Intangible Value Investing Using AI + NLP with Kai Wu of Sparkline Capital | S07 E32

Summary

• Investment Approach: Kai Wu of Sparkline Capital discusses evolving value investing by incorporating the rise of intangible assets such as intellectual property, brand equity, human capital, and network effects, which are increasingly significant in company valuations.
• Intangible Assets Framework: Wu outlines a framework with four pillars—intellectual property, brand equity, human capital, and network effects—to systematically track and compare intangible assets across companies.
• Accounting Adjustments: Traditional accounting methods are criticized for penalizing companies with high intangible investments by expensing R&D and marketing, leading Wu to advocate for capitalizing these expenditures to better reflect a company's true value.
• AI and Alternative Data: Wu leverages AI and natural language processing to analyze unstructured data sources, such as patents and social media, to quantify intangible assets and integrate them into a quantitative investment process.
• Market Dynamics: The discussion highlights the dominance of large-cap tech stocks, driven by their intangible assets, and explores the potential risks and opportunities in the current market structure, particularly concerning AI investments.
• Crypto Factors: Wu explores the application of factor investing in cryptocurrencies, identifying factors such as market cap, momentum, and intangible value, with a focus on utilizing open-source data like GitHub activity and blockchain transactions.
• Globalization and Trade: Wu emphasizes the resilience of multinational companies with high intangible assets in the face of trade wars, as these assets are less affected by tariffs compared to physical goods.

Transcript

I hope we're live. This is Value After Hours. I am Tobias Carlile joined as always by Jake Taylor. Our special guest today is Kai Woo of Sparkline Capital. How are you, Kai? Welcome to the show. >> I'm good. It's glad to be I'm glad to be on. I like the name value after hours. Well, the idea is that it's a more relaxed conversation, the kind that you'd have after going to a conference in the bar afterwards talking about the things in real shop. >> Yeah. I thought it was after market hours. Yeah. Well, I was like, "Wait a second. Market's not closed yet." >> No, I know. That's It's confusing. It's It's you know, it's it's in the middle of the day, but uh but it's after hours. >> Yeah. It's four o'clock somewhere. tell us a little bit about Sparkline Capital and your approach to investment. >> Yeah, so look, I'm I'm glad to be with you guys on this value podcast. Um, you may find me to be a bit of a black sheep, but um, you know, I do actually hail from the uh the the school of value. I started my career at GMO actually um and have been, you know, a value investor uh since basically entering the industry what over 10 years ago. um you know worked for uh Jeremy Grantham at GMO and then was part of a uh spinout of um a hedge fund spinout there um kind of the number two guy on a quant hedge fund um with another GMO partner and then started Sparkline in 2018 and kind of the idea here is to continue to evolve um the the idea of value investing um but to take into account um the changes to the economy that have um happened over the past century or so um namely the the rise of intangible assets um things like intellectual property, brand equity and human capital that um at least based on the data I look at you know are comprising a kind of increasing share of the value of companies especially US and especially large cap companies. >> How do you define intangible capital and how do you then track it and make it sort of uh comparable on an applesto apples basis between companies? >> Yeah. So there the kind of framework I use is I have these four pillars of intangible assets because intangibles are kind of by nature heterogeneous right it's hard to compare apples apples in general so I think having a framework where you create separate pillars um is helpful as a starting point um so the four pillars um are intellectual property so that's like patents but also kind of trade secrets and any kind of software code um knowhow that's held at the uh corporate level um the second is brand equity so think like Coca-Cola um LVMH. Um third is human capital. Um and then the final piece is network effects, right? And that's think of like you know um Google search or Facebook or even AT&T or New York Stock Exchange and anything where adding more nodes to the network increases the value for all users and that's becoming an increasing large component of value for the big tech platforms of course. So then what kind of adjustments to the accounting like where where is GAP wrong now in the new world? >> Yeah. Yeah. So you're bringing up an interesting starting point. So when I first started doing my work on intangible research um I I did start with the accounting statements, right? right to go to the 10 10ks and 10 Q's and and one of the inconsistencies that you of course are aware of is that um you know physical capital expenditures are capitalized whereas um you know R&D and marketing um intangible investment is expensed right and so that's effectively punitive towards companies that conduct homegrown R&D homegrown marketing now there's a weird thing where if you do an acquisition you get this goodwill item and you can capitalize that with like impairment things but for for the most part um you know companies that are um intangible intensive are are penalized relative to those that are more asset heavy more traditional businesses. So the first thing I did was you know following the work of several academics as well. Um so this is this is well known um did work on trying to kind of make um intangible intangible um capital expenditures kind of on a consistent playing field and so you can actually build up you can actually capitalize um R&D say each year and kind of amortize it over some um period to build up a balance sheet item which is you know intuitive because you know now Coca-Cola or you know any kind of like tech company whose book value is like 1% of their total value that kind of becomes somewhat corrected where like now you're giving credit to say Coca-Cola for having an asset on its balance sheet that it's the brand that is being created over hundred billion dollars of advertising or you know um for a tech company all the R&D spent you know building out whatever software um they they they have and I found that to be helpful right so if you think of your of your baseline as being say Russell 1000 value versus versus market and you say what what if I you know redefine you know the book value component of Russell 1000 value by adding the capitalized intangibles well you know the the performance has been less bad relative to the market over the past say decade um but it's still not amazing and so that led me to the next point where I said you know maybe we need to kind of consider metrics beyond accounting information I think accounting information has two um you know limitations first is that just the amount of disclosures mandated by the SEC are limited right so you have like headcount on the human capital front but I mean that's not obviously distinguishing between the CEO and you know their assistant um so there that there's that component and then there's the other piece just that like the um you know out the kind of output function input output function of say R&D or or even marketing is pretty like um you know nonlinear where two companies can put $100 million into developing a drug and and one ends up with OEMIC and the other with nothing or two companies can buy Super Bowl ads and one goes viral the other one doesn't. So, so you really need to focus less on historical cost which is you know by definition what accounting does and instead on the output of the kind of um production function that is R&D and marketing and that is what led >> So what does that show up then on like does that just show up as revenue at that point? >> No. So ultimately yes ultimately the end goal is to you know model something that will eventually trickle into revenue. One thing I found is, and this is a separate result, is that intangible investment, I'll tell you how I define that in a second, um, you know, tends to lead to profits and earnings per share growth with a pretty long lag. Um, you know, if you look at company, if you sort companies by the intangible value and then look at the how that filters into EPS over the next decade, the first year actually decreases EPS, right? Because you're obviously spending money to make money and then it J curves up such that