AMZN - Amazon.com Inc.

[rkbabang]

You guys all make the same mistake. If I could recreate every company out there with only the capital that has been put into it we'd have no Microsoft, no Apple, no company with any first mover advantage. This is not a valid argument.

 

+1.  I could give you $1T and you couldn't recreate Apple, nor Amazon, nor even Wal Mart.  Amazon could never outcompete Wal Mart by recreating Wal Mart.

 

You clearly haven't spent much time thinking about what $1T can do.  Here's how I would destroy AMZN with $1T.

 

1). I'd set aside $100 billion to make the 100 best employees at AMZN billionaires.  Good luck matching that comp Jeff.

 

2). I'd spend $30 billion replicating AMZNs PPE, this could be done in a couple years.

 

3).  I'd buy every Super Bowl ad slot for the next fifty years. This would only cost about $10 billion.  I'd also buy every Facebook ad for the next two years while I'm still unknown so that any time the 1 billion people using Facebook view a page, they see my ads and nothing else.  That'll cost me about $60 billion.

 

4). What id advertise on those ads is a pledge to give every American citizen $1,000 credit to spend, no strings attached, on my site where I would have the exact same merchandise as AMZN.  This would cost me $300 billion.

 

5).  I would provide free two day shipping to everyone, no questions asked.  This is probably a $100 billion NPV commitment.

 

6).  I would offer a low price guarantee. 

 

After doing all that, I have $400 billion left to spend.

 

Do you think anyone would still use AMZN on that scenario?

 

I can't wait to see it. From your calculations I deduct that you're no longer of the opinion that Amazon could be replicated with $20b?

 

OK, $1T was overstating my case, but as ni-co is pointing out the amount of money needed to replicate and replace Amazon would be way north of $20B.  Way more.

 

[cmlber]

You guys all make the same mistake. If I could recreate every company out there with only the capital that has been put into it we'd have no Microsoft, no Apple, no company with any first mover advantage. This is not a valid argument.

 

+1.  I could give you $1T and you couldn't recreate Apple, nor Amazon, nor even Wal Mart.  Amazon could never outcompete Wal Mart by recreating Wal Mart.

 

You clearly haven't spent much time thinking about what $1T can do.  Here's how I would destroy AMZN with $1T.

 

1). I'd set aside $100 billion to make the 100 best employees at AMZN billionaires.  Good luck matching that comp Jeff.

 

2). I'd spend $30 billion replicating AMZNs PPE, this could be done in a couple years.

 

3).  I'd buy every Super Bowl ad slot for the next fifty years. This would only cost about $10 billion.  I'd also buy every Facebook ad for the next two years while I'm still unknown so that any time the 1 billion people using Facebook view a page, they see my ads and nothing else.  That'll cost me about $60 billion.

 

4). What id advertise on those ads is a pledge to give every American citizen $1,000 credit to spend, no strings attached, on my site where I would have the exact same merchandise as AMZN.  This would cost me $300 billion.

 

5).  I would provide free two day shipping to everyone, no questions asked.  This is probably a $100 billion NPV commitment.

 

6).  I would offer a low price guarantee. 

 

After doing all that, I have $400 billion left to spend.

 

Do you think anyone would still use AMZN on that scenario?

 

I can't wait to see it. From your calculations I deduct that you're no longer of the opinion that Amazon could be replicated with $20b?

 

OK, $1T was overstating my case, but as ni-co is pointing out the amount of money needed to replicate and replace Amazon would be way north of $20B.  Way more.

 

That's a more reasonable statement.  But if you asked me to compete with Facebook with $1T, idk if I could.  Money can't buy 1 billion active users and the content they've all uploaded over the years.  Without content, it's useless to users, and without users, there's no reason for anyone to take the time to upload content. THAT is a moat.

 

Btw, my $20 billion number is actually substantially more capital than AMZN has put in the business.  Shareholders equity is only $13 billion. 

 

Since we agree $1T could kill AMZN, now we're only talking about how much capital would it take to compete, and have agreed it's somewhere between $20bn and $1T.  If you think my plan as excessive, we're somewhere between $20bn and $600bn.  I think the real answer is somewhere between $30bn and $100bn.  And there are many large retailers that will generate that kind of capital to spend over the next decade.  And if AMZN will generate enough cash flow to warrant a $1T market cap only capturing a single digit percentage of global retail sales (which you must believe, if you think this is a great long term buy), there's room for 10+ more AMZNs in the world.  So why wouldn't Walmart, Target, Tesco, insert large cash flowing retailer here, Google, Apple, insert other large cash flowing tech company here, etc. try to spend $100bn to create $1T? 

 

Google has said Amazon is a major competitor.  You don't think one of their moonshots could be a retail business?  They made a self driving car, you don't think they could build a warehouse that sorts packages and a website that takes orders?

[rkbabang]

You guys all make the same mistake. If I could recreate every company out there with only the capital that has been put into it we'd have no Microsoft, no Apple, no company with any first mover advantage. This is not a valid argument.

 

+1.  I could give you $1T and you couldn't recreate Apple, nor Amazon, nor even Wal Mart.  Amazon could never outcompete Wal Mart by recreating Wal Mart.

 

You clearly haven't spent much time thinking about what $1T can do.  Here's how I would destroy AMZN with $1T.

 

1). I'd set aside $100 billion to make the 100 best employees at AMZN billionaires.  Good luck matching that comp Jeff.

