Acadian’s John Chisholm Discusses Investments

third is overseas money. He has a fascinating background, UH an aspiring rocket scientist who worked at the m I t um Instrument Labs before taking a gig at State Street, and then eventually him and his partners launched Acadian about thirty two years ago. UH. They are a quantitative shop and have a very very UH interesting approach combining essentially fundamental factor models into a quantitative UH system and it's really very very interesting. They've put together quite a fascinating

track record over time. If you are at all interested in quantitative approaches, factor based UH investing, big data, artificial intelligence, the way to approach markets UH from a data driven perspective, then I think you're gonna find this conversation absolutely fascinating. So, with no further ado, my interview with Acadian Asset Managements John Chisholm. My special guest this week is John Chisholm. He is the co CEO of Acadian Asset Management. Previously

he was Chief Investment Officer. Acadian manages a six billion dollars in almost seventy countries around the world. UH. He began as an analyst at State Street Bank. Previous to that, he was a systems engineer at Draper Laboratories, which really is a great place to start. John Chisholm, Welcome to Bloomberg. Hi Berry, Thanks, it's great to be here. So, systems engineer at Draper Laboratories, which became known as the M

I T Instrumentation Lab. Is that right? How did you find your way from M I T to uh, the instrumentation Labs. When I went to college, my passion, what I was excited about, was really building or designing spaceships. This was in the early eighties, So so you were a rocket science I was a lot of the aspiring rocket scientists exactly. And so I got my undergraduate degree, but I found my senior year I was gotten really interested investing, and I was spending a lot of time

mostly just reading about investing, you know whatever. It was journal business publications, journals and uh. When I decided, okay, what do I want to do now? I thought, well, I probably want to go back to grad school. Do I want to do finance investing business or do I want to do um aerospace? And uh. I had had a an opportunity apply to several different programs, so I had both. I had a finance opportunity and an aerospace opportunity, and I thought, why don't I try to test both out.

I'll get a full time job here in Boston area. The only place there's not a lot of airspace jobs in Boston A Draper Labs is one that works on guidance systems. So if you've got a satellite or a missile, you try to figure out where's it's, where's it going to go? How does it get there? Before GPS of

headed guidance system. So uh, so I worked there full time, and I got a part time job, like sort of after hours job, working with a fellow named Gary Bergstrom, who was later one of my co founders at a Kadian. He had been a portfolio manager at Putnam and UH in the seventies and then he left sort of off on his own, consulting for money managers consulted at the time.

As big project when I was working with them was for State Street later State Street Global Advisors, and we helped building design their first international index fund and then later on some international active strategies. So that was sort of a part time job. UM I went back to that made me decide that was more interesting than the aerospace stuff I was doing at the time. That made me decide to go back. So that's what when you

said to go back, you really started. Your first full time job in finance was as an analyst for State Street. Is so, so the State Street job was a was also a part time That was also while I was at school um UH working for them for like there was some time off in January and then the spring semester. I worked for them as a potential employer. But in the end Gary's goal was to launch an asset management firm. UH myself and we had another colleague, Churchill Franklin, and

another colleague, Ron Frasier. They all came aboard. We all came together about the same time around when I graduated and uh and so we launched a Kadian as an active money manager at that point. So that's thirty two years ago. An active manager as well as a heavily influenced by quantitative strategies? Is that affair the work? We're a quantitative manager. We were all you know, my background aerospace engineering, UM all all quantitative. Gary's background, Gary gotten

a PhD from m I t uh. So we were all very quantitative. But quant at the time was not as sophisticated as what quant today is. Right, there wasn't any machine learning, there wasn't any big data. There was you know, little data. There were statistics, right, so you know what's the average payoff to value UM and ch how do we build a portfolio that captures that payoff? Very simple quantitative tools that we use back in the middle eighties. So do you consider yourselves today an active

manager or quantitatively? Like I think about firms like um D f A or any of the FAMA French based UM factor models, and they're somewhere between active and a quantitative screening approach. How how would you describe a Kadian? Describe us as active? So most of what we're doing is highly active, potentially high tracking our against a benchmark. We we have the flexibility so we can build low tracking our strategies. UM. This ties into this concept of capacity.

How much money can you manage and still expect to add the value your clients are looking for? And typically the more money you manage, the harder is to add value. UM So at lower levels of active risk, lower expected value added, you can manage more money. There's some clients who are happy hiring managers for that. They're also usually happy paying lower fees. Right, so you really have to trade off both from perspective of adding value and from

perspective of running a business. Where do you want to be? Most of our strategy is highly active, but we have some that are shading more towards UM enhanced index. When you say enhanced index, you're taking a basic index and then adding a little flavoring to it to move it away from the benchmark. Yeah, we're so, for example, we might say enhanced. The next would be if we have tracking or of less than two percent, if we have

one and a half percent. Tracking are just means what's the standard deviation of the expected returns versus the benchmark? UM one and a half percent tracking er would be enhanced index strategy, you might only expect to get one and a half percent UM excess return associated with that net of fees. If we had a more active strategy, we might expect we might see four percent tracking er. We'd expect to get about two and a half percent

active return neta fees. How do you avoid the challenge, as as Bill Miller described it, of UM active managers TUBJE active fees but are effectively our closet indexers. How do you clearly differentiate yourself from that group. So, so there's two parts to it. One is what what's under

our control? What we can do. We can build portfolios that are active in the sense that they are very different, they look different from the benchmark, they have UM high higher levels of tracking, or they have high active share UM. The other part of that isn't is the client's job. So if the client hires twenty managers like that, they're

still getting close to back back to an index. Let's talk a little bit about your quantitative approach, and I was noticing on your website you describe a fourth step process, and really what that is is strategy signal generation signal consumption and then process can can you explain what that means to to the perhaps the lay person who may

not be familiar with a quantitative approach. Absolutely, let me let me start with a signal generation part, because that's maybe the part that will be easiest for for people to start with the basic ideas. There's different characteristics companies have. Those characteristics can be on average predictive of returns. So, for example, one characteristic is just is how expensive as

a company look on whatever metric P ratio? Right, So you've got a company that has a P ratio of eight and one that has a P ratio of forty. That's all you knew about those companies? Which one would you want to own? If you look at the last fifty or sixty years globally, you'd say, I want to own the P eight company right on average is going to do better. The problem with that is you can have a ten year stretch where the P forty company

kills the P A company like we've just had. We've had you know, what would you rather own right the last ten years? We'd rather row on Amazon or would rather own P G n E? Right? I mean, so those are the two biggest streames you can you can really sorry I picked those out of it. Not hopefully it's not quite, but it's true. Generally growth stocks have done very well. Value stocks and stuff that joined up for a decade. So how do you it around that? UM?

