Jeff, I think the inescapable problem is that it’s too hard to overcome the 200 bps in fees, when you add the expense ratio and the foregone dividends.
Yep, agree it's a high price for insurance when you could add cash/bonds and achieve a similar risk/reward. That's among the reasons why they're not the right choice for me and I'd hesitate to recommend them to others. That said, the reason I thought it was important to try to estimate dollar-weighted returns is because that's probably the strongest argument you could make for these, ie that while they're imperfect they might confer some benefits in keeping people in their seats, so to speak, where they'll earn a better return on their dollar than they would if left to invest separately in stocks and bonds. The counterfactual is that investors don't appear to have struggled to use allocation funds effectively -- the dollar-weighted gaps have been smaller there. But on the flipside you could argue that perhaps that's so because the weaker hands have opted for or been put into vehicles like these where they're less prone to do harm to themselves. I don't know if it's the best analogy but you could probably make a strong theoretical argument that retirees shouldn't annuitize given that if they're reasonably well-funded at the beginning they should be in a position to let their investments do the work for them. But I think there's some other research that finds that annuities can be helpful in some circumstances in giving retirees more 'permission to spend' safe in the knowledge that they've got a guaranteed income stream. No guarantees in this case but greater assurance you'll earn an outcome within a range and so for some it'll be worth it and lead to a better final result than perhaps it might have appeared on paper.
Good question. In all likelihood I’d recommend the bond but it would depend on the terms of the buffer. From what I’ve seen the buffers have worked over the outcome horizon even if w benefit of hindsight a trad allocation would have worked just as well (but absent that hindsight would have courted risk of a larger drawdown).
Jeff, thanks for this analysis. I accept that investors get what is promised to them provided they buy and hold for the ETF’s defined interval period. But, on balance, for an investor with a long (20- or 30-year) horizon, would they be better off in a buffer ETF or a 60/40 allocation?
With that sort of time horizon I don’t think these are appropriate products. By definition those investors don’t need the sort of short term assurance (w re to range of outcome) that these products aim to confer. However for someone who has a shorter time frame and can’t afford the risk of a deeper drawdown in a trad allocation then I think you could make a case for these, albeit with caveats.
Jeff, I think the inescapable problem is that it’s too hard to overcome the 200 bps in fees, when you add the expense ratio and the foregone dividends.
Yep, agree it's a high price for insurance when you could add cash/bonds and achieve a similar risk/reward. That's among the reasons why they're not the right choice for me and I'd hesitate to recommend them to others. That said, the reason I thought it was important to try to estimate dollar-weighted returns is because that's probably the strongest argument you could make for these, ie that while they're imperfect they might confer some benefits in keeping people in their seats, so to speak, where they'll earn a better return on their dollar than they would if left to invest separately in stocks and bonds. The counterfactual is that investors don't appear to have struggled to use allocation funds effectively -- the dollar-weighted gaps have been smaller there. But on the flipside you could argue that perhaps that's so because the weaker hands have opted for or been put into vehicles like these where they're less prone to do harm to themselves. I don't know if it's the best analogy but you could probably make a strong theoretical argument that retirees shouldn't annuitize given that if they're reasonably well-funded at the beginning they should be in a position to let their investments do the work for them. But I think there's some other research that finds that annuities can be helpful in some circumstances in giving retirees more 'permission to spend' safe in the knowledge that they've got a guaranteed income stream. No guarantees in this case but greater assurance you'll earn an outcome within a range and so for some it'll be worth it and lead to a better final result than perhaps it might have appeared on paper.
So if we are talking about someone with a five-year horizon, would you choose a defined-buffer ETF or a 4% yield (roughly) on Treasurys?
Good question. In all likelihood I’d recommend the bond but it would depend on the terms of the buffer. From what I’ve seen the buffers have worked over the outcome horizon even if w benefit of hindsight a trad allocation would have worked just as well (but absent that hindsight would have courted risk of a larger drawdown).
Jeff, thanks for this analysis. I accept that investors get what is promised to them provided they buy and hold for the ETF’s defined interval period. But, on balance, for an investor with a long (20- or 30-year) horizon, would they be better off in a buffer ETF or a 60/40 allocation?
With that sort of time horizon I don’t think these are appropriate products. By definition those investors don’t need the sort of short term assurance (w re to range of outcome) that these products aim to confer. However for someone who has a shorter time frame and can’t afford the risk of a deeper drawdown in a trad allocation then I think you could make a case for these, albeit with caveats.