Aeropostale (ARO): Part Four - Shleifer and Ackman

As we discussed previously, Professor Andrei Shleifer's Inefficient Markets provides an instructive lens for viewing the past year or so of Aeropostale's business operations and ensuing price volatility.  

In short, investors watch a business, evaluate its performance, and form an opinion that tends to extrapolate its recent performance into the future. This tendency is called representativeness. 

Once they form an opinion, they stick with it even in the face of disconfirming evidence. This is called conservativeness, and it remains the case until multiple pieces of evidence finally overwhelm them and force a change of mind. Investors then fall prey to overreaction - classic Mr. Market - getting overly excited about a company or depressed about it and pushing its price up or down (whichever the news may warrant.)

The concept passes my reasonable man test for how people (I'll include investors in this category) behave, and so I'll graft it on top of what we've observed with ARO to see what we can learn.

First, let me step back a little for an aside (albeit a long one) that I think drives home the point. A few months ago I watched a re-broadcast of the Fourth Annual Pershing Square Challenge at Columbia Business School. It took place in April 2011 and features teams of business student presenting their best investment ideas to Bill Ackman and a panel of judges. (Thanks to Value Walk for posting the video. You can access it here.)    

One team pitched ARO at 26.00 and the judges (Ackman in particular) laid into them with their critique. As I've developed my ARO thesis, I've returned to this critique to challenge my own thoughts for the investment. (Another gratuitous hat tip here to the excellent blogger, Greg Speicher, who suggested this very thing in a recent post you can find here.)

The students highlight how ARO has several years of excellent - yet improving - operating metrics, making it a best in class business model for mall-based teen clothing retail. The team shares much of my view on the strength of the business model, and argues that - based on growth potential of the P.S. concept plus international growth plus ecommerce - this business is worth $34 per share.

The judges bring up pertinent objections such as...

1. The risk of being a copycat of Abercrombie & Fitch, especially if the Abercrombie style falls from teenager fashion grace. This lack of diversity sinks the business. So, how confident can an investor be that a.) style stays the same, and/or b.) ARO has the ability to adapt to any change in fashion?

2. The risk of shocks to the system disrupting a low margin business. Specifically, can the value retailer in this space with such low gross margins  either a.) absorb increased cost of goods sold while remaining an attractive investment with strong earnings,  or b.) pass the costs along to customers without losing its value-price halo? This is particularly important in regards to cotton price increases, a key input for a clothing retailer.

3. The best critique came from Ackman himself and belied an understanding of retail dynamics born of deep experience. After the team suggested that the worst case on the downside would bring a ten percent decrease in the $26 price, Ackman implied there is always much more risk investing in a retail business where existing investors have become acclimated to quarter after quarter of unabated good news (i.e., comp store increases). If they get an earning surprise or two, they would head for the exits and take the stock price down dramatically in the meantime.

Ackman's comments:

"If this is a stock that has never had a down comp and it starts comping down...something that's going to change is the multiple people are willing to assign to their earnings. So I think the downside is much greater than what you say. 

"I'll give you the most skeptical argument I can make of the company...which is: you have a business that doesn't do anything proprietary, they're basically stealing other people's intellectual property and just reproducing it with a lower quality product. Their margins are much slimmer and being squeezed by commodity prices.

"There isn't much opportunity for growth and there is question about the growth of their competitors who they've been leaching off over time. If you have same store sales declines and margins decrease, profitability is going to drop significantly because of the large operating leverage of the business...

"The combination of those factors, I think you could lose a lot of money on this stock." 

And look how prescient Ackman's comments were. ARO dropped from $26 to $21 just a month later on the announcement that Q1 2011 fiscal year comps were down seven percent. And management dropped guidance for the next quarter to boot. From the date of the presentation in April through October, the stock plummeted to just over $9 per share. A multi-year low for ARO. 

Ackman understood what Shleifer was saying, and the point underscores much of the thinking behind value investing. When you invest in a business whose owners have become accustomed to nothing but success, bad news in the form of a couple missed quarters can send the stock price tumbling.

It is now, in Shleifer's terms, an extreme loser. And extreme losers that survive over the long haul tend to do much better than extreme winners. According to Shleifer:

"...over longer horizons of perhaps three to five years, security prices overreact to consistent patterns of news pointing in the same direction. That is, securities that have a long record of good news tend to become overpriced and have low average returns afterwards. Securities with strings of good performance, however measured, receive extremely high valuations, and these valuations, on average, revert to the mean."

