Of course I'm not claiming that past performance predicts future results.
In contrast your example relies on the explicit assumption that future results of the 2 proposed portfolios will be identical before fees.
I will bet you any amount of money that 10 years from now the before expense (and after expense ratios for that matter) returns of the example portfolios will not be identical.
As to TLH being a gimmick, I can only tell you that in less than 6 months I have personally saved enough in taxes to cover more than 2 years expense ratios. That's real savings and it was effortless. So experience tells me that it is not a gimmick. For me it was unequivocally cheaper than vanguard this year.
Besides all of this betterment is a better user experience than vanguard and even if you take away the betterment fees and features, I believe their portfolio to be far superior. (Though this is clearly a matter of personal preference....I'm a believer in the size and value premia, and modern portfolio theory. To each his own.)
As to your pictorial analysis, of TLH it ignores the slice and dice nature of the Betterment portfolio. As an example Emerging markets would have been harvested last year in betterment despite the overall growth of the portfolio, with Zero opportunity in the vanguard all in one fund.
Finally note that you could not harvest in only 3 of the 14 years in the Betterment graphic, and you can carry forward losses from earlier years to offset income in future years. So one would have had to be very unlucky indeed not to be able to take advantage of a 50K, lump investment in a betterment stock bond portfolio.
In fact even if you had started in 2005, you would have been able to take a loss in 07 and 08 based on the size of the bear market.
This years 7% sub-correction was not a rare event, in fact the past few years have been abnormally lacking in volatility if anything. (QE perhaps?)
And if you continue to contribute to your taxable account (as most investors do), your ability to really TLH never expires.
You're missing the point. I understand they won't be identical, but I can't show someone how fees affect the account, if more than one variable is being considered.
I just read the fine print on that Betterment chart, and it's worse than I thought. Now it makes sense how they were able to tax loss harvest over this time, the portfolio, "deposits twice a month that start at $750 and increase annually by 5%". Over that 14 year period, the additional contributions totaled $364,423. Significantly more than the $50,000 initial deposit. This is the only way to take advantage of the volatility during this time period, as you can't tax loss harvest the same dollar again, unless the previous low is breached.
I don't doubt that tax loss harvesting is possible in the short term. If you expect each individual ETF in your portfolio to rise over time, however, it will not be there for the long term. But the higher ER will. Since the ER is percentage based, this is easily hidden, but let's see if I can highlight it. Milesdividendmd has saved enough in taxes to cover 2 years of fees. I presume he means the additional 0.15 ER that Betterment adds on, so let's make it 4 years of fees just in case. let's see what that looks like with a $100,000 deposit, and the worst case scenario (hehe), each ETF individually has an 11.5% yearly growth:
Year 1: $100,000 deposit, market goes up 11.5%, but is volatile enough that tax loss harvesting got you a $600 gain -$310 fee = $111,754
Year 2: 11.5% gain across all ETFs - Account is now $124,220 (after 0.0031% fee), you can no longer tax loss harvest, as there are no losses to harvest. Current year fee on $124,220= $386
Year 3: 11.5% gain across all ETFs every year from here on out - Account is now $138,076, with a $429 fee
Year 4: $153,477 with a $477 fee
Year 5: $170,597 with a $530 fee
Year 6: $189,626 with a $590 fee
Year 7: $210,777 with a $655 fee
Year 8: $234,288 with a $729 fee
Year 9: $260,421 with a $810 fee
Year 10: $289,469 with a $900 fee (total fees so far = $5,817)
...
Year 20: $833,426 with a $2,592 fee (total fees so far = $22,673)
How is this easily hidden? It's easily hidden, because if you invest another $100,000 in the second year, and get another $600 tax gain that year, you'd be up big-time! As far as you can see, each $100,000 invested, nets another $600 gain in taxes. To visualize this, imagine each $100,000 deposit, as a separate container. It's really easy to see the gain when your taxes are reduced, but not so easy to see the ER fee. I suspect this is why many people think they are turning a profit from their mortgage interest tax deduction. This strategy is essentially laddering you into higher and higher fees, which will never tip back in your favor. After the second year that each $100,000 container stays in the portfolio, the fees overcome the initial $600 gain, and it's all downhill from there. You can hide this if you keep depositing, but after a few more years even the $600 gain from that year will be usurped by the overall fees.
Even if you deposit significantly more than your initial deposit (as in the Betterment chart above), that does not affect the fees on the initial $100,000 container, you're just digging yourself deeper. I charted this out in Excel, hopefully it makes sense:
That's a single $100,000 deposit. Column A is the account value each year, and Column B is the fee incurred that year, 0.31 ER. I included a +$600 boost the first year form TLH. Now let's see what that looks like laddered, if you make the same $100,000 deposit each year, for 6 years:
and calculated the fees incurred:
Long story short, for the OP who will be making a single deposit, the extra fee very much seems to outweigh the gains. For someone continually making deposits, you're just making it worse.
Note, I attached the excel sheet used to make the calculations. I'm sure I messed up somewhere, but the overall story shouldn't change.