Keep connected
Our investment management team is made up of engaged thought leaders. Get their latest commentary and stay informed of their frequent media interviews, all delivered to your inbox.
Stocks stumbled into year end, skipping the traditional ‘Santa Claus Rally’ as investors were more focused on the hawkish comments from the Federal Reserve following their December meeting and the potential fallout from the threatened ‘Trump tariffs’ than they were on the potential economic stimulus and reduced regulatory environment from the incoming U.S. administration. Bond markets also sold off again on worries about a resurgence in inflation from the new administration policies. All those concerns were reflected in financial markets as December ended and January began. Large cap growth got hammered, bonds dropped further, and the major stock indices struggled. But even as the big growth stocks were reeling, other sectors rallied to mitigate those losses and signs of strong breadth through January’s first two weeks became evident. Financials and domestic small cap stocks picked up the rally flag right through January and helped push markets in the U.S., Europe and even Canada back to record levels.
The rally got a sharp scare in the last week of the month as DeepSeek, an artificial intelligence open-source model developed by a Chinese hedge fund, unveiled a competitive alternative to OpenAI and the other major AI software developers at a fraction of the cost that the incumbents had required. The upstart stunned the U.S. tech industry with an open-source AI model called R1 that provides the crucial model parameters that others can copy and improve upon, unlike OpenAI, which keeps its models locked in a black box. DeepSeek itself was built on the work of others. The company says its AI was built on Meta’s open-weight Llama model. Microsoft is also looking into whether DeepSeek built its AI with help from OpenAI’s technology, albeit without proper permission. Since launching ChatGPT two years ago, OpenAI has maintained that this business is fueled by a few key ingredients: chips, data and money. The more of those three elements you input, the more powerful artificial intelligence it will produce. Enter a tiny company that blows a hole in that formula, with an AI model that’s just as powerful and built for a fraction of the cost. The bullish view is that reduced costs to deliver AI will dramatically broaden its appeal and uptake. The more negative take is that the hundreds of billions of dollars ploughed into data centre and capacity expansion may have been overspending that may not yield a profitable return. Those spending initiative included the US$500 billion ‘StarGate’ initiative unveiled last week by President Trump alongside Microsoft, Oracle and OpenAI. In its quarterly conference call this week, Microsoft indicated they still plan to spend $80 billion on data centres this fiscal year to help it meet demand from customers for its AI products, though the company expects the growth in expenses to taper off in fiscal 2026. The sharp selloff in big name tech and infrastructure stocks on the morning of Jan. 27th showed the risks in the sector that has lead the market over the past two years and pushed valuations back to record levels. The “what goes up must come down” mantra was on full display as Nvidia dropped over 17% on that day, a decline in market value of almost US$600 billion, the largest single day loss in value for a stock ever! The deployment and growth of AI is far, far from over but the action in the past week must be a clear demonstration for all investors what the risks and volatility can be when momentum, valuation and sentiment get so far stretched in any one direction. The growth in technology did not end with the ‘bubble top’ in 2001, but it did take many of those stocks over ten years to get back to the levels they reached at that time, and we may be entering another such phase.
The overvaluation of stocks relative to bonds and traditional valuations can be seen in the Equity Risk Premium, or ERP, which dropped to zero and then actually went negative. This is a nearly unheard-of event where investors are willing to treat the stock market as a riskless asset class — attaching no risk premium at all compared to the T-bill market, which is perfectly riskless. A ‘zero equity risk premium’ implies that there is the same capital risk in the stock market as there is in the T-bill market. This is the first time in almost 25 years that this relationship has gone below zero, previously happening during the dotcom bubble period. Given that stocks are inherently a riskier investment over Treasuries, it’s hard to want to take an overweight position in the stock market at these levels when you can get paid more in a risk-free asset (10-year at 4.7% compared to the 4.6% expected earnings yield on the S&P500). Equity investors are paying to take on risk rather than getting compensated for it!
