Last summer Scott Devitt of Morgan Stanley put out a research note on Amazon.com in which he remarked that Amazon (AMZN) in 2011 reminded him of Wal-Mart (WMT) in 1991. I clearly recall pondering his observation but admittedly was quick to dismiss the notion that any reasonable comparison could be made between Amazon, trading above $225 that day with a forward price/earnings [P/E] ratio north of 130, and Wal-Mart, a regular dividend payer since 1974 and widely-held value portfolio stalwart.
Amazon shares did retreat from the $220s in early August but their breather was brief, as the stock soared to a new all-time high of $246.71 in mid-October on the back of highly optimistic expectations for the Kindle Fire, the Company's answer to the wildly successful Apple (AAPL) iPad. The fervor proved to be short lived however, as a disappointing Q3 earnings report derailed the AMZN train and shares slid for the remainder of the year to end 2011 at $173.10, 29.83% below their October peak.
As a trader and investor I love growth - I crave it actually - but I have never been one to pay through the nose for it. Selfishly, I was pleased to see Amazon shares in a downward spiral; with each successive downtick the Company's valuation was gradually approaching a level I could maybe begin to wrap my head around. I had planned to roll up my sleeves for some serious research when the stock dipped below $150. Needless to say, my due diligence process never even began. After spending the majority of January and February 2012 range-bound between $180 and $190, the stock broke out in mid-March surging from $182 to an intraday high of nearly $210 on March 27th.
With 2012 earnings expectations almost universally revised lower over the last three months, at $210 Amazon would trade at a forward price to earnings ratio of 164. This most recent run-up led me to ask myself: what were my fellow 'investors' seeing that allowed them to ignore such an atypical multiple for an almost seventeen year-old business? I wondered if the market had ever seen a growth story like Amazon actually come to fruition. Could the Company ever grow into its valuation?
Remembering Mr. Devitt's likening of today's Amazon.com to the Wal-Mart of twenty years ago, I decided to take an in-depth look at the 'comparable' financials underlying the two retail giants and see if I could indeed identify the similarities that the widely followed analyst alluded to last July. My goal was twofold:
- To mitigate concerns that the market is overpricing Amazon, even given its enormous future earnings potential, using Wal-Mart's valuation in early 1992 as a point of reference and
- To predict how Amazon shares could potentially perform over the next two decades by applying Wal-Mart's historical growth metrics as drivers.
Simply put, I hoped to finally learn (and possibly confirm) what everyone else seemingly already knows about Amazon's inherent value.
Comparing Sales, Growth and Margins: Wal-Mart in 1991 vs. Amazon in 2011
In 1991 (fiscal year ended January 31, 1992), Wal-Mart reported net sales of $43.89 billion, up 34.62% from 1990. After paying a $0.17 dividend, $1.40 in earnings per share [EPS] fell to the bottom line, resulting in a net profit margin of 3.67%. By comparison, in 2011 Amazon.com recorded net sales of $48.08 billion, up 40.57% over 2010, and generated net profits of $1.37 per share and is yet to ever pay a dividend. Although it grew sales slightly faster in 2011 than Wal-Mart did in 1991, Amazon.com's 1.30% net margin was considerably lower than Wal-Mart's comparable 3.67%. One may be inclined to consider Amazon's 2011 net margin an 'outlier' as in 2010 the figure was 3.27%, however the net margin consensus estimates among analysts that cover Amazon are 0.95% in 2012 and 1.53% in 2013. Given the significant impact of subsidizing (or fully absorbing) ever-rising shipping fees on net margins and the concern that the inevitably mandatory collection of sales tax could further erode profitability, relative margin weakness represents a rather significant hurdle when attempting to make the case for Amazon's future being a bright as Wal-Mart's past.
