FidelityConnects: The quant opportunity: Fidelity’s data-driven capabilities

[00:00:06.706]

Hello, and welcome to Fidelity Connects.

[00:00:08.541]

I'm Glen Davidson. What happens when human insights meet advanced analytics?

[00:00:12.812]

That's Fidelity's Quantitative Research and Investments team, or QRI,

[00:00:16.850]

a team where data scientists, mathematicians and portfolio managers work side

[00:00:20.887]

by side all focused on one goal, finding the next great investment idea.

[00:00:25.592]

Today we're giving you a look into their world.

[00:00:27.594]

How does our next guest turn data into investment ideas?

[00:00:30.663]

And what does it mean to have this capability at a private company like

[00:00:33.500]

Fidelity? Joining me now to unpack all of this and more is QRI's Head of

[00:00:37.437]

Advanced Strategies and Research, Gilbert Haddad.

[00:00:40.073]

Gil, great to see you here.

[00:00:41.241]

Good to see you too, Glen.

[00:00:41.975]

Thank you for being here. You're actually going to Montreal after this, what's

[00:00:45.412]

going on in Montreal?

[00:00:46.146]

I am. It's amazing to be here. I'm excited to be in Montreal.

[00:00:49.282]

We do sponsor a research centre at one of the top universities over there

[00:00:53.653]

and they're organizing this hackathon. They bring students

[00:00:57.991]

from all over the country, they compete for 24 hours on building investment

[00:01:01.795]

strategies and then the top winner gets a prize of $15,000.

[00:01:05.532]

It's pretty exciting for us.

[00:01:07.200]

It's a way for us to engage with the community.

[00:01:08.968]

It's a way for us to look at top talent and then make sure we participate in

[00:01:12.939]

the local economy as well.

[00:01:14.174]

Wow, that must be fascinating and an interesting way to source some future

[00:01:18.111]

employees as well.

[00:01:19.279]

Exactly.

[00:01:19.913]

Well, I wish you a good trip there, but first let's talk about you.

[00:01:23.083]

You have a PhD in mechanical engineering.

[00:01:24.918]

I was reading some things about you that earlier in your career

[00:01:28.888]

you used to assess equipment for the lifespan and

[00:01:32.892]

so on. There was a story you told about going to West Texas and sitting there

[00:01:35.862]

trying to figure out how to analyze a pump and how long it was going to last.

[00:01:39.466]

Now we're talking about QRI.

[00:01:41.167]

Tell us a bit of your story.

[00:01:42.469]

Absolutely. I started my career, or my graduate studies

[00:01:46.773]

where the degree was in mechanical engineering.

[00:01:49.275]

It's a cross-disciplinary centre between math, stats and engineering.

[00:01:53.012]

The idea was, how do you use machine learning to forecast the life of

[00:01:57.050]

industrial equipment? It was sponsored by an industrial company to do

[00:02:01.054]

it for jet engines and wind turbines.

[00:02:04.190]

Upon graduation I got hired by an oil company, oil service company,

[00:02:08.261]

that said, well, you've been doing these models on data on jet engines, can we

[00:02:11.965]

do it on oil equipment because the cost of downtime is very high.

[00:02:16.469]

Sure enough I joined the company, it was fascinating.

[00:02:18.638]

What I realized is the models are the same, the data is slightly different and

[00:02:22.575]

then the end goal is exactly the same, you're trying to forecast or predict an

[00:02:26.179]

outcome. It was fascinating.

[00:02:28.047]

We did it for frack pumps, drill strings, all sorts

[00:02:32.152]

of oil equipment.

[00:02:34.287]

From there on I got hired by a hedge fund and then the value

[00:02:40.193]

proposition, well, our data looks somewhat similar but the models can be

[00:02:44.097]

related.

[00:02:45.231]

Did you realize that and that's how the hedge fund found out about you or did

[00:02:48.768]

the hedge fund source you?

[00:02:50.470]

The latter.

[00:02:52.572]

It was sort of at the onset of alternative data.

[00:02:56.643]

The financial community started to source all sorts of interesting data sets

[00:03:00.914]

that are what we call alternative today or non-traditional data sets

[00:03:04.951]

giving you views on companies.

[00:03:06.653]

The idea can you use models to forecast the fundamentals of companies using

[00:03:10.390]

these alternative data sets turns out to be fascinating and I found my calling.

[00:03:14.994]

Since then I've been in finance.

[00:03:17.897]

Okay, so you're heading up .... you're CIO of many strategies within

[00:03:22.268]

the QRI group.

