Chit Chat Stocks - Amazon (Ticker: AMZN) Not So Deep Dive

Episode Date: January 17, 2023

Amazon might be the most well know online retailer in the United States. The company offers more than just an online marketplace, they also have physical stores. Additionally, Amazon offers Amazon Web Services. At the end of the month, we will publish an Arch Capital episode that will cover the company: Alphabet. Listen closely as Brett and Ryan go through the history, financials, and future prospects of Amazon. Enjoy the show! ****************************** This episode is sponsored by Stratosphere.io, a web-based terminal for financial data, KPIs, and more. Try it out for FREE or use code “CCM” for 15% off any paid plan. Sign up here: https://www.stratosphere.io/ ****************************** Subscribe to our Substack to receive free show notes and charts that go along with every episode: https://chitchatmoney.substack.com/ Want updates on future shows and projects? Follow us on Twitter: https://twitter.com/chitchatmoney Contact us: [e-mail] Timestamps Company Background | (3:14) Industry | (22:18) Management & Ownership | (25:22) Earnings | (29:45) Balance Sheet | (38:25) Valuation | (41:30) Our Analysis | (43:43) Disclosure: Chit Chat Money hosts and guests are not financial advisors, and nothing they say on this show is formal advice or a recommendation. Brett Schafer and Ryan Henderson are general partners and portfolio managers at Arch Capital. Arch Capital and its partners may hold securities discussed on this show. Learn more about your ad choices. Visit megaphone.fm/adchoices

Welcome to Chit Chat Money. On this show, hosts Ryan Henderson and Brett Schaefer interview industry experts and riff on the world of investing. As a quick reminder, Chit Chat Money is a CCM Media Group podcast. Ryan and Brett are also general partners at Arch Capital, and Arch Capital may have positions in the securities discussed in this podcast. Anything discussed on Chit Chat Money by Ryan or Brett or any other podcast guests is not formal advice or a recommendation. Now, please enjoy this episode. Welcome to Chit Chat Money. My name is Brett Schaefer, and I am joined by my co-host, Ryan Henderson. Today is our Tuesday not-so-deep-dive episode where we analyze

one stock by covering its business model, ownership, financials, and future growth opportunities. After listening to this episode, we hope you get a better perspective on the company that we cover. And today, as we chug through our big tech month of December, we are covering Amazon. January. Or January. What did I say? Did I say December? December. Yeah. Okay. Well, I don't know what month it is. It is January. And we are covering Amazon, the leading e-commerce and cloud computing company worldwide. But before we get to this episode, here are a few housekeeping Just for listeners, if you are a regular listener to the Not So Deep Dive episodes, subscribe to our free newsletter and get the show notes and charts delivered to your inbox that we're going to be talking about on this episode. There are going to be a lot of fun ones with Amazon that I think anyone interested in the company will find value in looking at these charts.

The link will be in the show notes, or you can search Chit Chat Money on the Substack app or the Substack website. Second, if you like watching these episodes instead, which I don't know, some people do. It's very fun. Also, when we share the screen, it's really the most important part. You can kind of see the charts we're talking about in real time. You can do so either on YouTube or Spotify. And third, if you enjoy the show, give us a review on either Spotify or Apple Podcast. Before we get to Amazon, today's episode is presented by Stratosphere, the best web-based

research terminal for company-specific metrics like KPIs and segment revenues. Stratosphere has clean data for KPIs, segment data that is triple-checked for accuracy, and beautiful data visualizations helping save you time and frustration digging through SEC filings. I can say for a fact, it really helped us research the Amazon episode. It helped us visualize what we're looking at when we're covering a lot of the Amazon KPIs because they are so complicated. and it helped compile some of the data

for the charts we make ourselves. It's perfect. It's a perfect addition for when we do our Not So Deep Dives episodes. We use Stratosphere as our investing home screen and you can too for free by going to stratosphere.io. That is stratosphere.io.

Okay, Ryan, with the big tech companies, it's a large task to decide what they do or to explain what they do. So why don't you try to break down Amazon? Yeah. And especially with Amazon, they're very discreet. They don't like to break out the financials of certain parts of their business. I think in some ways that might be to avoid regulatory scrutiny, potentially, but also maybe to ward off would-be competitors as well. Um, but Amazon breaks its business down into three segments. So they call it North America, international and AWS, but really under each of those hoods, I guess AWS is kind of exactly what it says it is, but under the North American international

hoods, there's tons of different divisions that function independent of one another and help drive revenue for the business. So, um, they, at the end, at the very end of their 10 K, they list out, I think it's seven different divisions, different segments. And they talk about, and they basically just say how much they generate in revenue. So I'll go through those seven. Obviously, Amazon takes a lot of big bets. They experiment a lot. So there's going to be more divisions than this, but I'll just go through these. So for starters, the biggest one in terms of revenue, the one that most people are familiar with is their online stores. In 2022, it accounted for 47% of their

revenue. It's primarily made up of exactly what you think, the online marketplace, which we all know and love. I'm sure we've all been consumers at some point for them. You can buy virtually anything in their store, either from Amazon themselves. So first-party items, which are actually... Amazon buys in bulk from vendors. In the case of first-party items, it's either something they actually built out themselves or they buy in bulk from vendors and then resell to the consumer. Or there's second-party, which is very... That's a very small percentage of the sales. And then third-party, which actually accounted for 58% of all units sold in Q3. So the most recent quarter, it's really driven by the third-party sellers. And third-party sellers,

they use Amazon for... They set up their shop. They control their brand on Amazon. but Amazon's typically, there's different cases where they can, they offer a lot of different services, but Amazon's usually offering the fulfillment. And so they're basically offering their infrastructure to these third-party sellers. There's other online stores as well here. And so I'm not 100% sure if this is included, but Movies sold on Prime Video. Prime Video is an online store. Ebook sold on Kindle. Kindle's got its own online store. I think subscription purchases made on Fire TV might be included here as well. So basically, Amazon has some other online stores, but it's really dominated by this online marketplace. Second one, physical stores.

