Here's a statement of the obvious: The opinions expressed here are those of the participants, not those of the Mutual Fund Observer. We cannot vouch for the accuracy or appropriateness of any of it, though we do encourage civility and good humor.
There's hysteresis built into the classification system. That is, if a fund has been holding 25% equity and moves to 35% equity, M* does not immediately change its category. It waits (three years or more) to see how permanent and how significant the change is.
The point here is that one can't look at an average over time and determine how the fund is classified. The classification is "path dependent" - it depends on how it got there. A fund that hit 30% from 25% and stuck there would likely remain in its 15% to 30% category, while a fund that hit 30% from 35% and stuck there would likely remain in its 30% to 50% category.
If the fund is bouncing back and forth between a tad under and a tad over 30%, there would be no reason to change its category. Even if those oscillations miraculously averaged exactly 30% equity.
The obvious problem being that there are funds that sit on 30% equity historically. Basic only allows a hard filter for either 30-50 or 15-30 screens. An investor searching for 15-30% equity may find a suitable fund sitting on the lowest edge of 30-50% category that gets screened out in basic. AOK is an example. In the last 5 years they dipped as low as 27% equity but classified today as 30-50% by M*. I have reclassified on my watch list as 15-30%. AOK is a good fund IMO. Seems to me this would affect the funds performance reputation as M* is quoted often. There is no perfect world, just worth noting. I am sure also applies to 50-70%. I will pay more attention going forward.
The fact that AOK has dipped as low as 27% does not mean that historically its equity allocation has been under 30%. You seem to be conflating single moments in time with "historical" positions.
What does "historical" mean anyway? "the last 5 years" is a minority of the lifetime of this fund that's been around since 2008.
I have reclassified on my watch list as 15-30%. It's currently at 31.75% equity (as of Dec 31), and rose as high as 32.5% equity as of July 31, according to its most recent annual statement.
There are always going to be problems running screens with "hard filters". I've posted on more than one occasion that IMHO it makes little sense to screen for funds that have never had a losing year. Which would you prefer: a fund that lost 0.1% in one year and made 10% or better in all its other years, or a fund that made 3% year in, year out? That's an example of a problem with any hard filter.
As far as AOK goes, according to Lipper there are only a total of 10 ETFs including both "Mixed-Asset Target Allocation Moderate" and "Mixed-Asset Target Allocation Conservative" categories. So it's no big deal to watch all these funds. http://www.funds.reuters.wallst.com/us/screener/screener.asp
M* Historical stock style for AOK shows: 2020 30% equity allocation 2019 30% 2018 27% 2017 28% 2016 29% More samples: 2015 31% Form N-CSR p.7 SEC filing 2013 30% Form N-CSR p.4 2010 22% Form N-CSR p.3 2009 21% Form N-CSR p.3
AOK clearly sits historically near the 30% low end of the 30-50% category. The iShares AOK website shows the benchmark index is S&P Target Risk Conservative Index. This Index construction shows a 30% equity allocation (2020 prospectus p. S-2). M* places it in the 30-50% category. Takeaway: keep your eye out for funds historically near the category weighting limits as it might still be within your risk tolerance and more importantly possibly could provide good performance metrics you are seeking.
"near" is good. It allows for fuzziness and the possibility of bouncing around the figure in a way that "on" 30% and "holding" 30% don't. holds 30% equity sitting on the lowest edge of 30-50% category
The iShares AOK website shows the benchmark index is S&P Target Risk Conservative Index. This Index construction shows a 30% equity allocation (2020 prospectus p. S-2)
As of July 31, 2020, the Underlying Index included a fixed allocation of 30% of its assets in Underlying Funds that invest primarily in equity securities and 70% of its assets in Underlying Funds that invest primarily in bonds. As of July 31, 2020, the Fund invested approximately 32.27% of its assets in Underlying Funds that invest primarily in equity securities, 66.98% of its assets in Underlying Funds that invest primarily in bonds and the remainder of its assets in Underlying Funds that invest primarily in money market instruments.
Actually that isn't quite correct with respect to the index holdings. The Underlying Index did not have a fixed allocation of 30% of its assets in ... equity securities as of July 31, 2020
As with anything else, if you want an authoritative definition, go to the source. According to S&P, the index had a 30% allocation in equities as of April 30, 2020. It rebalances only semiannually, at the end of April and at the end of October.