[00:03:24.837]

Our viewers will think, hey, Peter Lynch and all these other famous managers

[00:03:28.474]

always talked about fundamental research.

[00:03:31.477]

There's technical because many of our viewers also are familiar with the chart

[00:03:34.914]

room but the quant side, has that always been sizable at Fidelity?

[00:03:40.153]

Quant has existed in some form for decades at Fidelity.

[00:03:44.157]

It wasn't up until five years ago that the firm decided to

[00:03:48.228]

build a real quant organization, or a big organization

[00:03:52.498]

to be able to compete in the marketplace.

[00:03:54.867]

They hired Neil to lead this organization.

[00:03:58.605]

Since then, we've become a giant organization, 220 researchers

[00:04:03.142]

supported by 250 quant developers.

[00:04:06.779]

We have about $390 billion of assets.

[00:04:09.515]

From zero to five years, or within the last five years,

[00:04:13.519]

we now compete with very serious quant players in the market.

[00:04:17.890]

Neil, of course, Neil Constable, who's also done some of these webcasts.

[00:04:22.228]

You looked at Fidelity as a way to connect the research that you've done in

[00:04:26.833]

an investment and economic way, if you will.

[00:04:31.104]

I guess the private ownership side of Fidelity must have aided in your decision

[00:04:34.507]

as well as far as being able to expend expenditures and so

[00:04:38.478]

on wherever they need to go.

[00:04:40.713]

Absolutely. There is the private ownership and the fundamental data.

[00:04:43.883]

We talk about Peter Lynch and the legacy of 75 years of doing fundamental

[00:04:48.121]

research of the firm that's incredibly rich in terms of information

[00:04:52.492]

being produced.

[00:04:54.327]

When we're trying to bridge quant and fundamental together it's an undertaking

[00:04:58.564]

that requires support, that requires investment and requires patience

[00:05:02.869]

from the organization.

[00:05:04.637]

I think Fidelity has been a sweet spot and an incredible for us to be able to

[00:05:08.074]

marry quantum and fundamental together for us to get to the outcomes that we

[00:05:11.744]

have today.

[00:05:12.512]

And in doing that that outcome is alpha capture, what you're trying to see.

[00:05:15.982]

Explain that, please.

[00:05:17.917]

Alpha capture, let's talk about the concept historically.

[00:05:20.453]

Alpha capture dates all the way back to the very first hedge fund.

[00:05:24.157]

The idea is ... the gentleman, Alfred Winslow Jones,

[00:05:28.261]

used to go source ideas from sell-side analysts and then

[00:05:32.231]

would get the ideas and would put them together in a portfolio.

[00:05:35.868]

The concept there is you have fundamentally driven ideas from

[00:05:40.073]

experts, and when you put them together you can strip out all sorts of

[00:05:44.610]

factors that you don't want to incorporate in the portfolio, and then you end

[00:05:47.447]

up with a better portfolio than the individual alpha streams.

[00:05:52.185]

Fast forward over a couple decades the hedge fund industry started doing that

[00:05:56.889]

internally. You have what's called multi-manager hedge funds.

[00:06:00.393]

They have multiple portfolio managers and analysts.

[00:06:03.396]

At Fidelity we have all the raw ingredients

[00:06:06.733]

for us to do something even better than that.

[00:06:09.435]

We have 130 analysts, incredible coverage, we capture data

[00:06:13.639]

throughout the life cycle of their idea generation, we have ideas from

[00:06:16.676]

portfolio managers that allow you to do best ideas portfolios, and we have the

[00:06:20.546]

giant investment in alternative data.

[00:06:22.482]

You put all of these together you have ingredients to create this alpha capture

[00:06:25.518]

strategy that's truly unique and differentiated.

[00:06:28.221]

So it's taking all of those research reports that the fundamental analysts are

[00:06:31.758]

writing every day, making them available, but you or I might physically

[00:06:35.995]

go through those, you wouldn't because you've got a connection, but I'd flip

[00:06:39.599]

through them all but your group can actually process all of that

[00:06:43.703]

fast, any time of day, 24/7 and so on, and then make

[00:06:47.840]

a conclusion from that.

[00:06:49.175]

Exactly. Let's imagine a scenario.

[00:06:51.010]

You have 2,000 companies within our coverage which is some

[00:06:55.181]

sort of liquid universe within either U.S.

[00:06:58.418]

or U.S. and Canada. In 2,000 companies you have around 50

[00:07:02.455]

companies on any given day that have a change associated with them.

