nedsaid wrote:I don't know, I have been "quantsplained" here many times by people who believe investing is an engineering problem or a math equation. Don't get me wrong, the math is helpful and vital, but it has limitations, mostly due to irrational humans and their behavior in the investment markets. Can't tell you how many times I was told that I didn't know what I was talking about.
Ah yes, the volatility drag thread. You were saying that a "factor tilted" portfolio benefits from diversification but a plain vanilla "untilted" portfolio, such as a three fund portfolio, does not benefit from diversification, since a person with a three fund portfolio is not even trying. Not everyone agrees. In such a discussion, I'm not sure who should be called the quantsplainer and who should be called the quantsplainee.
What I said was that factor tilted portfolios attempt to deal with the "problem" of volatility drag, if it even exists at all. In my earlier posts in the thread, I questioned whether the phenomenon even existed. The quants in the group said that volatility drag existed in a mathematical sense, and I sort of agreed. After all, even I had the desire for a smoother ride in my portfolio. I further said that if volatility drag exists, an investor might not be able to do anything about it in actual practice. The factor tilting is an attempt to address volatility drag by putting together volatile asset classes that don't necessarily correlate with each other, hopefully some will zig as the others zag, the result being a smoother ride in portfolio performance.
A three fund portfolio is diversified across asset classes but as Larry Swedroe might say, not diversified by factors. US and International Stocks tend to take turns outperforming each other and much of this is a currency effect. This would help, hopefully, to reduce volatility drag on a portfolio. Stocks and bonds usually don't correlate with each other and this would help reduce volatility drag, but since the returns of bonds are less over time than for stocks, the cost will be lower returns for lower volatility. In theory, the protection against volatility drag for this kind of portfolio would be less than for a factor tilted portfolio.
A factor tilted portfolio not only has the US vs International dynamic but also Large vs Small and Growth vs Value. A DFA portfolio adds a dash of Momentum so to a degree you get Momentum vs Value. So you have more pieces doing the zig vs zag thing and in theory you should get less volatility drag.
In many other posts, I further explained that factor tilting worked great during the 2000-2002 bear market but during the 2008-2009 bear market, most everything crashed and thus didn't work the second time.
So pretty much, I am back to where I started. First, I have doubts that volatility drag is a real phenomenon. It seems to be an observation that markets are unpredictable and volatile. It is the old observation that a 50% drop in price then requires a 100% increase from there to get back to even. Knowing that, we would all like to have market returns but with less than market volatility. Second, even if volatility drag is a phenomenon that can be observed using math, that in actual practice it is difficult to do anything about it.
My point was that I was a volatility drag skeptic, but even in my skepticism tried to do something about it with factor tilting. The person who argued so fervently that volatility drag was a real effect then said that a three fund portfolio was enough. My argument was that my debate partner chose to not address the problem. I couldn't help but see irony in that.
In summary, a factor tilted portfolio is diversified across asset classes and factors. A plain vanilla three fund portfolio is diversified across only asset classes. The essence of the disagreement is the efficacy of factors and if volatility drag is a real world phenomenon.
A fool and his money are good for business.