• Home
  • /
  • Blog
  • /
  • Bogleheads® Live with Dr. Sunil Wahal : Episode 6

Bogleheads® Live with Dr. Sunil Wahal : Episode 6

Post on: June 19, 2022 by Rick Ferri

Dr. Sunil Wahal  explains the hidden costs of mutual funds, and answers audience questions. Dr. Wahal has published two papers that are germane to our discussion.

The True Costs of Mutual Fund Ownership
The Huge Tax Bills That Came Out of Nowhere at Vanguard

Sunil Wahal, Ph.D., is the Jack D. Furst Professor of Finance and Director of the Center for Investment Engineering at the W.P. Carey School of Business, Arizona State University. Before joining the ASU faculty in 2005, Dr. Wahal served on the faculty at Emory University and Purdue University.

Dr. Sunil Wahal

Listen On


Transcript

Dr. Sunil Wahal on hidden fund costs

Jon Luskin: Bogleheads® Live is our ongoing Twitter Space series where the do-it-yourself investor community asks their questions to financial experts live on Twitter. You can ask your questions by joining us for the next Twitter Space. 

Get the dates and times for the next Bogleheads Live by following the John C. Bogle Center for Financial Literacy on Twitter. That’s @Bogleheads. For those that can’t make the live events, episodes are recorded and turned into a podcast. This is that podcast. 

Thank you for joining us everyone today for the sixth episode of Bogleheads® Live. My name is Jon Luskin, and I am the host for today. My co-host for today is Dr. Sunil Wahal, finance professor and Director of the Center for Investment Engineering at the W.P. Carey School of Business at Arizona State University.

Today, we’ll be discussing the costs of mutual fund investing. I’ll rotate between asking Professor Wahal questions that I got beforehand across the Bogleheads® forum at bogleheads.org and Bogleheads® Reddit. But before that, let’s start by talking about the Bogleheads®, a community of investors who believe in keeping it simple, following a small number of tried-and-true investing principles.

You can learn more at the John C. Bogle Center for Financial Literacy at boglecenter.net

Before we get started on today’s show, a disclaimer. This is for informational and entertainment purposes only, and should not be relied upon as a basis for investment, tax, or other financial planning decisions. 

Professor Wahal, thank you so much for joining us today. Your research shows that mutual funds have less well-known costs, less visible costs. These less visible costs are in addition to the more transparent costs of expense ratios, trading fees, and - heaven forbid - advisor commissions. For those of us who aren’t investing nerds, know that an expense ratio is an annual fee charged by a fund – be it a mutual fund or an exchange traded fund - with the latter almost always abbreviated as ETF.

As an investor, knowing what your expense ratio is critical. That’s because numerous studies show that the greater the expense ratio, the greater the cost, the more likely that a fund will do poorly in the future. Said simply, smart investors focus on keeping costs low. 

 And Dr. Wahal, that’s why your research is so important, because you let us know that besides the costs that we can see, there are even greater costs in mutual funds that investors don’t see easily. Can you talk about how big those unseen costs are and what that means for investors? 

 Sunil Wahal: Well, first of all, let me thank you, Jon, and Bogleheads for the invitation. I think that what Bogleheads® does for the entire community of people, those who are fortunate enough to participate in it, is a wonderful thing.

 Financial literacy is something that we need a lot more of. And quite frankly, my profession, academia, does not as good a job as it should with respect to financial literacy. And at the end of the day, bills have to be paid, retirements have to be accomplished, and these things are hard without financial literacy. And the Bogleheads® does a tremendous job with that.

Let me answer your question in a few different ways. You’re absolutely right that the explicit costs of investing - things like expense ratios and trading fees - are very visible to an investor. At least an investor that is paying attention. The less visible costs are the ones that worry me the most, because they’re hard to measure.

