Minimum % for a "tilt" to matter in portfolio

Have a question about your personal investments? No matter how simple or complex, you can ask it here.
Post Reply
Sirjames
Posts: 34
Joined: Tue Sep 12, 2017 1:53 pm

Minimum % for a "tilt" to matter in portfolio

Post by Sirjames » Fri Mar 09, 2018 12:11 pm

I've heard anything less than 5% is likely not worth it... but is 5% even enough? Only the future knows what the future brings... so hindsight in 30 years will be 20/20.

I'm tilting about 5-8% "extra" over market caps toward domestic REIT (VNQ), international REIT (VNQI), small cap domestic (VBR and IJR), small cap international (last to start DCA in... deciding between SCZ, VSS, and/or DLS for later this month), and emerging markets (VWO and IEMG).. so our (me + wife) overall portfolio is therefore ~70-75% "plain vanilla" market cap indexing and ~25-30% "tilts" in five slices.

What are others' approaches to tilts?
Patience is the ultimate hedge

JBTX
Posts: 4281
Joined: Wed Jul 26, 2017 12:46 pm

Re: Minimum % for a "tilt" to matter in portfolio

Post by JBTX » Fri Mar 09, 2018 12:17 pm

Sirjames wrote:
Fri Mar 09, 2018 12:11 pm
I've heard anything less than 5% is likely not worth it... but is 5% even enough? Only the future knows what the future brings... so hindsight in 30 years will be 20/20.

I'm tilting about 5-8% "extra" over market caps toward domestic REIT (VNQ), international REIT (VNQI), small cap domestic (VBR and IJR), small cap international (last to start DCA in... deciding between SCZ, VSS, and/or DLS for later this month), and emerging markets (VWO and IEMG).. so our (me + wife) overall portfolio is therefore ~70-75% "plain vanilla" market cap indexing and ~25-30% "tilts" in five slices.

What are others' approaches to tilts?
That seems like a reasonable approach to me, as long as you aren't paying a bunch of fees on your tilt funds.

What would be interesting is to see how your potential allocations line up towards a vanguard life strategy fund. You may not find it is much different.

User avatar
David Jay
Posts: 5811
Joined: Mon Mar 30, 2015 5:54 am
Location: Michigan

Re: Minimum % for a "tilt" to matter in portfolio

Post by David Jay » Fri Mar 09, 2018 12:26 pm

I call anything 5% or less "window dressing" - it dresses up the portfolio but can hardly impact the overall performance. It's the math. If one sector outperforms the rest of your portfolio by 500 basis points (5%) over a given period, your portfolio sees a 25 basis point improvement.

10% is about the point where one begins to see a significant impact. Serious tilting is 20%- 30% (or more, as in the case of the Larry portfolio).
Prediction is very difficult, especially about the future - Niels Bohr | To get the "risk premium", you really do have to take the risk - nisiprius

livesoft
Posts: 63093
Joined: Thu Mar 01, 2007 8:00 pm

Re: Minimum % for a "tilt" to matter in portfolio

Post by livesoft » Fri Mar 09, 2018 2:24 pm

One can also think of an extra 10% allocation to equities to be a tilt. For example, suppose you decided your asset allocation should be 50:50 stocks:bonds. Then you decide to tilt more to stocks, so you end up at 60:40. That extra 10% to stocks might do you some good, but think about this: For a 1% per year benefit, that extra 10% has to go up 10% because 10% of 10% is 1%.

Now if you just decide to stick with 50:50 and move some of that 50% of stocks to another stock sector, you are taking away from Total Market Stock to add that stock sector. Since the two will be highly correlated, the gain will be expected to be less than moving bonds to stocks, so say the sector gives only 5% a year extra and you moved 10% of total portfolio, so your annual performance boost might be just 0.5%.

You can probably gain 0.5% by reducing the expenses of your investments, reducing the taxes on your investments, and ditching any advisor if you are using one. So before you do any kind of tilting, I would recommend that you demonstrate that you can do a total market weights portfolio first with very low expense ratios and very little, if any, taxes.