[00:07:06.392]

What does it mean? A company might have a meeting with a CEO or CFO,

[00:07:10.463]

another one might have a rating change, a third one might have an earnings

[00:07:14.100]

print, a fourth one might have a price target.

[00:07:17.236]

For you, Glen, if you're reading all of these, for you to

[00:07:21.307]

read the reports on these 2,000 companies it's almost impossible, and

[00:07:26.012]

then not just read, try to understand, analyze, and then for you take an

[00:07:30.082]

action in the portfolio.

[00:07:32.118]

It's almost possible. For a systematic strategy it allows you to capture

[00:07:36.255]

all that incremental information in a very efficacious fashion and then you

[00:07:40.226]

put it in a portfolio and capture that alpha.

[00:07:42.161]

This must be something, as you said, it's been the last five years as things

[00:07:45.898]

have really started to grow in QRI, and the sky must be the limit

[00:07:50.269]

because as technology, as AI, everything, is really becoming much more

[00:07:54.140]

prevalent but the future is going to be bright as well.

[00:07:58.344]

Absolutely. The technology has evolved dramatically but I think the

[00:08:02.515]

constant is change.

[00:08:04.550]

Things are always evolving, that's a constant. The other constant that

[00:08:08.521]

I have is we have data that's proprietary and differentiated that no one else

[00:08:11.958]

has. That's going to set us apart over the next 3 and 5 and 10 years

[00:08:16.262]

to come. Even if technology is changing it's going to allow us to

[00:08:20.500]

exploit the information set in a much, much better fashion.

[00:08:23.369]

Okay, you've intrigued me, you've intrigued our viewers because you just said

[00:08:26.405]

you have access to data that others don't have.

[00:08:28.708]

Now, it's nothing that we shouldn't have but you have the ability to get.

[00:08:32.678]

We better talk about that.

[00:08:35.181]

Let's talk about three sources of information, analyst data, portfolio manager

[00:08:39.118]

data and alternative data.

[00:08:41.821]

The first two are proprietary.

[00:08:43.656]

They exist only within the four walls of fidelity.

[00:08:45.658]

The third one is differentiated and we'll talk why.

[00:08:48.628]

Analyst, a company has done this amazing job at collecting the data

[00:08:52.598]

throughout the life cycle of the idea generation.

[00:08:55.735]

If you're an analyst you go meet with the company CEO or CFO.

[00:08:58.738]

We know who you met with, when you met with them and denoted right after the

[00:09:02.808]

meeting.

[00:09:03.075]

Do they document that and they lock it in the system?

[00:09:05.378]

Exactly. It's one system for everybody and it's the same format.

[00:09:08.981]

Then you go update your financial model in the company.

[00:09:11.817]

It's captured the EPS estimate or the forecast and the KPIs, all of them are

[00:09:15.755]

captured, pipe into database.

[00:09:17.924]

Then you can write the note on the security.

[00:09:19.692]

It has a rating from a strong buy to a strong sell.

[00:09:21.994]

It has a price target. It has a flag about if it's a disruptor or not, a

[00:09:26.933]

variety of other flags, and it has the long term thesis on the security,

[00:09:31.103]

and it has your updated views on what you've learned from your research on the

[00:09:34.640]

security. All of it is captured.

[00:09:36.943]

Then you go update your paper portfolio.

[00:09:39.545]

We capture that information.

[00:09:41.547]

The best part, that information is captured for more than 20 years.

[00:09:47.019]

If you ask me why, it's because data is in the firm's DNA.

[00:09:51.390]

It's just an incredible data asset.

[00:09:53.893]

The first thing you look, is the data efficacious?

[00:09:56.295]

The answer is absolutely yes. It's being used by our amazing portfolio managers

[00:10:00.399]

for years and years and years and they use it to generate alpha.

[00:10:03.469]

Now, imagine what you can do by testing ideas going back for 20 years

[00:10:07.573]

and measure that efficacy year in and year out.

[00:10:10.409]

That's the first data set.

[00:10:12.612]

The second data set is the PM data.

[00:10:14.947]

We have more than 50 portfolio managers, different stylistic ways of trading.

[00:10:18.117]

They have low turnover and long tenure.

[00:10:22.221]

It allows us to create the wisdom of the crowd.

[00:10:25.091]

On any given day we can answer the question, what are Fidelity's best

[00:10:29.061]

ideas in terms of large-cap growth ideas?

[00:10:33.599]

Very easy to answer. It allows you to source the ideas from the best portfolio

[00:10:37.269]

managers out there, put them in a signal.

[00:10:40.439]

The third one is the alternative data.