in 10 years it's it's a very, you know, valuable thing. you've now created the moat, you have a patent, you have a a trademark of of some brand that you can now charge premium prices for for the product. Um, but it takes some time for that to filter through to to earnings. Um, now, how do I define it? That's that's kind of where um I'm I guess I'm I'm a little bit different than than most folks in kind of the traditional value um school, which is, you know, my my background is is as a quant investor, you know, working for GMO. And, you know, I got pretty into like machine learning and alternative data um you know, I guess relatively early on. and um and ultimately what would become AI so natural language processing and and so a big component is to try to identify unstructured data sources alternative data sources to quantify each of the different pillars. Um an example might be to look at like patents right is an intuitive place to look for IP or trademarks for brand social media for brand. Um other examples might be like LinkedIn posts um sort of LinkedIn bios um so you can kind of build up a snapshot of the human capital um stock of a of a given company at each point in time. Um and so intuitively these these sources have a lot of um potential um value for quantifying intangibles. And of course like the challenge then becomes you know just the size of the scale and the fact that much of this data tends to be unstructured and trying to figure out how to work with that to kind of build structured factors that can then be kind of put into a more traditional quant process. Um which would be kind of effectively an ensemble approach similar to like combining a composite of price to book versus a sales price to earnings etc. any uh surprising correlations that came out of the machine learning process like things that you are counterintuitive? No, I mean actually I think the results are pretty intuitive, right? Like if you look at say a list of the companies with or the sectors say with the most um you know intellectual property intensive sectors, it's technology, communications, um healthcare. Um on the brand side, it's communication, consumer staples. On human capital, it's finance, um you know, tech, um you know, healthcare. Um and the network effects would be kind of communications would be the highest followed by some tech stocks too. So I think it tends to be somewhat intuitive actually the way kind of all it all shakes out. I think what's interesting is that you know we've managed to find a way to to quantify things in a systematic way right because you know most managers will say yeah of course intangible assets matter. Don't tell me they don't. Everyone knows that. Um it's just a question of how you go about um you know building up a systematic scoring process to identify these things um you know in a quantitative way. Why wouldn't the value of intangible assets just turn up in greater earnings or cash flow or why why can't you measure it just as a as a flow rather than having to sort of do these additional exercises to show where that intent >> you're trying to get out in front of it aren't you a little bit like and see what's going to when is you going to get those big paydays >> yeah I think it goes it goes with the J curve idea right which is that like a lot of intangible invest you think about Amazon being the the perfect example right a company that never made a dime for their investors um for decades and now is worth multiple trillions of dollars. Um you know they how did they do it? Well they basically reinvested into the business building out um goodwill customer goodwill through you know charging prices that were very competitive um for their you know building out the logistics networks the network effects and then ultimately AWS um which has been a big profit center um and so yeah I mean if you look at the the trailing earnings trailing free cash flows you know it looks okay um but you know still very expensive. Um if you look at it on an intangible investment basis actually they're making really interesting investments into these intangible assets. Um and for a long time it would have been a very attractive stock um you know on that basis. >> So you say that it reconstitutes these um value indexes so that you find what you would inexpensive relative to some measure. It makes them you're pumping up the measures of some of these so you can see how truly undervalued they are. >> Yeah. Yeah, I mean relative to like a more traditional price to book or even like a price to earnings kind of any traditional value metric, you're going to end up with a portfolio composition that is, you know, a little bit different. So less banks, less insurance, less asset heavy businesses, materials, energy, etc. And more technology, consumer brands, communications, healthcare um and and it will vary by by uh region. Like one thing we find is applying the same methodology to the US into the non- US um geographies, you end up with a lot more tech in the US because that's where all the talent and innovation is. and a lot more brand um outside the US which is you know if you're thinking about Europe um that kind of makes sense um and then through time as well through time things have changed obviously network effects are a lot more important today than they were 20 years ago because you know now we have six seven billion people connected on Facebook um back in the day we didn't have that um so these things can kind of change through time I mean also if you go back 100 years none of this is important or or very insignificant um and so it is kind of a dynamic process by which you know intangible assets have kind of become increasingly important um for for the uh day-to-day invest investor. >> I was looking through one of your doc documents and I saw you track the uh the number of PhDs. >> Is that a leading or or uh or is that a you want to let me tell you a fun let me tell you a fun story. So actually PhDs at least historically right um have been um you know a valuable indicator not just for and it's not just a tech it's not just a tech proxy right it's like adjusting for sector let's say the companies that are investing more in you know your high-end talent you know folks who are coming from I know MIT PhDs and machine learning those guys are you know pretty good and you know pretty soughta so the ability to to pull those people is an attractive thing one fun thing I did was I also looked at MBAs so MBAs are actually a negative signal >> a damn I know I'm wasting my time. >> Yeah, you're you're setting yourself up for that one. >> Yeah, stepped right into it. >> So, there is a positive correlation if you the more PhDs you have, the better your returns. >> Um, stock returns um adjusted for price. So, everything is normalized by market cap. So, in this case, like you know, it's not just number of PhDs because then you're just getting large cap stocks. You're saying, hey, per dollar invested, how much intangible capital am I getting out? whether it's PhDs or trademarks or patents. Um, you know, the whole idea is like a dividend yield. Um, you know, because we we're valued investors. If you build a portfolio simply of like the most intangible intensive companies, you know, highest R&D divided by sales, highest PhDs divided by total employees, you basically get tech, which is fine. But like I I don't know to what extent that's persistent moving forward because while tech has outperformed historically, right now it's kind of priced in. Everyone knows that these are good companies. And while network effects were for example um you know underappreciated for a long time, people kind of know that these things are not important now. And so now the key is finding undervalued network effects where it's not already priced into the stock. >> It would seem like also if you're you know if you're having to fish for these people with hund00 million signing bonuses at some point that's kind of neutralized a little bit, isn't it? >> Exactly. Right. If if if $100 million is coming out of the the stock, you know, in stockbased comp or whatever somehow, um then that's going to offset whatever value these people bring to the table for sure. >> Can you take us through some of the positions and how they would be turning up in your portfolio and sort of defying a more traditional classification of those positions? >> Yeah. What's sort of like the biggest surprise like jump from what we'd say like traditional value would say it was you know wildly overpriced or not good or something versus like the insight that you have from your process. >> Yeah. I mean so you know Toby and I were talking a little bit about this before the call. Um you