 

2). I'd spend $30 billion replicating AMZNs PPE, this could be done in a couple years.

 

3).  I'd buy every Super Bowl ad slot for the next fifty years. This would only cost about $10 billion.  I'd also buy every Facebook ad for the next two years while I'm still unknown so that any time the 1 billion people using Facebook view a page, they see my ads and nothing else.  That'll cost me about $60 billion.

 

4). What id advertise on those ads is a pledge to give every American citizen $1,000 credit to spend, no strings attached, on my site where I would have the exact same merchandise as AMZN.  This would cost me $300 billion.

 

5).  I would provide free two day shipping to everyone, no questions asked.  This is probably a $100 billion NPV commitment.

 

6).  I would offer a low price guarantee. 

 

After doing all that, I have $400 billion left to spend.

 

Do you think anyone would still use AMZN on that scenario?

 

I can't wait to see it. From your calculations I deduct that you're no longer of the opinion that Amazon could be replicated with $20b?

 

OK, $1T was overstating my case, but as ni-co is pointing out the amount of money needed to replicate and replace Amazon would be way north of $20B.  Way more.

 

That's a more reasonable statement.  But if you asked me to compete with Facebook with $1T, idk if I could.  Money can't buy 1 billion active users and the content they've all uploaded over the years.  Without content, it's useless to users, and without users, there's no reason for anyone to take the time to upload content. THAT is a moat.

 

Btw, my $20 billion number is actually substantially more capital than AMZN has put in the business.  Shareholders equity is only $13 billion. 

 

Since we agree $1T could kill AMZN, now we're only talking about how much capital would it take to compete, and have agreed it's somewhere between $20bn and $1T.  If you think my plan as excessive, we're somewhere between $20bn and $600bn.  I think the real answer is somewhere between $30bn and $100bn.  And there are many large retailers that will generate that kind of capital to spend over the next decade.  And if AMZN will generate enough cash flow to warrant a $1T market cap only capturing a single digit percentage of global retail sales (which you must believe, if you think this is a great long term buy), there's room for 10+ more AMZNs in the world.  So why wouldn't Walmart, Target, Tesco, insert large cash flowing retailer here, Google, Apple, insert other large cash flowing tech company here, etc. try to spend $100bn to create $1T? 

 

Google has said Amazon is a major competitor.  You don't think one of their moonshots could be a retail business?  They made a self driving car, you don't think they could build a warehouse that sorts packages and a website that takes orders?

 

Now that IS an interesting thought.  I have a hard time believing that Wal Mart or Target will make a real effort to take out Amazon.com.  It would be too disruptive to their current business.  If anyone could do it it would be a company like Google.  It has a brand everyone already loves, a website everyone already visits, it has the technical know how, and the cash to burn.  Certainly Google could more easily go into retail than Amazon could get into web searches. I've never heard anything about Google getting into retail in a big way however, would they be willing to spend the capital required to get into a fight like that?  Probably not, Google moonshots tend to be things no one else is already doing, or at least doing well.

 

 

[ni-co]

That's a more reasonable statement.  But if you asked me to compete with Facebook with $1T, idk if I could.  Money can't buy 1 billion active users and the content they've all uploaded over the years.  Without content, it's useless to users, and without users, there's no reason for anyone to take the time to upload content. THAT is a moat.

 

Btw, my $20 billion number is actually substantially more capital than AMZN has put in the business.  Shareholders equity is only $13 billion. 

 

Since we agree $1T could kill AMZN, now we're only talking about how much capital would it take to compete, and have agreed it's somewhere between $20bn and $1T.  If you think my plan as excessive, we're somewhere between $20bn and $600bn.  I think the real answer is somewhere between $30bn and $100bn.  And there are many large retailers that will generate that kind of capital to spend over the next decade.  And if AMZN will generate enough cash flow to warrant a $1T market cap only capturing a single digit percentage of global retail sales (which you must believe, if you think this is a great long term buy), there's room for 10+ more AMZNs in the world.  So why wouldn't Walmart, Target, Tesco, insert large cash flowing retailer here, Google, Apple, insert other large cash flowing tech company here, etc. try to spend $100bn to create $1T? 

 

Google has said Amazon is a major competitor.  You don't think one of their moonshots could be a retail business?  They made a self driving car, you don't think they could build a warehouse that sorts packages and a website that takes orders?

 

Still not a valid argument. Even if you "only" needed $100-200bn to recreate today's Amazon within a few years, Amazon would be 2-3x larger than today at this point and you'd need $100bn-$200bn more — it's a moving target. This is what makes it such a powerful business. I repeat: When Amazon was created there was no Amazon to compete against. You can't just assume this fact away.

 

This line of thinking leads nowhere. Not only is it impossible to estimate what it would cost to overtake Amazon while Amazon keeps growing and is competing against you. Even if you could, what's does the result tell you? That you shouldn't pay today's market cap? If it only were so easy! What does it help you when you come to the conclusion that you "only" need $100bn? There are very, very few companies out there that have enough financial firepower to spend $100bn just like so in 5-7 years. I don't even think that Google has enough firepower to do that — they have better things to do with their cash (hopefully). This is just not their competency.

 

If you want to think about threats to Amazon's business it would be far more fruitful to think about why they failed in China (so far). This is what I'm worried about. Not that CostCo, Walmart or Google enter/succeed in Amazon's business, online retailing—this opportunity is gone. They missed it 10 years ago.

