Amazon and the “profitless business model” fallacy (2013)
eugenewei.com
eugenewei.com
If wall street had been bashing Amazon, you'd think their stock would have been muddling along with low multiple over the past 20 years.
How old are you? I ask because if you want to see what people said about Amazon in 2008 look at what they say about Tesla today.
Neither earnings for a company that aggressively reinvests nor sales for a company that's a mixed bag of cloud computing (high margin) and retailing/logistics (low margin) make much sense as compass.
If you look at Amazon's price-to-FCF multiples[1], while Ben Graham would surely pass, it is not as expensive as their price-to-earnings make it seem, IMO (a non-holder of AMZN).
0: https://www.sec.gov/Archives/edgar/data/1018724/000119312505...
>[Segment title] Our Most Important Financial Measure: Free Cash Flow Per Share
> Amazon.com’s financial focus is on long-term growth in free cash flow per share. Amazon.com’s free cash flow is driven primarily by increasing operating profit dollars and efficiently managing both working capital and capital expenditures. We work to increase operating profit by focusing on improving all aspects of the customer experience to grow sales and by maintaining a lean cost structure. [...] This focus on free cash flow isn’t new for Amazon.com. We made it clear in our 1997 letter to shareholders—our first as a public company—that when “forced to choose between optimizing GAAP accounting and maximizing the present value of future cash flows, we’ll take the cash flows.” I’m attaching a copy of our complete 1997 letter and encourage current and prospective shareowners to take a look at it.
1: https://www.macrotrends.net/stocks/charts/AMZN/amazon/price-...
The key point is that timing of expenses vs revenue significantly impacts cash flow. A growing business can lose money on every sale and quite literally make it up in (acceleration of) volume till they grow large/efficient enough to have profitable unit economics.
For a brief example to hopefully illustrate why that matter, a pretty basic business would be to buy stuff, make a thing, and sell your widgets. If it takes you 6mo to turn every $1 into $1.30 in that business then you have pretty amazing unit economics but still might struggle with cash flow; if sales just double in a given year or two you won't have enough velocity of money to build the things people are buying without additional investments or cash reserves. Contrast that with a net-60 payment for your materials and a 30-day pre-order for your hyped up widgets. At any given point in time you'll have exactly (not actually exactly I don't think, I'd have to do the math, this might depend on how fast sales increase or decrease) twice as much money in the bank (because it's tied up in other people's accounts for a total of half the time) and a high enough velocity of money to nearly triple your sales every year (supposing you can find buyers for your product).
Additional money in the bank can have super-linear value for because of the extra options available. If you're struggling to pay rent and wages this month and a great deal on the lathe you need crops up then you might quite literally not be able to take advantage of that cheaper machine at the moment and be forced to pay more later when you can actually afford it. Or if you're able to efficiently utilize cash to grow your business you can just use those cash reserves to grow more quickly. In any case, cash flow is typically a good thing to have.
That isn't to say that profits and whatnot don't matter (especially on sufficiently large timescales), but they don't impact the day-to-day functioning of the business directly like cash flow does.
Important to note, that is free cash-flow from operations, not including cash flow from financing rounds and the like.
Amazon is the exception so, not the rule. And not every company prioritizing growth over profits will end up where Amazon is.
https://news.ycombinator.com/item?id=6620885
>However, things are not that simple. Sometimes some expenses which are about building for the future and investing into new growth are not capitalized. This is the case because for some expenses the benefits are so uncertain and difficult to quantify that the SEC requires that they are reported as ordinary expenses instead of capital spending. These types of expenses tend to involve R&D and may include certain administrative expenses associated with growth initiatives.
That shows why GAAP has legitimately different standards of profitability than an investor or HN comment might.
When people bring up “haha, Uber/Amazon has never even been [GAAP] profitable! What a joke!”, it’s usually to make the point that, “investors are dumb, this enterprise isn’t even capable of value creation because they can’t sell outputs for more than production costs, what a scam”.
But that doesn’t follow at all from GAAP-non-profitability. If a company would have been profitable but for regularly dumping fortunes into speculative projects, that’s still great news for investors. It means they could one day give up on that speculation and go back to taking those easy profits from value creation.
The reality is (at least in the US and probably elsewhere), that it's not an option to just decide to expense or capitalize everything and so the timesheets might be better viewed as the mechanism the company has chosen to remain compliant with the law and not something that's optional/subject to incentive pay.
