Subscribe
Share
Share
Embed
Around the Desk: This is where we at Avory think out loud, challenge narratives, and look for signal through the noise. Each episode, the Avory & Co. team dives into what’s moving markets, how companies are performing, and where opportunities may be forming. We break down earnings, macro trends, and investor sentiment — all from the lens of a concentrated, high-conviction portfolio.
*** The views expressed on Avory Podcast: Around the Desk are those of the hosts and guests and do not constitute investment advice. This podcast is for informational purposes only and should not be relied upon to make investment decisions. All investments involve risk, including potential loss of capital. Avory & Co. may hold positions in the companies discussed.
Speaker: All right, welcome back
to Around the Desk, where we
explore where the world is headed.
I am Shawn Emery, founder and chief
investment officer here at Avery and Co.
We actually just wrapped up
discussing something as a team
here, internal and external.
People were in the room, and I
thought what was discussed was worth
bringing here to the podcast, and
I'll do a little bit more of that.
You know, it really started you
know, as a conversation about one
company you know, but by the time
we finished, we actually weren't
talking about that company anymore.
I think we were talking more about
insurance markets, artificial
intelligence, and how I think technology
may fundamentally, you know, change
the way society transfers risk.
You know, before we jump into that,
one quick note, you know, we spend
a lot of time asking questions,
pressure testing our own ideas.
That's exactly what this podcast is about.
You know, it's a conversation about
frameworks, long-term trends, really
how we're thinking about where the
world is headed, where it's going.
You know, nothing discussed here should
be considered personalized investment
advice, legal tax, or insurance advice.
As always, our views can evolve as new
information becomes available over time.
Do your own research.
All right, let's start, you know,
really with the question, why are
prediction markets suddenly so valuable?
We saw the, you know, roughly
forty billion dollar valuation that
Kalshi was getting, and I think,
again, that is what brought up
this conversatine- conversation.
You know, th-they received tremendous
amount of attention recently.
Most people describe it as prediction
markets, a place really where,
people can trade on future events.
You know, things like, um, will
inflation be above a certain level?
Will a political candidate win?
You know, will interest rates move, you
know, this direction or that direction?
Uh, you know, obviously go into sports.
Who will win a particular championship?
Uh, and when most people hear prediction
markets or market, you know, they-- their
mind, uh, immediately, probably for the
right reason, honestly, goes to betting.
You know, sports, politics, speculation.
Honestly, that was, you know, I think
most people's first reaction too.
So there's-- You know, I think
that's the right reaction.
But then, as we sat here talking
more and more about it, um, you
know, I think ultimately we got
to a different question, which is,
what if prediction markets aren't
really about, predictions, let's say?
And ultimately, that
changed the discussion.
Because once you ask that
question, I think you start seeing
something much, much larger.
Um- You know, I took a bunch of notes
and I think, going through some of
these things I think are interesting.
First off, like, what is a market?
You know, I think we, we tried to look
at what is a market to, to really get
down to the, um, foundational elements
of, prediction markets and, you know,
that whole ecosystem more broadly.
You know, whenever we make Or evaluate
something, an investment, I think you
have to start with first principles.
That's what we do here,
you know, at Avery.
Not because it sounds intellectual,
you know, because I think
it's ultimately practical.
It-- if, if you understand what
something fundamentally is, I think
it's much easier to understand
what it could eventually become.
Again, the concept of know what
you own and why you own it.
So let me ask, you
know, a simple question.
Maybe you think about it in your own mind.
You know, what is a market?
I think most people would probably answer
a place where buyers and sellers meet.
I think that's fundamentally true.
But I actually think markets perform
a much more important function.
You know, they not only are a place
for buyers and sellers to meet,
but markets produce information.
Think about everything
markets already price today.
You know, if you look at, uh, you
know, the screens on TV, one I'm
looking at right now, you see stocks
pricing businesses, you see bond
markets pricing credit, uh, you
see commodity markets pricing oil,
copper, nat gas, currencies, prices,
uh, uh, you know, money, let's say.
Options are a really good
source of pricing and consuming
and understanding volatility.
Real estate markets price land, um,
and, you know, whatever, construction,
you know, may be on top or not.
You know, but every one of those
market I think exists because people
have different information, and those
different opinions, those incentives
that maybe the different constituents
have, different time horizons.
You know, some participants may know
shipping, others may know our agriculture,
some may know technology or politics.
