4% Inflation. Stretched Valuations. Why Is the Market Still Risk-On? | Tian Yang

Summary

---

Excess Returns Podcast — Episode Summary

Episode Title: 4% Inflation. Stretched Valuations. Why Is the Market Still Risk-On? | Tian Yang
Date: August 6, 2026
Host: Jack Forehand (Excess Returns)
Guest: Tian Yang (Head of Research, Variant Perception; Portfolio Manager, VPX ETF)

---

Overview

In this wide-ranging conversation, Jack Forehand welcomes Tian Yang to discuss why the market continues to run “risk-on” despite persistent inflation and high stock valuations. Tian, renowned for blending quantitative and qualitative frameworks at Variant Perception, explores the changing nature of macroeconomic signals, the influence of AI-driven capital expenditure (CapEx), and the real meaning behind headline inflation data. The discussion covers market tops, labor bifurcation, macro modeling, and building adaptive investment strategies in a rapidly evolving world.

---

Key Discussion Points & Insights

1. Creative Data Use and Filtering Out Noise

• Tian Yang highlights the importance of thinking from first principles when designing models:

  "Data is easier to access and more available than ever before. The key now is how creatively you use these inputs and most importantly, what you choose to leave out..." (01:14)

• He emphasizes causal relationships and sequence—rather than relying on black-box models or overwhelming data volume—as the core of Variant Perception's approach.

---

2. The Role and Evolution of Leading Indicators

• Defining Leading Indicators:
  Leading indicators are signals that precede large economic moves, like building permits foreshadowing residential construction, or order books indicating future manufacturing activity.

  “We tend to anchor our analysis on these lead-lag relationships…” (05:19)

• Adapting to a Changing Post-Pandemic World:
  Traditional signals (like the yield curve or consumer sentiment) have been less predictive due to fiscal dominance and economic bifurcation post-COVID.

  “People are frustrated with those traditional lead indicators... a lot of the models are static... the inputs stay the same and the coefficients stay the same.” (06:25)

• Dynamic, Mosaic Approach:
  Model inputs and importance are adaptive, accommodating shifts in what matters for the market at any moment.

  “It's not a fixed answer... we clearly have to acknowledge the economy is shifting.” (08:08)

---

3. Building & Interpreting the Macro Risk Indicator

• Construction:
  Variant Perception’s macro risk indicator synthesizes growth, inflation, policy, and liquidity using decision-tree models.

  “How can we boil it down to be something as simple as possible, but no simpler?” (09:55)

• Current Signal:
  The indicator has remained “risk on,” owing to resilient growth, mostly neutral policies, manageable inflation, and strong liquidity.

  “Right now the overall risk on message is because it thinks policy is pretty neutral ... inflation is a problem, but not overly so ... growth is fine ... liquidity ... is good.” (12:24)

• Amber Warnings, Not Red:
  Recent moves are interpreted not as headwinds but “tailwinds that are getting weaker,” with markets very efficiently pricing in first-order impacts, while second-order (like policy reaction) remain subdued.

  “Markets are very efficient at pricing first order impacts... once everyone prices in... do those first order impacts have some kind of second order impact, which is usually a shift in policy...” (13:30)

---

4. The AI Capex Boom: Impact & Downstream Effects

• AI CapEx as Key Macro Driver:
  Corporates and households are both “dis-saving,” fueling a positive feedback loop in the economy. The real risk arises if this spending slows.

  “Corporates are dis-saving at a historically epic rate... That’s what's been helping keep the economy very resilient.” (15:09)

• ROI Still Unclear:
  The return on AI CapEx is hard to measure in real time; the effects diffuse gradually, much like the digital productivity gains post-dotcom.

  “It’s very hard to measure ROI in real time.” (16:40) “[We’re] somewhat following the kind of mid-2000s playbook” (16:52)

• Policy and National Competition as Distinguishing Features:
  The US-China AI “race” means public sector involvement could prolong and amplify the cycle compared to past private-sector bubbles.

  “This is a sovereign existential race between the US and China where the state is going to be behind it.” (19:39)

---

5. Market Tops: Signals & What’s (Not) Present Now

• Behavioral & Macro Checklist:
  True market tops show behavioral excess plus underlying macro weakening (slower growth, tighter liquidity, increased policy risk).

  “There has to be a mechanism in which the crazy stupidity ends. And the mechanism has typically been monetary policy tightening...” (24:47)

• Today’s Ambiguity:
  Behavioral excess and high valuations are present, but economic and liquidity triggers are still missing; cash settlement events from big tech IPOs are just starting to test the environment.

  “Amber Warning ... Behavioral excesses obviously is there, valuations is obviously there ... the economy ... is kind of still okay ... liquidity hasn't ... tightened too much.” (29:05)

• Potential Market Supply Issue:
  The supply impact of large IPOs depends on whether cash is reinvested or leaves public markets.

  “If ... these IPOs ... money ends up being recycled ... then ... the market hasn't had to absorb it and liquidate cash. ... But ... people ... are probably going to keep it in private markets ... which means there's a net supply the public markets have to absorb...” (31:20)

---

6. Sector Dynamics, Capital Cycles, and Crowding

• Semiconductors Example:
  Despite the AI boom narrative, semiconductor companies themselves haven't yet over-invested, keeping their capital cycle healthy, while customers (hyperscalers) have ramped up spending.

  “Semis have ... remained relatively capital scarce ... the areas of overinvestment has been their customers ... who have ... gone out massively increase their capex ... but have not really generated the same returns ...” (32:34)

• Crowding Metrics:
  Combined with capital cycle indicators, crowding signals help evaluate whether a sector is over/under-owned and thus attractive or overbought.

• Cyclicity of Semis Debate:
  The “semis are no longer cyclical” thesis is moving mainstream—a sign the trade may be maturing.

  “This narrative that semis are no longer cyclical is definitely somewhere in the mainstream part ... there's way more ... talk about it's not as cyclical.” (35:46)

---

7. Inflation: Real Risks, Composition, and Implications

• Headline vs. Core Inflation:
  Headline (driven by volatile oil and externalities) exceeds target, but core inflation is muted due to soft housing, labor, and small business data.

  “Headline inflation is mechanically a reflection of ... energy prices... Underlying core inflation hasn’t been super strong ... on all those, the picture is a bit more subdued.” (36:10)

• Consumer Bifurcation:
  Tian describes a market split between higher and lower-income consumers, with growth driven by the affluent but inflation pressures largely dependent on the spending capacity of the less well-off.

  “...we've had this ... K shaped ... US economy ... low income wage growth is poor ... there's no, they don't really, they can't really absorb that much price increase from here.” (38:00)

• Oil’s Effect:
  Policymakers are likely to look through oil-driven inflation unless sustained, as most underlying factors are subdued and political (not just economic) constraints dominate.

  “We live in a world where politics trumps economics ..." (39:26 - 40:43)

---

8. Fed Outlook in the Walsh Era & Political Constraints

• No Hikes Expected Despite Data:
  Tian Yang argues that new Fed leadership under Walsh will prioritize systemic reform and political capital preservation over near-term rate changes, especially given the new President’s stated stance.

  "One of the implicit assumptions ... is that the energy shock will dissipate ... If this was a normal monetary policy cycle ... they should be hiking. But ... sovereignty and politics ... are way more important ..." (42:25-45:05)

• Main Agenda for Fed Reforms:
  Likely focus: moving away from excess reserves regime, restoring signaling, and carefully deregulating smaller banks—requiring nuanced coordination with Treasury.

• The Real Macro Influence:
  The lesson: who plays the game and how they play it, including understanding other investors’ constraints, can shape the investment landscape as much as economic fundamentals.

---

9. Bifurcation Everywhere: States, Labor, Technology

• SAHM Rule & Regional Disparities:
  Tian discusses how classic national indicators (like the SAHM rule) are now failing or must be adapted for a bifurcated US labor market—some states struggle while others benefit from AI or energy.

  "This idea of bifurcated K shaped is everywhere ..." (48:59-50:19)

• AI's Impact on Labor:
  AI causes more wage compression for “mediocre” roles but amplifies value for top performers. The adoption process is slow and frictional, but inequality in labor outcomes is likely to widen.

  "It reduces the value add for mediocre work that used to probably be more rewarded." (52:17-54:21)

---

10. VPX ETF: Adaptive Portfolio Construction

• How It's Built:
  VPX is a long-only US large cap ETF designed for higher upside/downside capture than the S&P 500 by dynamically tilting sector and stock weights based on Variant Perception’s forward-looking models.

  “... combine our cap cycle quality crowding, LPPL growth, all our models, macro everything to try and forecast forward returns for all the stocks and sectors ...” (56:27)

• Conviction and Active Risk:
  Willingness to aggressively overweight or underweight sectors when models signal opportunity; not designed for “benchmark huggers.”

