International | 10:30 AM
This story features MACQUARIE TECHNOLOGY GROUP LIMITED, and other companies. For more info SHARE ANALYSIS: MAQ
Contrary to the naysayers and Gen AI “bubble” doomsters, major research houses remain upbeat, pragmatic and forward thinking on the megatrend, keeping an eye on the details for investors about what matters and where to find the winners.
Part One: All Systems Go
-Bubbles, there have been a few
-Why this time is different
-Industries set to reap the Gen AI impacts
By Danielle Ecuyer
Gen.Ai sparks investing debates
Generative AI has catalysed the long-standing debate around investing in mega cycles and at what cost. The enormity of Gen.Ai on businesses and society remains an open question and much like previous life-changing technologies –electricity, vehicles, telephony, mobile and the internet– only through the passage of time will the full extent of the impacts including winners and losers be unveiled.
RBC Capital Market’s research remains unapologetic in its conclusion re Gen.Ai:
“while we think this will take several years to play out, the proliferation of GenAI is likely to re-shape the generations ahead similar to the way the PC, internet, mobile phones, cloud computing and social media has impacted us today”.
Simplistic narratives often congregate around share price performance, choosing not to examine the full context of what is driving those share prices.
Goldman Sachs confronts these big topics in its latest research in an aptly titled doorstop tome “AI: To buy, or not to buy, that is the question”.
The performance of AI stocks has drawn comparisons to previous major stock bubbles, from the Nifty Fifty in the 70’s, to Japanese equities in the late 80’s, early 90’s and the most quoted TMT bubble in the early 2000’s.
Equally, the same concerns around concentration risk, pull forward of earnings from the future into the present, excessive valuations, competition emerging, as well as over-investment are all part of the colourful spectrum of investing risks in the Gen.Ai megatrend.
Part of the explanation for the bubble narrative lays in the fact that at this initial stage of Gen.Ai, technology companies are at the front of the performance queue at a time when technology stocks have significantly outperformed other market sectors and driven most of the equity returns globally since the GFC.
Goldman Sachs refers to the phenomenon as “tech’s rational exuberance”, built quite simply on the fact earnings per share have “surged” at the expense of all other industries which have largely “stagnated”.
While unpalatable to the more value orientated investors, technology has been the winning trade and has been turbo charged with Gen.Ai.
This is the crux of the question for investors, how do they value the potential gains and identify the correct winners and losers?
Anyone who follows Nvidia would appreciate such questions are the most debated around the durability of the world’s leading chip/CPU maker for Gen.Ai with an 80% market share in data centres.
History often rhymes
Goldman Sachs breaks down history to make comparisons and concludes AI is still not a bubble across valuations. The broker’s analysis suggests this is the one factor that separates this cycle from previous ones like the internet mania boom and bust.
The average valuation of the seven largest technology companies stands at about half that of the major seven in the 2000 tech bubble.
More importantly, the financial structure of today’s tech giants far exceeds the previous bubble.
Taking the Mag7 versus the Tech bubble leaders (2000), net debt to equity on average is around 30% now versus circa 46% back then; return on equity at 46% exceeds the 28% in 2000 and net income margins of 28% against 16% also shine.
Financially, today’s companies are sounder, even if they represent a higher market concentration risk at 31% of the market versus 19% then.
The striking difference shows up in the capital intensity. AI is underpinning a major capex boom which poses a risk to the high rates of return the technology capital light models of the noughties brought forth through cloud and software, at a time when funding costs were repressed in a low-interest rate regime.
Given the huge spend in AI systems across data centres, PCs and other hardware, this refresh cycle is prompting many commentators and investors to request “show me the money” i.e. when will the tangible returns for stakeholders emerge in terms of return on AI investment?
Currently the hyper-scalers, such as Microsoft, Alphabet, Amazon, represent 23% of the total capex and R&D spend of the S&P500.
To answer the question, Goldman Sachs’ analysts draw on the Microsoft example from 2013-2016 when the company was spending up big on the Azure (cloud) infrastructure.
Margins were negative for a while, but Azure is now a growth engine for Microsoft. This cycle the analysts argue the company’s AI’s revenues of US$5bn-US$6bn have scaled more rapidly than the seven years it took for Azure.
While defending the Mag7 as part of the Gen.Ai trade, Goldmans stresses the importance of diversification for investors.
Just how many industries will benefit?
The broker shines a light on the ancillary industries which stand to benefit from the technology. Healthcare and biotech are highlighted as major winners, with the scaling of genome sequencing and data processing for the development of vaccines as two examples.
Healthcare analysts believe AI has the potential to accelerate data generation for drug development and diagnostics.
Banks and financial services could see benefits from multiple AI related services. An estimated increase in return on equity of around 200 basis points from AI adoption has been calculated.
Citi has run a similar ruler over the impact of AI adoption on banks’ profitability and estimates Gen.Ai could add US$170bn or 9% to the global bank sector profit pool by 2028.
This broker sees improved productivity gains from automation of routine tasks, streamlining operations as well as content and information management, coding and software.
RBC Capital Markets takes the Gen.Ai potential even further, quoting Bloomberg projections of 42% compound average growth rate in Gen.Ai revenue growth through to 2032 over US$1.3trn. Within that, over US$280bn stems from software expanding at 69%.
In highlighting sectors which will be impacted, it would be easier to list those that aren’t, the potential effects of the technology are so wide reaching. RBC points to healthcare; financial; aerospace and defense; industrials; software; internet; payments, processors and IT services; information and commercial services; global autos and auto parts; consumer; metals and mining and real estate.
This broker’s deep dive reveals those stocks which the analysts’ believe will be major beneficiaries: Accenture, Adobe, American Electric Power, AES, Amazon, California Resources, Crowdstrike (pre major outage); DigitalBridge, Eaton Corp, GoDaddy, Hubspot, Meta Platforms, Microsoft, Moody’s, nVent, ServiceNow, Shopify, Thomson Reuters, Verisk and Vistra Group.
In Australia, Infratil ((IFT), Macquarie Technology ((MAQ)), and NextDC ((NXT)) are highlighted.
Goldman Sach’s list of Ex-Tech Compounders’ includes Aristocrat Leisure ((ALL), ResMed ((RMD)), Cochlear ((COH)) and CSL((CSL)).
Part Two will look at the adjacent investment sectors of data centres, power generation and the clean energy challenge for GenAi.
Find out why FNArena subscribers like the service so much: “Your Feedback (Thank You)” – Warning this story contains unashamedly positive feedback on the service provided.
FNArena is proud about its track record and past achievements: Ten Years On
Click to view our Glossary of Financial Terms
CHARTS
For more info SHARE ANALYSIS: COH - COCHLEAR LIMITED
For more info SHARE ANALYSIS: CSL - CSL LIMITED
For more info SHARE ANALYSIS: MAQ - MACQUARIE TECHNOLOGY GROUP LIMITED
For more info SHARE ANALYSIS: NXT - NEXTDC LIMITED
For more info SHARE ANALYSIS: RMD - RESMED INC