AMZN - Amazon.com Inc.

[mrholty]

Are there any precedents for exceptional companies that grew very quickly, dominated their industry, and had a "lost decade" solely because they became overpriced? That list is very long.

 

This is exactly my thinking. 

 

I dabble a little selling on Amazon FBA.  Just 3 products that I import from China (I sell a PL item) alond with 3 others who all buy from the same factory and undercut the namebrand.  I was the last of the three and have the lowest sales.  I keep weekly rankings of my items and there has been no major shift for me in the past 3-4 months. (By itself this is unsurprising since the items are in toys and toys are notorious for being weighted around back to school and esp Christmas.  However, my sales are down 30% YOY in the past 6 weeks.  With my ranking relatively flat that tells me that volumes in this area are down. 

I don't know if this is a negative indictment on Amazon (free shipping went from $35 to $49) so maybe its just them or maybe the consumer is slowing down.  Just an insight I'd throw out there.

 

[Patmo]

why haven't Warren and Charlie (or even Ted and Todd) ever bought AMZN shares for Berkshire? Could it be that these extremely competent investors do not consider it a promising long-term investment?  ::)

No offense but that's a worthless argument.

This kinda slavish, blind adulation of well-known investors is what led people to end up going long VRX or ZINC.

 

You could have all of this argument in the whole world condensed into one cube. It would fit into your living room. What are you gonna do with it, you can fondle it, but it produces nothing. Instead you could have 10 exxon valdez and all the corn fields in the USA. What's it gonna be, a fondling cube or 10 exxon valdez? This argument ultimately has no value because it produces nothing, even though lots of people are willing to buy it.

[Liberty]

why haven't Warren and Charlie (or even Ted and Todd) ever bought AMZN shares for Berkshire? Could it be that these extremely competent investors do not consider it a promising long-term investment?  ::)

No offense but that's a worthless argument.

This kinda slavish, blind adulation of well-known investors is what led people to end up going long VRX or ZINC.

 

You could have all of this argument in the whole world condensed into one cube. It would fit into your living room. What are you gonna do with it, you can fondle it, but it produces nothing. Instead you could have 10 exxon valdez and all the corn fields in the USA. What's it gonna be, a fondling cube or 10 exxon valdez? This argument ultimately has no value because it produces nothing, even though lots of people are willing to buy it.

 

I'll allow it.

[Graham Osborn]

It's good to see that FANG is going strong again.  These are good companies with apparently decent moats (not so sure about NFLX but I'm sure others would disagree).  Let's dump everything else we own and buy them!  Which, of course, is what happened last time.. and the time before that.. and the time before that.. you get the picture.  I guess the people I told that to 6 months ago were wise to ignore me ;)

 

As to what Warren and his 3 T's are doing, I couldn't care less.  Having spent the last year scratching my head at various Berkshire buys, I've given up trying to establish a correlation between BRK buying and personal profit.  When BRK announces a buy in a "hot" name these days, I view it as just another sell signal - just as I would if Ackman or Icahn was the one buying.

[prevalou]

The question is: is AMZN price expensive ?

 

Amazon is impossible to value with any degree of accuracy. On a purely objective basis (using GAAP earnings), it has a history of not earning its cost of capital. So it is perhaps worth book value of $31. The present value of Chamath's $3 trillion is $2400. I am pretty confident it is worth between $31 to $2400.

 

This is where the concept of the outside view can help. At $700/share, you need a combination of sustained high revenue growth (which is very rare for a company this large) combined with exceptional margins (for a retailer). Are there any historical precedents for the implied growth and margins? None that I am aware of.

 

Are there any precedents for exceptional companies that grew very quickly, dominated their industry, and had a "lost decade" solely because they became overpriced? That list is very long.

 

--

To get a good return from the current price, you are literally betting on something completely unprecedented. It certainly seems plausible.

 

Base rates use by investors is what gives opportunities. They tend to generalize and there lies the mispricing.

 

"markets  are efficient" is the outside view, so lots of people won't try to beat it and go ETFs. There lies the opportunity

 

The outside view tells Amazon has to be compared to a retailer. But does a retailer invest 10% of sales in R&D ?

 

[Palantir]

The bear case for AMZN is that the addressable market is smaller than thought or competitors compete effectively such as WMT Online in the US or Flipkart in India, resulting in growth slowing down. The high valuation underperfoms/historical argument is nonsensical, and if AMZN achieves its growth it will be rendered moot. The only thing that matters is how AMZN performs, not what other stocks did in the past.

[KCLarkin]

Amazon doesn't spend 10% of revenue on R&D. Read the description of "Technology and Content". It includes buying and merchandise selection along with AWS operations. I don't think it is an accident that they lump a bunch of operational expenses in this category.