 

Since you seem to like FB so much better: I would be worried about the size of the advertising market (its share of US GDP has been a very constant average of 1.3%, going back all the way to the 1920s) and the limits to FB's growth in the long term. FB doesn't have this open field in front of them. And they have to compete against Google. Zuckerberg rightly is very paranoid not to lose what he has built up. This is why he has to pay $16bn in a defensive move to buy WhatsApp and this is why Google has to spend huge amounts on Android development. So far, they both excel at it. But in the long term, it's a situation I like much less than the one Amazon is in.

[cmlber]

That's a more reasonable statement.  But if you asked me to compete with Facebook with $1T, idk if I could.  Money can't buy 1 billion active users and the content they've all uploaded over the years.  Without content, it's useless to users, and without users, there's no reason for anyone to take the time to upload content. THAT is a moat.

 

Btw, my $20 billion number is actually substantially more capital than AMZN has put in the business.  Shareholders equity is only $13 billion. 

 

Since we agree $1T could kill AMZN, now we're only talking about how much capital would it take to compete, and have agreed it's somewhere between $20bn and $1T.  If you think my plan as excessive, we're somewhere between $20bn and $600bn.  I think the real answer is somewhere between $30bn and $100bn.  And there are many large retailers that will generate that kind of capital to spend over the next decade.  And if AMZN will generate enough cash flow to warrant a $1T market cap only capturing a single digit percentage of global retail sales (which you must believe, if you think this is a great long term buy), there's room for 10+ more AMZNs in the world.  So why wouldn't Walmart, Target, Tesco, insert large cash flowing retailer here, Google, Apple, insert other large cash flowing tech company here, etc. try to spend $100bn to create $1T? 

 

Google has said Amazon is a major competitor.  You don't think one of their moonshots could be a retail business?  They made a self driving car, you don't think they could build a warehouse that sorts packages and a website that takes orders?

 

Still not a valid argument. Even if you "only" needed $100-200bn to recreate today's Amazon within a few years, Amazon would be 2-3x larger than today at this point and you'd need $100bn-$200bn more — it's a moving target. This is what makes it such a powerful business. I repeat: When Amazon was created there was no Amazon to compete against. You can't just assume this fact away.

 

This line of thinking leads nowhere. Not only is it impossible to estimate what it would cost to overtake Amazon while Amazon keeps growing and is competing against you. Even if you could, what's does the result tell you? That you shouldn't pay today's market cap? If it only were so easy! What does it help you when you come to the conclusion that you "only" need $100bn? There are very, very few companies out there that have enough financial firepower to spend $100bn just like so in 5-7 years. I don't even think that Google has enough firepower to do that — they have better things to do with their cash (hopefully). This is just not their competency.

 

If you want to think about threats to Amazon's business it would be far more fruitful to think about why they failed in China (so far). This is what I'm worried about. Not that Target or Walmart or Google enters there business—this opportunity is gone. They missed it 10 years ago.

 

Since you seem to like FB so much better: I would be very worried about the size of the advertising market (its share of US GDP has been a very constant average of 1.3%, going back all the way to the 1920s) and the limits to FB's growth. FB doesn't have this open field in front of them. And they have to compete against Google. This is a situation I like much less than the one Amazon is in.

 

First of all, I was being generous with my range.  I think Google or Walmart or Target or Apple could recreate Amazon for under $50bn today.

 

Second, I don't think Amazons "advantage" is growing.  General Motors had a scale advantage too until they got destroyed by the Japanese.  Amazon doesn't have a scale advantage.  Walmart is 5x larger and the package delivery, the only part of the value chain that benefits from material economies of scale, is outsourced.

 

Third, I don't think anyone will recreate Amazon exactly the way it is, why would they, it doesn't make any money.  Profit, not revenue or market cap, attracts competition.  But I expect every retailer on the planet to invest more in their ecommerce businesses over the next decade and they'll make life more difficult for Amazon.  It will be a big, profitable business.  But $1T market cap?  I doubt it.  I'd be shocked if Amazons stock price is higher than this in 10 years, setting aside AWS which looks like it is possibly a good business.

 

Fourth, I'm contrasting with Facebook because it was brought up as comparable to Amazon, which it is not.  I've never invested in Facebook so couldn't tell you anything intelligent about the valuation, but I can tell you I admire the business quality and think it is one of the widest moats that exists.

[LC]

Read this article recently:

https://stratechery.com/2016/the-amazon-tax/

 

Concludes that Amazon retail operations basically exist to fuel AWS.

 

I'm not sure I think AWS is so great. What are the switching costs? If someone gives me $50B and I build a huge datacenter, what advantage does AWS have?

[rpadebet]

Read this article recently:

https://stratechery.com/2016/the-amazon-tax/

 

Concludes that Amazon retail operations basically exist to fuel AWS.

 

I'm not sure I think AWS is so great. What are the switching costs? If someone gives me $50B and I build a huge datacenter, what advantage does AWS have?

 

LC,

MSFT, IBM, GOOG and now AAPL would ask you to get in line. These guys have tons more than 50b to spend but are still unable to replicate AWS success.

 

I think the key to AWS is to think beyond the hard drives and processors. It's a bit like the iPhone, android Eco system argument, you have software, services, developers and third party services on AWS which serves to keep clients locked in and creates a virtuous cycle. That is hard to replicate once it reaches a critical mass. Look at the guys above competing again, they have all done the ecosystem thing once and won in their respective fields/era's, yet they are losing here. It's not that easy!

