[Editor's Note: This review is a guest post written by a regular WCI reader, Robert Kanterman, MD, a practicing interventional radiologist. This recently published book on momentum investing has been the subject of much discussion in the comments section of several posts as well as on internet forums. I had not read this book at the time the post was written and edited. We have no financial relationship. Bob also notes that he has no financial conflicts of interest, but would welcome some if anyone is interested.]
I am a 49 year old interventional radiologist in the St. Louis, MO area, and have been active in investing since I was a ninth grader, when I purchased Boeing stock with my Bar Mitzvah money. As a radiology resident, I was introduced to Morningstar and invested in Vanguard Wellesley, my first mutual fund purchase. By the mid-1990s, I became a community leader for the AOL “Sage Mutual Funds” site. I started indexing in the late 1990s and later had the opportunity to purchase DFA funds through an advisor that my practice contracted to oversee our retirement plan. I have always had an interest in the investing and personal finance space, and I credit the book, The Millionaire Next Door, published right after I finished training, for saving my financial life. (My wife may have another spin on that!)
I recently read (or devoured, is more accurate), Gary Antonacci’s Dual Momentum Investing. As a regular reader of the works of Bill Bernstein, Larry Swedroe, and the like, I approached the subject with a fair amount of skepticism. The word “momentum”, in the context of investing, takes me back to the go-go years of the late 90’s and painful losses. Global Crossing anyone?
Dual Momentum Investing – The Author's Story
The book begins with the author as a young employee of Smith Barney in the mid-1970s, first discovering the valuable attributes of indexing—low costs, diversification, and efficiency of the stock market. He promptly quit the firm, as he realized that he did not believe he could do well by his clients by trying to beat the market. By the 1990s, the author was questioning the Efficient Market Hypothesis, the foundational belief of indexing and Bogleheadism. [Although Bogle would argue the foundational belief is the Cost Matters Hypothesis-ed.] By this time, the “momentum anomaly” was just being discussed in the academic literature, and this sets the stage for the rest of the book.
Momentum Investing
Chapter 2 is entitled, “What Goes Up…Stays Up,” and discusses the historical underpinnings of momentum as a force of investing, “the tendency of investments to persist in their performance.” Momentum principles date back to the early 20th century and show up with increasing frequency in the academic finance literature, as well as in more mainstream resources like Value Line, Investors Business Daily, Dreyfus Mutual Funds, and even Fidelity’s first retail fund offerings in 1946, not to mention investing success stories documented in books spanning the 20th century. Commodity traders, early hedge fund managers, and even Nobel Laureates recognized the importance of momentum in predicting investment returns. In Chapter 3, the author discusses some of the contemporary concepts and controversies in Modern Portfolio Theory (MVO, CAPM, and the like), a deeper dive into some of the issues that have lead many of us to Bogleheadism and its variations.
Behavior is a Cause of Momentum
My favorite chapter may be Chapter 4, which delves into behavioral economics and how many of the principles thereof explain why momentum exists and works. In short, it is that “investors behave unexpectedly and irrationally in systematic and predictable ways.” In greater length, Antonacci describes anchoring and underreaction, confirmation bias, herding and overreaction, conservatism, and overconfidence. My favorite is the one that afflicts me the most, the disposition effect, holding on to losers too long and selling the winners early, to lock in gains. These behavioral shortcomings are pervasive, among individual investors, advisors, and professional money managers, costing the end user, the investor, dearly in the long run.
Alternative Strategies
The following chapters discusses various asset classes and strategies and why they result in the investor leaving money on the table. Antonacci says that bonds, “diworsification”, commodities, futures, hedge funds, private equity, active mutual funds, smart beta (our flavor du jour) and good old fashioned stock picking all have their shortcomings and limit the investor from reaching his/her return potential for various and sundry reasons. He also says that Factors like size and value may be overgrazed or artifacts of faulty investigation. So what’s left? Momentum.
Absolute and Relative Momentum
Absolute momentum is the process of researching the returns for an asset class over a look back period and comparing it to the risk-free return (treasury bills). If it is positive, there is absolute momentum, otherwise known as trend following. Relative strength momentum compares one asset with another, while absolute momentum compares an asset with itself longitudinally. Peer reviewed papers on absolute momentum profits show that it works for dozens of assets and classes over time, consistently profitable back to 1903.
Dual Momentum
And finally we get to the strategy that is the basis of the book: dual momentum, combining the positive attributes of absolute and relative momentum. The Global Equities Momentum Strategy, which [using backtested data-ed] outperformed the the S&P 500 (or your index of choice) for the last 40 years with less than half of the maximum drawdown, is finally revealed. I feel like it took 100 years to get to this point in the story—it did. Briefly, the strategy is as follows: Every month, look back at the 1 year returns of the S&P 500 and the ACWI ex-US index. If either is higher than the 1 year return on T-bills, buy (or hold) the one that is higher of the two. If not, buy (or hold) the aggregate bond index. Your position will be 100% either SPY, VEU, or AGG.
There are variations that can be used to either juice returns or reduce drawdowns. Step-by-step instructions are provided. I hesitated to reveal the strategy because if you do not read the background information, there is no way that you would or should consider employing it, and it will be of little value. The strategy is so easy to implement that it can be confidently applied to other asset classes and easily tailored to one’s native or ingrained investing styles and tilts.
Should you Read Dual Momentum Investing?
The book is extremely well-written, funny at times, incorporating anecdotes and pithy quotes to make points. I can tell by reading the book that the author would be enjoyable company at a dinner party or on a morning run. I have heard him interviewed in podcasts, and he speaks well, too. Even if you have no inkling to deploy the strategy, if you manage your own assets, as a reader you will get an overview of other strategies that you likely employ, their historical basis and academic foundations, their strengths and weaknesses and the behavioral tendencies that tend to thwart the individual investor, again and again. It is a quick, if not inexpensive, weekend read for the intermediate level (and above) self-directed investor, and it has certainly influenced my personal strategy going forward.
Have you read the book? What did you think? Do you incorporate momentum into your investing strategy? If so, how? Do you think it is worth including on the WCI recommended reading list? Comment below!
This is just begging for a pro/con series…
Dont we all kind of use momentum investing by not re balancing?
Also selling every month, would mean short term capital gains tax?
In the back-tested strategy, there were very few transactions (around one per year or so) and most gains were long term.
What’s the best place to find the “1 year return on T-bills” actual percentage?
Can’t get a consistent value searching.
Thanks!
I use the BIL ETF as a proxy. Because of the low interest rate environment, it has been a flat line for some time.
http://performance.morningstar.com/funds/etf/total-returns.action?t=BIL®ion=USA&culture=en_US
The current yield on the 1 year T bill is 0.27%.
http://www.treasury.gov/resource-center/data-chart-center/interest-rates/Pages/TextView.aspx?data=yield
If you want historical returns, try this site: http://pages.stern.nyu.edu/~adamodar/New_Home_Page/datafile/histretSP.html
Larry Swedroe
http://www.etf.com/sections/index-investor-corner/swedroe-look-beyond-price-momentum?utm_source=newsletter&utm_medium=email&utm_campaign=weeklynewsletter
thought that was interesting
“Controlling for past performance when constructing earnings momentum strategies reduces their volatilities. It also eliminates the crashes strongly associated with momentum of all types, without reducing the strategies’ high average returns.”
But seems like a lot of work
Its really not if you disregard volatility and the like (makes no difference in most variations). A simple version is this: Pick 3 or 4 etfs with some diversity, 1 preferably as strongly negatively correlated to the others (eg, 3 stock, 1 bond instrument).
Check 3 month performances once at end of month (whenever), and choose the best one and hold for that month. Do the same thing next month, repeat. Often you go long strectches in the same thing.
but wouldnt you get short term capital gains if you are selling every 6 months?
Definitely if you’re doing this in a taxable account.
Definitely and not recommended for anything but a deferred account due to that.
Nice review Robert, thanks for posting it. This book, if nothing else, is the most comprehensive discussion on the momentum anomaly that exists in markets. It’s very educational for those looking to understand momentum. It’s in my top few books that I recommend to others, but then again, I’m biased since I run a trend following strategy for clients.
The problem I think with the book, and almost all discussions surrounding momentum, trend following, etc.. is that those terms give off a connotation of hocus pocus chart reading and prediction. When in reality, these systems are reactive, and only use very simple charting to determine which asset or group of assets is acting “best” at a given time. From there, the real key to success is the risk management, also known as position management—ie cutting losers and riding winners.