Comparing Valuation Multiples: Wal-Mart in 1992 vs. Amazon in 2012
On March 30, 1992 Wal-Mart shares closed at $53.00 ($13.25 adjusted for 2 subsequent splits). Against 1991 earnings per share of $1.40, WMT was changing hands with a trailing P/E of 37.9 - well above the S&P 500's 16 P/E ratio at the time, but nowhere near Amazon's multiple on March 30, 2012 of 147.8x 2011 earnings. Per analyst estimates, we should even brace ourselves to see that multiple rise to 158.2x given expected 2012 earnings but then anticipate that it will drop considerably the following year to 76.4x 2013 earnings per share. Naturally those multiples assume that the share price will remain pinned at $202.51 for the next two years, which is as likely as me winning the $640 million Mega Millions lottery drawing tonight.
Consequently, I was unable to quell my uneasiness over Amazon's current earnings multiple by comparing it to Wal-Mart's in 1992. Instead, I would suggest that Amazon's inflated multiples imply that the market has essentially 'stripped out' at least three years of strong returns for an investor looking to establish a position today by already over-pricing medium-term earnings growth into the stock. Therefore, maintaining the thesis that Wal-Mart in 1992 and Amazon.com in 2012 are indeed legitimately comparable, I contend that buying Amazon today would not be like buying Wal-Mart in 1992; it would actually be like buying Wal-Mart in 1995, which would meaningfully reduce the long-term return on one's investment.
Further, with the price to earnings multiple at 147.8x today, eventually the market will [have to] stop bidding the stock up even if Amazon perpetually reports very strong earnings growth. As we learned from examples such as Qualcomm (QCOM), Cisco (CSCO), Microsoft (MSFT) and EMC (EMC), companies simply cannot trade at an exorbitant multiple forever. Paying-up for Amazon now - with a twenty-year investment horizon - virtually guarantees that at some point over the next two decades returns will be diminished via inevitable multiple contraction despite financial results that could potentially exceed expectations quarter after quarter.
Amazon.com in 2032: Can it look like Wal-Mart does today?
Although my initial comparison left me cautious on the prospect of investing in a far more richly valued Amazon.com, it is still possible that if Amazon is able to grow like Wal-Mart did from 1991-2011 over the next twenty years, I may be able to justify 'surrendering' the medium-term growth already priced into the stock in return for overall outsized gains.
In 2011, Wal-Mart recorded net sales of $446.95 billion, implying a 12.30% compounded annual growth rate [CAGR] since reporting net sales of $43.89 billion in 1991. Despite increasing revenues more than tenfold over the twenty-year period, Wal-Mart's price to sales ratio [P/S] decreased from 1.4x in 1991 to 0.5x today as its market cap only grew 242% from $60.90 billion in 1991 to $208.36 billion as of March 30, 2012. At $61.20 a share, WMT currently trades at 13.5x 2011 earnings per share of $4.52.
Applying Wal-Mart's twenty-year top-line CAGR of 12.30% to Amazon's 2011 sales of $48.08 billion, we would expect the Company to generate sales of $489.65 billion in 2031. Assuming Amazon never splits or repurchases its stock, there will likely be ~644 million shares outstanding in twenty years (estimated by applying Amazon's 2007-2011 shares outstanding CAGR of 1.40% to the 461 million share count in 2011). Before backing into the expected future value of the Company, its potential earnings power and the kind of growth one might expect from an investment in Amazon today, we must acknowledge the two most central determinants of its future valuation in the marketplace:
- Net profit margin and
- The P/E multiple assigned by the market
Although I ran scenarios where, in 2032, Amazon shares trade at 13.5x expected earnings (comparable to WMT today) and 15x expected earnings, I find it most reasonable to assume that the Company will trade at approximately 20x 2031 EPS, which still represents a premium to today's trailing multiples for tech bellwethers AAPL (17x), IBM (16x), CSCO (16x) and MSFT (12x). Even realizing top-line growth exceeding 1,000% over the course of two decades, the conversation as to whether Amazon.com would make a good investment today becomes unnecessary if the Company is unable to considerably improve upon its 2011 net margin of 1.30%. At 1.30%, expected 2031 EPS would be $9.88, which at 20x, implies a stock price of $197.62, which is roughly 2.4% lower than today's close! Even improving profit margins to 2.0%, at 20x expected 2031 EPS of $15.20, the stock will have only risen from $202.51 today to $304.02 over twenty years, which represents a CAGR of just 2.05%, growth that could possibly lag inflation. Assuming that Amazon can eventually bring net margin up to approximately match Wal-Mart at 3.50%, expected 2031 EPS would increase to $26.60, implying a $532.04 stock price at 20x earnings or a 4.95% 20 year CAGR - better but still not the return I would require for having my money invested in a relatively 'higher'-risk equity for two full decades.