[prevalou]

Amazon doesn't spend 10% of revenue on R&D. Read the description of "Technology and Content". It includes buying and merchandise selection along with AWS operations. I don't think it is an accident that they lump a bunch of operational expenses in this category.

 

Technology costs consist principally of research and development activities including payroll and related expenses for

employees involved in application, production, maintenance, operation, and platform development for new and existing

products and services, as well as AWS and other technology infrastructure expenses. Content costs consist principally of payroll

and related expenses for employees involved in category expansion, editorial content, buying, and merchandising selection.

Digital media content costs related to revenue recorded gross, including Prime Video, are included in cost of sales.

 

If it is not R&D what is it ? Is it typical of a traditional retailer ? By the way, it's more than 10 % of revenue

[KCLarkin]

 

Technology costs consist principally of research and development activities including payroll and related expenses for

employees involved in application, production, maintenance, operation, and platform development for new and existing

products and services, as well as AWS and other technology infrastructure expenses. Content costs consist principally of payroll

and related expenses for employees involved in category expansion, editorial content, buying, and merchandising selection.

Digital media content costs related to revenue recorded gross, including Prime Video, are included in cost of sales.

 

If it is not R&D what is it ? Is it typical of a traditional retailer ? By the way, it's more than 10 % of revenue

 

I have struck through the items that aren't really R&D. Other companies would label this category as "operating expenses". The biggest cost seems to be the depreciation of servers and other hardware.

 

Again, I don't think it is accidental that they used the phrase "technology costs consist principally of research and development activities" and then went on to describe activities that aren't R&D. They want you to treat operating expenses as if they are R&D.

 

--

For 2015, 2014, and 2013, we capitalized $642 million (including $114 million of stock-based compensation), $641 million (including $104 million of stock-based compensation), and $581 million (including $87 million of stock-based compensation) of costs associated with internal-use software and website development. Amortization of previously capitalized amounts was $635 million, $559 million, and $451 million for 2015, 2014, and 2013.

 

This would be the best available estimate of the "real" R&D expense.

[prevalou]

 

Technology costs consist principally of research and development activities including payroll and related expenses for

employees involved in application, production, maintenance, operation, and platform development for new and existing

products and services, as well as AWS and other technology infrastructure expenses. Content costs consist principally of payroll

and related expenses for employees involved in category expansion, editorial content, buying, and merchandising selection.

Digital media content costs related to revenue recorded gross, including Prime Video, are included in cost of sales.

 

If it is not R&D what is it ? Is it typical of a traditional retailer ? By the way, it's more than 10 % of revenue

 

 

I have struck through the items that aren't really R&D. Other companies would label this category as "operating expenses". The biggest cost seems to be the depreciation of servers and other hardware.

 

Again, I don't think it is accidental that they used the phrase "technology costs consist principally of research and development activities" and then went on to describe activities that aren't R&D. They want you to treat operating expenses as if they are R&D.

 

--

For 2015, 2014, and 2013, we capitalized $642 million (including $114 million of stock-based compensation), $641 million (including $104 million of stock-based compensation), and $581 million (including $87 million of stock-based compensation) of costs associated with internal-use software and website development. Amortization of previously capitalized amounts was $635 million, $559 million, and $451 million for 2015, 2014, and 2013.

 

This would be the best available estimate of the "real" R&D expense.

 

certainly AWS was created ex nihilo, and kindle, and Echo, prime video, etc.

[saltybit]

http://www.cnbc.com/2016/07/08/amazons-chinese-counterfeit-problem-is-getting-worse.html

[fareastwarriors]

http://www.cnbc.com/2016/07/08/amazons-chinese-counterfeit-problem-is-getting-worse.html

 

Yes it's a big problem. Even as a consumer, Amazon is overwhelming. I often buying the same thing from multiple sellers...It's frustrating when I just want one source, one seller...

[FourCastles]

  Up until a year-two ago, I was in the same camp as KCLarkin and others - I saw Amzn as an online retailer, albeit with huge technological advantage, running on 0.5% margin, and notorious for erratic quarterly earnings and maybe without shareholder focus. So I viewed it as risky and overpriced, with price dependent on the latest quaterly earnings which I could not predict. So I stayed away.

 

  With AWS, my view changed. AWS revenue is now 7bn (out of the total of ~100bn) and growing. I see the potential revenue of AWS in tens of billions over the next five years. Secondly, a retailer which can develop such a significant revenue stream in a totally unrelated field certainly has excellent execution capabilities and vision. So I decided not to attempt to guess what they will be good at in five-ten years, I just bought the stock and let's see how it goes. The continued development of the same-day shipping capability is also interesting. So they will have a good story to sell for at least several years, I would say.

 

  Overpriced as heck, for sure. But I just closed my eyes and bought. Now, I need to work on my entry points - I think I tend to buy at tops, as opposed to bottoms.