(The bigger question is: Why did these people want to believe that Tesla is a failure? I think the potential answers to that can teach us a lot about psychology, incentive structures, social media and what Nietzsche called The Use And Abuse Of History.)
A lot of people don’t really care if what they say is accurate.
This isn't Jeff bashing because he's obviously a hard working and very smart guy. But it's only obvious in retrospect that Amazon knew what they were doing.
I rarely buy anything besides accessories and hard to get groceries on there anymore.
Based on what, anecdotal evidence from squeaky wheels on the internet? I've heard the same claims, but the 10K and 10Q filings the original commentor linked to suggest the opposite.
- do i need it today- Costco, Walmart, Target, Safeway, Dicks Sporting Goods
- can i wait till tmrow- Amzn next day to my doorstep. No? resort to above options
- can i wait 2 days for a better choice- Amzn Prime
- No hurry- a week ok?- Amzn marketplace, eBay
Wal-mart is certainly trying to do just that. Amazon had to drop their the minimum order value for free shipping back down when Wal-mart took the offensive.
As far as I'm concerned, Amazon is already imploding. Their search engine has become nearly useless, returning whatever the hell it wants, rather than what you searched for. It keeps getting worse, not better. Have you compared the item counts when you sort by Relevance versus Price Low to High? Where did they disappear to? What the hell is wrong with their site?
I used to place thousands of orders from Amazon every year, now I'm down to dozens, just occasionally checking them before buying a large ticket item to see if they've got an incredible price on it. And even then I only go through and buy from Amazon when nothing leads me to believe it might be a scam / counterfeit / mislabeled / etc.
But, Google and Microsoft didn't wake up earlier. You mention it's only in retrospect Amazon knew what they were doing; but we also see Microsoft and Google didn't know what they were doing. It wasn't out of charity to Amazon, they simply lacked vision at the time.
Had major auto manufacturers taken Tesla seriously 5-6 years sooner instead of sitting idle, the Tesla doubters might be the smug ones at this point. But Tesla had the vision and the execution.
In both cases, the competition badly miscalculated while the company executed beautifully.
which is interesting - because in almost all cases, it's this unproven market that end up upending the business. IBM is an early case, and now electric cars.
So if an existing, well capitalized business would learn from this, they should buck the trend and invest heavily in risky ventures. However, shareholders, esp. conservative shareholders, would not want their existing, certain profit lines be endangered.
Facebook’s approach only works when regulators are asleep at the wheel, but standard oil style monopolies can just print money to cover these costs. Innovation is expensive and can fail regularly just look at Intel or most corporate R&D. Finally milking a dying business model can fail really quickly, just look at say blockbuster.
In a nutshell, the argument is that successful companies fail to enter new markets because they have good management, not bad one.
In Ford's or IBM's case, a sane manager would not go invest in an unproven market, they rather maximize profits and growth on the market they are most effective at; however, this comes at a cost of missing new opportunities, which makes this a dilemma.
Successful companies could try to create little startups inside the big corporation; however, this rarely works out since you really need the entrepreneural spirit, align compensation to success and freedom without the restrictions of the parent company. The best alternative might be to create a VC arm that can serve later on as merger as acquisitions, but even then, is not guaranteed the startup would want to merge to the parent company nor that it would be the best decision.
EVs on the on other hand will, most likely, replace ICE powered cars. With the big difference that existing ICE manufacturers will just switch production to EVs and will be just fine in the new EV world. I do see multiple case studies in that.
And as much credit as Bezos gets, the new CEO of Amazon was the one who (as far as I know) was the one who insisted that all Amazons web development be done in a way that made it easy to set up AWS.
Don't know who really thought of it, or designed it well enough to make it work, but hope all those folks were well, well rewarded.
I've heard this too, but I don't think it can possibly be true. That would mean that, at the start, any service or VM etc. you started on AWS would only work during non-peak Amazon hours. So your website or app would be down if the traffic to Amazon's website ramped up. I can't imagine many companies going for that offering. If that was the product that AWS originally offered I can't imagine it would have been successful.
Disclaimer: I work in non-aws amazon, and my imagination is my own.
While that's true, at the time I read this "peak Amazon hours" were a few hours on cyber Monday. In fact, that's probably still true.