[00:10:43.776]

Alternative data, as we said, it's non-market data.

[00:10:46.145]

Think about it, credit card, email receipt, satellite imagery, geolocation,

[00:10:49.949]

these are data sets that are available on the open market.

[00:10:54.320]

Now, what it takes is the ability to spend a lot of money

[00:10:58.257]

to acquire these data sets and having teams dedicated for

[00:11:02.461]

them to ingest, process and create insights.

[00:11:05.131]

It's a substantial investment.

[00:11:07.433]

Over the last five years  we've increased our spend significantly

[00:11:11.671]

there. We've increased our resources because the firm believes in

[00:11:15.875]

the value proposition of data in general.

[00:11:17.877]

That type of data is cell phone usage, aggregated, of course, not someone's

[00:11:21.814]

individual, I hope, and credit card usage and so on.

[00:11:26.018]

You can derive a lot of information from that at a particular vendor or

[00:11:29.455]

geolocation I guess it's called.

[00:11:30.856]

Exactly. We cannot use any PII or MMPI.

[00:11:34.460]

All of these data sets you can...

[00:11:35.761]

What's MMPI again?

[00:11:36.996]

Material non-public information.

[00:11:39.565]

I should have known that.

[00:11:41.267]

Or personally identifiable information for PII. We cannot use that in general.

[00:11:43.502]

There is very strict controls and processes around it but you can answer some

[00:11:47.239]

very interesting questions.

[00:11:51.110]

Right now we're in the fall season, a

[00:11:55.681]

number of vendors have these promotions about pumpkin spice latte on Tim

[00:11:59.552]

Hortons. If you would like to answer the question about what's

[00:12:03.989]

the sales from pumpkin spice lattes at Tim Hortons in a specific location,

[00:12:07.860]

that's a question you can answer. Well, how would you do that?

[00:12:11.230]

You can geofence a particular store.

[00:12:13.666]

You have the data for cell phones going into the store.

[00:12:16.936]

Then, of course, not tied to individuals, they're anonymized.

[00:12:20.372]

Then you have email receipts telling you

[00:12:24.477]

about who's buying and what they're buying and then their credit

[00:12:28.581]

card purchases. You can measure that before the season of pumpkin spice lattes

[00:12:32.384]

and there's ways for you to get that.

[00:12:34.553]

It allows you to answer questions that are very, very granular and detailed and

[00:12:38.791]

allows you to get an information edge.

[00:12:42.328]

That's tremendous. That's the alt data, the component that we spent significant

[00:12:46.065]

resources. We have these three data sources and they allow us

[00:12:50.035]

to create these signals. The signal is really a fundamental thesis that you

[00:12:54.006]

can test with data. That's all it is.

[00:12:56.375]

Then you put it in the portfolio.

[00:12:58.177]

That's just one small example, and we appreciate your Canadian example, by the

[00:13:01.714]

way, but it's interesting as consumers to know how we're being watched

[00:13:05.718]

and aggregated by where we are, by what we're spending and where

[00:13:09.655]

we're spending and so on. If I'm a ...

[00:13:12.525]

hypothetical ... if I am a portfolio manager at Fidelity I'm really in the

[00:13:15.661]

middle of a Venn diagram, aren't I, where I have access to fundamental

[00:13:18.497]

research, technical research, I can use all the charts that are online or even

[00:13:22.434]

walk up to the chart room, but I also have this QRI group that's giving me

[00:13:26.472]

this level of data that, really, other companies probably don't have.

[00:13:33.279]

You've got portfolio managers that must be all-in on what QRI provides

[00:13:37.683]

and then there's probably a few portfolio managers that say, I'm still trying

[00:13:40.953]

to figure this out but once in a while I'll need a data point.

[00:13:44.056]

Is that true?

[00:13:45.090]

It's absolutely true. Just like any technology there's an adoption curve.

[00:13:48.861]

You have the early adopters and then it follows the different four or five

[00:13:52.164]

stages. What we've seen, the important thing for us that we

[00:13:56.202]

gauge is we want to increase our footprint with the fundamental investors

[00:14:00.239]

by 20% year on year. What does it mean?

[00:14:03.342]

Next year I want to make sure I talk to 20% more portfolio managers and

[00:14:07.112]

analysts and increase engagement with existing ones

[00:14:11.250]

by 20%.

[00:14:13.018]

That's something we've done for the last four or five years and we want to make

[00:14:15.487]

sure to continue because there's true edge in that data and we want to

[00:14:19.458]

make sure that we can disseminate it to the entire firm and make sure

[00:14:23.629]

that everybody benefits from it.