With Shleifer's work we see a compelling case of human behavior grafted from psychology to investing in which human tendencies toward conservatism and representativeness cause overreaction. The prevailing sentiment becomes one of a trend continuing (i.e., even more same store sales declines) rather than a level headed view of the business's future prospects.

If I bought ARO at $26, as the Columbia students recommended, I confess readily that my own ability to assess the affairs of the business would be severely compromised by the system shock of seeing nearly 70 percent of my investment value disappear. Though I believe I have somewhat of a strong stomach, my hard-wired tendency toward representativeness (the bad trend will continue) would likely overpower my rational ability to be calm, cool, and collected in thinking about the future. It would be excruciating not to join the stampede of investors rushing for the exits. Indeed, even if I could have mustered the courage to survive the first two or three earnings disappointments, my resolve would have weakened with each ensuing miss.

So Ackman proved prescient: investors who came to expect even sunnier prospects for ARO as a best-in-class retailer threw themselves overboard as the business future earnings became more difficult to discern. 

That has now passed.

As we said, the job of investors at this point is to gaze deep into the business and decide whether the clouds of uncertainty can be penetrated.

Aeropostale (ARO): Part Three - A Behavioral Finance Case Study

In valuing Aeropostale in the previous post, we built an ugly scenario that assumed ARO's operating performance deteriorated significantly. I suggested that, were this to occur, the market would punish the shares through a broad sell-off that could easily push its price down by 50 to 60 percent from the current $18 level.

This is pretty much what happened in October 2011. After years of growing same store sales quarter after quarter in an unblemished winning streak, ARO went into a slide for several consecutive periods. Same store sales were declining, inventory wasn't moving, gross margins were taking a hit, and it was clear that another winning streak was at risk...growth in earnings per share.

Investors stuck with ARO through the first couple misses. Management deserved some leash, they had demonstrated a lot of success with this business model over the years. But from the spring of 2011 through the autumn, the one-time darling was quickly becoming a toad. The stock price plummeted from $26 per share to a multi-year low of $9.16. Even the most stalwart supporters were losing their resolve - irrespective of their views of the long-term business prospects - as they saw their holding value collapse by two-thirds. While they may have reached the same conclusions about the business as I did in my original post (here), few could stomach being part of a quick fall and then trudge through a recovery that would likely take many quarters if not years.   

For a construct to understand the dynamics of that sell-off, and the resulting opportunity to buy into a good business at a cheap price, I turn to the field of behavioral finance and Harvard economist  Andre Shleifer. 

In his book Inefficient Markets: An Introduction to Behavioral Finance, Shleifer builds on the work of Robert Shiller and Richard Thaler, pioneers in this burgeoning field that blends theory from economics and psychology. 

Considered simplistically, behavioral economics is a rebuttal of efficient market theory dogma...that belief in investors as rational agents using all available information to make rational decisions in a stock market that is thereby efficiently priced at any given moment. Nonsense, says the behavioralists. The market is an amalgamation of individual decision makers, the choices of whom are as influenced by whim, fear, and optimism we all experience as human creatures with emotionally-charged minds.

I picked up this book in spring 2008 and was immediately intrigued by Shleifer's use of basic psychology to theorize why strings of over-performance or under-performance by companies tend to lead to corresponding over- and under-valuations by investors. Allow me to frame the ideas with this long quote from the author: 

"When a company has a consistent history of earnings growth over several years, accompanied as it may be by salient and enthusiastic descriptions of its products by management, investors might conclude that the past history is representative of an underlying earnings growth potential. While a consistent pattern of earnings growth might be nothing more than a random draw for a few lucky firms, investors see 'order in chaos' and infer...that the firm belongs to a small and distinct population of firms whose earnings just keep growing. 

"As a consequence, investors using the representativeness heuristic might disregard the reality that a history of high earnings growth is unlikely to repeat itself, over-value the company, and become disappointed in the future when the forecasted earnings growth fails to materialize. This is, of course, what overreaction is all about." 

What forces might be at play that would allow this over- and under-valuation to occur? Shleifer turns to three heuristics from social psychology for explanation.