>Amazon invented AWS because they had idle computers most of the time to handle peak load, and they thought it would be a good way to monetize them.
That is forward looking.
On Tesla... I'm not sure it's a good analogy. I don't think you can point to Tesla and say they got lucky anywhere. What will GM/Ford/Toyota/Daimler/Volvo do differently now? Tesla is selling speed/sex appeal/engineering/electric - ie all those companies were already all competing on these dimensions. They can't seem to execute.
Google and Microsoft were plenty capable of executing, they just didn't see what the product they needed to deliver was.
[1]: https://www.greencarreports.com/news/1134120_tesla-is-losing...
Sounds just like the legacy auto manufacturers. Tesla is a great analogy.
Google also had zero enterprise experience. Things like MySQL, RFC 1918, VMs were not native by their devs.
So of course boards of these companies held off. Making that move could have tanked the stock as profit went down while they re-invested in their business.
This is the genius of Amazon. Wall Street doesn’t expect growing profit. Instead it’s happy with growing revenues. Allowing Amazon to reinvest into business that have an initial low margin until economy of scale kicks in.
Same was true of Amazon.
At the time, Amazon might not have had enterprise sales teams. But they had deep expertise in running the familiar suite of enterprise software at scale.
https://hurricanecapital.wordpress.com/2015/11/30/business-i...
>HB: So, let’s just establish this once and for all. Can Amazon make money?
>JB: Well, yes, and in fact we have in the past. And you have to understand, there are many ways of thinking about this, but the reality is that Amazon is a collection of several businesses and initiatives. And we have some very significant, very profitable, more established businesses that are free cash-flow generating. Very significantly. And fortunately, the way I think about it, we have lots of opportunities to invest in this new initiatives, and we take advantage of those opportunities. So, it’s kind of like we built this lemonade stand, you know, twenty years ago. The lemonade stand has become very profitable over time. But we also decided to use our skills and the assets that we’ve acquired over time to open up a hamburger stand, and a hot dog stand, and so on and so on. So we’re investing in new initiatives.
>JB: Look, you know. Warren Buffett has this great quote, he says: “You can hold a rock concert, and that’s okay. And you can hold a ballet, and that’s okay. Just don’t hold a rock concert and advertise it as ballet.” Investors come in all shapes and sizes. They have different investment horizons, different approaches, different beliefs about what the right kind of portfolio looks like. And, so it’s not when you know people use Wall Street as a shorthand. But there isn’t one type of investor, they come in different shapes and sizes. You have to be super clear about what kind of company you’re trying to build, what your approach is. We laid that out in our 1997 annual shareholder letter. We said we were going to take big bets. We said they were gonna fail. We said some of them hopefully were gonna work. We said we were gonna to invest for the long term, that we were gonna to take advantage of market opportunities as they arose. And there’s a certain kind of investor who is aligned with that approach. And so again, you could hold the ballet or the rock concert, and both can work. Just be clear about which one you are, and then people can self select.
The problem with this line of reasoning is it assumes you can cut off the loss leaders and just run the profitable lines. Generally speaking, you can't. The customers come to Amazon because everything is there. The second they think everything isn't there, they come less frequently, and those "amazingly profitable" categories become not so amazing or profitable. Rotisserie chicken is a loss leader for many supermarkets, and the supermarket makes a lot of money on those snacks near the register. They can't decide to be in the snacks only business tomorrow. Neither can Amazon.
The story about the ambition and the growth and all that is nice - that story is true for Facebook, Google, Amazon Web Services, Microsoft and so on. They all got profitable within a decade and continued with high growth, high profitability operations.
Amazon's doesn't have many loss leaders in physical ecommerce any more. They used them to build those businesses 10-15 years ago, but they don't need them anymore. The only loss leaders as a category they maintain are digital. And even that is pretty limited to Amazon Video/Studios at this point.
AWS is a genuinely different business, the rest of it is all bundled together. So, they have digital loss leaders. And they may not have many physical loss leaders, but the ongoing argument that they have all these lines which are profitable just doesn't mean anything when customers don't want an online shop for X specific thing. They want an "everything store".
It's survivorship bias to assume they had no real plans but lucked out - the gamble was part of the plan.
When you say "digital" do you mean like Prime Video? Or kindle books? Or something else? When you say devices do you mean kindles? I'm struggling to believe that the kindle book line is run at a loss. I'd understand specific titles might run at a loss and I'd understand if the devices run at a loss but kindle books overall? Prime Video "losses" are sort of my point - they can almost be thought of as marketing expense.
