Ahead of the curve with Ulrike Hoffmann-Burchardi
The desk believes that the U.S. economy is set to benefit significantly from the expanding AI market, a sentiment bolstered by NVIDIA's projections of a $600 billion AI spending this year, potentially growing to $3-4 trillion annually by 2030. Per the full note from UBS, this growth is not just optimism; it reflects robust underlying structural and cyclical factors, including a revised Q2 GDP estimate of 3.3% driven by strong consumer spending. The consensus view suggests a readiness among equities to capitalize on this positive economic trajectory, despite some historical skepticism regarding sustained high growth rates.
What the desk is arguing
The desk supports the bullish outlook for U.S. equities, underlining the potential of AI investment to catalyze economic growth. According to Ulrike Hoffmann-Burchardi at UBS, the AI infrastructure spending is projected to grow at an annual rate exceeding 38% over the next five years, which aligns with the desk's expectations for sustained economic momentum.
Key evidence from UBS indicates that recent GDP revisions have strengthened this outlook: the U.S. Bureau of Economic Analysis increased its estimate for Q2 GDP from 3% to 3.3%, emphasizing robust consumer spending as a driving force. This aligns with broader trends of technology investment fueling economic dynamism, hinting at a potentially fertile environment for institutional traders seeking growth opportunities.
Where it sits in our coverage
Our consensus target for the U.S. dollar against key currencies reflects both optimism and caution, settled at 1.075, with a range of 1.04 to 1.12. Specific firm targets include: - jpmorgan: 1.10 by Mar26 - bofa: 1.04 by Mar26
Current sentiment aligns with jpmorgan's optimistic positioning, while contrasting with bofa's more cautious stance. Given the desk's expectation for growth, it's likely positioned at the upper end of various forecasts, suggesting potential for upward adjustments in targets among firms leaning bullishly.
How other firms see it
Firms aligned with this view include jpmorgan and others expressing optimism about U.S. growth trajectories due to investment in technology sectors, particularly AI. On the other hand, firms like bofa exhibit contrarian sentiment, expressing concerns over the sustainability of such rapid expansion and its impact on inflation.
Close observation of currency pairs like EUR/USD and indicators such as Fed interest rate decisions will be critical, as they could provide insights into the broader implications of U.S. economic performance amidst global inflation narratives.
How firms align with this view
Aligned with the desk view
Contrary positioning
Key takeaways
- 01NVIDIA's projections highlight a potential $3-4 trillion annual market for AI by 2030, indicating significant future growth.
- 02Revised Q2 GDP estimates from the U.S. Bureau of Economic Analysis show an upward adjustment to 3.3%, driven by consumer spending.
- 03Historical parallels, such as 19th-century railroad capex growth, support the feasibility of sustained high growth rates in technology investments.
Market implications
Monitor the trajectory of the USD, particularly against the EUR/USD pair, which may reflect market reactions to evolving expectations around Fed policy shifts in response to the booming AI sector. Positive economic data could strengthen the dollar and drive equities higher.
Risks to this view
A significant deviation in economic data, such as disappointing consumer spending figures in subsequent months or a Fed pivot towards interest rate hikes due to inflationary pressures, could jeopardize the current bullish outlook and lead to increased volatility in FX markets.
Hello and welcome to Ahead of the Curve. My name is Ulrike Hofmann-Borchardy, CIO for the Americas and Head of Global Equities for UBS Wealth Management. In the thicket of news noise, where are the market signals from last week and what should we look for next week?
Signals last week were all about growth drivers for the U.S. economy, both structural and cyclical. First, NVIDIA's CEO, Jensen Huang, quantified the size of the AI opportunity. In NVIDIA's earnings call, he predicted three to four trillion of annual AI infrastructure spend by 2030.
If we assume 600 billion for this year, this would translate into an annual growth rate of at least 38% over the next five years. How realistic is this in our view? We tried to answer a similar question in our signal over noise note from May this year.