  “We take a lot of active risks if the model says it ... we're trying to take a lot of risk when you think it's appropriate to try and maximize the upside and downside capture in a portfolio...” (58:40-59:32)

---

Standout Quotes & Memorable Moments

• On Data and Noise:
  “The key now is how creatively you use these inputs and most importantly, what you choose to leave out.” – Tian Yang (01:14)

• On Market Tops:
  “It's not just that things look excessive and crazy and stupid. There has to be a mechanism in which the crazy stupidity ends. And the mechanism has typically been monetary policy tightening...” – Tian Yang (24:47)

• On Professional Investing:
  “Investing is as much about doing great fundamental analysis ... but equally it's as much about playing the game of investing ... understanding the players, the constraints, who are the buyers, who are the sellers ...” – Tian Yang (61:15)

• On Risk and Survival:
  “Investing is like playing tennis, except your goal is not to hit winners but to reduce your own force errors ... it's not what you don't know that destroys your portfolio. It's the thing you know for sure that isn't true that destroys your portfolio.” – Tian Yang (62:11)

---

Timestamps for Key Segments

• Introduction, Tian’s Framework & Data Use: 01:11 - 03:17
• Leading Indicators Post-Pandemic: 05:44 - 08:08
• Macro Risk Indicator: Construction & Current Reading: 09:43 - 12:24
• “Risk On” Explained & Market Headwinds: 13:20 - 14:54
• AI CapEx and Economic Impact: 15:09 - 18:58
• Tech Cycles & Government Involvement: 19:10 - 21:21
• Market Top Checklist & Current Status: 24:47 - 29:05
• Sector Crowding & Capital Cycles (Semis): 31:53 - 35:46
• Inflation: Core vs. Headline, Bifurcation: 36:10 - 40:43
• Outlook for Fed Policy under Walsh: 42:10 - 47:16
• Bifurcation & Labor Market Dynamics: 48:48 - 54:56
• AI’s Impact on Research & Productivity: 55:01 - 56:14
• Building VPX ETF & Adaptive Strategy: 56:27 - 59:32
• Closing: Contrarian Beliefs & Investing Lessons: 61:02 - 63:44

---

Final Lessons & Takeaways

• Think Causally & Adaptively: One-size-fits-all macro/investing models are obsolete in a rapidly shifting and bifurcated world.
• Behavior ≠ Mechanism: Even extreme behavior or valuations need a trigger (like policy tightening) to signal a true market top.
• Play the Game & Avoid Overconfidence: Navigating markets means understanding both fundamentals and psychological/structural constraints of other players.
• Survival over Bravado: Reducing unforced errors trumps maximizing winners—don’t let the thing you “know for sure” ruin your portfolio.
• Keep an Open Mind — Especially About AI: Both in portfolios and in the labor market, AI-induced change will be profound, but adoption, ROI, and disruption play out in unpredictable, sometimes drawn-out cycles.

---

For more about the ETF, research, or Variant Perception:

• ETF: VPX
• Research: variantperception.com
• Twitter: @variantperception

---

This summary skips host read ads, intro/outro, and sponsor-only sections for clarity and listener value.

Transcript

[Sponsor Announcer 1]

This episode is brought to you by Google Chrome. You think you know a browser, but Gemini and Chrome, that's new. It can help you with practically anything on the web, like restoring a vintage motorcycle from a 50 page restoration block. Or finally break down that long article you've had open for weeks. Gemini and Chrome is here for it, ready to make anything online make sense. There's no place like Chrome. Check responses set up required compatibility and availability various 18 plus

[Sponsor Announcer 2]

when you need to build up your team to handle the growing chaos at work, use Indeed Sponsored Jobs. It gives your job posts the boost it needs to be seen and helps reach people with the right skills, certifications and more. Spend less time searching and more time actually interviewing candidates who check all your boxes. Listeners of this show will get a $75 sponsored job credit@ Indeed.com podcast. That's Indeed.com podcast terms and conditions apply. Need a hiring hero? This is a job for Indeed Sponsored Jobs

[Jack Forehand]

welcome to Excess Returns. I'm Jack Forehand and I'm excited to be joined today by Tian Yang. Tian's the head of research at Varied Perception and also the portfolio manager of the VPX etf and we're going to talk a little background today. So, Tian, thank you for joining us.

[Tian Yang]

Thanks for having me. Looking forward to it.

[Jack Forehand]

You guys do some awesome work and you blend quantitative and qualitative frameworks, which is exactly what I like to do as well. So I'm really interested to dig into. We're going to dig into the economy in general, but we're also going to dig into the framework behind it and how you're getting to the conclusions you guys are getting to. But I want to start with a quote. You have a really great quote here that I think summarizes what you guys do, but I think is really, really relevant for what's going on in the market today. And the quote is data is easier to access and more available than ever before. The key now is how creatively you use these inputs and most importantly, what you choose to leave out. And that last part was really important, I think to me, in the world of noise is what you choose to leave out. So can you talk about what you mean by that? Steve?

[Tian Yang]

Yeah, so I think what we're implicitly trying to say here is that you have to think from first principles. What are like the causal reasons that data is statistically meaningful. So, you know, we all want to build models, especially in this age of AI. You know, pretty soon we'll have like, you know, superpowered AI to help us build Models. Right. So I think a lot of times though, even when you're doing that and processing data, you, you know, it's easy to, you know, throw things to a black box, find something that looks back, test really well and want to use it. I think from where we're coming from, we think a lot about is there something causal? Is there like a real world reason that this thing has a relationship and it persists? So like a very simple example we give is essentially the intuition behind the idea of evenly delicate is in the first place. Right. That there's a certain sequence in which things happen. In real world economies, you have to get a building permit before you can build a house. So if you keep out on building permits, that'll give you a sense of when people want to build a house. And there's lots of these examples. So I think that's probably more what we're getting at to think a lot about causal relationships, which actually just means sequencing what moves first to then cause something else to move. Is that something that is our first principles likely to persist through time? And let's build models around that, find data to proxy for those and then from there obviously build up the analytical framework.

[Jack Forehand]

You reference leading indicators in your answer. And I know that's a big part of what you guys do. And you know, many people tend to focus on sort of what the data is right now and maybe not necessarily what it's going to be in the future. So can you just explain to me, like, what do you think, what do you define a leading indicator as? Like what's important in a leading indicator for you?

[Tian Yang]

Yeah, so I think as investors will look a lot like GDP or inflation or Fed policy. Walsh has made a policy announcement. These are all what we would say is like real time coincidence. Things that happen, they may or may not move the market. But often though, by the time these data points move, there's a sequence of things that have happened ahead of time that you can actually track and look at. And in a way, when we say lead indicators, it's almost like instead of trying to get a crystal ball and forecast something, you're just standing back and observing if the data is shifting. So we take that building permit example I gave through to its conclusion. It's like, okay, let's say we have the building permits start to surge, right. This month, next month, and you get a few months in a row, building permits going up, then we know there's going to be a lot of construction activity. Right. Especially if it's residential building permit, then you know, they'll build a house in the next 6, 9, 12 months. After that, people are going to move in. Suddenly when they move in, what they're going to do, they're going to take home mortgages, right? They're going to try and borrow, they're going to go and make goods, they're going to buy a new fridge. And there's all this activity that comes afterwards as a sequence from observing this first turning point. And the idea is to go through different parts of the economy and look for these potential shifts. So it tends to be more intuitive. In traditional cyclical industries. So things like manufacturing, you track order books, that's very intuitive. Right. If you track order books relative to inventory levels, again that's very intuitive. If industrial businesses suddenly start reporting a lot more new orders and they also tell you that inventories are low, that's a pretty good sign that there's a lot of future activity to come that isn't necessary in the current GDP data but can be in the future. And so I think it's kind of taking that idea through on the growth side, on the inflation side as well. And even on things like policy, where you want to get ahead of it. Think about, okay, what are central bank policy mandates? Where is growth and inflation relative to their mandate, what's the curve pricing in and based on that, what's likely to be their policy response function? It's just a lot of these lead lag relationships that we tend to anchor our analysis of models.

[Jack Forehand]

Have you seen any shift in how leading indicators work post pandemic? When people talk about these, they're talking about different leading indicators than what you're using. But a lot of the traditional leading indicators you'll see all over cnbc, many people argue have not worked. They've been predicting recession for a lot since 2020. Have you seen a shift in how leading indicators work post pandemic? Did something change there?

[Tian Yang]

Yeah, definitely. So I guess you're talking about things like the conference board leis. Right. And famously some of them have like stock market as, as a leading efficiency.

