Is there a problem with glidepath (target date funds) for retirement savers?

Discuss all general (i.e. non-personal) investing questions and issues, investing news, and theory.
Topic Author
Browser
Posts: 4857
Joined: Wed Sep 05, 2012 4:54 pm

Is there a problem with glidepath (target date funds) for retirement savers?

Post by Browser »

Investors who own target date funds and make regular contributions may unknowingly be overinvested in bonds relative to stocks over time. The theory behind target date funds is systematically shift the allocation from riskier stocks to conservative bonds over time. The problem with this approach for those who invest periodically is that the fund is stock-heavy in the early years when the investor has not accumulated much principal, and it is bond heavy in later years when the fund principal is larger. This results in a much higher dollar-weighted allocation to bonds than the investor may be aware of.

Here's a simple example to illustrate. Let's say that an investor put $1,000 (inflation adjusted) per year over the 25 year period 1990-2014 into a fund that begins with a 70% allocation to stocks and a 30% allocation to bonds. Every 5 years the fund shifts 10% of its allocation from stocks to bonds as follows:

Year 1-5: 70% stocks, 30% bonds
Year 6-10: 60% stocks, 40% bonds
Year 11-15: 50% stocks, 50% bonds
Year 16-20: 40% stocks, 60% bonds
Year 21-25: 30% stocks, 70% bonds

At the end of 2014, the investor would have accumulated $84,418.

The average allocation to stocks and bonds over this period of time appears to be 50%. However, the dollar-weighted allocation to stocks is actually much lower while the dollar-weighted allocation to bonds is much higher. In fact, if the investor had simply invested his annual contributions in a portfolio of 20% in stocks and 80% in bonds in 1990 and maintained that allocation by rebalancing annually, he would have ended up with $84,380 - virtually the same amount. We see from this example that his effective allocation to stocks was only 20% over this 25-year time period; probably much lower than the investor thought based on the glidepath allocation recipe. Most retirement investors should have a much higher allocation to stocks than they think they have in their target date fund, based on correctly determining the allocation based on dollars invested in stocks instead of percentage invested.
We don't know where we are, or where we're going -- but we're making good time.
RadAudit
Posts: 4386
Joined: Mon May 26, 2008 10:20 am
Location: Second star on the right and straight on 'til morning

Re: Is there a problem with glidepath (target date funds) for retirement savers?

Post by RadAudit »

Browser wrote:Most retirement investors should have a much higher allocation to stocks than they think they have in their target date fund, based on correctly determining the allocation based on dollars invested in stocks instead of percentage invested.
Don't target date funds rebalance their holdings (stocks to bonds, bonds to stock) so the amount in the portfolio equals the percentages of the glidepath? In which case, it's not how much you invest; isn't it how the fund invests what you send them?
FI is the best revenge. LBYM. Invest the rest. Stay the course. Die anyway. - PS: The cavalry isn't coming, kids. You are on your own.
User avatar
Taylor Larimore
Posts: 32842
Joined: Tue Feb 27, 2007 7:09 pm
Location: Miami FL

Re: Is there a problem with glidepath (target date funds) for retirement savers?

Post by Taylor Larimore »

Browser:

It is notable that experts in ALL major companies (including Vanguard) have designed similar target-date glide paths. To my knowledge, they ALL begin with a high allocation to stocks at younger ages and end with a high allocation to bonds in retirement. The primary reason is that most retirees cannot afford to lose their life-savings in a stock market crash.

Target-date funds: Looking beyond the glide path in 2014

Best wishes.
Taylor
"Simplicity is the master key to financial success." -- Jack Bogle
Topic Author
Browser
Posts: 4857
Joined: Wed Sep 05, 2012 4:54 pm

Re: Is there a problem with glidepath (target date funds) for retirement savers?

Post by Browser »

RadAudit wrote:
Browser wrote:Most retirement investors should have a much higher allocation to stocks than they think they have in their target date fund, based on correctly determining the allocation based on dollars invested in stocks instead of percentage invested.
Don't target date funds rebalance their holdings (stocks to bonds, bonds to stock) so the amount in the portfolio equals the percentages of the glidepath? In which case, it's not how much you invest; isn't it how the fund invests what you send them?
Not sure if this is what you are asking, but here's a simple example of dollar-weighted vs. percentage allocation. Let's say that you invest $1,000 at two different times. At Time 1, the fund allocation is set to 80% stocks, 20% bonds. At Time 2, the allocation is set to 20% stocks, 80% bonds. For simplification, assume no growth in assets.

___________Porfolio Total___Stocks_______Bonds
Time 1...... $1,000............$800 (80%)....$200 (20%)
Time 2.......$2,000............$400 (20%)...$1600(80%)
Average......$1,500............$600...........$900

On an average percentage basis, your allocation to stocks over Time 1 and Time 2 is 50% [(80% + 20%%) / 2].
On a dollar weighted basis, your average allocation to stocks is 40% [$600/ $1500]

Most people believe it's a good idea to reduce stock allocation as one approaches retirement, but many may not be aware just how much their target date fund may be tilting their portfolio toward bonds based on the dollars they actually have invested over time.

Unfortunately, the literature on target date funds very seldom takes account of the risk and returns for investors who are dripping in their savings with periodic investing over their investing horizon. These results can be quite different than for a hypothetical lump sum investment made at the front end.
We don't know where we are, or where we're going -- but we're making good time.
miles monroe
Posts: 1290
Joined: Mon Jan 20, 2014 11:14 am

Re: Is there a problem with glidepath (target date funds) for retirement savers?

Post by miles monroe »

no one knows what the right answer is until that last check is written to the funeral home....

1. target date funds start with a high stock allocation and reduce it over time.

2. some people keep a 50-50 (or 60-40 or 40-60 or whatever) allocation thruout retirement.

3. and now some experts are advocating a U shaped glide path where you decrease your stock allocation to 20%-40% at retirement and then increase it 1% a year throughout retirement.
Topic Author
Browser
Posts: 4857
Joined: Wed Sep 05, 2012 4:54 pm

Re: Is there a problem with glidepath (target date funds) for retirement savers?

Post by Browser »

miles monroe wrote:no one knows what the right answer is until that last check is written to the funeral home....

1. target date funds start with a high stock allocation and reduce it over time.

2. some people keep a 50-50 (or 60-40 or 40-60 or whatever) allocation thruout retirement.

3. and now some experts are advocating a U shaped glide path where you decrease your stock allocation to 20%-40% at retirement and then increase it 1% a year throughout retirement.
To my way of thinking, the benchmark strategy should be a static fixed allocation with an allocation to stocks and bonds that represents the average allocation over time of the target date fund to which you are comparing. To simplify, if the target date fund slowly ramps down from 80% to 40% stocks in equal steps, the average allocation over time would be about 60% stocks, 40% bonds. You would compare the expected returns and risk of the target date fund to a static 60/40 portfolio to determine if the TDF is superior in some way. Unfortunately, most of those comparisons find that the functionally equivalent static fixed allocation generally ends up better in terms of terminal value and the variability of terminal values.
We don't know where we are, or where we're going -- but we're making good time.
User avatar
Kosmo
Posts: 1303
Joined: Wed Sep 05, 2012 11:54 am
Location: Philadelphia

Re: Is there a problem with glidepath (target date funds) for retirement savers?

Post by Kosmo »

Not sure what you're attempting to accomplish. By averaging dollar values you're equally weighting things that shouldn't be equally weighted. If you do this it will shift the average weight towards the higher allocation percentage as the total portfolio grows larger. Consider 2 scenarios:

Scenario 1) Add a Time 1a with a $1000 contribution at 50% stocks.
Browser wrote:___________Porfolio Total___Stocks_______Bonds
Time 1...... $1,000............$800 (80%)....$200 (20%)
Time 1a.....$2,000............$1000 (50%)...$1000 (50%)
Time 2.......$3,000............$600 (20%)....$2400 (80%)
Average......$2,000............$800...........$1200
Percentage basis allocation: (80+50+20)/3 = 50% stocks
Dollar basis: 800/2000 = 40% stocks

Scenario 2) Add a Time 2a with a $1000 contribution at 50% stocks. Same as above, but the sequence is different.
Browser wrote:___________Porfolio Total___Stocks_______Bonds
Time 1...... $1,000............$800 (80%)....$200 (20%)
Time 2.......$2,000............$400 (20%)...$1600(80%)
Time 2a.....$3,000............$1500 (50%)...$1500 (50%)
Average......$2,000............$900...........$1100
Percentage basis allocation: (80+20+50)/3 = 50% stocks
Dollar basis: 900/2000 = 45% stocks

You can clearly see that the allocation at higher portfolio totals dominates. You're completely ignoring the risk and expected return of stock vs. bonds, which is what drives people to select an asset allocation (or a target date fund starting point) in the first place.
Topic Author
Browser
Posts: 4857
Joined: Wed Sep 05, 2012 4:54 pm

Re: Is there a problem with glidepath (target date funds) for retirement savers?

Post by Browser »

Taylor Larimore wrote:Browser:

It is notable that experts in ALL major companies (including Vanguard) have designed similar target-date glide paths. To my knowledge, they ALL begin with a high allocation to stocks at younger ages and end with a high allocation to bonds in retirement. The primary reason is that most retirees cannot afford to lose their life-savings in a stock market crash.

