Jobs, consumer watch
From the NY FRB Survey of Services
(…) The employment index remained positive for a sixth consecutive month, though it fell five points to 7.0. The wages index came in at 33.3, a level similar to the prior few months, indicating that wage increases were steady. Both the prices paid and prices received indexes were similar to the prior two months, suggesting price increases held steady. (…)
Indeed job postings suggest rising labor demand since June. Next JOLTS could surprise again.
Confirming my own Consumer Watch findings, Goldman Sachs’ proprietary quantitative measure of sentiment around discussions of the consumer on earnings calls “suggests that sentiment around the consumer actually improved sequentially this earnings season and reached its highest level since 2022, suggesting that investor fears appear to rest on overweighting the more negative anecdotes from this earnings season.”
Housing Starts Fell Sharply in July
Total housing starts fell 6.8% during July, a sharper than expected decline driven entirely by yet another pull-back in single-family starts. After out-performing over the past few years, single-family construction is now turning lower amid softening conditions in the new home market.
By contrast, multifamily starts improved for the second consecutive month in July. However, multifamily permits declined and gave back much of June’s large gain, which suggests that multifamily development remains weak despite firming apartment market conditions and prospects for lower financing costs in the near future. (…)
Source: U.S. Department of Commerce and Wells Fargo Economics
“The sharp July decline in the South (-101k) more than led the overall decline (-91k) and was likely driven by Hurricane Beryl, while extreme heat may have also weighed on starts in the West (-36k). Starts also declined in the Midwest (-3k) but increased in the Northeast (+49k).”
Construction was started on an annualized 279,509 units in July, according to data released Friday by the Canada Mortgage and Housing Corp. That’s the highest level since June 2023, and exceeded the 245,000 units expected by economists. Starts rose 15.7% from the previous month, and the increase was mostly in multi-unit buildings. (…)
The country’s chronic shortage of homes has been further aggravated by an influx of temporary residents that has driven population growth to one of the highest levels among developed countries. (…) At the end of last year, CMHC estimated Canada needs to build at least 3.5 million additional housing units by 2030 to restore affordability.
US Office Market Shows Bottoming Signs, Moody’s Analytics Says Sellers are finally capitulating, volumes have steadied
A years-long price decline for US offices could soon be ending as sales volume has stopped falling and large properties are being resold well below prior deal values, according to Moody’s Analytics.
“We are seeing signs that the market is beginning to function in a healthier way,” the firm said in a report dated Thursday. “A significant increase in price discovery paired with reaching a bottom in transaction volume means we could be nearing a bottom in office pricing.”
While “significant pain” has yet to occur and distressed offices will need liquidation at likely significant losses, the report added, “Owners and lenders are in a much-better position to evaluate potential losses today — which is the first step necessary to the market recovering.” (…)
EARNINGS WATCH
From LSEG IBES:
463 companies in the S&P 500 Index have reported earnings for Q2 2024. Of these companies, 78.8% reported earnings above analyst expectations and 16.2% reported earnings below analyst expectations. In a typical quarter (since 1994), 67% of companies beat
estimates and 20% miss estimates. Over the past four quarters, 79% of companies beat the estimates and 16% missed estimates.
In aggregate, companies are reporting earnings that are 4.5% above estimates, which compares to a long-term (since 1994) average surprise factor of 4.2% and the average surprise factor over the prior four quarters of 7.3%.
Of these companies, 59.2% reported revenue above analyst expectations and 40.8% reported revenue below analyst expectations. In a typical quarter (since 2002), 62% of companies beat estimates and 38% miss estimates. Over the past four quarters, 62% of companies beat the estimates and 38% missed estimates.
In aggregate, companies are reporting revenues that are 1.1% above estimates, which compares to a long-term (since 2002) average surprise factor of 1.3% and the average surprise factor over the prior four quarters of 1.2%.
The estimated earnings growth rate for the S&P 500 for 24Q2 is 12.5%. If the energy sector is excluded, the growth
rate improves to 13.3%.The estimated revenue growth rate for the S&P 500 for 24Q2 is 5.4%. If the energy sector is excluded, the growth rate declines to 5.1%.
