Momentum

S2N spotlight

We are witnessing an exceptional 3-month performance in equity markets. The S&P 500 has returned more than 20% over the last 3 months.

Ever curious, I wanted to see how the S&P 500 performs in the future whenever it breaches a 20% 3-month return threshold. I was expecting this to be a good market top indicator, at least over a year. Well, it turns out there is a reason why momentum is one of those anomalies that make it very hard to apply rationality to.

If you look at the count, there were 389 days that the 20% threshold was breached. I think this overstates the signal, as it is natural for the signal to cluster for a number of days until a number of days roll off.

To avoid the signal clustering, I have deployed a 30-day cooling-off period once the signal triggers for the first time. This looks far more correct; we have a count of 38 now, all the way back to 1920. The results are robust.

I have become obsessed with using very long time frames to test a thesis's robustness. Here I go the other way and look if this momentum follow-through is present over a short time frame. I go back 35 years to 1990. I wasn’t disappointed, albeit that there are only 9 signals.

I am presenting here one signal that appears to tell a very strong story in favour of continued momentum to follow for another year in the S&P 500.

I come across these kinds of sound bites all over the financial press and on X in particular. On the surface they are quite compelling but lack context. It is similar to the current AI craze, where information is available in abundance but still requires a decent degree of knowledge and experience to appreciate its relevance.

Global macro investing needs more context than the time series of the S&P 500 on its own. However, this is still valuable input.

S2N observations

The crypto industry has just crossed above $4 trillion. That is massive considering just over 10 years ago this sector did not exist for investment dollars. I think it is clear that its purpose and value come from a concern around fiat currency and money printing.

The blockchain that launched this industry was what the faithful spoke about, and how web 2.0 was going to be the new world. I have to say AI has come and eaten blockchain’s lunch. I no longer hear about NFTs. I seldom hear about DAPPs. I know that blockchain has a purpose, but the whole industry was built on a solution looking for a problem.

AI is, without a doubt in my mind, a true productivity enhancer. I still wonder if the companies producing the AI engines that we are all using as part of our daily routines will be able to get a sufficient return on their investment to maintain the current valuations.

One thing is for sure: this is a new arms race; the amount of money being invested in infrastructure is staggering and will need to be paid for.

I want to add that I think there will be many companies playing in the agentic space that will disappear in the near future. What I am saying is that every day a new AI this or that is launched. I think the big boys & girls will eventually eat all the little boys & girls lunch.

I want to put on the record that I think the US dollar is going to soon put in a multi-month, possibly year-long rally. But not yet. If you see the dollar chart below, it looks to me like it still needs to make new lows.

One of the reasons I am calling for a rally, but not yet, is because the dollar has been oversold from a technical valuation point of view. The extent of the current discount is open for debate, but it is trading at a discount. Diverging gaps like you see above are not common and don’t last for too long either.

S2N screener alert

The Singapore Straits Times is up a record 13 days in a row. What the hell is going on in Singapore?

S2N Performance Review

S2N chart gallery

S2N news today