Target-date funds: Looking beyond the glide path in 2014

Best wishes.
Taylor
Thanks Taylor -

Most people would generally agree that it's safer to have more money in bonds and I wouldn't disagree in general. But how much did TDFs actually help during the stock market "crash" that lasted from 11/07 to 3/09? Vanguard's target date fund designed for people who would retire one year later in 2010 actually lost about 30% during that time, which was about the same as the loss for a portfolio that was 65% stocks, 35% bonds. I'll bet that was a surprise to those 2010 retirees! In general, target date funds performed very poorly during that time and Vanguard's was one of the better ones. So, I'm not completely confident that they are necessarily the best way to save yourself from a stock market crash if one happens and that it should be automatically assumed they will do so.
We don't know where we are, or where we're going -- but we're making good time.
User avatar
Taylor Larimore
Posts: 32842
Joined: Tue Feb 27, 2007 7:09 pm
Location: Miami FL

The best way to save yourself from a stock market crash?

Post by Taylor Larimore »

I'm not completely confident that they (target-date funds) are necessarily the best way to save yourself from a stock market crash

Browser:

What do you suggest is the "best way?"

Thank you and best wishes
Taylor
"Simplicity is the master key to financial success." -- Jack Bogle
Topic Author
Browser
Posts: 4857
Joined: Wed Sep 05, 2012 4:54 pm

Re: The best way to save yourself from a stock market crash?

Post by Browser »

Taylor Larimore wrote:
I'm not completely confident that they (target-date funds) are necessarily the best way to save yourself from a stock market crash

Browser:

What do you suggest is the "best way?"

Thank you and best wishes
Taylor
Well, let me say that the target date method seems to have lots of room for improvement if we go on the basis of past performance. I figure the experts are hard at work on it and I await their accomplishments. :)
We don't know where we are, or where we're going -- but we're making good time.
hnzw rui
Posts: 578
Joined: Tue Sep 22, 2015 2:26 pm

Re: Is there a problem with glidepath (target date funds) for retirement savers?

Post by hnzw rui »

Browser wrote:Most people would generally agree that it's safer to have more money in bonds and I wouldn't disagree in general. But how much did TDFs actually help during the stock market "crash" that lasted from 11/07 to 3/09? Vanguard's target date fund designed for people who would retire one year later in 2010 actually lost about 30% during that time, which was about the same as the loss for a portfolio that was 65% stocks, 35% bonds. I'll bet that was a surprise to those 2010 retirees! In general, target date funds performed very poorly during that time and Vanguard's was one of the better ones. So, I'm not completely confident that they are necessarily the best way to save yourself from a stock market crash if one happens and that it should be automatically assumed they will do so.
Seems to me like Vanguard's TD funds have done pretty well, though. I couldn't use Target Retirement 2005 (VTOVX) as I believe that's been consolidated into Target Retirement Income (VTINX).

According to Portfolio Visualizer

From 2007 to 2009 (retired)
Initial Amount: $1,000,000
Annual Adjustment: withdraw $40,000 (inflation-adjusted)
VTENX (Target Retirement 2010): $887,475
VSMGX (LifeStrategy Moderate 60/40): $819,789
VSCGX (LifeStrategy Conservative 40/60): $877,120

From 2007 to 2015 (retired)
Initial Amount: $1,000,000
Annual Adjustment: withdraw $40,000 (inflation-adjusted)
VTENX (Target Retirement 2010): $1,022,258
VSMGX (LifeStrategy Moderate 60/40): $975,787
VSCGX (LifeStrategy Conservative 40/60): $983,131

From 2007 to 2008 (accumulation, near retirement)
Initial Amount: $1,000,000
Annual Adjustment: contribute $5,000 (not inflation-adjusted)
VTENX (Target Retirement 2010): $863,308
VSMGX (LifeStrategy Moderate 60/40): $797,796
VSCGX (LifeStrategy Conservative 40/60): $870,073

From 2007 to 2009 (accumulation, near retirement)
Initial Amount: $1,000,000
Annual Adjustment: contribute $5,000 (not inflation-adjusted)
VTENX (Target Retirement 2010): $1,035,104
VSMGX (LifeStrategy Moderate 60/40): $964,994
VSCGX (LifeStrategy Conservative 40/60): $1,023,484

From 2004 to 2015 (accumulation, early 20s or 30s)
Initial Amount: $5,000
Annual Adjustment: contribute $5,000 (not inflation-adjusted)
VTIVX (Target Retirement 2045): $91,198
VTTHX (Target Retirement 2035): $90,329
VSMGX (LifeStrategy Moderate 60/40): $86,114
VSCGX (LifeStrategy Conservative 40/60): $82,668

For the accumulation stage, I kinda just averaged out the IRA contribution limits for the above period.
http://cashmoneylife.com/traditional-ro ... on-limits/
Last edited by hnzw rui on Fri Oct 09, 2015 5:25 pm, edited 1 time in total.
User avatar
Miriam2
Posts: 4387
Joined: Fri Nov 14, 2014 10:51 am

Re: Is there a problem with glidepath (target date funds) for retirement savers?

Post by Miriam2 »

Browser -
This is my understanding of Target Date Retirement Funds.

I think Target Date Funds are especially important to the vast majority of workers who have no idea how to invest and have no interest in learning. Vanguard, Fidelity and T.Rowe Price designed their Target Date Funds for them, so these workers will not be left overly stock heavy in retirement - at chance of losing everything - or overly bond heavy during their working years - making very little headway.

The funds are not perfect, but they are FAR superior to doing nothing. The average person often does nothing, and that is, unfortunately, what many secretaries and professionals in my office do - because they can't and don't learn the basics to invest and they see this long list of crazy sounding funds in our 401k/457b booklet. Or they are poorly advised by someone to put all their money in some Oppenheimer all-equity all world small cap fund, you know the fund types I speak of :shock:

I have my kids in the Vanguard Target Date 2055 - because if the worst happens to them and they can't invest on their own, at least they are adequately protected by Vanguard's expertise. :happy
Topic Author
Browser
Posts: 4857
Joined: Wed Sep 05, 2012 4:54 pm

Re: Is there a problem with glidepath (target date funds) for retirement savers?

Post by Browser »

Kosmo - I'm not sure a fully understand your point. In your second example, you show a series of allocations that violate the glidepath. The allocation to stocks goes from 80% to 20% and then back to 50%. This wouldn't be a glidepath allocation formula, which systematically lowers stock allocation over time while increasing the bond allocation. What we find is that the glidepath formula increases the bond allocation at the same time the portfolio is growing in value. The combination causes the dollar-weighted bond allocation to be higher than the average allocation to bonds over that timeframe. If I were to throw in an extra investment in stocks (as you did when inserting the 50% allocation) then, sure, that would increase the dollar-weighted equity allocation. How is that relevant?
We don't know where we are, or where we're going -- but we're making good time.
User avatar
Miriam2
Posts: 4387
Joined: Fri Nov 14, 2014 10:51 am

Re: Is there a problem with glidepath (target date funds) for retirement savers?

Post by Miriam2 »

Browser wrote: This wouldn't be a glidepath allocation formula, which systematically lowers stock allocation over time while increasing the bond allocation. What we find is that the glidepath formula increases the bond allocation at the same time the portfolio is growing in value. The combination causes the dollar-weighted bond allocation to be higher than the average allocation to bonds over that timeframe.
I think I understand what you mean - but don't you think Vanguard has built this into their formula? They changed their Target Date formula within the last 5-6 years, aren't they doing all kinds of these calculations and, with their expertise, coming up with what they believe is the best formula?
Topic Author
Browser
Posts: 4857
Joined: Wed Sep 05, 2012 4:54 pm

Re: Is there a problem with glidepath (target date funds) for retirement savers?

Post by Browser »

Miriam2 wrote:Browser -
This is my understanding of Target Date Retirement Funds.

I think Target Date Funds are especially important to the vast majority of workers who have no idea how to invest and have no interest in learning. Vanguard, Fidelity and T.Rowe Price designed their Target Date Funds for them, so these workers will not be left overly stock heavy in retirement - at chance of losing everything - or overly bond heavy during their working years - making very little headway.

The funds are not perfect, but they are FAR superior to doing nothing. The average person often does nothing, and that is, unfortunately, what many secretaries and professionals in my office do - because they can't and don't learn the basics to invest and they see this long list of crazy sounding funds in our 401k/457b booklet. Or they are poorly advised by someone to put all their money in some Oppenheimer all-equity all world small cap fund, you know the fund types I speak of :shock:

I have my kids in the Vanguard Target Date 2055 - because if the worst happens to them and they can't invest on their own, at least they are adequately protected by Vanguard's expertise. :happy
Can't say I disagree with this at all. I think many folks are in a real mess trying to manage their own portfolios these days and can do a lot worse than a TDF, as you have well-explained. In particular, you make the very good point that a lot of less knowledgeable people might be too freaked out to invest 90% in stocks early in their career (as many TDFs do) if it were visible to them -- but they should. I'm not talking to them as much as to folks with a little more know how and the time and willingness to DIY their retirement portfolios. But I am somewhat concerned that people without a lot of knowledge think, or are led to believe by fund companies and retirement plan sponsors, that TDFs are a kind of substitute for a defined benefit pension and that they can safely "glide" the individual to a known finishing point with their nestegg being expertly "de-risked" as they approach the landing zone. Nothing could be further from the truth, as many learned in 2007-2009 I'm afraid.
Last edited by Browser on Fri Oct 09, 2015 3:47 pm, edited 2 times in total.
We don't know where we are, or where we're going -- but we're making good time.
User avatar
Peter Foley
Posts: 5533
Joined: Fri Nov 23, 2007 9:34 am
Location: Lake Wobegon

Re: Is there a problem with glidepath (target date funds) for retirement savers?