This one's pretty straightforward. It's a tiny percentage of the revenue, 4%. But it includes all physical stores that Amazon operates. So that includes Whole Foods locations, as well as amazon branded stores if you're in a rural area you probably you might not know that these exist but um we live in seattle so we get to i think witness a lot of experiments in terms of physical locations so there's the amazon go stores amazon fresh and amazon style i think there's only two amazon style stores in in total but um they do a lot of experimenting in terms of concepts here they want you to have like i think you can do this as well already but they're trying to have this basically no checkout uh online or physical shopping experience so you go in you pick

yeah yeah they call it the and you'll see them reference this if you read any of their stuff they call it just walk out technology so that's what they're referencing uh what ryan is describing either way it's a small percentage of the business and then the the third one that's important to talk about here, and it's a big percentage of the revenue, is the third-party seller services. So I mentioned that third parties sell on Amazon's platform. They drive a lot of the unit volume. But those third-party sellers come to Amazon for pretty much two reasons. And I think everyone's probably seen one of those YouTube scammy looking ads. It's like, you can do fulfillment by Amazon too. Basically, you're going to Amazon if you're a third party

because A, the distribution capabilities are unlike anywhere else. And then there's a massive audience to sell to. And so it's really just trying to get your product in front of as many eyes as possible. I can't go into every single seller service that they offer because there's a bunch. There's storage costs, handling fees, shipping fees, professional plans, a bunch of different ways that they generate revenue from sellers. But basically, the gist of it is that they offer third-party sellers a number of different ways to sell on Amazon. And each of those ways includes typically either

fees, so like a dollar per item sold kind of thing, or some other sort of commission. They'll have professional plans where you could do like $40 a month, I think, to set up the shop. And then it's a smaller commission on each item. And so they really collect more revenue this way. I tried to kind of guess what the profitability was, but I really had no way to do it. I would, my gut says this is certainly margin accretive. So it's more profitable than their traditional first party business. But it's, like I said, they're very discreet. And so they haven't broken out how much they actually generate in earnings from that segment. Other three segments, and these three are kind of the epitome of what Amazon is. It's why it's attracted so much

investor attention. So the first one is subscription services. This accounts for only 7% of revenue, but it's basically made up of Amazon Prime subscriptions. So with Prime, most people already know this, you get access to a ton of different services for $15 a month or $140 a year. They've raised prices on this numerous times. They've added more value to it over the years as well. It's essentially leveraging the costs of all their other segments. So I would think that this should be quite profitable. However, they're also investing so much into it, which I think Brett will talk about later. And I'm not sure that they don't break out where the costs are included, but they're spending a lot of money to procure, for example, exclusive NFL rights for Thursday

night football. So they're really trying to bolster the offering. It's hard to do segment profitability because the NFL rights are supposed to be churn reduction for the whole prime membership, which you make money off in a variety of ways because more people are going to buy, say, they'll buy double the amount of things they would have on Amazon previously. And then that'll be gross profit accretive. But then where is that getting spent? It's a whole mishmash. And as we'll talk about, we kind of try to divide it into two segments, which Ryan is about to talk about the one of the most important one, AWS and then non-AWS revenue. Yeah. That's another way to break it up. But essentially, Prime is just leveraging

their physical infrastructure, I think still probably the biggest driver of Prime subscriptions and maybe Prime Video has grown in influence, but it's probably just the one day or really quick shipping, depending on where you are, which is a great way of kind of turning

a cost center into a revenue driver. The other one here, and this has grown really quickly, it's still growing quickly, is advertising services. So still accounts for a small percentage of revenue, but it's far more profitable. I've seen some estimates that it's 85% to 90% gross margins. They actually have the third largest advertising business in the world. So just behind

Google and Facebook or Meta. Basically, the bulk from what I can tell is driven by product promotions, both on Amazon's online marketplace, but also like authors promoting their books on Kindle, um, artists potentially promoting music on Amazon music. Um, they have a lot of different devices. They also sell ad space on fire TVs. They, I think they sell ads on Alexa. Um, so there are a number of different ads and they offer measurability to the advertisers, but really it's basically product placement that's driving the bulk of that, which is obviously, it's a great way to expand margins on your core retail business. Um, and then I think, uh, before you go on there, I think people should, the way we look at the advertising revenue, at least the sponsored listings, is they're essentially slotting fees that Amazon is trying to take advantage of, similar to like a Walmart or Kroger or tech to try to take advantage of in their stores, except in this case, it's an online store.

Exactly. And then I guess the last one, and this is probably the most important, AWS, it accounts for 13% of the revenue, but it accounts for the lion's share of profitability or profits. It needs very little introduction. It's the world's largest cloud platform. If you don't know what cloud is yet, that's kind of shocking. Basically, data centers across the globe, customers pay for computing power and storage. Very high retention business. We are not cloud experts. I doubt we're, yeah, we're probably going to focus 90% away from there today, but it

has 90% of the value potentially for the company over the next decade. Yeah, absolutely. Well, and if you are interested specifically in cloud, I recommend listening to our interview with Sean Wang, where he is more of a cloud expert and it kind of breaks down why it seems almost impossible. I use that word lightly. It seems almost impossible for AWS not to grow

over the next five years. And so that's obviously an important segment. And then the last one is other. This accounts for less than 1% of revenue. I think right now it probably includes things like Project Kuiper, Luna Cloud Gaming, maybe some devices in there as well. But occasionally something will start to grow really quickly and then it just basically becomes its own segment. Advertising used to be included in other, now it's broken out into its own. But that is the gist of the Amazon complex. And before you go on to the history, quickly, I want to share my screen. And just for any listener or viewer, I try to combine, and this is excluding AWS, so AWS would make this look even better, combine the third-party sellers, subs, and advertising businesses. these are potentially going to be the higher margin businesses than, say, the traditional marketplace. And it's where if you were modeling out the company, or excuse me, modeling out the financials over the next decade, you would try to say, okay, what could the margins be? Maybe they

could be higher than they were in 2018 or 2019. And the numbers I came up with, which just adding these all together, is that on the Q3 2022 trailing 12 month, they were... Oh, let me slide that they were 36% of revenue compared to just 30% in 2019 and then 23% in 2017. So a huge difference. And if you were betting that Amazon's margins were going to, say, expand over what they were in 2019 in, say, 2023 through 2025, this would be what you would be looking at. All right. Let's talk history. I'm going to go a different direction with this one. I think a lot of people are probably familiar with The Founding Story. If you're not, I would recommend reading Brad Stone's two books.