[Jack Forehand]

Yeah, like, and anything that, anything like at a high level, you see on CBC that like brings together these leading leading indicators that predict recession, even things like yield curve inversion or things like that. A lot of that stuff seems to have not worked as well post 2020.

[Tian Yang]

Yeah. So I think the way we do it is we have kind of a multi stage process. So we actually have what we would call like a causal discovery process. So we run a bunch of algorithms to try to Understand at any given point in time, which is a lead indicator that's been predictive at recent turning points, and we will just essentially overweight those so that your model can adapt. I think the reason people are frustrated with those traditional lead indicators is a lot of the models are static. The inputs stay the same and the coefficients stay the same. And so clearly we're in a world where there's more information, more need to adapt. So the idea is that you want your inputs to change and your coefficients you assign to those inputs to change as well through time. But I don't think the necessary first principle is wrong. Real curves do have a sound fundamental reason. It's just more, you have to try to be sensitive to when they might work less well. So yield curves in a traditional credit cycle, when inflation is not a problem, when fiscal policy is not a problem, and central banks hike or cut interest rates in response to worries about growth and inflation, they work really well. If we live in a fiscal world, a world where sovereignty dominates, there's US, China, competition, where governments need to get involved in FX markets to help the Japanese backstop the yen.

[Jack Forehand]

Right.

[Tian Yang]

There's a number of these other things that come in that if you're looking under the hood, you'll start to observe these things become less effective. It doesn't mean they won't go back to working at some point in the future. It just means right now the primary kind of mechanisms in which the macroeconomy operates have changed.

[Jack Forehand]

Is it almost like you have to look at these leading indicators as a mosaic and at certain points in time, certain leading indicators are more important than other leading indicators?

[Tian Yang]

Yeah, yeah, absolutely. That's effectively what our process is. We basically have like about a thousand manually curated inputs globally that we think are both theoretically leading. And also when you look at the data, the data revision is quite low in real time. They release monthly. Right. There tends to be high quality data. So that's the kind of manual part where we do the curation. After that, then you throw it to the model and the model tells you, okay, at this point in time, this thing matters. So yeah, I don't think it's like a fixed answer, but the idea is that we clearly have to acknowledge the economy shifting, manufacturing versus services, like online data. Another classic example is consumer sentiment, consumer expectations. That's become pretty much useless over time as the things have shifted. But there's alternative measures for the consumer that gives you good read. Right. And I would argue one of the reasons people might think it stopped working is because We've lived in this post pandemic price level as well as inflation kind of environment where it's not just the rate of inflation, but the fact price levels reset higher, wages haven't kept up and so consumers are telling you they're struggling with real incomes. But historically, because we've gone through a long period where there hasn't been a lot of inflation, historically that series was very correlated with job market prospects. So again when it was correlated with job markets, it makes sense why that gave you a very good lead on growth. Right now that it's all about inflation, there tends to be like more second order impacts on why, you know, why that series moving is no longer lead indicator directly for growth.

[Jack Forehand]

So you guys, you guys boil all this down to a macro risk indicator, which, which I think is very, very cool. And we're going to put up the chart here of the macro risk indicator. So can you explain first, before we talk about what it's telling us now, can you explain how you construct this?

[Tian Yang]

Yeah, so I think so. We came up with this idea because we wanted to abstract macro down for people that don't want to think about macro all the time. Right. Like macro sort of it doesn't matter until it does like once every five years, suddenly it matters kind of thing. And so that was kind of the original idea. How can we boil it down to be something as simple as possible, but no simpler? So we essentially came up with these four dimensions of macro, which is growth, inflation, the traditional kind of bridgewater style framework. And then we added in policy and liquidity as the four. And essentially how we come up with these components is they're actually decision trees that our model essentially goes down. So it'll check saying, hey, what's the US growth indicator doing right now? Okay, given this is going up, what's China's growth indicator doing now? And if given China's growth indicator is going up, then what's Fed policy doing now? Oh, the Fed is neutral and all the growth is doing good, then that's likely to give me a point for risk on. And it'll go for a lot of these decision trees to come up with the final score. So it's attempt to kind of formalize and make repeatable some of these relationships in macro. That is intuitive, but you have to kind of do them all at once to get understanding of that. So that's kind of how it's built, but essentially it gives you a number between 0 and 100. It moves smoothly over time. And the idea is to use this to dial up or dial down your risk. Essentially your risk exposure.

[Sponsor Announcer 1]

This episode is brought to you by Google Chrome. You think you know a browser, but Gemini and Chrome, that's new. It can help you with practically anything on the web, like restoring a vintage motorcycle from a 50 page restoration block. Or finally break down that long article you've had open for weeks. Gemini and Chrome is here for it, ready to make anything online make sense. There's no place like Chrome. Check responses, set up required compatibility and availability various 18

[Sponsor Announcer 2]

when you need to build up your team to handle the growing chaos at work, use Indeed Sponsored Jobs. It gives your job post the boost it needs to be seen and helps reach people with the right skills, certifications and more. Spend less time searching and more time actually interviewing candidates who check all your boxes. Listeners of this show will get a $75 sponsored job credit@ Indeed.com podcast that that's Indeed.com podcast. Terms and conditions apply. Need a hiring hero? This is a job for Indeed Sponsored Jobs.

[Jack Forehand]

So this has been correctly risk on for a long time now. Can you, can you just talk about what this is telling us right now?

[Tian Yang]

Yeah, it's still basically in the risk on regime. It's kind of been saying essentially since even since Iran war started that growth is basically resilient. It thinks the policy risk a bit overstated, so it kind of sees the policy landscape as pretty bifurcated. You'll have very obviously hawkish global economies in central banks like Japan or Korea. But he thinks a lot of Europe shouldn't be as hawkish, for example. And it thinks that it's touch and go for the Fed. Right. The Fed shouldn't really need to be hiking. So it's just combining a lot of these factors. And right now the overall risk on message is because it thinks policy is pretty neutral to slightly too risk on. It thinks inflation is a problem, but not overly so. It thinks growth is fine and it thinks the broader liquidity environment is good. And so when you piece it together, that's why it's risk on. And in a way, if you just look at the fact that equity markets are broadening out during the semi drawdown, I think that gives you a good clue on. Liquidity probably isn't as tight as everyone says it is.

[Jack Forehand]

One of the things a lot of people have been talking about is building headwinds, but you've kind of taken that the opposite way. You've talked about not necessarily building headwinds, but tailwinds that maybe are getting a little bit weaker. Can you Talk about that.

[Tian Yang]

Yeah, well, I think that reflects the fact our model has been going from super risk on in the last year to now being slightly less risk on as prices have adjusted higher. I think as a mental model, something I think about is that markets are very efficient at pricing first order impacts, right? So when things happen, markets can and people get a handle on it pretty quickly. And I think this is the underlying rationale behind very famous quips in the market. Like Bob Farrell has one of my favorite quotes about when all the experts agree, something else happens.

[Jack Forehand]

Right?

[Tian Yang]

And I think that's the attempt to get at this idea of everybody prices in the first order impact. But then what's interesting is once everyone prices in, it's not necessarily wrong. The key is once we all realize, do those first order impacts have some kind of second order impact, which is usually a shift in policy that happens. And so that's the way I think about using this. Right. This is giving us a sense of hey, first of all the impact is a risk on good. That's how people positioned. But then given this outlook, do you think policymakers want to lean against it and are they going to shift policy or not? And I would say right now it's kind of like, yeah, I don't see why policymakers want to massively lean against the fact the equity bull market economy is fine.

[Jack Forehand]

Right.

[Tian Yang]

They're not actively trying to do something against it. So then the model's probably valid. There's no second order impact resulting. Therefore, yeah, the outlook's broadly good. We're still staying fully invested in equity.

[Jack Forehand]

One of the things we've been talking about a lot of the podcast is this idea that AI Capex is dominating everything. It's definitely dominating the market and some argue it is or is not dominating the economy. But how important? Like if we look at the overall economy and everything that's going on, how important is this AI CapEx?

[Tian Yang]

Yeah, it's very important for sure. The way we think about it is we use the Calethi Levy framework quite heavily actually. So the basic concept is one person's spending is somebody else's income. So the way the economy functions is as long as I spend that creates income for someone else, then they're more likely to spend. And so the inverse of that is obviously tracking the savings rate. So as long as people are drawing down savings into saving, that's on net, creates a lot of growth and a lot of resilience. So we basically have an environment where corporates are dis saving at a historically epic rate. Are you? They're investing a lot more. And at the same time, US households also are maintaining a very low savings rate. So if you have an environment where both households and corporates are choosing to spend more, save less, then that's income for somebody else. Right. And it just keeps flowing around. And I think that's what's been helping keep the economy very resilient. The time to worry is precisely when these hyperscalers of people suddenly dial back their capex. Right. And suddenly showing that other margin, they want to save more. And if they want to save more, that's going to be less income for someone else. And suddenly that whole loop can potentially start to unwind. So, yeah, very important, but it's one piece alongside the household piece.

