Value Investing, Inflation and Expected Returns with Rob Arnott

[00:00:00] Welcome to Excess Returns where we focus on what works over the long term in the markets. Join us as we talk about the strategies and tactics that can help you become a better long-term investor.

[00:00:30] We're also investing the impact of passive flows in the market and much more. Rob ends with telling us what worries him but also what he's most optimistic about in terms of the future in the markets.

[00:00:39] All investors can learn from discussions with some of the industries brightest minds like Rob. As always thank you for listening. Please enjoy this discussion with research affiliates. Rob are not.

[00:00:48] Hi Rob. Thank you very much for joining us today.

[00:00:52] Deleted to be here. We always appreciate having you one talking to you about the economy. I think a lot of macro topics that are on investor's minds and we're going to do much of the same today that we've done in the past and some of the things I think that we're going to discuss our sort of where, you know maybe we are with inflation and the overall economy.

[00:01:15] We're going to talk a little bit about value investing, passive investing impact on the markets and it probably wouldn't be an investing podcast these days if we didn't at least drop AI in and see what your current tape is on that.

[00:01:31] Let's maybe start with inflation. So you know inflation has come down but it's not at the Fed's target yet and there seems to be a pretty wide divergence with where people are on no landing scenario, the hard landing scenario, the soft landing scenario or whatever scenario so you know what is your take right now on that.

[00:01:53] Sure sure well.

[00:01:55] Well market prices are set based on narratives based on a set of beliefs that is held by the consensus that represents the consensus view and betting on a narrative is a complete waste of time.

[00:02:15] The reason for that is very simple narratives are usually largely true but there are a hundred percent reflected in share prices already.

[00:02:24] So betting that the narrative will come true is the same thing as betting on nothing.

[00:02:31] Now that actually comes into play on both of the topics you alluded to inflation and the macro economy, the narrative on inflation is

[00:02:43] thank goodness that's over now we just need to wait for the Fed to push us through the last mile and get it back under control.

[00:02:54] And the narrative is largely true inflation is much more subdued. It's back down into sensible territory in the three, three and a half range.

[00:03:08] Okay, but I think it's awfully useful to ask where is the asymmetry. Now, 10 year break even inflation, the difference between tips yields and treasure yields for the 10 year bond is currently at 2.3%.

[00:03:27] So how do I think about a symmetry there is the coming 10 year inflation rate more likely to be 1% below that or 1% above that 1.3 versus 3.3.

[00:03:42] I think if you asked 100 people knowledgeable about markets and about the economy, you'd probably get 80 out of 100 saying well 3.3 is more likely than 1.3.

[00:03:57] So that's an asymmetry. You can have part of your portfolio positioned to do better in a regime of elevated inflation rather than a regime of rapidly subsiding inflation settling in at 2% in the next couple of years.

[00:04:17] The same asymmetry can be observed in the macro economy. The narrative was gosh, I hope we don't have a hard landing to gosh, I'm wondering if we're even going to have a soft landing to gosh, I think there's going to be no landing.

[00:04:40] And if that's the narrative, then the question is which direction is the asymmetry more likely now some macro economic variables that are interesting PMI is currently in its worst quintile ever inventories are in their worst quintile ever.

[00:05:02] The yield curve slow is inverted to an extent that's its worst percentile ever and rate of change of the cost of capital using a blend of short intermediate long term rates and using a blend of 2 to 3 year rate of change is in the worst S. Allen history.

[00:05:26] So cost of capital up yield curve inversion which historically is a remarkably good predictor recession PMI is an inventory is all looking bad. Okay, then why's the economy just chugging along just fine is the consumer.

[00:05:41] The consumer is propping up the economy.

[00:05:46] The way I like to think of this is that during the pandemic people took on the attitude I might as well spend like there's no tomorrow because it might be no tomorrow.

[00:05:57] And that pattern of behavior has persisted long after the threat of the pandemic has receded and so we still have very aggressive consumer spending once you don't have a drop of money into people.

[00:06:16] People's bank accounts anymore once their bank accounts run dry there's always credit cards well those are now at record levels and rising relatively fast so at some stage you get a come up and say gee I can't spend any more because I can't even borrow anymore.

[00:06:37] And so the question is will the consumer ramp down spending before the rest of the economy has a chance to turn.

[00:06:49] Maybe yes but the consensus in the market says I'm not worried about that that's not really an issue and so I think there's an asymmetry in both cases it's not that I'm saying the consensus is wrong.

[00:07:05] I'm just saying that there's an asymmetry that the consensus is more likely to blindly optimistic and too little attention is paid to the adverse tail which is not highly likely but not unlikely.