So let's say you believe on average, value is going to outperform, like some people might not believe that. But let's say I do. I want value in my portfolio, but I don't want to underperform for ten years in a row. What can I do? I can take other characteristics that I believe are also predictive of return and combine those with value. So I might say, for example, UM quality measures, right, I want companies that are well managed. How do you define well managed? Well? As a dozen definitions,

But let's say one definition is inventory turnover. Do you have companies that turn over their inventory uh more frequently than companies in the same industry? Right? And maybe that's a signal that on average has some payoffs associated with it for companies in many industries. So now I've got value and I've got this. Let's say I think people talk about a lot in markets as momentum, So I've

got momentum. Companies have good momentum um, they've been performing, they've been performing their peers for the last you know, six to twelve months. Maybe that's indicative that on average in the future, they're likely outperformed for the next, say, one to three months. I want to wrap that in. So you combine all these different signals and you've got

what you historically people call a multi factor model. Right, and so now, even if value does badly, maybe momentum and quality and these other things do well enough to allow you to still outperform, which is always the goal, uh for for us and for our clients, and and so that that's sort of the the genesis of signals. It's just there are different types of characteristics that we use to help predict company returns, and we then combine them. So when you say signal consumption, a couple of pieces

of that One is how do you combine these things? Right? Is the payoff to value the same as the payoff to quality will probably not um. So you have to figure out what do I expect to get If I'm looking at it doesn't differ by the type of company I'm looking at, is a tech company, maybe it has different drivers overturn than a utility, And so I have to mix the weight on those signals depending on what kind of opening I'm evaluating. Uh, And so that's part

of consumption. Then the second part of consumption is how do you implement that in a portfolio? Right, So, ultimately I'm gonna hold stocks and a portfolio, I'll hold Amazon or I won't. I'll hold pg n E or I won't. What's my weight going to be? Hopefully it will be last ten years, hopefully high on Amazon, lower zero. And But but the idea is you've got to then turn those expected returns that you're getting from the signal generation

part of your process into portfolio positions. At Acadian, we use a pretty quantitative approach to do that as well. We use what's called an optimizer that basically trades off the return expectations we come up with from the signals into and maps those into portfolio positions by trading those off against transaction costs. So, if I'm trading again Amazon, Samsung, a big liquid company, transaction costs are probably gonna be

pretty low, almost negligible. But if I'm trading a less liquid company, that may be more efficiently priced than the return up tratunity may be much greater. But I need to now account for what's it going to cost to get in the position and what's it going to cost to get out of the position someday. So what you're describing sounds a lot like traditional factor based investing. You're describing um value, describing a momentum quality. When we talk

about liquidity, I always think about cap size. How does your approach differ from factor investing? Or am I asking that question wrong? Are you effectively a farmer French factor type investor? Now that's it's a great question. Uh. And in some ways our approach is very much like factor investing. In other words, we consider these different signals. We consider them to be types of factors um. What's different is that we integrate the factors. So, for example, let's suppose

you've just been on momentum factor by itself. There's some periods if you built a portfolio that has the most attractive whatever it is, ten percent of uh, let's say us amentum stocks and then the least to try active or shorts, or you just went long the most attractive ten percent, there are some periods where your portfolio beta. Your sensitivity to market movements might be might be two, so you might have a huge amount of volatility in

the portfolio. And there's other times when your sensitivity to market movements might be very low, might be a bet of point five um. Why does that matter? It impacts how you can control risk. If you're doing these single factor portfolios, for example, unless you're very careful about how you build them, you are likely to take on all kinds of unexpected risks in the construction of those portfolios. With a multi factor approach, you're not beholding any one factor.

You've got all these different characteristics you can emphasize in the portfolio, and you can trade them off, and it allows you to manage risk better. Portfolio construction can be a lot better than what people do when they're doing When people talk about factor investing, if you look at a typical factor e. T f UM, it's not built in a very efficient way. It's more costly to investors in ways that the investors can't see things like how

they trade the portfolio. So they simply take a rank order of companies based on some factor, and they rebalanced. They buy some of the most attractive ones that have just gotten into that list and sell some of the ones that have fallen out. That can be a lot of turnover. And there may be times when you want to hold onto something that's become less attractive because it might be expensive to trade out of it and it hasn't fallen that far right, So it's important to do

to be smart about how you use these factors. And I'd say the keep one of the key things we do is we worry a lot about the engineering of our process. How do you put these factors together? How do you minimize the slippage the transaction costs while still getting exposure to the underlying factor in the portfolio? Quite fascinating. Let's talk a little bit about what it's like to run a global organization. You're headquartered in Boston. Congratulations on

the Patriots. Thank you. Um, you were at the game you mentioned, Uh, I was at the game, first time I've ever gone. So you don't know how long the Patriots that this might be their last Super Bowl in a while. Good idea to go. Yeah, now that makes perfect sense. Um, you have affiliates in London and Singapore and Tokyo and Sydney. What other countries are you located in I mean, I know you have clients in We've got clients probably thirty thirty five countries, but really they

coverage of Singapore, Sydney, Tokyo, London. We are thinking, because it breaksit, we may need to open an office in Dublin, UM, perhaps Amsterdam, but there's enough uncertainty there that we haven't actually pulled the trigger yet Amsterdam. So so if people I keep asking this, and I'm getting very different answers from very different people. If Brexit happens hard or soft, and London is no longer the central finance location for Europe,

where where does it go? Amsterdam doesn't really seem seem like the place UM Geneva stud guard. I mean, I can't Paris. Some people have floated, none of them seem to make sense. I think Frankfort's got the economic in many ways the commercial center of Europe. A lot of people don't love Frankfort. Because I live near Frankfort for a bunch of years. It's not always um doesn't have the cultural reputation that Paris does. For example. That being said, it's a very comfortable city to live in. So I

think Frankfort will do well. That all being said, you know, if if we did something with another office, we're we're just basically opening up an office to meet the regulatory requirements. We still keep London, that would still be a major center for portfolio management. Uh, you know, for our team, for client service team. So your client base is uh primarily institutional and public pension funds and other large investors. Are they mostly US located? Are they around the world?

