Reducing taxable income: Time to reconsider investments in broad indexes?

[samclem]

I'm wondering if recent changes may make it a good idea for some investors to consider abandoning broad index funds/ETFs in favor of more narrow sector MFs/sector ETFs in their taxable accounts. For example, rather than investing 50% of our equity money in the Vanguard 500 Index Fund (VFINX), we might choose to split the same amount among the 9 SPDR sector fund ETFs that cover the same stocks, but by sector (Financials, Energy, Consumer Staples, etc)

Why?
1) Ability to better manage taxable income. The quarterly/annual returns of a broad index will be a lot less volatile than the individual returns of the sectors. So, there will more frequent opportunities to tax loss harvest in taxable accounts.
The chart below (.PDF also at this link) is set up like the familliar Callan Periodic Table of Investment returns, but it is just for the 9 sub-sectors within the S&P 500. Between 2004 and 2013, the overall S&P 500 index year-end share prices had only one year when share prices declined (2008). The table below shows that over the same 10 years there were sectors within the S&P 500 that had losses in 4 of those years. Of course, if we looked at monthly or quarterly results we'd expect to see even more opportunities to harvest tax losses. In all cases, opportunities to harvest losses can be expected to be best if there are frequent new purchases of equities (to assure there are higher-basis shares on hand to allow short "dips" to be harvested)

So, investing in these sectors would have offered much more frequent opportunities to do tax loss harvesting. And this is probably a relatively modest example--more "exotic" asset classes (e.g. foreign equities held in regional sector ETFs rather than a larger, broad foreign index, etc) could be expected to be even more volatile. The >overall< risk won't be any higher (you'd be invested in the same stocks whether you've invested in a broad index or narrow sector ETFs), it's just the tax loss harvesting opportunities are enhanced.

While I keep saying "tax loss harvesting," obviously there is also potential to "reset" the cost basis on "winners" in years where a taxpayer might have some headroom below their expected future tax brackets. This can be expected to help reduce future tax liabilities.

2) It's easier now: With the new requirement that brokers track cost basis, and the very convenient pop-up menus that are common at broker websites for selecting which shares to sell, the "specific ID" basis method is a breeze to use. Gone are the days of manually entering this stuff onto spreadsheets, or writing letters to the fund companies (really for the IRS) to identify which shares are being sold. As lazy as I am, I'm now using the "specific ID" cost basis method in my taxable accounts.

3) It's potentially more important/lucrative to reduce taxable income than it has been in years past. The 0% Cap Gains rate has been made permanent for those of us "fortunate" enough to qualify, the OMAGI limits for qualifying for premium subsidies and assistance with OOP medical expenses, and even the revised qualifying criteria for Medicaid all increase the need to reduce taxable income for some of us. It looks like this might be such a way.

Other: As noted on the S&P marketing literature (at the link above), the SPDR sectors are not "cap weighted"--some are much, much smaller than others. To replicate the S&P 500, an investor would need to rebalance the amount invested in each sector occasionally. As the advertising info at the link shows, overall returns for the last 10 years have been significantly better for investors who didn't cap weight the sectors, but instead just plunked equal amounts (11.1%) into each one:
10 year cumulative return:
................ S&P 500: 107.37%
.......Equal Weighting: 140.76%
Comment: This strikes me as likely a bit of data mining, and possibly the result of the big declines that the heavily-weighted Financials sector took in 2007 and 2008. In effect, "equal weighting" amounts to a "sector bet" on the sectors that just happen to have lower capitalizations, and these are strictly an arbitrary result of the way each sector was defined when it was constructed. It happened to do well over the last 10 years, but I don't see any particular reason to expect that to continue.

Anyway, that's it. We've touched on this idea in previous threads, and I suppose there's nothing especially novel about it. Also, I haven't looked into various smart ways of implementing it (e.g these SPDR sector ETFs are probably the best known, but the very similar Vanguard sector ETFs (VCR, VFH, etc) appear to have slightly lower ERs). Also, I wonder if one could remain effectively fully invested and avoid "wash sales" by using the proceeds from the sale of a Vanguard sector ETF (e.g VHF--Financials) to buy the corresponding SPDR sector ETF (e.g. XLF). They are based on different indexes (MSCI and S&P, respectively), different CUSIPs, etc, so I'm pretty sure that's enough for the IRS.

 

[LOL!]