know I think Mag 7 right that's basically the entire game these days right? All people think about is Mag 7. Um and and it has to be fair driven the returns of the market um for the past few years. Um and so actually you know in in 21 um when we launched you know our product you know some of the largest holdings and this is something that's public so you can look it up it's not you know were actually mag seven stocks and you know the Wall Street Journal actually wrote an article on on us saying you know the value investor buying big tech stocks as if it were kind of like a weird thing which it kind of is right like you know maybe today Met is in in the index but like you know it wasn't always like that um and you know the the question is why right these stocks many of them traded at you know price to earnings ratios that were very high like Nvidia was at 100, Amazon was really high. Um you know why would we buy these things? And I think the reason is is that um you know if you add into the mix these intangible assets, right? Nvidia has has CUDA network effects. They have you know you know very well-known in a culture that's well known to be innovative. They they do hire a lot of PhDs and employ a lot of very talented folks. You kind of tally all these things up and actually on on a relative basis these stocks were at the time at least amongst the cheapest things um on the on the market. Now, obviously, fast forward, you know, four years and times have changed and everyone knows Nvidia is a great stock and the stock price is up like a million times. So, you know, maybe it's no longer cheap, but you kind of get the idea that, you know, when you adjust for these things, it it does change the relative um you know, it kind of reshuffles the deck a little bit. You know, generally in favor of intangible intensive industries like technology, but more importantly, even within those industries, um you know, certain names tend to benefit more from these adjustments than others. Um, of course, as you as you would expect, >> if you're not expensing something, so in in the traditionally a tech company would be they they expense whatever they or or it runs through a salary or something like that. So it's not capitalized and depreciated subsequently. It's expensed in the year that it's incurred. So that reduces their earnings and this and their book value as well. And so then if you make these adjustments, you increase their book value and you're reducing their earnings or you're increasing their earnings as well. increase earnings and then reduce return on invested capital typically. >> Yeah, that was that was where I was going. So, doesn't it make them look less valuable if you if you do it that way? >> It makes them less attractive on an um kind of ROIC or ROE basis, right? Because think about like I think Mobis has done some good work on this um where he looks like Microsoft and it like has insane RO. It's like all right, this thing has like an insane but like the capital base that Microsoft has is is not true. It's like it's a it's an illusion if you if you adjust for the intangible capital the R&D Microsoft has done. All right. Well, now the RO R ro ROE Microsoft is like a normal level, right? So, you kind of see that effect and I think it's important to also mention that, you know, for this this can actually operate in reverse, right? So, what really matters is is for the kind of earnings based approach is the rate at which you're investing in R&D relative to the depreciation or amateurization in this case of of R&D. So in the case of you you're a stable a steady state company stable company and you're no longer investing in R&D you're kind just like you know coasting um and you have you know a lot of cap R&D capital from prior years is now depreciating or now advertising this this actually will be uh the opposite effect right so we're really only talking about companies that are investing in in in growth and are kind of on this accelerating growth curve not those that are you know maybe R&D intensive but are kind of on the decline mature companies >> so can you see then like you know you can sweat the assets in in a more traditional sense of like underinvesting and like maintenance and you're just boosting kind of cash flow short term. Do you see the same thing when it comes to IP and um or intangibles where some companies are sort of sweating their intangible assets? >> Yeah, I mean you definitely want to see companies that are continuing to invest for the future. I think that's something you do want to see. You know, one interesting thing I've been looking at more recently is this, which is, you know, I'm obviously a fan of asset light businesses like, you know, Buffett and and most people, but you know, one interesting thing we've seen with the Mag 7 in particular has been this pivot from this beautiful asset light business model like Google search to what's effectively like a utility, right? Meta is spending one-third of 30% of their revenues on capital expenditures, which is the same as the average utility, right? They have a tech multiple where they're basically trying to build a utility. And there's many game theoretical reasons why this is happening. and and you know you can go back to prior investment booms and it it kind of makes sense that this is happening. It's not surprising. Um but one thing I did look at in the context of this kind of trying to understand the implications of capex is you know to what extent does changes in investment rates lead or passage outperformance or underperformance for a stock right and so this is a really well-known finding from hama and French the so-called investment factor which is this idea that companies that invest tend to overextrapulate the the prospects of the future and tend to then underperform so I did something interesting which I actually bifrocated that into two components what if you companies that invest heavily but they invest in intangible assets versus tangible assets assets and the companies that that are investing in in intangible assets do just fine. It's really the the companies that invest in physical capital expenditures that you know have historically at least um you know suffered again on average I'm not saying that this is going to happen to Meta. I'm just saying that on average this has been the pattern that that you know investment in physical infrastructure has not been you know rewarded by by the stock price right and this is of course what we saw in the dotcom boom right when you had companies like global crossing basically go to zero we saw it in the railro road boom as well now in this case it's a little bit different I don't think meta has any you know balance sheet issues or any shortfall of free cash flow but you know definitely you know it's it's a you know concerning development especially for a component of the market that's now 40% of the index Yeah, that was that was where my question was going. There was that I saw a chart on Twitter last week at the end of last week that said that the MAG 7 or whatever the capex the AI capex spenders are classified under they're spending something like 60% of cash flows which is like double what it was 10 years ago on on this buildout. And then the impact of that will be in about 3 years time that their return on invested capital is probably half where it is now as a result of those depreciation amotization starting to hit the the financial statements. What do you what do you think about the likelihood of that? >> Yeah, I mean look, I think there's a we don't really know what the return on this investment will be, right? like it it there is a state of the world where we achieve you know artificial general intelligence and you know there's only a few companies that have the service to do so and they somehow are you know they they you know are a cartel and they don't want to they don't cut prices so they kind of collectively earn all the rents. There's a state of the world that that happens. There's also one in which um you know artificial intelligence disappoints that it's not adapted adopted by the enterprise. There's no no real use cases. It's too kind of faulty and and even if it and even if it does succeed in the long run, there's a state of the world where there's a short-term hiccup in which case these stocks which are currently priced close to perfection will re rewrite and these companies will have to readjust their investment plans. Um, and then there's the base case. And I think, you know, Toby, what you're outlining is the base case that, you know, we're transitioning from, you know, think of like, I don't know, like Meta as having being the average of this amazing asset