They're called The Everything Store and Amazon Unbound. Both those are pretty good. They tell sort of an intimate story of what's going on internally as the company was evolving. But I'm going to discuss specifically what's happened in the last three years because it paints a picture of why the stock is down. I don't know how much from its highs, but it's down significantly as of late. So I kind of want to give more context around that. So in the first quarter of 2020, COVID hit. I think people probably know that. But unsurprisingly, this spurred a lot of demand for online shopping. And Amazon's business for all of 2020, Amazon's retail sales grew by 39% off of a $245 billion

base. And the year before, so fiscal year 2019, it was only growing by 18%. So it massively accelerated. And then you kind of couple that with stimulus checks sort of halfway through that. It's very hard for management to kind of predict what's going to happen in terms of consumer spending. And at the time, management said basically their fulfillment centers were working at close to 100% capacity. They were trying to lean more heavily on third-party logistics providers. And the logistics providers just said, basically, we're not going to or we can't do the volume that you want us to do. And so, Jassy said in a recent interview that they were forced to either expand capacity drastically, which is going to be costly, or fail to serve a

lot of their customers. And so, being Amazon, they chose to expand. And in the following two years, they doubled the size of their fulfillment network, which they built over the 25 years prior. They also developed a transportation network, which they were planning to build out over the next six to 10 years. They accelerated it into basically two years. And keep in mind, it takes 18 months or longer to get a fulfillment center actually up and running. And in COVID, it was taking around two years. So they really had to guess during 2020. So during the heart of COVID, what demand was going to look like two years out. And he basically said, Andy Jassy said in this interview that they thought, well, at least if we guess higher, eventually we'll grow

into it. However, we're now seeing, and I'm going to talk about this in a second, some excess capacity. So right around late 2021, start of 2022, shopping trends started to revert. People were, this is kind of right after Omicron, people were spending less, shopping more in person. And it led to excess capacity across their fulfillment network. And at the same time, nearly all their variable costs were skyrocketing. So ocean freight, the cost of container was growing really quickly. air that ocean freight went up like 10x there was no room at the ports remember the the porting stuff that was a big deal for like three months they had to in the seattle area didn't they have to lease an entire port in that was not the seattle one to get their distribution uh network flowing correctly i mean yeah that does this gigantic cost i think costco and other companies

bought their own freighters or maybe leased their own freighters so yeah that's expensive yeah the The other ones, air freight was expensive, trucking, so just like ground logistics, line haul, that was more expensive for any third parties. Fuel costs skyrocketed. Plus energy, yeah, or electricity even too, yeah. Energy at their data centers, it grew more expensive, especially around the globe. Um, so they were coupling less per fulfillment center productivity with skyrocketing costs. Um, and basically this led to a compression in margins and led to losses at, uh, across

their retail business. Additionally, they were overemployed at a lot of their fulfillment center. So when Omicron hit, there were a lot of people taking time off. And so they basically double hired for, to fill those roles. And so all this kind of happened at once where they've got rising costs across stuff they can't control, so variable energy stuff. They've got too many employees in their fulfillment centers, and there's weakening demand. So utilization rates or productivity of the fulfillment centers really kind of dropped during this period. However, we're kind of getting through, I would say, to the other side of that now. If you look across cost of containers, air freight, even natural gas, prices are starting to come down. They've come down a lot from sort of the peak, but they're still hovering above their 2019 figures. And then they've pulled back somewhat on CapEx, but you wouldn't be able to see it. So the CapEx is the same year over year this year, but more of it is going towards AWS because there's more demand for data centers, less is going towards the fulfillment network. Do you want me to pull up the chart of CapEx? Do you want me to share that just to see how things got aggressive in the 2019 period?

um yeah and this is a stratosphere chart you can this uh we screenshotted it from their website but this is something you can visualize yourself over there for free um at the click of a button so what is this right here ryan 2019 about 20 billion right right and then we go down to almost 70 i mean that's it's incredible yeah for full year 2021 it was basically like around $60 billion in CapEx. They're expecting around $60 billion for 2022 or a little over. But the mix is more towards AWS, which I tend to think of as more growth CapEx. They're pulling back spend on the fulfillment network. And they've kind of tightened their labor force. So they recently fired essentially 18,000 people on the corporate side, which I know is not a lot

compared to their overall workforce. But for the corporate side, it's not completely trivial. And I think they're in the process of getting their cost structure kind of right-sized, I guess would maybe be the word I'd use. That is the consulting term for firing people, yeah. Yeah, but also on the stuff they can't control, they're seeing that kind of revert as well. Um, so I, I think when we look at 2020 full year, 2022, um, you'll probably see a sort

of a, a sort of a trough in terms of profitability. That is the theory. And I think we'll probably discuss that near the end of the episode, but there are a ton of inputs there. So very hard to, for any analysts to go through and get any exact numbers there since it was a great history, great overview. Since that took a while because it's such a complicated company, I'm going to try to hit the industry and competition fairly quickly. And then management and ownership, very simple, too, because there's no red flags.

I'm going to separate the industry and competition into two segments, cloud computing and then e-commerce slash media. E-commerce slash media will also include devices, stuff like that. As discussed in the Microsoft episode, and if there's any number that any people need to take away who don't know the cloud before, I'm sure this will sound like a broken record here. But the cloud computing industry is large and growing quickly. And the CAGR or compound annual growth rate through 2030 is projected to be about 15%. Yes, 2023 could be slow. Yes, there might be another slow year in there. But on average, the projection is that there will be a 15% growth and hit a total size of $1.5 trillion. dollars. AWS, as maybe we should have talked about on the Microsoft episode, these AWS, the infrastructure providers are not targeting this entire market, but a big chunk of it. And they have 30% plus market share among the cloud infrastructure providers, which gives them a ginormous opportunity. I've had a bit for reference last quarter,

they had $82 billion in annualized recurring revenue. If you analyze annualize the quarterly revenue. The competitors here, one, Azure, that's the number one competitor. Two, Google Cloud, who's a little smaller with 10% market share. And then you got smaller ones like Oracle, IBM, Cloudflare, and then tons and tons of really niche providers. Now within e-commerce, the revenue opportunity is even larger, but probably with lower margins. Amazon divides its regions into North America and international. Just in the United States, the e-commerce market is estimated to be around $900 billion a year. And Amazon's North American revenue was, for reference, about $300 billion over the last 12 months. So let me say that again. Total e-commerce

market in the United States, $900 billion. Amazon North America revenue, which I believe includes Canada, potentially Mexico, but maybe just Canada, was $300 billion over the last 12 months. So they have a sizable chunk of the North American e-commerce market. Now, if we look to Europe, The e-commerce market is essentially the same as North America, but just slightly smaller. I think they're just a little bit behind on the growth trajectory in the market share there. And then outside of North America and Europe, Amazon has a few emerging market bets for its e-commerce marketplace. One of these is India, which I would guess, given how much they talk about it, how much they've invested, is their number one priority for international markets.

