Jack Shannon:  Unveiling the Future of Active Management and ETFs - podcast episode cover

Jack Shannon: Unveiling the Future of Active Management and ETFs

Mar 08, 202458 minEp. 194
--:--
--:--
Listen in podcast apps:

Episode description

In this episode, we are joined by Jack Shannon, Senior Analyst at Morningstar, to discuss the dynamics of active equity management, portfolio construction, and the importance of downside protection in long-term investment success. Jack shares his insights on various topics, including:

Topics Discussed

• The asymmetry of returns and the importance of preserving capital on the

downside

•The role of active equity managers and the challenges they face in deviating

from benchmarks

•The impact of fund flows on portfolio management and the potential risks of

passive investing

•The role of risk management in active investing and the difficulty of

identifying winning strategies in advance

•The changing dynamics of growth and value investing, and the importance of

understanding market history


This episode is a must-listen for anyone interested in active equity management,

portfolio construction, and long-term investing. Jack's insights provide

valuable strategies to navigate the complexities of the financial markets and

better understand the nuances of active management.

Transcript

Jack Shannon

A key contributor to long term success is being able to preserve that capital on the downside because, you know, we all know that sort of returns are asymmetric and, you know, a loss hurts a lot more than an equivalent gain. So, then equivalent gain helps. So I think that's where investors can, can really help themselves is looking at the downside, even though most people like to think about the upside, you're buying an equity fund, you want, you want to get those returns.

you don't really care about, well, it only lost 20 percent when the market lost 30. You're more focused. Most people are more focused on sort of the promise of the big gains. So I think people need to kind of just shift their thinking in terms of, of what to really look for.

Adam Butler

Welcome. We've got, Jack Shannon here today from, uh, from Morningstar. Jack is Senior Analyst for Morningstar Research Services and. Jack came to our attention through our mutual acquaintance, Brian Moriarty, who shared one of your more recent research pieces, I guess, on hit rates in large cap U. S. equity mutual funds and Mike and I both read it and thought, wow, this would be a perfect guy to have on the show. So I'm glad we could put the pieces together. Welcome.

Jack Shannon

Yeah. Thank you. I hope I live up to being the perfect guy.

Adam Butler

Well, there's all kinds of perfect guys. So, um, you'll be a perfect guy for the show. I have no

Jack Shannon

Yeah, hopefully.

Adam Butler

so maybe tell us what you do at Morningstar. how do you occupy your time? What interesting projects do you like to work on? What are you working on right now?

Jack Shannon

Yeah. so I, um, I'm in the manager research group at Morningstar. so what, what we do on a day to day basis is we, we primarily cover, individual strategies and individual investment firms. And so, you know, we meet with portfolio managers, we Question them on a, you know, a wide range of topics related to how they invest money, how they manage their team, how they manage their time, and all these things.

And ultimately, we sort of issue, forward looking ratings on investment strategies and firms and whether we think people should invest their hard earned money in these, in these strategies. So, so I focus on, active equity for the most part. So, you know, I've, I've had an interest in, in markets for a long time, but like markets in a very general sense of just like, you know, how much information is, is, is embedded in, in a single price is kind of amazing to me.

And so in this day to day job, why that's sort of relevant is, you know, you have portfolio managers and analysts who, are fundamentally saying. I think I have better information than the thousands and millions of participants out there, which is a very interesting, and bold, uh, stance to take. And so a lot of my research is sort of going down that, that road of exploring, You know, how good managers really are at having an informational edge.

So, so that's sort of the, the, non core stuff I do on a day to day basis, but that, that occupies a lot of the, the research time.

Adam Butler

So let's start at kind of a higher level. What is the goal of an active equity manager? would be the success criteria for an active equity manager?

Jack Shannon

so an active equity manager wants, wants to beat a, a benchmark, now they usually get to name their own benchmark, which is sort of the, uhhh, you know I was reading a book from Mark, Mark Spitznagel's Safe Haven where he described sort of the fund management business as being sort of a, you know, a self fulfilling game where it's like you get to set the rules and set your own performance standards and all that.

And it's true to an extent, but active managers ultimately want to beat a sort of a free passive benchmark, because investors could take their money invested in a, you know, a passive market ETF, for, 15 basis points, but instead you're going to say, well, if you, if you give me 70 basis points, I will give you an index beating return. and obviously the sort of the track record, over time is not very good for active managers.

and so, you know, a big part of the the hit rate piece that I, that I just published is sort of trying to get into, other than fees, sort of why have active managers not been better than just a passive, a passive alternative? Is it because they simply cannot pick winners at the same frequency as you get in a passive portfolio? Or is it something else? Is it that they're just not good at sizing and identifying their winners? in an appropriate manner.

So, so yeah, so the goal is simply to, to beat a passive benchmark, but oftentimes, especially in like the international categories and whatnot, there really isn't even an investable passive option. You know, if you're a foreign large growth manager, there's really, there's really not a, clean, um, You know, actual investable benchmark for uhhh for everyday investors. So, so oftentimes it's a benchmark that really an investor can't even access. So, a lot of different angles there, I guess.

But, um, yeah,

Adam Butler

mean, I asked a very broad question that was on purpose, right? But, what I, what I'm trying to get at, too, is. Do managers typically act like that's their objective to, I mean, and when we say beat the benchmark, do we mean generate a positive information ratio relative to their benchmark, generate higher absolute performance? How does risk factor in to that?

Jack Shannon

I, uhh so it depends on which shop you talk to and which managers. I think most, You know, I, a lot of, a lot do want to beat on risk adjusted metrics. So they want to, you know, have a sharp ratio, um, above the bench. but sharp ratio is, is, you know, adjusting for the volatility of the returns. But I think that's kind of a, I don't know. I think, I think claiming victory because you had a lower return than the index, but lower volatility is, is sort of a low bar.

so to me, you know, I, if I want to pay an active manager, I want to pay them to, to beat the benchmark. I don't want to pay them to, give me a high five and say, Hey, I didn't beat the benchmark, but at least I gave you more steady returns. because I could have gotten the benchmark at. You know, a third of the cost, uh, so you do have some managers who wanted to want to add, you know, add value on, on pure, Um, sort of risk adjusted metrics.

But, you know, when I'm evaluating a manager, I like to look at just sort of pure excess return potential because that's, you know, you can't eat alpha, you can eat. Excess return now.

