 |
Does Market Timing Really Perform?
by Paul A. Merriman Publisher and Editor
You are probably looking at timing results for the U.S. stock market in the 1990s. During bull markets, timing usually underperforms, and it’s not hard to understand why. When the market is going up, the only thing timing can do is tell you to be invested; that doesn’t give you any advantage over buy-and-hold. But of course no bull market goes up in a straight line for very long. And often when there’s a downward blip, it causes a sell signal, and you’re out of the market.
A real bull market can quickly reverse course and start going up while you’re still on the sidelines, and that means you miss out on some of the gains. Of course you could just ignore the sell signal and remain invested. But you never know at the time whether that sell signal is a "false alarm" or your tip-off that a big crash is on the horizon.
That’s why timing usually underperforms during bull markets. But during bear markets, timing often produces superior returns by getting you out of the market before you suffer the worst damage from falling prices. Our timing systems did exactly that in 1987, the market’s last whopper of a crash.
During the 1990s, timing has worked very well for bonds and international stocks. This doesn’t mean there’s anything wrong with timing the U.S. stock market. It just means that the patterns in these other markets have been more "normal" in recent years, allowing timing to get investors out soon after market peaks and get back in soon after the low points.
Well, the best answer is that you should continue following your timing strategy. Give it lots of time, certainly at least several years, before you even consider abandoning it.
However, that’s easier for me to say than it is for you to do. It depends on several things, including why you are disappointed. Do you have realistic expectations for that timing system? If you started using timing in order to "beat the market" over some short period of time, then I think that’s unrealistic.
Last year our timing systems were especially productive, and a lot of investors started using them. I’m afraid some people adopted timing for the wrong reason, looking to jump on the bandwagon of whatever was a hot investment idea at the moment. But if you flit from one strategy to another that way, you’re depending on luck to bring you good results.
Over the long term, successful investment doesn’t depend on luck. And whether or not you use timing, if you want to be successful, you have to make a commitment to a strategy and then stick with it long enough to let it work the way it is supposed to work.
I’m not saying you should never abandon a strategy if it gives you awful results. Sometimes in life there are times when the best thing to do is simply cut your losses and move on without looking back.
I realize I am straying away from your specific question, but I’m glad you asked it because it gives me the chance to make an important point about choosing a strategy. And that point is this: When you pick a strategy, decide right then how much time you are going to give it to either produce or be replaced. And decide at what point you are going to cut your losses. For instance, you could decide you are going to stick with something at least four years or until you lose 20 percent of your investment, whichever comes first. I’m not saying that’s the proper formula, but I want you to see what decisions you should make.
Make sure your stop-loss break-point is realistic for that investment. In the tables of returns that we have been presenting today (click here for an example), we show the worst interim periods for every strategy – the worst one-month, three-months, 12-months and so on.
Those numbers are there for a reason. If a strategy lost 30 percent over a 12-month period in the past, you should expect to live through something comparable in the future. So if you start using that strategy but decide in advance that you’ll bail out after a 20 percent loss, you are setting yourself up for a failure. Logically, you have to be thinking that either you will see the loss that’s coming better than your timing system will (and in that case, why do you need a system at all?) or that you will just be lucky and you won’t experience a loss as great as 20 percent. Either way, you are fooling yourself.
At the very least, you should give a strategy three to five years to show what it can do for you. If you’re not willing to make that much of a commitment, then maybe you should be a buy-and-hold investor. But that still puts you back in the same position, wondering how long you should hang onto an investment before you replace it with something else. If you don’t want to stick with a timing system long enough for it to work, what makes you think you’ll stick with a buy-and-hold system?
I see that kind of statement in the media from time to time, and they are not only scary, they are misleading and foolish. This is a silly argument that’s been around for a long time. It’s an easy "score" for lazy critics to capture lazy minds.
I bet that article didn’t tell you what would happen if you were in the market except for the 40 worst days. That would be just as valid, but it would not support the anti-timing point of view.
No market timing system is designed to avoid only the worst days or the best days, and we’ve never seen any claim that anybody ever did this. But just for fun, we decided to study the month-by-month returns of the Standard & Poor's 500 Index over a 50-year period. Here is a table that shows what we found:
*Compound rate of return
If you look at the dollar numbers, it’s astonishing to think that 69 percent of the returns of large U.S. stocks over 50 years would have been wiped out if you subtracted just the 10 best months.
