Where is the link between the graphical analysis (investors as a whole are bad at timing the market) and the subsequent quotes, (e.g., Bogle saying "I do not know anybody who has done it successfully and consistently. . . .")
It is a far cry from knowing that investors _as a whole_ don't market time well, but that has nothing to do proving that some skilled specific investors might not be good market timers.
For the record, I do think there's evidence out there that does show that even the "best" market timers succeed largely because of chance or luck, but it's not something you can get to from the data offered here. Because there are individuals who do succeed at timing the market. The question is whether they succeed due to skill or luck. . .
They're both part of the same story: that individual investors are bad at this, and professional managers aren't any good at it either.
Fair point though, that was definitely a conceptual leap of some distance there between the two.
The skill vs. luck argument is actually better investigated by looking at a separate data set: that of the persistence of performance over time for the same manager. We'll do a story about that sometime in the future.
How does outflow of capital from one asset to another measure investor's ability at timing the market? Yes, one pulls out on equity if one predicts that S&P500's going to drop but most of times decisions are made based on the volatility and risk tolerance. A retirement fund with a closer retirement date or income/growth fund will pull out if they find the risk/volatility level of the general market to be unacceptable.
A better measure of investor's ability to time the market is the performance of hedge funds as by its nature, the traders trades both sides of the market and performance is measured by their ability to predict both downward and upward trends.
Good data for the performance of hedge funds as an asset class that account for survivorship bias is somewhat hard to find. Off hand, Swensen in "Unconventional Success" recounts that for a single decade period that he was looking at, third quartile managers matched the market before fees. So after the standard hedge fund fees investors in those funds underperformed the market by about 1.6% (in his specific decade-ending-dec-2003 time series example).
Looking backwards, you can always identify fund managers that beat the market, but only in hindsight. One big part of this that's left unsaid is that yes it's possible to find managers to beat the market in hindsight, it's finding these managers ahead of time that's difficult & unlikely. Coupled with how much you'll trail the market if you try and don't succeed in finding outperformers ahead of time, it's a bit of a losing game to try.
There's an entire other field of study about the persistence of performance, but suffice to say that looking in the rearview mirror for last decade's outperformers doesn't help you find the next decade's outperformers.
And how are those folks faring who didn't time the market for the past decade? On average, their funds are at the exact same place they were in 2001.
I work for a company (who I won't mention simply because I don't want it to appear that I'm shilling for them), who has used market timing since 1972 and over the past 40 years have consistently outperformed the market using three fairly simple market timing indicators. Not by leaps and bounds - generally just a few percentage points - but still outperforming.
Thats true, but we must remember that for folks who _did_ try to time the market many will have actually had negative returns because of failing timing attempts. For example, there was a massive outflow of funds from Equities after the most recent correction (as there usually is when the market performs badly), and those people who pulled out of equities did not get to participate in the record-setting rebound that happened soon after.
Comparing performance over any given time period to some arbitrary standard (in this case, "flat" is assumed to be bad) doesn't tell the whole picture. On the flip side, there are many instances where funds performed admirably but in a time when the markets as a whole did even better. Just as I wouldn't commend a fund for gains in a bull market, docking a fund or portfolio for being "flat" when the market as a whole was flat over the given time period is unfair.
I don't doubt that your company has outperformed (after all, you have the data and I don't). But for what it's worth, was it a slam dunk to assume 40 years ago that your company would have outperformed over the subsequent 40 years? That's the problem: picking winners before they're winners. Will you outperform for another 30 years (my own investment horizon?). Hard to tell.
Er, every transaction has both a buyer and a seller – both of whom are investors. If one is timing well, the other is timing poorly – summed over both, the class of all 'investors' can never be winning or losing from timing. Only subsets of investors could.
Because we used the inflows & outflows of a couple large retail mutual funds as proxy for investor demand, you're right that the story actually is "retail investors are bad at timing the market".
There's a whole other set of data to dive into whether or not professional managers as a whole are bad at market timing (spoiler: they are terrible at it), but you're right in that this is not what this data set addresses.
Really the point the article is making is that on the whole, you and I, the investing public, shouldn't try to "read the tea leaves" and pull our money in and out of the market / move money around in the market because of what we believe will happen in the near future. What this data proves is that this doesn't actually work.
