Our conversation with Srikanth (Sri) Iyer, Managing Director, and Head of I3 (i-Cubed) Investments, at Guardian Capital.
From a head start in 2010 to today, Iyer talks about why factor and Smart Beta investing have not been working well, and boils down and shares 10 years of his group's discoveries, insights, and innovation in systematizing the growth, payout and sustainability of two classical dividend equity investment strategies:
1) Asset preservation through dividend growth.
2) Asset accumulation through earnings growth.
This podcast is also available on Apple Podcasts, Google Podcasts, Google Play, Spotify, Soundcloud or wherever you get your favourite podcasts.
Transcript:
Pierre Daillie: [00:00:24] Hello, and welcome to the Insight is Capital Podcast. I'm Pierre Daillie, Managing Editor of AdvisorAnalyst.com.
If you're at all interested in what's happening at the intersection of investing, and the innovation of artificial intelligence, it's pretty safe to say that you may be blown away by the inroads that are being made by Guardian Capital's I3 Investments unit.
Our special guest is Sri Iyer, Managing Director, and Head of I-Cubed Investments at Guardian Capital.
Sri joined Guardian Capital in 2001 to help lead the development and implementation of Guardian's proprietary systematic strategies. His career in the financial services industry began in 1995, when prior to assuming his role at Guardian, he joined Global Value Investors in Princeton, New Jersey, and was responsible for a variety of portfolio management and financial engineering roles.
Sri graduated from the University of Bombay with a B.Comm and earned the Chartered Cost and Works Accountant designation.
He received an MBA in Applied Finance from Rutgers Graduate School of Management. Without further ado, my conversation with Sri Iyer.
Good morning.
Sri Iyer: [00:01:29] Good morning. How are you?
Pierre Daillie: [00:01:31] Very well Sri, how are you?
Sri Iyer: [00:01:32] Pretty good. Can't complain.
Pierre Daillie: [00:01:34] It's good to see you. It's been awhile. I think we spoke  last year, right?
Sri Iyer: [00:01:37] Yeah. It was before for the things changed. That's for sure.
Pierre Daillie: [00:01:41] Yeah. How have you been?
Sri Iyer: [00:01:42] Very good. Actually, I think it's been this whole, I call it the great isolation has actually helped me more than hurt me because, obviously, we have been more into artificial intelligence machine learning, So though the world was coming to us. So for us, it was, almost a natural transition. So can't complain.
We're quite optimistic about the future to be honest. So , probably a very contrarian view today in this conversation, but I think I'm more optimistic than I've ever been.
Pierre Daillie: Yeah, I think for starters, being in the financial industry has been very fortunate during this time. Not only is the business sort of conducive to, being able to work remotely, especially with the tools that we have today with technology, I think when this whole thing first started, it was very unnerving. It was scary.
Sri Iyer: It sure was.
Pierre Daillie: [00:02:33] And then, you know, in a very short space of time, there was this incredible response from government and from central banks, to shore things up.  Oddly enough, I only realized, You know, maybe a month ago, I guess, because it was lost in the news, that quantitative easing was still going on at the same time concurrently as all of the fiscal stimulus, all of the, whether it's the fed stimulus response or the government's fiscal response. concurrently, quantitative easing is still running underneath it...
Sri Iyer: [00:03:05] And not just US, globally.
Pierre Daillie: [00:03:07] Exactly. yeah. what is the figure of 30% of global GDP?
Sri Iyer: [00:03:11] Yeah, I would say so. 30%. Yeah.
Pierre Daillie: [00:03:14] Yeah. To say unprecedented is an understatement. It completely eclipses anything that's ever been done or happened before.
Sri Iyer: [00:03:22] It's also put a lot of the myths away in the sense that the paradigm shift is happening.
I don't think so. In the realm that we we're just discussing. I think a paradigm shift, is happening in the way, capital is reallocated and how capitalism itself has shifted and it's proving itself extremely resilient. So the whole concept of capitalism has metastasized now.
So the COVID crisis is significantly more systemic than anybody can imagine. But we have been through the financial crisis. We've been through the tech bubble. Now we're going through what we call a consumer crisis. We have a tendency to systemically fear everything when something goes bad.
And so during the tech bubble, we felt everything was going to be bad, but actually a lot of other sectors did okay. During the financial bubble, we thought everything's going to go bad, but actually technology did okay. This cycle the consumer, industry as well, the old economy kind of industries that were in the cliff or precipitously waiting to end, just got accelerated so it feels a lot more than it really is.
But if I had said that this would have happened in the next two, three, three years anyways, would it feel as bad as it felt like, or that happened in three months. As Satya Nadella, the CEO of Microsoft said what we couldn't do in five years in technology, COVID did in three months. So his connotations are very profound in the sense that he's not talking about social evolution per se, talking about technology evolution.
And so you're seeing big data, AI and other components evolve significantly. But you needed an accelerator. So it's like pouring gasoline on it on a campfire. That's what happened is I just went and now the question is, how does this continue and how does this move forward in that has huge ramifications to the investment industry.
No doubt about it. Me and you and how we're adapting to the new normal. And this new normal looks very different than the old new normal, whether we call the tech bubble, the new normal, a financial crisis, the new normal. This seems to have a touched the social evolution ladder in many ways. there is a lot more, what I call thematic resonance to this systemic crisis then there has ever been in the past crises. So that resonance that we're seeing now in a global scale is not, it's not a small thing. So I think it'll be very interesting to see how the markets, progress forward. So we're using the traditional lenses to interpret market behavior and it's true: free money always makes things go up. You're think re seeing the market very narrowed into breadth?