[cmlber]

Overpriced as heck, for sure. But I just closed my eyes and bought.

 

I think you meant to post this comment in the "Signs of a bubble" thread under General Discussion.  :o

[FourCastles]

  Lol, possibly, I won't claim I am immune to bubbles but I hope my buy is based on some sound reasoning. First, notice that I did read the annual reports and have been watching AMZN for some quarters. I believe their online business will be doing just fine over the next several years and the AWS revenue stream will continue increasing, justifying the extra price. Now, I would love to pick the stock up for a much lower price but I believe there is growth ahead and if there is a correction, I will hold the stock until it recovers. I do believe the fundamentals are there for it to recover.

 

  So, in my view, not bubble thinking but perhaps lack of patience to wait for a better entry point. Famous last words? Let's see in a couple of quarters.

[KCLarkin]

Up until a year-two ago, I was in the same camp as KCLarkin and others - I saw Amzn as an online retailer, albeit with huge technological advantage, running on 0.5% margin, and notorious for erratic quarterly earnings and maybe without shareholder focus. So I viewed it as risky and overpriced, with price dependent on the latest quaterly earnings which I could not predict. So I stayed away.

 

To be fair, that's not really my opinion on Amazon. I have no opinion on whether Amazon is overpriced. But I don't like the risk/reward.

 

Let's be optimists. Assume that Amazon reinvests all earnings. Amazon will grow 20% per year for 20 years. At the end of 20 years, net margins will be 7% (double Walmart's peak margins). The company will be worth 20x earnings.

 

In 2035, Amazon has $4 trillion in revenue. $287 billion in earnings. And a market cap of $5.7 trillion. Shareholders? They get 14% per annum (assuming 1% annual dilution).

 

Or, you could say that $AMZN is worth $1485 per share (10% discount rate).

 

--

 

Now, let's be more conservative. Assume 15% growth for 20 years. 3.5% margins (Walmart's peak margins). 15x PE.

 

Under these conservative assumptions, you get 4% CAGR. $AMZN is worth $237.80.

 

--

 

Now, let's be pessimistic. Assume 10% growth for 20 years. 3.5% margins (Walmart's peak margins). 15x PE.

 

Under these pessimistic assumptions, you get -0.6% CAGR. $AMZN is worth $97.75.

 

--

 

So Amazon is worth $98 to $1485 per share. It's not overpriced. But the valuation risk is extreme.

 

 

[AccentricInv]

Don't forget AWS though, which has much higher margins. Should skew it up a bit.

 

Also funnily, maybe you're onto something. It's trading right in the middle of your range at $740.

[KCLarkin]

Don't forget AWS though, which has much higher margins. Should skew it up a bit.

 

Yes, this is the "x" factor. I'm guessing the consolidated net margins will be 3-8%. I doubt that AWS net margins will be higher than 10%. But you could make an argument that they will use the platform to offer higher margin services.

 

Let's use the average of Costco's margins (2%) and IBM's margins (15%). Even with blended margins of 8.5%, you need 15% annual growth for 20 years and a 20x terminal multiple to justify the current price.

[cmlber]

Don't forget AWS though, which has much higher margins. Should skew it up a bit.

 

Also funnily, maybe you're onto something. It's trading right in the middle of your range at $740.

i

 

On the other hand WMT had huge barriers to entry that are non-existent here.  If you benchmark to WMT returns on invested capital (20%, probably very generous) margins would be closer to 2%.  So 3.5% for the retail side of the business is probably too high.

[FourCastles]

  KCLarkin, completely agreed on the valuation risk argument. I guess it is very difficult to predict the rate of growth or the margins and so I would not necessarily use discounted cash flows to value, it is just too uncertain (as your great analysis shows). AMZN is transforming from an online retailer to a conglomerate with AWS (a very different beast, who knows what margins will be possible and what growth will be possible) but, even more importantly, with a real possibility of coming up with another AWS-like idea which may change the game five-ten years down the road. After all, Walmart is not experimenting with drone delivery, fully automated warehouses etc. Technology meets retail and that offers hope there will be breakthroughs which are beyond the likes of Walmart and leads to the current PE of 100-300. For instance, what if they improve their user interface to the point where it is augmented reality/virtual reality, and shopping there is easier than shopping at Walmart? Combined with same day delivery - bye bye, Walmart. But, indeed, how do you value that? Anybody's guess.

 

  On the risk/reward side, I agree as well, I would like to see more reward. I guess  in my mind, the probability of the reward scenario is higher than the risk scenario, and that's why I bought. Would love to pick up the stock at a better price to have more reward room but how likely is that? If there's a good correction over the next year, we may see a 30% drop. If there is not, we may see 30% growth. Do I wait for a drop or not? That's what I meant by saying I decided (consciously) not to wait for a better entry point. 