[Jack Forehand]

Yeah. The thing I've been thinking about a lot is sort of the downstream benefit of this, like the end user ROI of capex and how important that is. And it seems like maybe the market's not that concerned about that yet because we're a little bit down the road. But do you have any thoughts on that, like how important that is, that we start to see like ROI from this capex downstream?

[Tian Yang]

Yeah. So I would say there's a theoretical and a practical answer. And the practical answer is it's very hard to measure ROI in real time. Right. We're not going to be able to measure this in real time. So it's not like there's going to be a smoking gun you can point to which is like, oh yeah, this definitively proves AI is not creating any value because this company make no profit or vice versa. So I think it's more about thinking about the broader of sequencing on how these things play out. That's more of the way I would think about it. So as of right now, we actually think we're somewhat following the kind of mid-2000s playbook when you first had the initial dot com bubble that led to all that investment and they started to slowly diffuse out into the economy and start to boost labor productivity. We actually think we're seeing the first signs of that at the margin. Again, not a huge amount. But these things are obviously tough to measure in real time. I actually think there's a good chance it will start to diffuse out. The reason investors are concerned is because ultimately everybody agrees this is game changing. The problem is who actually makes a profit in the profit pool. And I think ultimately that's where the sequencing comes in, where right now the kind of bottleneck hardware leg of the trade is done. Right. Everybody realized these were the bottleneck they're going to benefit and all the money crowds into that and that's what, that's what's gone up I think from this next phase is potentially you're actually into more the Java's paradox leg of it where suddenly costs come down, people use it more, the AI adopters start to benefit, so the profit pool starts broadening out into those guys who benefit. The hyperscaler suddenly start taking advantage. Right. I would say that's one dynamic that's kicking in and that's going to carry on until we get another technological breakthrough. Right. Just like at the end of 25, you know, you could have made all the same arguments, man, this capex or like, you know, the bubble's over and then suddenly agent agentic AI is a thing and then that sets off a whole other leg with narratives. And so right now it's not clear what the next leg is. But maybe if we get to recursive self improvement or you know, world modeling or something else suddenly comes through, there's a big thing, suddenly everyone reassesses, that might be the next leg. But until you see that, I would say you're in that phase of the AI adopters starting to benefit until you get to the next big step, change in demand for compute.

[Jack Forehand]

Right? Yeah. It seems like it makes it so hard to predict because that next technological breakthrough, you don't see it coming, but then when it does, it could be a massive game changer. This technology is moving so fast it seems like it's hard to predict the whole thing.

[Tian Yang]

Yeah. Which is why I think it's important to be adaptive and open minded. So that's why I understand the major concern. There's a lot of very smart, good strategy is talking about is the railway bubble.

[Sponsor Announcer 1]

Right.

[Tian Yang]

Or the dot com like this. You know, there's a lot of that and I'm just saying. Okay, I would like to understand what that argument is, which is a classic capital cycle over investment argument. Right. But I want to also be open minded about, okay, what's the falsify in thing about it? And there's a couple of things I think that makes this cycle a bit different.

[Sponsor Announcer 1]

Right.

[Tian Yang]

The first of all, this is not just a private investment boom cycle funded by private sector credit. Right. This is a sovereign existential race between the US and China where the state is going to be behind it. So you got to almost think of it as like national balance sheets. Right. Just like Chinese companies, Chinese government people don't have a big, don't have a lot of trouble Thinking about that as one thing, it's all coordinated and then they plan money into AI, right? And tech. But I think you can get somewhere similar with the US max 7 US government. That could be a lot of coordination, a lot of, you know, private public initiatives because it's very important technology to win at. Right. And these are the things that can potentially elongate the cycle more than just purely looking at traditional private credit metrics. So I just think there's a lot of these forces at play that that means it's very hard to be declarative. Right. Like you know, Howard Marx is that very famous quote about it's always what you know for sure that just ain't true. That's what destroys your portfolio.

[Jack Forehand]

Right.

[Tian Yang]

It's not so much what you don't know. And a lot of the takes on AI, I feel like people veer from one to the other, right. It's like this is definitely a bubble. This could destroy all capital or suddenly it's the greatest thing ever and it's going to grow to the sky. I feel like you've seen that pretty aggressively this year as people shift their mindset to that point.

[Jack Forehand]

How do you think about the long term benefits of this? I mean you've got people in the tech community that are talking about know, deflationary growth in the world of abundance. This is going to create and levels of GDP growth you've never seen before. And then you've got other people who say like if you look at the history of these technologies, you know, GDP growth usually ends up about the same thing. Like in the long run it doesn't change that much. I mean, is that even a question worth thinking about or how, how would you think about that?

[Tian Yang]

Yeah, I mean as you say, I would generally put in the either too hard bucket or not really practically relevant for investors bucket.

[Sponsor Announcer 2]

Right.

[Tian Yang]

Like I, you know, we can debate about, hey, this is like electrification, it can boost productivity 1% of the next 50 years. I don't think those are very practical things for you as an investor today and what to do. However, if you think it through though, when we look at these historical cycles that you mentioned, a very common pattern is this idea of complementary assets. So ultimately the way humans and systems function does not change overnight. So it's very hard for our pre existing systems to fully utilize these new technologies. It takes a bit of time and so often during that transition it's that so called complementary asset that wins. And this is where your Amazon, Microsoft, Google's come in. Right. Because they have Your trust as your enterprise partner, your data is with them. You trust them to be safe, you trust them to take care of your privacy, your data security, all these things. And then suddenly, if they start, and suddenly AI is a great technology, you're still going to potentially use them to leverage AI for your own needs. Right. And so that's a good example of ultimately these kind of core businesses that didn't necessarily invent a new technology, but who has the main complementary asset that allows the technology to diffuse more widely and be adopted, they tend to win. So complementary asset is one that works really well, and the other one is always data. Historically, no matter what kind of boom bust cycles you've seen in tech, like the data part, we tend to just keep collecting more data, store more data, and data trends go up.

[Jack Forehand]

I want to pull up a chart. We talked about recession earlier and the fact that some of these things have been predicting recession, we haven't seen it. I want to pull up a chart from one of your presentations because you guys have so many amazing charts. And one of these is this idea that you talked about, this saving, and a lot of people are talking about this. Saving is kind of a risk. But you have this chart that says recessions are usually preceded by a rising savings rate, not a falling savings rate. So could you talk about that and why that's true?

[Tian Yang]

Yeah, this is where a micro phenomena does not necessarily work at a macro level, because obviously for people that save more at an individual level, it's generally considered the good, because you live within your means, you have resilience. But as I said before, when you choose to spend less, there's somebody else who has less income. So at a macro level, if we all try to save more at the same time, you know, the second order impacts all our incomes go down. Suddenly, if our incomes go down, we're gonna have to decide if we're gonna save more or borrow more. And if we choose to save more, that sets off this negative feedback loop. And even visually the way I present the chart, obviously, because since like Covid, I have to do log scales now because of the distortion, but you can kind of see this pattern that. Yeah. In the kind of lead up into when recessions have started, you've generally seen precautionary savings pick up. And that initial pickup in precautionary savings is what starts to draw down income for everybody else. And that's what potentially sets off the kind of economic slowdown. And obviously if it gets really bad, it goes into a recession, as you mentioned before.

[Jack Forehand]

We're going to have people probably calling the market top for a long time before we actually have a market top. But one of the things you guys have done that's really cool is you have a market tops checklist where you've looked at some previous market tops and you've looked for some things that are in common and you look through that checklist as you evaluate things like that. So could you talk about what's on that checklist?

[Tian Yang]

Oh yeah, yeah. I mean, I can send you updated one. It's a very long list. But the basic concept is there'll be behavioral signs you typically see at the top. And I think that's what everyone's obsessed with right now, circular financing. Obviously we see like the crazy leverage like retail getting high. That's definitely one aspect. You know, famous investors, right. Guru investors. Obviously we're recording literally the week after situational awareness made headlines. So that's definitely one piece at major market talks. But what you also tend to see is the economic piece, right? The growth slows down, the policy titans, liquidity tightens. And that's the macro piece that we wanted to flag that if you go back to things like 1972, the nifty 50 top, you go back to even 1929, these major historical tops dot com. It's not just that things look excessive and crazy and stupid. There has to be a mechanism in which the crazy stupidity ends. And the mechanism has typically been monetary policy tightening for about six to nine months where central banks will by however many means at the margin of liquidity starts to tighten, which starts to hurt people's ability to borrow. So margin debt and these every cycle is called something different.