Pierre Daillie: [00:05:59] Yeah.
Sri Iyer: [00:06:00] Everybody owns Tesla right now. so proof of concept, there's no longer a risk measure anymore concept is where you put all your money into the, I guess. The world has changed.
Pierre Daillie: [00:06:10] Absolutely.
Sri Iyer: [00:06:11] A lot more to it than just Tesla, to be honest.
Pierre Daillie: [00:06:14] For sure. For sure. What a catalyst though. I mean, it has accelerated not only the use or the adoption of technology, but it has accelerated social unrest; it has accelerated all of the inequalities; it has accelerated, economic reform. I recall, and it wasn't that long ago. It was just, this is, seven months ago, Andrew Yang, for example, talking about, I keep bringing it up in our conversations, but it just blows my mind that, when Andrew Yang was talking about universal income as an idea, it was so far-fetched, to so many, that's something like that could actually happen and then a month later, boom, there, it was right.
Sri Iyer: [00:06:51] So I think it's a question of, how do you redistribute wealth, if it indeed needs to be redistributed, I'm not passing judgment here, but if it needs to be, how does it get distributed and who distributed this? There's one camp that said that the system is better than people.
So we will decide how wealth is distributed while the other side says that the human nature and capitalism and innovation will trickle down and eventually have everything distribute itself. Both have proven to be wrong on the tails, but if we can come to something in the middle, I perhaps, things could go well, but as you and me, both of them, there's no such thing as middle today.
Unfortunately everybody's either extreme, left or extreme, right? The middle is the silent majority in many cases. So I think the world badly needs middle. That's the way I look at it.
Pierre Daillie: [00:07:43] Sri, welcome to the show. It's really great to have you. Yeah, we are rolling. I find, just before we get started, I find, having formal starts, can be a little awkward sometimes especially for me, I'm speaking selfishly, but so anyways, welcome to the show. It's really terrific to have you.
Sri Iyer: [00:08:05] You've always been very disarming. You always... I only met you once, but every time you have introduced and you have had a conversation, last time, I still remember the conversation; it's been fantastic and definitely slightly a different cut than most people I've spoken to in your side of the industry. So I look forward to our conversation today.
Pierre Daillie: [00:08:23] Thank you. Those are very kind words. Sri you're the head of a systematic investing strategies.
iCubed is the term that you've rebranded and, before we get started. I can't wait to talk to you because I think the subject matter of what you do every day is very exciting. I think it's bringing potentially a different kind of magic to the business.
I think a lot of the academics in the passive side of the business have done a, a really effective job of flattening the excitement, the excitement of the financial business, the macro, the stock picking, and I think what you're bringing, what you and your group are bringing, in terms of your efforts, it's very exciting because it really takes what people have been trying to do for all of history, and seeks to make it more efficient, less subjective, and, in that respect, it's a dimension of investing that really is a game changer. And of course it's the technology that we're adopting, that is emerging, in artificial intelligence, that's making a lot of these new developments possible.
It wouldn't have been possible just 10 years ago. The, maybe that's when things were kicked off or started to kick off in earnest. But before 10 years, really, we didn't have the bandwidth, the processing power, the ability to do a lot of this stuff. A lot of this work just wasn't possible.
I don't know if you agree with that or not, but so straight before we get started, I think, take us back a little bit. Please tell us a little bit about your background, how you got here from your beginnings.
Sri Iyer: [00:10:06] My background is actually in accounting. I'm a certified cost accountant and chartered accountant.
So I was into numbers all my life. This was in India and so I was like one of the youngest, charter cost accountants in India. So it's definitely not a job that's for sure. And then I moved to the United States when I was 19 years old. I did my new undergraduate and then I went from a master's degree, plant statistics and finance, and subsequent to that, I did my masters in business administration for Rutgers university in New Jersey.
And then I went straight into, asset management and research, in a firm in Princeton, New Jersey, which was basically a statistical arbitrage, a hedge fund for lack of words and my core job. It was obviously, I was only about. I'd say about a 24, 25 years, 25 years old. And I was a data grunt. Yeah.
Today we call it big data. We were doing big data when, before it was even called big data. And my job was to pull in virtually every global statistics you can possibly imagine. Of every country, whether it was the economic data, whatever, with the stock market data, whether it was the currency data on the bar market data.
And that was done because the world was very different than everything was different. So there are a lot of dispersion. So you have to first understand the country before you even think of investing globally. So you brought all this data in, and then you did a lot of curation of that data. And then you started to use that data into what we call it linear mathematics, more Bayesian-Gaussian statistics, and linear algebra. And you come up with data to figure out which countries are doing well in which countries are doing bad. Country selection mattered most, so it was what we called in the 1990s. It was called GTAA or global tactical asset allocation. And tactical asset allocation means, you have the ability to go long a certain country or region or assets, and short a certain country or region, or asset class.
So the prerequisite for that was dispersion. As the European Union formed, you started to see the European currencies collapse. Obviously to one currency. The bond markets collapsed. And so their dispersion, you were seeing between France and Germany or the Netherlands and the UK disappeared. So a lot of the tactical resonance of making money died. And then you fast forward that; you converge that with Asia, then you get a non-U.S., and then you bring in U.S. you get MSCI world. Now you have China merging with the world over the last 10 years. So you've got one big world market, which is called MSCI All Country World Index (ACWI) .
And your overall dispersion within that index has pretty much collapsed. So you're basically come back full circle now. To say that tactically, it's becoming very hard because the systemic influences whether it COVID or whether it be the bond market again, or everything else is becoming bigger and bigger.