[00:07:24] And what's interesting is both inflation that is higher for longer than expected and an economy that moves into slow down or recession.

[00:07:37] Both of those point in similar directions in terms of investment implications firstly mainstream stocks and bonds don't do well in the face of rising inflation expectations.

[00:07:52] Both of them don't do well in a recession well bonds don't do well in the early stages of a recession and then they find their bottom long before the stocks do.

[00:08:03] But this suggests a risk off posture not because we think economic or inflation risks are drastically elevated but because we think they're higher than the consensus gives them credit for.

[00:08:20] One last observation on inflation is we did a study a year and a quarter ago in which we looked at the 14 developed economies that were already developed in 1970 so that leaves out countries like Singapore and South Korea that weren't developed economies in 1970.

[00:08:42] Now if you look at those countries that have been developed for the last half century plus it gives you 53 years of data on 14 countries.

[00:08:54] The US has had three bouts of inflation above 8% in the last 53 years.

[00:09:02] The 14 countries have had 31 episodes you can't draw many useful conclusions out of three episodes you can when you're looking at 31 episodes.

[00:09:14] It was interesting is out of those 31 episodes four of them receded below 2% meaning that the inflation got back under control within two years four out of 31.

[00:09:28] The length of time it took for inflation to get under control ranged from 15 months as the fastest to 26 years as the slowest with a median of 13 years.

[00:09:41] Now that is wildly at odds with the consensus expectation. I think this notion that the last mile has been discussed at length by the Fed by pundits and prognosticators by Wall Street strategists.

[00:10:01] I think last mile is a euphemism that replaces transitory and has roughly the same meaning.

[00:10:09] Last mile presumes that you go from three and a half to three to two and a half to two and then settle in there and yes, that could happen.

[00:10:20] But it ignores the possibility. Here's a simple experiment if you take today's inflation and if you assume that inflation for the rest of the year is zero obvious every single month, okay, that'll take you to point seven. That's great.

[00:10:45] If you assume every single month matches the trailing three year average for inflation, which is just under half a percent of year.

[00:10:56] You finish the year at 5.6 now what's the market reaction going to be if we finish the year at 5.6 again I'm not suggesting that as my central expectation. I am suggesting that

[00:11:11] a central expectation that will be three is sure a little less at the end of the year is has asymmetric risk to the upside it's a lot more likely to be above.

[00:11:27] Let's say three and a half then below two and most people are anchoring on below three as the central expectation so there's opportunities.

[00:11:39] We will put a link to that paper in the show notes, but do you remember the title of that for those that are listening that might not be right for their computer.

[00:11:48] The most recent one was I believe was don't break out the champagne yet.

[00:12:00] Yeah, well and I think on the inflation side of it it's like thinking up to an energy prices is just a comment you know it's not like those prices went up and they stayed up.

[00:12:12] It's not like you know they came down so what people are spending at the grocery store and you know and with energy prices it's it's like the inflation got basically an acid floor.

[00:12:23] Inflation is receding it's not it's not reversing and it's actually stopped receding since last June.

[00:12:31] Inflate rolling 12 month inflation right now is higher than it was last June really pretty interesting.

[00:12:38] What do you suggest for people I'm thinking of like the 60 40 stock one portfolio you know it was a big wake up call in 2022 when the 60 40 had one of its first years ever then in 2023 it kind of came back.

[00:12:51] Nicely and I think those investors that were looking to diversify more sort of like almost that good performance in 2023 it's sort of got their mind awesome other asset classes that could be helpful so what do you think in terms of.

[00:13:10] You know other things that investors should be considering here.

[00:13:15] Well firstly you asked the question what markets are unusually expensive are priced for perfection and obvious example would be AI more broadly the tech sector.

[00:13:35] And then you ask the question what's abnormally cheap well the US stock market is at a Schiller PE ratio of 34 that means price relative to 10 year average earnings rather than trailing 12 month or.

[00:13:49] Price to forward earnings price to forward I view as being equivalent to price to fantasy because forward earnings have it happened yet and they might not.

[00:14:01] But if you look at price to 10 year smooth average earnings you're looking at price to sustainable long term earnings in the US is at 34 times okay when's it been higher than that has been higher it's been about the same as that.

[00:14:19] Briefly in 1929 it's been distinctly higher than that about 30% higher than that of the key to the tech bubble and it's been.

[00:14:32] That high no other times in history so twice in the last century that's a little alarming now okay broadener horizons let's look to develop economies around the world outside the US they're at 18 times that's a 40% discount that's very cool except wait a minute developed outside the US that's European Japan and we know they have problems you used to have.