[KJP]

Walmart is 5x larger and the package delivery, the only part of the value chain that benefits from material economies of scale, is outsourced.

 

I want to push back on both parts of this statement.

 

(1) Is Walmart really 5x larger in a way that would matter for economies of scale, assuming they exist?  I believe the relevant economy of scale we're talking about is in distribution -- more volume equals denser, more efficient distribution networks.  On the surface, Walmart appears to have a big advantage, because Walmart U.S. had $300 billion in sales last year (not including Sam's Club) while Amazon had only $63 billion in U.S. sales.  But Amazon's sales are growing much faster, and Amazon's distribution network increases it's throughput by providing fulfillment services to third-party sellers ("Fulfillment by Amazon" or "FBA").  Amazon does not record as revenue the gross amount of FBA sales; rather, only it only records as revenue the portion of the third-party sales it retains.  So, to make an apples-to-apples comparison of the volume of goods flowing through Amazon's distribution system versus Walmart's, you'd need to know the gross merchandise volume of FBA sales.  I haven't seen that number reported by Amazon.  But in its 4Q conference call, Amazon stated that third-party sales were about 47% of total units sold, and FBA constituted about 50% of third-party sales.  So, a very rough estimate would put the gross merchandise value (GMV) of Amazon's FBA sales at about 50% of its first-party sales.  If first-party sales were $26 billion in 4Q, that would put the GMV of FBA sales at around $13 billion.  All in, then, Amazon's 4Q first-party GMV and FBA GMV equaled about $40 billion.  In contrast, Walmart's GMV in 4Q was about $80 billion.  If these numbers are roughly right, then Walmart's U.S. distribution system appears to be only about 2x the volume of Amazon's today.  Moreover, the amount of GMV that Amazon distributes is growing very rapidly, particularly through FBA.  In comparison, Walmart's GMV is growing very slowly.  So, for how much longer will Walmart actually enjoy a scale advantage over Walmart in the U.S.?

 

(2) Why is "package delivery" -- which I assume refers to the movement of a package from a warehouse to the consumer -- the only part of a distribution system that benefits from economies of scale?  In other words, why are there no economies of scale at earlier stages of the distribution chain? 

 

More generally, what my comments highlight, I think, is the potential mistake in comparing Walmart to Amazon.  As Ben Thompson discusses in the article that has been posted several times on this thread, first-party retail sales on Amazon -- the business that competes head-on with Walmart, Target, et al. -- may never produce any profit.  But that business provides tremendous scale that provides the backbone for what the real profit centers will be:  services to third-parties through businesses like AWS and FBA.  That, to me, is the biggest problem with comparing Amazon to Walmart.  Amazon want's volume from its first-party sales; Walmart, Target, et al. need to make a profit from that activity. 

[plato1976]

Do we have many third party services on AWS now?

 

Read this article recently:

https://stratechery.com/2016/the-amazon-tax/

 

Concludes that Amazon retail operations basically exist to fuel AWS.

 

I'm not sure I think AWS is so great. What are the switching costs? If someone gives me $50B and I build a huge datacenter, what advantage does AWS have?

 

LC,

MSFT, IBM, GOOG and now AAPL would ask you to get in line. These guys have tons more than 50b to spend but are still unable to replicate AWS success.

 

I think the key to AWS is to think beyond the hard drives and processors. It's a bit like the iPhone, android Eco system argument, you have software, services, developers and third party services on AWS which serves to keep clients locked in and creates a virtuous cycle. That is hard to replicate once it reaches a critical mass. Look at the guys above competing again, they have all done the ecosystem thing once and won in their respective fields/era's, yet they are losing here. It's not that easy!

[gfp]

Netflix, Spotify Reddit, lots of government stuff - and thousands more. Is that what you were asking?

 

Do we have many third party services on AWS now?

 

Read this article recently:

https://stratechery.com/2016/the-amazon-tax/

 

Concludes that Amazon retail operations basically exist to fuel AWS.

 

I'm not sure I think AWS is so great. What are the switching costs? If someone gives me $50B and I build a huge datacenter, what advantage does AWS have?

 

LC,

MSFT, IBM, GOOG and now AAPL would ask you to get in line. These guys have tons more than 50b to spend but are still unable to replicate AWS success.

 

I think the key to AWS is to think beyond the hard drives and processors. It's a bit like the iPhone, android Eco system argument, you have software, services, developers and third party services on AWS which serves to keep clients locked in and creates a virtuous cycle. That is hard to replicate once it reaches a critical mass. Look at the guys above competing again, they have all done the ecosystem thing once and won in their respective fields/era's, yet they are losing here. It's not that easy!

[valueinvestor82]

Back to reality- so once this bubble gets a bit deflated (let's say- AMZN at $300) will Bezos issue twice as many shares to make sure employees are payed the same salaries? Just wondering...

[cmlber]

Walmart is 5x larger and the package delivery, the only part of the value chain that benefits from material economies of scale, is outsourced.

 

I want to push back on both parts of this statement.