like business model and this like, you know, kind of okay utility model, right? That's my base case, which is that like they're just kind of averaging into like a less good, you know, business, right? If they could spin out this this this data center thing, it would be less attractive. But like you know there's of course like these you know really extreme tales that's you know difficult of course to to kind of handicap that that probability. So it's hard to say. I mean my my my the way I think about things is just from a risk standpoint which is like you know there's basically been one risk vector so to speak that is now kind of driving markets markets up markets down and that is you know MAG 7 and more particular AI you know the AI you know kind of theme within MAG7 right if that's you if people are bullish then stocks go up and if they're bare stocks go down and like it's not a great market structure it's kind of a fragile market where there's one you know principal component driving everything and so to the extent you know you you are an investor. You you're thinking maybe, you know, I've done pretty well holding S&P 500, NASDAQ, you know, QQQ, whatever. Like, that's been great, but like at what point do I want to think about taking these chips off and and maybe diversifying into other things? And how do I do that in a way that's not just kind of buying companies that are, you know, oppositional to this that they're not negatively exposed to innovation, they're just like at least neutral or still positive, but maybe just less um, you know, concentrated in in a few names. I think it's funny to look at the reaction to the market to the spending now versus when Meta was doing it a few years ago under the metaverse you know name which people just hated and it sold off and I I guess that some of that reinvestment was happening uh before the you know it wasn't capex it was happening sometime further up the financial statement so it was like crushing their free cash flow as well so I guess there were two elements to it but I I feel like there was this fear that everybody had sealed themselves into the the rocket ship with Zuck for the for the run to the metaverse and then halfway through two years in he decided he didn't want to do it anymore. >> Yeah, I remember that when he did the when he when he um kind of called off the the investment splurge, you know, called for the year of efficiency, his stock, you know, did really well, right? So, I mean, there is a that would be the dream state for these companies, I guess, right? That's the other dream scenario where it's like one in which they all kind of collectively say, you know what, this was a kind of fever dream. let's just like step away, no more investment, you know, cuz I think I think one of the concerns I think is that a large component of the reason why these companies are investing so much is this kind of game theoretical arms race component. Whereas like if you're Google, you need to protect search at all costs. That's your entire business. And to the extent there's something that could potentially, you know, you know, impinge on that, you need to, you know, there's not a choice, right? I think Zuck has even talked about this specifically about like how, you know, this FOMO trade and you have to, you know, the downsides of not investing are too high, right? Right? So you have this kind of like um you know from a personal dilemma standpoint, you're better off all doing one thing, but they do they do the opposite because like they don't want one guy to be the one who raises the head like China or you know Soft Bank and then they kind of miss out. So like they all have to invest which it makes sense. You know it's rational for them to do it but it's like not necessarily the collectively the best thing if you're a shareholder in all those companies collectively. >> No. and they're now, you know, they kind of had their own little foms carved up where they were all like making a lot of money, but now they're all going to compete in the same space and maybe not as much profit pool there after when everybody's in this blood red ocean together. >> That's right. Yeah. I think it's funny to see h how much uh internet traffic seems to have fallen off most big websites which is clearly people not doing that traditional thing where they had go to Google go to search click the link read the page and now they use basically AI just gives you a summary of what the top something websites say and so >> it's likeformational like gathering >> yeah and so a lot of people use chat GPT or whatever their provider is as their search window and so there's No need to go through Google anymore. So I imagine that Google has seen that from their side as well. >> Yeah. >> And when you Google something, are you just reading the top little Gemini prepared section now? >> I do that. >> Yeah. I mean it just depends on like what specifically you're trying to accomplish. But yeah, I mean that that I think like there's a significant percentage of search traffic that can be just, you know, satisfied at that stage. Like you don't need to even ask further questions or or dig any deeper. uh it's it's going to upend the uh the business model for a lot of content providers out there. It's a very different approach. I don't know how they deal with it. I've certainly seen AC across a very large number of large websites like the traffic is down significant numbers in the last few years like 25 30%. Which would be very very concerning if I was running one of those. >> I saw the Substacks apparently doing pretty okay. >> Substack was the only one that was kind of like up and still performing really well. Yeah, we probably saw the same chart. Yeah, that's right. Substack was the top lefthand corner of the chart. Why do you think that is though? That's the real question. >> I'm assuming they're block I'm assuming they're blocked on chat GPT. >> Yeah, they I guess it's they've got they've got uh an email direct to consumer type business model where how often do you go to the substack.com to read it? Like often it's emailed to you and you just read it as a as an email. I mean again look maybe I'm an optimist here but like you know and I'm just kind of like warping things to my world view but like my my thought would be that you know much of the content is commoditized right like you can go you know they've already scraped Wikipedia and like just the random web and so like what really matters not to get kind of get the incremental improvements on models are like the kind of really truly unique content um and so like longtail stuff like because it's been shown right that like it's not about the amount of content it's about the quality like one good you data points one good piece of content is more valuable to training a model than you know a thousand kind of like you know whatever just noisy kind of you know low low quality right so to the extent that is true which maybe that'll be proven wrong um you know folks like I think Reddit for example is like a is a really important source of training data for a lot of these models um and because just because it contains as as we discussed like this kind of long tale of you know interesting opinions and and and whatever so like you know you you could envision a state of the world where that actually you ends up kind of continuing down that path where like you have even before AI you get all these like content mills, right? right? They're generally outsourced to like other countries where people just kind of put slop onto the internet and like use it to to generate SEO traffic like that I assume is going to be dead because now you can just do that on AI but at some point it's like well does you know that's not really incrementally useful to society or to the AI models more importantly and so you'd expect that you know people who are creating like you know genuine content would actually be um you know um advantaged in that world but I don't know maybe that's just me being optimistic. >> Yeah. How do you have like is how good can AI be if it's just trained on listicles? >> Right. Exactly. >> Buzzfeed Buzzfeed articles. >> Yeah. >> I will say like you know when I've used AI on my in my writing I haven't found it to be that useful and at least this is up in this is as of today um that that useful and like actually generating the writing. But one thing is really good at is generating like like headlines because like I'm pretty bad. So like I you know we'll try to post on LinkedIn or Twitter like oh yeah you know I just put out this new research piece like you know I'll you know say it in my own voice and it's pretty boring. it's like an academic like whatever