Adam Butler

We can debate that later. In terms of like a large cap manager, it seems to me like if the goal is to, like, like, like, Beat the cap weighted benchmark, for example, then they would cleave very closely to that benchmark and only deviate from it on names where they felt that they had an edge. Is that typically what you observe or is, you know, do we, do we mostly see?

Large cap managers who have fairly heterogeneous portfolios, they have, uh, you know, more concentrated basket of stocks that they follow and that they, have conviction in it's probably smaller than the S& P 500, for example. And they deviate quite substantially from the cap weighted weights.

Jack Shannon

you know, you're just sort of hitting on sort of the closet indexers that are out there, and there certainly are a lot of them who, you know, if you're a portfolio manager, you're ultimately getting compensated. Um, well, it depends on the shop, obviously, but a lot of it is, you know, deviation from, from the benchmark.

So, There, there could be an incentive there for a lot of managers to not want to take overly bold positioning because if you do you're sort of risking your own personal comp, and and I think that is a, certainly is a factor for, for some managers. And how they manage things.

But a lot of a lot of active managers want or sort of now marketing marketing themselves as specifically, you know, we're not like the benchmark, you know, if you want the, you know, the big tech stake with, you know, NVIDIA and Apple and Microsoft, you can get that in a passive fund. But we want to offer something different. And so if we're going to offer something different, we kind of have to take these bolder, um, Bolder positions relative to a market cap weighted portfolio.

so you see, you know, especially managers who have underperformed, uh, in the last five years because they were already underweight, those names, sort of a repositioning of, well, you know, we're, we're purposely underway. We want to be different and we want to generate sort of a, a, a different return stream than you would get in a, in a, in a market cap portfolio.

Adam Butler

But, I mean, do you think that, uh, Just how rational is it for an active manager with a large cap equity benchmark to deviate materially from that benchmark? I mean, to me it seems like the Like kind of the optimal approach is to have a portfolio where you hold, you know, effectively all the stocks in the S& P 500, in S& P 500 weights, and kind of proxy run a long short portfolio over top that effectively what that's going to run is. You know, you're going to have some names.

You've got conviction, more conviction, long in other names. You've got more conviction, short in, and a bunch of names that you probably don't have, you know, reason to have a materially different view than what is expressed in the benchmark weights.

And so, you know, it seems like it seems reasonable to me as well, that if an investor actually has a preference to not deviate too much from the benchmark, however you want to mention, um, measure that, let's say in tracking error, then the managers should be accommodating that objective in that way, rather than saying, well, you know, I want to deliver the best return I can for the lowest amount of risk, which to me seems like a reasonable objective for an equity manager, but you know, you end

up with that, with these potential misalignments between what a, an allocator is allocating to you for versus what your true objective is as a manager, right? I just think this gets conflated a lot. How do you see managers reconciling that?

Jack Shannon

I mean, I agree that that's a, um, that's sort of a worthwhile approach. You just don't really see it. you might see it in, um, you know, there are, quant funds or, or, you know, research funds where it's like a bunch of analysts putting stocks in, into a portfolio.

you might see the case where they're, they'll have, you know, these structured rules in place where they'll say, you know we'll, we have our sort of our, our, our risk budget, and we don't want any given stock to add X percent of our risk. So maybe their fundamental analysts hate Tesla. I think it's overvalued and think NVIDIA is overvalued, but they're going to own it in the portfolio just for pure risk management purposes.

so you see that a lot with quants and again, like these like broad research type portfolios, but there are some managers who, I'm thinking of Um, I'm blanking on the man, the, manager at Fidelity runs an international fund. but his approach is, to basically use the, market weight as just the, sort of the, anchor point. And then, will go plus or minus, you know, two, three percent one way or another. but it doesn't, it doesn't hold like the whole, the whole, uh, universe in the portfolio.

It's still a subset of stocks. but if it's like a, you know, if it, if it was a large cap domestic portfolio, it'd be like, well, maybe he's underweight Apple by three percent, but overweight Alphabet by Three, you know, just, we do see managers who do anchor to that market weight, but it's not while holding just the whole wide portfolio, the whole wide index that is.

Adam Butler

Yeah, I'm just trying to think about what what a rational manager should do given a client who has an objective. I mean, if a client says, look, I'm measuring you against the benchmark, then the client has said to the manager, Your risk or I'm measuring you on risk, which is tracking error to the benchmark, right? So from the perspective of. A large cap manager who's been, you know, told I'm allocating you because I want you to beat the benchmark.

The implicit statement in that is, the risk I'm measuring you on is your deviation for the benchmark. You're tracking your risk. And it could very well be that, that, you know, a lot of managers say, look, I've got no edge in the mag seven. They're followed by every Every research shop out there, they're massively liquid. They're as efficiently priced as any stock, and I'm just not gonna take an active view on that.

I'm gonna hold 'em all in equal weight or, you know, in, in their respective cap weights. But there's, you know, a hundred stocks near, closer to the bottom of the cap stack. that are less followed, where I bring some expertise and where I can take, you know, be underweight and overweight, 60 to 80 of those stocks. And that is how I'm going to try and generate my long term alpha, right? That manager would presumably have Very low active share, right?

So on that benchmark, then because he's only playing with, market, with, uh, stocks in the benchmark that have already at very low weights, those deviations from those weights don't matter much from an active share standpoint, but he could have a very large edge, right? So he might end up having a very high information ratio as an example, right? But that's, that doesn't seem to be the way that. That managers are evaluated and, why do you think that is?

Jack Shannon

yeah, I think it gets down to sort of the question of the allocators and what they're looking for. I think a lot of times when we talk with, with portfolio managers, the, the sense that we get is that especially on, you know, the institutional side, you know, they're wanted for sort of a more concentrated Uh, I guess more unique portfolio. so like, yeah, you know, I feel like allocators can get beta plus pretty easily.

And so when, when they're looking at these more active managers, I, it, it seems to me that there's more of an appetite for, more active share. I'm not sure that that necessarily means higher expected return or, or anything like that, but I think that's generally what, what managers are looking for or allocators are looking for. and so we can discuss whether that's a good thing to look for or not, but I think that's part of the reason why you see it.

Adam Butler

So, how would, uh, how would most people get access to beta and alpha separation in the way that you're sort of describing most allocators would look for?

Jack Shannon

Um, what do you mean in terms of, in terms of the institutional clients?