 So what are those costs? Well, I can point to at least three different types of costs. And let me show you how they manifest themselves. So, the three costs are trading costs, there are tax externalities - and I’ll define that a bit more precisely in a few minutes - and then performance externalities. 

Trading costs are probably the easiest of the three to describe. Because if you think about an investor investing in a mutual fund, the investor receives returns that are net of the trading costs incurred by the fund. It’s not like the investment manager of the fund family says, “oh, we’ll handle those costs for you.” Nope. Those are eventually borne by the investor.

 And those trading costs themselves have two components. Some are very explicit and relatively easy to measure, regardless of whether the investment manager reports them or not. Those sorts of explicit costs are things like commissions and stamp duties. You can go into the fund prospectus and certain SEC websites and see what those look like for investors.

They’re generally small: two or three cents a share commission. Commissions are not exactly huge in the United States. If you’re a price sensitive investor, as I expect most Boglehead® investors are, then you probably don’t want to ignore them. I consider myself cheap in the sense that I pay a great deal of attention to costs. And because I pay attention to those costs, even commissions that I normally think of as being relatively small can have an impact.

You shouldn’t ignore costs. And especially because they’re related to portfolio turnover. You tell me a fund pays two cents a share commission but has 100% turnover, that’s going to add up pretty quickly. That’s the visible part. That’s the part that can be measured. 

The harder part is implicit costs. They are fiendishly difficult to measure. They’re sizeable. They can be huge sometimes; 50 to 100 basis points (0.5% to 1%) is not an uncommon number to see in these sorts of things.

Let me add a word of caution here; if you hear an investment manager tell you that they have zero or negative trading costs, you might wonder precisely what they’re imbibing, shall we say. It cannot be true, and it has to do with the way things are measured. For example, imagine your favorite mutual fund buys 100 shares of a stock at the bid price.

Another 100 shares at the midpoint one minute later. And this fund says, “okay, we’re going to measure cost for you, the investor, and we’ll tell you what they are.” And it says, “okay, well, if I measure costs relative to the midpoint, I bought at the bid, so I actually incurred negative costs. And then the second one, I bought at the midpoint, so I incurred zero costs. And if I add those two together, I’ve got negative costs. Now the real question is whether the first trade moved prices so the second trade gets a much worse price. So, what’s going on here is that prices are moving in the future and a sequence of trades these things add up quite a bit.

Jon Luskin: Hi folks, your Bogleheads Live host here - Jon Luskin - jumping in for a quick podcast edit. Here, Dr. Wahal mentioned some technical terms. Let’s see if we can explain them for everyone who isn’t a super investment nerd. Dr. Wahal talks about moving prices and how mutual fund managers can ignore moving prices, thereby ignoring the increased cost to investors.

Let’s use an example of Apple stock to explain how this works. Let’s say that XYZ mutual fund is looking to purchase some shares of Apple stock. They’re looking to spend $150 per share. A seller comes along selling XYZ mutual fund some shares of Apple for $150. So, XYZ mutual fund has now purchased several shares of Apple stock for $150 each.

That sale price of $150 happened. And, it happened for all the world to see. Now, the entire world knows that some shares of Apple were just sold for $150. That’s because stocks are publicly-traded. Anyone, anywhere can see what the prices of stocks are as they’re being sold by the millisecond. Yet, even though XYZ mutual fund bought some shares of Apple at $150, they didn’t buy enough. And, they want to buy more.

So, XYZ mutual fund goes out to look for more shares to buy. But here’s the catch: every other trader in the world already knows that Apple just sold for $150. So, if you’ve got Apple stock for sale, you know that you can get at least $150 for it. And if you have Apple stock for sale, you’re probably going to ask for $151 to sell it.

So now XYZ mutual fund, who wants to buy more Apple stock, has to pay $151 when it buys that next round of shares. Because now, that is the going rate. So, to Dr. Wahal’s point, XYZ mutual fund can claim that they’re only ever paying the market price for Apple and therefore XYZ mutual fund didn’t add any additional costs to the investors in its mutual fund.

