Tuesday, June 12, 2007

Amazon: Why now may be the time to short

This write-up that AMZN is a compelling short at current levels. AMZN has shot up over 70% in the last two months, most recently on its "blowout q1" earnings report, a typically seasonally weak quarter in which AMZN reported EPS higher than their seasonally strong q4. Revenue growth has accelerated, and high hopes for new initiatives (digital distribution, web services, etc.) have fueled a speculative furor not seen in the name since the height of the boom. Analysts are excited by the prospect of amazon as a "media" company, rather than a online retailer of low margin products, trading at a over 100x earnings, with PEG multiples higher than GOOG and EBAY, both of which are more attractive, higher margin businesses.

For a variety of reasons, I believe these growth avenues are overblown, and that AMZN is more likely than not a leading online retailer with a strong online marketplace, low operating margins, and moderate but not spectacular growth going into the foreseeable future. I will argue that even if AMZN overnight became a leader in all areas they hope to grow into, the company would still be overvalued and likely to disappoint given current expectations. AMZN is overpriced on both an absolute basis, as well as relative to peers, and could see a upwards of a 30% drop based solely on a return to its prior lofty valuation levels, or upwards of a 50% drop if valuations were more in line with comps (which, arguably, are overvalued themselves), and potentially see further decreases as new ventures fail and their core retailing business receives continued pressure from offline and other online retailers, as well as the inevitable levy of an internet sales tax, which could wreak havoc on already tiny margins. Given the high PEG, as well as revenue and earnings growth going forward, I believe the risk of multiple expansion or multiples staying the same is relatively low, and that AMZN is an attractive low risk/high reward short.

Q1 "Blowout"
Amazon surprised analysts and everyone following the name. Though there were some legitimate business drivers for the gains (lower than expected margin compression, slightly better sales, etc.) the majority of the earnings surprise can be attributed to lowered R&D investment, a favorable tax situation, and foreign currency gains. These are not the kind of operational improvements that merit such an enormous increase in stock price, but instead served to artificially show earnings growth well above the real growth in the business.

Business Overview:
AMZN is the largest pureplay internet retailer, with $10.7B in sales in 2006, and has grown its revenue at about 25-30% per year for the last few years. Though AMZN is primarily known to US investors for their domestic presence, the company has become an international force, deriving 55% of sales from the US vs. 45% internationally. The company continues to grow outside its core media products into other categories (electronics, etc.), but media--composed of books, dvd, and cds, still account for 66% of revenue.

AMZN is entering other popular spaces, including distribution of online music, dvd rentals, and its much hyped Amazon Web services. These businesses are all in their infancy, but are being relied upon to deliver growth coming forward, which I do not believe will come for several years, if at all.

Amazon's non-retail businesses
Before delving to much into the economics of AMZN's core business, I think it is useful to discuss the prospects of Amazon's non-core operating businesses. I will argue that even if AMZN were to overnight become a leader in each of these new business, the total value of these business would not amount to more than $3B, or 1/10th AMZN's market cap.

Amazon Web Services
This business has received hype for sometime--despite being offered for the last couple years and not gaining much traction. The sales pitch here is that Amazon can sell its best in class technology and logistics solutions to other companies rather than those companies needing to worry about managing their own proprietary solutions. Outside their impressive e-commerce engine, which has been the bulk of the services hype, AMZN has a handful of other tools (a mediocre search product, selling excess storage an computer power, alexa, etc.) that have limited commercial value. Some analysts speculate that this opportunity could become larger than the entire business. I don't think any of these services--particularly the ecommerce service--will ever take off. Reasons include:
1) Amazon is essentially trying to sell its technology to its main competitors (offline companies coming online). This is a lose/lose situation. Either it works and you make your competition stronger, or it doesn't and your business suffers.
2) Unsurprisingly, companies are not too keen on the prospect of outsourcing anything to a large competitor. If you are best buy, for example, how would you feel about licensing technology and trusting your infrastructure to your biggest online threat? It makes no sense to put the core infrastructure of your business into the hands of someone with a clear conflict of interest. Frankly, I can't see this business taking off as long as AMZN also acts as a retailer.
3) AMZN's e-commerce solution, in particular is not friendly with other solutions. If you are a offline retailer building an online presence, you want to be able to integrate your storefront with your webfront with your catalogue. It is very difficult to do this with amazon's solution, and this will never be something they are good at, given that they do not have a retail presence themselves. If you cut offline retailers out of the market, who are you left with, other than a few small niche online retailers who have themselves spent millions developing their own systems?
4) AMZN's other services (search, storage, cloud computing, etc.) are of questionable value, and are not related to their core competency (ecommerce). They might be able to make a few million from selling some extra memory and bandwith to startups (as they do now), but in my mind these don't representative particularly valuable near term or long-term business models.