That has an e-commerce market of just $75 billion right now, but is expected to grow rapidly this decade. Then if we look at competitors, we could do a whole page. We could do 10 minutes of competitors here. It's really anyone in retail, anyone that's selling you anything. Or conversely, anyone that is looking to power online sellers, which as we know, the big competitor here is Shopify, who does... I think they hit $50 billion in quarterly GMB, which is basically $50 billion in online sales. So a sizable competitor now to Amazon. But yeah, if I listed everyone, that would take a whole page. Let's move to management and ownership. As well publicized in 2021, Jeff Bezos stepped down as the founder. Everyone knows him. He's not running Amazon

anymore. And he is now the chair of the board. He's actually the executive chair. So I think still has a little bit of influence, which we'll see. It's hard to tell though. We're not a fly on the wall on those discussions. The new CEO is now Andy Jassy. They say on the website, this is how they describe it, that he founded Amazon Web Services, which he basically, along with Bezos and the other parts of the S team, they came up with the idea. He started it and ran it until He took over as the consolidated CEO role. If you look at the board, pretty simple, 11 total members, Bezos and Jassy are both on the board. And then all the independent directors seem to just be random executives from other companies.

No red flags there. The board of directors, here's a good one. And this proxy statement, I'm going to say it was the favorite one I've read so far of any company we've covered. the board of directors do not get paid except for discretionary stock-based compensation grants now with some companies that might mean everyone gets paid every year but last year only two of the members got these grants and i'm assuming that might be new ones or the you know lead maybe independent director of say the compensation committee or something like that who had to do a lot of work um and if we look at the amazon leadership team they call it the s team it's huge

They have 28 members now across its divisions. And in the newsletter, I will link to a GeekWire post that goes through all of the different positions, vice presidents, executives, stuff like that. And if you look at it, I think it's a really good indicator of what they are focused on and what their priorities are. Because you have some people from Alexa, you got AWS people, you have India people, I think, or maybe it was international e-commerce and stuff like that. Now, if you look at compensation, the executive team gets paid, in their words, simple compensation, which again, we're fans of, right? Sometimes it's way too complicated over in the executive compensation area. First, they get a modest base salary. And this seriously does mean modest. No executive got paid more than $200,000 in cash last year.

Isn't that amazing? Trillion dollar business. No one got more than $200,000 in cash. Can you imagine reading this as, say, an analyst at an investment bank and you're getting paid more than Andy Jassy? You kind of go, huh? That'd be kind of crazy. I mean, well, we'll get to the stock grants here.

Yeah, it's not actually like that, but it is funny to look at it. They are very modest with their cash compensation. And the second thing they get are stock awards. I have a quote from it, but basically they are time-based restricted stock units and they vest and they are based on... So essentially the stock price goes up or down,

they'll get less or more, if you get what I mean. So if the stock price goes up, they'll get more in compensation. It's kind of what, yeah, that's all I guess matters really to investors. If you look at total exec comp in 20 or last year's proxy statement, it was $352 million. However, this includes the 10-year performance stock plans for executives that got promoted last year like Andy Jassy. So his $212 million in estimated quote compensation is the stated value of all the stock awards

at, say, the current stock price that he can earn over the next 10 years. So in reality, it's actually pretty modest compensation, I would say, for a company of this size. Ownership, pretty boring. Bezos owns 12.7%. And then essentially every other fund in the world, the index funds own it, and then all the large funds own stuff. Really, the proxy statement is great. No red flags I can find. I guess the only small thing that annoyed me was that Bezos gets paid... Oop, I misspelled his name. The company pays for his security every year. I think it was $1.6 million. That's fine, but he could pay for himself. He's the richest man in the world. But let's move on. Ryan, why don't you get the earnings here as we

try to speed through as best we can the complicated Amazon business? Yeah. I'll try to fight through this but um so i guess in terms of the last 12 months they've done 500 billion dollars in revenue it's growing 10 percent year over year the uh fastest growing segment is still aws um north america has still shown pretty north american i'm gonna call it commerce although it also includes like advertising it includes a lot of the divisions we mentioned earlier under that hood And it's growing at a really impressive clip, given that they are coming off one of the most explosive periods in online shopping or tailwinds that any business could ever see. We're getting a reversion to the long-term average, which was tough over the last year. Yeah.

The North America segment still makes up the majority of revenue, 61%. International makes up 24%. AWS makes up about 15, although, as I mentioned, AWS makes up the lion's share of operating income. And then when you actually look at operating income as a whole, only $13 billion in operating income over the last 12 months. It's coming down quickly. That's down 54% year-over-year. And I believe this is in the last 12 months, operating margins in North America were negative 1%. Operating margins in international commerce or international were minus 6%.

And then they had 30% operating margins at AWS. So basically at this point now, you can kind of see indicators that the costs are improving before the quarter actually shows. So the quarter is going to give you almost lagging results. So it's really important to pay attention to the conference call, I think. They are still basically unprofitable in both of their North America and international segments. However, that doesn't actually paint the perfect picture because North America and international, more so North America, that includes the high-margin stuff like advertising and prime and third-party seller services, I assume. So the actual core just retail operations of selling goods to consumers and shipping it to them must be incredibly unprofitable when you think about the buoy that they're probably getting from those other high-margin services. However, I mean, it's not like, you know, they built out those high margin services.

So it is, you know, it's a testament to them, but it kind of just shows the lack of profitability that they're seeing right now. Do you want to say something? Yeah, sure. So viewers can see this, but if we look at their consolidated operating margins, they, about 2018, if I'm looking at this correctly, hit about 5% and they were stable until later in 2020. And now the consolidated operating margin, even with AWS becoming a much larger part of this business, have gone back down to, over the last 12 months, about 2.5%. So, yeah, like Ryan mentioned, the costs have hit them hard recently. Yeah, and I guess right now also, just in terms of most recently what's going on, international sales were down pretty big. A lot of that's due to foreign exchange impacts.

however there's well there's also kind of a recession going on in europe also but the fx impact is only on revenue it's not really on income because they're taking those dollar or not dollars wherever the local currency is and they're reinvesting it back into their local geographies so they report it as lower on it on on their and or on their quarterly results but it's not really impacting their income but well i i would say yeah in the local areas but their corporate expenses are in some of the most expensive markets in the world, New York, Seattle, DC. So they have those corporate expenses that overhead is going to be a lot more expensive when the Euro goes down by, what was it, like 15% at one point last year?