As there's more government intervention, there's more fed intervention across the world. So our abilities in my journey from 1994 now have evolved from tactical asset allocation to bottom-up stock picking. The bottom-up stock picking, the evolution of that is, what we call Arbitrage Pricing Theory or APT, which is more multifactorial.
You have Growth factors, Value factor, Large Cap Bias and Size factors and Momentum, Quality. So you have all the linear factors, and they actually worked for a while for quite a bit, until you saw the statistical arbitrage wipe out all the factorial efficacy you could see for the last 8-9 years, value investing is not working.
Growth is now been overtaken by momentum. So we don't even know what Growth is. Everybody's saying Growth, but it could be momentum. It might not be growth. So the ability to truly extract Alpha or value-added from factorial investing like growth factors, value factors has not been working for actually about 15-20 years now. So we are trying to fix a round peg into a square hole, for years.
And then you saw the onset of big data and alternative data come through, which about 7-8 years ago at Guardian Capital, we were able to bring in a new sentiment analytics, that is non-linear, non-structured data sets and start to actually look at new sentiment as a driver of stock prices. Fast forward from there to now - now we use significantly more alternative data sets.
And now we have artificial intelligence and machine learning techniques that are now better suited to analyze the non-distributed data or non-normal data. So we're going from what we call, Bayesian-Gaussian, or some kind of Euclidean type math to now, we're going into non-linear classification type methodologies. Classification by definition is like the human brain.
So you could see the progression from linear, Bayesian-Gaussian, type statistics or Euclidean-type responses, to now classification type responses and this type of transition or movement had led to statisticians or quantitative- based money managers to start looking into the new horizons, which is now, machine-learning artificial intelligence.
And the convergence of what you mentioned of what we call big data or alternative data and machines, which are called computers, so far.
Computers haven't been called machines yet, but they'll be a time when computers will be called machines. They just need a mechanical response and a brain chip and they become machines like Terminator movies. So the construct here is if machines can use data and interpret data systematically and objectively and give you idea generation, then the response of a human being becomes that much more profound.
So the fear that we're seeing here in my journey that machines will replace me or models will replaced me is not true. Actually, machines make you better. They compliment the response in a very simplistic way. If a human builds models and the models are extremely biased, then your outputs are going to be extremely wrong.
Also. And when you say extreme bias leads to extreme precision because you're so convinced something is right, that you're going to force that upon somebody or force it upon yourself to make a decision. And the problem I always tell our clients is it's great to see that you could hire someone like ourselves.
We're very precise in our analytics, but what if I'm precisely wrong? You can't take that chance. So the artificial intelligence and the journey of my career has been from moving away from being very precise to becoming more accurate for what I do. So the evolution of I-cubed investments, our brand who we are, Guardian I-Cubed Investments is a combination of artificial intelligence and human intelligence. The days are gone when it's only Quant, and only human, or only fundamental - only quantitative. Technology and human intelligence have meshed right now. And that symbiotic relationship between human intelligence and machine intelligence, as a numerator and innovation at the denominator, you always have to innovate.
It doesn't matter who tells you what. You always have to have a research and development mindset because we don't know anything. So unless you are a seeker, it becomes extremely difficult to understand the behavior of stock market. They're not the same. Regimes are not the same, economies are not the same. So if you're looking at the three I's, human intelligence, artificial intelligence, and the third "I", no pun intended, Innovation, you take these three and multiply each other. You don't add, you multiply because multiplying gives you a better exponent. It gives you I-Cubed. And so this is not a marketing word, a lot of people have marketing names, and they have no connection to what they do, but iCubed investments truly represents the 28 year evolution and revolution of investment management, at least from the context of who we are at Guardian Capital.
Pierre Daillie: [00:18:19] Yeah, it's a very interesting Sri. I was going to ask; the area of quantitative investing, systematic investing, systematic strategies, I think more often than not most investors and many advisors are baffled, by these ideas. they, maybe they're having difficulty. It's not that they wouldn't understand if they devoted the time to understand what systematic strategies are, what things like smart beta and multi-factor strategies are. I think they're warming up to it. I liked the idea of rebranding because I think it puts more focus, on, first of all, it creates curiosity. What is I-Cubed? Why are you guys calling yourselves that. Number one; I think you've just explained that, but secondly, it really defined what you're doing in a much better way than the the jargon, of calling it systematic strategies.
I always have to stop and think; what does that, what exactly does that mean? And I have to remind myself every time I look at it, what does that mean? Because it's not part of, the traditional thought around markets. and it's a relatively new thing, but the adoption cycle is very long, for folks in the market and especially for investors, because as lay people, they don't understand a lot of the, goings on in the market, and a lot of the talk around factors and around smart beta. I've always found, I find that the jargon, the terminology of investing that has developed in the last 15 years, sounds really smart, but it doesn't really quite excite investors and, I've read enough Cliff Asness papers to realize that, as soon as I see a mathematical equation in the middle of a, an article, Oh no, it's going to be, this is very technical.
I don't know if I want to read, I don't know if I want to read on.
Sometimes, it's a little off putting, right? So if for those of us who aren't mathematicians.
Sri Iyer: [00:20:17] Sometimes a lot of these formulas are put in for marketing reasons, not necessarily for a true knowledge too. So I would strongly recommend sometimes just because there are formulas in there, it doesn't mean something profound is being written either.
Sometimes the simplest of solutions are the most profound. So our job is to take something very complicated and make it simple, not take something very simple and make it look complicated. And, that is the challenge we always have in our industry is to make sure something, as deep as we go into artificial intelligence and big data and idea generation, how do we crystallize all of that into actionable ideas?