For many of these reasons, Amazon has been eaten alive by GSIC in the e-commerce services space, which is the only area where I believe AMZN has a potentially valuable product. GSI, originally operator of small website Fogdog.com, has taken many of amazon's customers and won many contracts in head to head battles with e-commerce behemoth. GSIC reported $600M of revenue in the last fiscal year with its business here. Though AMZN does not break out this revenue, its other bucket for $263M for FY2006, so at the very least this business is half that of GSIC's and most likely more like a third or a quarter. Anyhow, for arguments sake, I will assume that this division is currently worth GSIC's EV ($1B), despite my belief that these businesses will continue to drain cash, and not be worth much of anything at all.

DVD rental
AMZN's foray into online dvd rentals is a much hyped, but largely irrelevant part of AMZN's future growth prospects. AMZN is a late entrant into the space, and it is unclear what additional value they bring compared with established competitors (NFLX, BBI, WMT, etc.), outside of arguably some benefit from their large customer list and distribution infrastructure. Despite speculation that they would launch in the US, they have only launched in the UK for the time being. I find it ironic that this business is attractive and an asset to a company like amazon, while the business on a stand-alone basis (with BBI and NFLX) is considered to be intensely competitive, unsustainable due to high churn, and at risk altogether from digital downloads. Forgetting all the concerns with the business model and the fact that AMZN is way behind the competition, I will assume that AMZN will become a leader in the space and that this opportunity is worth $1.2B currently, which is NFLX current EV.

Digital Distribution
Amazon is trying to eventually position itself to be a leader in digital content distribution. AMZN is late to the party, and has some formidable competition from the likes of Google, Yahoo, Apple, and a handful of small niche players (MovieLink, CinemaNow, etc.). To date, AMZN has focused on hyping its new digital music service. Lets forget for a moment that nearly all the big players have launched a similar service, or that apple is the undisputed leader in the space. After over a year of discussing ways of differentiating itself, including scrapped plans of releasing its own player, AMZN appears to have settled on a rather unspectacular solution--they will release DRM free music from EMI (for which APPL also has a contract), as well as 12,000 independent music companies. In other words, outside the 12k independent labels (who all likely have deal with APPL as well), there really isn't much differentiation, at least for the time being. Also, unlike other players, AMZN risks cannibalizing its own music sales.

Amazon also went ahead and released Unbox, a video download service that received some pretty awful reviews, particularly in comparison to Apple's storefront. This is another area where there are several established players, though no one has really been able to make the model work the same way it has worked for music. Even with broadband, movies can take a prohibitively long time to download, take up a large amount of server bandwidth, as well as a good deal of physical memory. Not to mention, unless you want to go through the hassle of hooking up your computer to your TV, you'll be confined to your computer screen for video watching. Like Unbox, AMZN's partnership with TiVo has generated little response, for similar reasons. Overall, I think this is another example of an expensive, early, and unattractive business. Despite all my reservatios, I'll assume an EV of $800M for the digital opportunity, or about 8x the EV of Napster.

Reviewing many of AMZN's most touted growth prospects reveals that, beneath the hype, AMZN faces significant competition in most areas, and even if we assume it becomes a market leader in each category, the total value of the opportunity is not particularly attractive currently.

