5 ms·

There are two rational justification for the (seemingly) persistent value factor: 1. It’s just a risk premium. The efficient market hypothesis says that all ri

by mruts 7y ago

There are two rational justification for the (seemingly) persistent value factor: 1. It’s just a risk premium. The efficient market hypothesis says that all risk adjusted returns are the same. i.e there is only one Sharpe ratio. Or put another way, there is no excess return without commensurate risk. Using this model, value out performs because of the increased riskiness of the investment. Reward for bearing risk is absolutely persistent so if this first reason is true, it should be persistent forever. 2. Behavioral bias. If this is true, people will wise up and the eventual equilibrium will price value fairly at some point. Either way, value shouldn’t be able to juice your Sharpe ratio forever. Either the first one is true and the Sharpe ratio is unaffected (just the mean returns, which don’t really matter unless you are unable to access leverage), or value is in fact not persistent, because no behavioral bias can ever be persistent, a priori. I’m not sure what the answer is, but value alone isn’t worth very much. AQR and other funds usually pair it with factors that have a negative correlation, such as momentum. This allows them to capture significant excess returns, compared to value alone.

kgwgk 7y ago
> no behavioral bias can ever be persistent, a priori Of course it can. Tripping twice over the same stone, etc. Casinos are full of people playing games with negative expected return after all. And if the stock market is like gambling, it’s mostly on the side of “glamour” (as opposed to “boring”) stocks.
- mruts 7y ago
  Well, you would expect the markets to reward the people who didn't have these biases, and hurt the people who still had them. The people with the bias will get poorer and poorer, and the people without it will get richer and richer. Those who get poorer and poorer will increasingly have less influence on the world economy, and those getting richer and richer will have out-sized influence. Eventually, the pool of suckers will disappear and the market will move onto some other bias to stamp out. You can see this very thing happening in capital markets. I work as a quant and many very simple strategies like value, momentum, quality, etc used to work remarkably well, just 15-40 years ago. Nowadays, you need more sophisticated factors, or combine these simple factors together in unique and interesting ways. There might be a sucker born every minute, but the sucker will never accumulate enough capital for anyone to ever care about him.
  - kgwgk 7y ago
    It's too early to declare that value is dead, a full market cycle will be needed at least. Looking at the Fama/French HmL series value strongly underperformed growth in the late nineties (the HmL series had a 40% drawdown). The current drawdown is not as large yet (but it's true that it has been going on for over a decade). "Are Growth and Value Dead? "1999 was the sixth consecutive year in which the S&P BARRA Growth Index outperformed the Value Index. Some investors have questioned whether Value will ever again be a successful investment approach. Some of them believe this is a “New Era” in which technology stocks are revolutionizing the way business is done: “New Economy” stocks will survive while “Old Economy” stocks (mostly Value stocks) will become extinct." https://www.northinfo.com/documents/111.pdf https://www.northinfo.com/documents/111.pdf Edit: see also pages 7/8 in https://www.yardeni.com/pub/style.pdf https://www.yardeni.com/pub/style.pdf Those charts confirm what I wrote above, growth crushed value 20 years ago but didn't quite kill it...
    - mruts 7y ago
      Current research from AQR seems to suggest that HmL isn't actually a persistent factor, just a behavioral bias without any economic basis (anymore?). Also current research by AQR suggests that that value is more structural, and is a real risk premia. I don't think anyone really knows whether value is a real risk premia, and if it's reason is structural or not. I think it's possible that many of the classic factors (value, momentum, carry, reversal, quality) are structural and do exist, but it's going to take a little bit more work than just creating a dollar neutral L/S portfolio exposed to E/P or whatever. If I had to guess, I would say that value is a real risk premia. That's not to say I would have put my money in it for the last 10 years, though..
      - kgwgk 7y ago
        > Current research from AQR seems to suggest that HmL isn't actually a persistent factor [...] Also current research by AQR suggests that that value is more structural, and is a real risk premia. I'm lost. HmL is the classic "value" factor (as in the 3-factor model of Fama and French: market, size and value). I don't say that's the best definition possible (clearly it's not) and I know other definitions exist (but HmL has the advantage of being available for a century). For what it's worth, I prefer cash-flow-based metrics and total shareholder yield (dividends + buybacks + debt reduction). AQR use an improved definition (using data as current as possible) but I don't think it's terribly different: "The bottom line is that while the standard approach to value was a reasonable and conservative choice that has served the field well, it is not the best possible choice." https://www.aqr.com/Insights/Research/Journal-Article/The-Devil-in-HMLs-Details https://www.aqr.com/Insights/Research/Journal-Article/The-De... In their latest publication ("Factor Premia and Factor Timing: A Century of Evidence", that I've not yet read) they say: "We follow simple value measures used in the literature to capture “cheap” versus “expensive” securities within an asset class. For individual equities, we use the book-to-market ratio following Fama and French (1992, 2012) and Asness, Moskowitz, and Pedersen (2013). For global equity indices, we use the aggregate 10-year cyclically-adjusted price-to-earnings ratio CAPE (value-weighted average P/E ratio for all constituent firms in the index)."
whatshisface 7y ago
I'll toss in a third: 3) A rational actor with $1 to their name can't eliminate mispricings because it takes more money than that to move prices. Just because you aren't the first person to show up with knowledge of a mispricing doesn't mean that it has been completely eliminated by the time you get there - those genius hedgefund managers that no mortal can beat may not have enough capital to price every stock in the entire market at exactly what they think it should be. >unless you are unable to access leverage It's also worth mentioning that you have to pay for loans. "Organic" risk that comes from buying risky securities doesn't have interest, while "synthetic" risk does.