 

(1) Is Walmart really 5x larger in a way that would matter for economies of scale, assuming they exist?  I believe the relevant economy of scale we're talking about is in distribution -- more volume equals denser, more efficient distribution networks.  On the surface, Walmart appears to have a big advantage, because Walmart U.S. had $300 billion in sales last year (not including Sam's Club) while Amazon had only $63 billion in U.S. sales.  But Amazon's sales are growing much faster, and Amazon's distribution network increases it's throughput by providing fulfillment services to third-party sellers ("Fulfillment by Amazon" or "FBA").  Amazon does not record as revenue the gross amount of FBA sales; rather, only it only records as revenue the portion of the third-party sales it retains.  So, to make an apples-to-apples comparison of the volume of goods flowing through Amazon's distribution system versus Walmart's, you'd need to know the gross merchandise volume of FBA sales.  I haven't seen that number reported by Amazon.  But in its 4Q conference call, Amazon stated that third-party sales were about 47% of total units sold, and FBA constituted about 50% of third-party sales.  So, a very rough estimate would put the gross merchandise value (GMV) of Amazon's FBA sales at about 50% of its first-party sales.  If first-party sales were $26 billion in 4Q, that would put the GMV of FBA sales at around $13 billion.  All in, then, Amazon's 4Q first-party GMV and FBA GMV equaled about $40 billion.  In contrast, Walmart's GMV in 4Q was about $80 billion.  If these numbers are roughly right, then Walmart's U.S. distribution system appears to be only about 2x the volume of Amazon's today.  Moreover, the amount of GMV that Amazon distributes is growing very rapidly, particularly through FBA.  In comparison, Walmart's GMV is growing very slowly.  So, for how much longer will Walmart actually enjoy a scale advantage over Walmart in the U.S.?

 

(2) Why is "package delivery" -- which I assume refers to the movement of a package from a warehouse to the consumer -- the only part of a distribution system that benefits from economies of scale?  In other words, why are there no economies of scale at earlier stages of the distribution chain? 

 

More generally, what my comments highlight, I think, is the potential mistake in comparing Walmart to Amazon.  As Ben Thompson discusses in the article that has been posted several times on this thread, first-party retail sales on Amazon -- the business that competes head-on with Walmart, Target, et al. -- may never produce any profit.  But that business provides tremendous scale that provides the backbone for what the real profit centers will be:  services to third-parties through businesses like AWS and FBA.  That, to me, is the biggest problem with comparing Amazon to Walmart.  Amazon want's volume from its first-party sales; Walmart, Target, et al. need to make a profit from that activity.

 

1). Those are good points.  But I was talking about scale as it would benefit purchasing power, given my view that there are minimal scale benefits on the logistics side of AMZNs business.  For purchasing power, the US isn't all that matters and third party sales don't provide any benefit.  So from a procurement perspective, WMT has 5x more scale.  Not to say that really matters.  I actually don't think it does, but it's the only reason scale might matter besides a cost advantage in logistics. 

 

2).  How hard is it to achieve optimal revenue per stop delivering truckloads of packages between 100 warehouses in the U.S.? Not very.  How hard is it to achieve optimal revenue per stop delivering individual packages to 1 billion households globally?  Very, very hard.  It's why DHL went out of business (with more "scale" than AMZN has today) in the U.S. ground game.  This distinction is why LTL trucking businesses are generally pretty good businesses, and truckload shippers are often just guys with a truck that take one off jobs.  It takes no scale to stop at one warehouse, fill a truck, and drive it to another warehouse efficiently.  It takes scale to stop at one warehouse, fill a truck, and drive it to 200 individual homes efficiently.  Route density is key in that scenario. 

 

Regarding your more general comment, I guess through rose colored glasses that's the right interpretation.  I'd say it's not WMT that HAS to generate $16 billion a year in FCF from its retail operations, but WMT that GETS to generate $16 billion a year in FCF from its retail operations.  How can that possibly be a disadvantage?  And how does scale in retail give Amazon an advantage in AWS?  Maybe they're running the retail business at breakeven and then can support the third party business, which is a good business.  But if that's the play, just buy EBay and save yourself $200 billion...

 

 

[KJP]

 

2).  How hard is it to achieve optimal revenue per stop delivering truckloads of packages between 100 warehouses in the U.S.? Not very. 

 

 

I think it's actually quite hard to achieve optimal costs in a distribution chain.  In fact, in Competition Demystified, Bruce Greenwald analyzed Walmart and identified one of its biggest advantages as its lower costs for "inbound logistics," meaning "the cost of bringing goods into its warehouses and sending them on to the discount centers." [p. 91]  Walmart achieved this cost advantage through local economies of scale. 

 

 

Regarding your more general comment, I guess through rose colored glasses that's the right interpretation.  I'd say it's not WMT that HAS to generate $16 billion a year in FCF from its retail operations, but WMT that GETS to generate $16 billion a year in FCF from its retail operations.  How can that possibly be a disadvantage?  And how does scale in retail give Amazon an advantage in AWS?  Maybe they're running the retail business at breakeven and then can support the third party business, which is a good business.  But if that's the play, just buy EBay and save yourself $200 billion...

 

My point about Walmart vs. Amazon is that Walmart has to compete against someone that doesn't need to earn a profit in the area where they are competing.  In my view, that's a dangerous position for Walmart to be in long term.  What will Walmart's FCF margins be in 10 years?  If you believe Walmart's FCF margins will be the same as today, how do you square that belief with Amazon's continued growth?