Netting out the $3B in non-core, immaterial businesses leaves us with a $27B market valuation on Amazon's e-commerce business. Using Amazon's upper end EBIT guidance for FY07 ($563M) gives us a forward valuation of 48x/EBIT. I use EBIT to net out the wild fluctuation in tax rate, which came in in the low 20%s vs. 40-50% historically for q1, and contributed to a large portion of the blowout q1. In addition to being an absurd multiple on its own right, this is well above EBAY (20-25x) and GOOG (25x-30x), both of whom I would argue have more sustainable competitive advantages, more attractive margins, and as good if not better growth prospects than AMZN. Even if we assume AMZN is worth the high end of GOOG's EBIT multiple, the stock would be worth about 40% less than it is today. If use consensus forward P/Es rather than EBIT, the numbers look even worse: 70x 2007 for AMZN, 33x for GOOG, and 23x for EBAY. Using GOOG again as the upper end of a market valuation would result in a drop of 0ver 50% from todays levels. Using EBAY (arguably a more fitting comp) would result in 67% drop.

Another way to look at this is that the market was valuing AMZN at about $45/share before their earnings announcement, which it beat largely due to a lower than expected tax rate, a large decrease in R&D investment, and favorable foreign currency gains. I think it's difficult to argue that these factors should result in adding about $25/share increase, and that as a base case it would appear reasonable that AMZN should return to $50, where it was trading before its earning release, or a decrease of about 30% from current levels. Basically, any way you slice the analysis, AMZN is trading at an unjustifiable high valuation compared to comps, where it has traded historically, and on an absolute basis

Though I am usually hesitant to short high growth names, I believe the risk with AMZN is relatively low. Given memories of the internet bubble, and valuations significantly above other internet companies with superior growth profiles, I find it hard to believe that AMZN could achieve additional multiple expansion. Also, given my low expectations from growth initiatives, I don't believe we'll see AMZN as anything more than a 20% grower in a low margin retail business which, if correct, will be rewarded with a much lower valuation than experienced currently. Likely worst case scenario in my mind is that AMZN grows into its valuation over the next couple years and the stock moves nowhere. One option for the risk adverse would be to do a pair trade with GOOG and/or EBAY and profit from the multiple compression while hedging out some risk that the market continues its irrational pricing of some internet stocks.

Potential catalysts:
-Continued failure of new initiatives
-Internet sales tax is enacted or gains momentum
-Increased competition from niche players and offline companies building out an online presence.
-Rotation away from the internet names
-Unexpected tax rate fluctuations in upcoming quarters

Disclosure: I am currently short AMZN.

Friday, June 8, 2007

Track my Performance, and yours too!

Using Coverstor's awesome portfolio performance tracking, I have went ahead and set up a little widget on the right side of my blog, about halfway down, that shows a few of my top holdings in a separate account I manage. This account is different and larger than my personal account, so the weightings of some stocks I writeup (e.g. eylogic and MCZ) may actually be smaller concentrations in this portfolio than I say they are when I refer to my personal holdings.

I just set up the account, so the performance tracker for the time being is largely useless, but it does display the majority of my holdings, as well as brief rationales for several. For those of you looking for a portfolio tracker, I highly recommend Covestor's--it updates automatically, and tracks things like max drawdown, portfolio beta, etc.

Thursday, June 7, 2007

The dawn of "Covesting"

David Jackson from seeking alpha posted a great article on Covestor, a newly launched site that allows investors to track their performance and the performance of other members. The eventual goal will be to charge users to "subscribe" to a members trades, and then split the profits with the investor. Covestor is focusing on a non-professional audience and allowing them to monetize their trading ability. I think this is an interesting approach--it's a bit like a platform that essentially allows anyone to start their own newsletter service, and simply pays a cut to Covestor for their platform.

Outside of the potential financial benefit, Covestor's portfolio tracking itself is pretty impressive--it automatically calculates your portfolio's beta, sharpe ratio, and alpha, and confirms your trades from your brokerage account without you needing to input anything manually. I am in the midst of setting up an account, and will post a link to my portfolio once it is complete.

Tuesday, June 5, 2007

MCZ FY07 Results – Turnaround appears intact

MCZ reported earnings of $.01 in Q4, of $.07 for the year. This is their 2nd straight quarter of strong gross margins, which came in at 29.1%--particularly impressive considering there was no major release of high margin product in the quarter, as far as I am aware. Management has done an excellent job improving gross margins on low margin product, and refocusing the business on more attractive products going forward. Management also kept expenses down, and has continued to paid down a good chunk of debt from cash flows. Looking ahead to next year, the company should benefit from launch of the InAir headphones, halo faceplates, as well as additional licensing deals that should be announced over the next year.