And you might just get that 0.5% from a combination of tax-loss harvesting, well-timed rebalancing, and even some market timing to temporarily higher stock allocation.

But it can all be worse. Just being in US equities when they do better than international or vice versa can blow away any tilts you want. Or you can be tilted the wrong way. Just ask the folks who own IJS this year and last instead of MTUM.
Wiki This signature message sponsored by sscritic: Learn to fish.

Sirjames
Posts: 34
Joined: Tue Sep 12, 2017 1:53 pm

Re: Minimum % for a "tilt" to matter in portfolio

Post by Sirjames » Sat Mar 10, 2018 12:54 pm

livesoft wrote:
Fri Mar 09, 2018 2:24 pm
One can also think of an extra 10% allocation to equities to be a tilt. For example, suppose you decided your asset allocation should be 50:50 stocks:bonds. Then you decide to tilt more to stocks, so you end up at 60:40. That extra 10% to stocks might do you some good, but think about this: For a 1% per year benefit, that extra 10% has to go up 10% because 10% of 10% is 1%.

Now if you just decide to stick with 50:50 and move some of that 50% of stocks to another stock sector, you are taking away from Total Market Stock to add that stock sector. Since the two will be highly correlated, the gain will be expected to be less than moving bonds to stocks, so say the sector gives only 5% a year extra and you moved 10% of total portfolio, so your annual performance boost might be just 0.5%.

You can probably gain 0.5% by reducing the expenses of your investments, reducing the taxes on your investments, and ditching any advisor if you are using one. So before you do any kind of tilting, I would recommend that you demonstrate that you can do a total market weights portfolio first with very low expense ratios and very little, if any, taxes.

And you might just get that 0.5% from a combination of tax-loss harvesting, well-timed rebalancing, and even some market timing to temporarily higher stock allocation.

But it can all be worse. Just being in US equities when they do better than international or vice versa can blow away any tilts you want. Or you can be tilted the wrong way. Just ask the folks who own IJS this year and last instead of MTUM.
Thank you all for your opinions thus far. To counter, for discussion-sake, if I am in fact minimizing costs already (low ER, commission-free), wouldn't there be a benefit over the years to buying low? For example, this year, VNQ remains down while the greater market is rebounding... I'm able to take advantage of that by bringing VNQ back up to its 5-6% proportion of my portfolio, and buying those shares on a "relative sale." Maybe I need to increase my tilts to 8-12% across the board to see a sustained benefit.

The more significant factor, as you mention, would be the equity:bonds ratio. I am going 100% for at least 10-15 more years (31 years old now) to accumulate equity shares and "time in" equity markets. Not worried about selling low in bad years.. I won't. Would welcome cheaper costs.

My portfolio general allocation, all low-cost and commission-free:

Within each category- overall %, category %
  • Domestic- 54.0%
Large cap 31.3%, 58.0%
Mid cap 11.9%, 22.0%
Small cap 10.8%, 20.0%
  • International- 37.0%
Developed 22.9%, 62.0%
Emerging 9.3%, 25.0%
Small cap 4.8%, 13.0%
  • Alternatives- 9.0%
REIT Domestic 5.0%, 55.6%
REIT Global 4.0%, 44.4%

12-14 funds/ETFs for the above in TSP 401K, Fidelity 403B, Fidelity Roth IRA, Vanguard Roth IRA, and an HSA

Cumulative weighted-expense ratio for all holdings= 0.085 -- I could get this down to 0.04-0.05 overall if I switch to broad domestic and international indexes across the board. I don't know that this savings of $50-100 per year should outweigh the "potential" missing out on the over-performance (20-30% over respective broad domestic or international index) of those tilts in certain years. Expecting the over-performance (-->would sell high) and under-performance (--> would buy low) of these tilts seems like a reasonable expectation according to the Callan table (https://www.bogleheads.org/wiki/Callan_ ... nt_returns). I see both sides of this so it's a tough decision..