Post by Peter Foley »

Browser wrote:
Not sure if this is what you are asking, but here's a simple example of dollar-weighted vs. percentage allocation. Let's say that you invest $1,000 at two different times. At Time 1, the fund allocation is set to 80% stocks, 20% bonds. At Time 2, the allocation is set to 20% stocks, 80% bonds. For simplification, assume no growth in assets.

___________Porfolio Total___Stocks_______Bonds
Time 1...... $1,000............$800 (80%)....$200 (20%)
Time 2.......$2,000............$400 (20%)...$1600(80%)
Average......$1,500............$600...........$900

On an average percentage basis, your allocation to stocks over Time 1 and Time 2 is 50% [(80% + 20%%) / 2].
On a dollar weighted basis, your average allocation to stocks is 40% [$600/ $1500]
Sorry - I must be missing something here. If this is the same target retirement fund there is probably a 30 year time difference between investment 1 and investment 2.

Browser wrote:
Most people believe it's a good idea to reduce stock allocation as one approaches retirement, but many may not be aware just how much their target date fund may be tilting their portfolio toward bonds based on the dollars they actually have invested over time.
This is true. Investors should also note that glide paths for target retirement funds differ greatly from fund sponsor to fund sponsor. Vanguard's glide pattern is much different than Fidelity's, for example.

I posted the PowerPoint Presentation that the Minnesota Bogleheads group created for our last meeting. Our main topic was Glide paths and target retirement funds. The PowerPoint is available on the Wiki. (I will edit this message to provide the link to the presentation.)


www.bogleheads.org/wiki/Bogleheads®_fin ... cy_project
sls239
Posts: 1208
Joined: Thu Oct 23, 2008 4:04 pm

Re: Is there a problem with glidepath (target date funds) for retirement savers?

Post by sls239 »

The glide-paths that are actually used are quite a bit different from your steadily increasing bond allocation your example and the difference is not nearly as great as in your example.

The actual glide-paths keep a high stock allocation for quite a long time, then steeply de-risk shortly before retirement. The experts chose, on purpose, a glide-path that was NOT linear.

The example you used is more like the age in bonds rule-of-thumb, so if you have an argument against anything, it is against that.
randomguy
Posts: 11295
Joined: Wed Sep 17, 2014 9:00 am

Re: Is there a problem with glidepath (target date funds) for retirement savers?

Post by randomguy »

Browser wrote:
Taylor Larimore wrote:Browser:

It is notable that experts in ALL major companies (including Vanguard) have designed similar target-date glide paths. To my knowledge, they ALL begin with a high allocation to stocks at younger ages and end with a high allocation to bonds in retirement. The primary reason is that most retirees cannot afford to lose their life-savings in a stock market crash.

Target-date funds: Looking beyond the glide path in 2014

Best wishes.
Taylor
Thanks Taylor -

Most people would generally agree that it's safer to have more money in bonds and I wouldn't disagree in general. But how much did TDFs actually help during the stock market "crash" that lasted from 11/07 to 3/09? Vanguard's target date fund designed for people who would retire one year later in 2010 actually lost about 30% during that time, which was about the same as the loss for a portfolio that was 65% stocks, 35% bonds. I'll bet that was a surprise to those 2010 retirees! In general, target date funds performed very poorly during that time and Vanguard's was one of the better ones. So, I'm not completely confident that they are necessarily the best way to save yourself from a stock market crash if one happens and that it should be automatically assumed they will do so.
I wouldn't say they performed poorly. They performed as expected. Most of them are in the 50-60%s stocks range in the years before you retire. That level of stock holdings is both needed (i.e you have a 30 year time frame) and volatile (when you have one of top half dozen worse time periods for stocks, you are going to lose a lot of money).

You are pretty much always stuck having to balance the risk of market volatility (holding stocks) and the risk of not enough return (holding too much bonds). Where you fall depends on your needs (i.e when you are down at the 2% SWR, you can just hold 100%bonds, at 5%, you better be holding 50%+ stocks), desires (are you ok spendind down the money or would you like to die with 2x as much as you started with), and general risk tolerance (i.e. if you lose 30% are you going to lose sleep).
Grogs
Posts: 1174
Joined: Tue Mar 24, 2015 4:55 pm

Re: Is there a problem with glidepath (target date funds) for retirement savers?

Post by Grogs »

sls239 wrote:The glide-paths that are actually used are quite a bit different from your steadily increasing bond allocation your example and the difference is not nearly as great as in your example.
Right. They tend to stay at 90/10 for a long time, then drop fairly steeply to 60/40 five years from retirement, then to 50/50 by the target date, where they stay for quite a while. This is also a considerably lower % of bonds than in the original post, where stocks have dropped to 30% by the target date. The problem of course is that it's a trade-off. Sure, a higher bond % is giving up some potential to continue compounding at the end, but it's also locking in some of the gains. A drop from $1MM to $750k 5 years out is painful, but not as much as a drop to $500k would be.

And to Miriam's point, I'm a perfect example. When I started my 401k, I knew virtually nothing about investing. There were all these bond funds, S&P fund, small cap, all-world index, etc that meant nothing to me, and then there were the LifePath funds whose description said something like "a fund designed for people who will turn 65 near the year 20XX." That made the choice pretty easy. Now that I've learned more about various asset classes, I'm still pretty happy with how the fund has done. It has almost certainly performed better than I would have by (essentially) just picking a random fund with no knowledge.
rkhusky
Posts: 17764
Joined: Thu Aug 18, 2011 8:09 pm

Re: Is there a problem with glidepath (target date funds) for retirement savers?

Post by rkhusky »

Frankly, I don't care what my average allocation is over a period of time. I am only concerned about the allocation at particular points in time.
leonard
Posts: 5993
Joined: Wed Feb 21, 2007 10:56 am

Re: Is there a problem with glidepath (target date funds) for retirement savers?

Post by leonard »

Ignoring taxes - how one invested in the past is a sunk cost.

What matters is matching current risk tolerance to portfolio stock/bond ratio. If a TR mirrors the investors risk tolerance, I don't see that they are losing out in any way by having the past (sunk cost) contributed money to a TR fund.
Leonard | | Market Timing: Do you seriously think you can predict the future? What else do the voices tell you? | | If employees weren't taking jobs with bad 401k's, bad 401k's wouldn't exist.
User avatar
Kosmo
Posts: 1303
Joined: Wed Sep 05, 2012 11:54 am
Location: Philadelphia

Re: Is there a problem with glidepath (target date funds) for retirement savers?

Post by Kosmo »

Browser wrote:Kosmo - I'm not sure a fully understand your point...How is that relevant?
I don't understand your point either. You seem to be complaining that an equal weighting (averaging) of unequally weighted numbers (allocation percentages) will skew the result towards the weighting at larger values (higher portfolio balances). But that's how math works.
User avatar
backpacker
Posts: 1620
Joined: Mon Sep 22, 2014 2:17 pm

Re: Is there a problem with glidepath (target date funds) for retirement savers?

Post by backpacker »

This is why I prefer age in stocks to age in bonds. Turns out, reverse glide paths increase returns and reduces risk. :twisted:
User avatar
bobcat2
Posts: 6076
Joined: Tue Feb 20, 2007 2:27 pm
Location: just barely Outside the Beltway

Re: Is there a problem with glidepath (target date funds) for retirement savers?

Post by bobcat2 »

Is there a problem with glidepath (target date funds) for retirement savers?
Yes.

Taylor Larimore wrote:
It is notable that experts in ALL major companies (including Vanguard) have designed similar target-date glide paths. To my knowledge, they ALL begin with a high allocation to stocks at younger ages and end with a high allocation to bonds in retirement. The primary reason is that most retirees cannot afford to lose their life-savings in a stock market crash.
The above assertion is false. DFA is a major company. DFA does not offer target date funds because the company considers target date funds to be an inferior product.

Here is a short DFA video explaining their view of target date funds.
Explaining What's Wrong with Target Date Funds
http://us.dimensional.com/services/dc-s ... nding.aspx

BobK
In finance risk is defined as uncertainty that is consequential (nontrivial). | The two main methods of dealing with financial risk are the matching of assets to goals & diversifying.
User avatar
Miriam2
Posts: 4387
Joined: Fri Nov 14, 2014 10:51 am

Re: Is there a problem with glidepath (target date funds) for retirement savers?

