Christine Benz: Hi, I'm Christine Benz for Morningstar.com. Deciding which asset types to place inside which types of accounts is an important consideration. Joining me to discuss this topic is financial-planning expert Michael Kitces. Michael, thank you so much for being here. Michael Kitces: My pleasure. Benz: So most investors have heard of asset allocation and understand its importance. There's another area of planning called asset location. Let's discuss first of all what that means? Kitces: So most people are familiar with this asset-allocation concept: We're going to own different investments of different types; hopefully they will be well-diversified with low correlations; some things will zig, while others zag. So the whole idea of asset location is to say, "Great, we've got all these different investments, but we also have a lot of different types of accounts, particularly in the individual world." So we've got brokerage accounts, then we've got IRAs, and we've got Roth IRAs. Each of those has different tax treatments. So the whole principal of asset location is to say, "Once we've allocated our assets, in which accounts do we put which investments?" Now the starting point for this was pretty straightforward. Stocks have preferential long-term capital gains treatment; we're going to buy and hold them for a really long period of time. So let's take that in the brokerage account … Benz: Taxable. Kitces: Yeah, taxable account because we're not planning on paying taxes on it for a long time anyway, since we're just going to hold on to that investment as a core, and beyond that we get these favorable, capital gains rates. Then let's take the bonds, which churn off all this ordinary income that they kick out every single year because we get paid bond interest, we will put those inside of the IRA and let those compound inside the IRA. We are going to get ordinary income treatment no matter what, whether it's bond interest or inside of an IRA, but at least, if it's inside the IRA, we can let it compound inside, get a little bit better tax-deferred compounding, and then at some point down the road, we will pull the money back out and pay the taxes. So that was sort of the basic principal of asset location. The problem is as the environment shifts, it doesn't necessarily work as well as a rule of thumb for where we are today. <TRANSCRIPT> Benz: So you have recently looked at that sort of conventional wisdom about asset location and kind of poked at that, and said, "Well for one thing, all bonds aren't completely tax-inefficient. For one thing, yields are really low." So no one is paying that much on income on their yields. The other thing is that stocks are not purely tax-efficient. So you do have to pay tax on dividends as they get kicked off. So let's discuss what's going on there. Kitces: There are almost two different effects that kind of overlay each other to create a problem with the traditional framework for asset location. The first is that, there are really two factors that go on in asset location. One is "just how tax-efficient is the investment?" So in theory, buy-and-hold stocks forever is sort of the ultimate in tax-efficiency: "I buy it once. I hold it for 30 years. I don't pay the gains until the end of 30 years, and I get this preferential rate." And bonds are at the other end of the extreme; all that income kicks off every single year. But there's really a second piece to it, and we can sort of make a graph of this. On one axis is the tax efficiency ranging from the really efficient stuff to the really inefficient stuff. But the second is the expected return of the investment. Because I can have something that's really tax-inefficient, but if it's all the way down here with a low return, as well, it turns out, it just doesn't matter that much. The difference between getting really effective tax-deferred compounding or not on a bond that yields, too, is almost nothing even after a decade or two. There's not enough return with a tax drag to make the compounding add up regardless of whether you shelter it or not. So returns become a really big driver of how much tax efficiency really matters, and that's one of the things that starts to undermine sort of our base of how we've traditionally done asset location which is in this ultra-low-bond-yield environment. It really doesn't actually pay very much at all to put a bond inside of a tax-deferred account. Now maybe with high-yield bonds or some things could kick off income, or yield, but with sort of our classic corporates and especially governments where the yield is so low, there's really not much of a kicker to keep the bonds in the IRAs. At the same time we have a problem on the stock end, as well. The problem on the stock end is while it's nice to look at this sort of theoretically and say, "Well, I bought a single stock and held it for 30 years. And it was a big growth stock, and it never kicked off a dividend. And I just got one giant capital gain at the end." It's not how it works in the real world for most investments, for lots of different reasons. First of all, we have dividends, granted wish we had a little bit more in dividends these days and from some companies, but we have dividends as a portion of the growth of the company that gets distributed out to us. Now we can reinvest [the dividend], but that's now a little bit of an ongoing tax grind where not all of our growth is compounding perfectly tax-deferred. Beyond that, sort of the practice we see in the real world is people don't really hold most of their investments indefinitely. Even when we're following a relatively buy-and-hold approach, indexes change, investment opportunities change. We've got a world of ETFs that didn't exist 10 years ago. We've got a world of indexes 10 years ago that barely existed 20 to 30 years ago. So the investment world shifts. What we own ends up shifting over time, and so it's pretty rare that anybody really goes 30 years without turning over their portfolio. And certainly from a lot of the analytics that Morningstar produces, we see really the turnover rate is actually much higher than that for most investors and most portfolios. So as we start turning over the capital gains, now we begin to have a little bit