Adam Butler

Well, no, I mean, in terms of, I mean, institutional clients can, right. Because they can get their, their betas and they can buy market neutral funds or, or what have you and, and use leverage, et cetera. Right? But it, it was my understanding that most. Retail investors typically, they have few options, but to kind of bundle their alpha and beta exposures together, right?

Jack Shannon

Yeah. Yeah. That's generally, yeah. I think that's generally true. and you think about where all retirement assets are, they're all, at least in the U S they're, you know, pretty concentrated in, in just sort of, target date funds where you've got your equity and your your fixed income and all that in, in one sort of fund. and, it's all usually passive. So it's all kind of, it's all beta basically.

and I think at least in the U. S. part of that is, you know, with the, with the history of, of active performance against, against passive and that it has done poorly.

If, if you have sort of a fiduciary standard, it's, I think people are a bit gun shy, Of, of recommending or putting active funds in, in a lot of sort of everyday people's portfolios, simply because of sort of fear that, you know, maybe you're violating a fiduciary standard by recommending something that in the aggregate has, has struggled against a, a cheaper alternative.

Adam Butler

How do you factor regimes into your thinking? So, for example, regimes where, equal weight portfolios outperform cap weighted portfolios tend to favor active management versus, you know, obviously the last 10 years, with the possible exception of 2023, obviously profoundly favored, cap weighted, uh, portfolios. How do you try to factor that into your analysis and, and recommendations?

Jack Shannon

That's a big piece of it. So every time we have a ratings committee, um, which is once a year, we update our ratings on, on every strategy we cover. and that's a big piece of the discussion is, you know, how, how do we contextualize a performance? Based on sort of how the market has behaved and whatever, given trailing five, trailing three, trailing 10 years. because that is, you know, it, it, there are good managers who can look, who can look, who can look pretty bad for a while.

and there are, there are, there could be bad managers who look good for, for, uh, for a while. so as long as we can sort of try to at least, uh, and that's why the The conversations with the portfolio managers so are so crucial, cause you get to sort of figure out how they truly think about portfolio construction and, and, and think about, um, risk management and all that. so through those conversations, we can get a sense of like, you know, were they purposely making.

You know, purposely positioning this portfolio so that it could succeed in this environment. Or is this sort of just a lucky byproduct of, of sort of a grab baggy philosophy type thing? so yeah, it's, it's, it's a big piece of, of what we do on a day to day, day to day basis.

Adam Butler

And where do you land? I mean, given that these regimes can last for a really long time.

Jack Shannon

it depends. So, um, you know, I was actually just looking at a, um. We don't, we don't actually cover the fund, but it's a, it's a super concentrated strategy that has, you know, it's been around since like 2015 and it's produced a great record, over, over that timeframe, it beat the Russell 1000 growth, beat the S& P 500. And it did that without owning any of the Mag 7. and so I was interested in like, okay, how, how could this have happened? so I built a you know, a simulator that would.

Bundle together, you know, 500 random 20 stock portfolios since they're a, they own 20 stocks. They keep them pretty much equal weight. so let's put together 500 random portfolios back in 2015 and see how they would have done. and then compare that to the actual results of the, of the strategy and the strategy ends up still being like the very top of the sort of the random distribution too. So that, that leads to sort of multiple, more questions though.

It's like, well, is it, is this evidence of skill, uh, because they've sort of beaten the, even sort of the most random of, of scenarios. Or is it the opposite that this is possibly the luckiest? Portfolio, we could have created. So I think it's, I don't think we ever sort of firmly land one way or another on, on a given regime.

but I think we, we do a lot of analysis like that to just sort of keep the conversation, you know, interesting and making sure that we're, we're sort of looking at it from all different sorts of angles.

Mike Philbrick

In a, in a 20 stock portfolio, yeah, that can get sort of. Very unlucky and very lucky, I suppose. And, and was there any indication in that particular example of what the turnover was? Was It you know, the idea of, of you've got

Jack Shannon

was about 0%.

Mike Philbrick

you beat those to death and,

Adam Butler

Zero percent, you said?

Jack Shannon

The, the annual

Adam Butler

Wow.

Jack Shannon

well, for 2023 was 0%. Yeah. They, they didn't, yeah.

Mike Philbrick

Interesting.

Adam Butler

Yeah. that's remarkable.

Jack Shannon

Yeah. And, but then to your, to your question though, it was, um, it was very, it's, it Was remarkably even, like contributions from the stocks. It wasn't that they had, you know, one huge winner that carried them. It was each stock like was, it wasn't, you know, it wasn't a, you know, a clean, you know, even just split, but it was pretty close, which was, which was, what was the sort of amazing part.

And so that's why I had to generate this sort of random scenario to be like, how, how sort of crazy, what is this? and Yeah, that, that, that's certainly an outlier. That

Adam Butler

how much do you guys use random portfolios? Always, I've been a huge fan of random portfolios as a way to evaluate skill in quantitative strategies, but also in mutual fund performance, etc. Ever since I read Patrick Burns papers back in, like, The mid 2010s. is that something you guys employ quite prolifically at Morningstar? I mean, it just seems like it makes so much sense as, as a key dimension to understanding manager skill.

Jack Shannon

I wouldn't say it's, it's something we use like super, super frequently. I, I have like a I have, I have some, coding skills and whatnot. So I, I like to build a lot of these sort of quantitative, Quantitative models and whatnot to do some of this analysis. you know, i, I find them extremely useful because I, to your point, it's, it's to me, that's the best, that's the best way to sort of assess whether or not this was a, true, true skill on display or, or, or really, really lucky.

And. You know, I think the more, I think we should do it more. I think the, you know, my goal is eventually, again, that was like the one we just talked about was a pretty simple one of just taking 20 random stocks and the S& P 500 and just letting them sort of run. but you could apply it to sort of any sort of any manager.

So if there's a manager who is, you know, You know, a quality growth manager and, you know, I only like companies with above median ROICs and I want to see the valuation below the median in the industry and I want to have this target turnover percentage. You know, you should be able to build a pretty, Pretty robust sort of random portfolio, um, simulation from that.

And just sort of see is if we know the generally what the manager likes to do, can we attribute the performance, whether good or bad to, You know, skill within that, that sort of narrow niche, or just that, that narrow niche ended up doing pretty well. And that's maybe more of the story than, you know, just sort of, uh, a raw look at it might,

Adam Butler

Well, then actually that raises a key question, right? So you've got a manager who uses a set of screens with, you know, a set of constraints and a target average number of stocks with whatever, different weightings. If that manager exhibits skill versus an unconditional group of random portfolios, but doesn't exhibit skill is kind of right in the middle of what you'd expect for any manager that uses the same selected criteria.