Yet, that isn’t the whole story because it ignores the fact that by purchasing some shares, XYZ drove up the price that it paid in its second round of purchase. And now back to the show.

 Sunil Wahal: So, really nobody gets away with negative costs. This is a zero-sum game. Trading is a zero-sum game.

So, funds incur positive costs and they’re very, very difficult to measure. That’s the first one that I described - trading costs. What about these tax externalities and performance externalities? Let me say first precisely what I mean by an externality. An externality is when you don’t bear the full consequences of your action.

A classic example is a factory polluting the environment. That’s an externality. So, the channel here is very simple because if you think about a mutual fund, a mutual fund is a co-mingled vehicle. Investors are lumped together. It’s like a big family. Notice I didn’t say happy family. It could be a happy family, but there are times when families are not necessarily happy. If one member of that family takes an action, it affects the lives of the other members of the family. So, for example, if one investor takes out a big chunk of money from the fund, the fund must sell those securities to account for the sell, by law.

So, those sells can generate capital gains tax distributions for other members of the family, other investors. And since the process of selling the underlying securities generates more trading costs, it’s also going to hurt performance. You see why I call these externalities? Other people bear the costs of one investor’s action. The whole thing is commingled.

You asked me how big these numbers are, how big they can get. On the tax side, it’s really hard to get precise numbers, at least on a fund-by-fund basis. The academic studies that are out there - there’s a very famous one by Dixon, Chauvin, and Salem and there’s another one by Salem and Zhang - they point out that if you’re a fund shareholder in the highest tax bracket, you could be paying as much as 1%. And so, for Boglehead® investors who are worried about two basis points here and three basis points there, that’s a big difference. What about the performance side? So, if the fund has to sell a bunch of securities, that’s going to influence future performance.

In this paper that I have with Albert Wang, we generate numbers using really, really precise daily data for about two decades or so. We can isolate the effects because a fund has an outflow on day T and then it has to generate sales of securities to pay for those outflows on T+1, T+2.

So, the performance differential between funds with high outflows and small outflows, that’s as much as one basis point a day. So, that’s about 20 basis points a month. It’s certainly nothing to sneeze at. And then the last thing I’ll say with respect to performance is that it doesn’t even stop there.

There’s a pernicious side to outflows. That’s a little bit tricky, and it’s harder to observe, but other academics and myself, we’ve figured out ways to tease this out from the data. So, imagine that you’re an investor in a fund, and you think that a large investor in the same fund is going to exit.

If you think that investor is going to exit and you think that that’s going to incur costs that are going to be passed on to you, you’re going to be tempted to head for the exit as well. This is a game theory sort of thing. And in the formal language we refer to it as strategic complementarity, but, you know, that’s a fancy way of saying people are heading for the exits expecting that other people will head for the exits. The evidence on this that comes from several different papers is that it can be as much as 19 basis points a month. None of these numbers are small. They can be significant. But what’s driving all of this is either outflows or expectations of outflows.

Jon Luskin: At the top, you mentioned these unseen, less well-known costs can be as much as 50 to 100 basis points. In nerd speak, that is saying half a percent or 1%, but that still doesn’t really tell us too much. So, what does that mean?

Well, imagine if you have a $1 million investment portfolio and now you have a 100 basis point or a 1% drag from these unseen costs of mutual funds on your portfolio. That means you’re losing $10,000 a year because of those unseen costs. That’s real money. That’s certainly going to impact someone’s quality of life, especially in retirement.

Thank you for sharing that, Professor. 

David can ask his question on mutual fund costs to Professor Wahal. 

David: In your opinion, given these implicit costs, if you added all these together and you embedded all this in an analysis of active management, how much worse does it make it relative to a passive approach? 

And is any of this nefarious? Is this a way to get the middlemen paid? Is there a nefarious aspect to this, or is it truly just a frictional cost? Does this make passive look even better? 