What what your mix so are right now? Are mixes price seventy thirty um US non US? Uh? We we from a business for spective you know, as a firm, you want to be diversified, you know, so having a fair amount of non US exposure with our clients is something we strive for. We think there's a lot of great growth opportunities in terms of just the the growth of pension markets, of of institutional investor markets in Asia, for example, there's still growth in Australia, um in in

Europe as well, there's pockets. You know, a few years ago Germany really didn't have defined benefit pension plans. Slowly evolving a little bit, So there's there's definitely opportunities. Don't most of Europe or don't doesn't much of Europe have UM some sort of retirement system covered by the government. How do you operate around that or is that their social security and it doesn't take the place of a

real retirement. Yeah, I mean there's a lot of variation, but a lot of Europe, you know, like the UK, for example, you've got public pension plans, just like in the US you've got uh State of California cows, cowpers and calsters. Well, in in UK you've got um local authority pension plans UM and you can sort of think of them as the equivalent of public plans. Here in the US you've got large companies based in the UK that have private pension plans, so there's some state provision.

But again, just like in the US, you have sort of security, but that doesn't exclude all these other types of pension plans. So in your list of countries, I didn't hear a whole lot In China. Is Hong Kong attractive or is mainland China possible or if it has, the government made it too challenging to set up shop there. Now the government's actually moving to liberalize, so it had been very difficult for a non local investors to invest

in in in Chinese assets and manage money for Chinese institutions. UM. We do manage money for Hong Kong clients, but that's sort of separate stile, a separate um regulatory structure. Is that why you're located in Sydney for that part of the world. Singapore is is we're serving right now, we're serving Asia extrapan out of Singapore. UM. But but China is liberalizing and there's plenty of non local managers now setting up shop to manage money for the Chinese institutions

in China. The challenges in China you need some scale. You need a partner because you can't touch the retail market without a local partner. And there's only really four big institutions you know that you've got that are like sort of the equivalent of Cowper's for example. UM. So that that that market, the institutional markets very narrow and the rest of the market, the retail market. You need

a partner. If we found the right partner, I think would be very excited about doing something in China that we're certainly doing some work there. We've we've had some people not fully full time based there, but spending a lot of time in the market. But that's something remains to be seen whether we'll find the right opportunity to really be a player there. So you previously were chief Investment Officer and now you're co CEO. I have so many questions about both. So are you still working in

a ce IO capacity as well? No, so, so I'm I'm still interested. I still go to Investment Policy Committee meetings. But you know, if you're if you're taking out a new role and you've picked someone to succeed you in your job, you really need to give them the ability to run that function. I've got a great successor, my successor, Brendan Bradley's or CEE i O UM started last year. As see. I always been with it Katian for a long time and got complete confidence in his ability to

manage the investment function UM. I still participate in some of the meetings and I I'm interested in the research and I talked talked to lots of the investment professionals. It's part of my job as CEO is being in touch. You know, what does an investment firm do? We invest

for our clients, So it's still important. But Brennan is managing and leading the investment team and you're a co CEO, which sounds like it has a whole lot of complications and issues that would come out of that dual CEO role. At least we've seen that with public companies. How do you navigate that? Is there a clear distinction between who is running what? Tell us a little bit about your co absolutely? Let me let me you that, um, because it is a great question. Uh. We we went through

a succession process. Our former CEO was stepping down, was retiring from the CEO role, and so myself and one of my colleagues rosted out were internal candidates for the role. We have a selection process where we had an equivalent of executive committee. Essentially you think of as eight individuals

running the firm making that decision. We were two members of that group, UM, and we we both shared our views on what's our vision for a Kadian, where do we want the firm to go, what would we like to do differently with with our executive committee, UM, it turned out what we were very well aligned in terms of where we wanted a Katie and to go. So when we looked at UM sort of are there situations I have a lot of respect for Ross, my current co CEO, he has I think a lot of respect

for me. He comes from a marketing client service background. I come from the investment background. And we we both wanted each other to remain at the firm, thought about how can we do that, and we looked at examples where there have been co CEO structures in the past and other firms, the ones that worked relatively well. UM, and there are some Um generally you had co CEOs with highly aligned visions and that we're able to work

together to provide a single voice of the firm. Right, So you don't want somebody coming to Ross getting one answer and then coming to me and getting a different answer. And we we thought, given the fact that we in fact do have highly lined visions, we do have areas of expertise that are complementary to each other. Uh, we thought this is something that not only could we pull it off, but it would actually be beneficial for Acadian. So we're now a little over a year into the role. UM,

we think we're managing the firm effectively. We're getting feedback from our team, UM that that's the case and UH and I think it's working extremely well so far. UM. We What we typically do is issue comes up, we will discuss it together. We'll figure out where are we what are we looking to do? Um. And there are times when you know it's an area. Ross has a lot of expertise in the fern him more, Um, there's times when it's an area. I've got a lot of

expertise in the for me more. It's great when we do need to be in two places at once. Right, he can be in Tokyo and I can be in Boston, or vice versa. UM. One of us can be doing to meeting with some of our clients, another one can be running internal meetings. So it really helps us, I think, be do a more effective job of managing the firm to have the structure we have. Quite interesting. So let's talk a little bit about what's going on in the

marketplace today. Are you still seeing the same sort of miss pricings and securities that perhaps we're so abundant a decade ago? The miss prices have changed a lot. So if we take any particular signal that we've used ten years ago and we look at the payoff to that signal today, it's lower today. Right. So typically, whether it's inefficiency is being squeeze out of the market, it's arbitraged by different types of investors, whatever it is. You know,