> On the x-axis we show the flow of capital into an asset class
For every buyer, there is a seller, so every single transaction nets out to exactly $0 flowing into the asset class. So exactly how does capital "flow" into or out of an asset class?
You would guess that they are confusing the change in valuation, but that is what they are using for the other axis.
You sell your stocks (asset class: equity) and someone buys it and gives you cash; then you go and buy treasury bonds (equity class: government bonds). Money flow out of equity into U.S treasury, otherwise known as "flight to safety."
The x-axis actually shows the flow of cash into mutual funds in the given sector, which we take as a proxy for investor belief in the value of investing in that sector.
I believe this is testing a false premise. The premise is, that when investors think the market is going to go up in the next month, they move money into mutual funds, and that when investors think the market is going to go down in the next month, they move money out.
I'm sure some proportion of investors move money in and out of mutual funds based on their market expectations, but I do not believe they do so with a single months time horizon, nor do I believe they are the majority.
I submit that a potentially quite large proportion of investors-- especially in mutual funds-- are not doing this, but are instead moving money into those mutual funds each month as a percentage of their paycheck goes into their 401k or into their savings plan. Further, some large proportion is likely investors moving from one mutual fund to another, while others are cashing out because they retire, etc. I bet a check on your flow dates to see if its correlated with an annual peak in employment change dates (e.g.: what month are people most likely to retire, or change jobs?) might find some correlation.
Worse, if I'm reading it correctly, this plot shows the return of the asset class over the course of a single month.
This means that you're assuming that these mutual fund investors are trying to time the market on the scale of the expected return within a month. This seems implausible to me given that mutual funds are managed and thus not well suited for timing (a lot easier to time the earnings announcement of Google, for instance, than guess what moves a mutual fund manager is going to make in a particular month.)
I suspect that most investors who choose mutual funds have at least a years time horizon, if not 5 years or more.
Your point about the single month as the timing period is definitely valid - in hindsight the article would have benefitted from having that be maybe 6 months or 12 months instead.
The premise isn't as clear-cut as what you laid out, in my opinion. In general people do start with the best of intentions; that is, their time horizon when they buy is usually something like "until I need the money". But, that's the generic case under stable market return conditions. In times of panic, folks who thought they were okay with risk find out they're not okay with it, and pull out (usually after much of the panic has already passed). In boom times people start to, like you said, "move money between funds" usually in a way that follows the recent price increases (gold recently, tech in 2001, etc). It's these movements that the article is written against - and you're right, it would have benefitted from a longer time series of returns.
Indeed, if you remember the oft-quoted Nasdaq Composite Index from the dot com boom - the level in 1998 was the same level as in 2002, but in the interim investors as a whole lost billions. That's a lot more than would have been lost if people had just regularly been investing the same amount each month into their 401(k), which is what we wish would have happened.
Active investors of all kinds definitely try, but again as a whole they've failed. I'll try to dig up the data and put it into an article sometime, but the fallacy that you should just pick a hedge fund and it'll outperform via market timing or securities selection is false.
And of course, as for picking the "right" fund which will subsequently do that - well that's the hard part.
[+] [-] hesitz|14 years ago|reply
It is a far cry from knowing that investors _as a whole_ don't market time well, but that has nothing to do proving that some skilled specific investors might not be good market timers.
For the record, I do think there's evidence out there that does show that even the "best" market timers succeed largely because of chance or luck, but it's not something you can get to from the data offered here. Because there are individuals who do succeed at timing the market. The question is whether they succeed due to skill or luck. . .
[+] [-] bolu|14 years ago|reply
Fair point though, that was definitely a conceptual leap of some distance there between the two.
The skill vs. luck argument is actually better investigated by looking at a separate data set: that of the persistence of performance over time for the same manager. We'll do a story about that sometime in the future.
[+] [-] noname123|14 years ago|reply
A better measure of investor's ability to time the market is the performance of hedge funds as by its nature, the traders trades both sides of the market and performance is measured by their ability to predict both downward and upward trends.
[+] [-] bolu|14 years ago|reply
Looking backwards, you can always identify fund managers that beat the market, but only in hindsight. One big part of this that's left unsaid is that yes it's possible to find managers to beat the market in hindsight, it's finding these managers ahead of time that's difficult & unlikely. Coupled with how much you'll trail the market if you try and don't succeed in finding outperformers ahead of time, it's a bit of a losing game to try.