That’s the beauty of the dual momentum system: it ensures you’re in stocks during bull markets and bonds during recessions. Yes, there’s a lag of a couple months (since the system is reactive, not predictive), and you’ll get whipsawed once in a while. But let’s be honest, average bull markets and recessions last way more than 2 months, and these systems just make sure you capture the meat of these huge trends without being on the wrong side of them which is where the most pain to a portfolio often occurs.
At the end of the day, these strategies are actually extremely simple, but not easy. It all comes down to psychological discipline. Cutting losers and riding winners is as easy as eating well and exercising, but still most people are fat with poorly performing portfolios. Reading books like this will help a select few find amazing results, but for the majority, it won’t really help much unfortunately.
Here’s the link the system’s performance, and it is updated each month. Again, it’s basically just equity index returns most of the time, except during recessions when it switches to bonds. Pay particular attention to how it does during recessions: ’74, ’00-’02, ’08. Simply sidestepping the majority of each recession increases the portfolio CAGR from 7-8%/yr to 16-18%/yr. http://www.optimalmomentum.com/trackrecord3.html
Of course, the only data you can REALLY rely on is the data going forward from publication. Backtested data is just that. Docs are well aware of the limitations of retrospective studies.
Good point. Retroactive studies showing the sun rising in the East may not be enough for some to make a bet it’ll rise again in the East tomorrow, if they don’t know WHY it rises in the East. No amount of backtesting will suffice for them. Yes, there is plenty of backtesting on the momentum return premium, showing it is evident in 212 years of US equity data, dating back to the Victorian age in UK equity data, in over 20 years of out-of sample evidence, in 40 countries, and in more than a dozen other asset classes. But it won’t be until you understand the WHY it works that you’ll be able to use it confidently and successfully. I bet the sun will rise in the East tomorrow, and that rising market prices will attract buyers, and falling prices will attract sellers. Always has, always will.
I’m not saying backtested data is useless, but it’s important to understand it’s limitations.
Very true. Extremely important to know what goes into the strategy and testing in order to understand how robust and valid backtests are. Out of sample tests, monte carlo simulations, etc…all can help too. Often times the best investment/trading strategies tend to have so few parameters that the entire system can be written on the back of a 3×5 card. They’re not curve-fitted. As such, they’ll certainly bend, but should never break, and remain alive through all economic environments.
Seems like it’s a little too good to be true. A graduating resident who prioritizes saving and puts aside 100k a year from 30-65 years old with 16-18% return would have a rather comfortable 200-300 million set aside for retirement. That’s an eye popping number that I have some serious doubts about. I’ll buy the book and give it a read though.
Everyone keep promising me millions but I dont see them
As the reviewer states, momentum is one of two well known “anomalies” in the fama/french efficient market hypothesis. It does work, consistently over long periods of time. Though I just read an article talking about how more and more institutions are climbing aboard so it may fall victim to itself.
I dont know about dual momentum, as it appears overly/unnecessarily complex. You can really make it as complex as you want, but truthfully a lot of people have done work in the area and even a simple etf paired switch between treasury proxy (say TLT)and index of your choosing (spy, qqq) really trounces the market. Its easy, requires minimal work and the formula does the timing. I know its sounds fishy, as it did to me, but since investigating it further I am employing a portion of the portfolio to it myself, its robust.
You can increase return/risk, decrease draw downs and volatility. For those that want to look into it the most common look back period is 3 months (what instrument performed best over that time, longer you have bigger misses and shorter you get whipsawed) and the holding period is monthly. You just assess and change (maybe, turns out you hold quite often) once a month at most and takes 30s. You can make it more complicated, but even a simple pair switching like the above would have moved you into treasuries in July of 2008 and you would have had a banner year. It doesnt do any better than follow the bull market at best, but shines in missing giant draw downs.
Worth a look. I do something more fun than simple paired switching, but thats a very indexer friendly way to do it.
I don’t think institutions climbing aboard will negate the alpha in momentum strategies like it would in arbitrage strategies. If anything, it’ll accentuate the trends, increasing vol, and blow more bubbles, creating even more momentum opportunities. The idea of cutting (or not being “in”) losers and riding winners will always work so long as there are human participants acting irrationally in the marketplace taking the other sides of those trades due to cognitive biases and heuristics.
The day I’ll get worried about the success of these strategies is when every single market participant acts perfectly rational and independently of each other. Only then would markets be perfectly efficient. But even then major market trends would still exist due to fundamental underlying economic trends…
I agree to an extent, and hope you’re right. Institutions that currently have momentum funds are usually not quite as good as an individual can be anyway. They usually have too large a universe of funds and try to make it more complex. Its probably hard to sell your service when someone can just look at it and replicate it easily, have to make it appear complex which seems to lead to under performance.
And, to be fair you can add volatility and hedges to the basics and do well, so its also ok to use a fund if you feel you need it. Thats just way beyond the scope here and really for most people.
I love the cover. “A proven way to consistently beat the market”. Mmm… I think I have heard that one before. I pass.
if its so successful why haven’t active investors used it to outperform the indices
Via email:
I have read the book and find it very compelling. Miles Dividend MD has a number of posts on his blog (he’s a practicing cardiologists) about dual momentum, the most recent post discusses some of the risks of the strategy.
http://www.milesdividendmd.com/looking-under-rocks/
I haven’t yet adopted dual momentum, but am strongly considering doing it with a portion of a tax protected account. I don’t want the added cost of capital gains taxes, even though there is only about one trade a year. I don’t plan to change the asset allocation of the rest of my portfolio – just use a portion of my bond allocation for dual momentum. That way in normal times that portion should be in equities and in times of financial stress should be back in bonds.
I’d recommend reading the book. It’s difficult to get a reasonable understanding of the strategy, and the simplicity of it, without doing so.
I think the biggest risk of dual momentum is the whipsaw, and the biggest risk of buy and hold is the bear market. Perhaps by having some of each is a way of diversifying these two risks.
I like the idea of diversifying those risks.
There are some excellent discussions on momentum on seeking alpha. I dont know about the need for the dual portion or its benefit over standard momentum…but theyre likely very very similar.
When I started (about 10% in advantaged account) I used etfreplay to familiarize myself with the fund universe, and play with backtesting and the different results on cagr, sharpe, maxdd, and volatility to match your personality. Im working on a spreadsheet version now instead. Helps to see things and what seemingly minor differences make to the strategy.
The biggest danger is curve fitting and forgetting the limitations of backtesting.
I haven’t noticed many whipsaws with dual momentum. The biggest risk to me would seem to be not sticking with the method.
what about the tax consequences
why own taxable bonds in a taxable account
obviously the theory would work best in ret accounts
You could use a muni fund, but I agree, far smarter to do a strategy like this in retirement accounts where the transaction costs are much lower.
WCI, do you have an article on momentum? what do you think about it? And how easy it is to do it?
I don’t think it is particularly hard to do. I’m cautiously optimistic about it. I certainly wouldn’t do it with my entire portfolio but I don’t think someone who wanted to do it with < 1/3 of their portfolio inside a retirement account using very low cost index funds/ETFs is necessarily stupid. I'm not currently employing a momentum strategy in my portfolio at this time.
isn’t it market timing which is a NO NO!!!!!!!!!!!!!
No its not. It does sound like that if you havent looked into it though and its an understandable but incorrect takeaway.
I believe in diversification of strategies in addition to diversification of assets. I have layered a dual momentum (DM) strategy on my traditional buy-hold-rebalance 60:40 allocation, within my self-directed retirement plan as follows. I took 5% out of the equity and 5% out of the FI allocation, for a total of 10%. So my allocation is now 55:35:10 (stocks: bonds: DM). Currently, the DM strategy has me in the S&P 500, so my real allocation is 65:35 (stocks: bonds). At some point, likely with the next lasting equity drawdown, the 10% will swing to bonds, so the allocation will be 55:45.
I have DM playing as a bit players in a couple other corners of my portfolio, but this is the primary role for me now.
In all, this will likely not have a major impact on the return on my portfolio other than modestly smoothing returns over time. If anyone knows how to back test this strategy combination, I would be interested to know how (or now the results of such). I am expecting the additional behavioral benefit of being able to “do something” in the next bear market, tempering the need to make an even more disastrous change.
Have you picked a date, like last day of the month, or last day of the third month, to decide which one are you going to put your money in?
