How to trade a split personality market

Preface: Explaining our market timing models 

We maintain several market timing models, each with differing time horizons. The “Ultimate Market Timing Model” is a long-term market timing model based on the research outlined in our post, Building the ultimate market timing model. This model tends to generate only a handful of signals each decade.


The Trend Asset Allocation Model is an asset allocation model that applies trend-following principles based on the inputs of global stock and commodity prices. This model has a shorter time horizon and tends to turn over about 4-6 times a year. The performance and full details of a model portfolio based on the out-of-sample signals of the Trend Model can be found here.



My inner trader uses a trading model, which is a blend of price momentum (is the Trend Model becoming more bullish, or bearish?) and overbought/oversold extremes (don’t buy if the trend is overbought, and vice versa). Subscribers receive real-time alerts of model changes, and a hypothetical trading record of the email alerts is updated weekly here. The hypothetical trading record of the trading model of the real-time alerts that began in March 2016 is shown below.




The latest signals of each model are as follows:

  • Ultimate market timing model: Sell equities (Last changed from “buy” on 26-Mar-2023)
  • Trend Model signal: Neutral (Last changed from “bullish” on 17-Mar-2023)
  • Trading model: Neutral (Last changed from “bearish” on 15-Jun-2023)

Update schedule: I generally update model readings on my site on weekends. I am also on Twitter at @humblestudent and on Mastodon at Subscribers receive real-time alerts of trading model changes, and a hypothetical trading record of those email alerts is shown here.

Subscribers can access the latest signal in real time here.



A bifurcated market

In the past few weeks, I have heard different variations of a similar message from professional investors when asked about their portfolios: “We recognize that the AI transformation is very real and AI-related plays could soar, we are staying with defensively oriented names in our client portfolios.”


Translated: We are afraid of being left behind in an AI bubble, so we have a part of the portfolio in those names. The core portfolio consists of high-quality value stocks.
Indeed, the stock market has become bifurcated.


The accompanying chart tells the story of bifurcation by style and quality. There are different ways of 


measuring quality. One simple way is to compare the performance of S&P and Russell indices in the same market cap range. S&P has a much stricter index inclusion criteria than the FTSE/Russell indices, which creates a quality spread as the Russell Index will have a higher proportion of unprofitable companies.
As the top panel of the chart shows, growth has outperformed value in 2023. Within the value universe (second panel), the high-quality factor is dominant. By contrast, low-quality is dominant within the growth universe (bottom panel).




That’s the essence of the split personality the stock market is exhibiting. Investors appear to be chasing speculative growth while holding high-quality value stocks in a barbell portfolio.


This time isn’t different

From a fundamental perspective, it makes perfect sense that AI will be disruptive to the way we work in the coming years, much like how the internet disrupt life in the late 1990s. The only question is how the AI stocks are priced and their upside potential.  


Large-cap technology stocks are already showing signs of extreme froth. These stocks, which are made up of the technology and communication services sectors plus Amazon and Tesla, comprise about 42% of the weight of the S&P 500. The ratio of the NASDAQ 100 to the Value Line Geometric Index, which represents the “average” listed stock, is at an all-time high and far exceeds the peak set in 2000 at the height of the internet bubble. If you think that comparison is too extreme, consider the NASDAQ 100 to Russell 2000 ratio, which set a new cycle high and hasn’t achieved these levels since late 2000.



These ratios are indications that tech valuations are already stretched compared to the rest of the stock market. If you don’t think the market is frothy, FT Alphaville (free registration required) reported that a company called Asset Entities is offering a forthcoming set of AI chatbot digital NFTs.

By contrast, the relative performance of cyclically sensitive value sectors are weak and all are in relative downtrends. (Consumer discretionary stocks were excluded from this analysis because of the significant weights of Amazon and Tesla, which are regarded as growth stocks).




Trading growth

The textbook approach to trading high-octane growth stocks is to employ a high turnover price momentum strategy. Buy the stocks that are rising. If they falter, sell them and go on to the next momentum candidate. While that should work well in theory, price momentum hasn’t been a dominant factor in recent price performance. There are several momentum ETFs available, and none of them are showing any signs of outperformance, which is a worrisome sign that the latest AI frenzy is faltering.



In a glass half-full or half-empty debate, bulls can argue that while the NASDAQ 100 to S&P 500 ratio looks stretched, the price momentum of the ratio (bottom panel) has barely started rising. If an AI frenzy is real, it’s barely started when compared to the 1990’s experience.


Growth and momentum investors should consider an important macro risk factor. My quality analysis shows that low-quality is dominant within the growth universe, which is reminiscent of the froth experienced the late-stage bull environment in 1999, when virtually every internet startup projected that it would be EBIDA positive within two years, indicating that they weren’t profitable then. When the economy fell into recession, the resulting credit crunch wiped away an entire universe of internet startups that were burning cash and needed continuing new financings to stay solvent. Should the economy experience a downturn today, the same effect is likely to devastate unprofitable tech startups, no matter how promising their technology might be.


Fast forward to 2023. While Q2 isn’t quite over just yet, the basket of unprofitable technology stocks tracked by Goldman Sachs is exhibiting its second consecutive quarter of double-digit gains. The last time this happened was during the tech run in 2020.