[00:14:25.097]

There is edge in the data so let's look at Fidelity from a competitive

[00:14:28.667]

advantage standpoint. The private ownership no doubt helps.

[00:14:31.937]

The size of the team that you're part of and the education that everybody has

[00:14:35.040]

no doubt helps. Is there commonality on the street or is this quite

[00:14:39.178]

exclusive?

[00:14:40.045]

This is very exclusive. The analyst data, the PM data are quite exclusive

[00:14:44.016]

because definitionally they exist only within the four walls.

[00:14:47.786]

For any company to have that massive of an equity research department is very,

[00:14:51.924]

very uncommon. There's maybe one or two players out there that have something

[00:14:54.793]

similar but nothing of that scale.

[00:14:56.896]

Alternative data, for a firm to be able to spend that much resources

[00:15:00.833]

and then head count is very, very unique.

[00:15:03.569]

This has been championed by hedge funds for the last 5

[00:15:07.840]

and 10 years. Fidelity has come in and we've come in very, very strong.

[00:15:11.911]

It's a race to the bottom, meaning that the more you can spend  the more

[00:15:15.714]

differentiated the data sets that you will get then the higher the alpha

[00:15:19.151]

potential that you have.

[00:15:21.921]

The players that fit within that category are very, very few.

[00:15:26.558]

Hello, investors. We'll be back to the show in just a moment.

[00:15:29.762]

I wanted to share that here at Fidelity, we value your opinion.

[00:15:33.332]

Please take a few minutes to help us shape the future of Fidelity Connects

[00:15:35.634]

podcasts. Complete our listener survey by visiting fidelity.ca/survey,

[00:15:40.739]

and you could win one of our branded tumblers.

[00:15:42.708]

Periodic draws ending by March 30th, 2026.

[00:15:45.778]

And don't forget to listen to Fidelity Connects, the Upside, and French

[00:15:49.381]

DialoguesFidelity podcasts available on Apple, Spotify, YouTube, or wherever

[00:15:53.419]

else you get your podcasts. Now back to today's show.

[00:15:57.623]

With the emotions in this market, the volatility that we see every day, I don't

[00:16:03.329]

know what the market's doing today but as you see, it's up and down all over

[00:16:05.931]

the place. This must add discipline to

[00:16:10.135]

invest.

[00:16:11.537]

Absolutely. The alt data is giving you views on the

[00:16:15.507]

fundamentals. Maybe let's break it down in terms of a concrete example.

[00:16:19.845]

If I'm looking at the revenue from a particular restaurant,

[00:16:24.717]

pick your favourite burger restaurant, if what you care about

[00:16:28.854]

is the year on year sales growth the data has

[00:16:32.891]

given you exactly how many burgers are being sold on any given day, you

[00:16:37.262]

can reconstruct that index and then you can measure year on year.

[00:16:41.066]

It has no views on what tariffs are going, in which direction, in which

[00:16:44.370]

countries, it's just telling you what are the sales today.

[00:16:47.172]

Core data.

[00:16:48.073]

Exactly, and it's telling you about the fundamentals of that particular

[00:16:51.143]

company, no emotions whatsoever.

[00:16:53.112]

You can use that information to inform your investment if you're

[00:16:57.149]

a portfolio manager.

[00:16:58.751]

Where does this not work?

[00:17:02.454]

Just like any data set it has its own biases.

[00:17:06.091]

When we talk about credit card data you don't get data for every single person

[00:17:09.561]

in the country. You get a very, very small sample in the small

[00:17:13.632]

single-digit per cents.

[00:17:16.402]

The first thing you do is you measure, is that representative enough of the

[00:17:20.105]

entire population? Once you get comfort you do that.

[00:17:23.208]

Now, of course, representative enough, it means that sometimes you're going to

[00:17:26.345]

have errors so that's one case where it won't

[00:17:30.482]

work. Other cases, where you get delays in terms of the data.

[00:17:34.153]

A third case is when the stock is trading on something that's completely

[00:17:36.922]

different than what you're modelling.

[00:17:39.291]

Imagine that the company is introducing a promotion and

[00:17:44.430]

the stock is trading on how efficacious that promotion is.

[00:17:47.299]

If you're measuring top line revenue and sales growth it doesn't really matter.

[00:17:51.270]

Even if you have an absolutely accurate forecast it really doesn't matter.

[00:17:56.075]

Companies come to see Fidelity all the time on the 14th floor at 245 Summer

[00:17:59.611]

Street in Boston, as they do here in Toronto.