And how does machine experience meet human experience? In the end as a portfolio manager, I still have to pick a stock. I still have to put it into the portfolio. I still have to make sure it beats the stock market. I still have to make sure that it has relevancy to the risk premium people asking for, for the Horizons Global Dividend, U.S. Dividend, or Canadian dividend ETF.
The job is to make sure that I give a sustainable dividend. Guardian Capital, our team, gives a sustainable dividend growth, over the longer period. So the sacred responsibility there for them, that ETF is to make sure that the risk premium duration, long duration is captured so that you're not worried behind your back, every time a COVID happens, can you still collect your rent? Can you still grow your rent? Is McDonald's still in your portfolio. Is Home Depot still in your portfolio. Has the ETF changed it's stripes. So the context of sustainable dividend growth. So what we call GPS or Growth, Payout, and Sustainability sounds good. Or what's underneath that and what could actually validate that word GPS and the combination AI,  HI has significantly revolutionized the ability for ourselves to have greater conviction in the names we own.
In the end, that's what matters. You need to have conviction in the portfolio. So from Beta to Smart Beta, to Active Alpha, to Concentration. That evolution is a function of dispersion, active share, and skill. Without dispersion in global markets, everybody has taken significant active share. So now we are talking about 10 stock, 15 stock portfolios when we were 10 years ago, 70 stock portfolio with highly concentrated.
So markets have evolved because we are trying to find solutions to make sure that the alpha is constant when certain variables within the alpha definition are falling. So there's no dispersion and a skill is constant. The only way you create more alpha is taking active share. But if you have skill, HI plus AI plus Innovation, then you do not need to take concentrated bets on anything.
You could have a diversified high conviction portfolio. In our case, we run about 40 stocks. So a 40 stock portfolio is pretty concentrated to be honest, but at least every stock in that portfolio is pretty high conviction. That high conviction response is a very strong antidote to the malice in the market today with beta, which is passive or smart beta, which was basically a flavor in the market, which has not resonated too much because certain flavors have not worked.
And so this change or this evolution in the market is going to be constant. And as we keep on top of it as money managers, our team, we are quite sure that as the innovation progresses from some of the areas that we are in, things would change even more as we go by.
Pierre Daillie: [00:23:51] Yeah. I just want to circle back for a second to what you said a few moments ago, which was about the simplest explanation is the one that makes the most sense.
Occam's Razor. Right? Let me add, let me add to that because I think just to take what we're going to be talking about, one step further in terms of understanding it better is that.
That's something that human beings that we do, that our brains are capable of doing naturally, which is collecting lots of information from the periphery, from what's right in front of us. so that could be the sort of, the different types of data that you're talking about.
Like normally distributed data and non- normally distributed data, so things that are in the center of the bell curve and things that are at the outlier fringes of the curve, but we're able to more often than not. We're able to develop an insight from all of that information.
What do all these little pieces of information inform me, then that's the power of the human mind, and to get a computer, to get an artificial intelligence, to do that is really difficult. It's very difficult to get a machine to do it. We don't because, we still don't understand difficult...
Sri Iyer: [00:25:13] I'm not able to disclose certain things about what we do, because it is proprietary in how we do this, but you're absolutely true. The evolution of artificial intelligence has two basic parameters right now. What we call domain knowledge that is, do you have knowledge in your domain?
You could have a PhD. in computer science or even machine learning AI come in today. But if he's not in your domain, he wouldn't know any clue what to do, because you got to still train a machine to think like a human. And if you have a kid and you want to add your kid to grow the way you wanted to grow only, you would know how to you can't just hire a nanny and say, make him grow, like the way I want him to grow because the nanny would not know... she'll be a great, surrogate mom or dad but the question is the nanny would not know what to imbibe into that kid to become who you want him to be, or what do you think his potential is? Machine learning and artificial intelligence, it's the same thing. Unless you have deep knowledge and deep history of success and failures, more failures than success, in my opinion, makes it better. You will not know how to teach a machine to think like the way you want it to think. So Guardian Capital with our team of engineers and data scientists, we have significant advantages in our domain expertise and Guardian did a good job taking care of it's talent, there's no doubt about it. The expertise of domain knowledge, combined with the ability to select features. Features are not factors. Factors are like growth, cap, size, momentum. Features are unique variables that machines will use in randomly classifying data in decision-making parameters.
So you're looking at features that are unique to a model. So if you build a machine model and I build a machine model, what features I use to train it and what you use will make a dramatic difference in the outcome. So that's a secret. So we will not reveal the features we're putting into our machines and our artificial intelligence system that gives you that classification and output signal that gives us a high conviction.
That we're predicting with a 60% certainty, 30 days before the fact that Royal Dutch Petroleum will cut its dividend for the first time since World War II. That Suncor, the blue-chip most-worshipped or in stock in Canada will cut its dividends. That conviction level and accuracy level is built from years of data.
Years of testing and making sure that you're in sample, that is your model results and your outer sample, the results actually played to the same logic. So the advantage is that i-Cubed investments, Guardian I-Cubed investments has, is in that realm. And I don't think that's going to be a bridged anytime soon, because we have had a massive headstart for about six, seven years now with a team of data scientists that were built in house.
So I'm being as abstract as possible here, but not saying anything,
Pierre Daillie: [00:28:18] You don't need to give away any trade secrets. What are iCubed investing strategies?
Sri Iyer: [00:28:25] I-Cubed Investment strategies today cover basically two solutions. The first solution, which was evolved through the demand from the Canadian market over the last 15 plus years, almost 18 years. Now I've been at guardian for 20 years. So 18 years when we started this mindset is the search for yield. And can we give yield to Canadians beyond the construct of oil and banks?