But those warnings are falling mostly on deaf ears as shown below in the chart of the positioning in equity futures by asset managers and leveraged funds as reported by the CFTC (Commodities Futures Trading Commission). At year end, net long positions in equity futures were near a record level at US$228.5 billion. Long positions in stock market futures have been rising consistently over the past 2 ½ years after hitting a record low of -US$80 billion in the fall of 2022, which (of course) coincided with a major stock market bottom!
The short and sharp ‘DeepSeek selloff aside, investors are seeing the glass as ‘half full’ when it comes to future expectations. We saw two clear examples of that this week in earnings reports from ‘Mag 7’ members Apple and Tesla. Tesla reported their first year over year decline in auto sales ever and operating margins dropped due to price cuts to deal with competition from Chinese upstarts. But Tesla’s quarterly results drove home the lesson that profit and sales numbers don’t seem to matter much for this stock anymore. Instead, it’s Elon Musk’s narrative that’s wooing investors. Even though the electric-vehicle maker’s fourth-quarter earnings fell short of analysts’ expectations pretty much across the board, the stock closed up 2.9% on Thursday, as investors shrugged off the disappointing report and instead focused on Musk’s upbeat tone on the robotaxi business, humanoid robots and artificial intelligence. Ditto for Apple, which saw 11% iPhone sales decline in China and an overall miss on iPhone sales expectations and services revenue. The stock was down about 3% going into the conference call, but CEO Tim Cook spoke positively about the rollout of Apple Intelligence on the next generation of iPhones and how that would lead to a major upgrade cycle. Sales will grow by a percentage in the low- to mid-single digits, the company said during the conference call. Though that’s in line with the roughly 5% that analysts have been predicting, it was enough to soothe the nerves of investors following the ‘mixed’ quarterly report and push the stock higher. Investors seem more than happy to look through any near-term missteps for these companies due to their belief that the long-term growth is secured. That level of unbiased optimism increases the risk that any ultimate disappointments in these growth expectations will weigh heavily on the stock prices. But for now it seems to be all ‘clear skies.’
Trump’s agenda will be tough to fully implement. This is not 2017! When Trump came to power in 2017, he inherited a relatively strong economy and moderate stock market valuations. He simply put his ‘foot on the gas’ with personal and corporate tax cuts and a substantial reduction in regulatory hurdles. The 2017 Trump tax cuts drove 3% growth in gross domestic product and added nine points to S&P500 earnings growth in 2018. This time around, it will be a monumental task simply to extend those tax cuts, let alone get incremental cuts. The deficit is north of 6% of GDP, versus 3.5% eight years ago. U.S. corporate taxes are now about on par with the rest of the world, and Republicans have only a slim majority in Congress. Inflation in 2017 was under 2% and had been in that range for years versus the more elevated levels today, which could easily start to increase again. Expectations on the potential budget cuts from DOGE are also overblown because social benefits and defence spending are more than 80% of the budget. Between interest rate charges, defence spending and Medicare there is not room to make the cuts promised without support from Congress in changing laws. The U.S. debt is now $36.3 trillion, against GDP of $29.4. The debt-to-GDP ratio is at the highest level since the Truman era, which inherited a mess to fund wars in Europe and the Pacific. The bottom line is that we should expect fewer tax cuts and tariffs than broadcast and that could give a reprieve to decimated bond investors. Reductions in regulations have already been factored into stock valuations, particularly the financial sector. The inability to deliver on these promises could lead to more confrontation with the Federal Reserve for lower interest rates, much like in 2018. Investors should remember that 2018 was not a very good year for the stock market!
The most important issue for stocks and interest rates will be the secular trend in inflation. For decades after Paul Volcker tamed inflation in the early 1980s, the 10-year yield trended downward in the most predictable and important pattern in global finance. Whenever the yield rose to threaten the pattern, a crisis — Black Monday, the Orange County derivative disaster, the bursting of the dot-com bubble, the Global Financial Crisis — would erupt, and yields would drop. While it’s never wise to make too much of drawing lines on charts, many investors and strategists have suggested that this latest rebound in yields show that a new trend is taking shape.