The estimated earnings growth rate for the S&P 500 for 24Q3 is 5.8%. If the energy sector is excluded, the growth rate improves to 7.3%.
Trailing EPS are now $232.60. Full year 2024e: $242.46. Forward EPS: $258.31e. Full year 2025e: $279.32.
Amid historically high earnings multiples, analysts are clearly setting a very high bar for EPS growth through 2025. Revenues are seen accelerating in a deflationary setting with profit margins rising incessantly.
Goldman Sachs is more reserved on margins:
Our 2025 forecast is below both the bottom-up and top-down consensus estimates. Our forecast for 4% sales growth is slightly below the consensus bottom-up estimate (5%). Analysts also expect net profit margins will expand by 128 bp, compared with our more modest forecast of 24 bp. Our 2025 EPS estimate of $256 (+6% growth) is 8% below bottom-up consensus of $279 (+15% growth) and 5% below top-down consensus of $269 (+12% growth).
Sales growth should decelerate modestly in 2025 to 4% from 6% in 2024. Nominal GDP growth and the strength of the US dollar are the most important drivers of corporate revenue growth.
Our economists expect nominal US GDP growth will equal 4% in 2025, compared with 5% in 2024. Our FX strategists expect the USD will weaken on a trade-weighted basis in 2025, supporting the revenues of firms with international sales exposure. Consensus bottom-up forecasts for a sharp acceleration in sales growth for the median S&P 500 non-Energy stock, which is more sensitive to economic growth and less international-facing, appear slightly too optimistic.
Our top-down model indicates that aggregate S&P 500 profit margins will continue to expand in 2025, but by less than consensus expects.
The key drivers of company margins include revenue growth and the difference between price and input cost inflation. If sales grow as we expect, operating leverage will lift margins. While price hikes have been slowing, so has the growth rate of input costs such as wages.
Index margins are also rebounding from margin contraction in 2023. Our macro model nonetheless suggest more modest margin expansion than in consensus forecasts. We expect S&P 500 net margins will reach 11.7% in 2025, slightly below the mid-2022 record of 12.0% and well below the consensus estimate of 12.8%.
Significantly, recent conference calls reveal that labor costs are no longer pressuring corporate P&Ls. BTW, Goldman this weekend trimmed the probability of a US recession in the next year to 20% from 25%, citing last week’s retail sales and jobless claims data.
Unsurprisingly, business sales are intimately linked with nominal GDP growth:
U.S. nominal GDP grew 5.4% and 5.2% YoY in Q1 and Q2 respectively, significantly outpacing business sales, up 0.6% and 2.1% respectively. This growth gap is not unusual and is generally due to YoY swings in oil prices which directly impact oil companies revenues.
In turn, business sales are intimately linked with S&P 500 revenues growth (Ed Yardeni). More stable oil prices should help close the growth gap in Q3 and Q4 but analysts’ 5.9% revenue growth rate for 2025 seems high.
Goldman monitors S&P 500 revenues ex-Energy deflated by the GDP price deflator (ex-energy).
On this basis, real revenues rose 2.4% year-on-year in Q2, roughly in line with the pace of Q1 and the average over the last six decade but down from +4.6% in 2023Q4. This adds to our conviction that economic activity is rising roughly in line with our estimate of short-term potential GDP growth.
Real revenue growth of 2.5% (60-year average) and inflation in the 2.0-2.5% range (current GDP deflator: +2.3% a.r.) get S&P 500 revenues ex-E rising 4.5-5.0%.
Goldman’s more subdued forecasts for both revenues and margins look more reasonable, suggesting that revisions might be downward in coming quarters.
[Recent earnings] revisions have been less negative than the typical historical pattern. Bottom-up estimates usually are revised down over time. 3Q and 4Q 2024 EPS estimates have been cut by 3% and 1%, respectively, since the start of earnings season. But 2024 EPS estimates have only been reduced by 2% since last March, compared with the historical average of -6% at this point in the calendar year. Similarly, 2025 EPS estimates have been raised by 1% since March 2024, versus the average of -2%.