Clearly, Amazon must focus on profit margin expansion if it intends to maximize shareholder value and see its shares generate substantial long-term returns for investors. While margins in retail are typically razor thin, the Company may eventually begin to generate material revenue in its more profitable Amazon Web Services segment and that could push overall margins higher. In a final example, let's assume Amazon is able to improve overall profit margin to 6.0% by 2031. Implied EPS would be $45.60 which at 20x would give us a $912.07 stock, a $587.58 billion market cap and P/S ratio of 1.2x. Twenty-year CAGR would come in at a respectable 7.81%, a long-term annual return that most investors would readily welcome.
And the verdict is…
So is Wal-Mart a harbinger for Amazon.com in 2032? I'd say yes, and no. While I agree with Mr. Devitt that there are numerous similarities between Amazon.com today and Wal-Mart in early 1992, I do not feel that those similarities in business life-cycles translate to comparable investment opportunities. The market has already priced in considerably more growth in Amazon.com than it was pricing into Wal-Mart shares in 1992. Thus, the potential for double-digit annual growth in Amazon's stock over the course of twenty years is extraordinarily unlikely for an investor that establishes a position at or around $202.51.
Employing the conservative assumptions that Amazon can achieve 3.5% net profit margins and will trade at 20x trailing earnings in 2032, we can expect a twenty-year CAGR of 4.95%. To highlight the extreme impact that paying-up for Amazon shares today would have on long-term returns, let's fantasize that we could buy AMZN today at WMT's 1992 trailing P/E of 37.9. 37.9x Amazon's 2011 EPS of $1.37 would give us a stock price of $52, which incidentally is only $1 less than Wal-Mart's March 30, 1992 closing price. If we could actually buy Amazon right now for $52, our twenty-year CAGR would more than double from 4.95% to 12.35% in the same scenario.
There has never really be a question of Amazon growing into its current valuation - it most certainly can and almost assuredly will - the question that must be asked is what will growing into its 147.8 trailing P/E cost new shareholders (those that establish a position around at or around today's closing price of $202.51)? As evidenced by the example above, investing in a lower multiple stock demonstrates that the cost of paying-up for a stock trading at Amazon's current valuation is quite sizable; the cost is so considerable in fact that I would consider it prohibitive.
Therefore, although I do believe that Amazon.com can certainly meet and quite likely exceed the top-line growth that Wal-Mart realized from 1991-2011, I cannot justify establishing an initial position in AMZN stock at these levels. While a material improvement in net margins could mollify my reluctance to 'invest where near/medium-term value has already been stripped out of shares,' I am generally skeptical of the Company's ability to meaningfully expand margins (to a minimum of 6%) across nearly half a trillion dollars in revenue under the assumption that traditional online retail will always account for the vast majority of Amazon's total sales.
So after all that, I may still be unable to understand why the market continues to bid-up Amazon shares but I am able to perfectly summarize the preceding 2,000 words with just two: Don't Buy.
[…is it also a short? Yes, but only as part of an active near-term trading strategy that includes well-defined and very precise entries/exits and employs tight stops.]