We disagree! The secular trends in inflation over the past 40 years have not ended. They were driven by the combined forces of globalization and technological productivity growth. While the globalization trend may have stalled over the last few years, the productivity gains from the implementation of new technologies has not. Over the last five years those long-term impacts were distorted by the policies of the pandemic, which necessitated a massive period of fiscal and monetary relief at the same time as there was a pronounced breakdown in global supply chains and other disruptions in production. We are now coming out the other end of that adjustment and expect that we will get back to periods of slower global growth and low inflation. Another factor that will keep inflation low is the fact that the largest economies in the world are mature now and growing more slowly, with China on the verge of joining that cohort after two decades of excessive growth.
Globalization drove the massive expansion of the world economy from the 1980s throught to the period of the pandemic. It seems a fallacy to believe that growth will continue to be as strong under a regime of ‘deglobalization.’ Countries benefit from trading with one another by focusing on making the things they are best at making, while buying from other countries the things they aren’t as good at making. In the era that followed, we created the most efficient supply chain the world had ever known. But then in 2008, we had a global financial crisis and a decade later we had a global pandemic unlike anything seen in over a century. The result today is that we are living in an era of deglobalization where the security of supply chains is deemed more important than the efficiency of supply chains. The U.S. could be on the verge of throwing another wrench into the machinery of global trade by enacting tariffs on its major trading partners.
Another interesting chart below which also argues for some caution in the stock market. The corporate insider buy-sell ratio is down to 0.22x, the lowest since data back to 1988. Selling has been centered in technology stocks after their huge run-ups the past two years. Buybacks may be all the rage, but company executives at the same time are dumping their own stock at a very rapid rate! Stock buybacks use the corporate money and can enhance earnings per share so they are an easier decision for corporate boards to make. The personal behaviour of those same executives have definitely been taking a more cautious approach and should be noted!
Of all the measures of the market’s priciness, among the most reliable is the cyclically adjusted price/earnings (CAPE) ratio developed by Yale University economist Robert Shiller, since it looks back a decade and adjusts for inflation. On that basis, American stocks are 83% more expensive than when Bill Clinton first took the oath of office, 145% more than when Barack Obama first did and a whopping four times Ronald Reagan’s starting point. They even are a third pricier than at the start of Trump’s first term
Below we show the excessive valuations investors are paying for the U.S. stocks. The smoothed CAPE multiple at 37x has only been this expensive 3% of the time in the past century — the late 1990s, and again in 2021 ahead of the big 2022 drawdown. This does indeed classify as an official “bubble.” No other stock market on the planet comes remotely close. Canada is actually at “fair value” or “neutral” on this measure. Many others actually have their valuations below historical norms, including Japan, much of Asia, and most of Europe. The major point for global investors who respect valuations, this is not the time to be strongly overweigth the U.S. stock market!
Updating our current investment strategy in light of all our commentary above, we continue to maintain a relatively defensive stance, with a slight underweight in stocks, but with a ‘barbell’ approach that has higher growth technology stocks on one side and high income, low valuation stocks on the other side. In technology, we exited the semiconductor and infrastructure names late in 2024 and switched to the software and user companies that will benefit from the implementation of artifical intelligence. In real terms we moved out of names like Nvidia, Micron, Celestica, Cameco and Capital Power and added to positions in Shopify, Amazon, Meta and Alphabet. We also reduced positions in Canadian industrial stocks that were most exposed to tariff threats such as Bombardier, MDA Space, Algoma Steel and Magna. In the bond market, we continue to find better value in the U.S. bond market, where yields on 20+ year government bonds are in the 4.7-4.8% range versus the Canadian market where similar maturities have yields of 3.4-3.5%. Moreover, the risk to the Canadian dollar versus the U.S. is still to the downside, so that also argues for the higher U.S. bond weighting.