 

Regarding AWS, Amazon itself uses a massive amount of computing power.  That need for computing power was the origin of the cloud-based system that developed into AWS and formed the beginning of the scale benefits inherent in cloud computing.  A dated article discussing some of the benefits to scale in cloud computing is:  http://www.cloudscaling.com/blog/cloud-computing/understanding-cloud-datacenter-economies-of-scale/

 

With respect to how AWS and FBA grew out of -- and benefit from the scale of -- Amazon's first-party retail business, see this article:  https://stratechery.com/2016/the-amazon-tax/

 

I personally have never worked in cloud computing or logistics, so I cannot evaluate the accuracy of these sources.  But I'd want to have factually supported reasons before I said they were wrong and that no real economies of scale exist.  I'd love to read any sources you have that debunk the economies-of-scale story in the sources cited above.

 

Finally, regarding E-Bay, look at the growth rates of Amazon's and E-Bay's businesses and the potential benefits from FBA.  I don't think they are comparable.

 

 

[cmlber]

 

2).  How hard is it to achieve optimal revenue per stop delivering truckloads of packages between 100 warehouses in the U.S.? Not very. 

 

 

I think it's actually quite hard to achieve optimal costs in a distribution chain.  In fact, in Competition Demystified, Bruce Greenwald analyzed Walmart and identified one of its biggest advantages as its lower costs for "inbound logistics," meaning "the cost of bringing goods into its warehouses and sending them on to the discount centers." [p. 91]  Walmart achieved this cost advantage through local economies of scale. 

 

 

Walmart has 5,000 stores with a particular concentration in the middle of the country.  Logistics between 5,000 stores and the associated warehouses and distribution centers is far different than logistics between 50 fulfillment centers.  Also, having 5,000 stores with exactly the right amount of inventory is a unique challenge to the physical world that is far more logistics intensive.  In a Walmart, if the item isn't at the particular store a customer is shopping at, the sale is lost.  In an online world, if the product happens to be at a fulfillment center that is relatively far from the customer, the sale isn't lost.

 

FYI, Bruce Greenwald, who you sourced, is massively short AMZN at a cost basis I think around half this price.

[cmlber]

 

 

Regarding your more general comment, I guess through rose colored glasses that's the right interpretation.  I'd say it's not WMT that HAS to generate $16 billion a year in FCF from its retail operations, but WMT that GETS to generate $16 billion a year in FCF from its retail operations.  How can that possibly be a disadvantage?  And how does scale in retail give Amazon an advantage in AWS?  Maybe they're running the retail business at breakeven and then can support the third party business, which is a good business.  But if that's the play, just buy EBay and save yourself $200 billion...

 

My point about Walmart vs. Amazon is that Walmart has to compete against someone that doesn't need to earn a profit in the area where they are competing.  In my view, that's a dangerous position for Walmart to be in long term.  What will Walmart's FCF margins be in 10 years?  If you believe Walmart's FCF margins will be the same as today, how do you square that belief with Amazon's continued growth?

 

Regarding AWS, Amazon itself uses a massive amount of computing power.  That need for computing power was the origin of the cloud-based system that developed into AWS and formed the beginning of the scale benefits inherent in cloud computing.  A dated article discussing some of the benefits to scale in cloud computing is:  http://www.cloudscaling.com/blog/cloud-computing/understanding-cloud-datacenter-economies-of-scale/

 

With respect to how AWS and FBA grew out of -- and benefit from the scale of -- Amazon's first-party retail business, see this article:  https://stratechery.com/2016/the-amazon-tax/

 

I personally have never worked in cloud computing or logistics, so I cannot evaluate the accuracy of these sources.  But I'd want to have factually supported reasons before I said they were wrong and that no real economies of scale exist.  I'd love to read any sources you have that debunk the economies-of-scale story in the sources cited above.

 

Finally, regarding E-Bay, look at the growth rates of Amazon's and E-Bay's businesses and the potential benefits from FBA.  I don't think they are comparable.

 

You're right that you can't be efficient in operating fulfillment centers for third-party vendors without scale.  But the amount of scale required to be efficient is tiny compared to the market size, which means there will be room for hundreds of efficient fulfillment businesses.  This is like saying American Airlines will dominate air travel because one guy with a plane can't possibly have the same cost structure.  You'd be right about the one guy with a plane, but what you're ignoring is that JetBlue, Southwest, United, Spirit, etc. all have scale also, and then you're stuck selling a commodity service with the same cost structure as multiple other competitors, even though you're way more efficient than the guy with a plane.

 

And why does Walmart need to make money in retail but Amazon doesn't?  That's the competitive advantage?  If that business model (retail as a loss leader for third party fulfillment) is the future of retail (worth $1T!), why can't Walmart just start doing third party fulfillment AND make $16 billion/year off the retail operation?

 

 

[Palantir]

But how can you say that AMZN doesn't need to make money when it is indeed making an economic profit as you alluded to a few posts above?

[ni-co]

I would argue against that, too. Amazon needs underlying profitability as much as WMT. But – and this is a big "but" – they are institutionally set up to reinvest it into growth right away.

 

Online retailing has a few fundamental advantages over offline retailing. The most important ones I can think of are: 1. infinite selection, 2. third party integration, 3. integration of digital products. 4. seems to be more controversial but I think it's less capital and labor intensive and therefore cheaper for the customer in the end.

 

When you guys think that WMT can just go out there and replicate those advantages like nothing just because they have large FCF, I can't help you and can only say: well, doesn't look to me like they do it well enough so far.

