nedsaid wrote: ↑Thu Dec 01, 2022 10:26 am
If you can keep returns about the same but reduce volatility, over long periods of time this can make a sizeable positive difference.
Then you haven't actually kept the assumed returns about the same.
The mathematical models you are talking about basically do that--implicitly assume higher returns for the less volatile asset mix, then "reveal" this assumption as they do the math involved.
The effect of what Vince is recommending is that US Treasuries are an excellent diversifier to stocks most all of the time. So for example, during the 2008-2009 financial crisis, US Treasuries were up when most everything else was down. You repeat that over many bull markets over a lifetime and this seesaw effect of Treasuries being up while Stocks are down makes a difference.
I would personally suggest that long-term
nominal Treasuries are not a reliable diversifier for stocks.
And actually, depending on what you mean by "diversifier", there may not be a strong case for anything being a predictably useful diversifier when you already have a globally-diversified stock portfolio. DFA seems to think maybe short-term global bonds could count, and I can understand why they might think that. But I would see it as a weak case.
Personally, I think the case for starting to phase in long-term
inflation-protected bonds a good length of time before expected retirement is not that they are reliable stock diversifiers, it is rather that they are an appropriate "risk-free" asset for such investors. Although not really fully risk-free, so call it more "minimized-risk". And there are reasons to believe some "de-risking" as retirement starts to approach can be helpful.
I note there is a liability matching version of this argument too. To me, it really mostly amounts to the same thing.
What used to non-correlate, correlated when things went really bad during 2000-2008.
Right, in the real world, correlations tend to vary a lot, and therefore you can easily encounter scenarios where the correlations you were hoping for don't materialize, or indeed reverse.
But if you can successfully reduce the volatility drag effect, it can have a big effect over time.
Again, this is mathematically equivalent to saying if you could predict the magic asset allocation formula that will beat the long-term returns on globally-diversified stocks over your investment period, that would be a good formula for you to use.
Which is true given that premise. But to my knowledge, there is really no such magic formula that will predictably have that effect. Many alchemists, here and elsewhere, have tried to identify one, through backtesting and such. But they don't appear to actually do any better out of sample than random luck would suggest. Which is what one would expect because their backtesting protocol is basically a form of what is also known as overfitting, and also because relevant financial markets, financial systems, legal systems, macroeconomies, and so on are constantly evolving. So, there is no consistent underlying mechanism that can be reliably investigated through a procedure like backtesting.
Oh well.
Bond Index funds and Core Bond funds are not the worst thing ever. They have their flaws but for the most part, 2022 as a big exception, did their job.
I'm just repeating myself, but this is basically just insisting that because there hasn't been a sustained bad scenario for nominal USD bonds in the last few decades, we can now safely disregard the risk that such a scenario will happen in the next few decades.
If that reasoning is compelling to you, I am sure nothing I can say will persuade you otherwise.
But to me, it is obviously an invalid way to reason.
Second, asset classes in a crisis don't always act as expected. . . . Fourth, each bear market is different, the causes are different and the ways to hedge them are different each time.
So the way I would put it is there are many, many different sorts of potential crises that can affect financial markets. In any given crisis, given the exact nature of the crisis, usually the behavior of the financial markets involved makes some sort of sense. But the problem is predicting in advance which sort of crises you are going to experience, and really no one can do that. In that sense, I very much agree with your fourth point, and think it really subsumes your second point.
And unfortunately, it tends not to be possible to hedge against all possible crises. Meaning if you try, you end up either cancelling out your hedges, and/or increasing some different kind of risk.
However, what we can still do is hedge against some of the
worse possible crises, given the total context of our financial planning. We can't hedge against the
worst possible crises, because those are "all bets are off" sorts of situations anyway. But we know some of the really bad but not total societal collapse scenarios that can happen, and we can specifically hedge against those.
Although for long-term retirement savers still deep in accumulation, a globally-diversified stock portfolio is already a really good start on such a strategy. Maybe a sufficient one, at least until retirement gets within a reasonably-moderate investment horizon.
Third, ALL asset classes have their unique risks, it often takes a bear market to expose them. Not sure all of the unique risks of US Treasuries are fully known.
Not sure what you have in mind with Treasuries--like, we know they have default risk, it is just assumed that risk is rather low--but I agree (as I did above) that there is no such thing as a truly risk-free asset. Moreover, risks are contextual, meaning the risks that logically matter most to you as a specific personal investor can be different from a different personal investor, and very much different from other types of investors like governments, corporations, universities, and so on.
Of course that is a large part of why Total Bond makes very little sense as a default investment. The bonds in question are not being bought just by personal investors like you. They are not in fact just being bought by personal investors collectively. They are being bought by all sorts of different entities, with their own idiosyncratic risk management strategies. And meanwhile, bond investors, including perhaps personal investors actually like you, are also buying fixed-income not included in Total Bond.
OK, so if you are buying bonds in an effort to manage the risks that you, as a specific personal investor, are logically most worried about, it makes no particular sense to assume Total Bond is the right choice for you. It could entirely exclude, in fact, the most helpful sorts of fixed income assets for a person in your particular situation.
I also get that 1982 through 2019 were a great period for Bonds. Disinflation and an almost 40 year trend of declining interest rates. This made a lot of Bond types look really good and masked some weaknesses.
I'd say the period in question starts before that once you add in the term of intermediate or longer bonds, but right, this was a generally favorable period for nominal USD bonds.
But still, they didn't actually do anything all that helpful for long-term investors deep in accumulation. Meaning you would have done better still just with stocks.
And then if you don't believe the next few decades must be equally favorable for nominal US bonds, if you instead think they face the same sorts of risks that materialized right before the beginning of that period, and have materialized in many other times and places . . . people relying on that observation about the recent past of nominal USD bonds will strike you as clearly violating the principle that you shouldn't assume recent good performance predicts future performance.
But others seem to think it makes perfect sense to rely on that observation for future planning purposes. So, there we are.