The macro outlook creates two-way risks to our EPS forecasts, but the bulk of these risks appear skewed to the downside. Prediction market odds of a Democratic sweep are now slightly above odds of a Republican sweep (28% vs. 24%). Our economists expect a Republican sweep would likely lead to the extension of some of the corporate tax provisions of the 2017 TCJA that are set to expire in 2025. Other tax cuts would also be possible, although they would be unlikely to rival the magnitude of the 2017 reforms.
In contrast, a Democratic sweep could result in corporate tax increases. Every 1 pp change in the statutory domestic corporate tax rate should shift S&P 500 EPS by slightly less than 1%, assuming no changes to other parts of the tax code such as the global minimum tax or buyback excise tax.
This somewhat dated Morgan Stanley chart (Oct. 25, 2023) shows the correlation between revisions and equity price trends:
Is it like 2000 again?
From the GMO Focused Equity team:
(…) Investment returns ultimately come from three sources: earnings growth, dividends received, and change in the earnings multiple. Our approach could be summarized as taking care of fundamentals by focusing on quality while controlling valuation risk by seeking to avoid overreach on the earnings multiple.
If we compare today’s markets with the market top in 2000, we see less to fear today on both the fundamentals and the multiples. In 2000, the top-10 stocks comprised five telecom companies, two networking beneficiaries, a software business, an oil company, and a holding company. At the June 2024 quarter end, we have five cloud tech businesses, two chip companies, a pharmaceuticals company, an oil company, and a holding company.
In 2000, the price-to-earnings ratio of the top 10 was 60x. Market participants were intoxicated by the ongoing communication revolution and heady fundamental returns of 22% per annum for the group over the preceding five years. The valuation was simply too rich. Even if the fundamental return expectation of 19% p.a. had been delivered and stock prices held, the top 10’s multiple would have still exceeded 25x by 2005.
In reality the outcome was starkly different – the fundamental return was a relatively puny 8% p.a. for the group, with earnings falling at the telecom companies after what turned out to have been disastrous capital allocation in 3G spectrum auctions. Total returns were famously poor.
Perhaps unsurprisingly, the stellar performer of the group – the only one to deliver a positive return – was the companyth e furthest from the TMT epicentre, Exxon.
2024’s top 10 have delivered similarly impressive fundamental returns of 19% p.a. since 2019, but the set-up is rather different to 2000, with a median P/E of 27x. If today’s top 10 were to deliver on 19% fundamental return expectations and prices were to stay the same this time around, the multiple would collapse to 12x by 2029. Implicitly, investors expect less from the mega-caps now than they did in 2000. In a real sense, the stakes are lower today.
A final “then and now” observation relates to the quality of the two top 10s. GMO’s longstanding systematic quality metric that we use for idea screening was in active service in 2000 (as were the writers of this piece). Based on an assessment of the level and stability of profitability, alongside balance sheet strength, 2000’s top 10 were barely better than the median business.
By contrast, today’s top 10 sits squarely in the high-quality camp, with the midpoint just shy of the 10th percentile globally. That fits well with our fundamental assessment of quality – most of the current crop have stronger barriers to entry, interesting optionality (e.g., in AI and digitalization), and a better record in capital allocation, too.
None of this is intended to be determinative and we are not making another “X-then-Y” case. We are pointing out that in an important way the stars have aligned differently this time around, even if the market is superficially similar from a technical perspective. In the short term, it’s anyone’s guess. But looking further out, whether today’s mega-caps turn out to be great or less-than-great investments will be a result of the evolution of their fundamentals and the consequent impact on their valuation multiples.
Unlike in 2000 when we held only a nonmaterial weight in the top 10, we invest with confidence in a number of these companies today.
What is different from 2000 is the importance of passive investing, boosting demand for the largest weights. The reverse will eventually happen…
BofA Global Research
AI CORNER
https://www.youtube.com/watch?v=7PMUVqtXS0A
Very interesting interview with Eric Schmidt. David asked Perplexity.ai to summarize it but still a great watch:
Based on the search results, here are the key points Eric Schmidt discussed regarding Nvidia, the chip industry, LLMs, agents, and technical aspects of AI:
1. Nvidia’s dominance:
– Schmidt admitted he had underestimated Nvidia’s CUDA software ecosystem, which is now crucial for running major AI models[6].
