A reader asked for my opinion on an article by Paul Merriman called The Ultimate Buy And Hold Strategy. I don't know much about Mr. Merriman other than he is a much bigger fish than I am in the world of stock market writing on the internet.
First a philosophical note; the notions of being able to set it and forget it or having a portfolio with very few moving parts are both very appealing for a lot of reasons. However, markets, companies, economies, global dynamics, cycles and probably a few other things are all ever-changing. For example most clients own Johnson & Johnson (JNJ). Every time I buy it for a new client I am hopeful that I can keep it forever. Chances are I wil be able to but who knows what the future will bring.
As for the article, Merriman makes it clear that this is a long-term strategy. From 30,000 feet this is crucial. You probably save and invest so that you will have enough to live on when you are no longer working. Beating or lagging the market by five percentage points 15 years before you retire means absolutely nothing. He also makes a great case for top down management.
The article goes step by step in building the portfolio using 60% S&P 500 and 40% Government & Corporate Bond Index as a starting point to get to the finished product which maybe the ultimate but it is not very simple, as I thought it would be before I read the article. The finished product has 10 holdings.
I am not going to outsmart Merriman but there are a couple obvious things that come up as possible road blocks to this idea working in the future. As I said the starting point is 60% S&P 500 and 40% Government & Corporate Bond Index studied from 1970 to 2005 to set a benchmark of 10.4% annual return and an annual standard deviation of 11.6%. The Ultimate has an average annual return of 13.2% and an annual standard deviation of 11.6%.
The numbers include the total return of bonds. I have to wonder whether a huge bias is embedded by many years of double digit bond yields and as yields worked down slowly, increasing bond prices. 25 years or so ago, long term bonds yielded 15% and shorter term notes were pretty close to that. Now it is possible that the increase in bonds leading up to the late 1970's would offset the effect I am talking about to some degree but weren't yields on the decline from 1982 to 2003 almost uninterrupted and wasn't that decline from the mid teens down to 1% for short dated paper and 3.1% for ten year paper?
Whatever the exact numbers, this effect is in the data series studied, might create a skew and should perhaps not be included in a forward looking model depending on the probability you would assign to bond yields going from the teens down to 3% at some point in the next 35 years. If my point here is valid it could change a lot of the assumptions that Merriman relies on to get started building The Ultimate.
Speaking of forward looking I don't think it does (do I have this wrong?). As noted above, markets and all things that influence markets are dynamic. The world today is very different than it was, say, 15 years ago let alone in 1970 when the study starts and will be much different 15 or 35 years from now.
I would have difficulty in going with a plan if it does not look forward. If this one does somehow look forward and I am missing how, I would rethink this part of my comment.
I will say I trust Merriman's intention behind the concept and it could work but keep in mind all ideas like this have flaws. Maybe I have found a couple and maybe I have not, but again, all portfolios have flaws.
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
20 comments:
Great response to his philosophy. One thing that I learned when I went to see Paul a year ago was his leaning towards value because of historical returns back to 1927 (especially in small caps)
So where is the rest of the world in this model? Just using US large caps figuring they capture the interantional growth?
Roger, I agree with your assessment that it is backward looking. That does not mean it will not work, but does bring into questions statements like: "if you change this percentage from x to y you will get z percentage greater returns."
I also did not see anything on re-balancing. I have seen a number of weighting strategies. The more sophisticated ones place as much emphasis on how often you re-balance, as they do on the weighting. It is this re-weighting that captures the buy-low sell-high aspect of portfolio management. If apportionment is the risk reduction part of portfolio management, re-weighting is the return part.
Today the Wall Street Journal had an interesting piece based on a recently released report about where the world's asset values lie circa 2005 (pay only unfortunately):
http://online.wsj.com/article/SB116839213664272112.html?mod=todays_us_money_and_investing
It noted that the U.S. takes in about 85% of the flows from countries that are net exporters of capital. Based on the numbers in their chart, US assets are weighted at about 40% of world total. Given that the US is only 30% of the worlds economy, and still has a shrinking productivity advantage (which it has consistently done over time since 1870), that is an awfully high risk premium paid to own the rest of the worlds assets. Even portfolio 6's relatively high 31% ownership of foreign stocks appears to be rather backward looking. Particularly when considering that a very large percentage of the worlds economy was not even encompassed within the free market economy for over 2/3 of their bench market period. It is also interesting to note that non-US bonds do not appear to be in their portfolios.
I do agree with their concept of using low cost funds. I do not see where they indicate how much the cost of their services would reduce the "expected" returns.
Roger, thanks for posting your thoughts on this.