And then, ultimately, I’d like to know which platform do you think pays the most attention to this and gives investors - since we can’t see it, since it’s invisible - which platform do you think gives the small guy like us the best product taking these implicit costs into consideration?

Sunil Wahal: Let me address the first one first. Do these costs make active versus passive differential even bigger? So, the answer I think is yes, but it depends.

What does it depend upon? When it comes to trading costs, excessive trading - and notice I said excessive trading, not zero trading - excessive trading is a cost that is borne by investors. And to the extent that active funds have excessive trading, then yes, that’s a cost that’s going to be borne by investors of those active funds.

And by definition, passive funds will do better. The reason I said ‘excessive’ and not ‘no trading’ was because some trading is always necessary and is required. And you made an important distinction when you used the word “passive” as opposed to “index”, because they need not be the same thing. And I think frequently people, in their minds, equate the two.

They’re not necessarily the same thing. An index fund takes an index defined by god, where god could be Russell, S&P, or whoever your favorite god is, and tracks that index. It need not be passive. Indexes are reconstituted by the definitions of god, but certainly tracking an index means that you trade less, and so those costs are lower.

That does not mean, however, that the costs associated with outflows are lower. They need not be. Really what I’m describing is the costs, the performance, and the tax costs that are associated with outflows in the mutual fund vehicle. An ETF, for example, would not suffer from those types of costs simply because of the ETF structure, you bear the cost of your own actions.

So the commingling, the family aspect of this, isn’t there. Broadly speaking, I would concur that these sorts of costs tilt more in favor of passive than active. So yes, I think your original conclusion is appropriate. But really, it’s about the vehicle that matters. Your second question was about whether this is nefarious or not.

That’s a really tough question to answer. Certainly, from a legal perspective, one would have to find more than a smoking gun to causally identify whether this is nefarious or not. I wouldn’t think so. Outflows, investors leave funds for a variety of reasons. There can be aspects inside trading that could be nefarious, but the flow relationship and the trading costs in general, I don’t think of them as being nefarious.

Your last question concerned what platforms pay more attention to these sorts of things than others. That is hard simply because there are not a lot of platforms that can put transparency associated with these things. Partly because they’re tough to measure. To the extent that a platform tells you “here’s what the turnover of fund X and ETF Y looks like,” that will help.

“Here’s what, at least the explicit trading costs look like,” that will help. “Here’s what the last year’s net flows look like.” All of these things will help. Unfortunately, I don’t know of any platform that is comprehensive, that puts all of that data together. It would be nice, but it’s not there. And part of it is, these things are hard to measure.

Jon Luskin: Wonderful. Thank you, Professor. In your presentation on the subject, you talk about how when there are redemptions - that is when folks who are in a fund want to sell the fund and want to get out - that means the managers of that fund, they have to sell existing positions, and they have to sell them no matter what price that they can get for them, even if it’s a bad price.

So that speaks to the fact that perhaps it’s not nefarious, it’s just a function of the structure of mutual funds. These mutual funds need cash because investors are headed for the door. They can’t wait to sell, unlike, as you mentioned in your presentation on the subject, those investment managers can wait to buy, getting the better price to buy a security. But when you’ve got to sell, you’ve got to sell it. And that means they’re going to take whatever price is available. And selling at a bad price, that is going to decrease the investor’s return, the investors that are inside that mutual fund. Is that right? 

Sunil Wahal: I think that’s exactly right. When I have to pay my kids’ college tuition, when it’s due, it’s due.

If I have to sell securities to pay that college tuition, that’s what I’m going to do. I don’t have any timing latitude in the matter whatsoever. It’s what has to happen. 

Buys are a different story. I can afford to be picky. Today is, I’m busy, I don’t have time to put my buy orders in place.

That’s okay. I’ll do it tomorrow. I’m not a market timer anyway, so it doesn’t matter to me. But sells are sells. You’re right. They have to be done. 