The first is REPRESENTATIVENESS. As social psychologists Tversky and Kahneman demonstrated long ago, people have the tendency to believe they see patterns in truly random sequences. We see consecutive quarters of earnings growth, and our mind convinces itself that this is a pattern that will continue into the future. Likewise, we see earnings drop and our minds create a future in which they just continue dropping. Though the events that caused the drop might be temporary (or even random), the human mind begins to see patterns that affect the valuation we assign to a business. In Shleifer's words:

"...when investors are hit over the head repeatedly with similar news - such as good earnings surprises - they not only give up their old model but, because of representativeness, attach themselves to a new model in which earnings trend. In doing so they underestimate the likelihood that the past few positive surprises are the result of chance rather than of a new regime."

Once we establish a model in our minds that the business is trending up or down, we run into the second heuristic, CONSERVATISM...

"...when [investors] receive earnings news about this company, they tend not to react to this news in revaluing the company as much as Bayesian statistics warrants, because they exhibit conservatism. This behavior gives rise to underreaction of prices to earnings announcements and to short horizon trends."

In other words, when we have an opinion, we tend to be protective of it, sticking to it for a while despite disconfirming evidence that should probably cause us to change our minds. In fact, Shleifer notes, it usually takes two to five pieces of evidence that are contrary to our opinion to get us to change our minds. 

But once our minds are changed, the third heuristic comes into play, OVERREACTION. Back to Shleifer...

"...over longer horizons of perhaps three to five years, security prices overreact to consistent patterns of news pointing in the same direction. That is, securities that have a long record of good news tend to become overpriced and have low average returns afterwards. Securities with strings of good performance, however measured, receive extremely high valuations, and these valuations, on average, return to the mean."

And the same is true vice-versa for bad news. In a nutshell, we witness a few events (either positive or negative) and our minds infer a pattern, believing that a trend has been established (representativeness). We interpret the trend and place our bets, sticking to our guns even in the face of initial evidence that our opinion is incorrect (conservativeness). Finally, after several instances of disconfirming evidence we do change our minds, but we do so in a way that pushes us to over- or under-value the company in question (overreaction). 

As any good economist, Shleifer then sets about proving his theory by going to the data and drawing on statistical models with greek symbols and formulas that escape my grasp. But he manages to conclude that, at any given time, the stock market is full of companies that are over- or under-valued based on overreaction to consecutive earnings surprises that stoked euphoria in buyers or panic in sellers. An investor would do well, he concludes, buying the extreme losers from this phenomena and staying away from the extreme winners.

Sounds like an interesting model to lay on top of what we've seen with Aeropostale over the past several months. In the next post, we'll do just that.  

Aeropostale (ARO): Part Two - Valuation Thoughts

In the last post we looked at several assumptions that, when accepted as more-likely-than-not correct, paint a picture of Aeropostale as a business with a market that's not going away and seems to benefit from barriers that would prevent competitors from encroaching. On those fronts, it shows real promise as an investment opportunity.

Now I want to look at a few scenarios for thinking through its value. When we initiated our ownership position in late-October, ARO was trading just below $14 per share. Today it's nearer to $18, and we'll use the more current price in considering a range for its intrinsic value.

For the following, we'll use these given assumptions:

1. ARO has 4.2 million square feet of retail space across its store base,

2. It will be taxed at 40 percent of operating income,

3. It has about 82 million shares outstanding, and

4. Its current share price is $18.

Now, let's plug in some variables to see how the business looks on a price-to-value basis.

Scenario I: Ugly Forecast   

Here we assume that ARO continues taking a licking. Sales drop 20 percent from its 2010 high of $626 per square foot. Gross margins match the lowest of the previous ten years. And SG&A expense are the highest in ten years.

At the conclusion of a year with this sort of performance, the market would certainly punish ARO further. Unless this was clearly the bottom of the business challenges, the drop may be warranted. If it were the bottom, it would represent an excellent buying opportunities if we continued to believe in the assumptions from the previous post (i.e., that ARO is a strong business in a niche market with long-term protection against competition). Indeed, the scenario reflects what happened to ARO in October 2011 when it dropped to a long-time low of a little more than $9 a share.

So, if performance continues to deteriorate and key metrics revert to multi-year lows, there is easily the potential that the investment loses 50 to 60 percent of its value. That would hurt, but (again) if you believe in the assumptions, you would see this as a temporary set-back and an opportunity to jolt your returns by buying more shares.

Scenario II: Rosy Outlook   

Here we assume that ARO reverts to its best performance over the previous ten years, though I only assume a 12-15x multiple on its earnings. (It's worth highlighting the multiple is probably on the low side of what the market would do. I believe buyers would be energized and optimistic, rewarding the stock with an multiple around of 20x.)