How will you judge whether that selection criteria itself adds value over the longterm? like how do you disentangle that skill versus.

Jack Shannon

Yeah, that's the That's the tricky part. because yeah, you should get credit for if, if you're investing in these sort of higher quality companies and lower valuations and it does well, well, that's a good thing that you, that you did that, right? and that is an active decision.

so, I think that, yeah, I think, you know, determining whether, you know, what gets more credit, um, I think is hard, I guess with the way, the way I would think about it is every manager tells you they invest in high quality companies. Every manager tells you they, you know, they're buying, you know, companies that they think are trading below intrinsic value. Every manager, uh, you know, wants to buy, companies with good sentiment.

Like, I think if we can, if we know the exact sort of screens they run, that makes it a bit easier, because then we can tell within the, within their own defined universe how they're doing. But, um, because if you're not picking stocks well in your own defined universe, I'm not sure I have confidence in you to, to continue to, and you outperformed. I have confidence in you to continue to do that unless I have confidence in these factors continuing to, to do well.

Which we generally try not to, take too, uh, too out of a stance on like a given factor or, or given style. But, Yeah, it's certainly tricky, but I think, um, I said, I'm a, I'm a, I'm a big fan of this sort of the random simulations. so yeah, hopefully we, we, we do it more and hopefully people find it useful.

Adam Butler

well, how do you think about this? The style style box? You guys still incorporate that in your, in your recommendations? Yeah.

Jack Shannon

We still do.

Adam Butler

is the thinking evolved on that over time?

Jack Shannon

Yeah. So all, all our ratings are sort of category dependent. So wherever the fund falls in its, in its category, we sort of stack each, we stack the, you know, we stack the deck depending on where it falls in its own category. So we don't take a stance on, you know, large growth is a better place to be than large value. We just say within large value, here's The good stuff. And within large growth, here's the good stuff.

are, there is a project underway now, um, with the style box, just because if you, if you looked at the style box recently, a core, you know, S and P 500 fund looks like it's in growth water now. so there's sort of the issue of like, Do we want the style box to be totally relative to, you know, if you want the Russell 1000 to be the core, the core, uh, benchmark, do we want the Russell 1000 to land dead center in the style box?

Or do you want to take maybe a more absolute approach and say, well, valuations have gone up across the market and maybe the Russell 1000 growth going in, Russell 1000 going into the side of the, you know, fence? Maybe that is a signal to somebody that, that, Hey, if you want pure core exposure, maybe you look a little further left on the style box.

So, I don't think we've, I don't think we've landed anywhere on sort of, on any changes in that respect, but that is something we've, we've seen recently is just sort of growth inflation, pulling everything to the, to the right of the style box,

Adam Butler

Yeah, it's just a question of whether a manager, you know, should get credit for choosing growthy stocks when growth, the prospects for growthy stocks is, is better and choosing value stocks from the prospects. It's just, you know, I remember my early days working in an asset management shop and you go through a 3 year period where, you The value guys would hide in their offices and all the growth guys would congregate at the water cooler, high five in one another.

And then you'd go through the next three years where, it would be, you know, the reverse, The value guys would be out of the water cooler and the growth guys would be hiding in the back office. And neither group had any influence or, you know, did, they weren't able to dictate at all the conditions of their own success. It's just, is their style in favor or not? Right. It just seems. It just seems so absurd.

Jack Shannon

Yeah. Yeah. It's funny you mentioned that. Cause, uh, you know, when I first started here, value, value was, was beaten down for, for a decade plus. And every time I talked to a value manager, you know, they're, it's almost, it was almost like an apologetic tone from them. And they're, you know, they're trying to say, look, you know, just, just give us a little more time, you know, things are going to turn around, blah, blah, blah, blah, blah.

And the growth managers are always, you know, they're beaming with confidence and, you know, so sure of themselves. And then, you know, 2022 comes around and, you know, growth managers Get um, Get hammered and value looks relatively good, even though they still didn't do well on an absolute basis.

But, but then you see the role is reversed where it's, you know, the growth managers are sort of apologizing and saying, yeah, you know, we got, we got a little ahead of ourselves on, on these and, and, you know, we learned our lesson and then the value managers finally get to say, see, we told you so. You know, we, we told you that the value would, would bounce back. so yeah, we, we, it's, it's interesting from my seat, just sort of seeing the, the psychology of, of.

Portfolio managers and having to deal with those, those ups and downs. Cause I think that is, that is a big sort of behavioral piece that they have to manage is that you can look, again, like I said before, you can look dumb for a while, and yet, you know, I have to have the fortitude to, to stick it out.

And on the flip side, you know, you can look very smart and very bright and make a lot of money in a short period of time, but you kind of have to understand that, you know, success can be fleeting and, that can, that can turn on a dime.

Adam Butler

Yes it's just a strange, there's a bunch of strange dimensions too, but one of them is that growth environments tend to be highly concentrated in a few stocks and value environments tend to be, have much broader. Participation in the markets.

And so for a kind of multidisciplinary manager, so someone say, who doesn't particularly favor growth or value as say a core manager, part of his job is going to be, you know, emphasizing growth or value, I guess, over time, but in those growthy markets, the amount of conviction you have to have in those. Few names that completely dominate the, the cap stack in the index and also often end up also dominating the return profile.

Like you sort of saw, in the recent period and in this sort of pre 21 period for five years, obviously leading up to 2000 as well. the amount of conviction you've got to have, like as a fundamental, equity core manager in your allocations. Is so out of proportion with what anyone could reasonably expect to have as their edge. For, for you, for a bottom up analyst.

To want to give 7 percent each to, to Apple and Microsoft and, you know, 4 percent to Nvidia, et cetera, like the amount of excess return that you need to assume for those, for those stocks to give those stocks, those weights in the portfolio. Like there's just no responsible way that you could, from a bottom up standpoint, be that confident, right? So. How do you, how do you think managers should navigate that paradox?

Jack Shannon

It's also, uh, in the U S it's tricky because they, um, oftentimes can't even hold it at market weight if they want it to. So there's the, diversification rules in a 40 act fund that, you know, the 25 5 rule, you can't have more than 25 percent of your assets in positions of 5 percent or more. And you literally can't own, if you're a large growth manager, you can't own, the top five stocks at market weight.