[00:15:02] Euraschoolerosis now it's Japan's chlorosis and in both cases the demographics are not exactly conducive to rapid economic growth so gosh shouldn't they be trading cheaper than the US absolutely should be the they'd be trading at a 40% discount I don't think so.

[00:15:22] The dividend yield on a broad EF a portfolio is north of 3% in the US it's about 1.3% it's tiny and so when you look at relative valuations non-US is much cheaper.

[00:15:38] The word is that when the US catches cold non-US markets get the flu all right that historically is true but the history where that was true was markets where they weren't already trading at a deep discount relative to the US.

[00:15:58] There's also emerging markets emerging markets are trading at 14 times their 10-year average earnings and the emerging markets are as stretched in terms of the growth versus value picture as the US is meaning that emerging markets value is is at less than 10 times earnings fundamental index which we introduced back in 2004.

[00:16:24] Has a stark value tilt it's best compared with value indexes and it's trading at a show or P ratio of eight and it includes the growth stocks it just rewrites them down to their economic footprint you can buy half the world's GDP for 8 times earnings.

[00:16:42] That's pretty cool and so I look at that and I think there's some interesting places to invest I've been called a permabare I'm a bear bears the wrong word I'm cautious on markets that are very fully priced and actually disinterested in those markets because they're long to us they would return isn't pretty good.

[00:17:04] I'm not a bear on things that are cheap I love things that are cheap.

[00:17:09] I want to ask you about your you guys have a great tool and research affiliates it looks at expected returns over I believe like a 7 to 10 year time frame it's one of the places I like to go when I try to get context of where we are can use to what that tools telling us right now I mean I assume it's probably a little more constructive on bonds than it was years ago but probably not not that pretty on stocks but what is that tool telling us right now the tool is our cast set allocation interactive.

[00:17:33] You can use a new website and rather than giving the link I'll just say if you use Google and type in asset allocation interactive the very first thing that pops up that's not an ad for somebody else is that tool.

[00:17:48] So asset allocation interactive it gives you forward looking long term returns for 130 different asset classes around the world and we're doing.

[00:18:04] Enhancedments to it that will add 18 additional countries before the year is out that will add value versus grow for each of these markets and so it's an exciting tool now.

[00:18:18] Now.

[00:18:21] Forecasting who wasn't who said forecasting is difficult especially about the future it is difficult if somebody asked me what do you think the stock market will return what do you think the S&P return will be over the next 12 months I haven't a clue anyone who says that they have a strong reason to believe it'll be X percent is probably a.

[00:18:45] Charlotte turn but 10 year returns are actually not that difficult you start with the yield then you ask how much will that income grow over the next 10 years.

[00:18:59] For bonds are fixed income it'll grow by zero for high yield bonds it'll shrink because there are defaults for stocks it'll grow with inflation.

[00:19:12] Plus a little bit of real growth over the long term history the real growth in dividends and earnings all over the world has tended to be about 1.5% per annum over very long periods of time so.

[00:19:28] Long term return for US stocks 1.3 from the yield 1 and a half from real growth in earnings and dividends that gives you a 2.8 real return add in to an a half for inflation and you're around 5.3.

[00:19:43] So 5.3% 10 year returns seems perfectly reasonable but there's a third component of return you've got yield you've got growth and you've got changes in valuation multiples.

[00:19:56] That sure P ratio of 34 is abnormally high historic norm is 18 we believe that in a more mature economy like ours with an aging demographics something in the neighborhood of 23 is more likely to be a normal so 34 to 23 has a big spread.

[00:20:18] Now maybe it's a new normal 34 is exactly where it ought to be or maybe it needs to meet her. Let's split the difference let's take it halfway so you're taking it down to 28 that's going to cost you on about 2% per year compounded over the coming decade and that in turn would take you down to the 3.5 range so anyway all asset asset allocation interactive.

[00:20:48] Currently forecast 4% because it makes a gentle nod in the direction of revaluation now I said I'm not a bear when things are cheap the same logic identically the same logic for heath it gives us an 8.5% return.

[00:21:06] For emerging markets gives us a 9.5% return 9.5% a year for the coming decades a wonderful rate of return.

[00:21:14] Now there's also value the spread and valuation between growth and values abnormally large it's in its most extreme death style ever in some cases in the most extreme stream two or three percent tiles per.

[00:21:31] So just snapping back to historic norms where the spread is in line with historic norms would for the US mean that value be it's growth by somewhere between 7.000 and 10.000 basis points.

[00:21:46] Spread that over 10 years that's 700 to 1000 basis points per year.

[00:21:53] Chop that in half because it might or might not have.