There's an entire other field of study about the persistence of performance, but suffice to say that looking in the rearview mirror for last decade's outperformers doesn't help you find the next decade's outperformers.
[+] [-] jonheller|14 years ago|reply
I work for a company (who I won't mention simply because I don't want it to appear that I'm shilling for them), who has used market timing since 1972 and over the past 40 years have consistently outperformed the market using three fairly simple market timing indicators. Not by leaps and bounds - generally just a few percentage points - but still outperforming.
[+] [-] bolu|14 years ago|reply
Comparing performance over any given time period to some arbitrary standard (in this case, "flat" is assumed to be bad) doesn't tell the whole picture. On the flip side, there are many instances where funds performed admirably but in a time when the markets as a whole did even better. Just as I wouldn't commend a fund for gains in a bull market, docking a fund or portfolio for being "flat" when the market as a whole was flat over the given time period is unfair.
I don't doubt that your company has outperformed (after all, you have the data and I don't). But for what it's worth, was it a slam dunk to assume 40 years ago that your company would have outperformed over the subsequent 40 years? That's the problem: picking winners before they're winners. Will you outperform for another 30 years (my own investment horizon?). Hard to tell.
[+] [-] gojomo|14 years ago|reply
[+] [-] bolu|14 years ago|reply
There's a whole other set of data to dive into whether or not professional managers as a whole are bad at market timing (spoiler: they are terrible at it), but you're right in that this is not what this data set addresses.
Really the point the article is making is that on the whole, you and I, the investing public, shouldn't try to "read the tea leaves" and pull our money in and out of the market / move money around in the market because of what we believe will happen in the near future. What this data proves is that this doesn't actually work.
[+] [-] PaulHoule|14 years ago|reply
[+] [-] bolu|14 years ago|reply
[+] [-] jpdoctor|14 years ago|reply
> On the x-axis we show the flow of capital into an asset class
For every buyer, there is a seller, so every single transaction nets out to exactly $0 flowing into the asset class. So exactly how does capital "flow" into or out of an asset class?
You would guess that they are confusing the change in valuation, but that is what they are using for the other axis.
[+] [-] noname123|14 years ago|reply
[+] [-] jonxu|14 years ago|reply
[+] [-] nirvana|14 years ago|reply
I'm sure some proportion of investors move money in and out of mutual funds based on their market expectations, but I do not believe they do so with a single months time horizon, nor do I believe they are the majority.
I submit that a potentially quite large proportion of investors-- especially in mutual funds-- are not doing this, but are instead moving money into those mutual funds each month as a percentage of their paycheck goes into their 401k or into their savings plan. Further, some large proportion is likely investors moving from one mutual fund to another, while others are cashing out because they retire, etc. I bet a check on your flow dates to see if its correlated with an annual peak in employment change dates (e.g.: what month are people most likely to retire, or change jobs?) might find some correlation.
Worse, if I'm reading it correctly, this plot shows the return of the asset class over the course of a single month.
This means that you're assuming that these mutual fund investors are trying to time the market on the scale of the expected return within a month. This seems implausible to me given that mutual funds are managed and thus not well suited for timing (a lot easier to time the earnings announcement of Google, for instance, than guess what moves a mutual fund manager is going to make in a particular month.)
I suspect that most investors who choose mutual funds have at least a years time horizon, if not 5 years or more.
[+] [-] bolu|14 years ago|reply
The premise isn't as clear-cut as what you laid out, in my opinion. In general people do start with the best of intentions; that is, their time horizon when they buy is usually something like "until I need the money". But, that's the generic case under stable market return conditions. In times of panic, folks who thought they were okay with risk find out they're not okay with it, and pull out (usually after much of the panic has already passed). In boom times people start to, like you said, "move money between funds" usually in a way that follows the recent price increases (gold recently, tech in 2001, etc). It's these movements that the article is written against - and you're right, it would have benefitted from a longer time series of returns.
Indeed, if you remember the oft-quoted Nasdaq Composite Index from the dot com boom - the level in 1998 was the same level as in 2002, but in the interim investors as a whole lost billions. That's a lot more than would have been lost if people had just regularly been investing the same amount each month into their 401(k), which is what we wish would have happened.
[+] [-] publicus|14 years ago|reply
[+] [-] bolu|14 years ago|reply
And of course, as for picking the "right" fund which will subsequently do that - well that's the hard part.