You know, millions and, and, and
really, like, billions of decisions,
but probably more hundreds of
millions, compressed in one constantly
changing signal, and that is a price.
Now, markets aren't always perfect.
We know that.
You know, sometimes they're very
emotional, uh, and you get, you
know, very big dislocations that
are much more emotionally driven.
Sometimes they're just simply irrational.
You know, you go back to '99, you go back
to, you know, some other time periods,
you know, across the chain, maybe,
you know, today in certain pockets.
Sometimes they're just
completely wrong, right?
You have a moment where something
happens, uh, markets react,
and they suddenly bounce back.
And, you know, I think the better, um,
uh, you know, uh, you know, takeaway
there is that, you know, it was emotional,
irrational and completely wrong.
But over, you know, longer, you know,
periods of time, markets have proven
to be one of the greatest information
systems humans have ever really built.
So I sat back and really just asked,
you know, another question: What,
don't markets then price today?
Like, what don't they do?
And all of a sudden we're sitting
there and, you know, you're, you're
talking about it, and you're talking
about stocks, bonds, currencies,
commodities, volatility, real
estate, , and then what's left?
And, you know, what is one thing I think
every business, every family, every
government, , deal with every single
day one way or another, whether it's,
minute things versus, you know, something
much more meaningful and important.
I think the answer there is
really just around uncertainty.
So look, w- we've maybe have seen
this before throughout history.
History teaches us something interesting.
You know, we almost underestimate,
certain technologies that
do come, meaningful ones.
We know Amazon looked like an online
bookstore when it first started.
We know Netflix, you know,
looked like a DVD rental company.
We know, um, we know, uh, NVIDIA, clearly
looked like a gaming company until there
was a breakthrough on the other side.
You know, none of these descriptions I
think were wrong at that point in time.
They were simply incomplete.
Books clearly weren't the
Amazon's, destination or, you
know, um, um, desires, let's say.
They were the entry point.
Same thing with, with Netflix, who
obviously migrated all the way to, uh,
creating their own content, are now in li-
live and, and comedy and things like that.
So naturally, you start to ask
yourself, are, are prediction
markets, some sort of evolution of the
past, and is it much more material?
And that's why we're talking about it.
You know, are sports, politics, elections
simply the easiest place to begin?
Because if they are, then I think we're
talking and thinking about a lot of this
stuff in a much different way than I
think is really the opportunity here.
Um, and, if you think insurance isn't
about insurance, then I think you start
to um, sit back and ask other questions.
So, you know, what business Already
exists to price uncertainty.
Are there any out there?
And, you know, there are probably
dozens of answers, but I think one,
one industry, you know, immediately
came to mind, and we're talking
about it, which is insurance.
And I think that's where today's episode
really begins, because the more I thought
about it, the more I realized, you
know, insurance companies aren't really
selling, you know, really insurance.
They're selling confidence.
Confidence that if something, you
know, unexpected happened, you
know, someone else will absorb
that financial consequence.
And how do you create that confidence?
And it's really about
estimating probabilities.
So if we think about everything an
insurance company does, studies, things
like, um, driving history that we see
more of, you know, weather patterns,
which is becoming more and more, obviously
popular as more data is out there.
You know, flood maps, wildfires,
crime statistics, uh, you know,
medical history on the personal side,
construction quality, uh, you know,
claims historically against them.
Thousands of, of variables really
and all trying to answer that one
question: what's the probability
of something happening here?
That's really the business.
So insurance isn't, you know,
really around just writing policies.
You know, it's about pricing uncertainty,
and that's when, something, uh, you know,
really around the concept of prediction
markets started to click a little bit,
which is prediction markets are really
trying to answer that exact same question.
They're simply using a completely
different architecture to do so,
gamifying it at the front end here.
But, as you think forward,
what could happen, right?
And I think that is where
the grand prize could be.
And it, it changed my thinking.
You know, at first I kept thinking
about prediction markets again versus
insurance companies, but the more
I thought, you know, about it, you
know, the less I, I like that, you
know, direct comparison because,
you know, maybe prediction markets
don't compete with insurance, right?
So, like, that's what it kinda sounded
like I'm, I'm referencing here, but
maybe they simply expand that market.
We've seen this time and time again.
Innovations sometimes are seen as
replacements, but really they're s-
you know, they end up being expanders
of TAM, total addressable markets.
And so let me explain a little bit.
You know, insurance is actually
a very simple business.
