Gold and silver stocks have never peaked at historically undervalued levels.
Miners are now trading at the cheapest fundamental multiples we have ever seen.
Where are we in the precious metals cycle?
There is no shortage of questions on why gold has significantly underperformed during such an ideal macro setting.
Here is thread 👇👇👇
Gold and silver stocks have never peaked at historically undervalued levels.
Miners are now trading at the cheapest fundamental multiples we have ever seen.
The median free-cash-flow margin is now above 25%.
That is the highest level we have ever seen.
We are experiencing the complete opposite today.
In fact, gold and silver miners just repaid the largest amount of debt in the last 25 years.
PM miners have turned gun-shy when it comes to acquiring new projects or companies.
Remember, mining companies tend to overpay for deals at the peak of the cycle.
We are barely seeing any deals being done today.
We have seen this price behavior happen twice before.
Also, January 2008, the metal hit new highs and continued to appreciate for another month until declining by 28% during the GFC.
The price is now 14% lower, and the entire financial media already claims that gold is dead.
Note, however, how the shiny metal tends to come screaming back after these pullbacks.
Yes, this ratio was higher during the covid crisis, but the current levels are almost as low as it was during other major bottoms.
My 2 cents:
Silver is the cheapest metal on earth.
Miners still look very oversold.
Last times we had such a divergence between the Philadelphia Gold and Silver index relative to its 200 day moving average, it also marked two important bottoms.
Such a phenomenon is unsustainable in our view.
Today, the setup today looks just like it did in the early 2000s ahead 10-year precious metals bull market.
These are important signs that a bottoming is taking place.
Ideally, one wants to see the risker parts of the market not only holding up their values but perhaps even leading the way to the upside.
Here is silver vs. 5-year TIPS yield:
Note how after the pandemic lows, gold front ran the potential risk of a rise in consumer prices and the entire precious metals market appreciated sharply.
The metal was up more than 75% from August 2018 to August 2020 and even reached historical highs during this period.
Now it is a real problem.
We think gold likely appreciated too quick and too fast becoming what some thought as an obvious trade.
We are now on the other side of this extreme.
See below global central bank assets reaching new highs while gold is still lagging.
For us, it’s a matter of time until gold follows to the upside.
H/T @crescatkevin
We should all know by now that the true cost of goods and services is growing at a drastically faster pace than CPI.
Precious metals are now at their cheapest levels relative to other commodities since 2009. The other 2 times this ratio reached such depressed levels also marked incredible buying opportunities.
Overall commodities are leading the way and look ready for another big move to the upside after consolidating.
Gold and silver miners have never looked this cheap relative to the S&P 500.
Their free-cash-flow yield is almost twice the overall market.
The value and growth proposition embedded in miners today is unmatched by any other time in history.
Consider the fact that the mining industry is also a very capital intensive business.
Because of a decade of exploration underinvestment, there have been no major new gold discoveries in the last four years.
This chart is also courtesy of @crescatkevin
The majors have not been replacing their reserves and the industry is about to face a supply cliff.
Courtesy of @crescatkevin again
Thus, supply is constrained, an incredibly bullish fundamental backdrop for gold and silver.
The entire precious metals industry is dirt cheap. Apple’s market cap is 4 times the size of the whole precious metals industry.
It is our job to identify the times when price volatility becomes unwarranted and use it as an opportunity to allocate capital accordingly.
We think this is the case today.
We did not make a firmwide commitment to partner with arguably the most successful exploration geologist to launch a precious metals fund if all we were aiming for was a couple years of strong returns.
Our view is that if there was ever a time to go up on the risk curve in exchange for upside return potential, that time is now.
In our humble our opinion, that could not be further from the truth.
https://t.co/BrUMzuypKA
More from All
How can we use language supervision to learn better visual representations for robotics?
Introducing Voltron: Language-Driven Representation Learning for Robotics!
Paper: https://t.co/gIsRPtSjKz
Models: https://t.co/NOB3cpATYG
Evaluation: https://t.co/aOzQu95J8z
🧵👇(1 / 12)
Videos of humans performing everyday tasks (Something-Something-v2, Ego4D) offer a rich and diverse resource for learning representations for robotic manipulation.
Yet, an underused part of these datasets are the rich, natural language annotations accompanying each video. (2/12)
The Voltron framework offers a simple way to use language supervision to shape representation learning, building off of prior work in representations for robotics like MVP (https://t.co/Pb0mk9hb4i) and R3M (https://t.co/o2Fkc3fP0e).
The secret is *balance* (3/12)
Starting with a masked autoencoder over frames from these video clips, make a choice:
1) Condition on language and improve our ability to reconstruct the scene.
2) Generate language given the visual representation and improve our ability to describe what's happening. (4/12)
By trading off *conditioning* and *generation* we show that we can learn 1) better representations than prior methods, and 2) explicitly shape the balance of low and high-level features captured.
Why is the ability to shape this balance important? (5/12)
Introducing Voltron: Language-Driven Representation Learning for Robotics!
Paper: https://t.co/gIsRPtSjKz
Models: https://t.co/NOB3cpATYG
Evaluation: https://t.co/aOzQu95J8z
🧵👇(1 / 12)
Videos of humans performing everyday tasks (Something-Something-v2, Ego4D) offer a rich and diverse resource for learning representations for robotic manipulation.
Yet, an underused part of these datasets are the rich, natural language annotations accompanying each video. (2/12)
The Voltron framework offers a simple way to use language supervision to shape representation learning, building off of prior work in representations for robotics like MVP (https://t.co/Pb0mk9hb4i) and R3M (https://t.co/o2Fkc3fP0e).
The secret is *balance* (3/12)
Starting with a masked autoencoder over frames from these video clips, make a choice:
1) Condition on language and improve our ability to reconstruct the scene.
2) Generate language given the visual representation and improve our ability to describe what's happening. (4/12)
By trading off *conditioning* and *generation* we show that we can learn 1) better representations than prior methods, and 2) explicitly shape the balance of low and high-level features captured.
Why is the ability to shape this balance important? (5/12)