The evolution of iCubed investments, Global Dividend Growth Strategy, or what we call GPS, Growth, Payout, and Sustainability of yield came about with the context of giving clients a sustainable yield, not high yield. Not some crazy yield we know what's going on with that in the market. Sustainable yield that can grow at a four to 5% or even 6% a year consistently.
So Global Dividends, U.S. Dividends, Canadian dividends, Emerging Market dividends is a stable of solutions that cover the Growth, Payout, and Sustainability of dividends with about a 90 to 95% up-market and a 60 to 70% downmarket. That range of movement with very low standard deviation, high quality resonance, at a 0.8 Beta allows investment advisors as well as clients through direct ETF participation to actually invest a core portion of their assets during the wealth preservation stage. The other side of what we do is; so artificial intelligence and machine learning idea generation comes from trying to predict the future one year dividend growth of a company, and also predict the probability that a company that does pay dividends could cut its dividends.
So what matters in a dividend strategy is whether the company will grow its dividend. Whether it will cut it's dividends. Everything else is relative noise. The second response to the market is what we have developed through artificial intelligence is on the accumulation side. That is when you're younger. And you're a little bit more ability to take on a little bit more risk and your duration is a lot more than say a 60 or 55 year old person who needs sustainable, asset protection, but participation type response. I'm not talking de-accumulation at the tail end where you need annuities and you're basically take your money back slowly. I'm not talking return of capital. I'm not talking any of that. I'm talking about accumulation of capital. For that we have used artificial intelligence machine learning to predict a company's forward earnings growth, and also the probability that a company's earnings will be worse than minus 10 percent. That's part of GPS, also a 25 and 30% of the GPS model uses earnings predictions because earnings prediction is the baseline for dividend growth. Without earnings, you won't get dividend growth, but we have also created solutions purely. That has all the biotechnologies, cloud computing, blockchain, electronic payments. The whole spectrum of big data is in that space, but a lot of them don't pay dividends, but when it comes to dividend investing and when it comes to a core portion of a client portfolio where they can go; my own personal portfolio, about 40% to 50% is in our Global Dividend Strategy is through an ETF like HAZ, but we also have our internal pool funds doing the same thing.
So that's what iCubed investment strategies is all about. Asset preservation through dividend growth. Asset accumulation through earnings growth. Both of these strategies cover Global, U.S., International, Emerging Markets, and Canada within this construct. That's all we do. There's no other products.
We have ESG based solutions, which are a derivation of that, but I'm just talking current over here, the kernel vein of what I-Cubed investment stands for, is these two classifications.
Pierre Daillie: [00:32:44] Two things. Number one, I imagine that given the disruption of COVID this year, that the question of whether or not certain companies in certain sectors would be able to make their dividends consistently during the course of this year became a huge question, with dividends at risk, with top line revenue at risk, closures; mind you, a lot of the big box companies, for example, like Walmart and Home Depot and Target, they were able to stay open during the entire time.
How did that work out for you in terms of, the predictive ability with regards to what are the two main, iCubed, systematic investing strategies, how important is alpha versus consistency?
If you're pre-retirement or pre-accumulation those two things are more important. I've read so many papers and listened to so many podcasts where people who have been in the business for 30 years, 20 years, the most important thing that they say they wish they had known at the beginning, that they know today, if you could go back and do it all over again is I wish I had stayed invested. and I
Sri Iyer: [00:33:59] I think if I may humbly correct you I would requote that I wish I would have stayed vested, not invested.
Pierre Daillie: [00:34:08] Okay. Alright.
Sri Iyer: [00:34:10] Vested would have allowed you to stay the course way more than invested as a connotation of in and out.
Vested has a connotation of owning a piece of property. So think of owning, McDonald's not as a stock. Think of owning a piece of property. Think of owning Home Depot as a piece of property, even think of owning Tesla today as a piece of property, not so speculative thing, that's gone up 800%. That mindset requires some degree of maturation.
That's what I would say. Yeah,
Pierre Daillie: [00:34:46] exactly. another thing is, the best portfolio to own, to remain vested in is the one that you can keep, not the one that makes you feel crazy.
Sri Iyer: [00:34:55] That's what we do, and we call it the regret free way of investing. You don't need to have regret in owning something.
And so I'll answer your first question. I think you got a two part question. The first question is, we were very, gratified by the outer sample validation of our models. In the sense, post COVID, the accuracy rate of prediction based on our artificial intelligence has been pretty significant and pretty unreal so far.
Now, remember the model hasn't even been trained on COVID data. The model has been trained for 14 years based on the past history. So it would have covered the tech bubble. We would have come into subprime and it would have covered everything in between . Now we're feeding it COVID data or the last, it has learned, we just finished a run.
So it is now the machine brain is learning from everything and anything that's gone on in the COVID world over the last four to five months. It's like a kid learning. And so it is learning as to what COVID has done and how I've learned from it and how I should learn from it. But the overall prediction, accuracy of cuts have been extremely strong.
So we were able to with high degree of consistency, which is another word for accuracy, were able to predict Royal Dutch Petroleum will cut its dividend. We exited be able to predict Darden we'll suspend the dividend. We able to exit. We're able to predict pretty well much, most, if not Oil companies in Canada, that'll cut the dividend.