You know, one person doesn't
want a particular risk.
Someone else is willing to accept
that risk for a price, obviously.
Everything else, you know, the
underwriting, the regulations, the,
the actuarial ter- you know, models
for age and, and death, um, you know,
the claims process, let's say, um,
those are all built on top of, one
simple, uh, transaction, let's say.
One party transfers risk, another
party accepts it, and that's
kinda like the history of it.
You know, that second party has almost
always been an insurance company or
a reinsurer or another institutional,
you know, balance sheet But generally
speaking, you know, a relatively small
number of organizations have absorbed,
uh, an enormous amount of society's risk.
You know, that system has worked
remarkably well but it also has its
own limits, you know, because the
amount of risk society can insure
is ultimately limited by the amount
of capital willing to absorb it.
Um, and that's where I think prediction
markets become incredibly interesting
because I don't think the big question
is, you know, can they predict the future?
I think the, the much bigger question is:
can they increase the amount of capital
willing to take the other side of risk?
Because if they can, then maybe the
story isn't about, again, about betting.
Maybe it's about expanding society's
ability to transfer risk from one
party to the other, and that's where
we're going next here, which is,
who will take the other side of it?
You know, why prediction markets could,
you know, maybe transform insurance,
extend that and not just betting.
And, you know, the market for risk, if you
really think about the market for risk.
So let's, um, you know,
maybe pull on that thread,
you know, because I think this is
where the conversation, gets a little
bit more interesting and changes.
You know, in phase one, you know, the
conversation was, you know, we ended with
a simple idea, insurance is not a policy.
You know, maybe insurance is one
party transferring risk and another
party accepting that risk for a price.
Historically, the party accepting that
risk has been, again, insurance companies,
just as a reminder, or again, reinsurers
or large institutional balance sheets.
But again, if that changes and
prediction markets, AI, tokenization,
dig-digital identity, modern payment
rails make it easier for participants
to take the other side of that, that's
where, again, things get interesting.
So, you know, let's use an example here.
Uh, you know, I, I, I live in Florida,
um, you know, because, you know, every
year I'm reminded, uh, of, of hurricanes.
You know, insurance companies
continue to leave the state.
Premiums continue to rise.
Coverage gets harder to find.
You know, some homes, um, you
know, become incredibly expensive
to insure, which means ex--
in-incredibly expensive to live in.
So it's a, it's a friction point.
And again, as a firm, we're always looking
for, you know, where the world is headed,
where are there natural friction points,
and who's creating, value, uh, along that
chain to remove some of that friction.
And so, you know- Uh, back to the Florida
example, you know, homes incredibly
impossible to insure in some instances.
Why does that happen?
I guess, you know, it's, it's not because
people suddenly stop wanting insurance.
Demand there didn't disappear.
You know, supply of capital, supply
of the desire to absorb that risk, um,
definitely went away and disappeared.
Uh, again, there simply just wasn't
enough balance sheets willing to
absorb, hurricane risk at a price
homeowners, you know, can really
afford for them to manage that risk.
So insurance doesn't become expensive
because, again, people stop needing it.
It becomes expensive when the supply
of capital willing to absorb the
risk, uh, becomes constrained.
And again, you can have two sides
of that, that equation going, uh,
in opposite directions, and that's
kinda what you're seeing locally.
Taking that Florida example, uh, I
have another example that I thought of.
Publicly traded company, but, you
know, again, have no interest there.
But the Generac, the company Generac,
um, they, generally speaking,
you know, benefits from hurricane
season when they're very active.
Um, you know, more storms often mean
more demand for backup generators.
Uh, a quieter earning season-- I
mean, a hurricane season would be a,
a quiet earning season, but that may
mean just like weaker demand for them.
And today that's, business risk.
You know, management has to forecast,
uh, not only, you know, the inventory
they need to have on hand, but they have
to forecast, you know, what the weather
patterns may or may not look like, you
know, across the world and country.
Um, you know, and management's
forecast at best, you know,
what they, what they see, right?
So using probability models to do that.
So investors accept volatility, but
what if, uh, you know, Generac, you
know, obviously can't simply live
with that level of uncertainty?
What if they could directly
hedge part of that exposure?
So imagine someone like Generac
participating in a market
tied to hurricane activity.
Now imagine a homeowner
on the other side of that.
The homeowner wants financial
protection if storms happen.
Generac may want protection if they don't.
