That would be Jeremy Siegel and Zvi Bodie. Both have contributed much to the investing stream of consciousness in their time but they have much different views on how to get there from here. That is the focus of a paper Geoff Considine wrote for Advisor Perspective.
Siegel wrote Stocks For The Long Run, is a proponent of owning equities and it is his research that is the foundation for the WisdomTree dividend weighted ETFs. Bodie is most known (I think) for a paper from 1995 where he advocates putting almost everything into TIPS and a little into equities similar to what Taleb advocates (a reader made that point the last time I mentioned Bodie's work).
The Considine article works through both sides of the argument. Quite candidly I have a tough time comprehending Geoff's commentary so I'm not going to delve too deeply into Geoff's conclusions. Some of the main points though are that a lot of this can be a function of circumstance. In looking at 50 different possible outcomes over long periods of times in which either all stocks would be best, all bonds or any number of combinations of the two could be the best there is no way to know ahead of time what will happen during your time of investing. This bit of luck could go a long way to determining your outcome. Ouch.
The risk of all stocks is that equities drop at the worst possible time leaving you unable to make it back thus leaving you with not enough money. The risk of all bonds is that inflation reduces the real value of your holdings thus leaving you with not enough money. TIPS give a better chance for keeping up with inflation (by definition) but you kind of run in place as opposed to growing assets.
Because of the strange decade we have had for US equities combined with a massive decline in interest rates over the last 28 years there will be studies showing that bonds having done better than stocks and will conclude that they will continue to outperform stocks going forward, we have seen some of this already.
There is obviously no way to know what the next ten or 20 years will bring. Perhaps bonds will outperform but it makes sense to zoom back some and look at the bigger picture. Ten years ago today the S&P 500 closed at 1360 so at 954 today it is down 29% in ten years not including dividends. Bond yields continued lower all through the decade. The idea of favoring bonds after a horrible, but not unprecedented, decade for stocks and a fantastic 30 year run for bonds has the horse has already left the barn feel to it. This would seem to have buy high (the bonds) sell low (the equities) written all over it.
My take on this has been the same for a long time. At times it makes sense to go heavy versus your target in equities and at other times you want to be underweight the target. Often if you should be underweight equities at a particular time you should be overweight fixed income BUT NOT ALWAYS. Each needs to be assessed independently and in conjunction with each other with an active decision made. Obviously passive/index investors would view it differently but if you know that every few years the stock market might drop by 30% (average bear market decline) do you want to try to avoid some of that? If you agree with me on that then you need figure something out that has a reasonable basis for working however imperfectly and if you think this is impossible then you shouldn't do anything to protect yourself.
Sorry if that comes off as harsh but most of the academic work I have seen comes closer to all or none than the way the real world works. Bodie can't be right all the time and neither can Siegel. This can't be a shock to anyone can it? If things are changing then it becomes a situation that requires an innovative solution so either figure something out for yourself or hire someone who think can.
Proud Member Of
Random Roger
© 2009 | Gadget Theme Blogger Template
Original WPtheme By Blogger Template | Converted By Gadget Lover
| Distributed by eFree Blogger Help
Original WPtheme By Blogger Template | Converted By Gadget Lover
| Distributed by eFree Blogger Help
16 comments:
I've used Considine's product after reading Stein and DeMuth's book. It seemed intriguing that owning individual stocks with low volatility could outperform (remember the days when people bought individual stocks?). So I gave it it's day in court through marketocracy as I have with many portfolio theories. My theoretical portfolios are comprised of small cap momentum (think IBD type stocks), Ultimate Buy and Hold and what I call "Steady Eddies" based on the low beta individual holdings.
Long story short, the steady eddy portfolio is doing the worst. Now maybe I did something 'wrong' but to me, this theory isn't holding up very well in practice. I feel that the 'superior returns' of low beta individuals is because of the increased risk. It may not be so obvious in a bull, but in a bear it's quite apparent.
I think I pointed out recently that 'ultimate buy and hold' held up the best even though asset allocation has been declared dead (chortle).
I wonder if Wisdomtree is based on the same research as Siegel's book. And I wonder if Siegel is having any doubts about that.
I really wish we longer term performance data on TIPS. I feel like I know very little about how they'll perform.
Seems like TIPS would cost the U.S. Govt more than nominal bonds. Given the current low interest rate environment, why does the Treasury even sell them?
I think we need a new concept. Buy and hold does not work in periods like we have now.
Buying bonds after decades of out performance is crazy stupid as you already indicated.
This point in time is a market timing period IMO.
Some time in the future will be a buy and hold equities for the long run period again, but not yet.
There is no one model for all decades. You are searching for a holy grail that does not exist.
SEG
Valuation can't be ignored. Buy and hold stocks usually works but something like the 10-year PE needs to be considered. Stocks for the Long Run has been kicked around a bit recently, but it does include tons of objective information - including the outperformance of lower PE ratio stocks.
Note that Bodie's paper basically predates commodity index etf's and reit etf's.
I've read other arguments by Considine and it's usually pretty sensible. Many people probably miss out on his logic - he usually breaks out the "Monte Carlo simulation" a few paragraphs in, and continues as if it were an oracle of future results. The fact is that this uses historical data and random number generation to create a probablistic prediction of returns. It's useful and generally a good way to approach the problem, but it's not a time machine running to and from the future.
I remember reading a book by O'higgins about 10+ years ago who said buy TIPS and not stocks. He's the same guy who made dogs of dow a popular investment. Well, after 10 years since then, he was correct since TIPS out performed stocks. Pretty good call!
