US Videos

Kinnel's Tips on Do-It-Yourself Fund Screens

Russel Kinnel
Christine Benz

Christine Benz: Hi, I'm Christine Benz for How can investors screen for the best mutual funds? Joining me to share some research on that topic is Morningstar's director of manager research, Russ Kinnel.

Russ, thanks for being here.

Russ Kinnel: Glad to be here.

Benz: Russ, for the past few years, you've written this piece called 43 Fantastic Funds. You screen for funds that you think are best-of-breed. Let's talk about the key ingredients that go into that screen. What are the kinds of things you are looking for?

Kinnel: I'm looking for some of the key drivers of good funds, so things like low costs, high levels of manager investment. I want a fund track record where the manager outperformed over their entire tenure versus a benchmark. I want really just those key kind of elements. I don't want to have a lot of other tests. The idea is, just have a few tests with a very high level, things like a Positive Parent rating rather than a bunch of tests at lower level. The idea is just a few key tests on the most important things and leave the rest out.

Benz: You published this article, and people can find it on But you heard from a lot of readers after you published the article, they wanted to create a screen maybe using some of their own inputs but also using some of your inputs. The problem is, is that not each and every one of your inputs is available to retail investors. Manager ownership, for example, is not something that someone can readily find. If I'm crafting my own screen for mutual funds, what are the data points that I should be sure to embed in whatever screen I'm running?

Kinnel: The good news is, you can screen on expense ratios, so that's a good starting point. Some websites let you screen relative to an average; some it's simply picking absolute levels. Either way, you can at least do some initial expense screen. That makes a lot of sense.

Benz: If I'm asked to plug in an absolute level, what would be a good starting point? Maybe we could take this by asset class, like domestic equity, bond, and perhaps, international equity?

Kinnel: Domestic equity, I might say, less than 1%; international, maybe less than 1%. Or if you want to put a fine point on it, less than 1.1%. Fixed income, I think, probably less than 75 basis points is a good spot. It's a little tricky. But if you want a one size fits all, something like those are a good idea.

Benz: Expenses would be near the top of the list because certainly all your research has found expenses to be really predictive of good or bad performance. Low expenses tend to predict better performance. What are the other factors that are widely accessible that you think do make decent components of screens?

Kinnel: I think manager tenure. It's not an explicit part of the Fantastic 43, but by saying you need a minimum of five years of outperformance versus a benchmark, I am essentially saying five-year manager tenure. Where you can screen on manager tenure, that's a good one. I think, not that as predictive value, but if you're looking for new fund ideas, a lot of them will let you screen out funds that are closed to new investors. I think that makes a lot of sense. Sometimes you can screen on share class or just some way to limit it to one share class per fund is helpful just as you are trying to narrow the list, so you don't get 400 results at the end of your screen.

Benz: This is something that you wrote about in the November issue of Morningstar FundInvestor. Let's talk about performance, because a lot of times at the top of screening tools that you might use, you might see that you can screen on three-year or five-year performance. I noticed when I looked at your Fantastic 43, you don't put a big emphasis on performance. Why is that?

Kinnel: There is one performance screen saying you have to outperform the benchmark over the entire tenure of a manager. When I go to build these screens, that option isn't really available. A lot of the screens I see out there, the fund screeners, they might have like an absolute screen number, which I think is a really bad idea. You could say, I want funds that returned more than 8% a year over the last five years, which to me is really silly because it doesn't tell you if that's really good performance or really bad performance relative to a peer group. It could be too short term. I don't really like three or five years.

I think a lot of the performance screens out there are not that good. But also, I think, past performance obviously just has some limits. You don't want to lean too heavily on past performance because it's not a very good predictor on its own.

Now, I think, risk screens, if available, are good ones. In the Fantastic 43, I said screen out those with Morningstar risk levels, just because the really high-risk funds, people tend not to do very well in them because they have extremes, and that leads to bad investor behavior. If you can screen on Morningstar risk level high, screen that out, that's great. Or you can just screen on any kind of standard deviation or other risk measure--just the idea is to screen out the really high-risk funds, not to screen for super low risk only, because I think then you end up taking on big opportunity costs by just focusing on really low-risk funds.

Benz: Right. In terms of the active versus passive divide, should I approach this job differently if I'm, say, an all-passive investor? How should my screen maybe look different than if I'm someone who is willing to consider active funds as well?

Kinnel: I think you should probably make a difference. One is, leave out the manager tenure screen.

Benz: Doesn't matter so much.

Kinnel: It's not too important. Passive funds tend to be really team-managed. In a lot of passive funds, the management part is not that hard. It's important that you have skilled, experienced people, but at the same time it's generally not the most important part of the process. Then obviously, performance, I would leave that out entirely, because again, if you are looking for low costs, you are not looking for something that beat the benchmark. I would leave out performance and manager-driven pieces. I think volatility is still a legitimate thing to look at. Obviously, fees should be doing a lot of the heavy lifting for you though.

Benz: Category average, for example, if I'm screening for passive funds. If I'm simply looking for funds that are below category average, that's probably not going to do a lot for me if I'm looking for passive products?

Kinnel: I think at a minimum you might set an expense screen of below 50 basis points or maybe even lower than that, and then you might want to rank on expenses when you get your results page so that you are starting by looking at the very cheapest funds.

Benz: OK, Russ. Thank you so much for being here to discuss do-it-yourself fund screens.

Kinnel: You're welcome.

Benz: Thanks for watching. I'm Christine Benz for