[Jack Forehand]

Right?

[Tian Yang]

Leverage starts to go down. And then what you see is though, even though leverage is going down, the concept stocks still rally. And then that'll be a sign of rotation. People start selling down everything else in their portfolio to buy the concept stocks because they only ever go up. And even though liquidity is tightening, they're still doing that. You've actually tended to see that at the major generational tops. And I think that's what's different about today. Today the average stock is actually making higher, higher lows. Right. We would define that as tracking index, like value line arithmetic. Right. You can just see it, it's just going higher, higher, low. So yeah, those are the typical things you want to see. It's not just the signs of excess. You need to see signs the economy is going to slow, liquidity is coming out and then. And obviously valuations are high, but I think we can generally accept valuations are pretty elevated. So you typically need to see all of that. But again, it's the sequencing. Often if all these things are in place, it's more a measure of gravitational energy in a way. But to turn into kinetic energy, normally there's a mechanism and it's almost always something about policy that forces a kind of a cash settlement into the market. There's always some kind of major event like at the.comtop it might be the AOL Time Warner like merger, some big event, big that has to be settled and then that settlement sets up a bunch of activities. Obviously this year we had a mini run with SpaceX IPO. Right now the unlocks are coming through. So all the people that have SpaceX now need cash. They need to liquid it into the market, right? So that's the first test. And then when our profit comes to IPO in October, whenever, when OpenAI comes to IPO end of year, those are going to be real tests. That's cash settlement in an environment where you have to see people willing to actually settle with cash, right? It's very easy to be like do stop for stock deals or do payment in kind or like hey, I pledge you compute or I backs up your loan. The test is always you got to get to the point where cash settlement needs to kick in. And I would say that's the thing to be on the lookout.

[Sponsor Announcer 1]

This episode is brought to you by Google Chrome. You think you know a browser, but Gemini and Chrome, that's new. It can help you with practically anything on the web, like restoring a vintage motorcycle from a 50 page restoration block. Or finally break down that long article you've had open for weeks. Gemini and Chrome is here for it, ready to make anything online make sense. There's no place like Chrome. Check responses Setup required compatibility and availability various 18

[Sponsor Announcer 2]

when you need to build up your team to handle the growing chaos at work, use Indeed Sponsored Jobs. It gives your job post the boost it needs to be seen and helps reach people with the right skills, certifications and more. Spend less time searching and more time actually interviewing candidates who check all your boxes. Listeners of this show will get a $75 sponsored job credit@ Indeed.com podcast. That's Indeed.com podcast. Terms and conditions apply. Need a hiring hero? This is a job for Indeed Sponsored Jobs.

[Jack Forehand]

So are you not seeing too many of those signs though? Not the triggering event itself, but the actual signs of a market top? Are you not seeing a lot of those right now?

[Tian Yang]

Yeah, I would Say it's like Amber Warning, right? Like, as I say, the behavioral excesses obviously is there, the valuations is obviously there. But the economy, like we talked about, is kind of still okay, right? It hasn't quite slowed down, liquidity hasn't really tightened too much. And these cash settlement forcing mechanisms, we're only just seeing the beginning of it. Right, but you're still seeing companies announce mergers and deals all the time, right? Hey, I'm going to use my expensive stock to go buy your cheaper stock. And you're seeing a lot more of those things and they're going through, right? Like you need to see these deals start to fail and then that'll be a pretty big warning sign. But so far we haven't quite had that.

[Jack Forehand]

You mentioned SpaceX and Anthropic and OpenAI potentially coming as well. How do you think about that? Like a lot of people think about, this idea of a lot of supply coming on is sort of a major issue for the market. And we've had some guests who say, you know, no, it's not. The market can absorb it. It's not going to be that big of a deal. Like, how do you think that through?

[Tian Yang]

So, so that some first principles. The way I think about it is when say you hold SpaceX shares, right, and suddenly you've decided, okay, the unlocks happen, I've made my ten hundred thousand X, what have you brought in and you sell, what do you do with the cash? So if I take the cash and I'm like, oh, you know what, I'm going to just park in S and P index funds or I'm keeping NASDAQ funds, then obviously that's fine, right? Because that money is going back in to the market and it's broadly participating. But what if I'm just like, you know what, I don't like public markets, I'm going to take my money and go do something else with it. I'm going to reinvest it back in private. So I'm going to sit on it, right? Like if that, if a lot of SpaceX holders start to do that, then suddenly is a net supply. But to your point, if, like, if these, after these IPOs of money ends up being recycled into, you know, the wealth management or whenever the, you know, Goldman called all these guys and signed them up to their wealth management service and suddenly the money goes in and gets placed into Goldman's whatever strategy and goes back in the market, then obviously the market hasn't had to absorb it and liquidate cash. So I Think that's the kind of unknown part. My suspicion would be there is some net supply into the market because I don't think the people cashing out are just going to blindly put it all back into listed equities.

[Jack Forehand]

Right.

[Tian Yang]

If anything I'm under the inflation people have generally done well that, done well in private markets, are probably going to keep it in private markets which means there's a net supply the public markets have to absorb and that would obviously start to add a bit of, you know, a bit of weight on top of the index.

[Jack Forehand]

You guys had a really, another really interesting chart in one of your papers here. This chart here I'm going to throw up which is this crowding score, 10 year percentile against capital cycle score. And you're looking at a bunch of different industries here and where they stand. Can you just talk about what this system is and what it's doing? Yeah.

[Tian Yang]

So our single stock framework generally has a few pillars. The most important structural pillar is the idea of the capital cycle. That again goes all the way back to marathon asset management. Ed Chancellor wrote the brilliant original book on it. And what we've tried to do is take inspiration from that and quantify and build our own scores. So we essentially try to understand which of the sectors that are seeing over under investment relative to the operational ROIC generated. So on our model semis have actually remained relatively capital scarce despite all the hype in the area. Because what's been happening is that they have not themselves overinvested yet.

[Jack Forehand]

Right.

[Tian Yang]

Even at the margin. If you see the memory guys they're only just starting to bring up capex but they've made phenomenal returns. The areas of overinvestment has been their customers, it's your hyperscalers, it's people like that that have actually gone out massively increase their capex to buying the products from them. But at the same time these guys have not really generated the same returns on that capex so far. So that's essentially what the capital cycle score is doing. It's trying to take all the different global profit pools, trying to understand within that sector, that profit pool what's been net investment and what's been the return generated and ranks everything. And that gives you a long short profile. And then for crowding it's basically built on top of the standard crowding. So people have like 13f all the filings and things like that except we have a couple of higher frequency crowding metrics. So we have like a fast money proxy and things like that that we Blend together to give you an overall sense of how are institutional managers speculative money, how they position overall. So obviously the sweet spot is that you ideally want to buy companies that are good on crowd, that are good on capital cycle, but don't on crowding.

[Jack Forehand]

Right.

[Tian Yang]

And that's kind of over time like a repeatable way to screen for ideas.

[Jack Forehand]

What do you think about these arguments? You mentioned semis. We've had a lot of people talking about this idea that semis aren't cyclical anymore. You know, they've got such a tailwind from this AI thing that, you know, semis have traditionally been always cyclical and you know, when they look kind of like they look right now is not necessarily been a great time to own them historically. But you know, some people are arguing that that's, that this is really not a good time to own them. And some people are arguing that the world has changed. These aren't cyclical companies anymore. What do you think about that?

[Tian Yang]

So I guess one of the mental models I think about a lot investing is this idea of like a diffusion of information, like an S curve, right. Like all the, all the best investments is ideally, you don't have to be the super early adopter, but you kind of catch it when it goes mainstream the idea and you just ride it until it's super late stage that, you know, your cousin who's got nothing to do with finance calls you up asking about it. And obviously we've seen that constantly in bitcoin gold, semis, right. All these things, this S curve diffusion. So I guess I don't actually have that strong opinion on it. I'm just more like, I think this narrative that semis are no longer cyclical is definitely somewhere in the mainstream part. I would say there was way more skepticism beginning of the year and that that narrative's gotten a lot more whole. So you probably. It's somewhere from the middle to the late part, right. There's obviously certain people I, I look, I'm like, they're very steadfast. This is cyclical as well. Or at the ending tiers, if you just see one or two of them give up, that's probably a sign that it's very late stage. So that's more the way I would think about it because it's slightly academic debate. Right. Like it's, it's really hard to know because this is a, a generational investment cycle. And like I say with all the sovereignty angle over the top. Right. Like it's actually not straightforward. So my mindset is more like, yeah, if how widespread is the narrative? If enough people think about it, then it's probably priced in.

