For "indexed" target date retirement funds, how do managers determine the glide path and/or asset allocation?

Discuss all general (i.e. non-personal) investing questions and issues, investing news, and theory.
Post Reply
Topic Author
VTI
Posts: 39
Joined: Sun Mar 14, 2021 12:56 am

For "indexed" target date retirement funds, how do managers determine the glide path and/or asset allocation?

Post by VTI »

Unsurprisingly, the asset allocations of Fidelity's target date funds differ between their active and indexed varieties:

Fidelity, actively managed: FFSFX - Fidelity Freedom® 2065 Fund

Domestic Equities: 52.53%
International Equities: 40.58%
Bonds: 6.40%
Short-Term Debt & Net Other Assets: 0.49%

Fidelity, indexed: FFIJX - Fidelity Freedom ® Index 2065 Fund

Domestic Equities 54.67%
International Equities 35.75%
Bonds 9.57%
Short-Term Debt & Net Other Assets 0.01%

How do the managers of index target date funds determine the funds' glide path and/or asset allocation? Those active choices can make a significant impact on performance and risk.

On a side note, those different asset allocations makes it difficult to compare options.
terran
Posts: 2019
Joined: Sat Jan 10, 2015 10:50 pm

Re: For "indexed" target date retirement funds, how do managers determine the glide path and/or asset allocation?

Post by terran »

Are you asking how they decided what the asset allocation would be at various points in time in the first place, or how the asset allocation will be adjusted now that they've decided? I don't know the answer to the first question other than to say "research" but for the second you can see how they plan to adjust the asset allocation over time if you click on the Prospectus link in the upper right near the Buy/Sell buttons. Of course, this can change as their research/philosophy changes as it did a year or two ago when they adjusted from approximately 30% international equities to now being more like 40% international. Long story short, you should take a look at the current allocations every so often and make sure they continue to be something you're comfortable with. Especially in a tax advantaged account you can always easily switch to other investment options if you're no longer happy with the asset allocation of the target date fund.
Northern Flicker
Posts: 7411
Joined: Fri Apr 10, 2015 12:29 am

Re: For "indexed" target date retirement funds, how do managers determine the glide path and/or asset allocation?

Post by Northern Flicker »

VTI wrote: Mon May 03, 2021 12:06 am Unsurprisingly, the asset allocations of Fidelity's target date funds differ between their active and indexed varieties:

Fidelity, actively managed: FFSFX - Fidelity Freedom® 2065 Fund

Domestic Equities: 52.53%
International Equities: 40.58%
Bonds: 6.40%
Short-Term Debt & Net Other Assets: 0.49%

Fidelity, indexed: FFIJX - Fidelity Freedom ® Index 2065 Fund

Domestic Equities 54.67%
International Equities 35.75%
Bonds 9.57%
Short-Term Debt & Net Other Assets 0.01%

How do the managers of index target date funds determine the funds' glide path and/or asset allocation? Those active choices can make a significant impact on performance and risk.

On a side note, those different asset allocations makes it difficult to compare options.
Maybe Fidelity doesn't want you to be able to compare easily the performance of their active and passive target date funds.
My postings are my opinion, and never should be construed as a recommendation to buy, sell, or hold any particular investment.
User avatar
nisiprius
Advisory Board
Posts: 43352
Joined: Thu Jul 26, 2007 9:33 am
Location: The terrestrial, globular, planetary hunk of matter, flattened at the poles, is my abode.--O. Henry

Re: For "indexed" target date retirement funds, how do managers determine the glide path and/or asset allocation?

Post by nisiprius »

Morningstar has a methodology for developing glide paths--as a service for advising companies on managing retirement plans they offer custom glide paths and such--and they give a lot of semi-detailed general talk about it in papers like Selecting a Target-Date Benchmark.

As always, you have the problem of people acting as if economics is a hard science when it is actually a social science. In a vague general way, the established way of thinking about it is that your future earnings are an asset with certain characteristics, and compared to stocks they are a relatively low-risk asset, so your ability to bounce back and earn your way out of any hole weighs on the low-risk side of the balance. But the older you get, the fewer future years you have so the less "human capital" you have on that side, so you should cut back on stocks and increase bonds to keep the balance. Everything else is the result of trying to put actual numbers on that handwaving balance. And then, of course, portfolio details.