and then like I'm like GPT make this better and it'll like give me like this like really kind of snappy headline like take out like the key stat well 40% of people are like you know and you put it up there and like I suddenly get so many more clicks so I don't know I mean obviously it is trained on social media data and it it does know like what makes a good headline um which I don't know so I guess like that just goes back to the the general idea right which is that like we should be using AI in areas that were that where like we're below average right because if AI always gives you the average outcome if you're really good at something, don't use AI. If you're really bad at something, then yeah, you should use AI. And I'm really bad at writing headlines. So, you know, AI is a great uh companion for that. >> It's it's got very mid tastes, I find. So, when I I always try to get it to do some writing for me, and it writes in this really like predictable kind of cliched midline way that is just so incredibly boring to read because it's like just a mishmash of everything you've ever read before, which I find really frustrating. But I think you can train it to give you good titles. I think it is quite good at that. >> Yeah. Yeah. And and I think, you know, to your point about it giving mid content, I think that has only become more more worse over time because a lot of the content we see online like maybe we don't really truly internalize that this fact is actually generated now, right? Like my guess is if you scroll through Twitter or whatever, like you just, you know, look through some stuff that a significant percentage of that content is actually AI generated. And so like it just starts to all blend into the background. Um and so, you know, part of what you're saying is maybe like five years ago, you would have been like, "All right, this is like a gamechanging thing. like this is a great this is content is amazing and now it's like everything you know they all use the m dashes and have like the word delve or whatever right you know >> all right well this is a generated that's a >> annoyingly I actually like dashes I've been using them for years maybe years I'm kind of annoyed that I can't use them anymore >> yeah I use them in in if you write in Google Sheets which Google whatever it is which I prefer because then it's you know available from computer to computer the m dash is just three of the dashes it's so easy to put in it's harder to do in word and yeah know now it turns out that's a a sign of AI writing which is a bummer >> right ruin for you >> JT top of the hour you want to take a chop at some veggies >> I will um you know I knew that Kai was was coming on the show today and I knew that he was interested in this topic I saw him post a tweet about it and I'd been reading it anyway so I thought it would I've been saving it for when he was coming on uh and we're we're talking about cash holdings Uh, and it starts with this question of why do some companies run their liquidity super tight and others sit on these mountains of cash? And so you might be thinking like, well, isn't having more cash always a good thing? Isn't it safer? Uh, and and as with most things in business, like the answer is it depends. It's complicated. Um, so cash can it can kind of be a blessing or a curse in some ways. So, you know, uh, too little is obvious. You know, the company risks collapse or is fragile to downturns. Uh but too much then you know it can risk dragging down returns on equity obviously but um it could also frustrate investors when it's sitting there and and probably most importantly it's tempting for company management to do these creative or destructive uh you know often ego-driven decisions you know M&A activity that uh you know when you're sitting on a fat wallet you can make some stupid decisions. So um there's this tension and and between safety kind of on one side and stagnation on the other. Um and that is at the the heart of this recent white paper from Michael Bobson and Dan Callahan that was called cash holdings data theory and alternatives. So we're going to do like a little minibook report today on that. Um so the the Michael looked at uh he analyzed over five decades of data across industries, life cycles of businesses, global markets and to understand how companies think about cash and what how have they expressed that and their conclusion is that there's no universal right amount. Instead cash is kind of a strategic lever to be run inside of a business. So um we'll start with some of the scale of it. At the end of 2024, the US public companies excluding financials were sitting on about two and a half trillion of cash. Uh that's no small change. It's about 4.7% of their total market cap and 9% of their total assets. Um so, and for reference, the average from 1970 to 2024 was about 7 and a half%. So, we're kind of in normal territory right now. Uh this cash is not evenly spread out across the corporate uh spectrum. A handful of giants account for a very disproportionate share of it. Just 10 companies, the top 10, hold one quarter of the excess cash. About a third is held by just 21 companies. An entirely half of that two and a half trillion lives on only 67 balance sheets. Um, so it's it's it's a power law. >> I thought you were going to say seven balance sheets then, but that's that's fine. Keep going. >> No. Uh and so it you know in Berkshire which is was technically excluded from the study because it's labeled as a financial you know they've got north of probably 350 billion at this point on their balance sheet. So um since the 1970 companies have steadily raised their cash balances from an average of about 6% of assets to in in the late 20th century to almost 10% uh in 200 uh1 to then 11 and a half in 2020. Um now not every industry hoards cash in the same way. Health care and technology dominate the the rankings. In 2024 the median healthcare company had nearly 40% of its assets in cash for biotech firms. Sometime you know a lot of times they're pre-revenue and so they'll they'll hold even higher. Uh sometimes more than the market cap entirely is held in cash. Um technology isn't far behind them with the medium at 23%. So think of software firms you know with big R&D pipelines. Um and then by contrast on the other side of the world you have utilities and energy firms which they rely heavily on tangible assets and they enjoy relatively steadier cash flows often especially on the utility side. Uh and they'll hold very little cash then and often less than 4% and in fact even sometimes utilities are forced to take on leverage from the rateayers perspective wanting to have a lower cost of capital. Uh so the split uh reveals a deeper story as well like it's not really just about risk management necessarily. It's also the nature of the business and the asset the cash is then supporting there's so there has to be some common sense to that. Um if your assets are all intangibles like codes and patents or algorithms all the stuff that Kai's been talking about um that those those things don't make for very good collateral. uh banks are pretty hesitant to lend against them and which means that companies that are kind of relying uh that are relying on tangibles they kind of end up self-insuring that by holding more cash. Uh and as you know over the last few decades the corporate world has shifted uh from building factories to building a lot more intellectual property. Um so let's see uh the life cycle of the company is the next thing. This this also plays a critical role in cash policies. As you'd expect, early stage companies, pre-revenue companies, they tend to hold a lot more cash. Their operations and investments, they they eat money and this financing, you know, u finance can provide a lifeline. But at that stage, like you need to hold a lot of cash and that the median then firms for early stage is 31% of assets in cash. At maturity, a company is ideally cash generative and and then the ratio then falls to about 8%. And then in decline actually cash balances arise again as sometimes you know there's devestatures that a company can do or a shrinking operation that might be in runoff and so you'll see cash balances balloon again. So there's kind of this U-shaped curve in the life cycle of the business as far as cash balances go. All right, quick quiz question. Uh was corporate debt to total capital higher in 1974 or 2024? >> 74. 