Taylor Larimore then said that he agreed with Laura's opinion about small value, although he suggested Total Stock Market in lieu of either Small Value or TM Small Cap (VTMSX). I figured I'd start a new thread rather than clutter that one up.In my opinion it is not worth holding small value in taxable. You can go for the small tilt with Tax Managed Small Cap but skip the value part unless you can hold it in a tax advantaged account.
It seems pretty clear to me that the general Diehard consensus (including Laura and Taylor Larimore) is that there are both size and value premia, and that one can expect to acheive greater return (and risk) by tilting to small value.
However, this raises problems of tax efficiency. Value stocks tend to have higher dividend yields, and VISVX--Vanguard's Small Value Index (where I invest my small value dollars)--has a large amount of REITs which are ineligble for the qualified dividend rate. Small stocks also tend to migrate up and out of the small-cap region, forcing managers to either realize capital gains or hold the stocks and reduce the fund's expected returns (since the size premium is lessened).
So the question is whether the expected returns of Small Value compared to TSM or Small Blend trump the tax problems. I looked at the distributions for VISVX and VTMSX. $10k invested in VISVX right at the time of each distribution last year would have resulted in $182.83 of income (ignoring reinvesting dividends, capital appreciation, etc.). The corresponding amount for VTMSX would be $74.76. Given my tax brackets (4% state and 15% federal--though I might even end up in the 10% bracket this year), that's only $20 difference in taxes paid. In other words, VISVX only needs to beat VTMSX by more than 20 basis points to come out ahead.
So will it? As the saying goes, "Past results are no guarantee of future performance," but looking at the backtesting spreadsheet shows that small value has averaged 146 bp better than small blend since 1972 (small value has also been less volatile). Also, as for the SV vs. TSM question, SV has outperformed by more than 4% a year bp on average, so I doubt any tax-efficiency problems would erase that gap.
In short, it seems almost impossible for the small/value premia to exist but be so tiny they aren't worth pursuing. I am sure there are arguments against a small/value tilt, but surely, if a premium exists, it must be more than few hundredths of a percent.
Any thoughts?