So you If you're a diversified manager, so you either have to change the diversification status of your fund, which I haven't got a great answer on why that's so difficult, but, a lot of managers don't seem to want to do it. I guess it has to go through a, you know, a shareholder vote and all that. And maybe there's a, some operational headaches and

Adam Butler

Massive platform

Jack Shannon

Non diversified funds, but yeah, so regardless, you know, they run into that, that, that problem, and like I said, I think a lot of them now are just like, There's a tendency of managers to you know, explain themselves as being unique when they're, when they have to be unique. Like, it's not necessarily a, I'm not sure it's an active choice they're making.

It's, it is a regulatory constraint that they have to be, And so, yeah, it would be, it would be interesting if, if you've removed that 25 5 rule, where their portfolio weights would go, and I suspect they would go up in those, in those seven names in the aggregate. there certainly are some managers who I think are legitimately, you know, have legitimate bearish views on a lot of those names.

but to our earlier discussion on, sort of tracking error and managers sort of being incentivized to stay positive. Somewhat close to the index. I think that that is a huge amount of risks they're taking on that they, that again, they don't really have a decision on. and so I think if they could close that risk gap, they certainly would like to.

Adam Butler

It's an interesting, point you make though, Mike, I hadn't thought of this about the fact that a lot of these active funds are actually not able to hold the MAG 7 and the, weights that the cap weighted index holds them in. so there's actually a, a structural compliance or regulatory disadvantage that these funds are facing at the moment. Is that a fair characteristic or characterization?

Jack Shannon

Yeah, they, yeah. So the 40 acts diversification rule is a 25 five rule, and I don't have the The exact stats on like the percentage of non diversified versus diversified funds in like the U. S. large cap universe, but it's very few are non diversified. Um, and the ones that are, have been non diversified for a long time, because they were sort of coming out of the nineties as these concentrated portfolios that were popular at the time.

And then saw their popularity wane and, now they've kind of come back in vogue. You have seen some, some shops, I think Fidelity and TRO have, have recently changed some funds from non diversified to diversified, or I'm sorry, from diversified to non diversified to sort of address this issue, but it hasn't been as widespread as, as, we thought. you know, cause on our side, it's like, again, we're not, you know, we're not on, you know, we're not overseeing any platform.

We're, we're just issuing our own opinions on things. And so we were always wondering, you know, it sort of seems like a free option, right? To. To just label yourself non diversified. And if you want to hold a diversified portfolio, great. There's no, there's nothing stopping you from, from doing that. but you are constrained on the other side where if you, if you are diversified, but you want to be non diversified, you, you can't.

but Adam, to your point on, on sort of the platform, issues, I, we've long suspected that's the case, but we haven't really, we just haven't really gotten enough, you know, firm information on, on that and sort of how big of a constraint it truly is.

Adam Butler

Yeah, fair. So, with all your data wrangling, do you have any, uh, suggestions on what investors might look for? You know, how first of all, what is the, what is the sort of future of Of active management in your, in your view and, and, um, how should investors think about the role of active management and portfolios versus other options? and, if they were going to choose to pursue active management, what are some good rules of thumb?

Jack Shannon

Yeah. So the, the future of active management I was, um, as I was reading a story in a Wall Street Journal today about a manager complaining about the passive flows, um, warping the market and, you know, creating all sorts of issues and contributing it to it being top heavy and all this. and to me, I see that and I say, wouldn't an active manager love that?

Wouldn't an active manager love uninformed, participants adding into, you know, coming into the market because if you're the, the informed investor with, You know, an informational edge, you should, you should like that. You should, you should be able to then see where the market's overheated and where it's not, and be able to act accordingly.

so if you talk to managers that, you know, they feel threatened by, by passives and not just from a, from a flow standpoint and from an asset standpoint, but they, a lot of them sort of characterize it as sort of, you know, changing the market overall. But I'm not sure I, I buy that argument.

we've seen, um, Uh, at our Morningstar conference, I'm hosting a panel on, um, On active ETFs, which is sort of the, I guess that the vehicle of the future for active management where, instead of in a mutual fund wrapper, you're getting it in an, in an ETF wrapper. There are obviously some tax advantages to that. you know, the, the portfolio managers, you know, there, there was always a question of sort of transparency with an ETF. You have to disclose your portfolios every day.

some portfolio managers didn't like the idea of that and thought that, hey, maybe people could get out in front and front run me and sort of hurt my hurt my performance that way. but I think a lot of that sort of fallen by the wayside and we, we see a lot of shops launching back ETFs. and I think there's, there's certainly gonna be more to, more to come. they're sort of specifically trying to keep them as distinct strategies for the most part, which is interesting.

So, you know, if you're a, if you have a large cap growth strategy, in a mutual fund, a lot of times we're seeing. firms launch a large cap growth ETF. That's slightly different than the mutual fund version. And I, again, getting back to that sort of fiduciary point we hit on earlier. I think a lot of that has is the reason behind that is if you have an ETF version, that's cheaper than. a mutual fund version, then you're probably just going to cannibalize the flows between them.

so, so that's where we're seeing managers go right now, just in terms of vehicles. And then in terms of just everyday investors and sort of how they can, sort of better their odds with active management. to me, it's all about, it's all about quality and it's all about downside. I think if, if you can find a manager who is consistently, investing in high quality businesses, and I know quality is a very nebulous term, we can, Talk about the different definitions for that or whatever.

But, generally I think, I think you, you can do well. There are, there are certain shops that lend themselves to being high quality firms that invest in high quality businesses. They do it for the long term. You know, they're not churning portfolios 200 percent in a year. and you see that in the downside protection. So if you, if you go on morningstar.

com or I don't know what other, sort of outlets out there that, that publish sort of downside metrics, but I think that's a key, A key contributor to long term success is being able to preserve that capital on the downside because, you know, we all know that sort of returns are asymmetric and, uh, you know, a loss hurts a lot more than an equivalent gain.

So, then equivalent gain helps, So I think that's where investors can, can really help themselves is looking at the downside, even though most people like to think about the upside, you're buying an equity fund, you want, you want to get those returns. you don't really care about, well, it only lost 20 percent when the market lost 30. You're more focused. Most people are more focused on sort of the promise of the big gains.

So I think people need to kind of just shift their thinking in terms of, of what to really look for.