[00:21:57] That gives you 3.5 to 5% a year now if you're starting point is US at 3.5 or 4 and yet let's say 4 for value now you're at 8% are people going to be ecstatic with 8%.

[00:22:13] No they'll be pleased.

[00:22:17] Take EFA 8.5% add for now you're 12.5. Wow, merging markets your 13.5.

[00:22:25] So value outside the US represents really an extraordinary opportunity right now I think the growth value cycle seems to have been dead for a long time value rolled over in 2007.

[00:22:41] Creatored during the global financial crisis snapped back big in the aftermath and then drifted slowly worse and worse and worse until 2018 when it fell off the proverbial cliff and you had an outright value crash from 2018 to mid 2020.

[00:23:01] We haven't come back all that far from that point Russell value from mid 2007 and in 2020 under performed the Russell 1000 index by 37 percentage points 37.00 basis points it's still 3300 basis points behind.

[00:23:22] So the value is at the moment incredibly cheap but the value growth cycle does exist and you do have when values really cheap.

[00:23:35] You have a tailwind for value because mean reversion can augment the already existing value effect if you have elevated inflation risk value computer a well so and if you have a softening economy value can protect the downside.

[00:23:51] So I see lots of really interesting opportunities to date they just aren't where everybody's pouring their money.

[00:23:57] Do you think there's anything to this idea that this mean reversion process is slowed down and that fundamental value investors have the length in their time frame you know people talk about all these other things that are causing flows in the market these days they're not fundamental related.

[00:24:09] Do you think there's anything to that like we're going to have to sit through longer periods of struggles to be fundamental value investors.

[00:24:15] I think there's there's definitely truth in that I think one of the main forces at work here is the move towards indexation which is a very powerful force in the markets these days.

[00:24:31] If money goes from non indexed portfolios to index portfolios much of the money doesn't move because the index owns the market but it really does it owns most of the market so what happens is that the non members in the index gets sold and the members in the index get bought.

[00:24:58] So every hundred dollars that goes into an index fund about eighty dollars stays more or less where it is because the index spans 80% of the market and about 20% of the portfolio moves from non members to members well that pushes up the relative valuation for members.

[00:25:18] And we saw that just two weeks ago S&P made a a blunder they reported a list of holdings for a narrow niche dividend index that they maintain that didn't include one of the stocks that's supposed to be in the index.

[00:25:42] And so the index trackers for that strategy and these aren't big index trackers because it's a niche index actually it was the other way around added that they added a stock to the list that wasn't supposed to be there and then a couple days later they realize their mistake and they took it out well the stock went from 84 to 90 and then back down to 84.

[00:26:07] That's a seven plus percent move for a little niche index strictly because the stock was added and then dropped and that's a big turnaround.

[00:26:19] So the spread and valuation for members versus non members of the big index is like the S&P in the Russell is much bigger by our measurements in the 30 to 50% range membership has its privileges.

[00:26:33] You're worth more if you remember and that in turn with the flow of money into index funds can push the valuation of these companies up higher valuation means lower future long term returns your more front end loading the return in the run up in price.

[00:26:53] And so none members ought to have higher long term forward returns the members of the index but that's more than offset by the flow of money into indexes as long as that flow is enough to push the prices higher and higher and higher.

[00:27:13] So that's a long winded way of saying this value cycle this anti-value cycle from 2007 to 2020 was undoubtedly augmented by the flow of money into index funds and augmented in a big way at some stage that pressure.

[00:27:30] If the flow of money isn't enough to keep a stock well above the level it would be as a non member then there's downward mean reversion pressure plus the upward pressure of the cheaper stocks being priced to give you a higher forward looking return and value has a chance to come back 2000.

[00:27:56] The growth value spread became the largest in history.

[00:28:02] A spread that was exceeded in 2020 briefly.

[00:28:07] But value came back big between 2000 and 2007 value outperform growth by over 10,000 basis points over a hundred percentage points in seven years just to pin this performance it's not impossible that we could see a similar rebound.

[00:28:26] In value relative to growth in the years ahead.

[00:28:30] I'm not predicting that i'm just saying that's an outlier possibility that could happen the opposite possibility of value underperforming by 10,000 basis points strikes is implausible in the extreme.

[00:28:46] How do you think about timing with a factor like value you know we know this can be an opportunity you know for long term investors would value cheap like this but we also know you're going to sit through a lot of pain a lot of time to get there so.

[00:28:57] What do you think like how do you think investors should think about this idea of timing and the ability to add to a factor like value in his cheap.

[00:29:03] Well most people think of it in terms of oh gosh these companies are value stocks for a reason they the crummy companies with crummy prospects and major headwinds and that's all true and that's why they're cheap.