We're able to even predict with about a 55% chance that Laurentian Bank in Canada will cut its dividends. If you look at, I did a, a seminar to the CFA society, Vancouver on nothing to do with product. Just hopefully I wish you had a chance. I'll make sure we send you that, the CFA presentation, we did a big data output to show exactly based on dividend landscape, where things are flaring out and where things are not working and what we saw during peak of COVID out in March, April, we saw consumer discretionary, energy, materials, flare out significantly while healthcare, biotech, consumer staples, and technology still have very strong, consistent dividend characteristics. So the probability of dividends cut for discretionary and materials all were in their 35-40% plus range for the whole index for the whole asset class, the probability of dividend cut for staples and other areas was pretty low. We also saw a significant flare up in the bastion of stability in dividends, which is real estate in the United States go from being the best to becoming the worst very quickly in a global scale within the retail space.
We also saw the naievety and the sensitivity and the vulnerability of the investment market and the investment crowd to dividend investing through banks. And we saw the lack of the ability of banks to reshape themselves out and resolve the scare of a higher chance of a dividend cut coming out of banks globally.
And the retribution that we saw to banks in Europe, where they were mandated by their governments and their Feds, and intervene to say you have to suspend their dividends. So you saw the disruption in bank dividends. You saw the disruptions in consumers' asset class, where you saw complete disruption to leisure restaurants, and any kind of commercial real estate, any kind of retail, real estate, you saw massive dislocation and devastation in that, which we have not seen the probability of dividend cut fall anytime soon, they're still pretty elevated. If I had an ability to show you charts, I'm going to show you charts.
The flare out of certain sectors or certain industries within global markets have continued to stay elevated while other asset classes like technology, healthcare, automation, consumer staples, and what we call bottom billions, that is, the middle class poor being helped by low oil and low interest rates to buy toothbrush and food.
The Unilevers, the Proctor Gambles, the Kimberly Clarks, the Costcos, all of those stocks. That's what you're vested. You don't invest it. That's the piece of land that you need to own in your portfolio to keep that sustainable dividend growth going and stay always in the stock market, not worrying about when to get in, or is it too late to get into Tesla now?
Should I go and buy it now and then lose 40% of the money on the way down. Those concerns go away. Even when you look at the regions in the world, United States by far by a mile is one of the strongest, most resonating components of dividend growth in global markets. It did flare out in certain industries, but now we're seeing U.S. Dividend Growth come back quite consistently.
Even today, this is as of June 30th. Our Global Dividend Strategy has more than 90% of the stocks in the portfolio have grown their dividends. That includes two quarters of COVID now. It might go down a little bit in the third quarter goes in because a very strong fourth quarter of 2019 is going to go away, but you're still seeing companies grow their dividends.
It's not like it's all doom and gloom out there. And so the ability to detect where dividend growth is coming from, the ability to detect and predict where patterns are frailing, which leads to a little bit more of a high conviction to when dividends will get cut, made our alpha for dividend strategies significantly higher than we would expect it, if we were just in the traditional, techniques of factor investing or anything else. As a peer group, we are beating. I'm talking about true good dividend peer group, some asset class that doesn't pay a dividend, but in a true institutional dividend asset class, we're beating the MSCI World High Dividend Index Bye about nine to 10% year-to-date.
Pierre Daillie: [00:41:06] Wow
Sri Iyer: [00:41:06] In Canada, HAL, which is a Canadian Equity Income, a Dividend Growth Strategy in Canada. We are on Lipper, on three or five years one of the best Canadian Dividend strategies in Canada today, that's HAL. The Global is HAZ. And HAU is beating SDY, which is one of the giant U.S. S&P 500 Dividend Aristocrat Index ETFs by close to 10% year to date.
All of this is coming from iCubed, Human intelligence, Artificial intelligence, which has combined through Innovation. So there has been some pretty good success with dividend growth overall, and there has been some pretty good success with the chance of separating ourselves from the pack when it comes to not chasing yield.
Pierre Daillie: [00:41:55] Yeah. I think this is an environment, Especially fraught with, investors possibly chasing yield and taking risks that they don't understand given where we are. for example, with government bonds, more than the traditional 40% sleeve in the portfolio, historically invested in governments and sovereigns and credit. I was really, looking forward to this conversation because, dividends are a traditional income investing strategy. It's important for investors to realize that chasing yield, in this environment could wind up being very dangerous, very, and so that's a great explanation. so basically that the answer to the second question was that, not only are finding the consistency of being able to, target and eliminate high chance of a dividend cut, but you're also, on the positive side, you're also continuing to target, owning the stocks that are likely to have dividend growth. So you're getting the alpha .
Sri Iyer: [00:43:02] Correct.
Pierre Daillie: [00:43:03] You're also getting the consistency. The worst possible thing that can happen for investors is first of all, not understanding the risk budget, like how much risk they're taking and not understanding at all what the term risk tolerance actually means.
Sri Iyer: [00:43:17] I think you have brought up a great point.
Consistency does not come without accuracy. If you're not accurate, you have too much variance or bias in your decision making. How can you get consistency? It's only when you have consistency, your conviction level goes up and you could stay to your knitting. So the ability to stay dividend growth, we call itself GPS or Growth, Payout and Sustainability.
We have done it for the last 10 years now to be consistently within that swim lane. And provide a long duration of about 14-15 year duration visibility of cashflow to clients, as well as very strong credit with a very low chance of the dividend cut to stay consistent with that, it allows you to create a capital preservation asset class that toggles between frailing fixed income, which does not give you income.
And highly levered or highly, momentum-based a little slow or low breadth market that we're seeing on this side. So if you look at the distribution of fixed income and equity, as I've said, we're coming back to, perhaps at the end of our conversation, we're coming back full circle that the world needs middle.