On the downside, management suggested they will not be launching a software title in the next fiscal year, which could make revenue comps difficult. I estimate that software accounted for about 12-13% of revenue in FY07 (including a whopping 28% in Q1), and its possible they could lose upwards of half of that next year as their current titles age. They are going to have to figure out a way to make up for that revenue--InAir and more licensed products could do the trick, aided also by a continuing recovery in the company's core hardware accessory sales, though it may be difficult to grow revenue more than a few percent, barring a knock out hit of some kind. The company will look to launch up to two titles in FY09, once the console transition is well behind us, and the installed base is more favorable to a MCZ release.

At 18x earnings (as of this writing), Madcatz does not appear particularly cheap. But looking ahead to next year and beyond, if the company continues to grow revenues at higher gross margin levels, earnings should ramp up quickly due to the operating leverage in the business. Revenue growth of as little as 3% would translate into earnings of about $.11 at a gross margin of 27.5%, assuming expenses in line with expenses this year. That said, the future is far from guaranteed—the InAir launch in particular is a totally new space for MCZ, and very little has so far been released about it. MCZ is and will likely continue to be a hit driven business, and its success relies on continually churning out new products. I will be looking to further announcements of licensing deals, more details on InAir, a Wii controller announcement, and q2 results (with 6 days of Halo sales) to see how MCZ should fare this fiscal year and beyond.

Disclosure: I am long shares of MCZ, and reserve the right to buy or sell shares without notice. This analysis is for educational purposes only, and should not be construed as investment advice or fact. Do your own due dilligence.

Monday, June 4, 2007

Position Sizing and when to sell

Figuring out the right position size for a stock, and deciding when to sell and when to double down is probably one of the more difficult aspects of stock investing, and an area I find myself struggling with regularly. It is one of those areas of investing that I believe is a bit more of an art than a science.

Position Sizing
A lot of position sizing usually depends on an individual investors risk tolerance and desire for diversification. I know several people who keep a very concentrated portfolio (as few as 2-10 stocks), and others with upwards of 30-50 positions. Each has its benefits, though I personally lean towards the more concentrated portfolio, as it is easier to keep track of, and allows for more $$ in places where you have the most conviction. I personally like to fall within about 15-20 for the long side of my portfolio, as it allows for a bit more diversification while still allowing me to heavily weight my top 5 holdings, which I usually like to have about 40-50% of my total portfolio. I don't like to have a position grow to more than about 15% of my total portfolio, unless there is a very strong margin of safety or unless I am unusually confident in the companies prospects. In general, my largest holdings are ones I have the most confidence in, and those with the largest margin of safety (I will rarely, for example, invest 10% in a speculative stock). Many of my smaller positions (<3% positions) are in companies I would like to own more of but are a bit too expensive for me right now (examples would include MSII and PN). Others are ones that have a high potential payout (200%+), but also a decent chance of being worthless, and often include puts. I have recently considered adding a few puts of high flying growth stocks that I think are likely to face issues at some point (CROX, JDSA, RIMM, and JMBA), but that I would not want to short for fear of being wrong and losing my shirt. I would likely have these as 1% positions, unless I had a very clear catalyst and timing in mind.

Position sizing is a constantly evolving question. As information changes (the stock price, company prospects, etc.), the position should be re-evaluated to ensure exposure is consistent with the risk/reward of the stock. In the case of MCZ, I originally started it off at as a 10% position. When the halo deal was announced and the stock shot up, I continued to hold the position, as my fair value of its upside potential was now more than when I bought. When the stock reached $1.40, it was beginning to near some of my more conservative valuations. I asked myself if I would buy the stock again if I did not own shares, and the answer was yes--but it seemed to me more fitting of a 5% position in light of my other holdings, vs. the 15% holding it had become.