of a compounding problem even on the equities side, where it's nice that we get pretty preferential treatment--and we do still get our long-term capital gains and qualified dividend treatment as long as we meet a pretty modest holding period--but an investment that goes for 20 or 30 years might have 25%-50% of its return kicked out as dividends every year anyways and might get turned over three, five, or 10 times over the span of a multidecade period of time. So when we look at that, suddenly we find, "You know what, getting tax-deferred compounding especially on something that's got a higher return, actually starts to look a little bit better in an IRA," even if we take capital gains and qualified dividends and convert them into ordinary income because we're getting really tax-inefficient as these investments start turning over and these dividends come out just in the natural course of investing that compounding starts to actually add up and outweigh the preferential tax rates if a time horizon goes long enough. And ironically that's only exacerbated in the new legislation we got a couple of months ago that lifted our capital gains rates and our qualified dividend rates higher, as well. Benz: For very wealthy investors. Kitces: For very wealthy individuals. Benz: So you're saying it's not as black and white. Does that mean though that the traditional asset-allocation advice is sort of backwards? Should people flip it? Kitces: What we find is it actually comes out a little bit backward for some people and that it does deserve to be flipped. The truth, it doesn't really deserve to be flipped per se because we're trying to do things like get the bonds out of the IRA. It flips around because what really becomes a priority is we need to cherry-pick whatever our most tax-inefficient investments are and really make sure those get inside of an IRA even if it knocks the bonds out in the process. So, as we view it now best practices in asset-location issues is, instead of just following rules of thumb, I call it "the work from the outside-in" approach. So if we make a giant list of all our investments, we could rank them in terms of a combination of expected return in tax efficiency or tax inefficiency. So one in the extreme might be the S&P 500 index; I truly intend to buy and hold as a core portion of my portfolio. The dividend is relatively low; there's no expected turnover. I don't get much better for tax efficiency than that. It's the top priority inside of my brokerage taxable account superseded maybe just by a growth stock I plan to buy and hold. Then at the other end of the extreme we find our most tax-inefficient investments; things that kick off ordinary income and lots of it. So this could be high-yield bonds, this might even be some types of REITs, this could be things like managed-futures funds, and this could also be actively managed strategies. So things where maybe I'm even generating gains and they're equities, but a lot of it's short-term gains, a lot of it's nonqualified dividends, things like that. We're still expecting a good return, which is crucial for asset-location issues, but they are really tax-inefficient. So we start working from the outside-in. Now, because everybody has got different amounts of money in their IRAs versus brokerage accounts, what happens in practice is when we work from the outside-in, at some point we run out of money, from one account type or the other, and basically everything ends out in the other one. And that's OK. But what that does is that ensures that whatever is most tax-efficient with a good return really ends out in the brokerage accounts and what is most tax-inefficient with good returns really ends out in the IRA. Ironically particularly in today's environment, it means things like bonds, they basically just land where they land. Maybe they'll end out in your brokerage account because you filled your IRA with other stuff, maybe they'll end out in your IRA because you filled your brokerage account with some high-return tax-efficient investments. But bonds will just land where they are and that's actually OK because again their returns are so low, it actually doesn't matter a lot where their location is compared with other investments that have higher priority. Benz: And what if yields go up again in the future then you might have a reason to [chose a specific location]? Kitces: Great question. So if yields go up again in the future, asset location is something that really can and should be a little bit more dynamic. That's another reason why we actually do this sort of outside-in approach. So my really tax-efficient stuff, things like equities for the long run with low dividends and high growth are going to have such a weighting toward a brokerage account that even if bond yields move around a little, stuff's still going to stay out there. But it is entirely possible that as the bonds improve they might dethrone some other investment that was in my IRA, in which case I'm going to sell out of the bonds and buy them in my IRA and sell out of the other investment and push it into my brokerage account. Well, that's actually easier. I can liquidate the bonds probably without not necessarily a lot of gains, in fact maybe even some losses to harvest, if rates are going up. And I can sell whatever is that was in my IRA account because I don't have tax consequences there and it's easy to relocate that out. So on the one hand we really would view asset location as something that should be more dynamic, and that's OK. And at the same time we actually can manage to allow that pretty well because the benefit of sort of managing things that are going in and out of the IRA is that a lot of those liquidations in now the IRA are tax-free anyways. We can sell something in the IRA and repurchase it here; that's not a problem. And many of the tax-inefficient investments we can relocate from a brokerage account to an IRA. And in a rising-rate environment that probably won't be bad anyways because you might even be harvesting losses along the way. Benz: Right. Lots to chew on here, Michael. Thank you so much for sharing these best practices. Kitces: Happy to help. Thank you. Benz: Thanks for watching. I'm Christine Benz for Morningstar.com.