Yeah. Yeah. But nevertheless, it's really not affecting their income nearly as much as their sales. So you're not going to see as big of a discrepancy there. And then just in terms of margins. I mentioned that both North America and international have gone negative. Margins have compressed a little bit at AWS too. There's a number of reasons for this. And I think Andy, or not Jassy, but the CFO on the conference calls talks about this. They're going to fluctuate. And so when they see big kind of demand, they're going to get more customer service reps. They're going to invest in more sales people. They're going to expand out to new data centers, right now they're also having higher utility costs. So they're seeing some

compression in the margins, but it's still basically fluctuating between 29% or 35%. I think over time, that's probably you're going to get north of 30%, but it's kind of a perfect period for them to be investing into it. Other things I'll mention, the number to pay attention to really in terms of profitability is going to be less, their free cashflow minus repayment of finance leases and finance obligations. I know that's kind of a word salad, but basically Amazon pays for a lot of its equipment through finance leases, meaning that they buy the assets over time instead of upfront. So you think about like a warehouse, something like that. So ignoring those finance leases can make free cashflow look better than it is.

So over the last 12 months, they've had basically negative $30 billion in free cash flow once you deduct those principal repayments. So you're seeing them basically really hurting on the cash flow line. However, a lot of that is from $60 billion in CapEx. A lot of that's going towards AWS right now. More and more is going towards AWS over time. I would think that moving forward, it's going to be predominantly AWS. And so any sort of increase in CapEx from here, there's some that's going to be maintenance, but a lot of it's going to be growth. You can kind of think of it as growing demand for their cloud platform.

The other thing I wanted to talk about... Yeah, Brett, you look like you want to say something. I was going to say, I wouldn't say maybe predominantly might be a little strong because there is CapEx in the e-commerce stuff, especially internationally, but it might be more of a 50-50 mix going forward, maybe even higher for AWS a few years from now. I believe north of 50 is already going towards AWS, but it's been sort of half and half

although with them shrinking the CapEx that they need for the fulfillment center right now I would think that AWS tends to grow its percentage but I want to talk about last time if you listen to the Microsoft episode we wanted to kind of figure out where we wanted to see how much of that just kind of try to take a guess

at how much of the CapEx for AWS or Azure in that case was going towards maintenance versus growth. And so I wanted to talk about the depreciation around the servers because that is fairly important. There was actually an announcement last year that Amazon is extending the useful life of their servers. And they're extending it from four years to five years. So that's actually going, that's going to help, that helps gap earnings, but they're going to be depreciating less on an annual basis. And the quote from the CFO said,

we've been operating at scale for over 15 years, but we continue to refine our software to run more efficiently on the hardware. So the way I look at it is that, and basically it's improving unit economics at their data centers. Yeah, it's a good, yeah, extending that life. Some people might think it's a accounting gimmick.

I see a lot of people talk about that, but I kind of think of it as a good thing. And also, they would pay less in taxes if they depreciate, if they kept the useful life of four years. Correct. And their first slide on every page is the cash flow statement. So we always know that their priority is cash flow, but they have to report the accounting the best way they think it's, yeah, the best way they know they can. Yeah, I would say they were more probably forced to do this than anything else. But let me just briefly talk about the balance sheet. I know I've been going long, maybe throwing too many numbers at you. Cash, let's just look at it,

59 billion. Basically, it's all money market funds and AAA corporate debt, which is interesting. Usually, I see more investments in US treasuries, but they like finding corporate debt, I guess. Um, some of this is debt that's, that doesn't mature for more than 12 months. And so as rates rise, they're not going to quite see the, the impact on their interest income as quickly if they had short-term bonds. So, um, they're only really getting this quarter. If you annualize the interest income, it's only about 2% on 2% interest on their cash base.

It should go up just because rates are higher as they progress through the maturities on the things they're invested in. As for the liabilities, the pure debt, so in terms of notes issued, they've got $63 billion, so almost equal to their cash balance. 90% of that was issued in the last five years. They've got varying maturities, but it's basically the weighted average maturity is 14 years from now, so it's going to be around 2036. But some of the maturities extend all the way to 2060, and the effective rate varies. If you annualize, like I did with the interest income, the amount they're paying on interest expense over the year, they're paying about 3.9% right now. So not too crazy. And that's going to go down over time as they pay back some of their older, more expensive debt. The other thing that's worth paying attention to,

they have operating and finance leases, as I mentioned. So they acquire assets or purchase assets through those finance leases. You can choose what to add. If you're looking at one of the aggregators like Stratosphere, you're going to see them add both the operating lease liabilities and the finance lease liabilities. So that adds about another $70 billion in debt. So you're going to see about a $70 billion discrepancy or difference between the market cap and the enterprise values. The enterprise value would be about $70 billion higher. If you exit out, it's not the biggest difference in the world, but if you exit out the operating

leases, you basically have a pretty flat EV to market cap. But you're going to go through that here in the valuation. So I'll pass to you. And yeah, I have this chart loaded up. They really, and again, there's a lot of ways you can calculate their net debt. they were basically net debt neutral or negative up until about 2017, 2018. Now, recently, it's really skyrocketed. And I think some of that is due to the pandemic because they wanted to accelerate their growth. But when you have to accelerate your growth, you can't fund it as much

internally. And yeah, their liabilities have just gone up quite a bit. And I wonder, as we look at this chart here, their net debt skyrocketed over the last couple of years. I wonder what it's going to look like a few years from now. But yeah, let's go through the valuation. Pretty simple. I got a couple of different metrics here. We just do market cap as we look at today, about $972 billion. And then if we use net debt, the way I calculated it, which is adding in some of the operating lease liabilities, again, look at all their commitments, look at all their debt. It's something we don't really want to cover comprehensively on this whole episode. The way I did it, I have an enterprise value at today's price of about $1 trillion or about $1.04

trillion dollars. I'm looking at three different operating metrics here. I'm looking at EV to operating income, which is 80. Then I'm going to look at EV to AWS operating income, and that will be 45. And then I'm looking at EV to free cash flow, which is negative 40, which right now kind of just tells you that they're not generating cash. I think the big takeaways from looking at these two is one. Actually, I think there's just one takeaway is that I don't think you can invest in this business and say, I'm only buying AWS today. It's very expensive still on just AWS profitability. You need to have a thesis on what the margins will be on the non-AWS segments. and there was maybe a time

like I would say a couple months ago where you could have said that no it's we're barely I mean it was let's pull up the stock I'll pull up the stock price while you're talking I thought it was at one point you were getting like 30 to 35 times AWS

but maybe if oh wait I'm not sharing my screen it's not too far away from the look at the stock can you see that So bottom to like 80. Yeah.