Adam Butler

Right. In terms of the flows, sorry, Mike, you look like you had a question. Um,

Mike Philbrick

No, I was just, I was just thinking through, you know, win rate doesn't really matter or isn't that effective in certain ways. Um, big bets aren't really that effective in ways. So is it, is it really the, it's the risk management? Is there, so there's a risk management, a downside. Is there any of the picking on the upside that adds value or like, how, how do you, how have you, can you elaborate more on

Jack Shannon

Yeah. So, so I did the research on the, on the hit rates and the big bets to see if, how can active managers win? Can they, can they win via a higher percentage of, of stocks being right? Answers? Eh, not really. Can't really say that's the case. you see a, you know, a loose linear relationship, but, um, in some categories it's, it's, you know, not really linear at all. It's just like a random scatterplot. Yeah. Same on the bets.

You'd think, well, I want an active manager to make these concentrated bets, you know, where they have the informational edge. And you're like, ah, you see that, and you're like, yeah, it doesn't necessarily help out too much either. so on the, so yeah, so it's, it comes down to just sort of consistency and, you know, I, I, think, I think, on the, on the upside, you do see funds, um, where over their whole track record, you look at 25, 30 year horizon.

where they're not good on the downside, but they're, they're good enough. They're so good on the upside that, that, that it, that it makes up for it. so there are some cases of that happening, but it's, it's, it's just sort of rare to find them.

and I think because if, if you are, if you are worse on the downside, while people don't pay attention to it, when maybe they're buying their fund at first, and those moments of pain and you're losing a lot more than, passive index, You know, people are going to start pulling money from your, from your fund. And I think it's, it's harder to survive when you have the, worst downside profile, but the higher upside profile.

Mike Philbrick

Right, the, the fund flows beat you up a little bit, but if you're, if you're an investor, that's I'm going to dollar cost average. And I love it. I love being down

Adam Butler

Uh, Uh,

Mike Philbrick

I love all I have all, and I love being down and I'm good. A dollar cost average is probably a better

Jack Shannon

Yeah. I mean, there are, there certainly are some, some, some growth funds that have, yeah, you look at their profile, like. Damn, how did this, how has this survived and how, but, but it's, you know, they have sticky client base who are willing to stick through those sort of turbulent times and, and, they do get the payoff eventually. they're just, yeah, they're, they're, they're rare.

Mike Philbrick

Yeah it's super hard to obviously identify those in advance as the one. I mean, this really sticks out in my mind too, is it's not a five year game. Like the people talk about your time horizon and investing being three to five years. There's all, it's, it's just randomness in three to five years, like literally 20 years, you can reliably count on stocks beating most other asset classes historically. And that's what you got to think. I'm making a 20 year decision here.

And if you can make it 30 years, boy, boy, it gets even better. but wow, what a, what a decision to try and make with a certain, call it dogma or religion or market cap value

Jack Shannon

Yeah. I've looked at, Um, so that's something I've, I've, nothing's been published on it, but it's something I've been really interested in is like, Looking at any sort of trailing risk metric, beta, standard deviation, down, you know, maximum drawdown, blah, blah, blah, blah, blah.

So just to think if I was, if I was an investor at any point in time, and I looked at all these risk metrics that are commonly, pumped out for in financial, websites and whatnot, would that tell me anything about the long run potential of the of the strategy? you know, would those high beta funds,

Adam Butler

Well,

Jack Shannon

know, end

Adam Butler

I mean, we

Jack Shannon

better or worse, whatever. And so, it's something I've done a lot of work on and it ends up, depends what time period you look at, but like those trailing risk metrics are not very great for when you, when you go over those long horizons, you know, 20 years, they're not going to tell you much in terms of either the downside or the upside. It's, it is just sort of, but to your point on the sort of the three and five years, you know, you go on Morningstar. com or you go on Fidelity or wherever.

The beta it's going to show you is going to be like a trailing three year beta, which is going to be on based on monthly, most likely monthly returns. So that's 36 data points that you're going to base your, your decision on. it's just, sometimes it's hard for me to think that those are super, super useful for an investor thinking of, you know, truly long term risk.

Mike Philbrick

Well, it's interesting though. If you, if you're in accumulation and you're doing that dollar cost averaging, you're, probably going to have some tailwind to that. although decumulation, you know, works in the exact opposite

Adam Butler

And I think you need to, it needs to be relatively evenly spaced out. You know, having a, having something where it kind of goes up for eight years and then gives you a 60 percent decline for two years, it doesn't really help the dollar cost average, right?

Mike Philbrick

of course. There's going to be, there's going to be this certain, the fingerprint of it that would be advantageous or disadvantageous over time. And I'm being, I'm doing a thought experiment. Let's be honest. The fact is that the mental capital that's required for someone to continue to invest in something that is, you know, the Nasdaq in 2000 to 2003 and continue to just she money in.

I mean, I guess you have seen a little bit of that in the, in the space of cryptocurrencies where there is a group of people who are just in it no matter what. They're, they're yolo-ing no matter what. And, and maybe there was a few people in the tech world like that. but boy, oh boy, that's, that's a hard road to follow behaviorally from, uh, from an investor's perspective, going through some 90 percent declines.

Adam Butler

definitely. I was just saying, Jack, that, you know, there's good evidence that that lower volatility stocks and lower volatility portfolios provide a tailwind, over time, but obviously that's far from a sure thing. And there's, there's going to be, good, long stretches where lower volatility portfolios underperform. So,

Jack Shannon

I think, um, so betting against beta, there is, I think AQR wrote a, wrote a, paper. on it. actually that was my, uh, so when we interview for Morningstar, we have to, we have to pitch a fund and do that as part of the interview process. And mine was AQR defensive style, which is based on the, uh, betting against beta paper, which is about low vol stocks over the long run.

Adam Butler

And quality. Yeah, awesome. Yeah, that's a, that's a good paper. And then the guys at Robico have done some fabulous work on that as well. We've published some on it. There's definitely something there, but, you know, again, you're going to go through 3, 5 year periods where those are going to look pretty silly too, relative to cap weighted indices. We're in the middle of

Mike Philbrick

we're five value investors. anyone?

Adam Butler

Oh, yeah, not value, value 10 to 15 to 30 years, but depending on the metric, but

Mike Philbrick

It's a great time to be accumulating.

