Amazon (AMZN): Playing Offense or Defense? Part A. Subsidized Shipping

Reported Earnings Overstated - The Impact of Restricted Earnings
Owner earnings are those that can be extracted from the company for the benefit of shareholders (dividends, buybacks, debt reduction), plowed back into the business to create even greater earnings in the future (growth capex, investments in expense infrastructure, acquisitions, etc.) or held as surplus cash. They are "unrestricted" in that management has significant discretion on how to use them without damaging the current earnings ability of the business. 

Reported (GAAP) earnings do not discriminate between the portion of earnings that are unrestricted and the portion ("restricted") that management has no option but to plow back into the business just to keep things current. Examples are replacing obsolete equipment, refreshing old stores, responding to a competitor's pricing tactics, or ramping up customer service because clients are threatening to leave without it. They are all necessary investments, but they don't create incremental earnings. At best, they prevent the erosion of existing profits. 

Reinvestment of the restricted portion of earnings creates the unpleasant sensation of running to stand still. You can expend a lot of energy without taking the business anywhere. 

You won't find a line on the income statement called restricted earnings. As Buffett said in his 1986 letter to shareholders, determining which portion of earnings are unrestricted versus which are unrestricted is tricky. Indeed, it's quite different from industry to industry. For a manufacturer in a highly competitive market, a large chunk of its reported earnings may not be available to reinvest for growth or to pay out to owners. Why not? Because every five years it must spend tens of millions to retool factory assembly lines to accommodate new designs, to engineer more efficient production techniques, or to begin construction of new products that replace obsolete models.   

Reported Earnings Understated - Making "Productive" Investments in Expense Infrastructure

But the idea can cut both ways. Most businesses will report earnings that exceed the actual dollars available to benefit owners. But some businesses (particularly those in growth mode) will report earnings that dramatically understate the amount of cash being plowed back to grow future earnings ability.  For example, they may report $1 million earnings but in reality they plowed $10 million into marketing to acquire new customers that will produce more earnings power in the future. That is certainly the case with GEICO that Tom Russo talked about at the Value Investing Congress last year (and which we discussed here).  

In that scenario, do we value the business based on the $1 million in reported earnings? Or do we value it based on the $11 million it would have generated if not for the investment in acquiring more customers?

Well - no surprise here - it depends.

The dilemma is that we don't want to use this idea to rationalize investments by bloating the target company's earnings. It can be a slippery slope...you can argue with yourself to exhaustion trying to justify buying a company with a great growth story at its current high price-to-earnings ratio. One must err on the side of caution.

When considering this sort of situation, the first filter I might apply to the decision is how confident you can be that the increased expenses that lead to lower reported earnings are actually investments with the high likelihood of paying off in the future. And in this thought process, apply a high burden of proof on the company. One is wise to evoke the wisdom of Richard Feynman...The first principle is you must not fool yourself, and you are the easiest person to fool.

Another useful filter is to ask whether the increased spending (or whatever caused the reduced earnings) comes from the company playing offense or playing defense. This is important. Is the increased spending a result of the business understanding its competitive advantage and investing heavily in it? Or is it a reaction to a competitive move in the industry where, if the company doesn't respond, its business is harmed? 

In the former, it's likely (though not conclusive) that the company is spending in a manner that creates future value for shareholders (again, like the GEICO example) even if it reduces reported earnings today. It's fair to consider those investments as unrestricted earnings and count them among owner earnings when valuing the business.

In the case of the latter, the defensive spending, I would argue that this is likely an example of restricted earnings that are rightfully withheld from reported earnings.  

Amazon.com: Increasing Spend = Offense or Defense?

Let's bring this back to Amazon. The business clearly operates in a market with growing demand for its products and services. Year-in and year-out, Bezos et al must make educated guesses about consumer demand one-, two-, three-plus years in the future and invest in their infrastructure accordingly. Sure, they could stop those investments today by declaring their wish to optimize throughput of existing assets, pushing more sales across existing infrastructure (fulfillment centers, technology, marketing efforts, personnel, etc.) and probably create sizable profits for investors. But that would be choking the golden goose, seriously affecting its ability to lay more golden eggs in the future. Instead they build out in anticipation of what's to come.

Here are several categories of that type of investing...were they offensive or defensive in nature?

A. Subsidized shipping to pull more shoppers to the web and away from traditional retail.

In the beginning, Amazon treated shipping as a source of income. Later, its goal was making shipping a break-even proposition. Now, the company proudly uses shipping as a loss-leader, accepting the glad trade-off that quick-and-cheap shipping translates into wider consumption from customers who would otherwise give the business to Target or Best Buy.

From 2010 to 2011 Amazon plowed $1.1 billion into subsidized shipping, increasing its net shipping costs 76 percent...far above the additional shipping revenue that came in with 41 percent overall sales growth.

Should we expect this to change? No. This is the ultimate offensive move in two ways.

First, consumers are quick to tally shipping charges into the total bill when comparing costs of buying online versus bricks-and-mortar. Amazon recognizes that much of its growth will come from prying shoppers from trips to Wal-Mart and the mall. One way to achieve this is to provide a lower "landed" price than what they would get when getting in the car to shop with competitors. Shipping is part of that landed price, and Amazon is willing to invest in making it cheaper and cheaper for buyers.

It's important to note that shipping is included in Amazon's overall cost of sales calculations. In 2011, COS was about 78 percent of revenue. For Wal-Mart, COS was about 76 percent of revenue...and Wal-Mart has a tremendous overall advantage in its purchasing in that it carries far less selection and buys in volume that's easily 10x that of Amazon. Its COS should reflect much cheaper product acquisition costs. Yet its advantage over Amazon is negligible.

In other words, Amazon is earning comparable gross margins despite subsidizing shipping. As Amazon improves other drivers of its COS (e.g., volume purchases leading to product acquisition cost discounts), I expect it will subsidize shipping even more. And as Amazon builds more fulfillment centers nearer to its customers, its costs of shipping will go down.

One can easily imagine a day when Amazon subsidizes the full cost of shipping, retains product cost advantage over traditional retailers, and provides overnight (or even same day) delivery. All the while maintaining a gross margin sufficient to cover its operating expenses and provide a tidy profit.

Second, the subsidized shipping presents a formidable challenge to other online retail competitors. Amazon is the trend setter. The more they set the standard for low-cost shipping, the more consumers expect it in all online transactions. If the competitor cannot provide it - and the consumer can purchase the same or similar item from Amazon for a cheaper price - the competitor loses the business.

This creates a powerful barrier to entry. Smaller operators that can't match Amazon's scale (and none can) will only be able to subsidize shipping by charging a premium purchase price. And if the buyer can get the same item at Amazon...

(As an aside, online retailers that find ways to compete with Amazon in this regard - Quidsi's diaper.com and Zappos both come to mind - are quickly neutralized. Amazon offers their inventory, undercuts their prices, attempts to replicate their service advantages, or acquires them. See the Business Week story of Quidsi here. The moral of the story: Amazon is deadly serious about preventing other retailers from gaining a toehold in their business...they want complete web retailing ubiquity.)

Conclusion: Definitely offensive. An investment in long-term competitive advantage that hurts competitors, garners greater share of online and traditional retailing markets, and leads to accelerated scale benefits. 

This post is getting too long, so I'll split it up. Next, we'll consider whether lowering prices is an offensive or defensive move.