[Jack Forehand]

Yeah, it's like when the skeptics initially appear, that's still pretty bullish for the whole thing. But when it becomes like everybody just accepts that they're no longer cyclical anymore, then you've probably got a problem in your hands.

[Tian Yang]

It feels like we're getting closer to that than not. Right. Like there's way more of talk about it's not as cyclical.

[Jack Forehand]

How are you thinking about inflation right now? That's something we've been talking about a lot on the podcast and inflation's been above target for a really long time right now. I mean, the Fed obviously hasn't hiked yet. Some people think that the odds are above 5050 now for September, that they might. How do you think about inflation? Do you see it as a big problem right now?

[Tian Yang]

So the short answer, we don't think it's being a big problem. We obviously understand it's going to be above target. And our headline inflation leis have been surging. Right. So our headline inflation allies are saying, hey, inflation is going to be north of 4%, but I'm not worried about it. And I think that's the gap that there's been. And I would say I've got a lot of pushback with our clients and people I talk to on this. So one thing we noticed that there's a meaningful divergence between headline and call inflation. Headline inflation is mechanically reflection of the fact we have energy prices. Right. And that's basically driven it. So if Iran war de escalates or the impulse shifts that will go away. Underlying core inflation hasn't been super strong in our opinion because core inflation is ultimately driven by housing, by the labor market and just by normal activity, small business, lower income, consumers. And on all those, the picture is a bit more subdued. Right. US Housing obviously transaction activities slow down, mortgage rates are back up, there's things that keep a little housing. House prices aren't really going up that much. So you don't really have the inflation piece from housing on the labor market. We think the US unemployment rate is roughly around where the natural rate is. So even theoretically there's not a huge amount of wage pressure, especially with AI and things like that. And even in the data, nominal wage growth is generally trending down. So again, you don't have a huge amount of second order wage pressures built up. And then we look at things like small business surveys, things like NFIB surveys. There's not much intention to raise prices or raise incomes and raise wages right from these businesses. So broadly, I would say most of the underlying things are muted. A very simple rule of thumb I try and describe is the idea that if you think about the K shaped consumer, the lower half consumers matter for inflation. If lower half consumers are doing well and they have real income growth, that means companies are going to pass on price increases to them, so that's going to allow inflation to go higher. But upper half income consumers tend to matter for growth because obviously they're bigger piece and they drive the overall growth number. So I think we've had this pretty bifurcated dystopian kind of US economy for a long time where your K shaped consumers get you into the bad sweet spot but not a good for society kind of sweet spot because low income wage growth is poor. Right. And lower income finances are so poor. There's no, they don't really, they can't really absorb that much price increase from here. So you don't get a huge amount of inflation pressure. It's the same reason you see things like New York Fed surveys on delinquency rate. Right. And those things are very high. But at the same time you haven't had a recession because the upper half of the cage shaped consumers being fine. And I still think that's the underlying dynamic. So yes, we're mechanically above target, but it's a classic kind of supply shock that policymakers should look through. And the underlying housing, labor market, small business, low income consumers, they don't suggest there's that much upside inflation risks.

[Jack Forehand]

So do you think to some extent you have to ignore oil a little bit in a situation like this where, you know, one tweet can change oil or one, one development in the war can change oil so much and that seems to have a big effect on inflation overall. Do you have to sort of pay a little bit less attention to that and try to get at the underlying drivers of what's going on?

[Tian Yang]

Yeah, it's, it's obviously difficult because of just how volatile it's become. But in general you would need oil to average a high for a while. So just because it touches the level is not necessarily the same thing. For sure, in terms of base effects, it's going to have some pass through. But these are the classic things that if it's a supply shock, policymakers should be looking for it. Right. I think the problem is because Covid and the post Covid almost double digit inflation is so fresh in everyone's memory and it was such a big failure of central banks and central bank credibility that or the central banks are going to default, almost are more hawkish than they would have needed to be now because they're very worried about credibility. So I think that's probably more the impact it's had that they're going to have to talk super hawkish. But in terms of following through and hiking, I think the bar is, I mean, my view is the bar is pretty high. Right. Because it's not. Again, we live in a world where politics trumps economics. Right. There's just a lot of other factors that means policymakers, everyone's not free to act in the same way they did 20 years ago. So yeah, I mean, clearly, mechanically there's an impact, but I think we're off the mindset. It's probably not macro relevant for inflation. It's going to be looked through.

[Jack Forehand]

You mentioned the labor market and you have a great chart. There's US Consumer Personal Finances versus Jobs Hard to Get, which is the conference for Jobs Hard to Get indicator. Can you talk about what you're doing with this chart and why you think it's important?

[Tian Yang]

Yeah, I think we hit on a little bit earlier in the conversation. I'm trying to showcase visually the underlying state of the U.S. household and U.S. consumers. So what's very interesting if you look is historically they're very well correlated right through multiple, multiple cycles. Because in general your personal finances are related to your ability to get a job and for your incomes. And what's been very interesting is that after Covid this completely diverged where most of the survey responders are telling you it's easy to get a job and they're not too worried about getting a job, yet they continue to tell you personal finances are struggling. And ultimately that gap is explained by price levels. Right? The massive reset in price levels and the inability of real income levels to keep up with the cost of living. So I think it's just, you can, this is a really stark way to understand the cost of living pressures, especially for low income consumers right now. And obviously there's multiple ways to measure it, right? Like amount of people working multiple jobs and all these things. So it's just telling you, yes, people can find a job, but it's not the best job. It doesn't pay you enough to sustain the previous standard living people expect. And this gap is obviously unprecedented in the history of the data.

[Jack Forehand]

I want to shift to the Fed because it's something we've been talking about a lot. We obviously just had a change in the Fed chair. And one of the things I found in researching you is I believe you're out of consensus view here is that you do not believe the Fed is going to hike this year. So can you talk about the position the Fed's in and what you think they might do?

[Tian Yang]

Yeah. So first of all I will have to say one of the implicit assumptions clearly is that the energy shock will dissipate, the impulse will go down. So by the time you get to September and late in the year, there's some leeway. But I guess the fundamental thing I'm driving at is I acknowledge if this was a normal monetary policy cycle and the central banks are supposed to look at growth and inflation, try and meet their targets, they should be hiking. I get that. But going back to my earlier comment on sovereignty and politics, I think they're way more important in the current investing landscape. The way I would read is I think Walsh is going to be incredibly important in the history of the US for doing reforms and you have to ultimately tie it together. Him and Scott Besson, right. Obviously they work together for Stanley drug committer previously. I think they've all generally been consistent in recognizing that there's been some very bad externalities from the post JFC policies, from the Fed in creating extreme wealth inequality, in introducing massive moral hazard, disabling the function of the the money market. Right. To give signal, flooding the system with reserves and all these issues. And I think Walsh is coming in to try and genuinely get a handle on this and reform things. Right. And if you look at the task forces, the people he's picked, I think that's actually long term, extremely bullish, extremely good. And that's ultimately going to be very important for credibility. Now to do that, there's going to be a lot of pushback. You're going to have to spend a lot of political capital. So then the question is what do you want to waste your political capital for 25 bips here or there in some of your first meetings when you know for sure the President, President Trump doesn't want you to hike.

[Sponsor Announcer 1]

This episode is brought to you by Google Chrome. You think you know a browser, but Gemini and Chrome, that's new. It can help you with practically anything on the web, like restoring a vintage motorcycle from a 50 page restoration block. Or finally break down that long article you've had open for weeks. Gemini and Chrome is here for it, ready to make anything online make sense. There's no place like Chrome. Check responses set up required compatibility and availability various 18

[Sponsor Announcer 2]

when you need to build up your team to handle the growing chaos at work. Use Indeed sponsored jobs. It gives your job post the boost it needs to be seen and helps reach people with the right skills, certifications and more. Spend less time searching and more time actually interviewing candidates who check all your boxes. Listeners of this show will get a $75 sponsored job credit at indeed.com podcast that's indeed.com podcast. Terms and conditions apply. Need a hiring hero? This is a job for Indeed sponsored

[Tian Yang]

jobs that probably one of the behind the scenes trades is trade offs is that he probably yet assured him he's very unlikely to hike when he was getting the job. So then you have to ask is the data so bad that he's going to fall that he can justify to President Trump why they hide? If not, then we should just he should on the side of not doing it. Don't bring attention to yourself. Get your things in place. These reforms are multi month if not multi year process. That's absolutely vital and you don't want to take risk getting derailed from that for the sake of 25 bips here or there. Let the markets do what they may. Sure. Flatten the curve. Market can flatten, steepen the curve after, after the meeting. Hey, if you think the Fed needs to hike more later, great. But I don't think that's the dead goal. That's not the most important thing. That's just where we're coming from. Yeah.