My picture will change in a couple of years when income will increase substantially, allowing us to annually max out tax-advantaged space and start a Vanguard taxable account (in which will likely keep it simple with 60:40 VTI:VXUS). Taxable account will be monthly contributions. At that point our Roth IRAs will be backdoor, and I may no longer have the energy to do multiple backdoor-conversions throughout the year to take advantage of the tilts as much.

Thanks again, appreciate your contributions to the site!
Patience is the ultimate hedge

livesoft
Posts: 63093
Joined: Thu Mar 01, 2007 8:00 pm

Re: Minimum % for a "tilt" to matter in portfolio

Post by livesoft » Sat Mar 10, 2018 1:06 pm

Since you brought up REITs and VNQ, let me write that I am on record here at bogleheads.org that I would not buy VNQ at the present time. I have owned VNQ in the past and will own it in the future. But I think there are better things to own right now than REIT funds. I will buy VNQ when it drops at least 3% to 4% in a single day.

For those folks who had a 5% allocation to VNQ over the past 3 years, their VNQ has underperformed Total Stock Market by 10% a year which has made their portfolio suffer a 0.5% annual underperformance. This is relatively huge. If they had a 20% allocation to VNQ, then that is a 2% per year underperformance.

So if you are going to tilt, then at least tilt to the asset class that is going to do well going forward. And of course the problem is that one cannot predict the future and guess with any certainty which asset class(es) will do well going forward.

Another example, small-cap value as represented by Vanguard Small-cap value index fund (VSIAX) has underperformed VTSAX (total stock market) by about 2% per year over the last 3- and 5-year periods. If one had a 30% allocation to SCV, then that would be 0.6% underperformance for 5 years. One might be able to make up the 0.6% underperformance in other ways such as rebalancing at opportune times and some market timing.
Wiki This signature message sponsored by sscritic: Learn to fish.

Sirjames
Posts: 34
Joined: Tue Sep 12, 2017 1:53 pm

Re: Minimum % for a "tilt" to matter in portfolio

Post by Sirjames » Sat Mar 10, 2018 1:29 pm

livesoft wrote:
Sat Mar 10, 2018 1:06 pm
Since you brought up REITs and VNQ, let me write that I am on record here at bogleheads.org that I would not buy VNQ at the present time. I have owned VNQ in the past and will own it in the future. But I think there are better things to own right now than REIT funds. I will buy VNQ when it drops at least 3% to 4% in a single day.

For those folks who had a 5% allocation to VNQ over the past 3 years, their VNQ has underperformed Total Stock Market by 10% a year which has made their portfolio suffer a 0.5% annual underperformance. This is relatively huge. If they had a 20% allocation to VNQ, then that is a 2% per year underperformance.

So if you are going to tilt, then at least tilt to the asset class that is going to do well going forward. And of course the problem is that one cannot predict the future and guess with any certainty which asset class(es) will do well going forward.

Another example, small-cap value as represented by Vanguard Small-cap value index fund (VSIAX) has underperformed VTSAX (total stock market) by about 2% per year over the last 3- and 5-year periods. If one had a 30% allocation to SCV, then that would be 0.6% underperformance for 5 years. One might be able to make up the 0.6% underperformance in other ways such as rebalancing at opportune times and some market timing.
Would you feel the same way for someone who is investing money they won't take out for 30-50 years? We all would bet that VNQ will have a much higher NAV in 20-30 years than it does now, therefore reinforcing the idea that a long-term investor is buying today's shares at a massive sale. Even moreso if it continues to underperform for another 5-10 years..

Or are you arguing that the "opportunity cost" of my invested dollars in VNQ for years (as opposed to a Total US holding) will not be overcome by the eventual rebound of VNQ?

Obviously the timing/magnitude of these things are impossible to predict...
Patience is the ultimate hedge

livesoft
Posts: 63093
Joined: Thu Mar 01, 2007 8:00 pm

Re: Minimum % for a "tilt" to matter in portfolio

Post by livesoft » Sat Mar 10, 2018 1:44 pm

Sirjames wrote:
Sat Mar 10, 2018 1:29 pm
Or are you arguing that the "opportunity cost" of my invested dollars in VNQ for years (as opposed to a Total US holding) will not be overcome by the eventual rebound of VNQ?
That it what I am arguing.