[00:29:18] Being a contrarian investor means buying what's out of favor what's unloved and buying what's unloved is inherently uncomfortable and has a very special additional bit of discomfort if you buy what's out of favor and you're right.

[00:29:40] You won't necessarily be right right away the chances of you buying at the exact bottle are slim to none which means you will look and feel like an idiot until the turn comes.

[00:29:53] And that means that it's much more uncomfortable than trend following strategies if you are a trend follower and you buy what's beloved everybody loves it.

[00:30:06] And so if it goes against you you think I got a lot of company it's too bad it's going against me but i'm.

[00:30:17] To use today's parlance i'm i'm gonna hold hold on for dear life i'm i'm gonna hang in there i'm gonna buy the dips.

[00:30:29] Buying the dips on something that's unloved out of favor and dirt cheap is not a comfortable thing to do but it's the only way you can assure that you have peak exposure at the bottom when it does eventually turn.

[00:30:43] You also have to be aware value traps and we've published some interesting papers on value traps recently on value traps are things that look cheap on their way to zero.

[00:30:53] I want to ask you more about that paper because that was actually on my list of questions i read that paper in preparation for the interview and you guys had some interesting thoughts on this idea value traps you know it's kind of like this holy grail of you besting you know people think if I could just avoid all the value traps then i've got this great return but it turns out a lot harder to do that in practice than it is in theory so what are your ideas on potentially avoiding value traps are at least trying to limit them.

[00:31:15] Well we actually do work with pymco in a product area called r a research affiliate equity and r a is built on the same foundation as fundamental index that is to say you wait the stocks according to how big their business is not how popular.

[00:31:34] Bluved and expensive they are so growth stocks if you want to own them in or a you're going to say i'm reweighting this down to its economic footprint value stocks you'll reweight up to their economic footprint because these reprice to the premium these reprice to the discount so you build in this dark value tilt by DM sizing growth and emphasizing value.

[00:31:56] Now the Achilles heel of fundamental index it's a wonderful strategy it has worked brilliantly since we launched it 20 years ago measured correctly measured against let's say a pharma French.

[00:32:13] Ritz risk and style adjusted alpha it's been relentless measured against cap weighted value indexes it's been relentless but it buys every value trap that comes along and it'll keep buying it all the way to zero.

[00:32:28] So in working on the RE strategy we thought.

[00:32:35] Do we really want to own the value traps well what can you do.

[00:32:40] Value traps have two common denominators measures of quality or lousy.

[00:32:46] And or they're in free fall.

[00:32:50] Free fall you can measure using simple momentum metrics quality you can measure using things like dead equity ratios and metrics of the aggressiveness of the accounting are cruelled that's.

[00:33:05] Sort of thing.

[00:33:08] If you filter out let's say the 20% of the stocks that score the worst on momentum and that score the worst on quality I would jokingly say you will miss a three out of 30 out of every 10 value traps.

[00:33:32] You'll miss some dirt cheap companies that subsequently bounce back big and you'll also miss the value traps so one of the really fun fact doids is that strategy has been live since year and 2000.

[00:33:46] 2004 so it's approaching its 20th anniversary.

[00:33:51] In that entire history it's owned over 1000 US companies not all at once of course and over that entire history it's had zero go bust.

[00:34:02] Well, that's something to have a deep value strategy that's never had a bankruptcy is pretty cool so yes you can eliminate the value traps you will also eliminate large numbers of cheap stocks.

[00:34:19] So we find that that strategy makes for more comfortable ride everywhere because you won't have these embarrassing value traps in the portfolio.

[00:34:35] But it actually adds value historically in the less efficient markets and the more efficient markets US large international it doesn't improve the returns but it sure makes the ride more comfortable in the less efficient markets small stocks emerging markets.

[00:34:56] It also adds value it does better than fundamental index so there's some interesting ideas out there on ways to deal with value traps I haven't seen the risk and return statistics on it but I'm just curious and interested in your thoughts I would imagine.

[00:35:12] Probably much lower drawdowns which the reason that that is important is because for investors a lot of the bad decisions come when strategies or markets are the bear market type territories and I think controlling those drawdowns can be important for investors actually realizing the returns the strategy like this.

[00:35:35] Right now cautionary note on drawdowns the biggest drawdown of the last quarter century was the global financial crisis and in that drawdown was more savage for value than for growth.

[00:35:49] More savage because the financial leverage of the value companies was greater and so the impact of the financial crisis on value was worse than for growth likewise the covid meltdown was hard around value then on growth.

[00:36:07] So the magnitude of the drawdowns isn't necessarily smaller the drawdowns relative to market indexes or more particularly a cap weighted value indexes those drawdowns relative to those indexes tend to be modest relative to the long term gains.