[Jack Forehand]

What do you think are some of the most important changes we might see under war? People talk about less forward guidance, they talk about less use of the balance sheet. What do you think are some of the major reforms or changes we might see under him?

[Tian Yang]

Yeah, I think it's going to be extremely difficult, but they need to try and get the system away from the excess reserve regime that central banks have operated in because it does create a lot of distortion in terms of how capital is allocated. Right. If you're a central bank, you just mechanically less banks run huge reserve balances, you pay them interest just giving out money all the time. And then so you create very different incentives for what financial institutions are doing. So I think to the extent they can reverse that, that'll be extremely important. But the challenge has clearly been when you try to do that, markets crash because of this moral hazard post GFC environment we're in. So they're going to have to find a way to finesse it. And I think this is where viewing the Fed alongside Treasury, viewing best in everyone, people who have an understanding of how Markets work, how the economy work, but also how market participants think and coordinating. I think that's probably their best shot. They can do deregulation of US bank at the same time, shift the burden of credit creation financing more into the private sector, but give the private sector some backstop along the way for now. But just transition that towards say smaller banks.

[Jack Forehand]

Right.

[Tian Yang]

Regional banks, you know, undo some of the worst parts of Basel 3.

[Sponsor Announcer 1]

Right.

[Tian Yang]

I think that there's things they're trying to do, but it's going to clearly take a lot of coordination. You know, like we've seen with, you know, how masterfully Besser managed his dollar yen intervention. Right. Just like it's just, you know, in a way it's almost ridiculous. He has that notepad where it's like to do buy yen and just make sure you guys see it. Right. So you can see that, that I think you're in the hands of people who have, have a genuine understanding how market participants operate and that potentially gives them the shot of putting this off.

[Jack Forehand]

Yeah. It seems like the challenge is always everybody has big plans and then you just don't know what they're going to do when they're punched in the mouth. Basically kind of like the back of the Mike Tyson quote. Like that's the challenge is will they be able to get through whatever goes wrong and stay the course.

[Tian Yang]

Yeah, it's hard, right. Because the history is against them. But I think you have a better shot if you know how participants think. Right. I mean we saw it in a way after Iran with oil. Right. I mean it's almost masterful in that they're like, okay, let's just create so much two way risk that all these lever guys get stepped up both sides and eventually nobody, you just create P and L loss and suddenly nobody wants to speculate on the market. Right. And you just force that risk down and doesn't move as much. In a way that's kind of almost what they're doing on dollar yen. Right. You just create two way risk so much two areas, you just stop everyone out. That buys you some time. So I think those are probably signs of nuanced understanding of market operations. That gives them an extra quiver to that bow that maybe historically policymakers have not had or have not been willing to use, but today they're willing to use it.

[Jack Forehand]

This next chart is the number of US States meeting the SAHM rule that currently sits around nine. But I'm just wondering if you could talk about maybe for people who don't know, just define what The SAHM rule is. And then talk about what this chart means.

[Tian Yang]

Yeah, so this is named after Claudia Sahm. So like, you know, a very prominent economist who came up with this as a proxy for recessions. And so she did this for the whole economy. But the general idea was if you see the unemployment rate rise, I think it's like 50 bips off the, off like a three year low or like off the lows in a short space of time. That's usually a sign of a recession. And historically it's been, you know, almost perfect hit rate. Right. For the economy as a whole. Very famously, this rule failed along with many, many recession rules in 2324. And so again, I think there's a number of reasons that happened, but ultimately the simple answer is fiscal. Right. We were in a new fiscal dominance regime that invalidated a lot of these indicators. And so what we wanted to do was at least take that, but apply it to the different US states to give you a sense of the underlying picture. Because, you know, this idea of bifurcated K shaped is everywhere, like the economy, everything is getting bifurcated and K shaped. So by breaking it out, you get a sense of how things are doing in terms of the breadth, not just the aggregate number. So this does show you that for the US economy, there's still a decent amount stress in certain states that have not necessarily participated in AI or in energy. Boom.

[Sponsor Announcer 1]

Right.

[Tian Yang]

And so that also feeds into things like consumer sentiments, that feeds into political pressures. And so these things are all somewhat linked. But I think the point of this chart was to say, yeah, the labor market is not obviously booming, it's just more, it's just very bifurcated.

[Jack Forehand]

It seems like that word bifurcating explains so much about what's going on. You've got people who have a lot of money versus people who don't. You've got the tech versus everything else in the economy. It seems like there's been so many bifurcations in what we're seeing in recent years.

[Tian Yang]

Yeah. And I think it's only going to get more so because the market will keep being more distorted. Because we live in a world where governments are going to actively intervene. Whether you want to call it industrial policy or things in the name of national security. There's just tons of these things that will keep happening and they're going to drive a lot of divergences and create winners and losers that the market will get that a free market normally would find a way to digest and normalize, but it's Just going to be so many more of these things that make it harder for these bifurcations to close them down.

[Jack Forehand]

Yeah, I would think AI too. I don't know if you agree, but I would think AI would also cause more bifurcation.

[Tian Yang]

Yeah. Like until you get a policy intervention. Right. Like you know, famous in Korea, they're talking about finding ways to tax it to distribute the gains more widely. Obviously that caused the market to sell off. So they're like, okay, maybe we have a sovereign wealth fund invest. But these are all going to be, I think questions that societies will have to grasp and deal with. But yeah, if you have these increasing returns to scale winner take all technologies, then it's obviously inevitable that if you leave them to their own devices they're going to create extreme winners and lose it.

[Jack Forehand]

On the point of AI, how do you think about AI in terms of the job market? We've put up this chart. You have us challenger layoff job cuts by reason and in AI is one of the reasons you list there. But you have some people calling thinking that even if this is a technology is going to change the world in many ways. Like we've got a pretty rough ride to get there in terms of maybe some job loss along the way. How do you think about AI, I guess both short term and long term in terms of its ability to create job loss.

[Tian Yang]

So again, these are super difficult. I think there's some very good work by Professor Bloom at Stanford that look at long term historical technology diffusions and the like. So if I were to just cite like some of the things that I took from reading his work, I think basically these things take time is the first point. Right. It's rarely instant. There's real world human friction and things that slow it down. So I think that's usually the thing where these things are dragged out. But in terms of even in our business, day to day, yeah, it's pretty amazing. There's certain things where I think it doesn't necessarily force you to want to replace a lot of workers, but I think it reduces the value add for mediocre work that used to probably be more rewarded. I think that's the, I think that's what's going to be really challenging that again, if the market mechanism allowed to operate, it's just going to, it's going to. If you're like say four and a half stars or four stars out of five or anything, I think you're probably fine.

[Sponsor Announcer 1]

Right.

[Tian Yang]

You're probably still going to be okay. But previously if you Were like three three and a half stars out of five or something. You used to be pretty integral to most organizations. Competence, get things done. And at least so far, especially with agency AI and these workflows, like AI seems very good at giving you three, three and a half job at everything. And I think that's the labor market wage compression that it's hitting. So I think it's going to be a problem. Right. And I don't know what the answer is, but if you want to look for signposts, things like youth employment is clearly a big thing. Right. You know, as we're recording this, India has their whole cockroach movement. Right. And like you see this in lots of countries where youth employment, you're basically going to create lots of labor market pressures that will ultimately lead to more of a social, political response one way or another. Yeah. So again, I think. I don't know exactly how it plays out, but those are the things I would say you would look for. And most likely this leads to more kind of political shifts against capital is probably the ultimate.

[Jack Forehand]

Yeah. To your point, I think for people who are really good at what they do, I think it's a huge leverage mechanism. The more I've been using it, the more I learn there's just certain times where you have to interject and your personal knowledge leads to the AI being 20 times better. And so I think you're right. I think for those people at the top who are really good at what they do and who have maybe some information that the AI doesn't have their ability to use that, I think it's just going to change the world in many, many different ways. And you know, for the people who don't use it, it's. It's probably a bad thing. I would say.

[Tian Yang]

Yeah. Just more bifurcate. I agree with you. It kind of is scary. Even the last few months, the improvement is scary.

[Jack Forehand]

How much does it change what you guys do in terms of the types of research you do? I mean, is it. Has it changed it a lot?