VNQ is actually traded quite a lot by presumably day traders, so some of them are making some money with it. I don't think it is a place for buy and hold investors. Even when I buy it, I do not intend to hold it a long time.

So instead of tilting to REITs, how about tilting with MTUM?

And since you are tilting with a 100% equity portfolio, how is it going for past 12 months? YTD? You can tell us whether it matters if one tilts or not (or as least whether it mattered in the recent past).
Wiki This signature message sponsored by sscritic: Learn to fish.

michaeljc70
Posts: 3932
Joined: Thu Oct 15, 2015 3:53 pm

Re: Minimum % for a "tilt" to matter in portfolio

Post by michaeljc70 » Sat Mar 10, 2018 2:02 pm

Interesting discussion. How many people will stick with tilts though if they underperform the market for years? That is the reason I got rid of mine a few years ago. It also keeps it simpler.

Sirjames
Posts: 34
Joined: Tue Sep 12, 2017 1:53 pm

Re: Minimum % for a "tilt" to matter in portfolio

Post by Sirjames » Sat Mar 10, 2018 2:33 pm

livesoft wrote:
Sat Mar 10, 2018 1:44 pm
Sirjames wrote:
Sat Mar 10, 2018 1:29 pm
Or are you arguing that the "opportunity cost" of my invested dollars in VNQ for years (as opposed to a Total US holding) will not be overcome by the eventual rebound of VNQ?
That it what I am arguing.

VNQ is actually traded quite a lot by presumably day traders, so some of them are making some money with it. I don't think it is a place for buy and hold investors. Even when I buy it, I do not intend to hold it a long time.

So instead of tilting to REITs, how about tilting with MTUM?

And since you are tilting with a 100% equity portfolio, how is it going for past 12 months? YTD? You can tell us whether it matters if one tilts or not (or as least whether it mattered in the recent past).
12 mos ago my wife and I were in a very different situation with our accounts.. I had my Roth IRA with an advisor paying commission fees and ERs of 0.8-1.1. Similarly shifted funds in 401k/403B. Wife's Roth was in about 50% cash and 50% VDE... Started an HSA.. Finding bogleheads saved us. Even YTD isn't what I want my tilts to be... Sold some VDE last year, but finding it hard to "sell low" on the rest of VDE that we're holding on to "until it rebounds." (I wish someone would just hack my account and exchange the remaining VDE for VTI!). The allocation that I listed is what I planned for our long-term plan to be.

I'm not familiar with MTUM... but do like the diversification potential there and its ER of 0.15 (although maybe some concerns with bid/ask spread). Also like the idea of removing "sector gambles" from my long term plan... but REITs were "supposed to be" the one sector that "acts differently." Does seem to have low correlation with VTI/SPY retrospectively. Will have to read more about momentum..seems fellow bogleheads feel pretty passionately for or against it, and think about your comment on "opportunity costs" remaining invested in REITs when "forecasted" to do poorly for some time.
Patience is the ultimate hedge

livesoft
Posts: 63093
Joined: Thu Mar 01, 2007 8:00 pm

Re: Minimum % for a "tilt" to matter in portfolio

Post by livesoft » Sat Mar 10, 2018 3:45 pm

When one tilts, one will be subjected to "tracking error" versus a Total Market Weights benchmark. If one finds that their tilts are hurting them, then it may lead to irrational behavior. For instance, I track my portfolio against VSMGX, VBIAX, and DGSIX which are 60/40 funds. If I get behind the performance of these funds, then I want to know it and I will do some market timing to get back ahead of them. Do you know what you will do?

Also, these 60/40 benchmark funds show the kind of spread in performance that is possible with a 60/40 asset allocation. The YTD performance of these funds is reported by Morningstar.com to be:
1.44% VSMGX
1.87% VBIAX
1.38% DGSIX

And if one finds that their tilts help them and they have a positive tracking error, then one may get a warm fuzzy feeling inside. But one cannot share that warm fuzzy feeling on bogleheads.org.
Wiki This signature message sponsored by sscritic: Learn to fish.