For me it is not going to make much sense. The reason is that splitting things up won't save me any on taxes if the main savings will be tax-loss harvesting.

a. As for managing taxable income, I don't see how the sum of the parts is different than the sum of total. The sum total dollar amount of dividends will be close whether using sector funds or a total market fund. The total market fund could possibly have a higher fraction of qualified dividends.

b. While I am a big proponent of tax-loss harvesting, I have enough carryover losses that new potential losses no longer drive any big investmg decisions.

c. I am not going to rearrange my holdings anyways because that would realize lots of capital gains when it is unnecessary to do so.

So I'm liking the status quo with just a limited set of asset classes (if possible): US total stock, US small-cap value, Foreign total, foreign small-cap. If I have new additions, I split foreign into more tax-efficient subsets: large-cap developed, large-cap emerging, small-cap, and small-cap emerging. The first two can be more tax efficient than than the latter two.

 

[clifp]

I agree SamClem I think slicing and dicing makes more sense than ever given the very high marginal tax rates at the 15% level and also the various ACA subsidies.

There are so many low cost ETFs out there that it is very easy to between VWO (emerging market) and SCHE (Schwab's emerging market index) or any other sector.

 

[panacea]

I like the concept. This same approach is ideal for converting IRAs to Roth IRAs (assuming the conversion has a tax cost). Splitting up conversions into different asset classes offers more opportunity to recharacterize and reconvert at a lower cost if one sector drops.

 

[samclem]

c. I am not going to rearrange my holdings anyways because that would realize lots of capital gains when it is unnecessary to do so.

This will be a challenge for me, too. But it will be a no-cost option for me when it is time to re-invest dividends and interest (granted, not a lot of money each year, but every little bit . . . ).
Really, the idea is probably most useful for those in the accumulation phase. Having constant inflows allows frequent high-basis purchases across many sectors, making it easy to TLH on even minor dips in share price. Also, for high earners (in the accumulation phase) the TLH is just worth more. And at the end of the working years you'd have a good assortment of "lots" with volatile share prices and widely varying cost basis that would provide a way to tailor taxable income during the drawdown years.

I like the concept. This same approach is ideal for converting IRAs to Roth IRAs (assuming the conversion has a tax cost). Splitting up conversions into different asset classes offers more opportunity to recharacterize and reconvert at a lower cost if one sector drops.

Thanks, that's a good angle, and one I could exploit at zero (additional) cost. Also, the tax savings from those losses would be at the earned income rate, so they are especially valuable.

I guess the questions for me will be:
-- How much higher will overall expenses be if I'm invested in these funds rather than the more efficient megaindex funds/ETFs?
-- If there's an expected payoff, is it worth the effort? Honestly, it's not a "hobby" I'd like to leave behind for DW once I shuffle off this mortal coil.

 

[CoolChange]

Many thanks to samclem for starting this thread since I am thinking along the same lines but was afraid that I was just making my life more complicated for no reason.

This is approach could be doubly beneficial for me since I continue to extend my accumulation phase (OMY syndrome which I cannot seem to cure) and likely bad timing/investing decisions (actually, lack of decision/action) leading to very large cash reserves currently.

 

[daylatedollarshort]

I am thinking about individual stocks for the same reason. Not market timing - just my own little index fund with a smaller number of stocks and no management or recurring fees.

 

[samclem]

I am thinking about individual stocks for the same reason. Not market timing - just my own little index fund with a smaller number of stocks.

That can be done, but be careful out there. A lot of people grossly underestimate the number of stocks it takes to really capture the performance of an index--its a lot more than a few dozen. From this (as always) enlightening and entertaining piece by William Bernstein. It was written over a decade ago, but I'd bet it is still accurate. Snippets:

To be blunt, if you think that you can do an adequate job of minimizing portfolio risk with 15 or 30 stocks, then you are imperiling your financial future and the future of those who depend on you. The reason is simple: There are critically important dimensions of portfolio risk beyond standard deviation. The most important is so-called Terminal Wealth Dispersion (TWD). In other words, it is quite possible (in fact, as we shall soon see, quite easy) to put together a 15-stock or 30-stock portfolio with a very low SD, but whose lousy returns will put you in the poorhouse.
. . .
In order to investigate this problem, I looked at the stocks constituting the S&P 500 as of 11/30/99, and formed 98 random equally-weighted 15-stock portfolios for the 12/89-11/99 10-year holding period. . . the TWD of these 15-stock portfolios is staggering—three-quarters of them failed to beat "the market."
. . .
The reason is simple: a grossly disproportionate fraction of the total return came from a very few "superstocks" like Dell Computer, which increased in value over 550 times. If you didn’t have one of the half-dozen or so of these in your portfolio, then you badly lagged the market.
. . .
So, yes, Virginia, you can eliminate nonsytematic portfolio risk, as defined by Modern Portfolio Theory, with a relatively few stocks. It’s just that nonsystematic risk is only a small part of the puzzle. Fifteen stocks is not enough. Thirty is not enough. Even 200 is not enough. The only way to truly minimize the risks of stock ownership is by owning the whole market.