Equities tend to outperform bonds. So between semi-annual rebalancing dates, statistically equities will be above the target allocation (and bonds below) on more dates than the opposite. Of course that wasn't true during the GFC, hence the low equity numbers for 2009-2010. Since then (a lot more years), equities have outperformed.
Regarding the annual statements - it would help if you would provide links to validate your figures. As it turns out, your "equity" figures in the early years are off because you didn't count real estate securities. For example, in the 2009 annual statement, instead of looking at the Management's Discussions section (p. 3) you could look at the actual details of the holdings in the Schedule of Investments for the fund (p. 16).
That shows iShares Cohen & Steers Realty Major Index Fund grouped under Domestic Equity (17.86%). Add in the International Equity holdings (5.81%), and one gets 24% equity. Still a far cry from 30%, but enough to illustrate why all figures are subject to verification.
(M* also considers real estate to be equity; it says that VGSIX is virtually 100% U.S. equity.)
Look at 2018. While M* may say that only 28% of the fund was in equity that year, the SEC filings tell a different story. The 2018 Annual Report says that equities amounted to 30% of the fund as of July 31. And the Semiannual Report says that equities came to 32% of the fund.
As I've explained above, the first year or so of this fund are outliers because bonds outperformed the equity market. (Also the fund was tinkering with its allocations as evidenced by the fact that it invested in real estate back then.) Likewise, you have implicitly labeled these years as outliers. Whether it's because they're not "near" 30% equity or they're not "on" the 30% edge, they're outside of your normal historical range.
Finally, as to the point that funds that bop around a boundary between categories could show up in a different category from what you're expecting: sure, I explained that in my first response. That goes for global vs. domestic, value vs. blend, etc. One can even use this to one's advantage as I illustrated in another thread. In looking for value funds near the value/blend boundary, one might search for value funds with a current blend portfolio style and for blend funds with a current value portfolio style. This doesn't catch all funds, but it's a good start.
I have followed this a debate by two astute commentators. I have to wonder if too much hair-splitting is not taking place. For my money (and I have none in AOK or any other allocation fund in the accounts I manage) I would pick a fund run by competent stock and bond managers and let them decide on the basis of market circumstances how far to overshoot or undershoot the percentage target. My total return will be little affected by a one or two percentage points difference. And, most importantly, my return will be totally unaffected by the « category » to which Lipper or M* assign it. The French have an expression for this: « Vous cherchez midi à quatorze heures, » or You’re trying to find noon at two in the afternoon.
"Historically" VWINX has held close to 40% equities (did I say that). It's a traditional 40/60 fund. Its extra equity all but guarantees a better longer term performance than a fund that's closer to 30/70.
(Though according to M*, AOK is actually closer to 30/60, meaning it has more cash dragging its performance down even further. M* says that 5% of the fund is in BISXX alone. You won't see this in the AOK holdings unless you dig for it. For example, 58% of AOK is in IUSB and 9% of IUSB is in BISXX. That would seem to account for the bulk of BISXX - but not all of the cash - in AOK.)
AOK current allocation: 30/61 (with 1% "not classified") VWINX current allocation: 38/57 (with 2% "not classified") Figures are from M*, "current" means most recent portfolio reported to M*.
Comments
Allocation Funds - 15 to 30% Equity
Allocation Funds - 30% to 50% Equity
https://www.morningstar.com/articles/306244/why-is-my-funds-style-box-different-from-its-categ
(This talks about style drift but it could just as easily been talking about a fund changing stock/bond allocations as changing large cap/small cap allocations.)
The point here is that one can't look at an average over time and determine how the fund is classified. The classification is "path dependent" - it depends on how it got there. A fund that hit 30% from 25% and stuck there would likely remain in its 15% to 30% category, while a fund that hit 30% from 35% and stuck there would likely remain in its 30% to 50% category.
If the fund is bouncing back and forth between a tad under and a tad over 30%, there would be no reason to change its category. Even if those oscillations miraculously averaged exactly 30% equity.
What does "historical" mean anyway? "the last 5 years" is a minority of the lifetime of this fund that's been around since 2008.
I have reclassified on my watch list as 15-30%.
It's currently at 31.75% equity (as of Dec 31), and rose as high as 32.5% equity as of July 31, according to its most recent annual statement.
There are always going to be problems running screens with "hard filters". I've posted on more than one occasion that IMHO it makes little sense to screen for funds that have never had a losing year. Which would you prefer: a fund that lost 0.1% in one year and made 10% or better in all its other years, or a fund that made 3% year in, year out? That's an example of a problem with any hard filter.