I've read a lot of articles by Merriman over the past decade and listen to his podcast on a weekly basis. That said, I'm sure Merriman would challenge you on your "looking forward" comment by presenting a pretty strong case against active management. He would tell you there's no way to predict what asset classes will outperform in the future - that looking at past performance over long periods and taking a balanced approach is probably the best anyone can do. Also note this strategy has outperformed the vast majority of active fund managers and the S&P 500 "benchmark" in real time since he began recommending it in 1983.
Your comments on past bond perfomance are very interesting. Two points here. 1) Merriman isn't necessarily recommending a 60/40 portfolio - he is just using it as an example in this article. Merriman actually recommends young people allocate zero to bonds, but add to bonds as they grow older to reduce porfolio volatility and risk. In his profession practice he evaluates an risk tolerance individually. 2) I wouldn't focus too much on the absolute return this backtested portfolio. I think all he wants to illustrate is the more bonds you hold, the lower your short term risk / standard deviation.
Data series: Granted, this article is based on a relatively short data series (36 years), but his thoughts on equities is largely based on past performance seen in the CRSP database from 1926. Again, I wouldn't focus on the absolute return of any backtested portfolio. The philosophy behind this portfolio seems pretty solid: diversify amongst many different equity asset classes - whereas the most advisors had traditionally recommended that the masses focus their holdings in large cap, U.S. stocks. His 50% U.S./ 50% Intl. is still a pretty radical concept to most.
One other thing (something not addressed in this article) that really convinced me create a similar buy-and-hold portfolio was the long term effect of using very low-cost vehicles. The long term effect of eliminating loads, commissions, and using vehicles with dirt-cheap expense ratios is simply stunning.
russell120
Merriman doesn't discuss his backtesting methodology in this article but I know he is a huge proponent of annual rebalancing.
Tom K,
That is avery thoughtful comments from you! From time to time one needs to be reminded that there is a difference in approaches between do-it-yourselfer and money manager. The latter emphasises value-added, keeping up the the Jones while the former basically can do anything he or she wants and is experimenting all the time. I do believe over 50% of investors can make use of a fixed portfolio such as Paul M or Stein suggested. Nobody can predict the future(forward looking). Look what happens to John Hussman and Bill Miller and the like(not to mention the blowup of many hedge funds). Investing is basically diversification and cash-flow management(position size and loss cotrol , for example. If you lose less in a down market you will do quite all right even your up years are not as hot as the market indices.
Keep in mind folks that Merriman's Ultimate Buy and Hold is just one idea - a strategy he thinks makes sense for a lot of do-it-yourselfers. He readily admits it isn't for everyone. Buy & hold only works if the investor is willing to ride out periods of significant drawdowns without panicking...and that's a pretty big IF.
The clearest long erm buy and hold strategy is to hold non-US equities. Given the pace of nominal GDP growth in the US versus the rest of the world, there is virtually zero chance that US equities will comprise 42% of global equity market cap, as is currently the case, in say 15 years time. US share of market cap is already down from the low 50's at the height of the 90's equity bubble, but there is plenty more juice in that "trade".
macro man,
I would definitely like to hear more opinions on what folks think the best U.S./Intl equity mix should be (long term). Nominal GDP growth, % of "World GDP", and % of global equity market cap seem to be important considerations. Are there others?
Also, it seems U.S. companies increasingly have large markets or operations outside of the U.S. How does this factor into the mix?
many good comments and questions, but while Merriman's buy and hold strategy probably needs future adjustments it is a better starting point than most people have. I also would agree that the future will probably reward foreign investments a lot.
More money is invested by institutions(pension funds, endowments) than individual individuals. Institutional investors, probably with the exception of Yale and Harvard are typically 60 equities/40 bonds. There is really nothing magic. For most people, this will be a good starting point for asset allocation. Increasing the international exposure to more reflect now only 40% is US is par for the course.
That might just be the most helpful article I have ever read about investing. Thanks
Tom K
In a perfect world where the ratio of market cap to nominal GDP were constant across countries, the answer would be easy. Obviously, that's not the world we live in.
The considerations that we need to consider include the following:
*Depth and transparency of capital markets; this is a significant issue for China, which is both large and growing quickly
* Nominal GDP growth rates; when investing for the long term, we want to consider how the global economic pie will be sliced 10-20 years from now, not just how it is today
* Home bias and trends thereof; virtually all investors, regardless of which country they live in, place a disproportionate amount of their financial assets in domestic markets. This is particularly true for retail investors, and particularly true in the United States. On the other hand, there is clear evidence that home bias is being reduced, not only in the US, where much more money flowed into international than domestic funds last year, but also in Japan, where the "Mrs. Watanabe's" are buying foreign equity and fixed income funds with seemingly reckless abandon (by Japanese standards)
The first of those issues would suggest keeping money in the US; the latter two, for keeping more money outside the US than is commonly done.