 

[haha]

I'm wondering if recent changes may make it a good idea for some investors to consider abandoning broad index funds/ETFs in favor of more narrow sector MFs/sector ETFs in their taxable accounts. For example, rather than investing 50% of our equity money in the Vanguard 500 Index Fund (VFINX), we might choose to split the same amount among the 9 SPDR sector fund ETFs that cover the same stocks, but by sector (Financials, Energy, Consumer Staples, etc)

Why?
1) Ability to better manage taxable income. The quarterly/annual returns of a broad index will be a lot less volatile than the individual returns of the sectors. So, there will more frequent opportunities to tax loss harvest in taxable accounts.
The chart below (.PDF also at this link) is set up like the familliar Callan Periodic Table of Investment returns, but it is just for the 9 sub-sectors within the S&P 500. Between 2004 and 2013, the overall S&P 500 index year-end share prices had only one year when share prices declined (2008). The table below shows that over the same 10 years there were sectors within the S&P 500 that had losses in 4 of those years. Of course, if we looked at monthly or quarterly results we'd expect to see even more opportunities to harvest tax losses. In all cases, opportunities to harvest losses can be expected to be best if there are frequent new purchases of equities (to assure there are higher-basis shares on hand to allow short "dips" to be harvested)

So, investing in these sectors would have offered much more frequent opportunities to do tax loss harvesting. And this is probably a relatively modest example--more "exotic" asset classes (e.g. foreign equities held in regional sector ETFs rather than a larger, broad foreign index, etc) could be expected to be even more volatile. The >overall< risk won't be any higher (you'd be invested in the same stocks whether you've invested in a broad index or narrow sector ETFs), it's just the tax loss harvesting opportunities are enhanced.

While I keep saying "tax loss harvesting," obviously there is also potential to "reset" the cost basis on "winners" in years where a taxpayer might have some headroom below their expected future tax brackets. This can be expected to help reduce future tax liabilities.

2) It's easier now: With the new requirement that brokers track cost basis, and the very convenient pop-up menus that are common at broker websites for selecting which shares to sell, the "specific ID" basis method is a breeze to use. Gone are the days of manually entering this stuff onto spreadsheets, or writing letters to the fund companies (really for the IRS) to identify which shares are being sold. As lazy as I am, I'm now using the "specific ID" cost basis method in my taxable accounts.

3) It's potentially more important/lucrative to reduce taxable income than it has been in years past. The 0% Cap Gains rate has been made permanent for those of us "fortunate" enough to qualify, the OMAGI limits for qualifying for premium subsidies and assistance with OOP medical expenses, and even the revised qualifying criteria for Medicaid all increase the need to reduce taxable income for some of us. It looks like this might be such a way.

Other: As noted on the S&P marketing literature (at the link above), the SPDR sectors are not "cap weighted"--some are much, much smaller than others. To replicate the S&P 500, an investor would need to rebalance the amount invested in each sector occasionally. As the advertising info at the link shows, overall returns for the last 10 years have been significantly better for investors who didn't cap weight the sectors, but instead just plunked equal amounts (11.1%) into each one:
10 year cumulative return:
................ S&P 500: 107.37%
.......Equal Weighting: 140.76%
Comment: This strikes me as likely a bit of data mining, and possibly the result of the big declines that the heavily-weighted Financials sector took in 2007 and 2008. In effect, "equal weighting" amounts to a "sector bet" on the sectors that just happen to have lower capitalizations, and these are strictly an arbitrary result of the way each sector was defined when it was constructed. It happened to do well over the last 10 years, but I don't see any particular reason to expect that to continue.