 

  Historically, AMZN has been returning 25% annual return over the last fifteen years. I can see them doing the same over the next five-ten years.

 

 

 

 

[dwy000]

  KCLarkin, completely agreed on the valuation risk argument. I guess it is very difficult to predict the rate of growth or the margins and so I would not necessarily use discounted cash flows to value, it is just too uncertain (as your great analysis shows). AMZN is transforming from an online retailer to a conglomerate with AWS (a very different beast, who knows what margins will be possible and what growth will be possible) but, even more importantly, with a real possibility of coming up with another AWS-like idea which may change the game five-ten years down the road. After all, Walmart is not experimenting with drone delivery, fully automated warehouses etc. Technology meets retail and that offers hope there will be breakthroughs which are beyond the likes of Walmart and leads to the current PE of 100-300. For instance, what if they improve their user interface to the point where it is augmented reality/virtual reality, and shopping there is easier than shopping at Walmart? Combined with same day delivery - bye bye, Walmart. But, indeed, how do you value that? Anybody's guess.

 

  On the risk/reward side, I agree as well, I would like to see more reward. I guess  in my mind, the probability of the reward scenario is higher than the risk scenario, and that's why I bought. Would love to pick up the stock at a better price to have more reward room but how likely is that? If there's a good correction over the next year, we may see a 30% drop. If there is not, we may see 30% growth. Do I wait for a drop or not? That's what I meant by saying I decided (consciously) not to wait for a better entry point. 

 

  Historically, AMZN has been returning 25% annual return over the last fifteen years. I can see them doing the same over the next five-ten years.

 

 

 

 

 

But if you can't put a value on it aren't you basically just investing based upon the greater fool theory?  Feels more like gambling than investing.  Good luck.

[KCLarkin]

So 3.5% for the retail side of the business is probably too high.

 

Yes. To be clear, my scenarios were using blended margins (1st party retail, 3rd party retail, Prime, digital, AWS, devices, white label products...).

 

--

 

Anyway, here is another way to look at Amazon. What is Par for Amazon? In other words, what combination of growth and margins gets a buy-and-hold investor 10% annualized?

 

Let's set Par at:

18% revenue growth

5% margins

20 year runway

20x terminal P/E

1% annual dilution

100% reinvestment

 

Amazon needs to outperform at least one of those assumptions before it is a "winner". You are playing one very challenging golf course.

 

[spartansaver]

So 3.5% for the retail side of the business is probably too high.

 

Yes. To be clear, my scenarios were using blended margins (1st party retail, 3rd party retail, Prime, digital, AWS, devices, white label products...).

 

--

 

Anyway, here is another way to look at Amazon. What is Par for Amazon? In other words, what combination of growth and margins gets a buy-and-hold investor 10% annualized?

 

Let's set Par at:

18% revenue growth

5% margins

20 year runway

20x terminal P/E

1% annual dilution

100% reinvestment

 

Amazon needs to outperform at least one of those assumptions before it is a "winner". You are playing one very challenging golf course.

 

And just for fun to take KCLarikin assumptions a step further, Amazon on a GMV basis controls 4-6% of the US retail market. If you grow the US retail market at 4% for the next 20 years Amazon will control 50-75% of the entire US retail market. I know growth will come from other regions, but the majority of Amazon's sales are currently residing in the US and it is much tougher to grow outside of your home market.

[KJP]

A question for Amazon followers:  Has the company outlined anywhere its long-term plans for the third-party SVOD subscription add-ons that are available through Amazon Prime?  It seems like nurturing third-party content providers, particularly in niche areas, could be a great (and very high margin) business that leverages Amazon's existing customer relationship and billing infrastructure without requiring significant additional investment in content.  There are also clear cross-selling opportunities for AWS related to hosting all of the third-party streaming content.  Leveraging its existing infrastructure to capture both infrastructure spending and royalties from third-parties in this fashion seems very similar to the third-party marketplace/FBA business. 

 

I'm asking because I'm looking into a company that offers a niche, subscription add-on through Amazon Prime, and I'm curious whether, at the end of the day, Amazon is likely to be a partner or a competitor. 

[moustachio]

Let's be optimists. Assume that Amazon reinvests all earnings. Amazon will grow 20% per year for 20 years. At the end of 20 years, net margins will be 7% (double Walmart's peak margins). The company will be worth 20x earnings.

 

In 2035, Amazon has $4 trillion in revenue. $287 billion in earnings. And a market cap of $5.7 trillion. Shareholders? They get 14% per annum (assuming 1% annual dilution).

 

Or, you could say that $AMZN is worth $1485 per share (10% discount rate).

 

Maybe its just me, but I don't think there are many large companies that you could do a discounted cash flow analysis at a 10% rate for 20 years and expect the stock to still look cheap, especially in this current market.