christmas, just curious what you do with anything differently to prepare for that and do you think that could actually even drive more holiday shopping online? Thanks.
Sure. In terms of your second one, second question, there are fewer days. There is not a lot that we do different. We certainly see when that happens. And there is some behavioral differences on behalf of customers just because of the shorter time period that we have certainly some more sizable days during that period, but there is not a lot to add to that. In terms of the transition for e-books, in terms of our total growth across Amazon both North America total, our global total, it's not a significant or meaningful impact to the overall growth rates and certainly this transition has been going on for you know some number of quarters now so it's not a lot I can help you with there.
Amazon.com brushed it off as "not a significant and meaningful impact". But we need to understand what's significant for Amazon.com. Amazon.com once brushed off an inquiry from the SEC regarding Kindle sales as not being material. Amazon.com seems to have taken an accounting threshold to it - as long as it's less than 10% of revenues, it's not material. Here, I'd estimate the impact from the accounting change to have been 3.9% of sales, or $666 million. The difference between a revenue beat and a miss. Not material. Even though Amazon.com breaks off things like the currency impact well below that threshold.
Anyway, it doesn't matter.