Anyway, that's it. We've touched on this idea in previous threads, and I suppose there's nothing especially novel about it. Also, I haven't looked into various smart ways of implementing it (e.g these SPDR sector ETFs are probably the best known, but the very similar Vanguard sector ETFs (VCR, VFH, etc) appear to have slightly lower ERs). Also, I wonder if one could remain effectively fully invested and avoid "wash sales" by using the proceeds from the sale of a Vanguard sector ETF (e.g VHF--Financials) to buy the corresponding SPDR sector ETF (e.g. XLF). They are based on different indexes (MSCI and S&P, respectively), different CUSIPs, etc, so I'm pretty sure that's enough for the IRS.

Good post. This is the reason why I originally migrated from mutual fund investing to individual stocks. Much greater variability, giving much greater control of realized gains/losses.


Will be interesting to see group consensus on this.

Ha

 

[NW-Bound]

That can be done, but be careful out there. A lot of people grossly underestimate the number of stocks it takes to really capture the performance of an index--its a lot more than a few dozen. From this (as always) enlightening and entertaining piece by William Bernstein. It was written over a decade ago, but I'd bet it is still accurate. Snippets:

...
The reason is simple: a grossly disproportionate fraction of the total return came from a very few "superstocks" like Dell Computer, which increased in value over 550 times. If you didn’t have one of the half-dozen or so of these in your portfolio, then you badly lagged the market.
. . .
So, yes, Virginia, you can eliminate nonsytematic portfolio risk, as defined by Modern Portfolio Theory, with a relatively few stocks. It’s just that nonsystematic risk is only a small part of the puzzle. Fifteen stocks is not enough. Thirty is not enough. Even 200 is not enough. The only way to truly minimize the risks of stock ownership is by owning the whole market.

While I agree that it takes a lot of individual stocks to minimize risks, I would like to point out that owning the entire market means that in go-go years you would be sure to enjoy the gains of the true destructive-innovators like Intel as well as frothy dot-coms. The other side of the coins is that you would suffer the crash of MCI-Worldcom, Bear Stearns, and the likes.

Many value-oriented MF managers lagged the market in the go-go years, then made it up in the lost decade.

Good post. This is the reason why I originally migrated from mutual fund investing to individual stocks. Much greater variability, giving much greater control of realized gains/losses.

Will be interesting to see group consensus on this.

Ha

I can predict with some certainty that the majority of investors will stick with indexing or MFs. Active investing takes too much work for most people, and the truth is that one can do decently with passive investing. I like the DYI method, because I like to see how things work. For me, it's not too different than trying to cook something at home instead of going to a restaurant.

 

[daylatedollarshort]

That can be done, but be careful out there. A lot of people grossly underestimate the number of stocks it takes to really capture the performance of an index--its a lot more than a few dozen. From this (as always) enlightening and entertaining piece by William Bernstein. It was written over a decade ago, but I'd bet it is still accurate. Snippets:

I am planning on what most posters here would consider lousy returns. They are already built into our retirement plan so I don't think it will be a problem.

My grandparents lived long lives and stayed financially secure throughout retirement, including years of high inflation, and they just invested in the basics from their era - dividend stocks, individual bonds, CDs, etc. I have been thinking of just investing more like they did. In fact, my grandfather lived an exceptionally long life and left a substantial inheritance, even though he never had an executive level type job or stock options and my grandmother never worked outside the home.

I don't really like the idea of potentially losing half my life savings in any given year. I'm going with the we've won the game why take much risk approach. Part of my GPs financial security was that they just didn't need much to live - paid off, modest house, one car, cheap hobbies, depression era spending habits and a part time, low stress retirement job for my grandfather, even though he didn't really need the money.

 

[clifp]

Good post. This is the reason why I originally migrated from mutual fund investing to individual stocks. Much greater variability, giving much greater control of realized gains/losses.


Will be interesting to see group consensus on this.

Ha

Well this plus I like to have something to add to the "what did you do all day thread ."

I don't think there will ever be a group consensus. I enjoy picking individual stocks and intend to keep doing for another 20 years or so. ACA gives me another reason to continue.

 

[walkinwood]

Seems like a lot of work to me, even though your reasoning is sound.

I wouldn't try it unless I saw some back-testing using a rigorous methodology. Index funds have tracking error - how does that work here. Do the individual tracking errors cancel each other out or are they cumulative? What about expenses?

 

[samclem]

I wouldn't try it unless I saw some back-testing using a rigorous methodology.