Jon Luskin: Yep. And if they have to sell them at a bad price, that’s bad news for the investor in those funds. So that’s unfortunate. But that is what you get when you sign up for a mutual fund, as shown by your research.

Folks, I am going to jump to a question that I got next from the Bogleheads forums. This question is from VTI, and he says, 

“Professor Wahal presents a compelling case that the hidden costs, especially outflows, could potentially dwarf mutual fund management fees. Given that, how do index mutual funds maintain such small tracking errors? Shouldn’t we expect to find tracking errors larger than the management fees?” 

And here, let me explain the nerd speak term: “tracking errors” is the nerd’s way of saying, “hey, the S&P500 index is going to get a return of X, but my S&P500 index fund gets X minus whatever that tracking error is.

So, the question here is, “hey, if index mutual funds should have tracking errors from these unknown fees, how come they’re not larger? 

Sunil  Wahal: It is an excellent question. Index funds are a great, great invention. They really, really are. Over the last 10, 15 - maybe longer - years, the vast majority of index funds have had net inflows, not outflows, even on a daily basis.

And the consequence of that is that because they’ve got inflows, the costs that I’m describing that are associated with outflows from a mutual fund just don’t manifest. They occur when there has to be net selling, but when there’s net buying, you’re not going to see it.

And that’s why they don’t generate so much tracking error. Now, in the rare case that when there are net outflows - so it is rare, but it does happen - we do, in fact, see hints of performance consequences. And remember that in index funds, there is a concerted effort to minimize tracking error. So, think about index reconstitutions. Imagine an index fund that’s tied to the Russell 2000, and the Russell 2000 is reconstituted. So, when it’s reconstituted, we know that index funds incur costs to minimize that tracking error. Sometimes those costs can be quite substantial. That’s why we don’t see it in the form of tracking error.

It’s because most of the time, we get to see situations where there are net inflows, not outflows. Remember, an index is not something infallible, It’s manmade. But it can make the fund a slave to this sort of manmade object. And that’s where the reconstitution stuff comes in.

But in general, you know, we’ve just seen net inflows. And I think for most investors, that’s beneficial.

Jon Luskin: It sounds like some of these unseen and really significant mutual fund costs really only apply to active mutual funds, especially when those active mutual funds have a substantial decrease in popularity, when folks want to get out of them.

But since we don’t really see that with index funds, that they’re in fact growing in popularity, that that’s necessarily not going to show up as one of those unseen but rather large costs of mutual fund investing. 

Of course, nothing is guaranteed. Perhaps if there is a severe event of all sorts of panic selling, then maybe we would see some of this show up. Alternatively, if indexing became unpopular and folks were heading out the door of the index funds, then possibly these unseen costs, fire sale prices would apply, creating some larger tracking errors. 

And then one clarifying point: Professor Wahal mentioned reconstitution. With a narrow enough index fund - the Russell 2000 is an index that invests in only small companies - if a small company gets big enough, they’re not allowed to be part of the Russell 2000 anymore. They’re going to move up into the mid-cap index instead. In that case, now the Russell 2000 index is being reconstituted. It’s kicking out what is now larger company. That means those Russell 2000, those small-cap index funds, have to make some trades. That’s reconstitution for those who aren’t investment nerds. 

Sunil Wahal: It was sometime back a few months back, where a couple of the Vanguard target date funds had to do a very large capital gains tax distribution. I remember Jason Zweig in the Wall Street Journal writing a nice article describing what had happened.

This is where people holding these Vanguard funds in their non-taxable accounts – it’s a pretty big illustration of that sort of thing.

The tax hit that investors take because of those withdrawals was pretty significant, even in a target date fund. A little anecdote that I was reminded of that I think it will be useful for people to take a look at. 