Post by Miriam2 »

Browser wrote:But I am somewhat concerned that people without a lot of knowledge think, or are led to believe by fund companies and retirement plan sponsors, that TDFs are a kind of substitute for a defined benefit pension and that they can safely "glide" the individual to a known finishing point with their nestegg being expertly "de-risked" as they approach the landing zone. Nothing could be further from the truth, as many learned in 2007-2009 I'm afraid.
Based on what I've seen, the Target Date Funds may actually become a "substitute for a defined pension" - but not because Vanguard is usurping the defined benefit pensions, but because employers are no longer offering defined benefit pensions. What else is there that is as good for the average worker? You're right that workers should not rely on Target funds like the reliability of a pension, but what else is there if employers are requiring employees to fund their own retirement.

There was some criticism of the asset allocations of the various Target funds when the great recession came in 2008-2009 when retirees and also the "older but newly out of work" workers found themselves in Target funds too stock heavy with losses in portfolio value.

But was that the "fault" of the mutual fund companies' AA, or was it the perfect storm of bad financial luck? Who knew it would be that bad? Should Vanguard design their Target funds for another great recession, that may not come any time soon, leaving fund owners too bond heavy too soon - just in case another great recession comes? Then be criticized for not preparing fund owners adequately when they retire with too low a balance, not enough equities, in a normal market retirement? I don't know, but I see the conundrum.
User avatar
nedsaid
Posts: 19275
Joined: Fri Nov 23, 2012 11:33 am

Re: Is there a problem with glidepath (target date funds) for retirement savers?

Post by nedsaid »

bobcat2 wrote:Is there a problem with glidepath (target date funds) for retirement savers?
Yes.

Taylor Larimore wrote:
It is notable that experts in ALL major companies (including Vanguard) have designed similar target-date glide paths. To my knowledge, they ALL begin with a high allocation to stocks at younger ages and end with a high allocation to bonds in retirement. The primary reason is that most retirees cannot afford to lose their life-savings in a stock market crash.
The above assertion is false. DFA is a major company. DFA does not offer target date funds because the company considers target date funds to be an inferior product.

Here is a short DFA video explaining their view of target date funds.
Explaining What's Wrong with Target Date Funds
http://us.dimensional.com/services/dc-s ... nding.aspx

BobK
I will come to Taylor's defense here. Target Date funds are actually a pretty good product but as in all things investing have their flaws. The thing is you could take even the best thought out strategies and come up with objections. One can shoot holes in almost anything.

Target Date funds are vastly superior to what most investors would come up with their own. You get an investment that gets more conservative as you get older. You get asset mixes that are the products of the best thinking of the mutual fund company. You get automatic rebalancing.

Most Bogleheads probably don't use Target Date funds and like me "bake their own." I use Target Date and Target Risk funds as a model for my own asset allocation. I certainly don't follow them exactly but it gives me an idea of whether or not I am on the right track or not. For example, my retirement portfolio has a similar asset mix to Vanguard's 2025 fund. I have looked at other fund companies' target date funds. I know that I am not way out there in left field somewhere.

I know Bob is real big an liability matching and he likes to use things like TIPS ladders and Single Premium Immediate Annuities. But again, nothing is foolproof. When I get near retirement, I will probably do a mix of strategies to deal with the sequence of returns problem. I will consider the things Bob has talked about.

Of course DFA will say that Target Date Funds are inferior. They want investors to go through advisors to buy their funds. Are DFA advisors going to tell people to be more aggressive as they get older? Are DFA advisors going to tell their clients not to rebalance? Are DFA advisors going to tell their clients not to diversify across asset classes? My suspicion is that they will do many of the things through their advice that the Target Date funds do for free. That is diversify across asset classes, become more conservative as you get older, and periodically rebalance.

Bob talks about matching strategies. You can buy Single Premium Immediate Annuities through Vanguard. You can build TIPS ladders at Vanguard Brokerage. You don't have to go through DFA to do DFA recommended strategies.

If I need let's say $40,000 in income above Social Security when I retire, that translates into needing approximately $600,000 in cash to buy Single Premium Immediate Annuities to replace that income. One can say, well you need to think about income in retirement rather than a "magic number" that you need in a retirement nest egg. Yes that is a good way to think about retirement planning but to buy the needed income there is still a "number" or a set amount of cash needed to buy that income.

There is a cost to those TIPS ladders.

Thinking about a "number" is not irrational though there might be better ways to think of income replacement in retirement. The thing is that my focus for years has been to make my nest egg as large as possible taking reasonable amounts of risk. If the idea is to match assets against future liabilities, I have to have those assets in the first place. To grow those assets large enough to match against future liabilities in retirement, I have to take a certain amount of risk. Very high savings rates by themselves are not going to get me there. And risk means stocks.

The problem is that all kinds of things can go wrong. But I am doing the very best I can with limited earnings and whatever I can save and invest. If I had millions of dollars to start with, I could design a much more perfect plan and match out everything perfectly. But I am not in that position and have to resort to imperfect plans and vehicles to get there.

There is a bit of mental gymnastics going on here. A bit less than meets the eye.
A fool and his money are good for business.
Rodc
Posts: 13601
Joined: Tue Jun 26, 2007 9:46 am

Re: Is there a problem with glidepath (target date funds) for retirement savers?

Post by Rodc »

It is true that the dollar weighted average allocation to bonds is higher than a naive glance at the glide path might suggest.

I am not sure why that is a problem. A rational person designing their own glide path may very well do the same thing because as they age they have less ability to take on risk because they have less time to refill the portfolio from income, they may get caught being tossed into early retirement during a down turn etc. (of course someone else due to having done better than expected might find they have so much money they can take more risk than they expected).

Obviously it is better to know enough to set your allocation based on your personal situation, personal goals, personal assessment of risk tolerance etc. but I don't think the general glide path idea is all that flawed for folks who want a simple product that requires little if any thought on their part.

That said, I am not particularly sold on these products being better than a fixed 60/40 lifetime allocation, or 60/40 to age 60(ish) then 50/50 or 40/60. That is two or three fixed allocation blend funds frankly works fine too, and allows from more personally tailoring than a single target date fund.
We live a world with knowledge of the future markets has less than one significant figure. And people will still and always demand answers to three significant digits.
User avatar
ogd
Posts: 4876
Joined: Thu Jun 14, 2012 11:43 pm

Re: Is there a problem with glidepath (target date funds) for retirement savers?

Post by ogd »

Browser wrote:The average allocation to stocks and bonds over this period of time appears to be 50%. However, the dollar-weighted allocation to stocks is actually much lower while the dollar-weighted allocation to bonds is much higher.
There are two issues with averaging over time like this.

One is that if you count one's future earnings as a weighted factor, the early allocation is in fact much more conservative. Now one can argue that earnings are more stock-like than bond-like, but I think they do have a bond-like quality at least partially; after all, you should be able to do something even in a bad economy to live on and save a bit. This varies between people, but in my mind it's the strongest argument for glide paths, stronger than the below.

The other is the instant-risk argument. Regardless of the past risk I took, I have to somehow solve the problem of how much risk I'm taking now. So I can calculate a glide path that optimizes something over the lifetime, but if the instantaneous risk spikes to huge amounts at age 50 for example, I'm not gonna be happy with that regardless of the fact that the average risk was lower or whatnot. That moment, at 50, would be too much exposure for that glide path to be acceptable to me. This is the problem with anti-rebalancing arguments as well -- you can integrate over lifetime to argue that I shouldn't be rebalancing, but when I'm facing a 90/10 allocation because stocks have done well, I'll throw out the integral and take some money off the table nonetheless because I've seen what 90/10 allocations can do within a year and I don't want to be staring at that kind of financial picture (which includes the partially stock-like quality of personal earnings) ever again.
User avatar
bertilak
Posts: 10725
Joined: Tue Aug 02, 2011 5:23 pm
Location: East of the Pecos, West of the Mississippi

Re: Is there a problem with glidepath (target date funds) for retirement savers?

Post by bertilak »

nedsaid wrote:If I need let's say $40,000 in income above Social Security when I retire, that translates into needing approximately $600,000 in cash to buy Single Premium Immediate Annuities to replace that income. One can say, well you need to think about income in retirement rather than a "magic number" that you need in a retirement nest egg. Yes that is a good way to think about retirement planning but to buy the needed income there is still a "number" or a set amount of cash needed to buy that income.

There is a cost to those TIPS ladders.

Thinking about a "number" is not irrational though there might be better ways to think of income replacement in retirement. The thing is that my focus for years has been to make my nest egg as large as possible taking reasonable amounts of risk. If the idea is to match assets against future liabilities, I have to have those assets in the first place. To grow those assets large enough to match against future liabilities in retirement, I have to take a certain amount of risk. Very high savings rates by themselves are not going to get me there. And risk means stocks.

The problem is that all kinds of things can go wrong. But I am doing the very best I can with limited earnings and whatever I can save and invest. If I had millions of dollars to start with, I could design a much more perfect plan and match out everything perfectly. But I am not in that position and have to resort to imperfect plans and vehicles to get there.

There is a bit of mental gymnastics going on here. A bit less than meets the eye.
Well said, nedsaid.