As for the capital leases - those are accounted for at the corporate level and mostly run through the cashflow statement. The numbers they quote as operating income for the non aws business don't include those capital lease costs. ie on page 17 of their most recent Q (https://www.sec.gov/ix?doc=/Archives/edgar/data/1018724/0001...) those reported operating earnings don't include the capital leases. If you look on page 3 of that filing (Principal repayment of finance leases) you'll see the (big) number there.
All that said - you still could be right. But if you are it's almost like they are going out of their way to deliberately make their numbers look bad on the non-AWS side. There is a big difference between not caring and actively making it look bad.
Because their earnings filings from 2000-2007 certainly suggest that they were making a boatload of money on books. Their Media segment was making a ton of money while their EGM segment (electronics and general merchandise) was losing a ton of money. And then EGM became profitable and they combined the segments together into a single physical retail segment that included their new focus on apparel and fashion.
Amazon started with books, made books profitable and then used that profit to move into the next category (VHS/CD/DVD) and then the next category, etc.
Prime video and amazon studios have been spending over 4 billion a year since 2014 (no idea what the current spend is). The entire point of those billions was to keep people as prime subscribers and keep them spending on physical retail. This strategy only makes sense if physical retail is profitable. It's an LTV play.
Devices were run at a gross margin loss and even worse capitalized loss if you include the capitalized R&D expenses. "We want to make money when you use our devices" - Bezos. The devices themselves were sold at a loss for many, many years.
Amazon used the same strategy of capital leases for funding warehouses and funding AWS data centers. They only started reporting on the capital leases in the last 5-10 years (can't remember the year).
I've been following Amazon's shareholder letters and releases for years. Physical retail has been massively profitable.
That's not the impression I get from this Bezos interview in 2000: https://www.youtube-nocookie.com/embed/6OVdcS4fuuw
[1] https://www.macrotrends.net/stocks/charts/AMZN/amazon/net-pr...
I'm not saying that's all you need to be successful. It's not so simple. Nothing ever is.
But 8 years later, and this essay can stand on its own accord, not just as a historical artifact of its time, but as stellar business advice.
https://deep-throat-ipo.blogspot.com/2018/05/amazon-walmartc...
I'll have to reread, more slowly.
Until then, some quick tidbits.
One anecdote that popped out was the correlation of Walmart super centers and crappy towns. Market segmentation via geography vs demographics. "The Twinkie Index" was the first time I heard this notion. Starbucks' real estate and marketing groups had came up with a infographic, contrasting cafe locations and per capita Twinkie consumption, for when people asked "When are you going to finally put a Starbuck in [insert town here]?"
Back in the 90s, Walmart's profit sharing (employ stock grants) was a role model. Cited as one of their competitive advantages. (How things change.)
I read that what made Walmart possible (feasible) was their early, enthusiastic adoption of bar codes.
For comparison, what are (were) Amazon's critical success factors?
Lowest cost of capital. Uniquely, was able to sell growth over profits story to Wall St. Superior tax avoidance (eg sales tax holiday). Compensating employees with stocks. It's hard to imagine anyone challenging Amazon while this disparity exists.
Amazon Prime, for sure. Brand loyalty, reduced cost of customer acquisition, and all that.
I have questions:
How did Amazon figure out that programmers are a consumable resource? Cattle instead of pets. From the very beginning, my friends and I marveled at how poorly Amazon treated programmers. Surely, that can't last, right? Whelp. I was totally wrong. And I still don't know why.
Why hasn't Walmart's Jet acquisition done better? Is Walmart unable to invest in Jet sufficiently to compete? Did Walmart and Amazon accept and settle into their comparative market segments? (eg Maybe Walmart's demo doesn't buy stuff online as much.)
Why hasn't Amazon's eBay-ification, the rise of fraud and counterfeits, harmed them more? Is it because of their "customer obsession" and generous return policies? With comingling and other bullshit strategic decisions, they're promoting fraud vs mitigating it. So weird.
Amazon and the "profitless business model" fallacy - https://news.ycombinator.com/item?id=6620598 - Oct 2013 (137 comments)
Why would any company take a profit when it can invest in it’s own future success and defer taxation?