As far as AOK goes, according to Lipper there are only a total of 10 ETFs including both
"Mixed-Asset Target Allocation Moderate" and "Mixed-Asset Target Allocation Conservative" categories. So it's no big deal to watch all these funds.
http://www.funds.reuters.wallst.com/us/screener/screener.asp
M* Historical stock style for AOK shows:
2020 30% equity allocation
2019 30%
2018 27%
2017 28%
2016 29%
More samples:
2015 31% Form N-CSR p.7 SEC filing
2013 30% Form N-CSR p.4
2010 22% Form N-CSR p.3
2009 21% Form N-CSR p.3
AOK clearly sits historically near the 30% low end of the 30-50% category.
The iShares AOK website shows the benchmark index is S&P Target Risk Conservative Index. This Index construction shows a 30% equity allocation (2020 prospectus p. S-2). M* places it in the 30-50% category. Takeaway: keep your eye out for funds historically near the category weighting limits as it might still be within your risk tolerance and more importantly possibly could provide good performance metrics you are seeking.
"near" is good. It allows for fuzziness and the possibility of bouncing around the figure in a way that "on" 30% and "holding" 30% don't.
holds 30% equity
sitting on the lowest edge of 30-50% category
The iShares AOK website shows the benchmark index is S&P Target Risk Conservative Index. This Index construction shows a 30% equity allocation (2020 prospectus p. S-2)
Quoting from that section of the prospectus: Actually that isn't quite correct with respect to the index holdings. The Underlying Index did not have a fixed allocation of 30% of its assets in ... equity securities as of July 31, 2020
As with anything else, if you want an authoritative definition, go to the source. According to S&P, the index had a 30% allocation in equities as of April 30, 2020. It rebalances only semiannually, at the end of April and at the end of October.
Equities tend to outperform bonds. So between semi-annual rebalancing dates, statistically equities will be above the target allocation (and bonds below) on more dates than the opposite. Of course that wasn't true during the GFC, hence the low equity numbers for 2009-2010. Since then (a lot more years), equities have outperformed.
Regarding the annual statements - it would help if you would provide links to validate your figures. As it turns out, your "equity" figures in the early years are off because you didn't count real estate securities. For example, in the 2009 annual statement, instead of looking at the Management's Discussions section (p. 3) you could look at the actual details of the holdings in the Schedule of Investments for the fund (p. 16).
That shows iShares Cohen & Steers Realty Major Index Fund grouped under Domestic Equity (17.86%). Add in the International Equity holdings (5.81%), and one gets 24% equity. Still a far cry from 30%, but enough to illustrate why all figures are subject to verification.
(M* also considers real estate to be equity; it says that VGSIX is virtually 100% U.S. equity.)
Look at 2018. While M* may say that only 28% of the fund was in equity that year, the SEC filings tell a different story. The 2018 Annual Report says that equities amounted to 30% of the fund as of July 31. And the Semiannual Report says that equities came to 32% of the fund.
As I've explained above, the first year or so of this fund are outliers because bonds outperformed the equity market. (Also the fund was tinkering with its allocations as evidenced by the fact that it invested in real estate back then.) Likewise, you have implicitly labeled these years as outliers. Whether it's because they're not "near" 30% equity or they're not "on" the 30% edge, they're outside of your normal historical range.
Finally, as to the point that funds that bop around a boundary between categories could show up in a different category from what you're expecting: sure, I explained that in my first response. That goes for global vs. domestic, value vs. blend, etc. One can even use this to one's advantage as I illustrated in another thread. In looking for value funds near the value/blend boundary, one might search for value funds with a current blend portfolio style and for blend funds with a current value portfolio style. This doesn't catch all funds, but it's a good start.
AOK vs VWINX chart beginning November, 2008.
Regards,
Catch
(Though according to M*, AOK is actually closer to 30/60, meaning it has more cash dragging its performance down even further. M* says that 5% of the fund is in BISXX alone. You won't see this in the AOK holdings unless you dig for it. For example, 58% of AOK is in IUSB and 9% of IUSB is in BISXX. That would seem to account for the bulk of BISXX - but not all of the cash - in AOK.)
AOK current allocation: 30/61 (with 1% "not classified")
VWINX current allocation: 38/57 (with 2% "not classified")
Figures are from M*, "current" means most recent portfolio reported to M*.