A chart of Morningstar's benchmark glide paths is revealing:

Image

The point here is that financial-economics math may give you the shape of a glide path, but risk tolerance tells how high or low the curve should be and there is a huge difference. Morningstar has a chart in a print book that I can't find online, but basically they plot actual glide slopes from twenty or so real-world target date funds, and it looks like a kid crayoning in a band between the red and purple lines. That is, their "conservative" and "aggressive" curves show the actual range of different decisions different fund companies have made about overall asset allocation.

I would give a lot to know how fund companies really do decide on the composition and allocation of all-in-one funds, but I think it's pretty ugly. Despite the façade of math and science, I feel pretty sure it is really someone saying "eh, so, I dunno, how about devoting some of the bond allocation to TIPS for the later years, maybe like 20%, whaddaya say?" "Well, our rival is doing this." "Well, our TIPS fund could use some more AUM..."

P.S. At one point, due to some kind of glitch in their glide-path construction rule, Fidelity's glide paths for their original Freedom funds had ripples in them, like smoothed and rounded stairsteps. And Fidelity was actually trying to pretend this was a feature that made their funds better than the competition! All I can find now is one image I posted earlier, but do notice the ® symbol on "rolldown," it was their trademarked name for their secret sauce.

Image
Annual income twenty pounds, annual expenditure nineteen nineteen and six, result happiness; Annual income twenty pounds, annual expenditure twenty pounds ought and six, result misery.
User avatar
David Jay
Posts: 10714
Joined: Mon Mar 30, 2015 5:54 am
Location: Michigan

Re: For "indexed" target date retirement funds, how do managers determine the glide path and/or asset allocation?

Post by David Jay »

If you are trying to find the differences between the two funds, here is the biggest difference between the two funds (FFSFX is 625% higher):

Fidelity Freedom 2065 ER .75%
Fidelity Freedom Index 2065 ER .12%
Prediction is very difficult, especially about the future - Niels Bohr | To get the "risk premium", you really do have to take the risk - nisiprius
hi_there
Posts: 784
Joined: Sat Aug 29, 2020 7:00 pm

Re: For "indexed" target date retirement funds, how do managers determine the glide path and/or asset allocation?

Post by hi_there »

This thread raises a question for me, which is why glide path, i.e. stock/bond allocation by age, is not widely studied from a market driven perspective. That is, we could index the universe of investors based on net worth and figure out the "market index" asset allocation, which based on invisible hand principals, should reflect optimal risk/utility for someone with X net worth and N years to retirement.

It is probably apparent to most people that static rules, like 100 minus age, are sub optimal. For example, someone should rationally prefer more bonds under high real interest rates and vice versa. An optimal measure should be dynamic and reflect constantly changing economic factors. If we believe that market forces efficiently drive stock prices, market weights, cross asset risk, and so on, then why not something simple like retirement asset allocation?
sycamore
Posts: 2044
Joined: Tue May 08, 2018 12:06 pm

Re: For "indexed" target date retirement funds, how do managers determine the glide path and/or asset allocation?

Post by sycamore »

hi_there wrote: Mon May 03, 2021 10:16 am This thread raises a question for me, which is why glide path, i.e. stock/bond allocation by age, is not widely studied from a market driven perspective.
Practically speaking getting the data needed for such a study might be difficult: enough investors use multiple accounts/custodians that just one custodian may not be able to provide a good idea of how investor's portfolios are really constructed. Just a hunch, I don't have any data myself on how easily the data could be gotten.
hi_there wrote: Mon May 03, 2021 10:16 am That is, we could index the universe of investors based on net worth and figure out the "market index" asset allocation, which based on invisible hand principals, should reflect optimal risk/utility for someone with X net worth and N years to retirement.

It is probably apparent to most people that static rules, like 100 minus age, are sub optimal. For example, someone should rationally prefer more bonds under high real interest rates and vice versa. An optimal measure should be dynamic and reflect constantly changing economic factors. If we believe that market forces efficiently drive stock prices, market weights, cross asset risk, and so on, then why not something simple like retirement asset allocation?
I think this approach would be problematic for one of the reasons TD funds are problematic: the glide path for any given investor should be based on more than just X net worth and N years to retirement. My risk tolerance and "need, ability, and willingness" to take risk aren't the same as the average investor. My age and years to retirement are but two factors used in determining my asset allocation.
hi_there
Posts: 784
Joined: Sat Aug 29, 2020 7:00 pm

Re: For "indexed" target date retirement funds, how do managers determine the glide path and/or asset allocation?