That might be fun to think about, but I wouldn't put money behind it. Suffice it to call this the upside scenario and leave it there.

Scenario III: Going By the Averages   


Not too hot, not too cold. This is the one I put my money behind. The plug-ins are conservative based on ARO's actual average performance over the past six or so years. The multiple range remains below the S&P 500 market average, despite ARO being a business with excellent returns on invested capital and a history of compounded earnings growth in excess of 20 percent.

To build an intrinsic value calculation for the business, I would take this "averages scenario" with $191 million net income, grow earnings at a conservative five percent compounded, and pay out excess cash in the form of share repurchases. At an 11x earnings multiple, it's not difficult to see the business worth nearly $60 per share five years out for a 27 percent compounded gain on the investment.

If you have the conviction that your analysis is roughly accurate and your assumptions sufficiently conservative, $18 per share is a great place to build a position.

Aeropostale (ARO): Part One - The Assumptions

We started building a position in Aeropostale in late-October but I'm just getting around to writing about it now as I'm revisiting the thesis and thinking through potential risks. Let's just jump in...

Assumption 1: The Mall Continues to Deliver Captive Market of Teenage Consumers

The mall is now and will remain fertile sales territory for retailers peddling wares to teenagers. While the traffic counts might ebb and flow a bit, these temples for commerce seem to have staying power with the teen demographic. In short, we assume they will continue going to malls and doing so with varying amounts of cash in their pockets.

Assumption 2: Mall-Based "Value" for Teen Clothing is a Distinct Market Niche

There is (and will continue to be) an important role for the value option for clothes in the mall. Indeed, it is a distinct market niche - as compared to full-price teen clothing retailing or value options outside of the mall - with its own set of unique properties and dynamics to consider.

If a business can offer acceptable alternatives to the full-cost, popular brands - and do so at significantly lower price points for similar lines of apparel - it will find eager buyers among teenagers and their parents. If nothing else, this is the point Aeropostale has proven in its 25 or so years of operations.

But it's a very difficult niche to fill. It requires a significant degree of operational precision from mimicking the popular designs, to executing quick supply chain turnaround, to driving inventory at incredibly high velocity in stores to overcome high rent expense.

ARO fumbled its way into this niche, as its history attests. It was a concept spun-off from a private label Macy's sponsored for affordable teen fashion. But it took more than a decade of trial and error (and continual capital investment) to truly understand the market opportunity and to refine the business operations to allow the company to exploit the niche through efficient operations.


Assumption 3: ARO is (and Will Remain) the Most Efficient Operator


The ARO business model can feel like a high wire act requiring constant operational precision to pull it off. Its gross margins hover around 35 percent versus about 65 percent for full-cost competitors like Abercrombie & Fitch. That means ARO must keep its operating expenses extraordinarily low (about 20 percent of revenue) to produce operating income comparable to full-cost competitors that spend upwards of 50 percent on SG&A.

The fat gross margins of the high-cost fashion brands hide all sorts of sins in bad inventory decisions, too much high-rent floor space, and expensive design departments. With 65 percent gross margins, you can make mistakes and get away with it. It's much more difficult to do that with 35 percent gross margins.

ARO chose this path and has spent years honing its operations to thrive in it. But the high wire act requires constant operational refinement with little room for error. We assume that ARO can continue as hyper-efficient operators of its business.


Assumption 4: ARO Will Face Pricing Challenges & Must Have Resources to Survive Battles

Assumptions one and two posit that ARO has a business whose market is not going to just suddenly disappear. Assumption three is an article of faith that ARO can keep on the same track that has provided its success to date.

Even though we assume ARO is not a direct competitor of the full-cost retailers (in some ways I'm mincing terms here, but let's go with it), that does not mean its immune to the problems they will inevitably face with their own inventory choices. In other words, ARO has to possess the financial wherewithal to lose sales when the full-cost brands miss on their merchandise selection and are forced to clear inventory at fire sale prices.

Indeed, this is the exact scenario ARO has faced for the past several quarters. When Abercrombie and American Eagle have to clear inventory, the value buyers go upstream for the deals. The full-cost brands try to avoid this, but it happens and most recently it has persisted through multiple selling seasons. Their businesses are hurting and the ripple effect has hit ARO.