Out of your 10% in DM, what actual ETF did you actually buy? What actual ETFs did you consider? Do you do like an annualized return for that period to decide if you continue or switch?
I am checking this perfchart at the first of the month (or first trading day):
http://stockcharts.com/freecharts/perf.php?SPY,AGG,VEU,BIL
The default is to 10 months, so you have to manually change the look back to one year (252 or 253 days) if you want to use Antonacci’s technique.
My accounts are at Fidelity where ishares ETFs trade for free, so I am using IVV, IXUS, and AGG in my portfolio.
so whichever one of those three ETFs has a higher return in last 12 months, you move your money to that?
Few questions: Have you actually moved your money or since SPY is doing so well, you just have been in one category?
Why look at a full year return rather than quarter or 6 months or 2 years?
I do wish WCI would do a post on momentum.
If you are considering employing the strategy, I strongly suggest that you read the book. It answers all of your questions and more.
Did find WCI response “remember the dangers of trend following. You will be out of the market at market bottoms, so you’ll miss the first rise off the bottom, which is often dramatic. You will also get the first drop off the top, which is also often dramatic. You also actually have to pay attention to your investments and current market performance, which does take some time which is valuable. I often don’t look at my investments for several months and basically only look at market performance once a month (when I write my newsletter.) The less closely you watch with a trend following system, the more of the drop from the top you eat and the more of the rise from the bottom you miss. You also have to beware of being whipsawed. In a whipsawing market, you’re basically buying high and selling low repeatedly. Backtested trend following data is designed to minimize the whipsawing, because if the whipsawing + expenses cost more than you’re losing with downturns doing simple buy and hold, the trend following comes out behind. However, you never know how well the backtested technique is going to do against the whipsawing going forward. Perhaps the whipsawing will become more violent as more and more people trend follow. “
Although this appears and sounds superficially like trend following…its really more like an algorithmic system that just gives you a signal.
Robert, I’m curious as to why you have AGG on this chart. As understand it, you compare the two equity funds to t-bills (BIL, as you have done), then if both are less than BIL, you switch to bonds. So no need to include AGG in your chart.
Just for interest. My master chart includes EEM, VNQ, and IJS, too.
I also use stockcharts to check the strategy, but have set it up to link directly with one click:
http://stockcharts.com/freecharts/perf.php?SPY,AGG,VEU,BIL&p=5&O=111000&B=BIL
This sets the lookback to 12 months, and sets the baseline to BIL’s return, so anything positive has absolute momentum and anything negative doesn’t.
Right now I’m paper trading this strategy to see how much faith I can put into it, but so far all my criticisms have been answered. Many by the author, who is extremely responsive to his readers!
With etfreplay.com you should be able to do a simple backtest without a subscription. You can also play with lookback periods and volatility (not worth it). I think you’ll find a 3 month look back will get you out sooner without such a whipsaw effect. I can try to link an example, but I have a subscription and am not sure it will work.
Dr. Khan, choosing different etfs will have a slight difference in outcome but the basic is to have at least 2 with negative correlation so that they smooth each other out, a stock index and a bond index for example. You can get more complicated of course, but it can actually reduce results and has a danger of curve fitting.
Here is a link to a very simple SPY/TLT paired switch from 2007-2015. I used TLT because the EDV (i use this one myself) was not available until 2008. Its important to note these strategies shine not by crushing the market during a bull, but avoiding huge losses in a bear.
http://www.etfreplay.com/members/backtest_rstactical.aspx
Thank you for the link. Without a paid subscription, I am not permitted to see the results.
One test that I would be really, really interested to see, would be to use the DM strategy but instead of using SPY and VEU (or VEA), using IJS, VNQ and EEM. It would seem to me that it would be optimal to capture the upside of the upside of small value, real estate, and emerging markets in a TM strategy, and sitting in cash or short term bonds when there was no upside momentum.
Of course, if there were indices that preceded these indices further back in time, it would be even more useful.
I will try to find another way to show it. Youre kind of on the right track for what I do. However, it seems sectors can be a bit volatile and dont necessarily work better than larger more broad funds. My current “normal” rotation strategy is a global one and the funds I have chosen (through lots of backtesting and risk of curve fitting) are:
QQQ-Nasdaq us stocks
EDV-bonds, negative correlation to stocks for most part
VWO-emerging markets
VGK-Europe (what I’ve been in for last two weeks and this months pick)
VT-global stocks
FM-frontier markets (just added as I feel they may have their day soon)
ZIV-inverse volatility
As stated, look back 3 months, i hold/reeval 2 wks. Tech specs are 70% on performance and 30% on volatility (a negative factor).
After a long learning/testing phase I’ve just barely started so theres not enough data to say anything one way or the other besides the premise is compelling and robust.
This tool:
https://www.portfoliovisualizer.com/test-market-timing-model#analysisResults
gets me pretty close. There are some fun models one can create and test.
I am working on what I call my “Quad Momentum” model (you heard it here, first, folks)–using SPY, EEM, EFA, and IYR (US stocks, emerging markets, developed international, and US real estate–for the ticker challenged. 😉 )
Thats a great mix but you’re missing one piece. While those are geo/asset diverse they are all equity based and thus are likely to be at the very minimum at least moderately positively correlated. That is, when spy goes down/up they will as well to a greater/lesser degree which somewhat limits the downside benefit of such a momentum/asset rotation strategy.
Negatively correlated instruments tend to perform contrary (not always of course) to the others and give you something to rotate into when those arent doing well. This would be something like TLT, EDV or your fav bond fund. Add one of those to your fund universe and see if it doesnt decrease volatility and max draw downs while simultaneously increase return.
Robert, according to Gary Antonacci is his blog post, this strategy works best with large caps.
http://www.dualmomentum.net/2015/01/and-winner-is.html
Will be see a mutual fund that employs this strategy
There are several now and growing daily. Most suffer from a too large universe of funds and weird trading rules. Some do well.
Which funds use this strategy?
MMTM
http://www.morningstar.com/etfs/ARCX/MMTM/quote.html
Actually, the existing momentum ETFs (MMTM, MTUM, MOM, etc.) use a relative momentum strategy. There is no absolute (trend following) momentum product that replicates the GEM strategy described in my review. The fact is that it is easy enough to do one’s self.
using just the three etfs US Stock World Stock and Bond could keep it simple
Using many other asset classes with this strategy is too complex and time consuming
Ordered the book out of curiosity
Have any of the BIG NAMES in finance endorsed this strategy
I’m happy to, on the first of each month moving forward, comment here on what the system would do (ie stay in position or change), and post the updated performance link. That way whoever is subscribed to this post can get the update.
I have run this system in all my retirement accounts and will for the next 40+ years. As long as this website stays running (and I stay alive), I’ll keep updating it.
Today, on May 1st, the system would still be 100% allocated to the S&P500. I’ll check back in on June 1st!
Why not just post a link to their website where they do this each month?
There’s lag on his timing of reporting, and I doubt he updates on the 1st of each month like the system dictates: http://www.optimalmomentum.com/trackrecord3.html
Also, he doesn’t, to my knowledge, actively give a “stay” or “switch” signal. People need to run the charts themselves to generate the signals.
Personally, since I’m running it in my own retirement accounts, and checking anyways on the 1st of each month, I don’t mind spending another 30 seconds updating it here. Plus, writing it here, will help me stick to it for the next 40+ years, which is the most important part.
You have a lot of faith in my ability to stick with this website for 40 years.
Donald, I don’t think you have to use the first day of the month. You could pick any day, say the day of your birthday, so long as you stick with that day each month for when you look back.
Interestingly, if you use a 6 month look back period, you’d be switching to international developed today, but if using a 12 month look back, you’d be staying in US equities. I live in Canada, so that’s in Canadian $, so the currency difference may make that not so in US $.
Today, June 1st, the Dual Momentum System (per exact parameters outlined in book) would remain 100% allocated to the S&P500. I’ll report again on July 1st.
Performance: http://www.optimalmomentum.com/trackrecord3.html
Today, July 1st, the Dual Momentum System (per exact parameters outlined in book) would remain 100% allocated to the S&P500. I’ll report again on August 1st.
Performance: http://www.optimalmomentum.com/trackrecord3.html
For those of you still keeping track at home, for Sept 1st I have the allocation switching from SPY to AGG. Correct?