Anyone paying out a dividend is saying that the shareholder is a better capital allocator by the marginal tax rate.
Philosophically should companies be allowed to do this?
Analogous to similar loop holes where individuals use their own wealth to raise debt for cash and avoid showing paper gains to minimize tax.
1) Because it's almost never the opportunity. 'Super High Growth' over a couple decades is a rare thing.
Imagine if AMZN had not invented AWS? It would be a different story.
2) The term you are looking for is 'Cost of Capitol'. Which varies between various parties for different situations.
If ABC Corp. can't yeild returns past a threshold, investors will want it back to do elsewhere.
Think another way:
If you had $1M to invest, what are the odds that it makes more sense to invest it in your own business vs. any other business?
Why would amazing 'Bet Everything' on AMZN instead of any number of other companies?
If a company is making decent profits, and a lot of powerful companies do, then it may be difficult for them to find other ways to get the same return.
If you're not building cash, how do you plan to avoid catastrophic, unnecessary damage to your business structure when a recession hits? Banks hate to lend during recession, and lending gets a lot more expensive when times are bad.
And banks generally are horrible alternatives when it comes to capital for acquisitions, unless you're a large corporation.
Dependent on venture capital? Another horrible, expensive alternative to profit.
You're seeing inefficiency in profit because you're not pricing in everything in the operating life of a business correctly.
1. Banks. Their primary customers are businesses.
2. Issuing stock.
3. Getting investment capital from rich people.
i wonder how the tax rules and accounting/SEC reporting rules could be tweaked such that the IRS and investors must receive the same level of reporting - aka, if investors see a profit/loss, the taxman must also see the same profit/loss
The flip side is that many companies get large enough that they can’t invest smartly any more. So they should give money back. (IBM, etc)
That doesn't mean perfect, but hard to name a greater company.
In historic terms, the world has never been more peaceful since capitalism 2.1
Whereas there are a million internet forums and mostly anyone can join any one of them for free. Maybe they don't have the same quality of conversation as here (although quite a few others do) but if the thing he hates (captialism, investment firms, etc) produces a culture where the quality of conversation is so high that nothing else can compete, shouldn't that be it's own signal?
That’s why after the country thing I included there aren’t many options. You have to play the game.
As far as i cab tell there are probably 20 anti capitalist sub reddits per pro capitalist sub reddit.
Again, a lot of those subs might have low quality discussion but if the reality is that certain topics attract low quality discusion then maybe that ideology simply isnt interesting to many intelligent people. And to be fair thwre qre definitely well moderated anti capitalist forums out there with good content.
Also plenty outside of reddit as well
> Buffet
Probably because it’s a trendy edgy thing to say online and casually in real life. Can you point me to some harder left subreddits or communities? Being casually or meme focused anti capitalist isn’t the same thing to taking it seriously. I know there’s stuff like that late stage capitalist one.
> 20 per pro capitalist
I believe we are using pro capitalist differently. I used it in an earlier comment for any community that will relate to serious status quo issues in society and be okay with capitalism and/or not make it a priority/superficially put it down at times. So all those big subreddits like their main news, political news, every country’s main subreddit, would all be pro capitalist.
> low quality
Exactly. The subreddit without low quality content was banned. There’s a funny thing of being anti capitalist but not wanting to upend the status quo as much. Those two ideas don’t add up of course. But that’s why low quality makes sense then.
> but if the reality is that certain topics attract low quality discusion then maybe that ideology simply isnt interesting to many intelligent people
I assume you will say you didn’t mean to point this specifically at anti capitalist ideology. Why write this sentence then? Why specify ideology vs topic if you weren’t at some level pointing a finger at anti capitalist ideology?
—
I know we are getting a bit contentious here. I’ll admit my side is primarily because of the last bit above where I’m perceiving my assumptions. I could be wrong.
I do want to point out I’m not anti capitalist in the same way the OP was. I don’t much care if capitalism is or isn’t the economic system in a society I dream up. I prefer focusing on status quo issues en masse. So I have little bias for specifically caring about anti capitalist communities being given their respect because of being anti capitalist.
If you don't want to be banned, you're welcome to email hn@ycombinator.com and give us reason to believe that you'll follow the rules in the future. They're here: https://news.ycombinator.com/newsguidelines.html.
[1] More explanation here: https://hn.algolia.com/?sort=byDate&dateRange=all&type=comme...