74. >> Uh well, you guys are correct, of course, if you knew you would be. Uh so 40% um debt to total capital in 1974 and 15% in 2024, which is a little bit surprising. You kind of have this narrative that like, oh, everyone's overlevered today. U maybe not true. Um now, why do firms hold cash? Economists and finance scholars point to four explanations. Uh number one, precautionary motive. Uh you know, it's kind of the simplest like cash is an insurance. When markets freeze, you know, companies with cash survive, co 19 revealed that in a big way, right? Like airlines, which you know, had been spending their free cash flow on buybacks, uh all of a sudden faced like literally zero revenue. Uh what do you do then? Um you know, you need a cash cushion. Um, number two, tax management. Before the uh 2017 uh, you know, Trump tax cuts, there were lots of cash that was parked overseas, these US multinationals. Um, and they weren't they didn't want to repatriate it and pay the taxes on it. So that led to this these massive offshore cash piles. Um, number three is option value. Um Alice Schroeder had said that um Warren Buffett thinks of cash as a call option on every asset with no expiration which I thought was a pretty eloquent idea. Um and in volatile markets ideally like a cashrich company has this anti-fragility to it where they can you know strike quickly and buy distressed assets or invest in opportunities that their rivals can't touch because they're flatfooted with not having enough cash. Uh and then the last one is agency theory. And this is probably like the most cynical uh explanation. So this would be, you know, managers may hoard cash because everyone else is doing it basically in their industry and they don't want to be look like an idiot and you know be cut out. U so they're reducing their own personal risk by doing whatever what everyone else is doing. So maybe that means running it at a red line of cash, you know, where you're really low and levering things up because all your competitors are. Uh and then there's some studies that show that public companies hold twice as much cash as private ones, which um suggests that like governance and incentives do play a role in this. Uh so so um we'll wrap this up here. Um it's already running a little long. So so the um you know pure averages aren't a policy. They're they're really kind of more of an alibi, which speaks to that, you know, what we're just saying with these principal agent problems. The real question isn't how much cash, but really what would you make what what would make you spend that next dollar or give it back. Um, I think that's like an important part. Um, and if the board can't really answer those cleanly, like the market will answer for you. Uh, and then lastly, you know, cash is a mirror. It like it shows, you know, fear or hub hub or readiness to act. Um, so if you read the balance sheet and the cash flow statements and and you can often see the strategy that's then revealed, like you can look through 10 years worth of cash flow statements and get a pretty good sense of what management thinks. Um, or sadly sometimes the lack of a coherent strategy, which also comes up. Uh, but when it's done well, it it probably rivals one of the most important levers that you have to pull in the investment equation. >> Good stuff, JT. Usually I'd give a I'd go around the horn and give everybody a shout out, but I my text window is not open today. I can't get it to open up, so I'm sorry about that. >> Uh I don't know if the text is appearing for everybody watching from home or not. So I hope it is in a well. I'll have a look. I'll check it on the >> on the recording. Um >> Kai, you did a uh a piece on crypto where you're talking about some crypto factors. Are there factors in crypto beyond u momentum >> volume? >> Uh yeah. Yeah, I did this paper um and um I got savaged by the the value community. So >> by the value community or the crypto community >> value community by by Cliff as the Financial Times. Oh no. >> Yeah. No, I mean look like and to be fair I mean look like isn't this what he does? Um no but look like basically the the idea um of this research was to say okay we obviously know there's a market factor in crypto like there's in every asset class um that that beta is you know not not perfectly correlated to the stocks and so that's its own factor >> um you know there >> how do you measure is it Bitcoin >> in this case I did a market cap weighted index of okay um of all crypto which is basically Bitcoin until more recently >> um and then you know there's a small cap factor so small versus large um kind of the same way you define it in FMA French with stocks. Um and then momentum which you mentioned um and then the final thing I did and um is is to kind of create this intangible value factor. So you know I mentioned intangible value in stocks and how that's created using kind of patents and how much price versus you know per patent. In the case of crypto it's a little bit different where the you know that the entire um you know value of these companies are generally being built in the open. So you don't really have patents, you have open source code. But what you can do is you can go on the GitHub repositories and say, "Hey, how many like developers are working on this? How much how many different commits do you have? How many lines of code? Um you know the nice thing is of course on the um blockchain ledger side is kind of by definition for each chain you can kind of see you know in real time basically you know how much activity how much dollar transaction volume is flowing through um the the ecosystem right? Right? So think of it as like basically open up opening visas um tank Q every second. Um and you can also look at daily active users. So kind of a SAS style metric. Um you know how many unique wallets are operating? You know how um how actively are they transacting um you know defining activity in different ways, right? So you can you can you can actually create at least in my mind kind of fundamental metrics the same way I mentioned across these four pillars, you know, human capital, brand, IP, network effects in in crypto. now that they're just the data sources are a little bit different and in some ways actually better um and so basically I built this and you know without going too much into details that was kind of the fourth I guess fifth factor or whatever right so then I said all right let's try to you know go around and kind of test each of these things and so you know market worked small cap has historically worked although not more recently and again it's maybe the same thing that's happened in stocks where mag 7 is dominating stocks and maybe bitcoin is dominating until more recently um crypto um momentum is interesting because you know it does seem to work historically. The one thing is that it operates on a faster cycle. So whereas like in equities you typically look at like one month 12 month momentum. So 12 minus one in in in crypto it's more on a matter of weeks which makes sense not maybe works in seconds too. Maybe there's actually trend maybe there's actually inversion on a second horizon. I bet you there is because of the micro structure. Um but at least on the weekly horizon say you know one month to one week or six weeks or whatever um to to one week you're you're seeing some strong trending behavior. you also find something interesting when you intersect I guess the fourth one is is is intangible value you do find a value factor or this intangible value uh factor um in in in the data you know so one interesting shows up interesting thing shows up when you like intersect these factors with size right so think of like morning star style boxes right you know small cap value large cap value small cap momentum large cap momentum um in in crypto at least momentum tends to work better in large caps right and that's a little bit counterintuitive because you know in in stocks it works better in small caps where it has historically Um, and I think that has to do with a few things, right? There's the argument that um, you know, the the largest assets are more macro assets these days. Like there Bitcoin is a digital gold, right? It's a it's a it's a um way of getting exposure to, you know, fiat or whatever, right? Um, and so therefore, that's where kind of you see a lot of the trends. There's also like a lot more narrative attention is being spent on the large caps whereas the small caps are kind of forgotten. And then the final piece just being leverage that you can lever up um you know Bitcoin and ETH and these sorts of assets and so you end up with these kind of cascading effects where you have like these virtuous and vicious cycles um due to leveraging and deleveraging when you don't really have that in small caps because you can't really lever up your like flower coin or whatever. Um >> thank god >> and then and then you know I found that intangible value tends to continue