When to sell
I think this is another area that can be a bit more of an art than a science. With certain investments, I usually have a catalyst in mind that I plan to sell around--for example, strong earnings in a particular quarter. With others, I plan to hold the stock indefinitely as it is a company I want to be invested in over the long haul (EYE.A and CASH are good examples).

I try as best I can to stay objective with my holdings, and always ask myself if I would still buy if I did not own it--particularly after big moves in the stock price or news that alters my investment thesis. Madacy is a great example of an investment in which I lost my shirt, but I believe I played well, all considering. The stock got whacked down about 60% after what appeared to be a one-time issue with their largest customer. Management bought back shares, and I decided to double my position, as I believed my thesis was still intact. Over the next couple quarters, however, it became clear that this was more than a one time issue, and that the operating business was significantly impaired. I recently closed the position at a loss.

Most important, I think it is useful to continuing evaluate what strategy works best for you and what doesn't, and to make decisions accordingly. There are rarely hard and fast rules in investing, and I think this is an area in particular where that is true.

Saturday, June 2, 2007

Underperforming CEFs as a hedging strategy?

I have been running my portfolios at about 0-20% net long (long exposure minus short exposure) for the last year, largely in accordance with Hussman's hedging strategy which has, over a full cycle, generally outperformed an unhedged strategy and with less volatility. I make some modifications--most notably shorting individual securities rather than his puts strategy, which I believe affords similar downside protection while allowing for the possibility of returns on the short-side, as well.

Unfortunately, I don't have enough ideas on the short side to fully hedge my portfolio, so I usually end up just shorting the indexes, or a ETF I believe will underperform (e.g. JKK or RWR). When doing this, I look for a sector that I believe is overpriced, and then find the worst ETF I can in that sector. This has had its intended effect, but a lightbulb recently went off in my head, and suggested a strategy that may be more effective. I believe shorting poor CEFs can be a more attractive way to hedge out market risk than shorting indexes, in the same way shorting a historically poor performing mutual fund would have been better than shorting indexes. If you are not familar with CEFs, I suggest reading this article and this. Otherwise, I have summarized:

Closed end funds are similar to mutual funds (open-ended funds) in that they are pooled capital managed by a portfolio manager. The difference is that Open-ended funds IPO to raise capital, and subsequently trade like stocks in a market based on supply and demand, whereas Mutual funds constantly issue and cancel shares as investors buy and redeem their shares. Because CEFs trade in a suply and demand market, the market price can become detached from the net asset value (NAV) of the underlying holdings. This is referred to as the premium or discount a fund trades at, and is usually driven by past returns and the dividend yield. This structure also means you can short CEFs, which you cannot do with mutual funds.

Shorting CEFs, in my opinion, has a few benefits over shorting ETFs:
1) Fees are higher, creating an increased hurdle to achieving returns above the market. It is not uncommon for smaller CEFs to have management fees of abou 1-2%, and expenses of about 2-3%, resulting in a 5% hurdle out the gate.
2) Most CEFs have less incentive to perform well than do mutual funds. When mutual fund shares are redeemed, the mutual fund company loses assets, but because CEFs raise money at the start of a fund they don't have the same concern. If investors are not satisfied, they do not get their money back from the fund, but on the market by selling shares to another investors. The company does not experience an outflow, and thus does not care much about a discount. Good CEFs do care about their reputation and ability to raise future capital, but luckily their are plenty of bad ones out there that don't seem to mind poor performance.
3) CEFs have less incentive to replace underperforming managers, assuming their fees are calculated as a percentage of assets (as most are). Many are happy to collect 2% of $50M indefinitely while exerting zero effort, rather than grow that a couple percent a year extra and have to work for it.
4) High volatility CEFs and/or those that underperform the market typically swing to a discount when the market heads south. In this case, you can profit both off the decrease in the NAV, but also in the increasing gap between the market and NAV price. Many high flying emerging market ETFs, for example, are trading at 5%+ premiums. In the event the NAV dropped 20%, and the 5% premium became a 5% discount, your profit would be 29% (a 20% from the NAV decline, + 9% with the premium contraction).

Next week I will do more work on profiling some of the more shameful CEFs that I plan on adding as short positions in my accounts as a hedge against market fluctuations.