So we're not, we're not that far away. I mean, I guess a little bit closer, probably we're in the thirties on AWS operating income, but still that's expensive for a more maturing business. Maybe that's cheap on AWS five years ago. It's certainly probably cheap,

but today they're not going to be growing at 30%, 40% revenue forever. And if they do, well, this will be a man. It'll be a big business one day. Okay. All right.

Anecdotal evidence. Yeah. We have to have it here. Ryan, what's your anecdotal evidence here? Yeah, I'm a customer across a number of categories, and I'm going to be for a long time. Not really a ton to – I think everyone's going to have some sort of anecdotal evidence, so I won't go too long on that. But I wanted to turn this into gut feel. um maybe sort of like a gut takeaway from looking at our research my gut says that if you ask that of all the big tech companies say meta microsoft apple amazon and alphabet yeah who would have who

would be a better business so who would have a wider moat in five years i'd pick amazon number one because it's a the highest likelihood or who will have the widest moat so or who has the highest likelihood of growing their moat the yeah why okay yeah yeah and i think it's just because of like no one wants to buy the business right now but i think the the investments that they've made have have deepened the moat yeah and it would be even if people know about the one in e-commerce I mean, you hear every dad investor going, I see those blue vans. I mean, they're everywhere. Like, of course, that's part of it, right?

But even in AWS, there's a fantastic boat, I think, that no one except Microsoft or Google Cloud in the West will be able to compete with. Specifically, they talk about the new chips they make. We don't want to go through all the details. I don't have the numbers in front of me. You can look at them on their earnest releases. They highlight this sometimes in their annual letters. they make their own chips now for some of their data centers. And the cost savings were actually

way, way higher than I would have expected, like 30%, 40% energy cost savings. I mean, that can be distributed. I mean, that's just pure economies of scale there where who is going to be able to replicate that? One, no startup is going to have the pure scale of AWS. And two, no one's going to have the scale, again, outside of Microsoft or Google Cloud to have the capabilities and the R&D departments or however you want to define it to make their own chips in-house or excuse me, design their own chips in-house.

Yeah, I agree. What about you? Yeah, there's no way I'm leaving Amazon as a customer unless they materially make the experience worse. Like it could stay the same as it is right now and there's no way I'm leaving. I start my buying experience on their website or app for everything and i want to see what your kind of anecdotes is here i start start my buying

experience there for everything except food and then clothing give or take depends depends what type of clothing i think the mode is so large that they it just as an example they have a dedicated they built in my apartment building they have their own delivery setup where the packages get put in these secure lockers i mean you can't replicate that but yeah what's your what's your is it is it your starting point for a lot of retail purchases i'd say it's for anything that isn't like um if i'm not shopping with the brand first in mind amazon's probably the place i go so like if i'm you know clothes is probably the easy one like if i'm buying something on lulu you know a lululemon product i'm gonna go directly to the website or adidas shoes if i know i want

adidas shoes i'm gonna go to adidas that kind of thing um but for all the other stuff like the gimmicky stuff the stuff i need that i don't really think about or don't categorize like i'm going straight to amazon i don't buy my i don't buy my food on there but um pretty much everything else yeah it's just kind of like the digital mall yeah they are good for dry goods as well get those subscriptions if uh i don't know you just get it once a month and uh they give you some insane discounts. All right. Future growth opportunities though. There are a ton here, but Ryan, what do you think? I think you had a fun one here. That's really new. Yeah. So buy with Prime for me. And I think it's actually really, and I might be overstating or

overestimating how important it is to the business, but I think it can be super valuable. So buy with Prime was launched in April of 2022. So April last year, it's only been available to merchants so far by invite only, but it allows US-based Prime members to use the Amazon checkout and delivery experience through another merchant site. So if you're on, let's take Lululemon, for example, I know they don't, I don't think they're a part of the program, but if you were on Lululemon and you were shopping online and you thought, well, I'm buying something it's going to take two weeks to deliver to my door. But then you saw buy with Prime in the checkout process. You can go in there and I think it would obviously be kind of a no-brainer if

you're a Prime member to use that. And so they just had this announcement by January 31st, they're rolling this out to all eligible US merchants. So now people can sign up now instead of getting invited, right? Yeah. And I think there's a bit of a process, but if you're a seller, then it's worth it. And so they had a report that basically said they had really good success with ByRuth Prime so far among the people they invited to the program. Merchants saw a 25% increase in shopper conversion. This is super valuable. It's a win-win-win-win. It's a win for everybody minus Shopify. The consumers get the checkout process of Amazon. They get the delivery of Amazon. The merchants convert more shoppers because they know that they're going

to get their items sooner. And Amazon fills all the excess capacity or a lot of their excess capacity that they have at their fulfillment center. So they increase the unit volume, which is one of the biggest hiccups that they're having right now in terms of profitability. So that's why I say it's a win-win-win. It's a masterstroke, really, I think. If they execute well, it's a masterstroke. Yeah. It was like...