[00:18:02.815]

Did they ever come in and say, what kind of research do you have on us?

[00:18:06.518]

Fidelity is asking them a lot of questions but they must be fascinated by the

[00:18:10.389]

amount that they could ...

[00:18:12.324]

and I don't know if you tell them or not but did they ever ask that?

[00:18:16.295]

Good question. I haven't seen it yet but also I'm not present--

[00:18:21.767]

They may not be aware.

[00:18:23.001]

--on all these meetings, or they may not be aware, too.

[00:18:24.970]

What I do know is the same level of data, it gets sold for

[00:18:29.241]

Fortune 500 companies through other channels.

[00:18:32.711]

There's a lot of companies that are interested in their competitive advantage

[00:18:35.347]

and how they're doing in the marketplace so they go procure these data sets

[00:18:38.484]

from the same companies that we procure data from.

[00:18:43.288]

The question I asked you about where doesn't it work, it made me

[00:18:47.259]

think that perhaps that's also creating a learning because sometimes maybe

[00:18:51.830]

the data doesn't work properly or there's something that was inefficient about

[00:18:55.534]

it but then that causes a restructuring and it's

[00:18:59.605]

better going forward. This is just like an EV car, sometimes things don't

[00:19:03.742]

work probably, they refine from there and they move forward.

[00:19:06.512]

It's probably something that's evolving by the minute as far as what resources

[00:19:10.782]

that you have and the abilities that you have.

[00:19:12.818]

Where do you see this going in the next few years?

[00:19:16.388]

I think today having access to that data has become

[00:19:20.592]

table stakes. For you to be able to compete on the same alpha as the

[00:19:24.530]

biggest player out there you need to have it.

[00:19:26.765]

Going into the next 5 and 10 years you have to use it slightly differently,

[00:19:30.736]

or you have to get data that very few people have.

[00:19:34.239]

These are the two ways for you to generate the alpha, either you have data that

[00:19:37.743]

no one else has or use it differently.

[00:19:39.878]

I think for us spending the cycles to understand the biases and then how do we

[00:19:43.782]

use that data better and update our models and augment it with new datasets

[00:19:47.719]

is the way I see the future coming for the next 35 years.

[00:19:51.723]

I hope this is a fair question.  Has there been a slice of data that's come

[00:19:55.594]

across your desk where you thought, I had no idea we could get to that level?

[00:19:59.198]

Has there anything really fascinated you that you can talk to us about?

[00:20:04.536]

There's nothing that we can get access to that we can use

[00:20:08.473]

that can be MMPI or PII.

[00:20:10.475]

Let me reiterate that.

[00:20:11.910]

There's a level that it can go to.

[00:20:13.779]

Yeah. For example, even if you can get 50%

[00:20:17.950]

of a specific credit card company

[00:20:22.621]

you cannot use that. The compliance department will not allow you to use it,

[00:20:26.191]

and then that's a legal liability so we never come even close to that.

[00:20:30.462]

More recently we've seen an incredible evolution in terms

[00:20:34.566]

of different vendors putting their

[00:20:38.570]

data on cloud infrastructure and platforms.

[00:20:41.573]

Then they're realizing that not only they can migrate their data to a cloud

[00:20:45.777]

they can start monetizing their data.

[00:20:48.547]

Monetizing your data is that cloud infrastructure connecting us to them.

[00:20:53.185]

Then having that conversation, hey, are you interested in selling your data?

[00:20:58.323]

Three times out of 10 I talk to these companies and they have no idea that they

[00:21:02.027]

can sell their data.

[00:21:03.528]

And then they have to set a price.

[00:21:05.063]

Exactly. We're talking about small-time retailers in Europe,

[00:21:09.835]

or grocers in some other geomarket where in their mind they never thought

[00:21:14.206]

that somebody is going to care about the specific sales

[00:21:18.143]

on their shelves in a

[00:21:22.114]

specific market.

[00:21:24.249]

For us it's incredibly insightful.

[00:21:27.286]

We use it to inform investment decisions for portfolio managers within their

[00:21:31.189]

respective geomarket.

[00:21:32.424]

But really interesting to find out that there's that data out there that

[00:21:35.260]

companies don't even know that they can monetize, as you mentioned.

[00:21:37.696]

You're creating an awareness there and, obviously, Fidelity's got kind of first

[00:21:41.333]

dibs at that for the right price.

[00:21:43.101]

Exactly.

[00:21:43.635]

Now, we've talked about this from an equity standpoint, can you talk about it

[00:21:46.271]

from a fixed income standpoint? This is just as useful from a credit analyst

[00:21:50.509]

side.

