Amazon’s scorched earth policy

I recently explained my reasoning for avoiding an investment in Amazon. In short – investing in Amazon requires the belief that something is going to magically change in the future. Somehow they will be able to convert sales into profits in a way that they have been unable to do for their entire 19+ years of existence as a public company.

Sales vs. profits

Amazon has been growing sales at a very impressive pace over the last ~20 years. Here’s a graph of their sales and profits from 1995 to 2013.

Look at that sales growth! That is what is driving Amazon’s valuation – their extremely high sales growth. The only way that Amazon could grow sales that fast is by taking market share from competitors. This means that sales that would have gone to, say, Macy’s instead went to Amazon.

But look at the profit numbers – as sales have gone up profits haven’t moved. Amazon is generating the same amount of profit from $70B in sales that they generated from $10B in sales. That means that Amazon is making LESS money from each sale than they used to (that is, their margins are dropping). This is exactly the opposite of what you’d expect a maturing company to do. Typically profit margins improve as a company grows. It benefits from improved economies of scale, better grand recognition, the ability to amortize fixed costs across more products being sold, etc.

Taking market share from competitors

Here’s what’s so interesting about Amazon’s market share growth – Amazon is essentially buying market share from their competitors by running their entire retail business for zero profit. In some quarters they make money on their retail business and in some quarters they lose money. In the last 5 years their lowest profit margin was -2.12% for the quarter ending Sept, 2014. Their highest profit margin was 2.82% for the quarter ending June 2016. On average their profit margins the last 5 years have been .33%.

Amazon’s competitors sell products and make profits from those sales. By taking market share away from their competitors Amazon now makes those sales but makes no profit from them.


The results are clear. Macy’s is shuttering stores. Wet Seal is going out of business. Aeropostale is shutting 100 of their 800 stores. Sports Authority has declared bankruptcy and is in liquidation. Pac-Sun filed Chapter 11.

Amazon is destroying the profitability of the entire retail industry.

The counter argument

I sometimes hear Amazon supporters say that this is all part of Amazon’s plan and that, with enough sales, profits will eventually follow. However, Amazon has been doing this for their entire 19 year history as a publicly traded company. This isn’t a temporary tactic. They are permanently driving the margins in the retail business towards 0%.

And what about the argument that Amazon is not turning a profit because they “choose to”? This argument goes something like this – the only reason Amazon isn’t generating a huge profit today is because they are investing in the future. Amazon is building out warehouses, improving delivery services, and experimenting with drone delivery so they can one day have same-day delivery. If Amazon wasn’t making those investments then they’d be generating higher profits.

The reality is that this is just a variation of “one-time charges” argument that so many companies make. They have a restructuring in one year that costs a lot of money, but the company argues that was a “one-time” charge that should be ignored when valuing the company. The next year they have to buy a new factory. Again, that’s a “one-time” charge that should be ignored. The third year they do yet another reorganization for another “one-time” charge. When “one-time” charges happen year after year they stop becoming “one-time” charges and start becoming the normal cost of doing business.

When Wal-Mart or Target pay to open a new store nobody considers those “one-time” charges. They are the normal cost of doing business and investing to increase future sales. The same thing applies to Amazon’s continuous

Similarly, Amazon has “one time” charges year after year. If they stopped making these “one time” charges then they’d likely see future sales growth stall, in the same way that if Target or Wal-Mart stopped opening new stores they’d see their growth suffer.


I suspect that Amazon will be a case study in MBA classes in 50 years. Whether it’s a cautionary tale of the importance of profits or an example of a total paradigm shift remains to be seen.

Amazon is putting their smaller competitors out of business but getting nothing in return. They are taking sales away from other companies in the retail sector but not profiting from those sales. What’s Amazon’s end game? Are they hoping to eventually have enough market share that they can raise prices and start reaping huge profits? Are they just running their retail operations as a charity?

I continue to love Amazon as a customer and hate it as an investor. If they want to price products so low that they can’t make a profit then that’s fine by me – I’ll continue to buy their products and avoid their stock.


Readers – what do you think about Amazon’s effect on the retail industry? Do you think they’ll be able to start converting sales into profits in a way they haven’t been able to do for the last 20 years? What’s their endgame?



9 thoughts on “Amazon’s scorched earth policy

  1. It’s always been interesting to me how two people can look at the same data and reach nearly opposite conclusions. Amazon is a very polarizing stock it seems, there are many people with your viewpoint and many whose viewpoint is more similar to mine.