ARO must have a very conservative balance sheet with little to no debt and sufficient cash resources to weather multiple seasons of bad sales. It's painful. But at some point it goes away and the selling environment normalizes. If ARO survives the onslaught, it rebounds nicely to previous levels of sales and earnings levels.

ARO saw its cash balances dwindle over the past year as it was forced to clear inventory at a cash loss then turnaround and buy more inventory for the following season. Management suspended its share repurchase program to conserve cash. They doubled-down on expense management. And while they never experienced an accounting loss, they benefited from maintaining sufficient cash on hand and no long-term debt.

They were positioned to weather the storm (which, frankly, may continue into the foreseeable future) and will have to maintain a similar philosophy for the inevitability of the full-cost retailers having inventory trouble again at some point.


Assumption 5: ARO Seems Protected Against New Value Entrants Into the Malls 

While I believe the occasional pricing challenges from the full-cost brands is the bigger threat, we must also consider the possibility of new value-based competition trying to elbow its way into this niche mall market.  I believe the ARO business works in part because it enjoys exclusive access to the niche. If a viable value competitor emerged, ARO would be a much less attractive long-term investment.

This exercise is about considering ARO's competitive advantages against new players. The over-arching barrier to entry is, I think, ARO's low-cost + low-expense + high-volume model. This is not the most attractive market to build a business (as discussed previously, the margin for error is incredibly slim) especially when a successful incumbent exists and would not cede market share without a fight.

That being said, I see these categories for hypothetical entrants...

A. The Fanatical Entrepreneur (see this previous post)

An entrepreneur with a fierce Sam Walton streak decides to start from scratch to compete for the value-minded customers in ARO's niche. It's easy to assume a rational businessman would take little interest such a venture. The margins are too thin. The infrastructure costs are too high to reach profitability without inviting a counter attack from ARO - an attack ARO is much better positioned to endure.

A rational agent is much more likely to compete with full cost brands, doing battle on fashion tasktes in hopes of commanding premiums, rather than fight it out with ARO on low price and efficiency.

Even a well-financed fanatic would struggle. My cursory analysis of the early ARO years suggests that the company needed almost ten years and more than 150 stores before it reached scale and enjoyed profitability. And that was in an ideal competitive environment (i.e., it wasn't going toe-to-toe with a competitor all too eager to do it grievous harm).  At $500,000 per mall store in capex investment, the fanatic would have to be prepared to put to work somewhere in the area of $75 million of capital over a period of up to ten years before seeing profit.

Conclusion: Low risk due to low probability of a competitor - even a fanatic - succeeding in this.

B. The Adjacent Mover

It's much more likely that an existing value retailer - already structured to operate with thin margins and high volumes - would attempt an adjacent move into the teen market to compete. Indeed, this is modus operandi for clothing retailers that are saturating their core brands...they look for a new "concept" store to open new markets and drive growth while taking advantage of existing infrastructure for back office, management experience, supplier relationships, etc to reach profitability quickly.

Gap did this with Old Navy (though in off-mall real estate), Abercrombie with Hollister, American Eagle with Aerie, and ARO is in process of doing it with P.S. for Kids.

While we have to view this as a constant threat, there still exists the protection of how unattractive margins would be for taking this investment risk. There is no obvious player well-positioned to do this, but I see it as the biggest threat for a new entrant.

Conclusion: Moderate risk but low probability due to ARO's entrenched position and likelihood to fight back.


C. The Desperado

Call this a corollary of the Fanatic...the Desperado might be a full-price clothing retailer whose very existence is threatened because it business model doesn't work in its current market. It knows clothing and teenagers, but can't manage to stay at the cusp of new fashion trends. It has infrastructure. It has real estate. It has seasoned management. But it needs a new model.

Would it choose to go up against ARO? I'm skeptical. Again, the margins make the market less attractive. Plus the need to maintain such efficient operations would cast doubt on any management team unable to compete in a wider margin business.

However, it was truly desperado, it might make a stab at it...transitioning its mall real estate quickly, leveraging its existing suppliers, and using current technology. And while its long-term prospects would be dubious, it could certainly make ARO's job harder for a time.

I suspect the outcome would be similar to the full-price brands dumping inventory into the value niche through clearance sales. It would be painful for ARO, but conditions would eventually normalize.

Conclusion: Moderate risk but low probability of it happening.

Okay, there are the core assumptions in a painfully long post. Next we'll look at some ways to think about valuing ARO.