Post by hi_there »

sycamore wrote: Mon May 03, 2021 11:00 am
hi_there wrote: Mon May 03, 2021 10:16 am This thread raises a question for me, which is why glide path, i.e. stock/bond allocation by age, is not widely studied from a market driven perspective.
Practically speaking getting the data needed for such a study might be difficult: enough investors use multiple accounts/custodians that just one custodian may not be able to provide a good idea of how investor's portfolios are really constructed. Just a hunch, I don't have any data myself on how easily the data could be gotten.
hi_there wrote: Mon May 03, 2021 10:16 am That is, we could index the universe of investors based on net worth and figure out the "market index" asset allocation, which based on invisible hand principals, should reflect optimal risk/utility for someone with X net worth and N years to retirement.

It is probably apparent to most people that static rules, like 100 minus age, are sub optimal. For example, someone should rationally prefer more bonds under high real interest rates and vice versa. An optimal measure should be dynamic and reflect constantly changing economic factors. If we believe that market forces efficiently drive stock prices, market weights, cross asset risk, and so on, then why not something simple like retirement asset allocation?
I think this approach would be problematic for one of the reasons TD funds are problematic: the glide path for any given investor should be based on more than just X net worth and N years to retirement. My risk tolerance and "need, ability, and willingness" to take risk aren't the same as the average investor. My age and years to retirement are but two factors used in determining my asset allocation.
I agree that investing is based on a series of personal attributes as you said. However, there are hundreds of target date funds that determine asset allocation based on somewhat arbitrary or static formulas. So, there is indeed a widespread attempt at offering a standardized asset allocation product to a broad universe of retirement investors. There should be some kind of dynamic benchmark for these asset allocations, and I am proposing a market weight as a way to approach this problem.

In terms of data collection, we probably don't need to sample the whole universe of investors to get interesting results. Most studies on investing are done based on a subset of investors that are considered to be representative of a broader population - for instance, clients of a single large brokerage. Studies on demographical asset allocation must surely exist, but are just not popularized for whatever reason.
User avatar
vineviz
Posts: 9837
Joined: Tue May 15, 2018 1:55 pm

Re: For "indexed" target date retirement funds, how do managers determine the glide path and/or asset allocation?

Post by vineviz »

hi_there wrote: Mon May 03, 2021 11:27 am I agree that investing is based on a series of personal attributes as you said. However, there are hundreds of target date funds that determine asset allocation based on somewhat arbitrary or static formulas. So, there is indeed a widespread attempt at offering a standardized asset allocation product to a broad universe of retirement investors. There should be some kind of dynamic benchmark for these asset allocations, and I am proposing a market weight as a way to approach this problem.
I can't tell what you're driving at exactly, but there ARE benchmarks for target date funds. I know that both Morningstar and S&P construct and maintain multiple versions.

But dynamically sampling the market cap weighted average of holdings for similar investors would be an immense undertaking with limited benefit, since it's actually pretty easy to estimate what YOUR allocation should be at any particular point in time.
"Far more money has been lost by investors preparing for corrections than has been lost in corrections themselves." ~~ Peter Lynch
Northern Flicker
Posts: 7411
Joined: Fri Apr 10, 2015 12:29 am

Re: For "indexed" target date retirement funds, how do managers determine the glide path and/or asset allocation?

Post by Northern Flicker »

nisiprius wrote: I would give a lot to know how fund companies really do decide on the composition and allocation of all-in-one funds, but I think it's pretty ugly. Despite the façade of math and science, I feel pretty sure it is really someone saying "eh, so, I dunno, how about devoting some of the bond allocation to TIPS for the later years, maybe like 20%, whaddaya say?" "Well, our rival is doing this." "Well, our TIPS fund could use some more AUM..."
If you look at Vanguard Target Retirement Income, VTINX, and separate out the bond subclasses of the bond index fund into treasuries, MBS, and credit, you will find that the fund targets holding approximately equal allocations of treasuries, TIPS, and non-US bonds. Given that TIPS are 17% of the fund, I don't think that was an arbitrary choice.
My postings are my opinion, and never should be construed as a recommendation to buy, sell, or hold any particular investment.
Post Reply