 

There is an undeniable shift from offline to online retailing going on, and it's far from over—it's a secular trend. If WMT is not able to secure a market share proportional to their share in the offline market I would be really worried as an owner because, with a large market share in a shrinking market, they will inevitably shrink and, in the long run, they will be finished because they will lose their scale advantages. And if you tell me that they can do it cheaper or as cheap as Amazon and that Amazon is essentially faking it, then where does Amazon's (profitable) growth come from? All the nice FCF will help WMT as little as it helped Blackberry to fend off iOS and Android. They are going to be disrupted.

 

With regard to AWS vs Google's or Microsoft's cloud I read an interesting post by a longtime ex-Amazon, now Google engineer recently. It's from 2011 but gives an interesting insight into the institutional strengths and weaknesses of both companies. I can imagine that Google's service has become a lot better since then but I know too little about it to be able to judge it. What became clear to me is that there's more to it than just plugging together a bunch of servers. You need to build a cloud platform. Your customers are very diverse and it needs to be a very flexible.

 

Share issuance: seems to be pretty constant to me. Their share issuance is minimal compared to their fundamental growth. I don't see a Salesforce-like scenario here.

[NewbieD]

Netflix, Spotify Reddit, lots of government stuff - and thousands more. Is that what you were asking?

 

 

Spotify has recently switched to Google: http://uk.businessinsider.com/spotify-switches-to-google-cloud-platform-2016-2?r=US&IR=T

 

Edit: according to the article it was mostly hosted on their own servers before, not AWS.

 

[KJP]

 

You're right that you can't be efficient in operating fulfillment centers for third-party vendors without scale.  But the amount of scale required to be efficient is tiny compared to the market size, which means there will be room for hundreds of efficient fulfillment businesses. 

 

I don't know what the economies-of-scale curve looks like in fulfillment, nor do I know what the what the minimum efficient scale is.  If you have seen an analysis of that, I would like to see it.  I don't think Amazon discloses enough information to determine whether its cost per fulfilled unit is falling over time. 

[cmlber]

 

You're right that you can't be efficient in operating fulfillment centers for third-party vendors without scale.  But the amount of scale required to be efficient is tiny compared to the market size, which means there will be room for hundreds of efficient fulfillment businesses. 

 

I don't know what the economies-of-scale curve looks like in fulfillment, nor do I know what the what the minimum efficient scale is.  If you have seen an analysis of that, I would like to see it.  I don't think Amazon discloses enough information to determine whether its cost per fulfilled unit is falling over time.

 

So your default when you don't know if there are economies-of-scale and there is absolutely no evidence in the financials that there are economies-of-scale is to assume that they exist?

 

Think about what fulfillment is.  It's four walls and some shelves located where land is cheap and plentiful that takes in truckloads of inventory from third parties, sorts it, and brings it to a dock where it's loaded on a UPS truck when ordered.  Sorting packages is all they are doing.

 

If you want to know the minimum efficient scale, google "Fulfillment center" and see how many competitors there are that you didn't even know existed (but must be making money because they wouldn't all exist otherwise).  Here's what comes up (in order):

 

Ad for shipfusion.com - E-Commerce Fulfillment - Warehouses in US & Canada

Ad for innotrac.com - eCommerce Fulfillment - Fulfillment & Customer Care

Ad for shipwire.com - Order Fulfillment Centers

Ad for rakutensl.com - Fulfillment Centers

shipwire.com - Fulfillment Centers and Order Fulfillment Services by Shipwire

Fulfillment House Wikipedia Page

What is a fulfillment center? Dictionary page

Amazon Fulfillment - Locations     

eBay Fulfillment Center

pickandship.com - Specialty Fulfillment Center - Order Fulfillment

fulfillmentcenter.com - Fulfillment Services since 1984

Ad lynxfulfillment.com

Ad warehousingandfulfillment.com

Ad meyerfulfillment.com

 

There are literally hundreds of legit competitors if you take the time to Google.

[KJP]

 

So your default when you don't know if there are economies-of-scale and there is absolutely no evidence in the financials that there are economies-of-scale is to assume that they exist?

 

 

I'm not assuming anything.  I've said all along I don't know.  And I'm not long Amazon either, and I never said I was.

 

What I'm trying to do is take the bull and bear cases and find data that either supports or disproves them.  That's why I asked you if you had any.  Apparently you do not.*  Or maybe you do and don't want to share it. You're obviously under no obligation to tell me anything, and you're free to tell me to go jump in a lake. 

 

All I'm trying to do is understand whether your stated beliefs are based on actual data, your personal experience as a logistics or cloud computer professional, or something else. 

 

*I don't think the mere existence of other third-party fulfillment companies tells us very much about the economies-of-scale curve of the business.  And I'm not sure why you assume I didn't know about them. 

[cmlber]

 

So your default when you don't know if there are economies-of-scale and there is absolutely no evidence in the financials that there are economies-of-scale is to assume that they exist?

 

 

I'm not assuming anything.  I've said all along I don't know.  And I'm not long Amazon either, and I never said I was.

 

What I'm trying to do is take the bull and bear cases and find data that either supports or disproves them.  That's why I asked you if you had any.  Apparently you do not.*  Or maybe you do and don't want to share it. You're obviously under no obligation to tell me anything, and you're free to tell me to go jump in a lake. 

 

All I'm trying to do is understand whether your stated beliefs are based on actual data, your personal experience as a logistics or cloud computer professional, or something else. 