Uh, one side is worried
about hurricane damage.
The other side may be worried about, you
know, weaker gener- generator demand.
Um, again, neither participant
is really gambling there.
They're managing their risk.
They're transferring uncertainty
from one side to the other.
That's a completely different way
to think about prediction markets.
They're not just places where
people speculate on outcomes.
They're actually potentially
mechanisms for matching participants
with opposite risk ex-exposures.
Uh, and again, one party
wants to reduce risk.
Another is willing to accept it.
The market creates the connection.
That is very different from how I think
most people are talking about or even
thinking about prediction markets today.
Um And, you know, that's
obviously super, super important.
Now if we go in the opposite direction,
instead of thinking about billion-dollar
catastrophes, you know, let's think
about even, you know, tinier risks.
You know, historically, many risks simply
were not, economical to, to insure.
Um, you know, not because, again,
they didn't care, uh, it's just
the, the economics didn't work.
You would have to underwrite, you know,
that costs money, administrative burden
costs money, claims fraud detection,
settlement costs, you know, all that
stuff along the chain, um, costs money.
And, you know, maybe the premium
is only, you know, a couple, twenty
bucks, a hundred bucks, fi- five
hundred bucks, whatever it is.
But, but the operational cost could
be, higher than the, the premium
itself, you know, if you go down
the list of things you have to do.
So, those markets that are small
in nature, really never develop.
They don't grow as, as big unless,
again, technology is infused in this.
So you take artificial intelligence,
AI, you know, imagine, automated
underwriting, automated claims, you
know, fraud detection settlements,
suddenly, you know, millions of tiny
insurance markets become, possible.
We're seeing that with some publicly
traded companies I don't wanna name,
but, in general, uh, you know, that
is the path forward it feels like.
Uh, you know, a single shipment
from one place to another, a
weekend of rainfall for a farmer.
Again, we're not-- now we're
not talking hurricane season,
we're talking about a weekend.
Uh, you know, a construction
project delayed by weather, which
costs money either to the builder
or to the home buyer, developer.
Um, you know, a single, contract
between one party to another, but
maybe there's some sort of event
that could derail that situation.
Uh, you know, a one-day event, uh,
a local, tournament of sorts in, in
sports where you're the operator of
that venue and again, you're hedging
your weekend, uh, you know, with,
uh, rain and, and, and whatever.
Um, again, so theseâ¦
to insure some of these things would've
cost, you know, there's really no market
for that if you just think about some
of those examples I just mentioned.
And again, another like example of
that, you can find these micro risks
that have multiple parties that would
be involved if something was out there.
So As, putting some of this
together, which is, thinking more,
I realized I was, again, still
describing insurance the wrong way.
Most people think about
insurance as a product.
You buy a policy, a
premium, again, a claim.
Underneath the surface, I think,
again, insurance is really about
one thing, and it's capacity.
Um, and I'll explain that.
I had, you know, some notes,
but, you know, imagine there
were only two insurance companies
left in the entire country.
Would insurance become more expensive?
Um, you know, probably certainâ¦
there'd be a lot of, I, I, I'd feel
pretty, uh, high about that probability.
Uh, would some people become uninsurable?
You know, probably as well, and
again, none of this would be because,
demand changed or wanting, you
know, people wanting protection.
Uh, it was just capacity, financial
capacity, balance sheet capacity.
There wasn't just enough of it.
So the opposite, imagine instead of
a few balance sheets, you suddenly
had thousands and thousands of large,
small, medium, micro balance sheets
that could again hedge back and forth,
uh, and that changes the economics.
Uh, again, every time we see capacity
expand, we see new markets emerge,
more businesses get financed, again,
venture capital startups and, and such.
Uh, and I think ultimately that's some of
these questions worth, worth asking here.
Now, one important point, you
know, this doesn't all, you
know, happen in isolation.
Um, I think one mistake we, we make when
we think about technology is assuming,
you know, this one breakthrough of
sorts, you know, changes everything,
like we're living in that moment today.
But historically, you know, it
doesn't just really work like that.
Technologies merge.
They reinforce one another.
I think that again is happening today.
You think of AI, you think of
tokenization, digital identity,
modern payment infrastructure,
cloud, APIs, distributed computing.
You know, you have now these agentic
tools, uh, really around software.
Individually, you know, they're all
interesting clearly, but together
they create possibilities that
I don't think existed before.
I like to use the Uber example.
