Source: Company Information; Leverage Shares analysis

• The online stores segment (40% of net sales in 2023), which grew at 5% Year-on-Year (YoY) last Fiscal Year (FY), is currently trending to cede these gains by the end of the current year.

• The third-party seller services segment (24%), which grew a massive 19% YoY last FY, is trending to gain only 2% this year.

• Advertising services (8%), which grew at a solid 24% YoY last FY, is trending to grow 4% this year.

• Subscription services (7%), which grew 14% YoY last FY, is trending to grow 8% this year.

• Physical stores (3%), which grew 6% YoY last FY, is trending at 4%.

In comparison to business of «buying and selling», Amazon Web Services (or «AWS») only accounted for 18% of net sales last FY. However, it also delivered a whopping 66.8% of total operating income – thus making Amazon «primarily» a cloud computing business which continues to gain prominence as AI spends continue apace. At the halfway mark, the segment-wise sales contribution metrics remain largely unchanged except for online which ceded 2% to AWS.

However, not everything is doom and gloom in the business of «buying and selling». The company’s «International» division has gone from a drag of (i.e. «a negative») 7.2% to total operating income last FY to a positive contribution of 3.9% as of H1. However, the «North America» division, which had recovered strongly last FY relative the massive 23% drag in 2022, declined from delivering 40% of income last FY to 34%. All told, both divisions together have improved in income contribution from 33.2% last FY to 37.4% in H1 2024.

Massive investments were made to drive this level of operational improvements by richly drawing upon the vein gold described by AWS’ continued YoY success over the course of several years. While no division is running negative in income performance, the fact remains that a strong payoff from all the successive years of investment – as evidenced by the massive growth in segments outside of AWS last FY – might be a blip. In the business of «buying and selling», competition is plentiful and unconsolidated, i.e. there’s no single giant to slay.

The company’s content business is a particularly interesting ball to unravel. As a «tech» company, the company’s debt is financed at a lower rate of interest than a «media» company’s is since the latter is considered to be in a «riskier» field of business. This means that the company can ostensibly produce «original content» more cheaply. In the area of «original content», two original series stood out: «Fallout», based on a popular award-winning series of video games, and «Mr. and Mrs. Smith», a reboot/retooling of a popular movie released in 2005. «Fallout» attracted a staggering 2.9 billion minutes of views within five days on Prime Video while «Mr. and Mrs. Smith» accrued 964 million minutes in the first three days of the pilot episode’s availability. Both series contributed 33 of the 62 Emmy nominations the company earned in the current year.

The classical argument that the company could have made about bundling content with the business of «buying and selling» is that those who come for the drama stick around to buy goods. This argument might have found some (but not strong) support in the YoY trends in Subscription Services vis-à-vis other businesses related to «buying and selling». However, what weakens the argument is that both these massive shows were released in H1 2024 and there is no evidence that those who came onboard stuck around and ordered substantially more wares from the company’s platforms than before.

The greater the number of unique customers, the more sophisticated the aggregate behaviour. Both customers and merchants (both «brick-and-mortar» and online) have numerous avenues to interact with each other; Amazon currently finds itself hitting a ceiling on «inviting» more sales for simply that reason.

In June, the company had noted¹ the challenge posed by Chinese platforms Shein and Temu, which have been making steady inroads in the company’s addressable markets all across the Western Hemisphere and often at cost to the company’s network of merchants in place. Even company CFO Brian Olsavsky noted during the earnings call that the company missed revenue expectations on account of consumers on average choosing to buy cheaper products, leading to lower average selling price (ASP).

Towards that end, the company is working on plans nearly two months ago that would be launching a «discount store» in the vein of Dollar General Corporation and Dollar Tree, Inc wherein unbranded apparel, home goods and other products mostly priced below $20 would be available. At least in terms of setup with a roadmap towards addressing a sizeable market, this would likely be quite a costly endeavour. Fortunately, AWS continues to deliver: after a 13% YoY growth last FY, it is trending to grow by approximately the same rate this year and will continue to deliver valuable dollars to fund these «dollar store» analogues.

With a large number of learning models such as Anthropic Claude 3.5 Sonnet, Meta Llama 3.1, and Mistral Large 2 launched on Amazon’s Bedrock AI computing platform along with the likes of DoorDash, Nasdaq and Workday building applications within Bedrock, Arm-powered Graviton4 instances delivering better cost performance on AWS and a host of new agreements with Discover, ServiceNow and GE Healthcare – among many others – the company’s AWS division will likely continue to prosper and deliver. Given an almost-exclusively «corporate» clientele with specific expectations and a higher likelihood of «repeat business», AWS has the opportunity that AMD has just begun to realize (and which was covered in a recent article²).

And, finally, in what would likely need substantial availability of computing as well as financial resources if brought to completion, Amazon-owned Zoox has announced that its prototype robo-taxis are expanding their testing grounds from San Francisco, California and Las Vegas, Nevada to include Austin, Texas and Miami, Florida.

The key factor that drew the company’s stock into the ranks of the «Magnificent Seven» was the explosive growth across segments through FY 2023. If revenues continue to pile in, the costs associated with building – which led to such massive instability in the «passthrough» Earnings Per Share (EPS) in YoY terms gain justifications. If all else stays the same, EPS will likely close 54% higher than last year but stagnation in business segments’ growth imply that sustained EPS growth of a similar magnitude in the years to come will likely be questionable.

This is the rubric that has been driving away institutional growth investors from continuing to hold or buy into the stock. With the loss of sustained passthrough value generation comes the fall in traded volumes and a fall in valuation. Choppiness can be expected around future earnings dates and index rebalance dates since ETF issuers are among the largest blocks of investors holding the stock.

Footnotes:

1. «Amazon plans to launch discount store in bid to fend off Temu and Shein», CNBC, 26 June 2024
2. «AMD Q2 Earnings: Another Nvidia in the Making», Leverage Shares, 1 August 2024