Sirjames
Posts: 34
Joined: Tue Sep 12, 2017 1:53 pm

Re: Minimum % for a "tilt" to matter in portfolio

Post by Sirjames » Sun Mar 11, 2018 12:46 pm

livesoft wrote:
Sat Mar 10, 2018 1:44 pm
Sirjames wrote:
Sat Mar 10, 2018 1:29 pm
Or are you arguing that the "opportunity cost" of my invested dollars in VNQ for years (as opposed to a Total US holding) will not be overcome by the eventual rebound of VNQ?
That it what I am arguing.

VNQ is actually traded quite a lot by presumably day traders, so some of them are making some money with it. I don't think it is a place for buy and hold investors. Even when I buy it, I do not intend to hold it a long time.

So instead of tilting to REITs, how about tilting with MTUM?

And since you are tilting with a 100% equity portfolio, how is it going for past 12 months? YTD? You can tell us whether it matters if one tilts or not (or as least whether it mattered in the recent past).
Below is a thread from a couple years ago regarding REITs as a long term investment to buy/hold/rebalance. Wanted to tag it here for any other readers who stumble upon this thread and, like me, find additional facts and opinions on this topic beneficial. Thanks again for your input, livesoft.

Anybody regretting holding on to your REITS?:
viewtopic.php?t=203059
Patience is the ultimate hedge

User avatar
raven15
Posts: 363
Joined: Sun Nov 30, 2014 8:01 pm

Re: Minimum % for a "tilt" to matter in portfolio

Post by raven15 » Sun Mar 11, 2018 12:59 pm

The correlation and volatility of the proposed tilt matter. Tilting VTI 10% towards VBR may only have a modest impact. However, put that same 10% into a gold miner equity fund instead and it will look like an extreme, scary, and questionable tilt. Gold mine stocks have low correlation to the broader market and around 45% annualized standard deviation. Similar for an extreme duration bond fund like EDV. For these kinds of things tilting 5% or less of your portfolio can make sense, whereas splitting your US stocks 50/50 between VBR and VTI may have a similar effect to that 5%.
It's Time. Adding Interest.

livesoft
Posts: 63093
Joined: Thu Mar 01, 2007 8:00 pm

Re: Minimum % for a "tilt" to matter in portfolio

Post by livesoft » Sun Mar 11, 2018 1:09 pm

Sirjames wrote:
Sun Mar 11, 2018 12:46 pm
Below is a thread from a couple years ago regarding REITs as a long term investment to buy/hold/rebalance. Wanted to tag it here for any other readers who stumble upon this thread and, like me, find additional facts and opinions on this topic beneficial. Thanks again for your input, livesoft.
Another thread to consider about REITs that has come up again recently:
I thought Swedroe was a fan of REITs?
Wiki This signature message sponsored by sscritic: Learn to fish.

User avatar
mhadden1
Posts: 428
Joined: Tue Mar 25, 2014 8:14 pm
Location: North Alabama

Re: Minimum % for a "tilt" to matter in portfolio

Post by mhadden1 » Sun Mar 11, 2018 1:27 pm

Apart from the listed tilts, it looks like OP is tilting;

-- drastically away from FI with 100% equities. Higher expected return, but, stocks can underperform bonds for long periods.

-- toward domestic, but not as much as I perceive that a typical BH does.

These tilts are likely to be more significant than the typical 5-10% small/value/REIT.

Also, a batch of uncorrelated tilts can easily cancel each other out. :( Or reinforce each other positively. :happy Or negatively. :(

Full disclosure: I have a REIT tilt, 10% of stocks, that I do not carefully rebalance. My affection for tilts in general has waned over the years, and yes, partly because of recent VNQ performance. I beg for the mercy of the BH court.
Oh I can't, can I? That's what they said to Thomas Edison, mighty inventor, Thomas Lindberg, mighty flyer,and Thomas Shefsky, mighty like a rose.

Post Reply