Everything is going to the tails. The middle combines artificial intelligence with human intelligence. The middle combines earnings growth with dividends to create dividend growth. So when you go to the middle in the construct, you come up with better accuracy. And when you come up with better accuracy, because you're not always in the tails, you come up with greater consistency.
Tails are important to give you ideas. The construct has to be middle so that you can provide quality and consistency over the long term. Sorry, I'm being very philosophical here, but there's merit to, hopefully there's a lot of math and there are a lot of engineering behind what I'm saying here.
If my team of engineers are listening, they would perhaps appreciate it. What I'm saying here is this, because they're going through the journey of discovery and idea discovery, trading machines, feature engineering, as well as, our portfolio management team, trying to communicate what I just spoke to and clients like advisors to say what we actually do, hopefully we did a decent job today to give you guys some ideas as to what our construct is and what our thinking is overall.
Pierre Daillie: [00:45:45] Yeah, absolutely. I'm blown away by the, the concept of what you're doing. I'm still wrapping my own head around it.
Sri, I'm curious. Have you identified any circumstances under which systematic strategies are challenged or ineffective.
Systematic strategies are challenged and to a degree in effective because factors are not working. So I would like to separate ourselves from the traditional systematic way of investing, which is more the Cliff Asness and the factorial based investing.
If you take growth factors, value factors, and you go with a GARP strategy - Growth at a Reasonable Price, you go 60% growth, 40% value. That's not working that's because the factors itself don't have a lot of information. What you put in matters more than the technique. So systematic by definition means you're disciplined.
So the word systematic has connotations in the context of what goes in. We're still systematic, but we are completely in the realm of artificial intelligence, machine learning and other components that traditional math, that the Bayesian Gaussian Euclidian type models, they are not doing that well. That's why smart beta has become closet indexing in many ways.
And you're basically getting market systemic risk. Even blind thematic investing is a lot more systemic and just a momentum chasing rather than really understanding what the theme is. Are you buying thematic real estate or are you speculating on thematic? There's a big difference. So systematic by definition has both linear Bayesian connotation that is doing a lot worse while the artificial intelligence machine learning and the integration of human intelligence, artificial intelligence has morphed intel itself into something significantly more successful. So I wouldn't throw the baby out with the bath water in the context of the techniques.
But I would surely advise clients that naively investing in factor investing might not deliver the returns that you will expect.
Yeah. If anything, the last 10 years has shown investors the bias that has occurred with growth and momentum
Sri Iyer: [00:48:03] Very much
Pierre Daillie: [00:48:03] And then, the sort of there's the simplistic view that value as a factor will have a comeback and that, that remains to be seen. I think we've started to see some signs that value will have a resurgence. And, but as you said, it's one of those areas ,  is it like a broken clock?
When is it going to be right, correct. As opposed to, will it be right? Right now, the market's certainly favoring growth and momentum until that breaks.
Sri Iyer: [00:48:31] These are labels humans put on things, value, growth, large cap, small cap in the context of artificial intelligence, these labels have no relevance whatsoever.
We don't define a stock as growth. We don't define a stock as a value. We don't define a stock as large cap, small cap, mid cap. We don't define a stock as being us or Canadian or European or Asian or Chinese or Indian. We're just looking at the principle components of what makes a business tick and what makes a company do this well in a market and grow it's cashflow it's revenue and pay out the fair share in the form of dividends or reinvest in itself in the form of stronger earnings for the future.
Under this definition, does it really matter if a stock has growth or value in large cap, small cap, mid cap. It has no that's humans, classifying risk, not humans, classifying alpha. If you tell a machine growth value, large cap, small cap, it has absolutely no idea what that is. It comes back and tells you that none of this works, I'm not going to be training myself with this data.
Remember artificial intelligence is not linear. You can't force it. You give it will learn. So it learns that it tells you that these kinds of labeling this, we call it labeling classifications are irrelevant in the context of return and risk management going forward. What is technology today?
Everything is technology. Not because MSEI, says that something is technology. That's like us going back in our days when Nortel is Nortel, industrial or technology. By the time you finished that debate, you lost all the returns and you bought it right at the peak of it. So classifying a security by what we just had a conversation here is not as relevant on a go forward basis. You need to build yourself a portfolio of global gorillas, agnostic to much of what our of what our conversation was about, and truly expose your portfolio to sustainable dividend growth, sustainable earnings. Without having any bias towards the past, that is artificial intelligence that puts those into a competitive edge.
Pierre Daillie: [00:50:51] It's true. the only way that, an AI or a machine would be able to know what growth is if you told that what it is correct. and then it would be subject to the limitations of your explanation. Correct? so you would end up with biases. Based on how you inform the machine that make it inherently wrong because it's so abstract when, whereas what you're saying is more of an agnostic approach to selecting companies that meet all of your criteria .
Yeah, you can do that without limiting yourself to all of these definitions. Wow, it's fascinating Sri. I think You've done a really good job of explaining it. I think. it's a subject that's hard enough for the majority of people to wrap their heads around yet.
I think we understand it in a basic way, but when you get down into the details, it's very complicated . The simplest things are the hardest things. They're not the easiest things. I think, when you, if you ever sit down with somebody who's been managing money for 40 years or 50 years, the only reason , that they could possibly make it look easier is because they've had 40 or 50 years to boil it down to the simplest terms.
it's not something that happened overnight. And that's what I was getting at, which is that to teach a machine to do that will take,  more and more and you need a a bigger machine with more processing power and the faster it becomes, the faster it learns. But the sheer amount of information that we contain in our brains, a lifetime of information.