Since we're not striving for any change in before-tax investment returns, but only a way to reduce taxes, the back-testing would have to very specific to individual cases (tax bracket for cap gains, time held/basis at purchase, etc), so it would be tough to model. I guess someone could do it, but it is clear that using these sector funds (or individual stocks) will provide more opportunities to do TLH than investing in broader indexes.

Index funds have tracking error - how does that work here. Do the individual tracking errors cancel each other out or are they cumulative? What about expenses?

Tracking error doesn't bother me much, as long as it is small and random (not systematic). Regarding costs: The ER's of these index sector ETFs can be very low--most of the Vanguard offerings are 0.14%. But that's not as cheap as buying broader index (VGD TSM ETF : 0.05%). So, costs would be higher by about 0.1% per year on this portion of the portfolio (about $100 for a $100,000 investment)

 

[RobLJ]

SamCl:
I've pondered this for almost 15 years and do a version, in which I have too many core funds and have branched out in narrow areas (Biotech, Latin America, Materials, Floating Funds, Emerging Market and foreign bonds in bond allocation) to target areas that I think are under-valued or longer term values (based on a theme). China and Biotech were theme/value picks that took 5 years of accumulation before they paid off, but when they paid, they paid big. Biotech is not the norm, at all, but it and health over the last year are about 4% of my holdings but 13% of returns. China functioned similarly before I sold most in '06 and also Europe, also before the Crash.
My lesson--which may not be correct--is to harvest gains, at least by harvesting gains to reduce risk in steps half-way or completely back to your original position. I sold all of China in '07, based on a risk assessment, so China is not a good example, although I'm grateful for the returns. I moved back into it in a small way late last year and will do so again in Europe in a few weeks and next year.

To be clear, I have much larger core positions in the normal stock areas (Large Growth, small value, international) and add smaller positions in targeted "undervalued" areas that I assume will take 3-6 years to pay off, if ever.
Don't do what I say; do your own diligence. Also, if it isn't 3% or more of your allocation, it may not be worth the trouble, unless you are dollar cost averaging monthly, which I did originally in China and bio-tech 8 and 11 years ago. When these did pay, however, they paid quickly and well.

Lowell Herr at Seeking Alpha has a series of very sophisticated posts on a version of this based on portfolio theory. I think his approach is too theoretical and complicated, but it is a fascinating read and actually is a more justifiable version of my more informal approach. He does not use a core + periphery approach that I do, but a smaller percentages of less-correlated ETFs, which drives down costs, which is an elegant (in my mind, at least) approach. He is closer to what you are suggesting than my approach above.

I'm wondering if recent changes may make it a good idea for some investors to consider abandoning broad index funds/ETFs in favor of more narrow sector MFs/sector ETFs in their taxable accounts. For example, rather than investing 50% of our equity money in the Vanguard 500 Index Fund (VFINX), we might choose to split the same amount among the 9 SPDR sector fund ETFs that cover the same stocks, but by sector (Financials, Energy, Consumer Staples, etc)

Why?
1) Ability to better manage taxable income. The quarterly/annual returns of a broad index will be a lot less volatile than the individual returns of the sectors. So, there will more frequent opportunities to tax loss harvest in taxable accounts.
The chart below (.PDF also at this link) is set up like the familliar Callan Periodic Table of Investment returns, but it is just for the 9 sub-sectors within the S&P 500. Between 2004 and 2013, the overall S&P 500 index year-end share prices had only one year when share prices declined (2008). The table below shows that over the same 10 years there were sectors within the S&P 500 that had losses in 4 of those years. Of course, if we looked at monthly or quarterly results we'd expect to see even more opportunities to harvest tax losses. In all cases, opportunities to harvest losses can be expected to be best if there are frequent new purchases of equities (to assure there are higher-basis shares on hand to allow short "dips" to be harvested)

So, investing in these sectors would have offered much more frequent opportunities to do tax loss harvesting. And this is probably a relatively modest example--more "exotic" asset classes (e.g. foreign equities held in regional sector ETFs rather than a larger, broad foreign index, etc) could be expected to be even more volatile. The >overall< risk won't be any higher (you'd be invested in the same stocks whether you've invested in a broad index or narrow sector ETFs), it's just the tax loss harvesting opportunities are enhanced.

While I keep saying "tax loss harvesting," obviously there is also potential to "reset" the cost basis on "winners" in years where a taxpayer might have some headroom below their expected future tax brackets. This can be expected to help reduce future tax liabilities.