– He emphasized Nvidia’s key role in AI infrastructure, predicting massive investments flowing to the company[1][7].2. AI investments and infrastructure:
– Schmidt revealed that major tech companies are planning unprecedented investments in AI infrastructure, potentially reaching up to $300 billion[1][2].
– He suggested that a significant portion of this investment would go to Nvidia for AI training chips[1][2].3. AI agents and context windows:
– Schmidt discussed the potential impact of large context windows, AI agents that can learn and improve themselves, and text-to-action capabilities[5].
– He predicted these technologies would have a massive impact on the world, potentially surpassing the effects of social media[5].4. LLMs and AI development:
– Schmidt noted that the gap between frontier AI models and the rest of the field appears to be widening[2][5].
– He compared the current state of AI to the early days of electricity, suggesting that realizing its full potential will require time and significant organizational innovation[6].5. Chip industry and competition:
– Schmidt mentioned Nvidia’s upcoming B200 chip, which is so advanced it requires clean room conditions for assembly[7].
– He noted that competitors like AMD are struggling to compete with Nvidia’s ecosystem, particularly in software[7].6. AI stack and future developments:
– Schmidt described a future where large context windows, self-improving AI agents, and text-to-action capabilities would converge, leading to rapid development in AI applications[5].
– He suggested that these advancements could enable the creation of complex applications, like a TikTok competitor, in a matter of minutes[5][6].7. Technical challenges:
– Schmidt highlighted the massive computational power required for AI development, mentioning that the U.S. might not have enough power to support these efforts alone[5].
– He discussed the potential need for partnerships with countries like Canada or Arab nations to access necessary resources for large-scale AI projects[6].Overall, Schmidt’s comments paint a picture of rapid advancement in AI technology, with Nvidia playing a central role in providing the necessary hardware and software infrastructure. He emphasized the increasing importance of large-scale computing power and the potential for AI to dramatically reshape various industries and applications.
Based on the search results, Eric Schmidt had several key points to make about China, competition, geopolitics, and technology:
1. Urgent need for U.S. action:
Schmidt emphasized the necessity for the U.S. to take decisive action in competing with China technologically, stating, “We have to do whatever it takes”[1]. He believes the U.S. is currently unprepared for the challenges ahead in the AI competition with China[4].2. China’s strategic plan:
According to Schmidt, China has a strategic plan to dominate crucial sectors such as semiconductors, AI, clean energy, and biotech[1]. He noted that China announced its intent to master and eventually dominate these areas, which he sees as the engine for trillions of dollars of new businesses and global markets[4].3. Technological leadership gap:
Schmidt stated that the U.S. is “one or two years ahead of China, not five or 10” in AI development. He pointed out that China is well ahead in areas like face recognition[2].4. Global platform dominance:
Schmidt emphasized that the competition with China is now about “global platform dominance” rather than traditional military assets. He believes that the country controlling global platforms will gain advantages in both national security and economics[4].5. Specific technological concerns:
He highlighted China’s progress in quantum communications and its efforts in quantum computing, expressing concern about potential threats to encryption and secure communications[4].6. U.S. strategy and allies:
Schmidt advocated for a “global competitiveness strategy” for the West to win technological battles. He stressed the importance of working with Pacific partners and suggested that India has become the U.S.’s most important ally in the tech race, as he believes the U.S. has “lost” China in terms of technological cooperation[1][3].7. European challenges:
Schmidt was critical of Europe’s tech innovation environment, particularly Brussels’ regulatory approach. He saw limited hope for European tech prospects, mainly in France[3].8. Urgency and organizational approach:
He stressed the need for urgency in U.S. policy, regardless of whether it focuses on public funding or private-sector initiatives. Schmidt believes that to compete effectively, the U.S. needs to organize itself around these technological challenges[4].
In summary, Schmidt views the technological competition with China as a critical geopolitical issue, emphasizing the need for urgent and strategic action from the U.S. to maintain its technological leadership and global influence.
Our own research led by David suggests that China is actually ahead of the U.S. in LLMs. We also think unwise to invest in companies investing in LLMs: there will be blood. Wiser to do bookend investing in AI: infrastructure (chips, equipment, energy) and apps.