    We tend to glorify individuals who save nearly everything and reinvest, but a lot of people don’t apply the same standards to publicly traded companies. For some reason, a lot of investors seem to feel a lot more confident with companies either returning profits to investors, or hoarding cash, rather than companies who invest the excess. For example, Apple is sitting on $216 billion in cash! I absolutely love Apple as a company, but I’ve always been scratching my head on why they’re not investing that money into either new products, ventures, or improving their cloud services. There are several companies that pay dividends with excess profits, which is essentially saying to the investor that the investor can find higher returns than the company can. What I find fascinating about Amazon is how they invest everything back into the growth of the company and how much growth these investments have returned. It is astonishing to me.

    At any point in the past, someone could have asked the same question of why Amazon isn’t profitable and how they should consider going after a higher bottom line rather than reinvesting. But in retrospect, it’s easy to see what would have happened to Amazon had they made this decision. For example, what if they stopped with books? What if they never rolled out AWS? What if after the Amazon phone, they’d shut down their consumer electronics division and didn’t release the Echo? What if they had never done Prime?

    They just keep innovating. It’s what defines the company and it’s the reason they’ll be able to stay ahead of their competitors.

    If you just look at nothing other than the AWS, it’s quite amazing. They don’t really have any competitors and their innovation is outpacing all of the other niche players. AWS runs 10x more cloud capacity than the next 14 players combined. Really it’s a two horse race with Microsoft being second, but I don’t really envision a case in which there isn’t room for two big players. Amazon has shown that they can continue to be innovative over the last ten years and I expect this will continue.

    Another thing that’s quite astonishing for Amazon is relatively low CPA for an online ecommerce company. Amazon is getting a lot of it’s traffic via DTI (direct type in). There are more product searches performed directly on Amazon than Google. This is quite unusual for an online ecommerce company. A lot of other online companies have a tremendously high CPA (cost per acquisition). A large portion of this could be attributed to the growth in prime members. Prime has had a 50% CAGR over the period from 10/12-2/16 and as of today has an estimated 80MM paying prime customers, or roughly 40% of US households. Prime members account for 53% of all Amazon purchases. Why? Mostly because of free shipping. Amazon has been expanding same day shipping and Prime members can get free two hour shipping on select items within 1000 cities. Prime is continuing to expand globally, so should see a lot of user growth in this area.

    Then there’s Prime video which rolled out globally late last year (except China). Amazon is now the third largest streaming service behind Netflix and YouTube. It’s pretty incredible that they were able to achieve this level of market share this quickly.

    Finally, there’s FBA, fulfilled by Amazon. I think this is one of the more interesting aspects of the business. It’s an entire ecommerce platform to sellers which allows them to take advantage of the entire Amazon ecommerce infrastructure including storage of items and shipping. In most categories, third party sellers account for > 80% of sales versus Amazon selling items directly. This is important for a couple different reasons that are incredibly important: content/selection and pricing. These are the two components that drive any two-sided marketplace. More selection and better pricing drive more consumers, which drive more sales, which drive more suppliers, which drive more selection and the loop continues. There is the thinking that Amazon is destroying margins, but in a lot of ways, it’s the Amazon platform that’s destroying margins because 3rd party sellers can take advantage of Amazon’s efficiencies from their numerous investments in this area. This is why Amazon had 36.9% of online holiday market share last year. To give you an idea of how this compares Best Buy was second with 3.9%.

    Then there are the moonshot bets Amazon is making: drone delivery (which is in testing right now, but will only be a matter of time before rolling out globally, perhaps 10 years) and their automated stores (Amazon Go, also in testing).

    Then there is the company’s skill set which is tremendously unique in the marketplace. They are experts in hardware, software and logistics which is proving to be quite a formidable skill set.

    I suppose it’s a matter of taste and investing goals to choose a company like Amazon over a company like Coca Cola. However, if you’re looking for growth, I’m not sure why you wouldn’t bet on Amazon. They are poised to take on even greater market share and unless others are investing at a much higher pace (unlikely), they’ll continue to crush it.

    1. Greg,

      Thanks for the detailed response. It’s clear that you’ve thought a lot about Amazon and their business model.

      I agree with everything you’ve written, but I think there’s one important thing missing in your analysis – profits.

      It doesn’t matter how many businesses Amazon moves into if those businesses don’t make money. Take Prime for example. Prime membership keeps growing (50% CAGR, as you pointed out). At $99/person you’d think Prime would be a huge source of profits. Here’s the problem – Prime members tend to order more stuff. That should be a boon, right? After all, more stuff equals more money. Unfortunately, “free” 2-day shipping isn’t free to Amazon. They are eating the $5/shipment cost of shipping. If somebody is placing 10 orders/year at $5/order for shipping and paying $99/year for Prime then Amazon would be making $49/year.

      Unfortunately, Prime members order more things and don’t care about grouping their shipments together to reduce shipping costs. After all, shipping is free to the customer. As a result, Amazon is actually LOSING money on Prime.

      I absolutely agree with you abut AWS – that business unit is driving profit for the entire company. But is that advantage sustainable over the long term? What can Amazon do that Google or Microsoft or IBM can’t? I don’t know the answer.