by mruts 7y ago

kgwgk 7y ago

> no behavioral bias can ever be persistent, a priori Of course it can. Tripping twice over the same stone, etc. Casinos are full of people playing games with negative expected return after all. And if the stock market is like gambling, it’s mostly on the side of “glamour” (as opposed to “boring”) stocks.

mruts 7y ago
Well, you would expect the markets to reward the people who didn't have these biases, and hurt the people who still had them. The people with the bias will get poorer and poorer, and the people without it will get richer and richer. Those who get poorer and poorer will increasingly have less influence on the world economy, and those getting richer and richer will have out-sized influence. Eventually, the pool of suckers will disappear and the market will move onto some other bias to stamp out. You can see this very thing happening in capital markets. I work as a quant and many very simple strategies like value, momentum, quality, etc used to work remarkably well, just 15-40 years ago. Nowadays, you need more sophisticated factors, or combine these simple factors together in unique and interesting ways. There might be a sucker born every minute, but the sucker will never accumulate enough capital for anyone to ever care about him.
- kgwgk 7y ago
  It's too early to declare that value is dead, a full market cycle will be needed at least. Looking at the Fama/French HmL series value strongly underperformed growth in the late nineties (the HmL series had a 40% drawdown). The current drawdown is not as large yet (but it's true that it has been going on for over a decade). "Are Growth and Value Dead? "1999 was the sixth consecutive year in which the S&P BARRA Growth Index outperformed the Value Index. Some investors have questioned whether Value will ever again be a successful investment approach. Some of them believe this is a “New Era” in which technology stocks are revolutionizing the way business is done: “New Economy” stocks will survive while “Old Economy” stocks (mostly Value stocks) will become extinct." https://www.northinfo.com/documents/111.pdf https://www.northinfo.com/documents/111.pdf Edit: see also pages 7/8 in https://www.yardeni.com/pub/style.pdf https://www.yardeni.com/pub/style.pdf Those charts confirm what I wrote above, growth crushed value 20 years ago but didn't quite kill it...
  - mruts 7y ago
    Current research from AQR seems to suggest that HmL isn't actually a persistent factor, just a behavioral bias without any economic basis (anymore?). Also current research by AQR suggests that that value is more structural, and is a real risk premia. I don't think anyone really knows whether value is a real risk premia, and if it's reason is structural or not. I think it's possible that many of the classic factors (value, momentum, carry, reversal, quality) are structural and do exist, but it's going to take a little bit more work than just creating a dollar neutral L/S portfolio exposed to E/P or whatever. If I had to guess, I would say that value is a real risk premia. That's not to say I would have put my money in it for the last 10 years, though..
    - kgwgk 7y ago
      > Current research from AQR seems to suggest that HmL isn't actually a persistent factor [...] Also current research by AQR suggests that that value is more structural, and is a real risk premia. I'm lost. HmL is the classic "value" factor (as in the 3-factor model of Fama and French: market, size and value). I don't say that's the best definition possible (clearly it's not) and I know other definitions exist (but HmL has the advantage of being available for a century). For what it's worth, I prefer cash-flow-based metrics and total shareholder yield (dividends + buybacks + debt reduction). AQR use an improved definition (using data as current as possible) but I don't think it's terribly different: "The bottom line is that while the standard approach to value was a reasonable and conservative choice that has served the field well, it is not the best possible choice." https://www.aqr.com/Insights/Research/Journal-Article/The-Devil-in-HMLs-Details https://www.aqr.com/Insights/Research/Journal-Article/The-De... In their latest publication ("Factor Premia and Factor Timing: A Century of Evidence", that I've not yet read) they say: "We follow simple value measures used in the literature to capture “cheap” versus “expensive” securities within an asset class. For individual equities, we use the book-to-market ratio following Fama and French (1992, 2012) and Asness, Moskowitz, and Pedersen (2013). For global equity indices, we use the aggregate 10-year cyclically-adjusted price-to-earnings ratio CAPE (value-weighted average P/E ratio for all constituent firms in the index)."

whatshisface 7y ago

I'll toss in a third: 3) A rational actor with $1 to their name can't eliminate mispricings because it takes more money than that to move prices. Just because you aren't the first person to show up with knowledge of a mispricing doesn't mean that it has been completely eliminated by the time you get there - those genius hedgefund managers that no mortal can beat may not have enough capital to price every stock in the entire market at exactly what they think it should be. >unless you are unable to access leverage It's also worth mentioning that you have to pay for loans. "Organic" risk that comes from buying risky securities doesn't have interest, while "synthetic" risk does.