[00:21:50.976]

Absolutely. When you look at the cap structure of the company it has the equity

[00:21:54.880]

side, it has a credit side.

[00:21:57.716]

The research, being a company that's so privately owned,

[00:22:02.954]

a collaborative setup that has this giant investment in research and data,

[00:22:07.259]

it allows you to get a view from both sides of the cap

[00:22:11.530]

structure. The same data can be used to inform the bonds as well.

[00:22:15.600]

We have many, many examples

[00:22:19.571]

but in the recent five years, especially

[00:22:24.810]

throughout the COVID period, you have these companies where bonds sold off

[00:22:28.580]

because of a COVID shock.

[00:22:30.682]

But then you can go look at the sales for this particular company, retailer or

[00:22:34.353]

cruise line or whatever company you're looking at, and then you form a view.

[00:22:38.623]

Is the bond setting up for the right reason or not?

[00:22:41.660]

You can use that for the credit side as well.

[00:22:45.764]

We do have proprietary data within Fidelity that we've been experimenting with,

[00:22:49.368]

for example, is migration data, is when people

[00:22:53.405]

change their address on the website you know

[00:22:57.442]

they went from this address to the other address.

[00:22:59.411]

Now, you can imagine a world where that could be relevant for the muni

[00:23:03.548]

desk because you understand the demographics of people going from

[00:23:07.686]

this particular zip code to that zip code and then that has direct impact on

[00:23:11.123]

the tax revenue for municipality.

[00:23:16.428]

There's no limit to how much insights you can get from

[00:23:20.365]

this data.

[00:23:20.932]

I can tell you have a PhD in engineering because you talk about all this with a

[00:23:23.869]

big smile. You love this, it's just constantly coming at you.

[00:23:27.139]

You did not come up here from Boston to this studio in Toronto to tell us that

[00:23:31.176]

quantitative is the way to go, fundamental is lousy and don't go there.

[00:23:34.713]

You actually came here to tell us about the complementary nature, is what I'm

[00:23:37.783]

deriving from you.

[00:23:39.551]

Absolutely, absolutely. I'm here to tell you that and always that's

[00:23:43.622]

my pitch to everybody. Quant plus fundamental is an incredible

[00:23:48.059]

amalgamation of information for us to derive alpha for the future.

[00:23:52.664]

It's durable, it's scalable, it's differentiated and Fidelity is uniquely

[00:23:56.368]

positioned for us to bring these two together.

[00:23:58.870]

Our viewers are probably saying, wow, you've got all these resources at

[00:24:02.207]

Fidelity for the fundamental analysts but you actually also have a fund,

[00:24:06.711]

Advanced U.S. Equity, I'd like to talk about which came out in January which

[00:24:10.415]

marries up fundamental and quantitative.

[00:24:13.018]

We should talk about that because it's important our viewers understand you're

[00:24:16.888]

there as a resource, QRI, quantitative research

[00:24:21.193]

and investing. There are funds that are deriving information

[00:24:25.363]

from the QRI side that are run by the QRI side.

[00:24:27.933]

Let's talk about U.S. Advanced equity.

[00:24:30.202]

Absolutely. Advanced U.S. Equities as a fund has

[00:24:34.539]

the same spirit, how do you take fundamental insights from the firm and then

[00:24:38.310]

package them together in a systematic fashion for you to remove any biases that

[00:24:42.447]

are specific to a style.

[00:24:44.850]

The way we look at it is we take all the research data on any given day,

[00:24:49.087]

we create these signals and then we put them together in what we

[00:24:53.124]

call a portfolio that has high idio risk.

[00:24:55.961]

High idio risk, it means there is no style factor, there is no growth,

[00:25:00.165]

value or momentum bias and there's no sector bet.

[00:25:03.835]

It's the purest representation of what...

[00:25:06.171]

They negate emotions and biases, as you said.

[00:25:09.441]

That's a great way of putting it.

[00:25:11.443]

Then you put all of these ideas together in an unbiased fashion that

[00:25:15.614]

allows you to extract the best ideas from fundamental research on any given

[00:25:18.817]

day.

[00:25:20.519]

The way it's constructed it selects the highest conviction

[00:25:24.456]

ideas. It has anywhere between 50 to 60, 70 names on any given day.

[00:25:29.394]

It does well when the fundamentals are doing very well.

[00:25:32.430]

It's going to struggle when the fundamentals are struggling.