Typically the payoff to these characteristics decreases over time. So as a result, we're in a way, we're on a treadmill. We need to keep on finding new ideas to replace the old ideas that aren't working as well anymore. So when you say these the payoffs to these ideas um decline over time, is that all these ideas? Is that a function of we had a giant reset with the

financial crisis. Hey, anytime the markets lose of the value, you have to think that some value is going to be created and a lot of babies get thrown out with the bathwater. Or is it just the nature of every good idea eventually runs its course. I think it's really a little bit of the reset idea. So there's no there's there's definitely a pattern that differs a little

bit the rate of the client. And some of these things accelerated during the financial crisimmediately after the financial crisis, and the payoff to value is the biggest one where it's clear really been the worst ten year period for value globally post GFC that we've seen in the long term history, whether it's the U S history or longer history that's different. That all being said, a lot of

the factors, it's an average thing, right. There's some signals that still work, you know today, not much worse than they worked ten years ago, but the average signal, the payoff decreases a little bit every year. M that's quite interesting. Quant has been around for thirty forty years or so. Do you think things are very different based on the rise of you mentioned earlier, big data and artificial intelligence.

How has that affected how Acadian approaches quant investing or is that just something that is a background noise that affects the market overall. Yeah, I think there's um there's no question that the machine learning, art and big data, artificial intelligence, those are early days, right. Those things are starting to impact investors and how people invest. But we're still in the early days of that in quant, let alone in finance in general. And there's a lot more

to come. But I would say things have changed a lot since the eighties and nineties, the the cerphification, the not not so much big data, but just any kind of data now is a lot more available than it was then, So we have a lot more information and quants can do things today that they couldn't do twenty years ago. Fundamental investors could maybe do them for a

small group of companies, quants couldn't. Today we can look at all these like we have industry specific information about lots of companies that we just didn't have access to twenty years ago. Is it the technology and database or is it actually the specifics of the data itself that's changed so much. It's both. So the technolog no question, the technology and the database access um, the power and speed of databases and of software and processing in general

has increased tremendously. That makes a lot of things easier to do. Machine learning, those algorithms can be very computationally intensive, and you hardware you had five years ago, you couldn't do these things today. Today you can do them on your on your laptop. In some cases might take a while, but there's things you can do on your laptop. If not, you go to Amazon Web Services and scale up processing power and you've got everything you need in terms of

the sort of the processing computational aspect of things. So that's changed a lot, but also the data itself today is much broader than it was. I got started. You got a p you gotta p B, you got a price to cash flow. You've got a market cap, a price, and maybe a divident deal thrown in, and that's there. That's that was your data. That was like four ish and then UM, shortly after you started getting analyst data electronically.

What what do you think about some of these alternative data points that people are pulling from either satellite data, Hey, here's all the ships moving oil around the world, UM, or parking lot activity to determine how well read tailors are doing. Is is any of that potentially useful and valuable to investors? Or is it just um a bunch of of geeks playing with some new tech toys and

and kind of having fun with it. It's both, so that on the latter point, you know, we we have an analyst, and if we have a satellite data project, we have no problem getting somebody volunteer, put their hand up and say I'd like to work on this. This will be fun, right, So that's true. It's potentially valuable. Now, whether it's actually valuable to any individual in any particular investment firm depends on their style and their process. So

let me let me tell you what I mean by that. UM. If you've got a satellite data, let's say you're getting your parking lot, your infrared images UM, and you're getting information about you know, parking lots UM. And that's if you're following retailers and UM. Investing in retailers is a big part of what you do that can be useful in predicting over the short run revenues. You've got to have a lot of infrastructure, you gotta know all the locations.

You've got to be able to aggregate that in quasi real time, and satellite coverage at high resolution at a quick successive short time intervals is still expensive. UM. So you've got to figure out is it worth it to your process to do that. If you only one percent of the portfolio have that you have invested in retailers, maybe it's not really going to move the needle that much, right, So so let's talk about some other things that don't

involve satellites. This has been a let's call it a typical UM political environment for the past couple of years, not just the Trump presidency, but Brexit and the financial crisis and the rise of the tea party. How does a quant shop manage those sorts of non market inputs or does it all just come out in the wash and it's not really all that important. There's there's two sides of this. One side is what I call risk management.

Right can you if you can observe some of these risks and you don't observe them in a standard quant risk model because they're the quant risk models are typically backward looking. They're not forward looking. So you've got to, as a professional investor, think, what are some of these risks that maybe aren't priced into the risk models that are looking at the historical data, but that could impact the portfolio UM and let me give you an example

of such a risk. One one risk is UM. We we managed one strategy that's a low volatility equity strategy. So we're trying to do there is reduce the risk of equity markets UM capuited benchmark. Let's say in the US, it might have a twelve or fourteen vol and we might want to produce a ten vol for example UM and what that means is somebody gets the same risk returned that that they get on count a capuited benchmark, but they get it with less risk. That's very helpful

from asta allocation perspective. When you do that today, though, what you're doing is you're taking on a lot of interest rate risk because these lower risk companies typically tend to be higher dividend companies. Companies are more sensitive to interest rates. So if you're worried about a rising interest rate environment, your historical risk model wouldn't say, constrain your

exposure to interest rates your sensitivity interest rates. But going forward you might want to do that um in a low volatility portfolio so that your volatility doesn't come out much higher than you expect or your return is much lower than you expect if interest rates do in fact continue rising. So that's the risk management piece. You want to anticipate certain risks and build that into your risk

controls that you apply to your strategies. The second pieces can you use are there are other signals that help you navigate from a return perspective, These kinds of you know, macro events and uh you know, for example, volatility itself can be an earlier warning signal. Right, every major devaluation of currencies and emerging markets and many market breaks were preceded by periods of rising volatility, rising by untill. He

also sometimes predicts more benign environments. But the point is, if there's a signal there, maybe there's ways to predict these these environments, and so are the top down part of what we do tries to look at these macro events um or potential macro events and figure out how can we anticipate those and how can we position the portfolios based on that anticipation. So you mentioned a rising

rate environment. Lots of folks have been focused on the Federal Reserve and focused on are we gonna take a pause? I imagine that your shop doesn't spend a whole lot of time struggling with that that it should end up in the data, and it's not the sort of thing that you have to play macro tourists or or am

