Can You Get A Meaningful Factor Tilt With Present Investment Products?

[Park]

https://www.factorresearch.com/research ... or-returns

"We’re going to focus on the largest smart beta ETFs in the US focused on Value and Growth. All of these have more than 10 years of trading history and at least $4bn in AUM. We calculate the excess returns for each smart beta ETF by deducting the benchmark returns from the smart beta returns. We also calculate the factor performance for Value and Growth by creating long-short portfolios comprised of the top and bottom 30% of the US stock universe. The portfolios are constructed dollar-neutral and include 10bps of transaction costs. The table below provides an overview of the 14 smart beta ETFs used in the analysis.

https://www.factorresearch.com/wp-conte ... a-ETFs.png

The next chart shows the excess returns of the Value smart beta ETFs compared to the Value factor performance. The factor is constructed by taking a long-short portfolio of stocks ranked by a combination of their book-value and price-earnings multiples, which is a fairly standard definition for the factor. We can observe that the factor generates much higher returns than any of the smart beta ETFs, which can be explained by the factor taking concentrated long and short positions while the smart beta funds have only overweights and underweights across all stocks in the universe. Investors who are familiar with Value factor returns and are using smart beta ETFs to harvest factor returns are likely to be disappointed by the excess returns from the ETFs."

https://www.factorresearch.com/wp-conte ... eturns.png

Edited to change the thread title from "Will You Get Increased Return From Your Factor Tilt?" to " Can You Get A Meaningful Factor Tilt With Present Investment Products?"

Edited a second time to include the following comment. The third link might initially give the impression that VBR and VTV have outperformed other value ETFs. However, the data for the other 5 value ETFs and the value factor (long/short) use a start date of 2000. The data for VBR and VTV is from 2004, when both funds were started.

[willthrill81]

Smart beta seems dumb.

If you want to tilt toward the 'mainstream' factors, you can do so easily and inexpensively without such funds. But no one knows if YOU will experience increased returns from tilting your portfolio over your investing time frame. And if you do tilt, we can say with a high degree of certainty that you will underperform the market at some point, and that underperformance can last for years, maybe a decade or longer. If you aren't prepared for that, then don't tilt.

YMMV.

[Park]

Smart beta seems dumb.

If you want to tilt toward the 'mainstream' factors, you can do so easily and inexpensively without such funds. But no one knows if YOU will experience increased returns from tilting your portfolio over your investing time frame. And if you do tilt, we can say with a high degree of certainty that you will underperform the market at some point, and that underperformance can last for years, maybe a decade or longer. If you aren't prepared for that, then don't tilt.

YMMV.

I agree with what you've written. However, the question is whether, with the investment products available to the retail investor, whether you will get a tilt that is of significance.

[willthrill81]

Smart beta seems dumb.

If you want to tilt toward the 'mainstream' factors, you can do so easily and inexpensively without such funds. But no one knows if YOU will experience increased returns from tilting your portfolio over your investing time frame. And if you do tilt, we can say with a high degree of certainty that you will underperform the market at some point, and that underperformance can last for years, maybe a decade or longer. If you aren't prepared for that, then don't tilt.

YMMV.

I agree with what you've written. However, the question is whether, with the investment products available to the retail investor, whether you will get a tilt that is of significance.

No one knows about the future.

With regard to the past, funds like Vanguard's small-cap value fund (VISVX) have generated higher returns than the market over their lifespan, in that particular case, by over 2% over the last 20 years.

[YRT70]

I agree with what you've written. However, the question is whether, with the investment products available to the retail investor, whether you will get a tilt that is of significance.

Wouldn't that be possible with a small cap value fund?

[Park]

With regard to the past, funds like Vanguard's small-cap value fund (VISVX) have generated higher returns than the market over their lifespan, in that particular case, by over 2% over the last 20 years.

Since its inception in May 21/98, $10K in VISVX would have grown on a pretax basis to $53,155. The comparable number for VTSMX (total US stock market index) is $40,283.

To calculate CAGR, I had to assume a 21 year period. For VISVX, that results in a CAGR of 8.28%. For VTSMX, that results in a CAGR of 6.86%. So an outperformance of 1.42%.