Sri Iyer: [00:52:30] Let's not forget people, by the way. I'm the weakest link in the chain of i-Cubed investments, the strong parts of the chain are. Our data scientists and our portfolio managers, Fiona Wilson, Global Risk Management Specialist, a PM with me for more than 10 years, Adam Cilio one of our Senior Data Scientists, Masters in Mathematical Finance, Industrial Engineering.
Yvonne Gin, Mathematician, Smith College with a Master's in Mathematical Finance Bingsen Yan, Masters in Artificial Intelligence. Manny Saravanan and Master's in Applied Computing and Computer Science. I got a Masters in Applied Statistics and Finance, CA, MBA. That's our team. And we're growing. There's a lot of talent in Toronto. U of T Computer Science, Waterloo, Queens. They're producing some brilliant kids coming out of a university college. Not everyone's going to go to Silicon Valley going forward. I think Toronto is going to become a big. Big data hub and intellectual hub, in my opinion, in the next 10 years.
And you're going to see global talent come here. Already starting to see that. So the landscape is changing where brighter minds are coming in and at i-Cubed Investments, we welcome some of the smartest minds to join our team. So we're very excited for the future. I'll end it with that. Thank you very much.
Pierre Daillie: [00:53:44] Hey, before we let you go. We've been in this permanent staycation, at least in our spare time. I don't know about you, but we've been busier than ever. Working longer hours than ever. Am I correct in saying it's been the same for you? That, Y
Sri Iyer: [00:54:02] Yeah, I don't commute. I don't take the gold train that saves me about two hours a day. That has allowed me to do more yoga and I'm taking care of my health a little bit better sharper, but, you get closer to people more now because you all were compensating for the fact that you're not with them.
And so that actually opens up a lot more interpersonal dialogue with my team. We are very tight team and we every day, actually my meeting at 10 o'clock, it's 11 now, as soon as I finish this, I'm heading into our team meeting. And, the portfolio management team, the data science team, we all get together.
Yeah. Every day for about an hour and a half. And we discuss market conditions. Each one is a Marine and a specialist. So each of the bringing in their domain trends and talking about something where the models are, project management, alpha discovery. So it's been, I've been very excited.
I see this whole COVID thing as being a opportunity, personally, for me to discover some part of myself, which I never did. And then on top of that, I think, my team overall has been extremely happy. So these are nuances that you don't see
Pierre Daillie: [00:55:08] yeah.
Sri Iyer: [00:55:09] in a rat race. But once that rat race just suddenly stopped, you start really understanding what it's all about, inform managing people's money, what it's all about, and truly trying to beat the stock market and what it all goes to.
So I would say. I wish everybody the best of luck going forward, because a lot of good things coming forward, not bad things, a lot
Pierre Daillie: [00:55:32] in many, in many ways. I think one of the big silver linings of this period has been it's been a gift. If you get, if you embrace it that way, there's an opportunity, getting closer to the family , getting closer to, your inner circle of associates.
it's really, it's been quite a revealing time. And so on a lighter note, what have you been, what have you been streaming? What have you been reading? and I'm talking in your free time. but if you have any, is there anything you've watched recently that you really enjoyed or Netflix or,
Sri Iyer: [00:56:03] I don't want to be too nerdy, so I do a lot of reading in my space, as the company and our senior management, it has, has been seeking more solutions from our team. So I've been trying to spread my knowledge base a little bit wider. So I've been doing a lot of reading, and we're running seven mandates of the team. So there's a lot going on, but as far as, As far as binge watching and Netflix, I do watch a lot of these dystopian, scifi type movies and, scifi tag, movies, channels, or shows like, I would say something really good is Dark.
If people haven't seen Dark it's just very dystopian. It's amazing. If you really need to, use your brain to watch the whole thing.
Pierre Daillie: [00:56:45] Yeah, absolutely.
Sri Iyer: [00:56:48] Clearly any of these futuristic Sci-Fi, dystopian type movies I do watch, I do like to watch historical, so I've been, watching some of the European history type, historical or whether it is, the Romans or whether it is the, the Turks or whether it is the Mongols, those kinds of shows while they're being Hollywood based, at least they give you some context to what the past was like. So I've been watching those as far as books go. I've been reading a lot more health books, not any science books or any statistics or financial books. So probably just take care of your health, how to eat well, how to exercise. I got into stregth training, so I've lost about 10 to 12 pounds squat deadlift. I have started yoga everyday in the morning for about a half hour. Yeah, it's been a lot more inner engineering going on rather than outright, as much as we are the engineering guys and stock markets and doing everything else. It seems to be a lot more inner engineering, as much as our engineering let's put it
Pierre Daillie: [00:57:54] that way.
Definitely. You have more energy. you have more mental clarity.
Sri Iyer: [00:57:59] Yeah. a lot more calmer.
Pierre Daillie: [00:58:01] Yeah. Yeah. I think it's, I think we've definitely learned to be more patient.
Fascinating conversation. I have to say, I've really enjoyed talking to you today. Thank you so much. You've been very generous with your time. I hope that we can do this again in three or six months time and, continue the conversation just to see, where we left off, how things have changed, what's happening, how things have moved forward on your end. thank you.
Sri Iyer: [00:58:30] Thank you very much. It's our pleasure on behalf of Guardian I-Cubed Investments.
We'd like to thank you for this opportunity. Thank you.
Pierre Daillie: [00:58:38] My pleasure. My pleasure. Thank you so much.
Sri Iyer: [00:58:40] Cheers.
Note: Guardian Capital is also sub-advisor to Horizons ETFs suite of Active Dividend ETFs.
Copyright © AdvisorAnalyst.com