2) It's easier now: With the new requirement that brokers track cost basis, and the very convenient pop-up menus that are common at broker websites for selecting which shares to sell, the "specific ID" basis method is a breeze to use. Gone are the days of manually entering this stuff onto spreadsheets, or writing letters to the fund companies (really for the IRS) to identify which shares are being sold. As lazy as I am, I'm now using the "specific ID" cost basis method in my taxable accounts.

3) It's potentially more important/lucrative to reduce taxable income than it has been in years past. The 0% Cap Gains rate has been made permanent for those of us "fortunate" enough to qualify, the OMAGI limits for qualifying for premium subsidies and assistance with OOP medical expenses, and even the revised qualifying criteria for Medicaid all increase the need to reduce taxable income for some of us. It looks like this might be such a way.

Other: As noted on the S&P marketing literature (at the link above), the SPDR sectors are not "cap weighted"--some are much, much smaller than others. To replicate the S&P 500, an investor would need to rebalance the amount invested in each sector occasionally. As the advertising info at the link shows, overall returns for the last 10 years have been significantly better for investors who didn't cap weight the sectors, but instead just plunked equal amounts (11.1%) into each one:
10 year cumulative return:
................ S&P 500: 107.37%
.......Equal Weighting: 140.76%
Comment: This strikes me as likely a bit of data mining, and possibly the result of the big declines that the heavily-weighted Financials sector took in 2007 and 2008. In effect, "equal weighting" amounts to a "sector bet" on the sectors that just happen to have lower capitalizations, and these are strictly an arbitrary result of the way each sector was defined when it was constructed. It happened to do well over the last 10 years, but I don't see any particular reason to expect that to continue.

Anyway, that's it. We've touched on this idea in previous threads, and I suppose there's nothing especially novel about it. Also, I haven't looked into various smart ways of implementing it (e.g these SPDR sector ETFs are probably the best known, but the very similar Vanguard sector ETFs (VCR, VFH, etc) appear to have slightly lower ERs). Also, I wonder if one could remain effectively fully invested and avoid "wash sales" by using the proceeds from the sale of a Vanguard sector ETF (e.g VHF--Financials) to buy the corresponding SPDR sector ETF (e.g. XLF). They are based on different indexes (MSCI and S&P, respectively), different CUSIPs, etc, so I'm pretty sure that's enough for the IRS.

 

[photoguy]

I have a very basic question. Does the capital loss carry-forward *have* to be taken against capital gain in future years? Or can one book a large capital loss in say year 1, pay capital gains taxes in year 2 (with no offset), and wait until year 3 to use the loss?

 

[Rustward]

I have a very basic question. Does the capital loss carry-forward *have* to be taken against capital gain in future years? Or can one book a large capital loss in say year 1, pay capital gains taxes in year 2 (with no offset), and wait until year 3 to use the loss?

It must be applied each year. You cannot choose to skip a year. If you do not apply the loss in a tax year, then you will have no loss to carry forward to future years.

Sorry. I did not write the tax code.

 

[photoguy]

Thanks for the explanation Rustward.

I guess it's not going to be possible to make a cookie jar of losses (I'm primarily thinking for ACA limits).

 

[Midpack]

Not claiming it would be as tax efficient as the OP's approach, but I use Vanguard Tax Managed Funds and tax loss harvesting to reduce tax consequences. And I know some folks here hold Vanguard Tax Exempt funds as well. Other options for folks who may not want to go all out with sector slice and dice.

 

[haha]

I wouldn't try it unless I saw some back-testing using a rigorous methodology.

I doubt you ever will, so you are set as is.

Ha

 

[pb4uski]

samclem, I'm actually thinking about doing what you describe, but not for the tax benefits.

I'm doing it more based on a view that if rebalancing forces one to buy low and sell high, that to rebalance annually across S&P sectors would result in higher returns over time that simply holding the S&P 500 index.

Also see http://www.early-retirement.org/forums/f28/something-im-thinking-about-65096.html

I need to think about the tax benefits some - it might make it juicer as I was more worried about rebalancing taxable gains so I was initially planning to focus on equities in my tax-deferred accounts.

 

[samclem]

I'm doing it more based on a view that if rebalancing forces one to buy low and sell high, that to rebalance annually across S&P sectors would result in higher returns over time that simply holding the S&P 500 index.