      Ultimately, Amazon has done a great job increasing sales but they haven’t been able to generate profits from those sales. And ultimately profits are the only thing that matter.


  2. Profits do matter for sure, but for Amazon, this is a deliberate choice to not be profitable, but rather plough all remaining cash back into capex and new businesses. Amazon is comprised of dozens and dozens of businesses, each with their own P&L. Some are very profitable, some are new and operating at a loss, which is not uncommon for new businesses. For each quarter, Amazon makes the deliberate decision of keeping net income flat and spending the remaining on either capex or new businesses. This deliberate decision is much different than some of Amazon’s businesses operating at an indefinite loss, which is not the case.

    This is all part of the original strategy, which (in my opinion) is exceptionally brilliant. Google “Jeff Bezos Napkin Sketch” for more info. The napkin sketch is all about growth in a closed-loop manner. The thinking was that growth would lead to lower cost structure, which would lead to lower prices which would lead to better customer experience which would lead to more traffic, which would lead to more sellers, which would lead to more selection, which would lead to better customer experience which would in turn lead to more growth. The loop then begins again. These network effects are exactly what have driven Amazon’s growth over the last 20 years.

    So, the capex expenditures goes back into more distribution centers and airplanes, which in turn improves prime, which in turn grows prime users, and the loop continues again.

    And as far as AWS goes, there’s a reason why Amazon runs 10x more cloud capacity than the next 14 players combined. It’s in large part because they’ve taken all of their profits over the last ten years and have reinvested nearly every dime back into the business–building out more and more data centers around the world. So what can Amazon do that Microsoft, Google, or IBM can’t? It starts with spending 10 years pouring profits into a massive capex infrastructure and doing it so well that it would be extraordinarily difficult to follow, both in capex terms, software infrastructure terms, and adoption. The switching costs are quite high for businesses. Once they’re up and running in AWS, all of their infrastructure is built on top of the Amazon API. It would be very costly for companies to up and switch. Google certainly has the cash required and expertise to go after Amazon, but I’m sure they’d have to pay quite a big premium for CPA to get users to switch. They’d also have about 10 years of API building to catch up on (which can’t be done overnight), and then all of the infrastructure tools would have to integrate with Google. If any, Microsoft is the only other one who has shown they are willing to invest, and they’ve invested heavily, but have still lost ground to Amazon.

    To give you an idea on the scale of the infrastructure, Amazon has an estimated 6.4 million servers worldwide, each in a data center with 102 Tbps bandwidth. The “napkin loop” applies to this situation too. More servers/infrastructure means economies of scale which means reduced infrastructure cost which means better pricing for customers which means more customers, etc. etc.

    So the question would be, if Amazon is deliberately choosing not to be profitable, in order to reinvest all cash into new businesses and capex, when would they cease to do this? This is anyone’s guess, but my personal guess is they will continue to do it as long as they possibly can.

    While unconventional, this has been a winning formula for Amazon for 20 years. It is why the stock has seen 46,800.14% growth over the last 20 years.

    1. Amazon’s policy over the last 20 years as a publicly traded company has clearly resulted in a lot of efficiencies. And whether they are a retail company, a logistics company, a technology company, or a web hosting company, at some point they need to make profits. Or, I should say, I want to invest in companies that make and grow profits. Amazon has shown that it can sell enormous volumes of products but it hasn’t shown that it can do so profitably.

      Maybe we should value them as a technology services company rather than a retailer. I am onboard with that thinking. But that still leaves the problem of Amazon’s PE ratio of around 165, which is insanely, ridiculously high.

      It’s entirely possible that you’ll be right and Amazon will be worth the current stock price. For me, I file it under “not interested”. The risk is too high at these prices.

  3. I think the point you may be missing is that Amazon is not a company for retailing of products, it is a pure logistics company.

  4. you are looking at the wrong numbers…. in the same chart instead of comparing with net income.. compare with ebitda instead… and then you will see why amazon market cap is shooting up…

    1. I’m curious why you think that is a more valid way to analyze Amazon than other income metrics? After all, interest and taxes are real exposes. Depreciation and amortization are less “real” in that they aren’t actually cash leaving the company, but I’m still curious why you think it’s best to exclude them from an analysis of Amazon. Should EBITDA be the metric used to evaluate all investments?

      As a side note – I absolutely agree with you that Amazon’s sales growth has been absolutely tremendous. I’m just saying that higher sales need to be accompanied by higher profits, otherwise what’s the point?

      1. EBITDA is the most accurate measure of a company’s ability to generate earnings. Most private equity funds use a “multiple” of EBITDA to decide what they’re willing to pay for a company.

Leave a Reply

Your email address will not be published. Required fields are marked *