[00:36:31] I want to shift and talk about what is probably the furthest thing possible from value which is a bubble is an AI.

[00:36:37] You've done a lot of work over your career about bubbles about how they work you know what goes into them and you know we have a lot of people talking about right now AI potentially being a bubble so I'm wondering if you apply the criteria you've looked at like what do you think about AI right now relative to being a bubble.

[00:36:52] Sure Brad Cornell formerly of UCLA and Caltech that they're working with as well to motor on coin the expression of big market delusion a big market delusion is a special kind of bubble it's a macro bubble that's in a particular industry that's new.

[00:37:18] So the big market delusion the internet back in the late 90s electric vehicles back in 2020 and 21 AI today what is a big market delusion like I said at the beginning prices are set based on narratives that the narrative for the dot com bubble was Internet's going to change everything it'll change how you buy and so good.

[00:37:48] And services it'll change how you communicate how you transmit information around the world how you socially interact how you get your news how you do your research it's going to change everything true it did.

[00:38:04] The narrative went on to say these are the dominant players in internet and dot com world and they will be the dominant players in 10 years not quite as true this ruptures get disrupted.

[00:38:20] The other part of the narrative that was not true was this is all going to happen really fast how much did our reliance on the internet change between 2000 2005 a lot.

[00:38:38] 2005 to 10 not a lot 10 to 15 not a lot 2000 to 2015 oh my goodness yes it's a cumulative effect it takes time so the red and I wrote a paper suggesting that electric vehicles were a big bubble excuse me a big market delusion back in early 2021 there were nine electric.

[00:39:08] Vehicle specials companies that made only electric vehicles and Tesla at the time was priced at 24 times sales.

[00:39:18] That's a huge vulnerable forget price earnings ratios 24 times sales and at the Tesla was the second cheapest of the nine companies on price to sales ratio one was even priced in north of 10 thousand times annual sales.

[00:39:38] Because it sales were near zero and so.

[00:39:43] We said this looks like big market delusion again not that he is not that there's anything wrong with the use not that they are a big market that's on its way that's going to change the world.

[00:39:59] But that the notion that these disruptors weren't going to subsequently be disrupted themselves was a little naive and the notion that this was going to happen overnight was a lot naive.

[00:40:14] And so the aftermath of that is that the nine companies per all of them underperform the S&P.

[00:40:23] And they underperformed by anywhere from 15% to 99%.

[00:40:30] In almost exact proportion to their starting price to sales ratio.

[00:40:36] Now we in September came up with another paper in which we suggested that the Nvidia AI the paper was in video and AI.

[00:40:52] Revolution excuse me breakthrough or bubble.

[00:40:56] And our conclusion was it's both as is often the case to break through for sure it's going to change the world as we know it big time.

[00:41:09] But the narrative is these are the players that are dominant in this industry they'll still be dominant in 10 years.

[00:41:19] The disruptors will get disrupted and this change is going to come shockingly fast.

[00:41:25] So the parts of that narrative that are dubious are the rate of change of adoption of AI and the possibility of disruptors getting disrupted.

[00:41:40] So as we look as we look at this whole landscape is AI going to change everything yes it will.

[00:41:49] It's going to just place millions of jobs.

[00:41:54] You guys aren't needed I'm not needed if AI takes over our work.

[00:42:00] But I did tell our all hands meeting about a year ago.

[00:42:06] I spoke with the whole company and I said not a single person here is at risk of losing your jobs to AI.

[00:42:15] You say there's no risk at all you might lose your job to somebody who knows how to use AI better than you do so get on it learn how to use it and learn how to have it leverage your time and efforts.

[00:42:29] And it was an interesting reaction but AI is going to change everything.

[00:42:37] A couple of interesting examples early days of chat GPT I test it to write a children's bedtime story with brave nights and unicorns in a row, a little 500 word bedtime story that in the child's book author would have been proud to put their name on.

[00:42:57] It was beautiful it was sweet it was all again and it captivated the attention this was done in the space of about five seconds by a computer cool I then asked it to do short bio of me.

[00:43:14] And I didn't know that I graduated with an MBA from University of Chicago and started my career at Goldman Sachs but I guess I must care.

[00:43:22] So it's good makes things up.

[00:43:28] I was recently reflecting back on Sunsou's famous book The Art of War and I hadn't read it in decades and so I asked chat GPT write me a summary.

[00:43:42] 2000 words or less of Sunsou's art of war and it wrote less than a thousand words and it was six inked and it was thorough and it mirrored perfectly mirrored my recollections.