According to Morningstar and my perf chart, the SPY (and IVV) are still above zero and above BIL, over the last 365 days, by a hair, so I am still in IVV. Are you looking at the 200 day numbers?
I was looking at the 252 day chart on Stock Charts. I see what you mean when you add in BIL. Here’s a link comparing SPY, VEU, AGG and BIL (this defaults back to 200 but just change it to 252).
http://stockcharts.com/freecharts/perf.php?SPY,AGG,VEU,BIL
I’m a bit confused as to exactly which T-bills Antonacci is comparing SPY or VEU to. The 1-3 month BIL etf? 1 year? 5 years? 10 years?
And if you’re ultimately going to select from either SPY, VEU or AGG, why not just use those 3 for the sake of simplicity?
If you read his book, you’ll see that Antonacci uses short term TBills because absolute momentum looks at the excess return of your risky asset by subtracting (or comparing it) to the risk free rate (TBills). If you don’t do it this way, you are looking at relative momentum not dual momentum.
It’s looking likely to be a switch from SPY to AGG on October 1.
Havent we all learned that being out of the mkt on just a few days every yr truly kills your end result
If this method is a no brainer you would think there would be mutual funds expounding it as well as academians promoting it
Its really the first time I have heard of it and will be reading the book for enlightenment
For most being 100% in stocks is beyond their risk tolerance
Trying it with a small % of your ret plan MIGHT be worthwhile but we all know past performance is not a guarantee of future performance
I think that you are correct, and I do not think that we should ever consider other ideas or concepts–ever–because what we know to be true and correct is never shown to be faultless or improvable. 😉
That saying is of course in a very specific context. The 10 best/worst days tend to be clustered together very closely, something called volatility clustering. Volatility rises in bear markets do to behavioral and psychological phenomenon. They also tend to occur after the market has already been going down in a bear market. So if you’ve already ridden all the way down, for certain do not sell.
However, if one were to miss the whole week where the best days occurred, your returns go up, and if you miss the 10 worst days your returns are phenomenal. Google “where the black swans hide” for a great analytic short paper on it.
In the systems described, you would have been out of the market and in bonds months, months, before the worst and best days, and would have been back in shortly after the uptick which was still way down from where you sold. You will not exit at the exact highs and enter at the perfect lows, but you miss the worst of the bear market, while catching the majority of the bull.
Being 100% in stocks was beyond my risk tolerance until I read this book. If you can eliminate most bear market exposure, then 100% in equities offers the best long-run return without excessive discomfort.
I guess the question is whether you feel you really can eliminate most bear market exposure. Be sure to read up on “gapping down” before you assume you’ll be able to sell in a bear market. For example, you may plan to sell if the market drops 5% and run your backtests based on you getting a price 5% below the high, but in reality, you may find you cannot actually sell at the price and the only price you can sell at is 20% down. Combine those issues at the top with similar issues at the bottom and you realize that there is more to this market timing stuff than most back tests suppose.
Look at this chart from 1987, for instance. You would have sold low and bought high. There are plenty of times when this sort of thing doesn’t work out well, and if there are more of those in the future than there were in the past, you’re not going to find that you’ve “eliminated most bear market exposure.”
I have no idea what the future holds, but I’m not sure I would assume that using technical indicators is going to allow you to “eliminate most bear market exposure” or avoid “excessive discomfort.”
I was referring to the long term methodology in the DM book rather than a stop loss method like you seem to be describing. 1987 was an extremely rare flash crash situation rather than a real bear market. It was the only flash crash that I could find looking over charts back to the 1920s. Yes, the DM method had to buy back in with the market 10% higher than where it got out then. But this was the only significant whipsaw I know of, and it pales in comparison to the equity erosion it would have prevented in the 1929, 1967, 1974, 1990, 2000, and 2007 bear markets.
You makes your bets and you takes your chances.
even if it has worked using past data, HOW many here would like to be 100% in equities nearing or in RETIREMENT!!!!
When WCI head hancho jumps into it I would like him to tell us
I then would do a small allocation to test the hypothesis
I dont think anybody would or should be comfortable with that allocation. This may just be a portion of your strategy to see how it fares or if it smooths out the portfolio. I have 10% currently in this type of strategy, not too much to feel queasy but enough to see how it goes.
That’s really not all that different from Bogle’s Tactical Asset Allocation tactics. He basically said it was okay to vary your stock/bond allocation by 10% or so. Now, whether you do that based on valuations or on momentum or just don’t do it at all is up to you.
Not doing it yet. I see a new investing strategy every month that backtests well. If I changed my strategy once a month I would never be successful. The most important parts of investing are coming up with a reasonable investing plan and STICKING WITH IT! Stay the course! Endure to the end! etc. I’ve got a rather high bar for changing my investing strategy because my strategy # 1 has worked for me, # 2 has worked for millions of others, # 3 has a sound theoretical basis, # 4 backtests well (what do you think every new strategy is tested against- buy and hold, no?), # 5 has very low costs, # 6 is simple to implement, # 7 is easy to avoid behavioral errors etc.
That said, I certain’t can’t prophesy that I’ll come out ahead of a Dual Momentumer.
“If something sounds too good to be true, then turn around 180 degrees and run! Because probably it is” … A wise man said.
used to be a theory about the DOGS OF THE DOW
hows that working today
or a theory where you buy/sell in May or whenever and do the same later in the year
Lots of crazy theories trying to beat the mkts
ONLY BUFFETT has done it over a very long time but he will not going forward(his words)
I’m an investment advisor and I’ve seen a whole lot of accounts and different approaches (including the Boglehead style) which works better or worse depending on how skillful the person is at operating it.
Momentum investing actually works really well for investors who know the research and do it in a disciplined way. It actually works a little better than typical buy and hold investing (of the same quality asset) because it benefits from the mob like mentality of the market at various inflection points and if you are disciplined you can use it too.
With that said, most people don’t do it because it doesn’t fit with their subjective “style” of investing. Some people are growth investors, some are value investors, some are single style (i.e. tech or oil) investors, some are mutual fund investors and some are ETF investors. Investing is like Buddhism — there are higher levels of thinking and awareness. Most people start out in a single style but those who are willing to push themselves and grow actually have the most productive portfolios. You may be skeptical of a various approach but really good investors (i.e. not Warren Buffett just those who do better than their peers) keep open minds and grow in their investing thinking. Whatever you choose to do I hope you maintain an open mind in your thinking and continue to grow your investment skill set because those investors get results that compound to portfolios much larger than their peers who have more rigid fixed mentality. This doesn’t mean you should change your style it just means that you should be open to it if you can find something that is predictably better.
I agree and very well-stated. When I first started investing seriously, I found active mutual manangers with excellent performance records. I studied more, and this led me to investing in index funds. More research led me to greater asset diversification, small and value tilting, cutting tail risk, etc. Additional reading has brought me to momentum. Along the way, these elements have been incorporated into my investment strategy.
The thirst for greater knowledge and portfolio optimization never ends.
There are lots of people doing well and beating the market, verifiable so through certain newsletter watchdogs. I dont know why people misunderstand what “the market” is and that not everyone is trying to beat it or should even reasonably be compared to it (ie, if you have a 60/40 mix you shouldnt feel bad if you dont beat the market in bull runs, its not designed to). Another is hedge funds, people completely misunderstand the usual meaning and reasons for them….hint, its in the name, hedging. Most of those folks dont care about crazy upside they are hedging downside risk for capital preservation. Whether or not thats a rational viewpoint is another thing. Its unreasonable to compare the two as if they have the same design and intentions however.
While “lots” of people beat the market, the percentage is dismal, especially when the long-term track record is examined. Consider this article from the WSJ:
It has always been difficult for investors to consistently beat index funds. It has been nearly impossible lately.
And there’s a double whammy: The small number of advisers who outperform the market rarely can keep doing so.
One big culprit, experts say: the rise of sophisticated computer-trading programs.Consider the 51 advisers out of more than 200 on the Hulbert Financial Digest’s list who beat the market in the decade-long period that ended April 30, 2012, as measured by the Wilshire 5000 Total Market index, including reinvested dividends.
Of that group, just 11—or 22%—have outperformed the overall market since then. That’s no better than the percentage that applies to all advisers, regardless of past performance. Over the past year, on average, the group has lagged the Wilshire index by 6.2 percentage points.
Google “Man vs. Machine: The Great Stock Showdown” to read the entire article.
Hey Jim,
That is a good point. Beating is the market is hard. But I don’t think that is actually the goal.