But that’s only half the story. It’s even more astonishing to see what would have happened if you had a way to get rid of the 10 worst months. That would have made you 3.3 times as much money as you’d have from investing in the index in all the months. You’d have $1.9 million instead of $579,000.
Now you and I know that in real life, there’s no way that market timing can do either one of these things except through incredibly good or bad luck. That’s why I think the whole argument is foolish. The people who write these articles hope they can scare people away from timing. Their arguments don’t mean timing is either good or bad. Their arguments do show how important timing can be.
It’s certainly possible to get very high returns for one calendar year. And it’s possible to do it for two years in a row, though that is quite a bit less likely. In theory, I guess it’s possible you could even get returns like that for four or five years. But I think the chances are very, very small. And for any period longer than that, your chances are probably better if you "invest" heavily in your local lottery.
I think these claims just defy common sense. A 50 percent annual return, sustained over time, would produce astonishing results. At that rate, you could start with only $250 – that’s less than most people’s monthly car payments – and grow it to $831,000 in 20 years. In 30 years that $250 would grow to $48 million. And over 50 years, that $250 would be worth $159 billion.
I know that the prospect of extremely high returns is awfully tempting to a lot of people. I suspect that as long as there are human beings on the planet, there will always be a place for greed.
But I happen to agree with the advice Mark Hulbert, whose Hulbert Financial Digest has been tracking investment returns of newsletters for nearly 20 years. Mark says that whenever you see such claims, you don’t even need to read all the fine print. Just throw those advertisements away. You deserve to do something better with your time. And you deserve to do something better with your money.
Mathematically, yes it can be true. If you keep making that $2,000 contribution every year for 30 years, and if you got a 20 percent return, you’d have $2.4 million. If you got 22 percent a year, which is not likely but certainly possible, you’d have $3.5 million. And if you got 24 percent a year, you’d have almost $5.3 million.
The question of course is whether you can do that. Our hypothetical studies show that if you had used market timing with aggressive strategies over the past quarter of a century, you could have achieved compound rates of return of 20 percent and higher. There’s no way to know what those returns will be in the future, but it’s certainly possible that level of performance could continue until you are ready to retire.
But even if you get a lot lower annual return, putting $2,000 a year into a Roth IRA for 30 years and investing it aggressively is a great way to produce a retirement fund. If your Roth IRA grew at "only" 14 percent a year, you’d have $713,624 in 30 years. In 35 years, you’d have almost 1.4 million. That’s enough to produce a substantial tax-free retirement income without depleting your capital.
That’s a good question, and you have to be skeptical of hypothetical results. When you see hypothetical returns, you know that something, usually the timing system, has been optimized to respond to past patterns. And those patterns won’t ever repeat themselves exactly.
So hypothetical results really give you much less guidance about future returns than you want. But that’s also true of actual results. That’s right, even factual results that really happened don’t tell you what the future will be like. But there’s one thing you can say for sure, and that is: The future won’t be just like the past. Whether you’re dealing with a "real" past or a hypothetical past, things just aren’t going to turn out in the future quite like they did before.
But there’s one thing that past results, either real or hypothetical, are very good for, and that’s to give you some parameters of how much risk you are taking with a strategy. If the past shows you could lose 40 percent of your money in a year, and that was the case with the Standard & Poor's 500 Index back in the 70s, then you had better figure it could happen again. I’m afraid that fact is lost on a lot of people these days who believe that they can make their money grow at low risk with only an S&P 500 Index fund.
The short answer is that the farther back in time you can go, the better your data. And the better your data, the more realistic your conclusions will be.
We can design timing systems that in the past would have done just about anything we want. Say you lost a lot of money in the market yesterday, and you decided you wanted to make sure that never happened again. You could probably design a timing system that would have gotten you out of the market in time to avoid whatever losses you took yesterday.
But obviously, you would want a system to protect you from a lot more than the problems of a single day. And when you try to devise a system that will deal with many different patterns, you have to make compromises that are likely to reduce your returns. I have studied hundreds of timing systems over the years, and it turns out that no single system is very good at identifying all the dangers and opportunities that you would want to be aware of.