[00:43:58] It was so good that I circulated it to my whole management team said I know you haven't most of you haven't had time to read art of war but it's useful.

[00:44:11] Here's a synopsis that I think is just brilliant and it was written by chat GPT.

[00:44:17] How do you think in this kind of gets back to what you're talking about with researchers fill it's like how do you think AI changes our business like if we look out 10 years in the future and do you think we need there's going to be way less analysts on Wall Street for example because one analyst is going to be able to do the job of ten or.

[00:44:30] If you thought about like well how it changes the investment management industry.

[00:44:34] Well certainly in me Wall Street analyst who isn't doing a deep dive into how can this leverage my time is immediate.

[00:44:46] I'm and it will people in the programming community talk about being over employed now this is more urban legend than reality but it does exist where somebody has a full time job as a programmer works from home and realizes okay I'm working from home.

[00:45:06] I can get another full time job how cool is that I can get two salaries and I'm going to use AI to do a rough draft of my computer programs and all then.

[00:45:19] That it make sure works and do the final edits and oh my goodness this is going so smoothly I can get a third job in a fourth job I haven't heard of anyone with more than four but you know if you're a programmer making whatever programmers make these days earning four times that means you are considerably more than the head of IT at your company.

[00:45:48] And no one has to know that you have four jobs alright like I said that's more urban legend than reality but it does exist and that's a beautiful example of leveraging your time financial analysts can do the same.

[00:46:04] I would note that that people like working with people so I was speaking with one of our largest clients that has a platform of mutual funds I shouldn't say who because I didn't clear this with them in advance but they have a they have a model portfolio and you can opt to put a million bucks into the model put for you.

[00:46:34] Or you can opt to have a financial advisor tell you where to invest but the and the model portfolio costs half as much and doesn't give you a conversation when markets are going haywire and but it performs better it performed better during the financial crisis it performed better during the code crisis.

[00:46:59] And people were redeeming out of the model portfolio in the financial crisis and the covid aftermath but they weren't redeeming out of the financial advisors who were performing less well which is really interested people like talking to people so I don't see a.

[00:47:21] Replacing people I see a at least not in the next decade I see AI leveraging our capabilities now there's certain areas if you're talking graphic arts i can do amazing graphic arts.

[00:47:41] If you're a graphic artist and you can increase your productivity tenfold easily using AI and by the same token as autonomous vehicles get past their hiccup be early learning stage the notion of okay I got to get to work tap on the iPhone whatever's the new version of.

[00:48:11] Auto pilot Uber what kind of card do you want to take you some compact on up to luxury click what you want it arrives in two minutes it was you to work eventually people are banned from the roads because they're too dangerous and when that happens you don't even need traffic lights cars can go right past each other at full speed through intersections.

[00:48:39] And forget about 55 speed limit these things will be perfectly safe at 90 so you're going to wind up with autonomous vehicles that just revolutionize transportation i was in an Uber about six months ago I said are you ready for autonomous vehicles.

[00:49:06] And the guy said what do you mean I said cars drive themselves and he said well i wouldn't use those said but your customers would.

[00:49:15] And those cars don't need a drive they said when's this going to happen i said five or ten years he said i'm going to be doing something else by then.

[00:49:25] He which is the great attitude to have about all this I did ask chat cheaply so I posed a question who's rob are not and what are his greatest accomplishments and I pick removes the goldman stuff and the stuff that was wrong i think it's a pretty accurate.

[00:49:42] I'm sorry you but what I find interesting and what i'd like you to comment on is it says that you've published over a hundred academic papers i don't know if that number is accurate about a hundred to be okay so and what where I want to go on that is like research affiliate affiliates and such a deep.

[00:50:04] Research emphasis and culture and you thank you and just talk about the importance of that for the for you and the firm and sort of what your belief system is over there.

[00:50:21] That's that's a really nuanced question i'm that AI is not new I was doing neural networks back in the eighties didn't work very well so I didn't do much time spend a lot of time on it but.

[00:50:35] But AI has been around for a long time high frequency traders almost all hft work these days is done with AI and has been for years so it's a computer matched against a computer doing that frequency trading.

[00:50:51] AI is massively data dependent if you have thousands of samples of data AI is useless.

[00:51:05] If you have millions hey I starts to be useful if you have billions or trillions now you're talking that's where AI can run circles around humans.

[00:51:17] hft is heavily rely on AI because you've got billions of samples of data and so when the bid aspect is here the computer can say I think the next.

[00:51:31] Tick is going to be on the high side so if somebody wants to sell i'll come inside the bid aspect and i'll buy it for a little more because I think the next tick is going to be up and it's right just often enough in to create a robust alpha engine.