[Tian Yang]

So I think we've tried to vary. We've generally adopted a lot of different data science, machine learning techniques anyway, going back. So I would, So I would say it's not like we're forced to adopt it from brand new, but like you say, I think we're in the mindset of man plus machine. Right. Like you have a human AI loop that. Yeah. The creative aspect, I think you still probably need a human a little bit like just telling the AI on its own to go, come up with something, it doesn't seem as effective as giving it a more clearly defined task. It's being amazing at falsifying things. So it's amazing figure out if there's a bug in your code or there's something you didn't think about when we built a model. Also like you give it your thesis and you can test it and tear it apart, right. And see what you're missing. That I think has been extremely good and you know, productivity enhancing. But if I would just sit down from scratch like hey, give me an idea and I don't give it something, then it can be a little bit tricky. So that's probably more the stage we're at. I would say agents, especially internal agents using own data and models, very powerful cowork, very powerful. And then the day to day is good, but like the day to day, if you're doing it via browser or whatever, like yeah, you do need to babysit it a little bit.

[Jack Forehand]

I want to ask about your VPX ETF because I think that takes a lot of the things we've talked about today, your framework and how you think about markets in the economy and it translates it down into an actual portfolio. So can you talk about how you build that portfolio, how you run vpx?

[Tian Yang]

Yeah. So VPX is a loan only US launch cap etf. So we designed this to try to maximize upside capture and minimize downside capture to S and P using the same stocks essentially. And the reason we thought we might have an edge in doing this is that a lot of traditional core allocation strategies are either obviously pure passive or they tend to be more static sector or factor allocations through time. And I think one of our thesis is that the world is now going to change so much that a lot of these slightly more static tilts and factor shifts that worked in the past 10, 15 years might not work in the future. And so essentially what we've tried to do is combine our cap cycle quality crowding, LPPL growth, all our models, macro everything to try and forecast forward returns for all the stocks and sectors and then try to maximize theoretical for returns. Right. And so it's a slightly different methodology to traditional kind of small beetle factor investing. And so that was kind of the theory in terms of how exactly works essentially is we basically use our capital cycle models to decide sector tilts, so which sectors are we over underweight. And then within those sectors we'll use all the other single star specific names to kick out the worst names. And then for the rest of the base and it would just hold the rest. So it's kind of very much this idea of addition by subtraction. And ultimately yeah, I just think of as the product is for someone who's like, hey, I have a lot of SPY or VTI in my portfolio, but I'm a little bit nervous at this point. Like, you know, half the index is in like very, you know, in like 10 names, whatever it is. And you know, there's lots, you know, the AI thing could carry on. Maybe it carries on for another year, but maybe it's over tomorrow. Right. And then is there a way for me to just get long term beta in a similar vehicle but that tries to adapt. So that's kind of how we thought about it and why we launched the product.

[Jack Forehand]

So you mentioned you'll kick out stocks. Will you kick out sectors as well or will you have all the sectors just at various weights?

[Tian Yang]

Yeah, we will kick out sectors. So I think there's. The market has a lot of obviously alternatives. So we take a lot of active risks if the model says it. So for example, when we launched in March this year, we ran extremely large tech overweight. That would have probably made everyone pretty scared if you weren't aware of the model. But then we rolled it up to May and then basically went underweight tech. And that was quite good for that market adjustment period. Right now we're running a pretty, we have a pretty big overweight on energy and semis again. So we're trying to take a lot of active risk if the model says so. And in periods when the model thinks there's less divergence, obviously we'll dial down the risk.

[Jack Forehand]

What I like about that is we live in a world where people don't want tracking error, where people are a little bit afraid to be different. And it sounds like you guys are definitely based on your conviction are willing to be different when it's called for.

[Tian Yang]

Yeah, I think that's the point. I think that's why we think that product potentially has space. Right. Obviously if you want benchmark hugging or you know, you can buy index or something simple and you'll be okay. I think this is more trying to be, trying to take a lot of risk when you think it's appropriate to try and maximize the upside and downside capture in a portfolio so you know it's even. You don't even have to sell your spy. It could be like, hey, you have $100 spy. You can put $5 into this 95 spy. But then it'll give you and try to improve the overall US Core allocations

[Sponsor Announcer 1]

Upside Downside this episode is brought to you by Google Chrome. You think you know a browser, but Gemini and Chrome? That's new. It can help you with practically anything on the web, like restoring a vintage motorcycle from a 50 page restoration block. Or finally break down that long article you've had open for weeks. Gemini and Chrome is here for it, ready to make anything online make sense. There's no place like Chrome. Check responses set up required compatibility and availability various 18

[Sponsor Announcer 2]

when you need to build up your team to handle the growing chaos at work, use Indeed Sponsored Jobs. It gives your job post the boost it needs to be seen and helps reach people with the right skills, certifications and more. Spend less time searching and more time actually interviewing candidates who check all your boxes. Listeners of this show will get a $75 sponsored job credit@ Indeed.com podcast. That's Indeed.com podcast. Terms and conditions apply. Need a hiring hero? This is a job for Indeed Sponsored

[Jack Forehand]

Jobs Chan, this has been great. I really appreciate you taking the time. As we wrap up each episode, we have two standard closing questions we ask. The first is what is one thing you believe about investing that the majority of your peers would disagree with?

[Tian Yang]

So obviously it's hard to define majority, but I would say we're true believers in this idea that investing is as much about doing great fundamental analysis, right, understanding how the world works, but equally it's as much about playing the game of investing. That's a lot of times understanding the players, the constraints, who are the buyers, who are the sellers, and I think doing both at the same time. I would say this is obviously antithetical to old school Buffett or Buy Horn. But equally, I would say 10 of analysis will say, hey, why do all that fundamental analysis. But I think we truly believe there's a heavy element of playing the game. But you can't play the game unless you do your homework and actually understand what's going on. So I don't know if that's a majority, but I would say that's something that, yeah, I would say some people would definitely disagree with.

[Jack Forehand]

Yeah, no, I think a lot of people. That was a great answer. And our final closing question is based on your experience in markets, what is the one lesson you would teach the average investor?

[Tian Yang]

Yeah, so I think I hit him a little bit with the Howard Moscow earlier. It's from his book the Most Important Thing. I read that quite early on in my career and I think it really did make a a big difference to me. And there's a couple of things he had in there, one of which was this idea of investing is like playing tennis, except your goal is not to hit winners but to reduce your own force errors. And there's a very similar kind of concept, I think, for long term investing that average investors should think about. And the other quote was something I said earlier, which was it's not what you don't know that destroys your portfolio. Right. It's the thing you know for sure that isn't true that destroys your portfolio. So I would say that's probably like the best thing to hold on to. Like living to find another day and creating your portfolio structure around that is underrated.

[Sponsor Announcer 1]

Right?

[Tian Yang]

Just like, you know, obviously we record this because of situational awareness. Even if he's right, like clearly he's not going to monetize as much as he would now right after this big drawdown and being stopped out.

[Jack Forehand]

Yeah, I feel like in these types of periods we're in right now, like people don't think about living to fight another day as maybe as much as they maybe should. They think about maximizing their gains as much as possible. So that's a really great lesson. If people want to find out more about you, about your etf, about what you guys do, where can they go?

[Tian Yang]

Yeah, so the ETF ticker is VPX. So like S&P SPX, but VPX in terms of our research and framework. Variantperception.com and we're also on Twitter as well. You can find us there if you search for variant perception.

[Jack Forehand]

Well, this has been great. Thank you for joining us. We appreciate the time.

[Tian Yang]

Yeah, thank you for having me. Enjoyed it.

[Jack Forehand]

Thank you for tuning in to this episode. If you found this discussion interesting and

[Tian Yang]

valuable, please subscribe on your favorite audio platform or on YouTube.

[Jack Forehand]

You can also follow all the podcasts in the Excess returns [e-mail] if you

[Tian Yang]

have any feedback or questions, you can

[Jack Forehand]

contact [e-mail] no information on this podcast

[Sponsor Announcer 1]

should be construed as investment advice.

[Sponsor Announcer 2]

Securities discussed in the podcast may be

[Tian Yang]

holdings of the firms of the hosts

[Jack Forehand]

or the Vuoricore short moves with you with everyday versatility and classic athletic fit. It's the one short for everything your day brings. Invest in your happiness and get 20% off your first purchase at vuori.comcore20 that's V-I.comK20 exclusions apply. Visit the website for full terms and conditions.

[Sponsor Announcer 1]

Hello.

[Tian Yang]

Look what T.J. maxx dragged in.

[Sponsor Announcer 2]

The Devil Wears Prada 2 is now streaming on Disney plus and Hulu.

[Tian Yang]

We are digital.

[Sponsor Announcer 1]

We are downloadable.

[Tian Yang]

We are streamable.

[Sponsor Announcer 2]

The fashion event of the year is certified fresh.

[Tian Yang]

Pull yourself together.

[Sponsor Announcer 2]

We have work to do. Critics say it's smart and witty and the perfect sequel.

[Tian Yang]

That's all.

[Sponsor Announcer 2]

Get Runway ready for The Devil Wears Prada 2 on Disney plus and Hulu. Rated PG 13.

[Tian Yang]

Their clients.