Ferg's Finds

This is a short weekly email that covers a few things I’ve found interesting during the week.

Article

I learn something new every time I jump on this site: A Dozen Things I’ve Learned from Jason Zweig about Investing.

Podcast/Video

I like to listen to Pippa Malmgren whenever trying to rap my head around some Geopolitics: MacroVoices #382 Dr. Pippa Malmgren: A Tale of Two Civil Wars

Quote

“We don’t know what lies ahead in terms of the macro future. Few people if any know more than the consensus about what’s going to happen to the economy, interest rates and market aggregates. Thus, the investor’s time is better spent trying to gain a knowledge advantage regarding ‘the knowable’: industries, companies and securities. The more micro your focus, the great the likelihood you can learn things others don’t.” Focusing on the simplest possible system (an individual company) is the greatest opportunity for an investor since a company is understandable in a way which may reveal a mispriced bet. Howard Marks puts it simply: “We don’t make macro bets.”

Tweet

I like the parallel David Perell draws to Ancient Jewish Law with a Venture Capitalist, stating all their worst investments were made when every single partner on the investment team agreed.

It reminds me of this quote:

“When everyone thinks alike, everyone is likely to be wrong” -Humprey Neil

Charts

China isn’t leading the energy “transition”; it is leading the energy “addition”.

The below framing, “electricity instead of fossil fuels,” gives me the shits (electricity is a transmission medium). In China, 2/3rd of electricity is fossil fuels (63% coal + 3% gas).

Might as well call them coal cars as that’s where the additional base load power to charge them is coming from.

Something I'm Pondering

I'm pondering how we love to extrapolate technologies that fit the current narrative and refuse to acknowledge those that don't.

My favourite example is the IEA’s consistent underestimation of the S-curves for oil demand in China and India.

I’ve only seen AI portrayed in a good light regarding energy, i.e. it will make energy systems more efficient.

I’m interested in how much of this efficiency eventuates vs. the massive power requirements to run AI.

Take this article: How Data Centers are Enabling Artificial Intelligence (AI)

The power density of racks going from 4kW to 60kW seems like a big deal! Hyperscale data centres with power capacities of up to 100 megawatts (which is a small power station or large renewable park).

I hope you’re all having a great weekend.

Cheers,

Ferg 

P.S. If you’re interested in my story and why I started this Substack, you can read the story here.