So the best way to respond to a variety of market patterns is to use several independent systems, each of which is effective at identifying certain types of situations. I don’t trust any one system enough to rely on it exclusively, and that’s why we use multiple systems.
To get back to your question: The most important thing a timing system can do for you is let you invest in an asset class while protecting you from the occurrences that are relatively rare but quite devastating. In the U.S. stock market, nothing in recent memory fits that description so well as the severe bear market of 1973 and 1974. So make sure you have a timing system that’s been tested back to the early 70s and one that would have protected you from the bulk of 1973 and 1974.
Let’s think for a minute about why you should or should not use timing. There are good reasons and bad reasons.
If you use timing as a way to beat the market, that’s the wrong reason. If somebody convinced you that some timing system is the next thing to a sure bet, that’s also the wrong reason.
One good reason to use timing is to have some active – instead of only passive – protection from the major losses you could sustain in a bear market.
Another good reason to use timing is if you believe the market is bound to crash sooner or later, and you know that you won’t be likely to see it coming in time to get out. A mechanical timing system can help you do that.
It’s just not practical to report timing results on an after-tax basis because everybody’s tax situation is different. What tax bracket are you in? Are you subject to the alternative minimum tax? Do you have capital loss carryovers from previous years? All these variables would be necessary to report after-tax returns as they would apply to you.
You are right, taxes can certainly reduce your returns from timing. But the financial damage may not be as bad as it seems.
To find out how taxes might actually affect an account, we did a hypothetical study assuming $100,000 was invested at the start of 1972 and continued through 1998. We compared an investment in an aggressive timed strategy using enhanced index funds at a level of 2 to –1 and an all-equity buy-and-hold strategy using Dimensional Fund Advisors’ asset class index funds.
The results were startling. If you had invested in the timed account over this 27-year period, you would have paid $2.5 million in taxes – about 25 times as much as his original investment! If you had done the buy-and-hold portfolio, your total taxes over 27 years would have been "only" about $414,000. Yet at the end of 1998, the timed account was worth $4.16 million, the buy-and-hold one $3.89 million.
So even after paying enormous taxes, the timed account did a bit better than the buy-and-hold account. I think this shows that even with taxes, market timing can work in a timing account. However, there’s no question that market timing works best in a qualified retirement plan such as a 401(k), a 403(b), a profit-sharing plan or an IRA.
In those accounts, there is no tax consequence, regardless of how many trades you make. In a traditional IRA, what ultimately matters to the IRS is the same thing that ultimately matters to you: how much money your account is worth when you make withdrawals. In a Roth IRA, it’s even better, because you won’t ever pay federal income taxes on your profits.
Finding a good benchmark for a timed account isn’t easy. Comparing timing results with buy-and-hold results is like comparing the results of U.S. stocks to those of international stocks.
Try to find a benchmark with similar risk characteristics and volatility. For instance, we believe that a 100 percent equity account with timing has about the same level of risk as a buy-and-hold account split half and half between stock funds and bond funds.
Another good benchmark, with or without timing, is your own personal needs. If you absolutely must have an average of at least 12 percent a year (or whatever the number is) from your investments, that is a good benchmark for you.
You are in luck. Twice a year, each winter and each summer, The Hulbert Financial Digest publishes an exhaustive report on the performance of more than 450 portfolios recommended by investment newsletters, including FundAdvice.com. Mark Hulbert, the editor, is independent, credible and a very thorough student of newsletter performance. Sometimes our portfolios stack up very well against the others he tracks, and sometimes we’re down in the pack. But we trust Mark’s unbiased reporting, and we especially appreciate his insistence that returns be adjusted to reflect the risks that investors had to take to achieve those returns.
To learn more, visit www.hulbertdigest.com on the Internet or call 1-800-HULBERT.
Source: http://www.fundadvice.com
Article viewed 550 time(s). Read more: 1 | 2 | 3 | 4 | 5 | 6 | 7 | 8 | 9 | 10 | 11 | 12 | 13 | 14 | |
||||||||||||||||||||||||||||||||||||||||||||||||||||||