Jon Luskin: That is a really good follow-up to the point that I just made as, hey, we’ll probably never see these additional large costs for index funds unless there’s a severe market panic and/or if index funds become unpopular. You just made the case, this just happened because of a restructuring with the different share classes that Vanguard was doing. So again, the mutual fund even in the index fund format, can be problematic. Not always the best way to structure an investment vehicle.

Ross, I’m going to make you a speaker, and you can ask your question on mutual fund costs to Professor Wahal. 

Ross: Professor, I want to ask about ETFs. They’re often presented as a panacea to the problems of costs with mutual funds.

I’m wondering if you could sort of break down the costs that ETFs might help with and then the costs that ETFs might not help with. 

Sunil Wahal: Where do ETFs fit into this picture? Do they incur the same types of costs? Are they better? Are they worse? Your comment, Ross, about sometimes they’re presented as a panacea is also appropriate. 

Nothing is a panacea. Costs are costs. They have to be borne by investors and by society as a whole. That said, progress in financial science and in financial products takes place because some structures are better than other structures. So, the ETF structure, in general, is an improved structure because of the externalities that I described earlier. That an investor in an ETF bears the costs and benefits of his himself. It’s not shared, it’s not a commingled setup like a mutual fund or a family setup.

So, if one investor leaves, it doesn’t have consequences for other investors. That doesn’t mean, of course, that the ETF doesn’t incur costs. Other things have to happen. The in-kind create and redeem procedure that’s in ETFs is also a better structure. It reduces how much needs to be traded, and the arbitrage mechanisms that underlie the ETF structure reduce costs.

Can they shrink them to zero? The answer is no, absolutely not. Costs will not go to zero. But they can be minimized. 

Imagine you’re driving a car and your car tires are rated at 40 PSI. If you drive that car with 30 PSI, you’re going to have more friction. It might give you better control, but it will reduce your gas mileage.

The frictions associated with a mutual fund are more significant than with an ETF. It doesn’t mean that it’s going to be frictionless. It’s just a better structure, but not a zero-cost structure.

Jon Luskin: jackofspades from Bogleheads Reddit asks, 

“What should retail investors look for or know regarding share lending when deciding what’s best for them?”

Sunil Wahal: I think the biggest question, the biggest piece of information that an investor ought to know, is where’s the security lending revenue going.

Is it being kept by the investment manager – the fund family – if you will? Or is it being distributed to the shareholders of the fund? That’s the single most important aspect of share lending revenue. 

So, ask yourself, “who’s getting it?” That disclosure is clear. It might take a little digging to get to it, but the disclosure is there, and I think that’s what investors need to know.

My preference is that the share lending revenue goes to fund shareholders, but that’s me. 

Jon Luskin: I’m biased. But I think that also makes sense. 

Here is another question. So, frequently folks will take a conventional approach to investing, not using low-cost, tax-efficient vehicles.

They’ll be in some kind of high-fee, actively managed mutual fund. And then they’ll see the light and decide, “hey, I’ve got to do something different.” Of course, the catch is now that they’re staring down some significant unrealized capital gains in these high-fee, tax-inefficient mutual funds. And that just means if they sell it, they’re going to pay a large one-time tax hit.

You can spreadsheet out the value of selling that high-fee, tax inefficient mutual fund today, compared to lower investment costs going forward. You can just do a little spreadsheet analysis. Look at the expense ratio of the existing fund compared to the expense ratio of the low-cost fund, calculate your annual savings, see how many years it’ll take you to break even on that tax bill you’ll pay for that one-time realized gain selling that fund, multiplying the realized gain by your long-term capital gains bracket. 

But that is incomplete when it comes to your research that shows there’s much more than just the higher expense ratio in making an updated, more complex, but more accurate calculation. Professor, how should one estimate those additional unseen costs of mutual funds? 

Sunil Wahal: There are at least a couple of aspects to this. The first one is to recognize that as much as we love precision in things, you’re not going to get precision in any of these things.