A better way to look at it is to think about investment returns is in a distribution. In today’s environment is you are earning about 1/2 of the return on the S&P 500 you are around the 50th percentile. So if you take the last three years, if your portfolio is earning about 9.4% per annum you are about normal (the S&P earned 18.85% per annum compounded the last three years). If you do a little better than that on your long term growth assets you might be making some progress on your investments as well as you savings and earnings.
There are a lot of good reasons to earn 9.4% or less on your money. They can be that you need your money soon, you’re personally highly emotionally sensitive to volatility ect. but if that isn’t your situation you do have to ask yourself is 9.4% a good return on my long term assets in an environment where a stock index fund earns 18%? If I did the stuff that took me to 10% or 11% could I stomach it and is it worth it? We all talk about how important it is to save on fees because that small amounted compounded grows to a huge number. Those fee differences are a mere fraction of 1 percent each year (but, like anything, a big number when compounded over thirty years), imagine if the differences in approach were worth 2%, 3% or 4% a year — that compounds to a huge difference as I know you know from your prolific examples of the power of compound interest. Most of the lower returns received by investors are due to their asset allocations, in my view, and the mix of assets they pick even for long term investments basically bakes in a very average return on their money. You have to ask — if you had more of a growth orientation in your asset allocation and it added 2-4% a year compounded to your money is that worth doing?
Now the knee jerk response you’ll hear from people is that anything that is more productive involves more risk. If they mean short term volatility they are right — if you needed this money soon higher earning assets are volatile and can fluctuate without reason and you probably shouldn’t be holding estimated tax money in the SPY or the EEM. But for retirement money – that has thirty years before you’ll touch it — making the argument that that is too risky for a primarily growth allocation seems kind of silly. It raises the point that maybe people with productive assets aren’t taking on too much risk – maybe those with unproductive assets are taking on too little.
Which brings me back to the original point — if you are earning 9.4% on your investments compounded over the last three years are you really making much investment progress? Now you know people are making financial progress due to their high savings rate, high income ect. but its possible that if you 9% on your money and the average investor earns 9% on their money that you are really kind of standing in place from an investment point of view? Its sort of like getting a 2% raise when there was exactly 2% inflation that year — you don’t have any more purchasing power — you are just in the same place you were a year ago.
Overall, this is what I’ve found most interesting about the indexing approach. Indexers have a similar looking distribution as mutual fund investors, direct stock investors, ect. The tool is clearly better than the typical mutual fund – but why isn’t the distribution better? I mean these are really smart people right — the 20% of the population who figured out the merits of indexing before everyone else did?
And then I came to this conclusion — maybe the indexers are rich because they are frugal not because they are exceptional investors. Frugality alone can almost make you rich. Frugality combined with a high income will definitely make you rich.
Therefore, they probably didn’t become rich by indexing but because they were just frugal people who saved a lot and applied that same kind of cheapness to investing where they became indexers because it is one of the lowest priced options.
I’m personally convinced that many of these people would be better off from an investment point of view with a good financial advisor (even investing purely index funds) but maybe their cheap nature keeps them from wanting to pay for it — which is interesting because the richer you get the more likely you are to pay for a financial advisor (40% or people with $250k pay for one whereas 70% of those with $10 million pay for one). This group of people are double outliers — they both invest largely in asset that isn’t quite dominant yet and I suspect they generally don’t pay for much advice while most people at their level financially do. And it is especially ironic because mathematically they could be much richer (with little effort and not that much money on their part) with good advice but not everyone in this group will pay for it. There is a lot of irony there.
Summing it up, I think it is kind of silly for someone who earns 9% on their money (I’m not saying you – just a typical run of the mill investor who asset allocates) to scoff at a mutual fund manager who earns say 15% when their benchmark earned 18%. That same investor would be much better off earning the 15% return than the 9% they got. Now don’t get me wrong I like indexing and you can use the tools fine – but if we are going to be evidence based investors here we have to acknowledge the fact that people’s portfolios don’t come anywhere near index rates of return and it’s quite possible people are messing up their long term growth prospects by being too conservative in their asset allocations and strategies.
It’s good to use what works for you in the best – but I think unless you’re earning 85% of so on your growth assets relative to an index, I think you have room for improvement. You can do it yourself or you can outsource it but we should always remember that what counts is the result actual investors are getting and we should be focused on ensuring that their assets are productive and working hard for them while they don’t need them. This can be done with index funds, stocks, mutual funds, reit’s, private investments – whatever they want to use — but I think if you are under 85% of the upside capture on your long term growth assets it is at least worth your while to continue to try to learn and see if you can improve upon whatever you are doing. You have to be your own judge of what works and be objective and honest with yourself – but we shouldn’t shun improvement because even a couple percentage points compounded over a long period of time on a decent amount of savings can lead to huge improvements in the financial position of an investor.
Of course if an approach adds 2% a year long-term it’s worth doing. I’m far higher than 85% of the upside capture of my long-term growth assets. It’s far closer to 99% given that I’m nearly 100% indexed. I get 100% of the bond return, 100% of the domestic stock return, 100% of the international stock return, 100% of the SV stock return etc.
If you think momentum/trend-following will add 2% to your long-term returns, then you should go for it. It’s a free country and there are many roads to Dublin.
I understand these realities, but regular folks beat the market as well (they are consistent, dont trade and just let things ride). This is getting much harder as everyone is piling into index funds. One prominent fund manager is the Vanguard Health Care fund, hes old now but has been killing the market for decades (nice to be in a great sector for sure). It happens, as opposed to no one does and all is hopeless.
That wasnt the point at all though, but rather people misunderstand “the market” when comparing things, what it really is, what certain funds are for, and mostly hedge funds as you wouldnt expect them to beat the market except in a down turn, as their name implies they are hedges. Its a different philosophy for people with more money than I’ll ever have to worry about.
But was mostly that people with varying asset allocations shouldnt look to the s&p 500 and feel either great or terrible just because their portfolio is doing better or worse. Unless you have a 100% american stock portfolio or index fund, they should not be expected to be the same. More about just measuring the right stuff. Just like when people compare the nominal price of gas today to some random past time, without considering inflation or the average mpg of that days cars. Usually results in assuming gas is more expensive now, since they are forgetting that the important measure is the cost of a fill up and how frequent that is, and incidentally, is cheaper than ever.
It’s can be good to be indexed but that is often not enough to get the job done. See many indexers are getting 98% or 99% of their benchmark returns per each asset class they invest in, but the way that they mix their benchmarks or index funds means that they are earning about the same return as the average investor – which means that they aren’t making much or any progress on their investments relative to everyone else that they will compete against for goods and services in retirement.
So what would be important in your case is the mix – how much are bonds, how are much are small value, how much are international index funds and how much are domestic index funds. I’m not asking you to answer that – I’m just saying that the assets you use and the allocation you give to each asset has a bigger impact on productivity relative to the small amount you save using an index fund versus say a mutual fund.
Additionally, even more important than the individual index funds you use is the price you paid for each of those funds relative to the cash flows or profits on the underlying companies. I believe that only 25% of the return of a given index fund can be attributable to buying that index, and the other 75% of the return is attributable to the price at which you bought the index. So for example on the S&P 500, only about 3% of the return may be attributable to the fact that you are in large cap domestic stocks and the other 10% of your return (if you get that much) may be attributable the premium or discount you paid to acquire that asset. If pricing is more important than asset classes then it means we have to pay close attention to pricing in our investing decisions.
Momentum matters too but I think it is less important than pricing. Nonetheless, the empirical evidence shows that it has worked for a long period of time and so open minded investors should at least consider it. In my experience, I’ve personally found it to be one of the more successful strategies you can employ and it doesn’t take as much effort or money to implement as you would think.
As you’ve acknowledged anything that increases portfolio productivity over the long term is worth doing and the benefits of doing it (when done right) can be immense to a good saver. I haven’t said that you will earn 2% a year from momentum strategies, but what I will say to the WCI community is that it is a good approach rooted in empirical evidence and it is worth a close look.
There may be many roads to Dublin but some of them are highways with no traffic on them and some are gravel roads full of potholes (and lots of variations in between). When you actually arrive in Dublin depends on how skillfully you applied the factors that make productive portfolios to your situation. When your peers arrive in Dublin you might already be on your way to Tripoli (I couldn’t resist).
You sound like you prefer a valuation-guided timing approach to a momentum-guided timing approach. That’s fine. Another reasonable road to Dublin.