I giving you guys too much credit? No, you know you're giving us just the right amount of credit here the very the you need to play the game that you're good at um and so we don't want to do we don't want to try to do things that we've got other people who are much better at it than we are and predicting UM rates using it the FED what what's the FED going to do? That's not all as a quant manager, that's not really what we're

good at, right, So you're absolutely right there. What we would do is we would say, let's just look at what's happening with you know, the short term rates, long term rates, what's happening with the yield curve. Those can be signals that we use in a model, but we're not trying to really forecast the direction of interest rates per se through FED statements or through other kinds of

actions like that. Um, it just means trying to do what what where we think our edges and really trying to focus on that in terms of the things we actively do in the portfolio. So you mentioned your models. When I was perusing the various offerings you have for institutional clients, there are thirty something different models, maybe even more. How do you develop different ideas? How do you express them in a portfolio? Is it strictly math or there

are other guiding principles that affect that. The first first step is always is there a you think it as a story, it's really hypothesis of why a particular characteristics related to return? Why is it? How can it be used? To predict returns. What is it? What's the inefficiency that we're capturing? And if we have that, then the next step is, okay, now let's spend some time looking at the data and figuring out how do we best um create how do we best capture that inefficiency, how do

we best measure it? Uh, So, you know, we might have an efficiency related to momentum um. And back in the seven back in the eighties, you know, you had some papers about price momentum and they basically said, Okay, the best way to capture price momentum at the time is uh, sort of a twelve month trailing risk adjusted return price return. That's your best momentum measure. Since then,

a lot of things have changed. We've got a lot better understanding what drives momentum, you know, what's what are the inefficiencies we're capturing with it, and a lot more ability to um turn that into different kinds of signals. And today, in addition, we've got machine learning, so we can put in all the historical prices and say, okay, machine learning algorithm, what do you think the best predictor

of return is based on past price moves? And when you do that, you have to be careful because machine learning is one way to do what's called overfitting a problem, you know, where you you have a great solution of the past, but it doesn't work in the future. UM. One of my colleagues, Michael bat Nick, once observed, the best track record of any model is the last ten years something something to that effect. Does that sound about So every model, you know, every model implicitly has some

potential for some degree of overfitting associated with it. We try to guard against that. We have various, you know, statistical procedures that we follow, UM in various research procedures we followed to try to avoid that, but it does creep in no question about that. We have been speaking with John Chisholm. He is the co CEO and former

Chief Investment Officer for Acadian Asset Management. If you enjoy this conversation, we'll be sure and come back and check out the podcast extras, where we keep the tape rolling and continue to discuss all things quant You can find that at iTunes, overcast, Stitcher, Bloomberg dot com, wherever final podcasts are sold. We love your comments, feedback and suggestions right to us at m IB podcast at Bloomberg dot net. You can check out my daily column on Bloomberg dot

com slash Opinion. Follow me on Twitter at rit Halts. I'm Barry Ri Halts. You're listening to Masters in Business on Bloomberg Radio. Welcome to the podcast. John, Thank you so much for doing this. I've been looking forward to this for a while. We were having a conversation in my office and on the way out the door, someone said who you who you say interviewing today? I said, oh, John Chiselm of a Kadian asset management and the person said, oh,

I've never heard of them. Uh do they manage any money? And my auntswer was spitting distance from a hundred billion dollars and that sort of shock some people. How do you, um, how do you feel about being a little below the radar and why are you sort of poking your head out from from below the radar? So in general, it's we think it's good to run a little bit below the radar. Right. There's there's elements of First of all, you can only manage so much money and still add value.

You just have to be careful managing capacity. Um. And we also if you're a big name in the industry, uh, you get more press attention. That's in one way that's good, but in another way it can also be detrimental depending on what's the type of attention. UM and a lot of investors, a lot of institutions especially, want managers that are very careful to focus us on maintaining their ability to add value for clients by not getting too big. Right.

We all know managers that have grown and grown and then at some point they just couldn't add value anymore. They just got too big to happen. Then they shrink and shrink, as we've seen with a number of famous hedge fund managers the past decade or so. Exactly. We we just like to be maybe a little bit um less volatile in terms of our business than that. And it's just best. It's best for our team, it's best for our clients. UM. And those are really the key

considerations typically. Now sticking the head out part is it is important, I think to have some degree in name recognition because a we want talented people, and if your potential employees don't know you are who you are, then you may not be their first place of employment of choice. UM. When there's an opportunity that might be a great fit for them. So there's an element of that, and also we're doing a number of new things that we haven't been doing before. One of them is we've built a

multi asset strategy. So historically we've been primarily an equity firm. We have a multi as strategy today that has about a little over a year live track record. It's done very well relative to many of its peers. It's a very quantitative approach, is very consistent with our philosophy, but it invests in equities, fixed income, currency, commodities, and options, and the goal there is to create a income return stream that's much more stable than what you get from

an equity market beta. You know, that doesn't go up and down every time the market equity markets go up and down, but that provides a fairly consistent typically, for example, one version of strategy cash plus five return. So not quite risk parity, but it's not risk parity because we're

not necessarily investing in equal risk proportions. It's it's really you can think of it more as it's related to this concept that there's certain efficiencies that operate not just inequities but also in in other asset classes um but it also relates to specific expertise in these other asset class is that there's individual drivers saying commodities UM that are fairly unique there and there you can capture them through these return models and in turn gets some significant

value added from that area, which you don't get in a lot of these so called alternative risk premium strategies. So you mentioned capacity, you're at eight six billion. How much more capacity is there? You are you in broad areas and equities and countries that have a lot of a lot more headroom or do you do you see

limitations not too far down the road. It varies. So the emerging markets, for example, strategies is closed new clients when if we if a client withdraw some money, will add some money for any existing client more capability to invest, but we're closed their frontier markets is closed. Emerging markets small cap is closed. Are non US small cap again subject to some reallocation when there's flows out UM is

also closed UM. But we have capacity in areas like global like our manage of volatility strategies, this multi asset strategy. So we what we do is we have a very specific process to measure how much money can we invest and still meet our investment objective in each strategy, and when we hit that number, we close the strategy. If we've got headroom, we tell the clients here's how much headroom we have. There's how much we expect to be able to add before we have to close the strategy.