I see some important concepts in most of bobcat's writings:
  1. Liability matching is the best way to think about and plan for retirement. As a fairly recent retiree it has certainly hit home that "cash flow is king" once retired. But as you say, it costs assets to produce that cash flow so one really needs to think about growing/accumulating those assets to give one flexibility in doing so.
  2. Making a distinction between guaranteed and expected cash flow is very important. Guaranteed cash flow is more expensive.
  3. Bobcat often discusses products and techniques for generating the difficult, guaranteed part, of cash flow.
  4. I am still trying to digest bobcat's latest discussion of "liability driven investment (LDI) strategy" so I may have missed the point. It might be addressing at least part of what you say in that it discusses how to be sure you have "enough" to pay for that guaranteed income when you need it. It may boil down to the idea that paying the cost for a reliable income stream is a liability in itself and that too needs to be part of the plan. Or am I getting too recursive here?
I think stressing the accumulation/growth part of investing is most important for younger investors. Actually it is perhaps the only important thing early on, although being aware of item number 2, above, is wise.

Once one is within a few short years of retirement, planning on how to reliably produce the cash flow becomes the most important thing but that sneaks up on you and you don't want to be surprised!
May neither drought nor rain nor blizzard disturb the joy juice in your gizzard. -- Squire Omar Barker (aka S.O.B.), the Cowboy Poet
Rodc
Posts: 13601
Joined: Tue Jun 26, 2007 9:46 am

Re: Is there a problem with glidepath (target date funds) for retirement savers?

Post by Rodc »

ogd wrote:
Browser wrote:The average allocation to stocks and bonds over this period of time appears to be 50%. However, the dollar-weighted allocation to stocks is actually much lower while the dollar-weighted allocation to bonds is much higher.
There are two issues with averaging over time like this.

One is that if you count one's future earnings as a weighted factor, the early allocation is in fact much more conservative. Now one can argue that earnings are more stock-like than bond-like, but I think they do have a bond-like quality at least partially; after all, you should be able to do something even in a bad economy to live on and save a bit. This varies between people, but in my mind it's the strongest argument for glide paths, stronger than the below.

The other is the instant-risk argument. Regardless of the past risk I took, I have to somehow solve the problem of how much risk I'm taking now. So I can calculate a glide path that optimizes something over the lifetime, but if the instantaneous risk spikes to huge amounts at age 50 for example, I'm not gonna be happy with that regardless of the fact that the average risk was lower or whatnot. That moment, at 50, would be too much exposure for that glide path to be acceptable to me. This is the problem with anti-rebalancing arguments as well -- you can integrate over lifetime to argue that I shouldn't be rebalancing, but when I'm facing a 90/10 allocation because stocks have done well, I'll throw out the integral and take some money off the table nonetheless because I've seen what 90/10 allocations can do within a year and I don't want to be staring at that kind of financial picture (which includes the partially stock-like quality of personal earnings) ever again.
That is a good point. And even if one think the integral over time approach is good, one still has to integrate from now forward, not over a life time. If one is 60 they would not integrate starting 35 years earlier. They would start with what already happened, where they are today, and estimate forward.
We live a world with knowledge of the future markets has less than one significant figure. And people will still and always demand answers to three significant digits.
User avatar
nedsaid
Posts: 19275
Joined: Fri Nov 23, 2012 11:33 am

Re: Is there a problem with glidepath (target date funds) for retirement savers?

Post by nedsaid »

Bertilak, BobK is a very knowledgeable poster and I appreciate his contributions to this forum. His posts do take some time to read and digest because there is so much there. I certainly don't disagree with the liability matching strategies and he has caused me to really think about some things. Particularly when you are older and have accumulated quite a nest egg.

I am not saying Bob is wrong. There certainly are flaws in Target Date funds and in other threads have pointed these flaws out myself. The problem is that in the investing world and in the world at large, almost everything is flawed. We have to use the good but imperfect to achieve excellent though not perfect results. There are flaws and drawbacks to everything.

Can one do better than Target Date Funds? Probably. Will most investors take the time and trouble to "bake their own" and come up with something better? No. Gosh, it is hard enough to get people to save in the first place and even harder to get them to get them to invest. For people with 401k plans and who have no idea what to do, I recommend for them a Target Date fund corresponding to their expected retirement date or a Moderate Risk Target Risk Fund. For most people, a target date fund is as good as it gets. I also recommend that investors consider annuitizing or "pensionizing" a part of their nest egg upon retirement.

I have talked to a lot of people about investing. Most folks don't have the passion that I do for it and I find that their eyes glaze over after a few minutes. Target Date Funds are perfect for those with glazed over eyes or the "deer in the headlights" expression on their face. That is about as much as they can handle.

It is too bad that pensions are for the most part gone. It is too bad that savings rates and investment choices aren't done for us. But we have to life with the world as it actually is and not how we wish it would be. In the absence of pensions, why not invest our workplace savings like the pension funds do and why not create a pension with part of our nest egg?
A fool and his money are good for business.
User avatar
bertilak
Posts: 10725
Joined: Tue Aug 02, 2011 5:23 pm
Location: East of the Pecos, West of the Mississippi

Re: Is there a problem with glidepath (target date funds) for retirement savers?

Post by bertilak »

nedsaid wrote:Can one do better than Target Date Funds? Probably. Will most investors take the time and trouble to "bake their own" and come up with something better? No. Gosh, it is hard enough to get people to save in the first place and even harder to get them to get them to invest. For people with 401k plans and who have no idea what to do, I recommend for them a Target Date fund corresponding to their expected retirement date or a Moderate Risk Target Risk Fund. For most people, a target date fund is as good as it gets. I also recommend that investors consider annuitizing or "pensionizing" a part of their nest egg upon retirement.
Agree 100%.
May neither drought nor rain nor blizzard disturb the joy juice in your gizzard. -- Squire Omar Barker (aka S.O.B.), the Cowboy Poet
User avatar
bobcat2
Posts: 6076
Joined: Tue Feb 20, 2007 2:27 pm
Location: just barely Outside the Beltway

Re: Is there a problem with glidepath (target date funds) for retirement savers?

Post by bobcat2 »

nedsaid wrote:If I need let's say $40,000 in income above Social Security when I retire, that translates into needing approximately $600,000 in cash to buy Single Premium Immediate Annuities to replace that income. One can say, well you need to think about income in retirement rather than a "magic number" that you need in a retirement nest egg. Yes that is a good way to think about retirement planning but to buy the needed income there is still a "number" or a set amount of cash needed to buy that income.
The above is incorrect.

If a 65 year old male had purchased a real life annuity in 2007 that produced $40,000/year in income, it would have cost about $625,000. Had a 65 year old male purchased a real life annuity just two years later in 2009 that produced $40,000/year in income, it would have cost about $950,000. This is because interest rates fell between 2007 and 2009.

Life annuities have long duration. Their prices and payouts are very sensitive to changes in interest rates. You can’t set aside a given portfolio value and think you have locked in a level of annuitized income. That is folly.

See the following thread for more on this issue.
viewtopic.php?f=10&t=174991

I have a couple of questions for Nedsaid.

If you asked me about the value of my db retirement plan would you consider it reasonable, if I replied its present value is $390,000 and currently the AA of the plan is 45% equity/55% bonds and they intend to keep that AA?

If you asked me about the value of my SS retirement benefit would you consider it reasonable, if I replied its present value is $415,000 and the SS surplus is invested in Treasury bonds?

A target date fund is supposedly a retirement plan. So why it is reasonable to state its value in terms of its present value and AA now and its AA at retirement? Yet that's what we do. If asked the value of my target date fund retirement plan, I reply its current value is $450,000 and its current AA is 65/35 and at its target date the AA will be 40/60.

Shouldn't all of these retirement plans have their stated value in the same metrics? So either SS and db plan values should be stated in terms of present value and AA instead of annuitized income available, or the value of target date funds should be stated in terms of annuitized income available.

Stating the value of target date funds in terms of annuitized income available is a proposed change in information that 401k plans need to provide to participants. Companies that sell target date funds are fighting this rule change. Reporting this information would show that the amount of annuitized income a target date fund is producing over time is bouncing all over the place. This is because of both changes in portfolio values and changes in interest rates over time. It would show that target date funds are very risky in terms of producing reliable retirement income.

BobK

PS - I believe that target date funds are some improvement over the options in most 401k plans. If 401k plan offerings were high quality options and few in number, I think that would be just as good. A 401k plan that offered the follow four funds would be just as good in my opinion - a broad based US stock fund, a broad based international stock fund, and ST and LT TIPS funds. The current DC plan design is a poor design for being the primary retirement asset. The design of DC plans needs to be reformed with the focus on the output (income) and much less focus on input values such as AA. In short, DC plans need to be redesigned to perform more like DB plans.
In finance risk is defined as uncertainty that is consequential (nontrivial). | The two main methods of dealing with financial risk are the matching of assets to goals & diversifying.
Topic Author
Browser
Posts: 4857
Joined: Wed Sep 05, 2012 4:54 pm

Re: Is there a problem with glidepath (target date funds) for retirement savers?