The problem with a valuation-based approach is that you have to choose which one, and there are times where using any of them is wrong. For example, anyone using any kind of a valuation based approach would have avoided most of the bull run in equities from 96-2000. With momentum, you tend to get in too late and get out too late. With valuation, you tend to get in too early and get out too early. An agnostic, fixed asset allocation, buy and hold approach avoids both of those issues, keeps costs low, avoids many behavioral issues, and certainly requires far less time and effort.
A lot of what you said is very true.
But you’ve left out something critically important: for many the fixed allocation, buy and hold approach even with low cost likely earns millions less than these other strategies over time. The reason is that the buy and hold strategy ensures average returns over time if you are buying evenly which will almost always be worse than if you buy at a discount or if you pursue another strategy that enhances returns. The reason why the returns are average is that over time you’ll buy some assets at a premium, some at a discount and some at average prices – and it will kind of average out to average price.
But and just humor me for a second — what would happen if you just tried to buy assets at good prices instead of average prices?
The data is out there. A compounded 20 year return the median cyclically adjusted p/e ratio of 16 has historically earned 6.8%. By contrast when the cyclically adjusted p/e ratio was lower say 10 to 12 the returns on those years was 10.02%
So here is the average case:
$100k compounding at 6.8% for 30 years grows to $719,676. Pretty good right?
Well, not after you compare it to the alternative…
And here is the buying at a discount case:
$100k compounding at 10.02% for 30 years grows to $1,754,483.
That slight difference due to pricing is worth one million dollars! Even if you had to pay a one percent fee to get that you would be so much better off paying the fee than not using the strategy (you’d be $600k better off than if you pursued the first strategy and bought on a fixed allocation).
Now over the long term, you are way better turning your $100k into $1.7 million (or $1.3 million with a fee) than you are having a mere $700k. And it’s all done using the same investment – so there is no difference in what you are investing in — the difference is when you are investing in it.
A million dollar difference in outcomes is worth paying attention to and it shows that a valuation strategy over time is 2.5x more productive than a standard buy and hold the S&P 500 strategy. That evidence is pretty much overwhelming and if you are evidenced based – you can’t ignore it (although I guess you can choose not to act on it).
With that said, the valuation approach makes over 4 trips to Dublin in the time the standard price insensitive approach makes over 2 trips.
Now some people will call this timing the market and of course it is. But it isn’t that hard to buy the S&P when the cyclically adjusted p/e ratio if 10 to 12 or perhaps don’t buy it when the cyclically adjusted p/e is 26 (as it is today). You aren’t guessing what the market is going to do tomorrow. Rather you are looking at what people have historically paid for a good asset and if you can get a really good deal on it you pounce. If you can’t there is bound to be another asset with a good price on it you can buy. By contrast, a fixed allocation tells you to ignore prices which seems kind of silly when the evidence shows overwhelmingly how important prices are in the outcomes.
Aside from costing their users perhaps a million dollars (which is a pretty hefty penalty if you ask me) in foregone profits versus valuation based approaches, agnostic fixed asset allocation approaches can also suffer from behavioral issues. People may want to quit the strategy during periods of underperformance and/or drawdowns. That isn’t unique to value or momentum strategies, it is a universal issue in investing and people just have to have the toughness and perhaps the good guidance to follow through. With that said, they should follow through with the most productive approach they know of that they can actually implement. Financial Advisors may expand what you are capable of implementing because you are paying someone who has seen it all before and may have done greater research, have superior information, and therefore have greater natural confidence than you in whatever approach is being undertaken.
At the end of the day these reasons are why many people actually benefit from hiring financial advisors because the good ones know these facts and aren’t as easily tricked by market hype versus market reality and, if they themselves are disciplined, they can often help you stay the course even with tougher to follow strategies.
Remember that people who want to sell you an index product have an incentive to harp on price because that is something their product has an advantage in. But the net cost of using the product, especially if not used well, can more than cancel out the benefit of the lower initial cost if you don’t buy it at the right price. Similarly, some companies want you to keep on buying their product over and over again and so the buy and hold methodology is something they preach – but if it fails so clearly in an empirical test (to the tune of $1 million dollars in foregone gains) versus a very basic value methodology why buy the hype?
By the way, I couple of disclosures here. I don’t think index funds stink – but I think if you buy them wrong they do. Nonetheless, I own some index funds and some direct ownership of securities as do most of my clients. Also, I’m sure some people won’t take me seriously because of my investment background – but the empirical evidence speaks for itself all you have to do is look at it. And if you do better as a do it yourself investor, which many of you are, I don’t benefit in any way economically from that – but I still hope it happens because I know you worked hard for your resources and I hope they work hard for you. I also think many people would benefit from hiring some but certainly not all financial advisors. I think you should search out those who are 1) frugal so they aren’t under personal financial pressure to cheat you 2) rich by saving and investing so you know they know how to do it and 3) a teacher — that if you want to work with them for a while just to learn they will be cool with that.
Best of luck to everyone.
Obviously if you buy something at a cheaper price you’re going to do better than if you buy it at a more expensive price. However, the data is pretty clear that of all the people out there trying to do that in order to beat the market, on average they aren’t able to, at least after their expenses of doing so. It turns out it’s a loser’s game. Do some win? Sure. But the vast majority do not.
It’s interesting that you say that – because the way that you’ve framed the question you’ve guaranteed a certain answer. But I think we are mixing concepts here.
I’m not saying that you should try to “beat the market” – I’m saying that it is worth your while to get better as an investor to move from say the 50th percentile to the 70th or the 80th percentile. In order to do that, you only need to improve your returns by 1 or 2 percentage points net of all costs. And by definition it can’t be a loser’s game since 30% of people are in the top 30th percentile or better at investing. Does that make sense?
Also look at the actual numbers – the distribution is really tight meaning that a 1% or 2% improvement in results net of costs makes your investments much more productive than a lot of your peers. It’s never been about beating the market, it’s improving your spot on the distribution by having a more productive portfolio. That’s it — very simple concept — and if you look at the data it isn’t even that hard to achieve because a tiny difference in net investment return does the trick.
When you view investing in this relative way, it can’t be a loser’s game. You can “lose” at the game by playing it poorly or you can “win” at the game by being productive. Or you can be in the middle – which works for some but that may be a million dollar decision. But the difference being in the middle and being on the right side of the distribution is a million dollar difference for a good saver with 30 years to grow their money. So it isn’t something we should brush off — I mean that is a big difference in people’s lives wouldn’t you say?
Switching back, here are the results for 2014: I’m using index investor data since it is easy to get but in my experience the distribution for mutual fund investors and stock investors is similarly tight (just a few percentage points in return differentiates large numbers of people).
2014 Investor Results (some of this is estimated but I think it is generally accurate):
The median index fund investor probably earned around 8%.
The 60% percentile investor earned around 8.7%,
The 70th percentile investor earned around 9.1%,
The 80th percentile investor earned around 10.4%,
The 90th percentile investor earned around 13.1%
The 95th percentile investor earned around 13.7%.
I believe that the return on the S&P 500, with dividends included was 13.8% in 2014. So 95% of investors “lost” according to your definition. But according to my definition the top 50% of investors are winning to some degree and with a little effort they may win even more. In can be so easy to dismiss these results because what is 1 or 2 percent after all? I mean that isn’t worth doing it right over, correct? But because of compounding that ends up being a massive number and so investment productivity plays a huge role – as you’ve acknowledged – in where people end up. And pricing plays a big factor in that – so maybe people should pay attention to it? Certainly they aren’t losers playing a losers game if they just apply well known principles that are known to be effective over the long run.
If you are an 8% earning median investor, you may be able to be a 9% earning (net of cost) top 30% investor without taking crazy risks of doing anything wild — just applying proven principles and learning on your own or working with people who are already knowledgeable about what works. I mean stop and think about that for a second – a one percent difference in return net of costs moves you from someone treading water to someone who is adding to their relative position due to the investments. That is a great thing to do, it isn’t that hard, and its hard to argue that a 1% change is beyond people’s ability since I just showed you exactly how to do it – and you acknowledged that it works.