It took a combination of,
devices, all the components
inside of, you know, a smartphone.
It took, you know, three,
four, 5G technology.
It took payment rails to allow, It took
trust and comfort from society to allow,
you know, someone like Uber to exist.
And I think ultimately what you're seeing
here is, is some of the, the groundwork
for, um, what I'm talking about, which
is not prediction markets, but more
so insurance-like, features where risk
and very different parties that maybe
didn't think about micro risks or even
macro larger risks in certain ways
now have an opportunity, and I think
that's what's happening here, um, in
uh, the, prediction markets of sorts.
So look, if I had to, to, to summarize
the framework, really of this episode,
it would be prediction markets may not
be valued because they let people bet.
They may be valuable because they expand
society's capacity to transfer risk.
Again, that's a very different thesis.
Uh, it means the opportunity is not just
sports, not just, uh, you know, politics
and elections and speculation, which
again, is what most people talk about.
I think the opportunity is the creation
of entirely new risk markets, markets
for things that were previously too
expensive to insure, too specialized to
really underwrite, um, and, you know,
by size, like obviously too small to,
to justify, you know, going after it.
So that is, um, where
I'm thinking about it.
The-- I think that's, directionally
right, which is ultimately
as a firm where you wanna be.
Um, I think the story is much
bigger for someone like a Kalshi.
People are starting to embed
it in their platforms as well.
So Kalshi, I think is, is creating
popular aspects of this, but I think
ultimately this is going to move
into other different categories.
So again, Kalshi may be the catalyst, but
I think pr- prediction markets may be,
um, a much bigger product opportunity.
And if you go back and study, you know,
financial markets, many of them were not
particularly liquid in the beginning.
You know, over time, I think participants
showed up, market makers showed up.
Think about what a market
maker actually does.
They're not necessarily trying
to predict every outcome.
They're willing and able to make markets.
They're buying and selling.
Uh, they're quoting prices in real time.
They're absorbing some, some of
that imbalance and in exchange,
they earn a small spread across,
you know, many of these millions
and thousands of transactions.
And so without, you know, market makers
and, you know, sometimes they get a bad
rep, financial markets simply would not
function, um, the way they do today.
So I wonder if something similar
eventually develops here.
Another, , I think, um,
challenge is, is regulation.
Insurance is heavily regulated.
Clearly, um, prediction markets
are regulated, but much m- lighter.
And, it's should be regulated,
you know, much heavier.
Uh, you know, when your home floods
or your business burns down or a
hurricane destroys your neighborhood,
you need confidence that someone
can actually, pay that claim.
And so, again, it all comes back
to balance sheets on the other side
and ensuring that, , people can
pay, so, capital matters, you know,
oversight matters, you know, consumer
protection and contract designs in
these things, uh, all that will matter.
And I don't think prediction
markets eliminate all of that.
I think they, they create a much,
uh, you know, healthier arrangement.
But again, I think it'll, it'll
evolve on the regulation side.
The last, challenge is, is just trust.
Consumers do not really adopt, you know,
new financial, infrastructure overnight.
Specifically if you're talking about
business, trust can take some years.
Ultimately I think that's what's
being built here, uh, is, is trust.
, Speaker 5: You know, so here is, uh, I
think the whole thing in one thought.
You know, Kalshi looks
like a betting company.
Prediction markets definitely look
like speculation as they sit today.
But if you really pull on, you know, that
thought, I think, you know, the story is
not really about predicting the future.
It's more about who is willing
to take the other side of risk.
Insurance is, is about
capacity in some ways.
And if AI prediction markets,
tokenization, identity, modern
payment rails all come together
as we articulated in the evolution
of things like Uber, then all of a
sudden you can expand that capacity.
More people can build, more businesses
can, you know, hedge, um, and more
risk that used to be uninsurable
become much more manageable.
And, you know, that is not
a bet on a single company.
That's not what we're doing here.
It's really a bet or a thought
on a much larger evolution in
how society transfers risk.
And that is what we're watching
here around the desk and what
we're trying to bring to you.
But with that, you know,
thanks for joining me.
If you're-- you know, if you enjoyed the
conversation, we'd really appreciate if
you subscribe, share the episode with
somebody who enjoys thinking about,
you know, things like technology,
markets, and longer term trends.
And, you know, we'll bring more frameworks
and thoughts to you to explore next.
But until next time, keep
those questions coming, and
we'll keep the thoughts coming.
See you next time.