Post by Browser »

It's worth noting that Arnott, et. al. found that, if you measure the success of glidepath in terms of ending portfolio value (higher being better), the range of ending portfolio values (smaller being better), or the downside ending portfolio value (higher being better) then a static allocation apparently beats a declining equity glidepath, and in turn is beaten by a rising equity allocation glidepath. Now I'm not sure I would actually want to increase my equity allocation over time, but I think I'd be comfortable simply holding a static allocation such as 50/50 or 60/40 and not using a declining equity glidepath. I've not seen any data to contradict Arnott's findings that you would probably be better off to do so. It seems shameful to me that the providers of glidepath funds have not, as far as I know, presented any compelling data to support using these funds; as opposed to merely holding a static 50/50 or 60/40 allocation until one nears or reaches retirement age. Where's the data folks?
We don't know where we are, or where we're going -- but we're making good time.
User avatar
ObliviousInvestor
Posts: 4212
Joined: Tue Mar 17, 2009 9:32 am
Contact:

Re: Is there a problem with glidepath (target date funds) for retirement savers?

Post by ObliviousInvestor »

Browser wrote:It's worth noting that Arnott, et. al. found that, if you measure the success of glidepath in terms of ending portfolio value (higher being better), the range of ending portfolio values (smaller being better), or the downside ending portfolio value (higher being better) then a static allocation apparently beats a declining equity glidepath, and in turn is beaten by a rising equity allocation glidepath. Now I'm not sure I would actually want to increase my equity allocation over time, but I think I'd be comfortable simply holding a static allocation such as 50/50 or 60/40 and not using a declining equity glidepath. I've not seen any data to contradict Arnott's findings that you would probably be better off to do so. It seems shameful to me that the providers of glidepath funds have not, as far as I know, presented any compelling data to support using these funds; as opposed to merely holding a static 50/50 or 60/40 allocation until one nears or reaches retirement age. Where's the data folks?
David Blanchett has a piece from earlier this year that you might find interesting:
https://www.onefpa.org/journal/pages/fe ... -path.aspx
Mike Piper | Roth is a name, not an acronym. If you type ROTH, you're just yelling about retirement accounts.
User avatar
Miriam2
Posts: 4387
Joined: Fri Nov 14, 2014 10:51 am

Re: Is there a problem with glidepath (target date funds) for retirement savers?

Post by Miriam2 »

Bobcat2 - you're back on the prowl :happy
bobcat2 wrote:A target date fund is supposedly a retirement plan. . . .
It would show that target date funds are very risky in terms of producing reliable retirement income.
1 - I'm not sure this is accurate - "A target date fund is supposedly a retirement plan."
It's not a retirement plan. It is an all-in-one mutual fund that a person can invest in as part of their retirement plan.

2 - I can't see this - "It would show that target date funds are very risky in terms of producing reliable retirement income."
Very risky compared to what? Compared to people placing their hard earned pay in some type of Lord Abbott high yield ER 1.5% plus 12b1 fees and Oppenheimer small & micro cap growth ER 1.25% and Templeton European emerging markets ER 1.45%?

3 - I watched your link to the Zvi Bodie presentation for DFA funds "How Target Date Funds Glidepath can lead to a Crash Landing".
Is this really accurate - a crash landing? Or is this scare tactics (does DFA benefit from this sort of talk?).

Zvi Bodie said it's a crash landing because, unlike an airplane's glidepath, "there is no expected time of arrival and no destination." I don't see the problem. Actually, even Bodie admits that Target Date funds in 401k's are a "big improvement over giving no guidance whatsoever, the unlimited choice with no guidance" which is what so many 401k plans do with their long lists of incomprehensible funds.

The Vanguard target date glidepath - designed for the average person - glides down to THE DATE, which Vanguard uses age 65. At "Early Retirement" of age 65-72, the AA is 50/50 and glides to 30/70 at age 72. At "Late Retirement" age 72 and beyond, the AA fixes at 30/70, which is the Target Retirement Income Fund [VTINX].

Do you mean this fund is "very risky in terms of producing reliable retirement income?" What is better - Nedsaid's "bake your own?" The average person can't and won't bake their own.
BTW - love the imagery, Nedsaid!
Last edited by Miriam2 on Sat Oct 10, 2015 8:57 pm, edited 1 time in total.
User avatar
bobcat2
Posts: 6076
Joined: Tue Feb 20, 2007 2:27 pm
Location: just barely Outside the Beltway

Re: Is there a problem with glidepath (target date funds) for retirement savers?

Post by bobcat2 »

Hi Meriam2,

Target date fund advocates promote the funds on the basis that they de-risk retirement assets. The truth is that in terms of providing reliable retirement income, target date funds are very risky at their target date. People may well make even riskier decisions on their own, but the fact remains that target date funds are risky in terms of providing targeted levels of retirement income.

I'm guessing that many people investing their 401k assets in target date funds believe they are safe. Those investors are sadly misinformed. :(

BobK
In finance risk is defined as uncertainty that is consequential (nontrivial). | The two main methods of dealing with financial risk are the matching of assets to goals & diversifying.
User avatar
Miriam2
Posts: 4387
Joined: Fri Nov 14, 2014 10:51 am

Re: Is there a problem with glidepath (target date funds) for retirement savers?

Post by Miriam2 »

bobcat2 wrote:Target date fund advocates promote the funds on the basis that they de-risk retirement assets. The truth is that in terms of providing reliable retirement income, target date funds are very risky at their target date.
The "target date" meaning age 65 (at least for Vanguard) which is 50% stocks - is that the problem date you're talking about? That it should be less than 50% stocks at that date?
You're not talking about age 80, for example, where the AA would be 30% stocks? That's ok?
User avatar
iceport
Posts: 6054
Joined: Sat Apr 07, 2007 4:29 pm

Re: Is there a problem with glidepath (target date funds) for retirement savers?

Post by iceport »

bobcat2 wrote:Stating the value of target date funds in terms of annuitized income available is a proposed change in information that 401k plans need to provide to participants. Companies that sell target date funds are fighting this rule change. Reporting this information would show that the amount of annuitized income a target date fund is producing over time is bouncing all over the place. This is because of both changes in portfolio values and changes in interest rates over time. It would show that target date funds are very risky in terms of producing reliable retirement income.
I think I understand your rationale here — essentially, to demonstrate one aspect of how a target date fund is not a retirement plan. However, I do see a problem with the proposal. Outside of this and similarly focused forums, folks who invest in target date funds are precisely the least likely investors to ever read their statements closely or frequently enough to see or understand the variability they need to see and understand.
"Discipline matters more than allocation.” |—| "In finance, if you’re certain of anything, you’re out of your mind." ─William Bernstein
User avatar
Miriam2
Posts: 4387
Joined: Fri Nov 14, 2014 10:51 am

Re: Is there a problem with glidepath (target date funds) for retirement savers?

Post by Miriam2 »

Vanguard has a relatively new paper on their Target Date funds and glidepaths:
https://personal.vanguard.com/pdf/icrtdf.pdf

LadyGeek - I see our Wiki has an old version as a reference www.bogleheads.org/wiki/glide_paths
User avatar
bobcat2
Posts: 6076
Joined: Tue Feb 20, 2007 2:27 pm
Location: just barely Outside the Beltway

Re: Is there a problem with glidepath (target date funds) for retirement savers?

Post by bobcat2 »

Miriam2 wrote:
bobcat2 wrote:Target date fund advocates promote the funds on the basis that they de-risk retirement assets. The truth is that in terms of providing reliable retirement income, target date funds are very risky at their target date.
The "target date" meaning age 65 (at least for Vanguard) which is 50% stocks - is that the problem date you're talking about? That it should be less than 50% stocks at that date?
You're not talking about age 80, for example, where the AA would be 30% stocks? That's ok?
A portfolio value at retirement has little to do with the amount of reliable retirement income that portfolio will produce. There is a mismatch of goals and methods. Setting an AA as the target of the retirement plan has even less relationship with reliable retirement income. What disengaged investors want out of their retirement plan is reliable retirement income, not a particular asset allocation in their retirement year. What they want is reasonable, but they don't get that from target date funds.

See this linked thread for more details on one aspect of the problem.
viewtopic.php?f=10&t=174991

IMO at a minimum a target date fund should show throughout pre-retirement how much annuitized income that fund will produce. Take the case of a 58 year old who has a TDF and wants to retire at age 65. Then the current fund statement should show how much a deferred annuity purchased today, and beginning payouts at age 65, would produce as retirement income. In that way the investor could see how well she is prepared for retirement by adding that income to the income she will be receiving from SS. But instead of that useful information, she gets to see what her AA will be at age 65. What is this individual, who knows little about investing or retirement planning, supposed to do with knowing her AA at age 65? :?

But even this improvement leaves aside the fact that the amount of income that will be available at retirement will vary widely due to changes in interest rates and portfolio values. TDFs are very risky for investment products that are promoted as being low risk retirement products for people approaching retirement.

BobK
Last edited by bobcat2 on Sat Oct 10, 2015 3:09 pm, edited 1 time in total.
In finance risk is defined as uncertainty that is consequential (nontrivial). | The two main methods of dealing with financial risk are the matching of assets to goals & diversifying.
User avatar
nedsaid
Posts: 19275
Joined: Fri Nov 23, 2012 11:33 am

Re: Is there a problem with glidepath (target date funds) for retirement savers?

Post by nedsaid »

bobcat2 wrote:
nedsaid wrote:If I need let's say $40,000 in income above Social Security when I retire, that translates into needing approximately $600,000 in cash to buy Single Premium Immediate Annuities to replace that income. One can say, well you need to think about income in retirement rather than a "magic number" that you need in a retirement nest egg. Yes that is a good way to think about retirement planning but to buy the needed income there is still a "number" or a set amount of cash needed to buy that income.
The above is incorrect.