Your comment about buying cheap being “obvious” reminds me of a funny story early on my career. Clients used to always call me and ask me various things about what Warren Buffet said on CNBC and they were always very interested whenever he was on tv. And this went on for years with lots of clients, but the strange thing was these people who called me never actually ever owned in Berkshire Hathaway stock. They were mesmerized by this guy and yet they never ever bought a piece of that company – which was obviously doing quite well. Because one of my jobs was to go through client past holdings to see if they ever owned certain shares of stocks that were part of class action lawsuits, I ended up seeing every holding every client had since electronic records began in the 1990’s. Do you know how many times of time I saw Berkshire Hathaway stock over the tens or hundreds of thousands of holdings I saw — maybe 5 times. It was so obvious but everyone forgot to buy it or just never got around to it.
Sometimes we forget to do the obvious even when it is staring us in the face.
I disagree that adding 1% to your returns with no additional risk is easy. In 2014, a bull market, the top earning investors were more invested in US stocks than the bottom earning investors.
The only risk-free way I’ve found to add 1% to your returns is to fire the advisor charging you 1% to do something you can do yourself.
That all depends all on how you define risk. If you define risk as volatility – you win hands down — stocks, real estate, pretty much all growth assets are more volatile than cash. But you advocate owning them, and so do I – why do we do that?
The answer is of course, that they are also the source of growth and a little bit of temporary volatility is a fair price to play for several hundred thousand or perhaps a million dollars. The important thing for everyone to remember is the volatility is usually temporary but the foregone gains are usually permanent.
I would argue that true risk in investing comes from paying too much for an asset. If you buy things at huge premiums to whatever people have or can pay for them, they can be really difficult to offload while not losing money. In that way, an S&P 500 index fund bought in 2009, at a discount to what people pay for it, is a safe long term investment, in my opinion, because you’ll probably be able to sell it for more than you bought it for in the same environment. You might reap the benefits of that investment for years and it might have increased your returns in 2014. That would be, in my judgement, a safer route than an investor who bought say a lot of long term bonds in 2014 which fetched premium prices at the time. So I don’t think you can look at return and conclude that one person’s portfolio is more risky than the other. It may be more volatile – but a riskier looking portfolio might actually have a lower chance of a true risk of loss if they bought the asset at a good price than a less risky looking portfolio full of lower volatility assets bought at premiums.
As for firing the advisor — it’s the same analysis as valuation or anything else. If an advisor costs you 1% a year they have to produce an extra 1% of gains net of costs from what you would have done on your own to break even for you. If they help you produce 2 or 3% of gains that you would not have had on your previous strategy and cost you 1% you are a net winner and they are a net winner. Everyone comes out on top.
For people around the median (at the 8% return range), I think a lot of them could have advisors that create a net benefit for both sides net of cost – and many smart and rational investors do this. The consequences are very high for making mistakes and the benefits are similarly wonderful for doing it well and once people figure that out they tend to invest heavily in getting better whether that be by your reading your site, finding other good investors to talk to and hiring a competent advisor. But I can tell you that people who do this for a living have an absolute advantage over those who don’t simply because they get to see so many portfolios and so many ways of doing it that they would have never thought of on their own.
I think there is a joke in legal circles that a lawyer who represents himself has a fool for a client. Not everyone is so good at self evaluation nor is everyone an expert or willing to put in the time to be a good investor – despite the obvious benefits of doing so. If you can get good on your own, which I fully encourage people to do if they want to, then you most definitely have a profitable hobby. But if you don’t get good you might have a very expensive hobby. It all depends on the person.
Buying stocks in 2008-2009 seems like a very safe investment NOW looking back, but I assure you that it didn’t feel like you were making a very safe investment at the time. It felt like the world might implode. We didn’t know if we’d go from the 50% drop we’d already seen to the 90% drop seen in 1929-32.
I agree that theoretically an investment manager can add 2-3% to your returns. However, most of that 2-3% comes from getting a reasonable asset allocation, choosing low cost investments, and avoiding bad behavior. If you can do that much on your own it is very difficult for an investment manager to even pay for himself, much less add value.
I have to admit Jim — they way you frame the issues sometimes surprises me. A big part of me when I hear your arguments wonder why you don’t just pull back when the facts stop supporting your argument. I mean the evidence really supports 90% or so of the indexers arguments – but the final 10% they try to extend it out is where a really good evidence based argument just fails, in my opinion. And it doesn’t fail on opinion it fails on the lack of empirical facts.
Let me just focus on the 10% where I think your argument lacks factual support.
First, do you believe that there are zero behavioral drawbacks to a fixed asset allocation? Do you think emotional feelings about investments only impact those with other strategies and indexers have a force field of security around them that makes them never question their strategies? Won’t some of those investors abandon course when they are experiencing big drawdowns or not participating in a growing market as well? Is there evidence that shows fixed indexers abandon the strategy at lower rates? Have abandonment rates of other strategies ever been studied?
Second, with the people who are disciplined enough to stick to a strategy – do you believe that some are historically empirically much more effective than others? From everything I’ve seen valuations provide one of the most effective strategies – and I’m not being dogmatic here I have no personal preference for one strategy over another (I’d be a fixed asset allocation indexer if the evidence showed it to be the most effective way to grow money) I just pay attention to what has worked well for people in the past. The empirical difference in asset class returns can be as much as 4% per annum from someone who buys at a good valuation compared to someone who buys at an average valuation. If you look at the difference between someone who buys at a good valuation and someone who buys at a bad valuation the difference is about 10% per annum. What do you not like about a valuation approach and do you think using it would benefit your portfolio? Do you think it would benefit other people’s portfolios?
Third, how would you quantify the empirical difference between a low cost investment and a high cost investment? So for example if you went from a fund or an advisor that cost 1% to a similar set of investments that cost of 0.06% what is the annual empirical benefit of that decision? My belief is that the value added by saving money is exactly what you saved 0.94% compounded over time, nothing more and nothing less. Every little bit counts but isn’t this a small factor compared to the 4% to 10% improvement that people get from buying at good prices. I mean if it cost me 0.94% to earn 4% extra – I’ve netted out a benefit of 3.06% for myself. Empirically can we see it any other way?
The reason why I ask that is that I sometimes think the maniacal focus on cost you see amongst indexers is prone to miss the forest from the trees. And that isn’t personal opinion – that’s empirically based — you are never better off foregoing 4% in gains to save 0.96% in fees – the math just can never support that argument. A dollar saved can never cost you more than a dollar of forgone gains or else you would be better off not saving the dollar. Maybe you can explain it to me if I’m missing something – or maybe we are on the same page and you agree that there are limits to the benefits of choosing low cost investments.
I like your site and the arguments you’ve made. But I just can’t find any facts that support the concept that 1) fixed asset allocations how lower abandonment rates than other strategies and don’t come with their own behavioral issues with implementing them 2) Valuation enhancements even on an indexed portfolio doesn’t add HUGE compounding benefits over time with lower risk of permanent loss 3) Price savings can’t contribute more than 1% a year per annum to investment returns whereas other strategies that you may miss out on by price saving generate 4-10% a year in extra gains at lower risk per annum and 4) there is evidence that people who are indexers or any other point of view can’t apply or utilize valuation when guided by those who understand it.
Your argument about not paying for something you already know how to do is totally logical and I agree with it. There is 100% empirical support for that one 🙂
Thanks for your comments.
All right, we’ve been talking vaguely up until now. You seem to want to get specific. You claim to have a strategy that beats a buy and hold strategy by 4% a year. Let’s hear it. What is it?
The strategy is to buy the S&P 500 index fund when the cyclically adjusted p/e ratio is between 10 and 12 and hold it for 20 years. The strategy should work because you are benefitting from normal regression to the mean and while it involves some market timing it is a clear rules based approach to investing.
The results are if you would have bought the S&P 500 when the cyclical p/e was under 12, you would have earned around 10% a year real. If you would have bought it at an average price, which is something that would probably happen if you were price insensitive, you would have bought the S&P 500 at a cyclically adjusted p/e of 16 – which historically that has earned 6% a year in real terms. The mere 4% a year difference, is worth about a million dollars over time in the example I laid out in a previous comment.
That should hopefully lay out the approach of just buying things when they are on sale – which should enhance an index return over time.
I saw you skipped over some of the points not related to this strategy. I’m asking about them seriously because I actually wonder if there are facts that you have that perhaps I and others just aren’t aware of. These are the questions from the previous comment and it would be great if you could answer them and what kind of facts have lead you to believe what you believe.
1) Is there a reason to believe that fixed asset allocations have lower abandonment rates than other strategies? Has it been established that valuation based strategies have high abandonment rates? Has anyone been able to establish scientifically that investors can handle one strategy but not the other?