For me, on balance, the latter two considerations are more important when thinking of very long term horizons.
One major issue with "buy and hold" is that your entry point becomes pretty important. Personally, I would feel much more comfortable buying EM equities here and now for the long term (despite the run-up of the last few years) than, say, an EAFE type index, where a good deal of the market weight is allocated to regions (such as Europe) with a dramatically overvalued exchange rate vis-a-vis the dollar.
The same issue holds for fixed income as well, where relative valuations strongly argue, in my view, for an underweight bond/overweight equity mix on the standard 60/35/5 benchmark.
My $0.02 (or should that be €0.02 or £0.02?)
Macro man, thanks for your insight. There sure is a lot to think about. We didn't even touch on the odds of social or political upheaval in many regions of the world.
I might be stating the obvious but Merriman really promotes the DFA family of funds available through authorized financial advisors. His site has a lot of interesting articles. The Index Funds Advisors (IFA) site also has many variations on the buy and hold using DFA (primarily) but has comparisons of trying to do the same with other proxies. I also believe Merriman would be very open to adding asset classes as they become available and I guess relevant to him. He also talks of rebalancing. I don’t think his buy and hold should be interpreted as buy and stick your head in the sand. I think it is buy and stick with what has been historically solid, diversify, and don’t actively tinker. I think it is appropriate for some people.
> "don’t think his buy and hold should be interpreted as buy and stick your head in the sand. I think it is buy and stick with what has been historically solid, diversify, and don’t actively tinker."
I couldn't have said it better.
tomk, Is your diversied buy hold portfolio in a taxable account, while the tactical allocation is in nontaxable one? Nice to see such flexibility, whether they are tax specific or not. Bonds are the asset class that I need to accumulate, or I feel that I am supposed to. Every year I talk myself out of it. And, Roger just gave me the ammo.
anon 7:57, both are in tax-deferred, but obviously the buy and hold portfolio would make sense for a taxable account.
There's another advantage to dividing your investments between two or more strategies: Psychologically it helps you maintain discipline. When equities are on a tear like they have been for the past several months, the buy & hold portfolio is going to do extremely well. That counterbalances a strategy that uses market timing (market timers almost always underperform during bull markets). Conversely, when the market is taking a beating, you're more likely to be holding a lot of defensive sectors and cash in a tactical allocation portfolio. Psychologically this helps you to not tinkering with your buy & hold portfolio - which is likely to be taking a beating.
Strategy diversification is something I rarely see discussed. Most investors decide to use just one strategy, but rarely can they stick with it through subsequent market cycles. They jump from one strategy to the next every 2 to 3 years hoping that they just found the next best thing. Also, investors are overwhelmed with so many inputs they can't cope. A written strategy can help filter out the noise or inputs that don't meet the criteria of your strategy.
Merriman's strategy is, as he said in the article, not his own. It is based on the efficient market theory and the maximization of return for a given amount of risk (standard deviation). The idea is to create a portfolio based on the entire stock market and a risk-free asset (bonds). The "automatic" part of the strategy is periodic rebalancing based on past return and risk. It sounds like he may also throw in the fama-french 4 factor model for some good measure.
How do I know this? I studied under Fama and have met with DFA. This is taughted in any entry level PhD level finance book, or even an MBA book.
I am not slamming merriman, but I believe that he takes an inherently elegant and simple process and attempts to make it seem more complex than it is.
I really don't think Merriman's portfolio is complex. You can look at various flavors of the same concept at IFA: http://www.ifa.com/portfolios/
These porfolios (20) are just as "complex" as Merrimans. IFA and Mark Hebner use DFA funds which are largely based on the Fama and French research. Note how the IFA funds shift allocation based on age and risk tolerance. This is what Merriman does with his paying clients - The Ultimate is just a down-n-dirty version for do-it-yourselfers.
I'm not a student of finance but I've read about Fama and French and listened to a few interviews by both of them. The three factor model basicly adds size and book value/price to beta - two factors CAPM was missing.
From Wikipedia: "Many studies show that the majority of actively managed mutual funds underperform broad indexes based on three factors if classified properly. This leads to more and more index funds and ETFs being offered based on the three factor model."
I don't know how you can adequately capture both the "value" and size factors and manage their portfolio weights without owning at least 10 funds. Call it complex, but if you want a balanced diet you need to eat from all the food groups.
Post a Comment