If a 65 year old male had purchased a real life annuity in 2007 that produced $40,000/year in income, it would have cost about $625,000. Had a 65 year old male purchased a real life annuity just two years later in 2009 that produced $40,000/year in income, it would have cost about $950,000. This is because interest rates fell between 2007 and 2009.


Nedsaid: I went by memory. Within the last year or so, I went to immediateannuities.com and priced out what $100,000 would buy in monthly income for a 65 year old single male. It came out to about $550 a month. That is what I based my figures on. Interest rates have not changed substantially since then.

I just checked, $100,000 for a single male at age 65 with a 10 year certain comes in at $553/a month. $553 X 12 = $6,636 a year. So if I take the $40,000 and divide it by the $6,636, I come up with 6.027. So my $600,000 figure is pretty darned close.


Life annuities have long duration. Their prices and payouts are very sensitive to changes in interest rates. You can’t set aside a given portfolio value and think you have locked in a level of annuitized income. That is folly.

Nedsaid: Bob, I realize that the monthly payouts depend upon the level of interest rates at the time of purchase. Higher interest rates cause higher payouts and lower interest rates cause lower payouts. Once you buy, you are locked in at that interest rate and locked in at your monthly payment. So of course, the $600,000 figure I cited would vary with the level of interest rates.

See the following thread for more on this issue.
viewtopic.php?f=10&t=174991

I have a couple of questions for Nedsaid.

If you asked me about the value of my db retirement plan would you consider it reasonable, if I replied its present value is $390,000 and currently the AA of the plan is 45% equity/55% bonds and they intend to keep that AA?

Nedsaid: You would do a present value calculation based on the monthly payments you would receive and the current interest rate. The values you cite sound reasonable to me, I assume you got the asset mix from the annual report you got from the pension.

If you asked me about the value of my SS retirement benefit would you consider it reasonable, if I replied its present value is $415,000 and the SS surplus is invested in Treasury bonds?

Nedsaid: That sounds about right. Again you would do a present value calculation. You raise a good point, most people don't realize the value of Social Security. Jack Bogle has made comments about "Social Security as a bond" when trying to figure out an appropriate asset allocation. He has said in interviews that many retirees are too conservatively invested. That is surprising coming from him but I did hear him say it in a Morningstar interview. In fact, he thought that 65/35 is an appropriate balanced portfolio for many investors. Of course, the fact that Social Security is inflation adjusted would also add to the present value calculation.

A target date fund is supposedly a retirement plan. So why it is reasonable to state its value in terms of its present value and AA now and its AA at retirement? Yet that's what we do. If asked the value of my target date fund retirement plan, I reply its current value is $450,000 and its current AA is 65/35 and at its target date the AA will be 40/60.

Shouldn't all of these retirement plans have their stated value in the same metrics? So either SS and db plan values should be stated in terms of present value and AA instead of annuitized income available, or the value of target date funds should be stated in terms of annuitized income available.

Stating the value of target date funds in terms of annuitized income available is a proposed change in information that 401k plans need to provide to participants. Companies that sell target date funds are fighting this rule change. Reporting this information would show that the amount of annuitized income a target date fund is producing over time is bouncing all over the place. This is because of both changes in portfolio values and changes in interest rates over time. It would show that target date funds are very risky in terms of producing reliable retirement income.

Nedsaid: This is a good point. I think what would happen if 401k plans were stated in terms of income generated, many people would realize that their retirements are seriously underfunded. $150,000 might sound like all the money in the world until you start living on it. They would also be aghast at how volatile the estimated income streams are.

But seriously Bob, really all you are saying is that markets are risky and that someone is taking risks in guaranteeing monthly income. That is why most companies have gotten out of the pension business. I was in a cash balance pension at a time when 8% was guaranteed for the year and the market went down 50%. I found out later the fund had a 70/30 allocation. You can do the math on that one, as I recall the market losses were close to 1/2 billion dollars. Ouch!! They froze the pension. What they were doing was guaranteeing the inflation rate plus three percent. Even a relatively conservative fund should do that but no one counted on the financial crisis.

A lot of public employees would be aghast at how underfunded their pensions are. Last I heard, Illinois was 50% funded. You can state the balance in terms of dollar income all you want but the fact is the public employees are taking a lot of market risk. The risk is there whether it is apparent as it is in a 401k plan or hidden behind the curtain as it is in a pension plan.

Investing is risky no matter whether you state the value of a retirement account in terms of present value or in retirement income. As you stated yourself, the values of both could be quite volatile. Like it or not, you will be taking investment risk until you annuitize the portfolio, match your liabilities with bonds, or a combination of both. The question is what to you do in the mean time?


BobK

PS - I believe that target date funds are some improvement over the options in most 401k plans. If 401k plan offerings were high quality options and few in number, I think that would be just as good. A 401k plan that offered the follow four funds would be just as good in my opinion - a broad based US stock fund, a broad based international stock fund, and ST and LT TIPS funds. The current DC plan design is a poor design for being the primary retirement asset. The design of DC plans needs to be reformed with the focus on the output (income) and much less focus on input values such as AA. In short, DC plans need to be redesigned to perform more like DB plans.
Nedsaid: Yes Bob, but you still need the actual assets to buy the income or to buy the matching bonds. You are playing a bit of a semantic game here.
A fool and his money are good for business.
Rodc
Posts: 13601
Joined: Tue Jun 26, 2007 9:46 am

Re: Is there a problem with glidepath (target date funds) for retirement savers?

Post by Rodc »

Browser wrote:It's worth noting that Arnott, et. al. found that, if you measure the success of glidepath in terms of ending portfolio value (higher being better), the range of ending portfolio values (smaller being better), or the downside ending portfolio value (higher being better) then a static allocation apparently beats a declining equity glidepath, and in turn is beaten by a rising equity allocation glidepath. Now I'm not sure I would actually want to increase my equity allocation over time, but I think I'd be comfortable simply holding a static allocation such as 50/50 or 60/40 and not using a declining equity glidepath. I've not seen any data to contradict Arnott's findings that you would probably be better off to do so. It seems shameful to me that the providers of glidepath funds have not, as far as I know, presented any compelling data to support using these funds; as opposed to merely holding a static 50/50 or 60/40 allocation until one nears or reaches retirement age. Where's the data folks?
I have not read Arnott's paper but have looked at some of this in some detail myself and I have read Wade's rising glide path paper. The differences are all just noise and largely driven by assumptions about things that are not at all known. Glide path that averages 60/40 is going to do about the same as a static 60/40 on average. I don't think anyone advertises any differently. The advantage is mostly psychological, but since most of us are not Spock that is not all bad. I prefer to hold a static allocation for many years rather than shift 1% or so a year, but that slow shift does help some because being gradual people don't get hung up on what date to make shift. I have not noticed providers claiming any magical powers to their glide slope funds, just that they are simple suitable buy and hold products. And while not perfect are pretty good, at least the low cost ones. I don't think the low cost providers have anything to be ashamed of.

http://home.comcast.net/~rodec/finance/ ... nBonds.pdf
viewtopic.php?f=10&t=149402&p=2234204&h ... e#p2234204
We live a world with knowledge of the future markets has less than one significant figure. And people will still and always demand answers to three significant digits.
Rodc
Posts: 13601
Joined: Tue Jun 26, 2007 9:46 am

Re: Is there a problem with glidepath (target date funds) for retirement savers?

Post by Rodc »

ObliviousInvestor wrote:
Browser wrote:It's worth noting that Arnott, et. al. found that, if you measure the success of glidepath in terms of ending portfolio value (higher being better), the range of ending portfolio values (smaller being better), or the downside ending portfolio value (higher being better) then a static allocation apparently beats a declining equity glidepath, and in turn is beaten by a rising equity allocation glidepath. Now I'm not sure I would actually want to increase my equity allocation over time, but I think I'd be comfortable simply holding a static allocation such as 50/50 or 60/40 and not using a declining equity glidepath. I've not seen any data to contradict Arnott's findings that you would probably be better off to do so. It seems shameful to me that the providers of glidepath funds have not, as far as I know, presented any compelling data to support using these funds; as opposed to merely holding a static 50/50 or 60/40 allocation until one nears or reaches retirement age. Where's the data folks?
David Blanchett has a piece from earlier this year that you might find interesting:
https://www.onefpa.org/journal/pages/fe ... -path.aspx
interesting paper, so far only skimmed. Thank you.

The results are consistent within the error bars of all the other such studies. I note that to compare scenarios one to another they need 3 significant figures when the reality is they have at best one. I also note that like all studies I am aware of they make no attempt to estimate the accuracy of their models. If they did they would all report that within the error margin they found no differences worth noting.
To put the overall success rates in perspective, the probability of success for the 4 percent initial withdrawal rate over 30 years, for the 40 percent base equity allocation with a constant glide path and moderate nominal returns and inflation expectations, is 71.5 percent (versus 73.5 percent for the decreasing fast glide path, and 68.2 percent for the increasing fast glide path).
The most honest representation of the data is that for a 4% initial withdrawal rate the success was around 70% for each method.
We live a world with knowledge of the future markets has less than one significant figure. And people will still and always demand answers to three significant digits.
User avatar
nedsaid
Posts: 19275
Joined: Fri Nov 23, 2012 11:33 am

Re: Is there a problem with glidepath (target date funds) for retirement savers?