And I've seen, um, some of your long short UH portfolios are one ninety over thirty or something like that. Am I? Am I getting that more or less right? Yeah, we have a variety. We have some that are pure um market neutral, so they're equal sides long short. We have some that are one thirty thirty so d long short, and then we have some other variations as well. We have a we call it diverse fight Alpha strategy that's

a slightly different ratio as well. But essentially, all these strategies the ideas take advantage of the inefficiencies on the short side and the unattractive companies that we follow. And alright, so one thirty thirty is the long short as opposed to fully market neutral, which is its leveraged. So effectively, we have a global leverage market neutral strategy that's about

two long short. Quite quite interesting. Um, let me go through some of the questions we didn't get at get to during the broadcast portion before I get to my favorite questions, UM, And there was one that I thought was kind of interesting, and I pulled this off of either your website or something you had written. Quote documented recurring behavioral errors drive irrational actions in financial markets, behaviors

that are often contrary to investors best interests. How does your firm uh use your understanding of this to help manage money? So this goes back to how do we come up with these signals? So, for example, one behavior Lewer is investors typically are overconfident in their ability to

predict future growth rates. So if you're buying growth stocks in the tech bubble, UH, and you're looking at companies that are growing their earnings at or more a year or higher UM, those companies were trading in some cases at multiples north of hundred on earnings on current earnings. And if those companies had continued growing their earnings at those very high rates for twenty years, that would have

been a reasonable price to pay. What happened is investors didn't realize that, yeah, they can grow their earnings at that rate maybe for one year, three years, four years, it's very hard to do that for twenty years UM, and so that overconfidence, I think is one of the key drivers of why you see in the law term value UM working effectively. What's happened the last ten years that's interesting is two things. One is that actually there were some companies that actually did grow their earnings for

at really high rates for a long time. So typically people think of the Internet services, you know, the Googles and Amazons and so on. Those companies have been tremendously successful for a while. Albeit you're starting to see a few cracks in those growth rates now for some of these some of these companies. And the other thing that's

happened is just a general repricing within valuation. So UM you had a certain level of dispersion where value was so successful from say two thousand one to two thousand seven, that the dispersion of valuation multiple shrank and as a result, you didn't the expensive companies weren't really that much more expensive then the slower growing and expensive companies, and that gave a little bit of tail wind growth over that period.

I think we've pretty much worked off all that, all that dispersion UH, or rather the tightening of dispersion, so we're back to more normal level dispersion now, so at a kadie and we'd expect going forward that you're more likely to have at some point soon whether soon as you know, next month, next year, but not not in six or seven years. Sometimes sooner than that. We expect to see value reassert itself, and so we continue to have some component of our of our factors focused on valuation.

So q FO fair to say that was value reasserting itself. Yeah, I think actually a Q four was just for us for a Kadian particular or not a great quarter um And it was partly that actually value in some markets didn't pay off well, but it was also partly um uh. Smaller companies in general, especially in the US and emerging

markets did poorly the relative larger companies. And we have in our portfolios a fair amount of exposure too smaller and medium sized companies because typically that's where we see the general inefficiencies any kind of fact or we see those as being greater in that area than they are in the very large cap companies. So what hurt us in the fourth quarter a little bit of value, but primarily just the risk of small versus large, biting us you hinted earlier at E T F s UM sometimes

being less efficient than other ways of expressing the same strategies. However, go back a decade or two and there were certain strategies that you can only get through expensive alternative investments. You're paying two and twenty for certain strategies that you can now pay on a fifty basis points and an eight dollar transaction fee. So what do you make of this landscape and what does this mean for quantitative strategies

eventually migrating to some of these low cost products. I think that we've seen that trend and there's a good reason for it. Right. Investors should be looking for what's the if I want to get a certain return and risk stream, what's the least expensive way for me to do that? And it's been great for investors the fact that there has been pricing pressure on the esset management side of the business. That's actually a great thing for investors, right.

It forces the investment managers to be more efficient, It pushes the overpriced products away from you know, it makes them less viable, and it allows strategies that can be run inexpensively. But still provide value to do well in the marketplace. So great for investors, tougher for asset managers. Is not as easy to make money now as asset manager as it was, you know, ten or fifteen years ago.

We've seen margins for the esset management business gets squeezed a little bit over the last few years, no, no doubt about that. What as long as we're talking about UM indexes and ETFs and price squeezes, what do you make of the argument that some of this movement away from active management into passive is distorting prices. I don't think we're there yet, UM, in terms I do believe. Look,

there's a value to price discovery. If you had a hundred percent of of every know, all assets were pastively managed, UM, you wouldn't have a mechanism of price discovery. But you don't need you don't need, you know, as assets active management to get the price discovery process to work. UM.