And it's good for the website builders that we just talked about in December. Unless you rely on payments like Shopify, it's going to be good for them as well. They don't feel threatened that everyone's going to build their internal Amazon store. You can have your internal Amazon store, and then the same process is on your external one as well. And there's also now a part two where I think they talked about rolling this out. They're integrating Amazon reviews onto... Sellers can integrate reviews into the checkout process from Amazon. So if you have a shop on Amazon and you have a bunch of good reviews, you can put that on there so that shoppers kind of feel validated, which should only increase the conversion further. yep and that kind of this highlights why amazon is always focused on the customer first and the customer is not the merchant it is the people buying the things where shopify is always focused on i think the merchant as the customer which is a is not the best winning proposition over

the long haul because in this case it also benefits the merchant in this case it does yeah which is why i think it's a masterstroke all right i'll hit my future growth up and that is... I mean, it should be AWS, but we don't need to cover it again. I'm going to choose advertising. Ad revenue includes sponsored display and video advertisements across Amazon's property. So this includes sponsored listings, but also advertising on Prime Video. But mainly, it's the e-commerce website, right? So since breaking out this segment, we've learned that advertising revenue has grown at 46% annually since 2019, and now makes up 7.1% of overall sales at high margins, like Ryan mentioned, it's probably 85% gross margins. Advertising revenue

will be clear help, I think, for consolidated operating margin expansion over the next three to five years. If this goes from 7% to say 10% to 12% of revenue, that's a lot of leverage they can put on here. But as we talk about in the highlights and lowlights, as a nice segue here, they do like to spend, and we'll get to the downside of Amazon, and it's a big one, is they spend and spend and spend on experimental products, projects, excuse me. So highlights on the lights, Ryan, what do you like and dislike about this one? Yeah, the most obvious one is the infrastructure that they've built over 27 years, roughly. Just the physical footprint, the logistics capabilities,

or the logistics network enables them to offer prices and delivery times that no one can compete with. It's not even close. I mean, Walmart is maybe close, but I think they've kind of fumbled the bag. So I don't see how anyone can replicate them in retail. I think the moat's clear, at least on the e-commerce side. And you're saying in North America, though. International, it's a different story.

Yeah, North America. They've also done a great job adding margin-accreted business lines, which has further enabled them to invest more in the infrastructure, which so far has been at the expense of the bottom line. But by having the more margin of creative line items, they could deepen the moat. You're doing everything you can to say flywheel

without saying flywheel. Yeah, basically. Third one, I like the management team. I like Jassy, which apparently is controversial to say now because people want Bezos back, which I think is just unfortunate timing probably for Jassy. I think Jassy is a clear and rational thinker. I think the CFO is pretty good too.

He's the only one that's on conference calls now, and he does a really good job communicating the important points. And they also strike me as honest. They don't seem like they're trying to hide anything. Yeah, and it comes back to the process statement and the culture. or excuse me, the culture is exhibited in the proxy statement. All the highlights we saw from that, why we think it's the best one we've looked at so far, that leads into one-on-one with their honesty and how we think they operate with high integrity. I will say, I one time was randomly

paired up in a golf round with a VP at Amazon, and they said that Andy Jassy was the smartest person they've ever met. So take that. That's your investing thesis right there. But I think, yeah, you got dealt a tough hand for these years, the last two years, right? But it seems like they've done a pretty good job to navigate this and come out strongly on their side. But what's your low light here though? Well, I think we're basically going to have the same ones, which is they spend too much. I think they waste a lot of money. They have a weird obsession with devices um weird huge and it's not it's not small it's ginormous you're gonna you're gonna give the number out here yeah there's rumors that alexa burns about 10 billion

dollars a year the alexa segment um they're also buying irobot for 1.7 billion um they i mean the fire phone was obviously a flop but there's a bunch of devices that they've tried out that are just kind of like you know it makes you question stuff and and it's also like it's not your core competency apple is going to crush you google is going to crush you google slash android samsung is going to crush you how how do they think they're going to win outside of maybe fire tv at the time at the time the fire phone might have looked like the right bat because it maybe wasn't clear who the winners were going to be but the there are some things that they do that seem stupid from the jump where it's, and I'm going to list, I'm going to list some this way.

Yeah. So that's, that's a low light. The other part that's a low light for me is you see some of these numbers around how much they're spending on certain divisions. And this is now, I didn't even put this in here, but the accounting being so complex and the, I guess, mask that they put over the numbers, like they're not, they don't want to show you what the margins are in each division make it harder for me to tell. It makes it harder for me to value the business because all I see is those big headline spending numbers on these random projects. And I don't know what kind of margins they can generate. They could generate potentially. Or what ROIs they're getting. Yes. You have to put more trust in the management team at this point.

All right. My highlights. One, talked about already, corporate philosophy and culture is fantastic for setting up incentives to drive long-term value creation for all stakeholders, and that includes shareholders. Second one, with the growth of the higher margin businesses, like I mentioned, going from 23% of sales in 2017 to 36% over the last 12 months, Amazon should theoretically have room to expand its consolidated margins. If we look at the 2019 margin profile, it was 5% consolidated operating margins with these segments growing as a percentage of revenue, you would think it could get to 10%, especially when you add in AWS. And one, I'll add my last highlight, it's just AWS. We haven't

talked about it that much, but again, it's an obvious highlight. The business is probably going to be worth a trillion dollars within a couple of years. Not much else to say there. Lowlights though, one, media and content division. I'm not sure that other than the lock-in they get from the super fans for various sports, which again, paying for the sports content rights are going to be probably breakeven anyways, and the value isn't actually in the sports rights. It's across the prime subscription. I don't think the content spending is going to drive better long-term economics for the prime business. Oh, but I didn't finish this thought. Inverting the situation, and I knew what I was going to say here. I'll finish it for the newsletter.

if you inferred it and said okay, there's no Prime video how many people would churn from Prime subscriptions? I don't know how many. I would certainly, it wouldn't matter to me. Yeah, you don't have the boys

I guess that's a great TV show, but so what? There were some data minuscule data around basically they just gave out some points on the latest press release around how many

Prime members, both the Lord of the Rings and the Thursday Night Football attracted. And it seemed like it drove a lot of Prime members, but I don't know how. Yeah, but both of those are a billion dollars a year in costs each. So,

yeah. But it might have been worthwhile. We have no idea is the problem. Yeah, that is the issue. I think it can be valuable if it can be valuable but I think

if the only way it is is if they raise prime subscription prices which they have and they probably will they probably will again

yeah alright my other low light and this is the big one Ryan mentioned it already is the wasteful spending across so many R&D projects now if I talk to an executive

there if I had the lucky chance to talk to one of them I'm sure they would argue that these projects will create value over the long term

you just can't see it today they have experiments that are going to be the next AWS I think that's what they would talk about, or the next advertising division or whatever. But my big question is, why do they need to do so many all at once? I'm going to list them off here as quick as possible. All of the projects they are working on outside of their core businesses, I'm sure I'm going to miss some. And these are just from the past two years of press releases.