 

*I don't think the mere existence of other third-party fulfillment companies tells us very much about the economies-of-scale curve of the business.

 

If you don't think the existence of hundreds of competitors operating profitably at tiny scale is evidence of minimal economies of scale, idk how to convince you.  You're confusing economies of scale (which exist in literally EVERY business) with economies of scale that are not easy to obtain (which exist in only a handful of good businesses). 

 

There are huge economies of scale in operating independent pizza restaurants too.  If you sell 1 pizza per day, your cost per pizza will be insanely high and you'll go out of business.  But since the market for pizza is gigantic, it's very easy to take enough share to get the benefits of "scale," so nobody says this is a scale business...

 

Read the 10-K from 2012 of Innotrac (before it was taken private), one of the 5 competitors that appears BEFORE Amazon when you google "Fulfillment centers."  They were profitable in 2012 with $107 million in revenue.  They say on their website that they now ship $4.5 billion in GMV.  So with 2.5% of Amazon's GMV, which itself is only 0.4% of global retail sales, you can be profitable?  Does that look like a business where scale is difficult to achieve?  You need 1 basis point of global retail sales to be profitable. 

 

On the actual package delivery side though, we've seen that at 4% market share, DHL was bleeding cash and saw no path to profitability.  So, do you still disagree with my statement that the actual delivery of the package is where scale really matters?

 

I'm not sure either if there are no economies of scale.  If I had to guess, I'd say Amazon has a few basis points of cost advantage due to its scale in fulfillment, but, the evidence is overwhelming that the scale advantages aren't huge. 

[KJP]

 

So your default when you don't know if there are economies-of-scale and there is absolutely no evidence in the financials that there are economies-of-scale is to assume that they exist?

 

 

I'm not assuming anything.  I've said all along I don't know.  And I'm not long Amazon either, and I never said I was.

 

What I'm trying to do is take the bull and bear cases and find data that either supports or disproves them.  That's why I asked you if you had any.  Apparently you do not.*  Or maybe you do and don't want to share it. You're obviously under no obligation to tell me anything, and you're free to tell me to go jump in a lake. 

 

All I'm trying to do is understand whether your stated beliefs are based on actual data, your personal experience as a logistics or cloud computer professional, or something else. 

 

*I don't think the mere existence of other third-party fulfillment companies tells us very much about the economies-of-scale curve of the business.

 

If you don't think the existence of hundreds of competitors operating profitably at tiny scale is evidence of minimal economies of scale, idk how to convince you.  You're confusing economies of scale (which exist in literally EVERY business) with economies of scale that are not easy to obtain (which exist in only a handful of good businesses). 

 

There are huge economies of scale in operating independent pizza restaurants too.  If you sell 1 pizza per day, your cost per pizza will be insanely high and you'll go out of business.  But since the market for pizza is gigantic, it's very easy to take enough share to get the benefits of "scale," so nobody says this is a scale business...

 

Read the 10-K from 2012 of Innotrac (before it was taken private), one of the 5 competitors that appears BEFORE Amazon when you google "Fulfillment centers."  They were profitable in 2012 with $107 million in revenue.  They say on their website that they now ship $4.5 billion in GMV.  So with 2.5% of Amazon's GMV, which itself is only 0.4% of global retail sales, you can be profitable?  Does that look like a business where scale is difficult to achieve?  You need 1 basis point of global retail sales to be profitable. 

 

On the actual package delivery side though, we've seen that at 4% market share, DHL was bleeding cash and saw no path to profitability.  So, do you still disagree with my statement that the actual delivery of the package is where scale really matters?

 

I'm not sure either if there are no economies of scale.  If I had to guess, I'd say Amazon has a few basis points of cost advantage due to its scale in fulfillment, but, the evidence is overwhelming that the scale advantages aren't huge.

 

I don't know which of those competitors operate profitably and which don't.  All I asked was a simple question:  What is your view based on?  I thank you for your responses.

[rpadebet]

If your fulfillment centers are closer to end consumer compared to your competitors fulfillment centers, don't you think you can deliver packages quicker/cheaper assuming all competitors use UPS at same price? Can that lead to increased sales at your company compared to competitors? Does it now cost you more to sort that marginal package? If you are using robots to do it and comps are using humans, do you have a variable cost advantage?

 

Also considering you ship tons more through UPS compared to your competitors, and since UPS is gaining their scale through you, is it possible to negotiate favorable shipping rates with UPS compared to your competitors?

 

I think this is how scale helps on the demand side of things....

 

[ni-co]

If your fulfillment centers are closer to end consumer compared to your competitors fulfillment centers, don't you think you can deliver packages quicker/cheaper assuming all competitors use UPS at same price? Can that lead to increased sales at your company compared to competitors? Does it now cost you more to sort that marginal package? If you are using robots to do it and comps are using humans, do you have a variable cost advantage?

 

Also considering you ship tons more through UPS compared to your competitors, and since UPS is gaining their scale through you, is it possible to negotiate favorable shipping rates with UPS compared to your competitors?

 

I think this is how scale helps on the demand side of things....

 

It's kind of funny that we are still having this discussion after 20 years of crazy growth for Amazon in the face of hugely profitable competitors. It's pretty obvious that the logistics industry is the next one being disrupted by the same model. I get why people have a hard time valuing Amazon but it's irrational to say you can't value it and therefore it's overvalued.