I think there's been various academic work on this. Andrew Low right in your backyard, and and there's folks at Harvard and and generally they what they come up with is that you can have a greater level of passive management than we have today and still get the you know, the social benefits if you will, of the price discovery process. Quite quite interesting. I know I only have you for a limited amount of time. Let me jump to some

of my favorite questions we ask all our guests. UM, tell us the most important thing that people don't know about, John Chisholm. Wow. Um, that's a tough one. And you know it's funny because I know you gave me the questions in advance. So that's that's the one where I looked at and I was like, I don't know if I have anything there, and I skipped it. So I did not pre didn't think. I did not come up with an answer to that that particular question. UM. I would say a couple of things. One is, um, I

love I love asset management. That's probably I love investing. That's probably not a that's something that's some of the people who work with me know pretty well. But it may be something that you know, the you're listening audience maybe doesn't appreciate as much. Uh. And um. The other thing that maybe it's something that's maybe not really directly work related is um. Uh. Two things in terms of leisure activities, I love ultimate frisbee. Ultimate frisbee is a

great sport. I don't know if you know what it is. Of course I know. I went to college at stony Brook. Ultimate frisbie was a huge thing on campus back in the nineteen hundreds when I went to school, so so same. You know, I actually went to high school at Bronx Science and the Bronx here, and uh it was it wasn't. I was not the Ultimate team, but that's where I started playing with some of the guys on the team.

And then I played a little bit in college, played after college, and you know, now there's a over in Boston area. There's no over forty league. I still get a chance to go out and play every now and then. It's not a contact sports, so you know, it's not like rugby exactly. That's that's the beauty of it, I think, is that you get a great exercise. It's a lot of fun, it's very social and um and you don't kill yourself. It's not like I played basketball typically once

a week as well. And I'll tell you after alsop and left and right, I'm limping, you know, for like three or four days, so I can start walking well again and that does not happen after ultimate that. That's very funny. So the next question was a question I used to use as a throwaway to just do a mic check, but the answers have been so interesting I decided to ask it while we were recording. Tell us what was your first car? The make, model, and year.

If I'm not a car guy, but I do remember, it was a Mazda GLC cost about used to cost about seven eight hundred dollars and it ran about like it cost seven or eight hundred dollars. This is probably eight four ish. Um. It had a nice little stick shift in the wherever and we call those, by the way today those are millennial anti theft devices. I like that. I like that description. Um. And it was probably I was probably because I had an eighty four is already used.

It's probably like a nineteen eight. I don't I don't even know, but probably nineteen eighty or something like that or seventy nine. That's interesting. Um, tell us about some of your mentors who helped guide your career. Yeah, I'd have to say there's there's really some of some of my partners at a Katie and some of my co founders, so Gary Brookstrom. Um, you know he Uh I started my first part time job in asset management was working

with him, and uh so he was very important. We were a development stage company, so there were lots of you know, syncratic things we didn't have, like an HR department we didn't have. But but Gary was really was also really passionate about investing. Um. He's retired now, but he's still invests. So I would say Gary. My other colleague, Ron Fraser, who was a portfolio manager at Putnam before he came to join us as one of the fourth

Health co founders. Ron is a true gentleman, an investment professional. Taught me a lot about how to treat other people. UM, And so I would say that would be another another one of the folks that I learned a lot from when I first came into the business. Quite quite intriguing. Um. What about investors who influenced the way you approached the

world of investment? I think Ben Graham, you know, sort of the value part of that at Principle and again even though value hasn't been great the last ten years, UM, just the way he thought about, Um, how do you make an investment decision? Uh, you know, a lot of things came from Ben Graham. Um, I'd say he's important, Um, and then I would say there's people outside of investment, outside invest Mary who but who have lessons for investing.

So Michael Lewis, you know when he wrote back in the eighties, he wrote Liars Poker, and and uh, that book actually, even though it's not technically an investing book, it's certainly not a textbook, but has a lot of interesting information that someone who's coming into the investment industry for the first time, you know, it's a great book to read, or was certainly at the time a great book to read. So I found that that's another example of something where you can learn a lot even though

it's not technically an investment book. Speaking of books, let's talk about some of your favorite books. What do you read for fun? What do you read for work? Investing? Non investing? Fiction? Nonfiction? Yeah, so I like two days nonfiction books can be great. Um. I mentioned Michael Lewis Liars Poker. His new book is a book about actually the transfer power between them you know it. Okay, So I've read that and again that's the stories there. It's

just it's interesting because it's his way. He has this way of getting into you know, sort of getting into the detail situation learning it's enough about the milieu and talking to enough people, and then it's both humorous and as you say, har flank, but it's it's educational too. You know, you've learn a lot. So so that that would be an example of a nonfiction type of nonfiction book.

I love the way he finds these eclectic characters and the story is always unwound through these unusual people, the person from the weather channel, and it's just that's a fascinating book. And then the the personal stuff, I would say would be UM. I read some occasionally, not not it's not a huge volume nowadays, but consistently over the last years. I'll try to find some science fiction stories.

And by sense fiction I mean not um fantasy. I guess this is the aerospace engineering me not sort of the fantasy version, but the sort of hard science like the three that would be one or um Alice. Redemption. There's a Redemption Space series that I'm currently reading, Redemption Space. Who's the author there? Okay, you Alistair, So I've I've just finished the first one of the series. There's a but about a six book series, and I'm embarked on the second and I'll have to get back to you

on Alice. Uh if you could do Redemption. Uh, let's see what Google has to say about this Redemption arc by Alistair Reynolds. Reynolds, sorry Reynolds, the second book in the Revelation Space series. Revelation Space is the name of this series. Okay, absolutely, And so you're going to go sci fi a feeling very predictable that that would be an example of that's when I'm reading it's a dollar series. I think they started. I started writing those around you know,

twenty years ago or eighteen years ago. But that's an example of the kind of sort of it's a little bit harder science fiction with a lot of speculative stuff, and it's kind of fun um to just think about technology and the impact technology can have in the very long term. And I find certain types of science fiction writers. Another example would be that's a little older, would be Larry because I knew you could things exactly. Uh, those those books, the whole series of Ringworld, um from Niven.