Sobering post, Roger. My portfolio has evolved into a mostly large-cap and mid-cap US and foreign ETFs and a few mostly US and foreign large-cap ETFs and a few large-cap US and foreign stocks portfolio with a hand full of low-cost Vanguard mutual funds, a low-cost Vanguard short-term bond fund, an international and a US small-cap ETFs, and my 401k (all US large, medium, and small-cap stocks). It was down 46% from the Oct 2007 high on 3/9/09 and has recovered to being down only 24% as of the close yesterday. Think I will move more to bonds if/when it gets back to down 10% or so.
A thought on yesterday's post about collectibles. I have, framed and hanging on the wall in my computer room, a black-and-white photo copy of an AT&T 1-share stock certificate dated May 10, 1996 that passed through my hands in 1996 from my father's estate. May 10 is my son's birthday. The certificate is displayed between a picture of my father (who died in 1996) and my son (who was born in 1984). My father worked for AT&T subsidiary Western Electric from circa-1950 thru circa-1976 or '78 when the Western Electric plant he worked at shut down (I remember he told me once it was difficult finding a job at age 51 or 53; but he found one at a new IBM plant). Like T's post yesterday, the certificate is worthless, but priceless to me. The guy at Kinko's refused to make a color copy but looked the other way while I made the black-and-white copy on their copy machine; guess he feared I might might to market a high-quality color copy.
Thanks for your blog. I read it daily and get a lot of good process thoughts and ideas from it.
Some of Bodie's work now discusses using TIPs, but buying call options on the market for equity exposure.
So you put 10% into the call option to control your equity exposure and invest the rest in TIPs.
Of course, call options get extremely pricey during high volatility, but the theory is pretty strong.
The one fear I have with TIPs isn't just that they haven't been around that long, but that the government can always fudge the CPI data. If you look at the changes made to the CPI over the past two decades, each move has depressed the inflation numbers below what they would have been before the changes. So, if the government faces some real costs due to the increase in CPI, then they will probably change the game. Not trying to be a conspiracy theorist here, but it is the truth that these numbers change to fit the needs of the government.
when your 75, the long run is over. Everyone seems to speak in long term like everyone has a lifetime. Good luck to those of you like me where fixed income has to be at least 75% of your portfolio.Where is a good website for the old farts?
Bottom line, it all depends on how you define "long-term".
I think the way Siegel often frames it is sheer nonsense. Human beings don't live to be 200 years old with 100 year investment time horizons.
On average, we are talking a 70-80 year lifespan with peak earnings years of maybe 35-55 so roughly most people have a 20-30 year accumulation phase so that is the relevant "long-term".
So what is typical 20-30 year return? Well, alot of variation, depending on your starting and ending dates. As always, dshort to the rescue with a great chart:
http://www.dshort.com/charts/SP-returns-roller-coaster-shifted.html?SP-Composite-20-year-real-returns-with-dividends-shifted
In my opinion, although starting valuation probably doesn't mean diddly squat over 1-year or maybe even 5-years, that chart seems to support that is the exclusive determinant of "long-term" returns.
If you buy stocks near P/E peaks (late 1920s, 1960s, late 1990s) subsequent 20-year returns suck. If you buy stocks near P/E troughs (1940s, 1970s, early 80s) then 20 year returns are great. March 2009 was likely a good buying opportunity for 20 years. The rally has probably put us into back into average
Isn't it time for a new kind of ETFs to become available that are based on the indexes that stay fully invested or go into cash based on 200dma or some such indicator. Seems like a lot of people see value in this and perhaps practice it in some form or another - often perhaps without discipline (referring to myself).
I would consider such a product if the criteria for transitioning between stocks and cash seemed reasonable.
Here is a site where you can back test TIPS and compare the performance to adding other types of assets to a portfolio.
http://www.riskcog.com/portfolio.jsp#5mlb5u0
Bill B, the Ultimate B&H was written up recently with comparison to an "optimally weighted" portfolio composed of the same assets classes:
http://www.riskcog.com/lazy_portfolio_comparison.jsp
Mike C
Great chart! Is this saying 100% fixed income may be best or can market timing work? I'm inclined at age 75 to be 100% in fixed income. Or only depends on the time frame.
http://tinyurl.com/kqhklt
Hi 6:51, as a general rule I view being 100% in any asset class as an approach that increases risk. Fixed income does have low price volatility, but the other types of risk faced by that class are very very real. There is interest rate risk if rates jump, re-investment risk if rates drop, negative real returns if inflation picks up, default risk in a credit crises etc.
If you had held the total bond market for the past 4 decades the compound return would have been 7.86% with the worst year being 1994 at -2.86%
With a diversified portfolio designed to hit that same compound return the historical worst year would have been +1.38% in 2008. (Portfolio Allocation: 49.8% TBILL , 20.9% LTGB , 12.2% ST Trsry , 12.2% Intl Value , 4.9% GOLD)
Or to put it another way if you are willing to accept losses on the order of -2.86% in a year then you could have earned a higher return with a more diversified portfolio e.g. 11% compound. (Portfolio Allocation: 30.1% GOLD , 18.0% SCV , 15.8% LTGB , 13.7% ST Trsry , 9.9% 5 Yr T , 5.2% EM , 4.3% TBILL , 3.0% Intl Value.)
Of course going forward the bond market or diversified portfolios may do better or worse than the have in the past.
Matthew,
I guess my thought was that a bond portfolio would provide more downside protection than a diversified portfolio or one with equities. My interest is not how much I can make but how little I have to loose and still earn more than inflation.
Sorry, I didn't mean to be opaque with my last post. My point was that a 100% bond portfolio does not provide the most downside risk in nominal or inflation-adjusted terms.
Post a Comment