to work pretty well works better in small caps than large caps which is consistent with the way that we've seen things work in in equities. Now there's one kind of interesting wrinkle here um which I brought up in the paper which is you know one of the things that people talk about and you know going back to cliff asnness is this idea that like you know small caps have generally not worked because of like this junk component that a lot of small caps are unprofitable and kind of junky um and so you know if you kind of filter that out it's it tends to work again right so I kind of did the same thing where I said look like you know these intangible you know if you look at intangible value one thing that's interesting is that by definition you're looking at price to x where x is some kind of fundamental metric and so by definition anything the divided by zero goes away, right? So, you're kind of like removing the the junk. It's almost like a quality filter, I guess, in the small cap universe. And so, that kind of explains a little bit, you know, in my mind at least, you know, why um you know, why why there is less of a small cap effect today, right? Just that like a lot of the these um these uh tokens are basically vaporware. Like there's no actual users on these things. they they get launched in 2021 in the boom and then like no one uses them anymore and they kind of just sit there as a zombie in the same way that like you know if you're a Japanese company and you you never get delisted after 1990, right? Like you're kind of just like sitting there as a zombie. Um which is fine, but like you know the problem is that you can never do list these things are always going to be on the blockchain by definition. You just once they're created they're there. Um, and so you need ways to kind of like filter out some of the the, you know, speculative tokens, the vaporware, and and in doing so, um, you know, you actually can restore, you know, a lot of the, um, you know, kind of perceived missing premium. >> Has anybody else ever done factor work on on crypto? >> I saw that um, Kraken and CF Benchmarks came out with a piece recently. I'm not exactly sure what it it said um, but I did see that they put something out on that topic. Um, there have been a couple academic papers that have looked at these sorts of things. I don't think they're widely circulated. Um, but you can I think I cite them in my paper. Um, so they're out there. Um, and um, and yeah, I mean I I think it's, you know, my guess is that like what we're seeing in crypto is kind of like what we saw in equities, right? where it's like, you know, you start out with like just Bitcoin which like, you know, and then and then over time you build out these index funds like Bitwise has, you know, Bitwise 10, these sorts of products, you know, and then you start to see, you know, um, you know, kind of smaller cap things get launched and then, you know, eventually, you know, you have you you have asset pricing models that kind of come up in order to help investors understand what's going on in the long tale of assets and that's when like kind of the style of premium, right? Basically, AQR built their business, you know, pick on cliff on this whole idea within equities, right? And I think it may not be me, but I think somebody will will um you know, eventually over time kind of build this the same edifice out um in the crypto space. Why wouldn't they? >> And what was what was Cliff's uh critique of the paper or of the >> I don't think he likes I don't think he likes crypto. >> Okay, >> just stop there. >> Yeah, >> cuz somebody gets the Nobel Prize in economics for this. So, it's got to be it's you or one of these other guys who published their paper first. You're the fam of French of uh of crypto. >> There you go. >> I hope not. So um back to back to the uh the equity side. Yeah. The the phenomenon that I have observed and I've been tweeting out a bit recently is just this uh dominance of large cap over small cap and it turns up anywhere you want to cut it. Like if you look at the S&P 100 which is the largest 100 burst versus the 500 and in this instance the 500 is the small you still see the same effect where you get the 100 massively outperforming the 500. You see it in equal weight versus market capitalization weight you get almost exactly the same shape. So does any of the work that you do does it put that into any any context? Does it explain that market capitalization? Does it sort of normalize for it or how do you how do you deal with that? >> Toby wants to know when it's going to stop also. >> That's exactly what I'm asking. >> That's exactly what I'm asking. So look, I mean, yes, small caps have underperformed large caps. But you know, if you were um the counterargument, right, to what you said is that well actually large caps have had an earnings growth that has justified their, you know, their appreciation, right? Like the Mag 7 have actually, you know, had earnings growth. we'll see whether it persists. That has been pretty impressive and has kind of justified the increase in their price. Whereas small caps, international stocks, emerging market stocks, you know, these kind of unfavored categories have not, right? Like I I I I focused more on international stocks. I did a whole paper on this where I found that, you know, over the past what 10-15 years um in kind of real US dollar terms, you know, so just for inflation, international stocks have basically had no had no earnings growth, which is pretty pretty disappointing. And you know to some extent that's I think at least the reason why they've they've underperformed. So not only have the earnings been tepid but like you know investors are tripling that moving forward and hence they have experienced no um say margin valuation um expansion whereas in the US they have because the index has become quote unquote higher quality you know more growthy um and um as a result investors are extrap extrapolating that moving forward right and that same story I believe can be applied to small caps right um which is small caps companies have underinvested in intangible assets think of the example of like JP Morgan and a lot of these big companies that are hiring all these API engineers and launching blockchain initiatives and you have like the regional banks which I don't know what they're doing right they're not doing well so like I think you you see this kind of across the economy both in terms of value versus growth um US versus non- US large cap versus small cap where the folks who have invested in tangible assets over the past two decades have actually experienced higher growth and like you can show this like I I mentioned that like study I did on the cross-section where I said you know there's a J curve right where you companies that invest um in intang iuals have then subsequently gone on to outperform th those that have not and that can be applied on the on the aggregate level to different groups whether it's like US versus Japan or US versus Europe or US large versus US small or value group versus growth. So a lot of the the kind of like um you know perceived discrepancies that we're seeing in in growth rates across um groups I think can be explained you know at least partially if not fully in many cases by this intangible um discrepancy. Now multiples have also changed right to and perhaps they're overshooting in some cases and I I think in most cases they are overshooting um and so in other words said differently right I actually debated cliffness on this too is like the value spread value growth spread right like it it seems like at that alltime wides and you know will that lead to future stock returns that are you know extraordinary for the value versus growth like I think you know there's kind of two components one is that if you just are intangibles the gap's less big but there's still a gap right so it maybe explains half the gap in this example Um, so I think that like, you know, it it does help explain what's going on. Um, and primarily through the earnings channel, but that the multiples are still a little bit like, you know, kind of overextrapulating. I think it might, I think, in this case. >> Is it a compositional problem? Because accounting doesn't deal with software particularly well because software is unlike a physical edifice. It's something that it's this expense runs through the salaries of the engineers who put it together and there's not a lot of like material that goes into it. And so that gets that's where all the problem is. To what extent do you see intangible sort of predicting returns outside of tech? >> Um yeah it's interesting. So you can run these intangible value like