[00:51:50] And it's when when if i'm the seller somebody's giving me a better bid than I would have had without the hft people but that's short term long term there's too few samples and so that's where classic scientific method comes into play and I say classic scientific method because.

[00:52:13] Scientific method is much distorted in the world Anthony Fauci said I am the science actually said that i'm.

[00:52:28] No science is about starting with a hypothesis and using data to test it.

[00:52:39] And very important try to prove yourself wrong one of the most damning critiques in science is that your ideas confulsifiable you can't prove it wrong if you can't prove it wrong it's useless hypothesis.

[00:52:58] So when it comes to longer term investing over months years decades that's where you have few enough data samples that thinking about the way the world ought to work starts to come into play and you can ask the question should value win well you're buying out of favor companies that are on love any notion of risk premium should give you a reward for that.

[00:53:28] So the word will be positive in financial terms and negative income for terms.

[00:53:35] And so then there's the question of relative valuation is the spread and value between growth and value abnormally large if it is then.

[00:53:47] So that's the sense that any meter version should help you.

[00:53:52] And so then you use empirical data to test is my hypothesis correct and can I find chinks in that armor can I disprove my own hypothesis.

[00:54:04] That's not the way a lot of the quality community works a lot of the quality community dives into the data and says what relationships can I find in the data.

[00:54:13] So millions or trillions of samples and you're doing high frequency trading that's fine.

[00:54:19] That's the right use of the data if you've got billions of samples if you're dealing with factors, factor investing for instance where you're dealing with typically thousands of samples of data then that's the wrong use of science you should start with the hypothesis and tested don't just mind the data and whatever pops out of the data.

[00:54:43] Say oh, this worked historically, therefore will work in the future that circularity has been deadly for the quant community.

[00:54:53] So in closing we've since we've had you on you can't answer standard closing questions so we wanted to sort of end with what worries you the most about the future but also on the flip side what makes you the most optimistic.

[00:55:08] Oh.

[00:55:09] Oh.

[00:55:12] Firstly, I tend not to worry much.

[00:55:15] I pushed to name what worries me most it's the existential threat of humankind doing something really stupid that derails humankind is possible.

[00:55:32] But why why waste time thinking about that because if it happens it happens as it doesn't happen you wasted your time thinking about it either way it was a waste of thought.

[00:55:42] So existential threats would be my biggest worry but I don't sweat it because doesn't matter.

[00:55:49] There's so much good news in the world.

[00:55:56] The world is getting to be a better and better and better place slowly gradually but fast enough that intergenerational it shows up we live longer than ever before we live healthier than ever before.

[00:56:09] The infant mortality is the lowest it's ever been in history.

[00:56:16] The risk of dying a violent death is an order of magnitude less than it was 100 years ago and a hundred years ago was the best it had ever been.

[00:56:25] And it's an order of magnitude less now that's cool and that's worldwide.

[00:56:30] So as long as humankind doesn't do something incredibly stupid I have enormous optimism that the intellectual curiosity of mankind and the willingness to think outside the box.

[00:56:47] Well most people don't think outside the box but enough do to move things forward and that's fun.

[00:56:55] There's a wonderful book called Ten Global Trends that every smart person should know it's kind of a cutesy title but it goes through trend after trend after trend in the world and it's almost all good news.

[00:57:10] We're worried about deforestation not true there's more forest land today than there was 20 years ago which is more than it was 50 years ago which is more than it was 100 years ago urbanization is allowing for us to grow.

[00:57:26] So there's all sorts of wonderful things happening if you're patient unfortunately politicians are in votes and media earns eyeballs by promoting fear and anger.

[00:57:37] And not spending time on the good news which is ampled.

[00:57:43] It's stuff I just got to ask before we let you go here what is that cool painting behind you.

[00:57:49] Oh that's a that's a litho from Warhol of his endangered species series let's see there you can get very cool so it's yeah it's it's a cool piece it's um I love it.

[00:58:05] Nice well thank you very much Rob we really appreciate it.

[00:58:10] All righty take care.

[00:58:13] This is Justin again thanks so much for tuning into this episode of excess returns you can follow Jack on Twitter at app practical quant and follow me on Twitter at J.J. Carbano.

[00:58:23] If you found this discussion interesting and valuable please subscribe in either iTunes or on YouTube for leave a review or a comment we appreciate Justin Carbano and Jack forehand are principles at the Lydia Capital Management.

[00:58:35] The opinions expressed in this podcast do not necessarily reflect the opinions of the capital no information on this podcast should be construed as investment advice.

[00:58:42] Securities discussed in podcast may be holding stuff clients of a little.