Post by nedsaid »

Here is another problem with "guaranteed income" or "matching" strategies. You could get all your future liabilities matched perfectly with annuities and bond ladders only to run into a period of sustained inflation. Sustained inflation can just destroy the purchasing power of a guaranteed income stream and it doesn't take that many years to do it. That is why people own volatile asset classes such as stocks and REITs. It is an attempt to beat inflation. Unfortunately, to beat inflation you have to deal with perhaps a lot of volatility.

I suppose the most foolproof way to beat sustained inflation would be a TIPS ladder but I suppose a market scenario is out there that could defeat even that. We have found out that TIPS can be pretty darned volatile. It was something that I didn't expect but markets have a way of surprising you.

Bob, you have raised an excellent point. That is that portfolio value and the "number" is not the end all in retirement planning. You are correct in saying that investors should take into account the amount of retirement income that can be generated. You raise a great point about liability matching. But I think you took a great point and then took it too far. You create the impression that there is a perfect way to deal with all of this when in fact there are pitfalls to just about anything.

The reality is that you still have to deal with inflation. 4% withdrawals from a portfolio of stocks and bonds is one way to deal with it. But as has been discussed in many threads, 4% may not be sustainable in this era of very low interest rates.

So I will probably do a number of things to deal with all of this. Annuitize a portion of my portfolio, delay Social Security as long as I can, take withdrawals from a stock/bond portfolio, and maybe tighten my belt when markets are bad.
A fool and his money are good for business.
Rodc
Posts: 13601
Joined: Tue Jun 26, 2007 9:46 am

Re: Is there a problem with glidepath (target date funds) for retirement savers?

Post by Rodc »

The point that target date funds are not retirement plans while true seems off topic which is about the allocation between stocks and bonds over time. I would think most people even if they don't fully grasp the risks understand that mutual funds do not come with some guaranteed income. And while true that is hardly unique to target retirement mutual funds.

Having a 401k or other retirement savings/investing plans report the level of income one would get if one bought a real annuity is not a bad idea, but is independent it seems to me of whether one invests in target date funds or other funds.

Also, while likely worthwhile, since few people actually buy annuities from what I read and certainly not with their entire portfolio, it is instructive but only partly germane to most people.
We live a world with knowledge of the future markets has less than one significant figure. And people will still and always demand answers to three significant digits.
User avatar
bobcat2
Posts: 6076
Joined: Tue Feb 20, 2007 2:27 pm
Location: just barely Outside the Beltway

Re: Is there a problem with glidepath (target date funds) for retirement savers?

Post by bobcat2 »

The topic - Is there a problem with glidepath (target date funds) for retirement savers?
The topic is not the allocation between stocks and bonds over time.

The criticism is valid. The problem with TDF glidepaths is the concept does not lead to the goal that retirement savers have in mind. The goal of retirement savers is to have enough reliable income in retirement to support their desired retirement living standard. Retirement savers, who no little about investing, are not interested in the AA of their retirement plan in the year of their retirement. Nor are they interested in the AA path between now and retirement. Nor is there any reason they should be interested in that micro level of investing input strategy. What's worse though is setting an AA at retirement has next to nothing to do with acquiring the reliable retirement income they aspire to. This is a design failure.

If 401k plans still had their original purpose of providing discretionary income beyond the retirement income produced by DB plans and SS, they would be an excellent choice for producing additional retirement assets. But as a replacement for DB plans, TDFs by design fall woefully short of meeting a retirement's plan goal of producing reliable retirement income. Quibbling about which 'glide path is best' is a classic case of missing the forest for the trees.

BobK
In finance risk is defined as uncertainty that is consequential (nontrivial). | The two main methods of dealing with financial risk are the matching of assets to goals & diversifying.
User avatar
bobcat2
Posts: 6076
Joined: Tue Feb 20, 2007 2:27 pm
Location: just barely Outside the Beltway

Re: Is there a problem with glidepath (target date funds) for retirement savers?

Post by bobcat2 »

nedsaid wrote:Here is another problem with "guaranteed income" or "matching" strategies. You could get all your future liabilities matched perfectly with annuities and bond ladders only to run into a period of sustained inflation. Sustained inflation can just destroy the purchasing power of a guaranteed income stream and it doesn't take that many years to do it. That is why people own volatile asset classes such as stocks and REITs. It is an attempt to beat inflation. Unfortunately, to beat inflation you have to deal with perhaps a lot of volatility.
Guaranteed income in the form of real life annuities and TIPS ladders are not affected by inflation. SS is a real annuity and is not affected by inflation. Private real annuities are also not affected by inflation. TIPS bonds in a ladder and held to maturity as the steps in the ladder mature, are not affected by inflation. It's true that the portfolio value of the TIPS ladder will vary over time, but the TIPS in the steps of the ladder will mature with known real values.

BobK
In finance risk is defined as uncertainty that is consequential (nontrivial). | The two main methods of dealing with financial risk are the matching of assets to goals & diversifying.
User avatar
nedsaid
Posts: 19275
Joined: Fri Nov 23, 2012 11:33 am

Re: Is there a problem with glidepath (target date funds) for retirement savers?

Post by nedsaid »

bobcat2 wrote:
nedsaid wrote:Here is another problem with "guaranteed income" or "matching" strategies. You could get all your future liabilities matched perfectly with annuities and bond ladders only to run into a period of sustained inflation. Sustained inflation can just destroy the purchasing power of a guaranteed income stream and it doesn't take that many years to do it. That is why people own volatile asset classes such as stocks and REITs. It is an attempt to beat inflation. Unfortunately, to beat inflation you have to deal with perhaps a lot of volatility.
Guaranteed income in the form of real life annuities and TIPS ladders are not affected by inflation. SS is a real annuity and is not affected by inflation. Private real annuities are also not affected by inflation. TIPS bonds in a ladder and held to maturity as the steps in the ladder mature, are not affected by inflation. It's true that the portfolio value of the TIPS ladder will vary over time, but the TIPS in the steps of the ladder will mature with known real values.

BobK
Real annuities are another thing I might consider. There is a trade-off in that your monthly payments start out lower. When buying a Single Premium Immediate Annuity with no inflation adjustments, in effect you are getting your inflation adjustment up front. Believe me, the insurance companies have this figured out.

Social Security is a real annuity which is why I hope to delay taking it as long as possible. By delaying Social Security, you are in effect buying an inflation adjusted annuity with your nest egg dollars.

TIPS ladders I think are a good solution but people would be aghast at their volatility. A TIPS ladder might be the best solution if people don't peek at the portfolio value.

There are pitfalls and trade-offs to everything. There is no perfect solution for retirement income but there are some pretty good options.

The other thing that wasn't really addressed is why are both private and public pension plans in trouble? This tells me that pensions are riskier and more expensive than advertised. My understanding is that pensions used to meet their guarantees by matching future liabilities with bonds. Somewhere along the line, the light went off that pension plans could achieve higher returns if they invested in stocks too. Higher returns meant lower contribution rates. The problem is that stocks caused the underlying portfolios to become more volatile.

We could make pensions a lot "safer" if they went to a matching strategy with 100% bonds. But that means higher contribution rates and nobody wants to hear about that. There is also a sequence of returns problem with stocks and that applies to pension plans too and not to just individuals.

If achieving guaranteed income was so easy, why have private companies mostly gone out of the pension business and why are some public pensions in trouble? If this was so easy, many private employers would be offering this but instead they are running away as fast as they can. That should tell us something.
A fool and his money are good for business.
User avatar
nedsaid
Posts: 19275
Joined: Fri Nov 23, 2012 11:33 am

Re: Is there a problem with glidepath (target date funds) for retirement savers?

Post by nedsaid »

bobcat2 wrote:

Life annuities have long duration. Their prices and payouts are very sensitive to changes in interest rates. You can’t set aside a given portfolio value and think you have locked in a level of annuitized income. That is folly.

See the following thread for more on this issue.
viewtopic.php?f=10&t=174991
Bob, you raised an excellent point about matching the duration of a Single Premium Immediate Annuity you hope to buy in the future with the duration of the bonds you set aside to purchase that annuity at a later date. This is something that I honestly hadn't thought about and this is a very subtle but important part of a matching strategy.

I also saw that you estimated that an SPIA would have a 15 year duration if purchased around age 65. The bonds that I would buy now at age 56 to purchase that annuity in 9 years would need to have a 15 year duration. If at age 65, rates are higher than now, the loss of value in the bonds would be offset by the increased amount of income I could buy with the annuity. If rates were lower than today the increased value of the bonds would offset the decreased amount of income I could buy with the annuity. So I see what you are talking about.

You are right that just setting aside $600,000 would not necessarily enable me to buy the $40,000 a year in income. The amount I would need to buy the $40,000 of annual income at age 65 would vary with interest rates.

A problem with this is that it is a subtle part of the strategy and I can tell you that many financial advisors would not even have thought about this. Most insurance agents selling annuities would have no idea what you are talking about.
A fool and his money are good for business.
Post Reply