One, Fire TV and Wi-Fi devices. Two, Alexa and Alexa connected devices. Three, Kindle and reading devices. Four, Ring and other home security products. Five, consumer robotics, like the Astro Robot and the acquisition of iRobot. Six, just walk out technology and Amazon Go slash fresh stores. Seven, Whole Foods. Eight, Prime Air drone deliveries. Nine, Amazon Music. That also includes podcasts. They acquired Wondery, which is a podcast studio, and a live audio service, which apparently had 300 employees. Ten, acquiring MGM Studios. Eleven, amazon care they're also acquiring one medical 12 project kuiper which is a satellite internet service similar to starling hasn't launched yet uh what are we on 13 luna cloud gaming service

i mean what are you getting uh 14 amazon games which is a video game production company or and published slash publisher 15 amazon style physical fashion uh slash apparel apparel store It's a fashion slash apparel store in person. It's not physical fashion. 16, Amazon Pharmacy. What did they acquire? PillPack, right? They have a pharmacy division. 17, electric vehicles with a riveting commitment and riveting investment. 18, self-driving vehicles. that these are all just from the past two years these are all ongoing today here uh we can't talk about this forever but what's one that needs to go in your opinion a one that you think should stay as i'm sure we have the perfect perspective without any internal data on what ones are working or not but for fun what one do you think needs to go and what one do you think needs to stay i'm gonna go luna i think that's a waste of money and they already fired they fired a bunch

people that's probably gonna yeah it's probably gonna go it's gonna get shut down here soon what one needs to stay what one do you think again we're not we don't know but one one in your opinion do you think has the most promise oh well i think i think having an electric vehicle fleet would save them money yes i think along that could be yeah and i mean they're committed to that now And it also, I think, I remember reading that, I mean, there's probably benefits like ESG-wise, maybe a lower cost of capital for doing stuff like that. Yeah. All right. I'm going to go, I would have said Luna Cloud. I'm going to go Alexa. I think it's a total bust. And then I think for valuable one, I'm going to say Project Kuiper. I think if you added that on as a feature, as Amazon Prime, that's quite valuable for someone. You can get internet in remote spots. outside of your home. All right.

Yeah, of everything else, that seems to have the most, like if you're looking at what could be an AWS, that would probably be the biggest one. Project Kuiper? Yeah. Yeah, although they're behind Starlink,

but we'll see. All right, bold case, Ryan, what are your thoughts? Well, we've kind of alluded to it this whole time, which is the higher margin products or services that they have are making up more and more

of their revenue. So if they're able to basically just get back to some level of profitability within their commerce, this is going to have, I think it could have 10% operating margins. And it's obviously growing. There's a lot of businesses within it that are growing, but overall, in aggregate, it's growing quickly. i i mean i think i think the the ultimately it's a 10 operating margin business with a much higher revenue line i don't think you have to think that much more than that the problem is sort of what's it going to take to get there and when you look at i would say basically you just have to get higher productivity and utilization out of your fulfillment centers or fulfillment network which

just going to come from a combination of increasing unit volume and maybe maybe diminishing the capacity that you actually have so to kind of find that balance i think there there's a number of reasons that they're going to be able to get the unit volume up one of which would be by with prime yeah uh i think i have a similar bull case very simple it's just recovering margins back to slightly above 2019 levels. I think that is very doable with AWS, ads, subscriptions, and third-party sellers being a larger part of the business today and will likely be an even larger part of the business in 2024, 2025. At 10% operating margins on a $600 billion revenue base, which for reference, what are they? They're just over-

20% more revenue. Yeah. That's just 20% growth, which is very doable within a couple of years, if not 2023 or 2024, that's $60 billion in annual operating income. Current enterprise value is $1 trillion. Those are the numbers right there. Let's go back. Let me just show, because a lot of it's around variable costs. That's one of the difficult things that people are having a hard time assessing. Let me show, and God, Zoom always this is the chart of cost of containers and you can see basically it's come down significantly it's it's it was around 1500 from 2017 all the way till 2020

it shot up to 10 000 11 000 dollars i'm not sure what the metric here is but basically I think it was up 7x, 8x. And now it's come back down 80% to a little over their 2019 level. And then it looks very similar for the air freight. And natural gas prices have come down as well. I think there's a number of variable costs. And I don't know how they buy. I don't know what contracts they have with other logistics providers or energy suppliers, anything like that. but I have a hard time seeing those costs not come back down and margins revert. Yep. I agree. I agree. Yep. All right. Bear case.

It's probably short term more than anything, um, which it was basically if it takes a while for retail to get profitable again, or there's any sort of a consumer slowdown, which means less utilization rates across the fulfillment network, uh, And they have another year like this year, and I think this could be a losing stock. I don't really have any long-term concerns, I guess. yeah i the only one is that they just this is my bare case is that they even if they give say a good effort they find it impossible to rein in their spending um i kind of as bezos famously said was this a one-way door decision to be so experimental with all these

things at the same time. And can they bring that in and become profitable? I think it's likely they can, but I worry that they can't. And then we just, they keep running at breakeven forever. And that's... Maybe this day one thing was like one of the worst jinxes. Maybe it's so ingrained in the culture that they think any sign of positive cashflow is ruining cashflow for the future yeah it's too much that's at one it's too much at one point yeah day two is here yeah well i'm more interested this is probably the one that i'm watching the closest of all the big tech businesses that i don't own which is just the only one we own is google so um i mean it's we've talked about it between the two of us a number of times, feels like a great moat, a widening

moat, and AWS is a wonderful business. That's kind of maybe the thesis in a nutshell. Yeah, I'm in consensus with you there. Price isn't too bad. There's a lot of moving parts, but you can see a path because of the 2019 numbers where once we get through all this stuff, margins will be fine. And again, yeah, the moat is so strong. So you have that combination of management, durable, competitive advantage, attractive price.

That's the three criteria we look for. And that's probably why we're very interested today. Actually, as a full disclosure, we don't own it today. It's possible we could own it in the future. That's why the disclosure on these shows are stock pitches. They're just stock analysis. But let's go through the disclosures. First up, the stock for next week is going to be Meta Platforms.

and then after that, Salesforce and Alphabet. Okay. That's going to do it for this episode. Thank you all for listening. Remember, we are not financial advisors. Anything we say on the show is not formal advice or recommendation. We are general partners at Arch Capital and clients may hold securities discussed in this podcast. Thank you all for listening. We'll see you next time. Thank you.