Also see http://www.early-retirement.org/forums/f28/something-im-thinking-about-65096.html

I need to think about the tax benefits some - it might make it juicer as I was more worried about rebalancing taxable gains so I was initially planning to focus on equities in my tax-deferred accounts.

Thanks for the link to that previous, related thread. Your OP there mentioned the recent better performance from an equal weighting rather than a cap-based weighting of the sectors, and I'd just comment as I did earlier here: It seems plausible that was just a reflection of the fact that an equal-weighting amounts to under-weighting the Financials, and they got clobbered in 2007 and 2008. If that's right, the overperformance of the "equal weighting" approach is just a random artifact of the way these sector funds are put together and there's no reason to expect that performance to continue.
I know it is common knowledge here, but just to restate it: rebalancing among asset classes does not result in improved returns. Returns would be enhanced if we never rebalanced. The primary reason to rebalance is to reduce volatility. Rebalancing usually results in the selling of higher-return assets (e.g. emerging market stocks) to buy lower-return assets (e.g. domestic bonds), and this hurts returns. Now, I don't know if that same observation holds true between domestic equity sectors, it would depend on whether they have fundamentally different expected returns (and risk).
Anyway, it appears that if you pursue the sector-based approach you are exploring that there could be some tax benefits, which is a good thing!

SamCl:
I've pondered this for almost 15 years and do a version, in which I have too many core funds and have branched out in narrow areas (Biotech, Latin America, Materials, Floating Funds, Emerging Market and foreign bonds in bond allocation) to target areas that I think are under-valued or longer term values (based on a theme).

I'm digesting Lowell Herr's writings, thanks for that. Your less rigorous approach would scratch the itch many of us have to [-]make a bet[/-]strategically invest in niche areas that look promising. One thoguht that occurs to me: Keeping the bets narrow (rather than, say, putting all the dough in Mr Smartguy's Surefire Rising Star Fund) would have the advantage of being more likely to generate losses that could be selected and harvested for tax purposes, rather than having them mixed in with the "winners" and thus not exploitable.

I think I'm in danger of letting the "tax tail" wag the "return" dog.

 

[Animorph]

I think this is where I first started with rebalancing, though he's showing smaller intervals than others I was looking at:

Rebalancing

"

CONCLUSIONS:
1. The expected return of a rebalanced portfolio is not accurately represented by simple arithmetic weighting of individual asset returns. This is particularly true of assets which have high standard deviations and are poorly correlated. This may be of some importance to those utilizing optimization technology which depends on linear expected return determinations. Standard MVO techniques underestimate the benefit of high variance/low correlation assets.
2. The capitalization weighting suggested by the CAPM does not appear to be maximally efficient.
3. A formula is presented which accurately predicts rebalancing benefit. The inputs to this equation are portfolio composition, individual asset standard deviation/variance, and asset pair correlation/covariance. Since these parameters vary with sampling interval, it may be possible to use this formula to predict the optimal rebalancing frequency. Further, the results of the analyses presented suggest that shorter rebalancing intervals may not always be optimal. 4. Optimizers should understand that sampling interval will significantly affect both input and output data, and that he or she is optimizing only for standard deviatons (or other risk measures, such as semivariance) over a given sampling interval. "

 

[samclem]

There's a good discussion on the purpose and value of rebalancing on the Bogleheads board (link). To steal from one post by "SmallHi" there:

Technically speaking, the rebalancing bonus (or loss) and the diversification return (or return due to diversification, RDD) are 2 different things.

The rebalancing bonus can be positive or negative, while the RDD cannot (on a pretax basis).

The rebalancing bonus (or loss) is simply the difference in return between two assets (or more) that are rebalanced periodically, vs the same combination that is not rebalanced at all.

For example, since 1926, a 50/50 unrebalanced portfolio of S&P 500 and T-Notes compounded at +9.5% a year. Had the same investor rebalanced annually back to the 50/50 weighting, his return would only be +8.4%. This is a rebalancing loss (not a bonus) of 1.1% a year.

Now, if we take a 50/50 unrebalanced portfolio of S&P 500 and EAFE since 1970, we find the non-rebalancing investor compounded at 11.1% a year. The investor who rebalanced annually back to 50/50 made 11.5% a year. This is a rebalancing bonus of 0.4% a year.

Sometimes anual rebalncing results in gains, sometimes it doesn't.