Investment Office Logo

The Lystrosaurus

One of the foundational ideas of the Zeitgeist is that measuring linguistic similarity is a powerful way to observe what we are being told matters by those who publish most of the words we read in a given day.

Rusty Guinn
Epsilon Theory, The Zeitgeist, June 17, 2020

Epsilon

An illustration of a particularly dapper-looking Lystrosaurus georgi specimen

 


 

One of the foundational ideas of the Zeitgeist is that measuring linguistic similarity is a powerful way to observe what we are being told matters by those who publish most of the words we read in a given day.

It should be intuitive that the source of that similarity is sometimes reducible to topics. If a single event is dominating headlines, then language that describes that event is going to cause measures of similarity to rise. This is useful, but not especially interesting. You don’t need us to tell you when a topic is dominating headlines.

That is why when we write about narrative in terms of our measures of linguistic similarity we tend to either control for topic (i.e. we look at measures within topical sub-sets of news) or focus on the evolution of topic behaviors over long periods of time. We think these are powerful ways to observe when a story and its associated vocabulary have become common knowledge.

Sometimes, however, concentrating on a single topic can make it easy to miss the connections of a narrative across multiple disciplines. In other words, there is practically no information (by which we mean something that would make you change your mind about something) in the observation that people are talking about the same things. There is some information in the observation that they are using the same language patterns to talk about it, since that implies some measure of other-regarding behavior. But there is a lot of information in the observation that multiple otherwise unconnected disciplines or lenses for looking at the world are applying the same language to those different angles of a connected problem.

You can think of it as the linguistic equivalent of the discovery of the fossils of a handsome specimen like that lystrosaurus pictured above on the Indian subcontinent, China, Africa and Antarctica, a discovery that permitted us to draw new connections between an entire range of scientiUc and cultural topics that went far beyond a pig-like creature from the early Triassic. Like, say, plate tectonics, geology, evolution, and cross-cultural similarities in mythologies and legends.

But just in case the analogy feels like I’ve been playing a bit too much quarantine science teacher (guilty as charged), let me tell you in more practical terms what I’m talking about.

 


 

Yesterday I got three emails.

The Urst was from a long-time friend, a specialist in technology and VC law, who sent me a link to this piece from the Wall Street Journal.

If Inflation Is Coming, the Market Isn’t Ready

The second was an email from a CIO at a Top-5 university endowment who circulates a daily list of what he is reading to friends and colleagues. This one is behind a research paywall, but you get the topical gist from the headline even if you don’t subscribe.

The State Of The Stock/Bond Relationship

The third was an article that topped the Zeitgeist last week, but didn’t make the cut to produce a full-blown article from us (at the time, anyway).

Why It’s Time to Rethink Bonds

If you don’t subscribe to the WSJ, all three of those links may be behind a paywall for you, but it doesn’t matter. I don’t think any of the articles is particularly informative, or at least I don’t think that any of them provides any new or novel insight. What is fascinating to me is that within a week, a professional market research shop, a personal finance writer and a financial markets journalist all took on the question of “the role of bonds.” What is fascinating to me is that within that same week, a bright private markets specialist (but public markets layperson), one of the ten or so most important asset owner CIOs in the US and an NLP algorithm all told me that “the role of bonds” is something that was on their mind and the minds of others.

We aren’t predicting. We are observing.

I can’t tell you how to gaze through the fog of a deflationary shock to predict what the Fed’s unprecedented intervention will mean in the medium- and long-run for prices. I can’t tell you if and when the macro regime will become one in which bonds cease to diversify stock exposures like they have for the past 35+ years. I can’t tell you whether financial advisors and individuals change the way they think about the role of the bonds in their portfolios.

But I can observe that enough people are thinking about it – and enough people know that other people are thinking about it – that common knowledge is forming around the question. If I can be permitted one pretty uncontroversial prediction, it is that the narrative, NOT the reality of inflation or correlation matrices in the real world, will be the force that causes investors to change their behaviors and portfolios.

We’ve been ringing the bell for asset owners and advisers to figure out how our industrialized investment ecosystem and crystallized processes may need to be adjusted to handle change in the narratives of inflation and the stock/bond relationship for a couple years now.

We rang the bell here:

Epsilon1

We rang the bell here:

Epsilon2

 

And we rang the bell here:

Epsilon3

 

And while the uncertainty and opportunities of COVID-19 and politics may be (appropriately) front of mind for investors, while the reality of inflation may feel miles away, we are ringing that bell again.