Recession risk is elevated. As a reminder, Bloomberg recently reported a warning from JPMorgan strategists based on the divergence between equity and bond market expectations:

“Bond markets are still pricing in a sustained period of elevated macroeconomic uncertainty, even if there has been some modest decline over the past three months,” strategists including Nikolaos Panigirtzoglou and Mika Inkinen wrote in a note. “By contrast, equity markets look ‘priced for perfection’ with the S&P now above a fair value estimate looking through the rise in macroeconomic volatility since the pandemic.”



Value opportunities

On the other hand, if you are a value investor who isn’t convinced of the NASDAQ and technology hype, where can you hide and find opportunities? The accompanying chart of regional relative returns tells the story. The U.S. equity market violated a rising trend line and it is consolidating sideways. By contrast, Japanese equities staged an upside relative breakout from a long base, indicating strong upside potential. Tactically, Japanese equities are seeing strong fund flows and the Nikkei Average just rose to multi-decade recovery high. Investors may want to wait for a pullback before committing to a full position.

Eurozone equities also staged an upside breakout and they have pulled back but remain above the breakout level turned relative support. I pointed out in my recent publication (see A global market review: Risks and opportunities) that Chinese sector rotation is signaling a cyclical rebound in spite of the dire headlines. European exports are highly sensitive to the Chinese economy and should benefit from Chinese economic strength.



Here are two specific examples of European sensitivity to the Chinese economy, though they don’t necessarily buy recommendations without further due diligence.
BASF is a classic example of a German industrial with high sensitivity to China. The share prices of BASF (in USD) closely tracked Dow Chemical, another commodity chemical company, before the onset of the Russo-Ukraine War. Even though BASF has significant operations in China, the conflict devastated the margins of the company’s European operations because of the high cost of natural gas feedstock. However, the bottom panel shows that the BASF/Dow Chemical ratio has been recovering and it has been in a slow but steady relative uptrend ever since.



LVMH is another European company with strong sensitivity to the Chinese economy, as its outlook for the sales of its luxury goods depend a great deal on the high-end Chinese consumer. The shares have pulled back and the relative return pattern in the bottom two panels is showing violations of rising relative trend lines and tests of relative support, which are signs of technical caution.




However, an analysis of insider activity shows an astounding level of insider buying on weakness (green dot = buy, red dot = sales). Hermès, another European luxury goods producer whose charts are not shown, has a similar technical price chart pattern and positive, though less enthusiastic, pattern of insider buying.



In conclusion, the U.S. equity market is becoming very bifurcated. Leadership is composed of a handful of frothy growth names while value and cyclicals are laggards and signaling recessionary conditions. Investors who want to trade growth stocks can use the price momentum factor. Investors who are seeking better opportunity should consider Japan and Eurozone equities.


4 thoughts on “How to trade a split personality market

  1. Interest rates and AI.
    We are chasing shadows.
    First, interest rates….big focus on them and the Fed and inflation. But how much do they really matter? They do of course because it is the cost of money. Back in the 70s, did 5% interest rates kill inflation? Seeing as how they had to go much higher, it was not enough, so what is so special about 5% now?
    I think it is more about credit, is credit expanding or contracting? Do we have good debt or bad debt? It’s a mix of course, but bad debt defaults which destroys credit. Increased interest rates slows credit expansion because people borrow less as rates go up, or am I missing something? Rates go up if borrowing demand rises enough, but the other side of the teeter-totter is that rates going up of their own accord should deter borrowing. So what happens when we have record debt and increased rates? More defaults, less borrowing, credit contraction coming?
    AI sounds great, so does robotics, they should improve productivity…at the cost of jobs. It’s why we have a service economy…the production jobs have gone away, AI and robotics will reduce those jobs, and we may lose service jobs as taxis, and food deliveries etc get taken over by smart robots. Sure we can subsidize the jobless, and create more reality shows, but this does not make wealth, and who will buy all the stuff AI produces when they are supported by the earnings of AI. Something about laws of thermodynamics is buzzing here. Well, we can always try more debt. Borrow to pay the unemployed masses. Look at what has happened to federal debt since the 80s, the % of the working age population that is employed (not unemployment)…AI will not make those numbers better.
    Inflation has many inputs and what matters most right now may not matter in 10 years. On shoring and energy costs matter, as do resources…these are not likely to go away…technology helps against inflation.
    Equity prices are influenced a lot by money flow. Even though for every buyer there is a seller, there is a demand or selling pressure (if that is the right term) that influences the direction of prices. Right now there is a lot of sentiment to buy AI, but maybe next week we see panic to get out for whatever reason…you know, the elephant freaking out because of a mouse.
    But we have a lot of debt, and I suspect a lot of it is bad debt. This cannot go on forever, but it can go on for a longer time than we think.

  2. Cam what do you think of the buy signal on the monthly MACD of global indices like Dow Global and Wilshire 5000? You shared it in other occasions as an useful tool for marking long term entry / exit points. Thanks.

  3. The best performing momentum ETF is Fidelity FDMO. It rebalances every three months rather than six. MTUM just rebalanced in May so it was in the wrong stocks until then. MTUM is now in the AI stocks.

  4. My thinking on AI stock timing is that institutional portfolio managers must show MESFT and NVDA plus other AIs, in their client portfolios for the mid-year June 30 update. So expect lots of buying and little selling until then. After the end of the month, is when a correction, if there will be one, will happen.

Comments are closed.