Adam Butler

that's right. That's

Mike Philbrick

this, this is not, if you're a value investor, this is the, this is the, the heaven, the nirvana that you look for in order to be able to just accumulating these value stocks over and over again at these

Adam Butler

That's right. 20, 20 years of accumulation with no payoff.

Jack Shannon

Lower that cost basis.

Mike Philbrick

it's a rough game, man. It is a rough game.

Adam Butler

But that's, you know, to your point about the flows, the passive flows, right? And I, you know, I, we've had discussions on this before, but, you know, others have raised this idea that, well, if there's all these. passive flows that are kind of no edge flows, then why wouldn't active managers be all over this and trying to orbit? And I think the reality is that you're, you may be right at some point, right? Like you're sort of betting that you're going to be right at some point.

That the passive flows are going to stop. And, but as long as passive flows continue into stocks with disregard for, you know, any anchoring to fundamental, metrics, then you're going to remain offside and your degree of offsidedness is going to grow over time, right? So it's one of these, the market can stay irrational for many, many, many more years, and you can stay solvent or stay in business as an active manager. I think that's the. I think that's a fair point.

Mike Philbrick

and potentially, you know, when you're fighting the dumb crowd, sometimes the dumb crowd's pretty smart.

Jack Shannon

Yeah.

Mike Philbrick

know, I mean, I've never seen a company like Meta grow at the size it is at 25 percent last year. That's astounding.

Adam Butler

Yeah.

Mike Philbrick

I just don't know you. I'm, I'm gobsmacked. I mean, this is the, this, this set of companies is not the 2000 set of companies. it's a very different set of companies. So it is, and the, you know, the whole software eats the world and, and the change in the way you can lever these businesses. I think you're right. There's a lot of people fading that trade until they're all out of business. They've sold everything. They're underweight as late as they can be. and just didn't make it to the end

Jack Shannon

I actually did a,

Mike Philbrick

and maybe they are right. I don't know

Jack Shannon

I did a study comparing those two periods just in terms of how, managers were trading during this sort of tech bubble and the, what I call the COVID bubble. the idea of the, the the idea in my head was that, oh, managers must have learned from that. Like, you know, there's, there must be enough institutional knowledge at a firm where they don't get sucked into buying, you know, ridiculously valued companies that had never earned a single dollar in their history.

And of course it ended up being slightly worse than the tech bubble, just in terms of buying activity. Like I'll have to look at the numbers, but something like four and a half, percent of all large growth assets at one point were in companies that had never earned a profit. and that's with historically well below like one percent, like a tenth of a percent is a usual amount.

and so you see this big spike in 2000, You know, late 90s, 2000, then you see nothing, nothing, nothing, big spike in 2020, 21, and then nothing since. Um, that's, that's an interesting, angle that I, again, I want to explore. I don't know how, how exactly to do it, but just, um, sort of market memory and, and sort of how much, you know, managers, managers or, or, any participant in the market really has, you know, actually internalized history. Yeah.

Mike Philbrick

the collective memory. Cause we, we had the nifty 50 as well, prior to the tech bubble. Then we had the reason reason tech bubble followed right along the resource boom. You know, we had the bricks and all, all the rush into Canada and Australia and South Africa. I mean, I don't know. I don't know. I think it's, it's a feature.

Adam Butler

think it's that the man, the managers who have those long memories. Will not participate in those markets and get left behind. And so it's, you know, they just get run out of the business, right? Like too much experience in this business actually ends up being detrimental because you just, you know, you keep looking at this crazy market saying. You know, it's, this is ridiculous. I'm not going to participate this time. I learned my lesson last time and it goes on and on and on.

And eventually you just get hopped and someone with that will participate, um, replaces you and has a good run for two or three years. And then they learned the hard lessons and it all just repeats.

Jack Shannon

Yeah,

Mike Philbrick

Yeah, there's got to be a collective memory though that, you know, sort of the, the collective memory of the participants of the marketplace, um, with interest rates, even, I mean, I was in the transitionary period of that going from a period where, when I grew up into 1981, 82 was rising interest rates and nothing built. And I bought a Canada savings bond that paid 19. 5%. That was a pretty good deal. And then watched it go from 19 and a half to zero.

And, uh, you know, I remember an old, an old sod of a broker who would get his old paper out from 1981 and ask the clients, what, what, here, here's the newspaper from 20 years ago, what do you want to buy? And he kept it, they'd always be like stocks. And he's like, no, it's just long duration bond over here. You see this 30 year bond yields, 18 percent or yields. It wasn't 18%. It was about 15%.

You know, 30 year bond for 15 percent and rates went, you know, straight down, but totally outperformed stocks, both risk adjusted and absolute. So even if you're given all the information, investors will make the wrong decision.

Adam Butler

Yeah, I don't think there's any institutional memory lasting longer than a couple of years at all, just because the people that for

Mike Philbrick

a couple of years, but there's definitely, you know, there, there's, there's certainly a youngening, I think, going on at the moment. You know, there was, uh, in the advisor space anyway, there's, and even in the institutional space, you're seeing. More and more of a, you know, the, um, the boomer crowd being sort of moved on for the most part. and, uh, millennials moving into that Gen X is, is in there a bit, but, um, you know, the youngening is a real thing. So that, that collective

Jack Shannon

I've asked firms about how, um,

Mike Philbrick

I mean, the number of times I heard.

Jack Shannon

Yeah. I've asked some, I've asked some mutual fund companies how they handle that because, you know, they're, a lot of their analysts with 10 years of experience, they graduated college in 2012 and they didn't see anything but growth, growth, growth, growth, growth, and it worked, worked, worked, worked, worked. Um, And then again, 2022 was a disaster.

And, you know, a lot of them acknowledge that we, yeah, we actually have to spend time to, you know, educate the analysts on sort of market history overall, and that, you know, the more enterprising, I guess, analysts are naturally interested in, in market history, I think. like I, I like to just read it on my own anyway.

But if, if you, if you, if you do treat it just simply as a day job and not something you're naturally curious about, it is something you can overlook and get burned by, I feel like.

Adam Butler

I hear. Yeah. But too much knowledge of history can also be detrimental for for very, very long stretches of time. so it's,

Mike Philbrick

I also think there's nothing, there's nothing like living it. I could read it. I could read it. I could read it. And until you've got rubber under the road and chinks in your armor and stab wounds and scars.