If you wanted to play devil's advocate, you could make a case that without the internet bubble of 1999, the outperformance may not have existed or even been underperformance.



Based on CAPE data since 1880, the only time that US stock have come close to the internet valuation peak is 1929 or today . The above link shows Shiller's older CAPE data in dark blue and his newer CAPE data in green.

VTSMX has a 30 day SEC yield of 1.97%. For VISVX, it's 2.24%. VISVX has a turnover rate of 17.6%. For VTSMX, it's 3.4%.

So VISVX, based on dividends, is slightly less tax efficient. But if the rules change, such that ETFs have the same exposure to cap gains as mutual funds, VISVX's tax efficiency would be more of an issue. Under those circumstances, Vanguard's mutual funds would not have the benefit of having ETF tax efficiency.

When you use portfolio visualizer on VISVX, loadings on market, size, value and momentum are 0.96, 0.53, 0.57 and -0.08 respectively.

The comparable numbers for VBR are 1.01, 0.58, 0.35 and -0.05 respectively.

The comparable numbers for IWD (Ishares Russell 1000 Value ETF, a large cap value ETF) are 0.93, -0.12, 0.36 and -0.01 respectively.


My impression is that a tilt to small cap value might result in outperformance of about 1.4%. This assumes that tax laws don't change. This also assumes that you use investment products with a relatively strong tilt.

With DFA small cap value funds, you might get more than 1.4%, but there is the added cost of advisor fees.

[Park]

I tried to add this to my previous post, but that didn't work.

https://www.factorresearch.com/wp-conte ... SP-500.png

The above link shows the 1 year correlations of three large value funds (IWD iShares Russell 1000 Value ETF, IVE iShares S&P500 Value ETF, VTV Vanguard Value ETF with the S&P500 from 2001 to 2017. It's basically 0.98 over the time period, except around 2001, when it got down to around 0.83.

[packer16]

You bring up some interesting points about factor investing vs. lets say simple index investing with Taylor's 3 fund portfolio. Successful factor investing is dependent upon said factor premium over your hold period but also a fund that can adequately capture the theoretical "factor" premium. I am not a factor expert but I think some of the DFA funds are said to have the most tilt/$ invested.

Packer

[willthrill81]

You bring up some interesting points about factor investing vs. lets say simple index investing with Taylor's 3 fund portfolio. Successful factor investing is dependent upon said factor premium over your hold period but also a fund that can adequately capture the theoretical "factor" premium. I am not a factor expert but I think some of the DFA funds are said to have the most tilt/$ invested.

Packer

But of course, the problem with DFA funds is that you need to earn a premium just to overcome the drag of the advisor fees needed to invest in DFA funds in the first place.

[packer16]

You bring up some interesting points about factor investing vs. lets say simple index investing with Taylor's 3 fund portfolio. Successful factor investing is dependent upon said factor premium over your hold period but also a fund that can adequately capture the theoretical "factor" premium. I am not a factor expert but I think some of the DFA funds are said to have the most tilt/$ invested.

Packer

But of course, the problem with DFA funds is that you need to earn a premium just to overcome the drag of the advisor fees needed to invest in DFA funds in the first place.

That is the corundum, fees vs. exposure to factors with fees being absolute and whether the premium shows up during your hold period uncertain.

Packer

[rascott]

You bring up some interesting points about factor investing vs. lets say simple index investing with Taylor's 3 fund portfolio. Successful factor investing is dependent upon said factor premium over your hold period but also a fund that can adequately capture the theoretical "factor" premium. I am not a factor expert but I think some of the DFA funds are said to have the most tilt/$ invested.

Packer

But of course, the problem with DFA funds is that you need to earn a premium just to overcome the drag of the advisor fees needed to invest in DFA funds in the first place.

That is the corundum, fees vs. exposure to factors with fees being absolute and whether the premium shows up during your hold period uncertain.

Packer

Merriman's Best in Class ETFs track the DFA funds very well. Which is what he has stated his goal to be....to offer a DFA like portfolio without the advisory fees.