Instead of estimating the dollar spend, we ask ourselves how much AI brain activity will we need by 2030? And for this, we computed the number of floating point operations per second that we believe AI will process. We predict 20 zettaflops, which would require five times the current install base of Compute.
This is not directly comparable to an annual dollar spend as the price per flops will come down and the install base for Compute comprises multiple years of revenue, but it does suggest quite a similar growth rate. Now the question is, is there any precedent for such a sustained growth rate? This is what psychologist Daniel Kahneman calls the outside view necessary to guard against overconfidence.
One historical analogy that comes to mind is the railroad capex in the early 19th century. In the 1830s, railroad infrastructure investments averaged 31.5% per year and peaked at 2.6% of GDP in 1854. So sustained 30% plus capex growth rate is not unprecedented.
Secondly, on the cyclical side, last week brought more positive news on the US economy. The Bureau of Economic Analysis revised Q2 GDP estimates up from 3% to 3.3%, boosted by stronger consumer spending and private investments. So despite a weaker labor market, the US consumer and economy show resilience.
There's one piece of news that hit late Friday that made headlines, but will likely only have a modest positive effect on equity markets. And that is the decision by the Court of Appeals to reject the use of emergency powers for tariffs. The small impact stems from three reasons.
One, the tariffs stay in place. Two, the government can seek Supreme Court review. And three, the US administration has already pivoted to sectorial tariffs, which are not subject to the Emergency Powers Act.
So as we move beyond last week, where do we look for signals next week? First off, some statistics on September. September seasonals are the worst of the year.
When you look at monthly S&P returns, since its inception in 1923, September is the worst month of the year with about 1% loss on average. It is then followed by a mediocre October and the seasonally strong months of November and December. So statistically, September tends to be volatile.
Turning to data next week, the most important macro data point will be non-farm payrolls on Friday. We expect a number below 100,000, which should be enough to firm up a rate cut of 25 basis points on September 17th. The most relevant AI data point next week will be Broadcom earnings on Thursday.
In focus will be its AI revenue, which has become its growth engine and is now a third of its overall business. It is the company with the second largest AI semiconductor revenue after Nvidia in the US. Though it's important to keep proportions in mind, Nvidia derives more than 10X the revenue from AI compared to Broadcom.
And another distinction, Broadcom builds custom AI chips, whereas Nvidia's AI chips are general purpose in nature. And after a solid print from Nvidia, Broadcom earnings are more directional guide for custom chips than overall AI demand. One last thought on valuations.
It's one of the most frequent client questions I get combined with the question when the correction is coming. Valuations by themselves are not a catalyst for correction. One example is the NASDAQ in the late 1990s.
It reached a first new valuation high in late 1998 at a PE of about 60. It then took another 15 months for the market to make new high after new high or 40 turns of PE until it reached a peak valuation of over 100 times PE. What ultimately catalyzed the correction was a combination of increasing fed fund rates, telco spending strains, and dot-com cash burn that emerged in the first quarter of 2000.
At this stage, with rates trending lower and CAPEX still on the rise, our advice is to use any September volatility to add to US equities, in particular financials, AI, power and resources, healthcare stocks, and US quality bonds, such as investment-grade bonds and agency mortgage-backed securities. With this, stay well and stay ahead. UBS Chief Investment Office's investment views are prepared and published by the Global Wealth Management Business of UBS AG or its affiliate UBS.
This material has no regard to the specific investment objectives, financial situation, or particular needs of any specific recipient and is published for informational purposes only. As a firm providing wealth management services to clients globally, UBS AG and its subsidiaries offer both investment advisory services and brokerage services. Investment advisory services and brokerage services are separate and distinct, differ in material ways, and are governed by different laws and separate arrangements.
In the USA, UBS Financial Services, Inc. is a subsidiary of UBS AG and a member of FINRA SIPC. For information, please visit our website at ubs.com forward slash working with us. For a full legal disclaimer applicable to the independent investment views produced by UBS, please visit our website at ubs.com forward slash CIO dash disclaimer.
Sources & References
How we cover this story