Adam Butler

I don't even think there's any motivation to read it. If you read it, but you kind of just skip over the parts where, you know, they were rough. You know, value oriented grinds because you're having so much fun in the current growth environment. I mean, I remember, I remember those times. I remember feeling exactly the same way and you know, you read the histories if you're forced to, but really you're not paying attention because it just is not salient to you at the moment.

Mike Philbrick

even, even, I mean, I think I've seen the same game run written about extensively in reminiscence of a stock operator in the stove, whatever it was. I've seen that fricking gamut run through four different marketplaces. I've seen the whole thing done. Like I, I just, I, I see it over and over again. I'm like, wow. We, they just, they, they, you know, doll it up a little bit, but anyway, I, you know, there's, there's, uh, there's basic, boring investing.

Uh, I think, I think you were, you were talking about that, Jack, you know, just kind of basic, boring stuff kind of wins in the longterm trotting along. And, uh, and, and maybe, maybe

Jack Shannon

it's it's no fun at a cookout when, you know,

Mike Philbrick

we've been chatting for about an hour, but, but I wonder what is boring. Yeah.

Adam Butler

He was Mike wasn't being, dismissive of

Mike Philbrick

no. I mean, but what, what are the key, what are the key features of boring of sort of boring that, that you would, that I know you alluded to that in the papers, you know, this is boring. It's just boring. It's kind of this, but what maybe, maybe as a final parting gift, what are the key features of boring that tend to work and just are, you know, things that people can look for?

Jack Shannon

It's the low turnover sort of buy and hold approach. Um, you know, we, we talked about sort of those portfolio simulations that you can, you can run. And if, again, if you look at five years or three years, it's not going to tell you much because of market cap. Weighted portfolio would have beaten anyway, but go back and start in 1995 and pick any random selection of, of companies and just sort of let the market take them and it's, it's the most boring thing you could possibly do. Right.

Set it and forget it for? 30 years. but it Turns out that doesn't do that bad. And, in fact can do better than most of the active stuff that you, that you, you end up paying for. so yeah, I, I think just the low turnover buy and hold, you know, don't look at your, your, your, your balances every week, don't look at them every month, maybe check in once a year. Uh, make sure that your portfolio manager didn't commit fraud or something, but, Other than that, it's, it's, it's a really long game.

and, and I'm saying this and I'm not that old. Um, so I, I kinda, I, I kinda recognize the, uh, the inherent sort of contradiction and, you know, a young person telling somebody that, but, that's how I sort of view things. I don't, I don't check any of my, any of my balances, any, any. Real frequency at all. and who knows, I'll check it in a couple of years, make sure nothing, nothing's too off the rails. But,, yeah, that's

Mike Philbrick

Benign neglect.

Jack Shannon

right. Yeah. Yeah.

Mike Philbrick

The key to investment success in a world that's overhyped and over, marketed is benign neglect. Is to just ignore all of it.

Adam Butler

as long as the initial portfolio wasn't informed by whatever mania was in place at that

Jack Shannon

100%. Yeah. Yeah. cause there's a fund in, uh, in America, it's called like, Voya Corporate Leaders Trust, um, was put together in 1935 as a static portfolio. Hasn't traded a thing since. And over that, 1935, that might not be the right date, but over its entire history, it's beaten the S& P 500 and hasn't traded a share.

Adam Butler

It's amazing that all the companies are still in

Jack Shannon

Well, not all of them are.

Adam Butler

Maybe they all

Jack Shannon

So it's, it's a big big Union Pacific stake in it right now. because, you know, some companies have, you know, been, you know, bought out, some have, some have died. And so You end up then getting, you know, market weightings within this small portfolio. and so, yeah, it's, it's become a very concentrated portfolio, but it's still beaten. The broader market over, you know, almost a hundred years. So there's something to it.

Adam Butler

yeah, sure. Well, um, are you working, working on any particular project right now? You wanna give a teaser for.

Jack Shannon

Um, nothing, nothing that's, you know, going to get published anytime. Immediately soon. Um, I, I mentioned sort of the risk stuff I've been, I've been looking at of just, um, how informative are sort of trailing risk metrics in terms of forward risk. you know, the classic betas and standard deviations and, and, all that jazz that, That we all see when you pull up a fund on a, you know, on a screen.

But, yeah, that's a, that's a, that's a more, you know, this hit rate stuff and the big bet stuff is very straightforward when you're getting into the world of risk, there's so many sort of other angles that people, people want to explore that makes it a bit bigger of a project, but, Hopefully that one sees the light of day. but if not, you know, everything sort of comes to me through, uh, you.

know, like talking with portfolio managers and that they'll say something and I'll say, like this hit rate stuff that came up because a manager said he had hit on 70 percent of his stock picks. And I thought that was a ridiculous number. And so I said, I want to see how rare that, that is. And I, and I did The study and there was one manager who had 70 percent or more and it wasn't, this guy.

Um, so it's, uh, yeah, it's, I just, I, I like being here and having, having sort of a unlimited, uh, access to our database to just sort of explore what's ever top of mind at The time. So something will pop in my head and you'll, you'll see it at some point, I suppose,

Adam Butler

I hear you. The best articles are motivated by someone who says something that you think is silly or that you strongly disagree with.

Jack Shannon

right. I got to prove that guy wrong.

Adam Butler

Yeah. 100 percent yes. Completely concur. Okay. Well, this is great. A little over an hour, um, covered a lot of ground. Jack, uh, where can people find you other than at Morningstar? In addition to Morningstar.

Mike Philbrick

On the streets of Chicago.

Jack Shannon

you can bump into me on the South side of Chicago. That's where I live. But, uh, no, other than that, um, I don't really have a social media presence or anything, but, um, yeah, you can just Google me with Morningstar and you'll see a list of all my articles pop up if you're, if you're interested. Um, I, I have a LinkedIn page. I don't really look at it, but, people can find me there, I suppose.

but yeah, and I do, you know, I like to hop on these podcasts too, so maybe they'll see me across, you know, some other ones or, or on this one again at some point.

Mike Philbrick

Sounds good.

Adam Butler

Awesome. All right. Well, thanks so much for sharing and for the, for the great chat. And, maybe we'll catch up with you in a year or, so. See what you've, what you've touched on.

Jack Shannon

Yeah, I,

Mike Philbrick

cooking?

Jack Shannon

that's right. I appreciate it. Uh, again, thoughtful discussion. Um, I enjoyed it as well.

Adam Butler

Perfect. All right. That's a wrap.

Transcript source: Provided by creator in RSS feed: download file