[00:25:35.400]

In the genesis this is a way for you to get access to the best fundamental

[00:25:38.870]

research from Fidelity in a product.

[00:25:42.674]

Very interesting. I want

[00:25:46.611]

to go back to what you were talking about as far as going to Montreal for the

[00:25:49.881]

hackathon. Could you talk about of the few hundred people that are

[00:25:53.852]

part of QRI what kind of education and backgrounds

[00:25:57.989]

do you typically look for?

[00:25:59.925]

We typically look for people that are excited about the

[00:26:03.929]

markets, that are eager to work with data

[00:26:07.866]

and models, that are comfortable with working with problems that do not have a

[00:26:11.570]

textbook solution.

[00:26:13.405]

Generally speaking, these are coming from STEM backgrounds,

[00:26:18.009]

science, technology, engineering and math backgrounds.

[00:26:20.979]

Some of them go for MBAs, some of them go for CFA's or charters.

[00:26:26.151]

Generally speaking, these are STEM backgrounds that are coming to join us on

[00:26:29.621]

this journey for us to achieve the outcome that we hope to achieve.

[00:26:33.024]

A lot of them probably, you said there's got to be an interest in the markets

[00:26:36.628]

but many of them probably don't even think about that but then they meet

[00:26:39.030]

someone like yourself and say, I can take what I've got from

[00:26:43.001]

a STEM background and actually put that into the investment world.

[00:26:46.071]

There's a need. That's really interesting.

[00:26:49.641]

Absolutely. If you're trained as an engineer unless somebody tells you or makes

[00:26:53.645]

it an intentional decision to go take courses in finance you don't get

[00:26:57.749]

exposed to that entire world.

[00:26:59.884]

It's very, very exciting because you get to work with real data, real problems

[00:27:04.122]

that have direct impact on investors and shareholders.

[00:27:06.925]

It is very exciting.

[00:27:08.126]

Many universities try to encourage their engineering students to do humanities

[00:27:12.297]

so that they can actually write and there's a creative side.

[00:27:16.034]

It's interesting what you just said, they should also encourage, and maybe they

[00:27:19.037]

do now, a direction towards finance.

[00:27:22.607]

I'm curious how many actually do that but that's another story.

[00:27:26.077]

We're wrapping up in a minute or so, what would you like our viewers to take

[00:27:29.881]

away from our discussion today?

[00:27:31.816]

I think we live in very, very exciting times.

[00:27:35.420]

The marriage of quant and fundamental I certainly believe, and

[00:27:39.591]

highly convicted, that's going to be the future of investing.

[00:27:42.594]

Fidelity is uniquely positioned for us to capture and be a top player in

[00:27:46.698]

that space.

[00:27:47.799]

Wonderful. Thank you for a succinct summary as well.

[00:27:50.035]

So nice to see you and safe travels to Montreal.

[00:27:51.636]

Thank you so much, Glen.

[00:27:52.470]

Good to see to you, Gil.

[00:27:53.872]

Thanks for watching or listening to the Fidelity Connects

[00:27:57.809]

podcast. Now if you haven't done so already, please subscribe to Fidelity

[00:28:01.946]

Connects on your podcast platform of choice.

[00:28:04.749]

And if you like what you're hearing, please leave a review or a five-star

[00:28:07.585]

rating. Fidelity Mutual Funds and ETFs are available by working with

[00:28:11.556]

a financial advisor or through an online brokerage account.

[00:28:14.926]

Visit fidelity.ca/howtobuy for more information.

[00:28:18.630]

While on Fidelity.ca, you can also find more information on future live

[00:28:22.467]

webcasts. And don't forget to follow Fidelity Canada on YouTube, LinkedIn,

[00:28:26.604]

and Instagram.

[00:28:28.606]

We'll end today's show with a short disclaimer.

[00:28:31.476]

The views and opinions expressed on this podcast are those of the participants,

[00:28:35.313]

and do not necessarily reflect those of Fidelity Investments Canada ULC or

[00:28:39.250]

its affiliates. This podcast is for informational purposes only, and should not

[00:28:43.254]

be construed as investment, tax, or legal advice.

[00:28:45.790]

It is not an offer to sell or buy.

[00:28:48.093]

Or an endorsement, recommendation, or sponsorship of any entity or securities

[00:28:52.430]

cited. Read a fund's prospectus before investing, funds are not guaranteed.

[00:28:57.235]

Their values change frequently, and past performance may not be repeated.

[00:29:00.805]

Fees, expenses, and commissions are all associated with fund investments.

[00:29:04.642]

Thanks again. We'll see you next time.