You’re going to have to take estimates of trading costs and estimates of expected underperformance due to outflows. I could give you a ballpark number. Let’s say 50 basis points for small-caps, 10 basis points for large-caps, right? It doesn’t really help you because a small-cap fund with 100% turnover and a billion dollars in assets is very different from a small-cap fund with 20% turnover and $5 billion in assets.

They have very different trading costs, and the outflows - let’s say a $100 million outflow in each of those two funds - one with a billion dollars under management and one with $5 billion under management, are going to have very different consequences. 

The bottom line is I can’t give you a number that says here’s what you stick into your slide rule or, as you described it, spreadsheet out.

One has to take broad averages, or one can impute - sort of reverse engineer, if you will - how big those costs would have to be for me to flip the decision one way or the other. What I will tell you is that I think the thing that is the most important is to recognize that the performance effects - not the trading costs, but the performance effects - are driven by flows.

So the thing that you have to watch out for very carefully is whether a mutual fund is experiencing large outflows. If that’s the case, I think it tilts the analysis quite a bit. It changes outcomes. 

I’ve done a bunch of work in the last couple of decades on institutional investment management. On the institutional side, people pay attention to flows.

Plan sponsors, consultants. They pay a lot of attention to flows out of not only just the funds, but also the fund families themselves. Outflows create costs. So, unfortunately, Jon, I’m not going to give you a number that says here’s what you plug into your spreadsheet.

Here’s how it tilts the analysis one way or the other. Because it really is going to be a case-by-case basis. Let’s suppose the marginal investors - I don’t know what their tax rate is, right? - so it might affect, the tax rate might be relevant. If the investor is working with a good financial advisor who is providing proper, honest, and good advice, then the advisor should be able to - I really like your phrase - spreadsheet out sort of various scenarios. But I think that’s the right approach. 

 Jon Luskin: For retail investors, for do-it-yourself investors, where can they find information on outflows? 

Sunil Wahal: Morningstar carries that sort of information.

I believe there are other data vendors that have it. It’s all there. It’s in public domain. We use in our research data sources that we purchase, but Morningstar has it, Lipper probably has them, Bloomberg has them. So, I don’t think it’s hard to get. I think it’s relatively straightforward information to gather. 

Randy: I’d like to just confirm what was said earlier. If an individual has a position in an S&P500 index mutual fund, the takeaway seems to be that they would do better off in the same fund as an ETF. Is that correct? 

Sunil Wahal: Right. If the equivalent investment is available in an ETF, I think that the individual would be better off in an ETF. I can tell you that in my own personal portfolios, wherever that has been an option, I have always moved everything to the ETF. It’s just a cleaner solution.

Jon Luskin Folks, that is it for the time that we have for today. Thank you to Professor Wahal for joining us today and thank you for everyone who joined us for today’s Bogleheads® Live. 

On our next Bogleheads® Live, Eric Balchunas will be discussing his new book, “The Bogle Effect: How John Bogle and Vanguard Turned Wall Street Inside Out and Saved Investors Trillions.”

The week after that, our guest is Dr. William Bernstein. Until then, you can access the Bogleheads® forum, Bogleheads® Wiki, Bogleheads® Reddit, Bogleheads® Facebook, Bogleheads® Twitter, Bogleheads® YouTube, Bogleheads® local chapters - shout out to my San Diego group - Bogleheads® virtual online chapters, the Bogleheads® on Investing podcast, Bogleheads® conferences, and Bogleheads® books. The John C. Bogle Center for Financial Literacy is a 501(c)(3) nonprofit organization. At boglecenter.net, your tax-deductible donations are greatly appreciated. 

Thank you again, everyone. Look forward to seeing you again, where Eric Balchunas will be discussing his new book, “The Bogle Effect.”

Until then, have a great week.

About the author 

Rick Ferri

Investment adviser, analyst, author and industry consultant


Tags


You may also like