2) Is there any evidence to show that buying that at a discount does not in fact lead to portfolios that compound at higher rates? Is there evidence to show that buying things at a discount creates a greater risk of long term loss?
3) Is there any evidence to show that investors benefit from lowering cost by an amount in excess to the exact amount they saved on fees or expense ratios. For example, it is known that if you cut your costs from 1% to 0.06% you will increase your compounding rate by 0.94% per annum provided you achieve the same rates of return under both strategies. Are there any empirical reasons to believe that if an investor takes a strategy that earns 4% of more than what they were doing but has to pay an extra 0.94% a year to do it – is there evidence that shows that they have not increased their compounding rate by 3.06% per year despite the extra 0.94% in fees?
4) Is there any evidence showing that an indexer or someone who follows any other strategy could not in fact successfully utilize or apply valuation when guided by someone who understands it?
I hate to bring up the same four questions from the last comment – but you seem to me like you really believe deeply in your positions and I’d love to know what kind of facts or evidence has lead you to those beliefs.
Thanks again – and I really am trying to learn here — I don’t know everything about indexing but the arguments you’ve raised – I’ve seen raised in lots of places (where I was not part of the debate) but I never saw the facts that backed up those beliefs. I am hoping that maybe you might have them and could supply them to us so that we can learn.
Thanks again for such a great blog!
I skip over a lot in your comments Dan. Are you familiar with the acronym “TLDR?” I’m getting 30-50 emails a day and another 30 or so comments on the blog. Your comments range from 500 to 1200 words in length, about the length of a decent blog post. So if I’ve skipped something, it’s because I’m skimming.
Here’s the issue with using a PE10 strategy or Shiller PE or whatever you want to call it. Not only is it not anywhere close to a perfect system (i.e. sometimes returns starting from a higher PE are higher than returns for a similar period starting from a lower PE) but it also ignores the fact that you may go a very long time with out hitting whatever your goal PE is. There is no doubt I’d rather buy stocks at a PE of 10 than a PE of 20. But I’ve got an issue in that I have money to invest every paycheck. I have to do something with it and my choices are limited. I can buy stocks, real estate, bonds, gold, or cash. Right now, cash basically loses 1-2% after inflation each year. If the last time I thought the PE was attractive enough to buy was in 2009, and my plan is to put my money in cash when it isn’t, then I’ve now had 6 years where I’ve lost 2% a year instead of making 5-15% a year.
1) Yes, because you don’t have to look at your portfolio or the markets as often. I don’t know of a double blinded study though. Believe what you want and invest as you like.
2) Not that I know of. Wouldn’t you prefer to buy at a discount if the discount is available?
3) Somewhat. For example, index mutual funds have lower ERs, but they also have lower transaction costs (fewer buying/selling events so fewer spreads to pay) and tax costs due to the lower turnover. Plus you’re not running manager risk. Of course it is wise to pay 1% to get 4%. That’s a simple math question. The real question, however, is whether or not there is a way to reliably do that. Mutual fund managers, hedge fund managers, brokers, advisors etc all have a vested interest in convincing us that they are worth their fees. The data, however, suggests otherwise. It suggests that active managers can’t reliably outperform the market by more than their costs of doing so.
4) Not sure what you mean by indexer. Lots of people who try to use valuations invest in index funds, and lots of those who don’t also invest in index funds.
By the way, I realized that I didn’t totally answer your statement. You don’t have to have an exclusive momentum or valuation based strategy in a portfolio. You can run multiple uncorrelated strategies side by side in a portfolio if you want and it might even create an additional diversification benefit to do so.
What I wonder is that if you run a momentum strategy and a valuation strategy will they just cancel each other out and give you a total market strategy with higher costs and hassle?
They very well could. I’ve never tried to break it down. I can tell you this much the value stuff I am doing now is dragging down results and the growth stuff (which is where the momentum is) is working well.
I would say that there is not a huge hassle factor or cost factor of having them take place simultaneously. You can actually buy and hold momentum investments to a degree (I learned that at an investments class at Yale) so you don’t actually need to constantly churn them. I’ll tell you the first time I was introduced to it – I couldn’t believe it would work — but then I tested it and it absolutely did. I guess there are benefits to always expanding your knowledge base.
But not unlike your previous point about having a strategy that underperforms for a period of time — it hurts to own value investments right now and they are generally not working in the current period. But I’ve been around long enough to know that eventually everything switches and probably my growth stocks will be killing me while my value investments get their day in the sun.
Interesting idea, that would be a total bummer. The valuation strategy requires a whole lot more work than a momentum/relative strength one for sure as well. Further, it seems momentum is much more of an anomaly than value investing according to the data which is interesting.
Even if you did a small portion of your portfolio in a momentum style that could shave a couple of bear market losses out, that would end up a partial to full percentage point over time and be diversified. We’ll see how it goes, Im giving it a try with smaller amounts of money. Letting it high fly for now but constantly looking at “decent” returns with lower downside. Probably have to wait for the bond market to unwind a bit before that strategy will be good again, if it even can be for some time.
Yeah valuation is more work. You can make it more manageable by just doing a valuation approach on the major index funds – but there can be lots of times, like today, where there isn’t much to buy from a valuation point of view. You have to do quite a big of digging in todays environment. Remember 2008 though — you could have bought almost anything based off a good valuation and gotten better than normal returns (which is probably why this bull market has been so good).
But I agree the momentum approach is much easier to implement in the current market. Best of luck with it and I think it is really good that you are starting small and seeing how it works for you – we always seem to learn best when we have real money on the line 🙂
I do valuation as well, but its certainly easier in down times, and I hope to be in a position (mostly mentally) to follow through when it comes. I bought some oil stocks when they were down, buy country etfs when theyre getting hammered, etc…Its like reverse momentum really.
Indexes are awesome in a bull market, or before a turnaround/after correction. I think its hard to beat a secular bull.
I didn’t think I would say this but I actually liked your answers. They were very fair and balanced and I thought presented a fair treatment of the issue.
I think worry about fees is overblown. The evidence on fees really shows two things: 1) most fund advisors don’t beat that market – neither do 95% of individual investors – so they have something in common. 2) It is possible to actually use low fee products and invest badly and in the long run the price you pay for the asset is a bigger determinant of outcome than the fees you pay (because asset pricing differences create a bigger range of results than fees which are actually kept in a pretty tight range from the most expensive to least expensive options).
I usually don’t chime in much on your blog despite being a regular reader but I might have to in the future if I see people spinning the fee issue in a way that is untrue 🙂 And of course people should be skeptical of my view too – but they should also be skeptical of Vanguard who might want to convince that if you just lower your fees you’ll have it made in the shade – which isn’t true for most people and can lull them into a sense of complacency and lead them to ignore other factors that are highly important to having a productive portfolio.
Thanks again for taking the time to answer my questions – it was very helpful to understanding your point of view.
One thing about fees is that it is one of the few things that you actually can control. Might as well do so.
To throw some more fuel on the fire, the offer adds to very compelling dual momentum strategies on his website–neither is discussed in the book. I surmise that you can take any number and kind of asset classes, especially if they are not highly correlated, and apply the dual momentum strategy.
Meet Dual Momentum Fixed Income:
http://www.dualmomentum.net/2014/12/dual-momentum-fixed-income_23.html
and Dual Momentum Social Responsibility:
http://www.dualmomentum.net/2015/03/sustainable-momentum-investing-doing.html
Enjoy!
If you want to see someone doing a very interesting style of momentum/tactical asset allocation check out Frank Grossmans Logical-invest.com. They have several strategies they do through a newsletter format though they tell you enough to do it on your own as well.
Whats great about the site is there is a lot of the backtested data which looks insane, but then there is also some forward data so you can see in real time how it fared and compare the promo to reality. He has some extremely educational articles/comments on seeking alpha. Id read those for more info.
Thanks for the link, Zach. There’s a lot of great information on that site and some of those strategies certainly look intriguing. Have you tried to implement any of those strategies in your portfolio?
Ignore Frank Grossman. He wrote some articles in 2013 showing great results over a 7-10 year back test. He’s gotten crushed since then and then modifies the system to take into account what has happened since. He’s the epitome of curve fitting. Go to his 2013 articles on seeking alpha and read the comment section. He gets blasted and has not posted since.
Dual momentum has a longer look back (1973) and has done well since being published.