studies both within you know across the market where you allow it to go bottoms up and choose a sectors or you can constrain it to say all right I'm just going to go kind of pick the pick the most intangible value stocks within each sector right or within each industry or within each country. And in general, you find that um you know, intangible value works in both the new economy and the old economy. So it works in in in software obviously as you might expect, but it also tends to work in even sectors that are kind of more traditionally asset heavy businesses. I think the reason why is just that there aren't any growth investors who are even trying to price these things in those sectors, right? Growth investors are all kind of all over tech and they just have basically left the old economy for dead. And as a result like if no one's looking at it even having a smaller edge you know even if the edge is only on a smaller share of the balance sheet that's still meaningful. Now what's really interesting actually I looked at this and I think this is actually worth mentioning is the counter of this is the opposite of this which is traditional value. Take the most basic FMA French price to book vector. If you are careful in how you constrain the universe to only be um kind of old economy or or low to be more precise like non like low intangible intensity intensivity companies actually it's done just fine right so traditional value has done just fine if you exclude the you know sectors of the economy that are you know the regions the the the stocks that are more intangible intensive of course the problem there then the counter argument to that the problem with that is that that share is becoming diminishingly small in the US right like um that used to be most of the market and over time it's becoming smaller and smaller because intangible assets matter for every company now um you know maybe less for some than others but um is is still important >> um but I I think it does at least you know make a kind of point on principle whether or not it's useful in practice that you know maybe it is the case that you know you know that that some some of these factors have have been underperforming because they the omission of of these intangible components which when added um you should in theory correct It's interesting because one of the observations that Cliff made was that price to book has worked did work pretty well through that period of time when value was really getting hammered specifically because price to book was the less good value factor and so the ones that were >> the worst value factor the one least representing the weakest. Yeah. >> Yeah. I think that I think that speaks to the cyclicality part right like you know like the these types of like value factors like there's two reasons they outperform. One is just like there's a recycling effect, right? that like you know um VA these stocks prices rise more than their earnings underperform you know and that growth stocks underperform because they you know grow less than the market's pricing in right that and that effect seems to have diminished at least based on what I look at and and you know over time whereas like the the cyclicality component where like there are groups of stocks tech versus energy US versus emerging um value stocks defined this way versus growth stocks that do go through cycles right and I think it is undeniable that value stocks are in a disfavor sector right now and at some point that rubber band will likely snap back and these stocks will likely experience a cyclical rebound now will the trend line continue to be you know if you go back to the French chart since what 26 there's just been this nice like 5 percentage point of like you know smooth outperformance for value versus growth and obviously the past 15 years it's been kind of falling off that it may rebound but will that necessarily mean that the trend line will continue to be up that I think is is um you know more open to the debate >> uh do you have any view on I saw you wrote one of your pieces on trade wars. Do you have any view on the on the tariffs? >> Yeah, so I wrote that paper um maybe a week after liberation day or two weeks after liberation day. And kind of my my point there is, you know, a lot of people were um starting to kind of get concerned about geopolitical risk and to the extent that most people have home bias and they're mainly invested in US stocks were effectively dumping their multinationals to buy domestic stocks, right? like RH is the best example of a stock that went down like this 50% just like a single day um to go buy utilities and safe stocks. And so I wanted to do was ask the data and say over the long you know arc of history um was that would actually have been a good strategy. And what I found is that multinational stocks whether importers, exporters or pure multinationals at both import and export that that class of companies has massively outperformed their domestic peers whether in the US in Europe whether for sector um just kind of on all dimensions that these companies have done better and you know one explanation right is selection bias that if you're an A player you want to compete on the global arena where the TAM's bigger if you are kind of a B player you want to kind of hide in your local market and like you know use um you know tariffs and kind of export controls or whatever import controls to kind of protect your your niche, right? So, that could be one explanation, but it could also be the fact that, you know, being able to um be connected in in a free trade ecosystem is actually helpful, right? You know, the idea of outsourcing to lower your costs um to access customer bases across the the globe. Um there's a lot of tax advantages that multinationals can access that those who are, you know, stuck in one country don't necessarily have the ability to do. Um so, anyways, I thought that was a really interesting finding. The other kind of finding I you know homeed in on you know related to intangible assets is this idea that yeah look intangible assets. So first of all multinational companies tend to be more intangible. There's like a positive correlation. Not all of them but like sometimes like it's a non non-per imperfect but positive correlation. But the other thing is that like when you think about tariffs like what is tariffs really doing? They're making it really hard for a good that's being exported through over a border um to to to pass without like a tax. Intangible assets by definition don't have to pass borders, right? you can't tariff an intangible asset. And so it kind of became a really interesting, >> you know, way of of building a portfolio that is still long globalization and still long free trade, which to me is capitalism's golden goose, right? I like free trade. Um and um but to do so in a way that is, you know, more insulated to some of the risks that we've seen with, you know, um governments being able to kind of like capriciously, um put tariffs on, you know, their their trading partners. Um and obviously like you know think of a company there's obviously shades of gray whereas like you have pure intangible you know um revenues like you know software licenses and then you have pure physical goods and then you have the in between which are like you know Apple's iPhone which is like um of course a physical asset but a lot of the value there is kind of the IP and so you know there's some kind of transfer pricing you know gets really complex um but there's you know of course an in between category as well but like you know staying kind of the most intangible um uh intensive companies tend to actually you know, relatively well positioned for trade wars. >> Yeah, it's fascinating stuff. Um, we're coming up on time, Kai. So, if folks want to follow along with what you're doing or get in touch with you, what's the best way of going about doing that? >> Oh, um, you can go to my website. It's just sparkline capital.com and you can read my research there. And I have a um a form you can kind of um submit um any kind of like messages to me or you can email me directly. Um um you can find my email actually on any of my white papers. um um which are on my website. Um it'll be in the upper leftand corner. >> And you're in Twitter, too. >> I am. My handle is Yeah, my handle on Twitter and LinkedIn is the same. It's C Kai Woo. C K I W. >> Good stuff, JT. Any final words? >> No, thanks for coming on, Kai. It's good to see you. >> Yeah, no, fine. I'm I'm glad I could join Value after hours. >> Thanks, Kai. Uh we'll be back, folks, next week with any luck uh with the chat fixed up again. So, I hope that's I hope that's working then.