He was amazing. Yeah. I had a great way of coming up with these ideas and then sort of making I mean, I think the quality of the of his writing over time varied a little bit, but certainly the examples like Ring World, um, A Moot in God's Eye that you co wrote with Jerry Purnell, those are examples of books that, uh, you know, there's a lot of creative thinking, and they're entertaining stories as well. They start with the framework and then the characters and the and

the plot really move along. Any other ones do you want to mention before we move on that that's a that's a really good collection. I'm a giant Larry Niven face, and I had a feeling. I had a feeling you were you were heading in that direction. Um, so what what excites you right now? What about the world of investing has you really enthusiastic looking forward to the future. I think I mentioned earlier were early days with respect to things like machine learning and big data, and I

think there's a potential for significant transformation. So you've got this historical division of quant and you know traditional or fundamental investors, where the quants go broad but maybe not that deep, and the traditional investors go very deep but they may not be quite as broad. Um. I think we're at a point where we're gonna be able to start going broad and deep because of these these kinds of both the on the data side and then the

ability to interpret the data using machine learning. You have to be very careful with machine learning. It's prone to overfitting, so you've got to build in some safeguards to avoid that. And we're still learning best practices. You know, what are what are the best techniques to use UM in driverless cars? People talk about neural nets. Neural nets you know, can easily find the best fit to historic data UM, but not always guaranteed to outperform UM just a standard linear

statistical model historic with with future data. And so I think there's a lot of opportunity there for us to to learn and do better in that area, and that that kind of stuff is very exciting both for me and it turns out when you talk to young people coming into the quantitative research area, that's that those are the kinds of things they're excited and working on. I can imagine. So tell us about a time you failed and what you learned from the experience. Yeah, I mean,

there's there's probably plenty of plenty of areas. One area would be uh in many ways. I was kind of lucky. I went to a good school. Um, I was good at taking exams. UM got a job that you know, we had the aerospace job, but the investment job, and that turned into a company. And I've been very fortunate in the people I've worked with, so I've always things

are kind of work, been kind of successful. And when it came time to go through the CEO search process, UM, you know, one of the things we did is we took these I guess you administer different kinds of not just personality exams, but they're sort of invent tories of your managerial leadership capabilities, and so what when you take one of these, they ask you to rate yourself, and then all your peers and all your colleagues at the company do the same thing and even sort of compare.

Here's where I think I am. And I'm doing this as a gesture, but I'll explain in a minute for your audience, here's where everybody else thinks I am. So it's very humbling to find out I had a very high opinion of my strategic thinking and UM, my ability to bring people to a consensus or to pull behind a decision, and UM, some of my colleagues observed that there were aspects of my decision making that, you know, they didn't appreciate as much potentially as I would have

thought they might have. Um And so that was humbling, but it was also great because you know, really hearing other people's honest feedback is something that not everybody gets easily. And this is sort of an anonymous process, so it's a little filtered, but I can sort of see, here are some areas where I actually, you know, could be doing better than I than I was. UM. So I an area where I what I think of as failed as um my self image was miscalibrated relative to where

everybody else was. On the plus side, that's a that's a learning opportunity because you can say, Okay, here's some things I could work on. I can try to do better. And if even if I can't do better, because I am who I am, maybe it's good to have an appreciation for some of my shortcomings. I love the way you phrase that, as an engineer would my self image was miscalibrated with the with the rest of the rest

of the office. That's funny. Um So, if a millennial or recent college grad came to you and said they were considering a career in either quantitative research or asset management. What sort of advice would you give them. I'd say, if you're if this is something that you're excited and you're interested in, Absolutely it can still be a tremendously exciting, rewarding career. I do think it's very different than the

environment that I faced three years ago. Right when you're entering something that's sort of new and green field, you know, there's not a lot of established players. Um, you've got a lot of opportunity. I mean, it could go completely astray, in which case you have to go to Plan B. But um, you've got a lot of opportunity. We've got a more mature industry now, there's lots of established competitors, and so it's harder to come in and have an immediate,

big impact on a firm or established investment process. Um. You know, it's going to take more work and it's gonna take some time. UM, so you've got to be prepared for that. You know, if if you're if you want to, um, you know, develop the next great idea, there's still scope to do that, but um, you're doing it within the context typically of of a bigger existing process and firm. Another area though, might be fintech. So fintech, you know, the retail investors I think still are not sure.

Fees have come down somewhat. You've got lots of index funds, you've got a t f s, but they're still not served as well in terms of the sort of advice and planning portion as they potentially could be. And so you know, some of these fintech companies, I think there there are some potentially disruptive ideas that either we are seeing or some of them may pan out, some of them may not, but that may be an interesting area as well to consider beyond pure investment management. Quite quite intriguing.

UH and our final question, what do you know about the world of quantitative investing today? You wish you knew when you were starting out thirty years ago. There's a couple of lessons. One is the importance of risk control. I mentioned um, if you're just betting on a single factor, single signal, there could be a lot of risk associated with that exposure and a portfolio you need to manage

that risk effectively. That's really important. A second thing is the payoffs to factors can change a lot over time. I think I intellectually, I think my I and my colleagues appreciated that. But there may be ways to manage those uh, the expectation of those payoffs using models that help predict how well values going work or quality or

momentum is going to work and uh. And so the importance of having such models and incorporating them into your process is something I'd love to appreciate, Say before two thousand eight, for example, quite fascinating. Thank you, John for being so generous with your time. We have been speaking with John Chisholm. He is the co CEO and former

chief investment officer for a Kadian asset Management. If you enjoyed this conversation, we'll be sure to look up an inch or down an inch on Apple iTunes or wherever final podcasts are sold and you can see the other let's call it two hundred and thirty or so such conversations we've had. We love your comments, feedback and suggestions. Please write to us at m IB podcast at Bloomberg dot net. If you enjoy this conversation, go to Apple iTunes and be sure to give us a five star rating. UH.

Check out my daily column on Bomberg dot Com. Follow me on Twitter at rid Halts. I would be remiss if I did not thank the crack staff who helps put together this conversation each week. Medina Partwana is our producer and Charles Volmer is our returning audio engineer an all time champion. Taylor Riggs is our booker slash producer. Uh Attica val Brunn is our project manager. Michael Batnick

is my head of research. I'm Barry Ridholts. You've been listening to Masters in Business on Bloomberg Radio