Bloomberg has an article that takes a different approach to why bonds with duration are a bad way to go, making more of a fundamental case related to demography causing inflationary pressure leading to higher interest rates from here. If interest rates go up then bond prices will go down as a direct cause and effect. The further out a bond's maturity date, the more sensitive the price is to rising rates. A bond/note/bill that matures in a year or two has very little sensitivity to rising rates. It would take an extreme move up in rates to cause a big move in the price of a two year instrument, very extreme, but if that happened, the time needed to bail you out would be very short as opposed to be far underwater on an issue that matures in 2035 or 2040.
The argument I've been making has been simpler. The 40 year, one way trade is over. Bonds with duration are now more volatile than they used to be and that volatility is less reliable than it used to be making them a less effective diversifier for the equity portion of a diversified portfolio.
The article suggested allocating more to equities, perhaps emulating the Norwegian Sovereign Wealth Fund which the article said is 70% equities, 30% bonds. It also talked about going heavy into some sort of commodity exposure.
The point from this chart is that no matter what sort of allocation decisions you make with regard to what to do with the 40, as in the 40% traditionally allocated to bonds in a 60/40 portfolio, don't get too far away from a "normal" equity allocation. The Bloomberg article included a couple of quotes about dialing down the equity exposure in retirement which has been the default approach but that chart shows why dialing down is a bad idea.
The way we've talked about that here is to think about setting aside an amount of cash to correspond to some number of months of expenses you feel comfortable with. This is of course one way to mitigate sequence of return risk. The simplest example would be the person to retired at the end of 2007 and then 12 months later, the stock market was 39% lower. That person would be in some trouble without a cash buffer. With a cash buffer of two years or whatever, that sort of decline doesn't have to derail anyone's retirement plan. I don't think more than three years of planned spending set aside in cash is necessary because of how long bear markets tend to last before starting to recover but there doesn't have to be a single right or single wrong answer.
We spend a lot of time here trying to explore the what to do with the 40 question. We've generally found several different ways to avoid interest risk and the volatility that goes with it but course that doesn't ensure outperformance. What I believe these ideas do is remove a source of potential negative surprises. I can't imagine why investors, as opposed to traders, would buy fixed income and hope for a lot of unpredictable volatility. Again, we're just removing the volatility.
It is fun and interesting to spend time trying to improve what we've learned so far about what to do with the 40 and I do think there is more to learn. Part of how to go about this is to dig into cross asset relationships and how things blend together.
Portfolio 1, the blue line, combines a couple of negatively correlated fire crackers rife with tracking error yet blending them together looks a lot like an absolute return strategy as captured with the red line. The point is not to mimic Portfolio 1, no one in their right mind should put that portfolio on but the idea is instructive to think about how a fund/strategy that contributes to a bottom line result for a cohesive portfolio as opposed to having a collection of individual holdings that you hope all go up. When I worked at Fisher Investments way back when they talked about having a valid portfolio versus a collection of investments, there's a difference and collections are far less robust and resilient.
That brings us to an interview with Felix Zulauf in Barron's. As the comments pointed out, he's a perma bear but that doesn't mean there isn't something interesting to pull from the discussion. He talked about the version of the carry trade where you short yen to buy a higher yielding or riskier asset. Specifically he talked about shorting the yen to buy the NASDAQ 100 ETF (QQQ). This can be replicated with the ProShares Ultra Short Yen ETF (YCS) which is a 2x inverse fund.
The tracking error of YCS has consistently favored YCS which doesn't mean it will continue to do so. Where Portfolio 3 should zero out, it's pretty close to doing just that. In five of the backtested years, the change was less than 1% in either direction but in 2023 and 2024 the net change was in the three's.
The first two portfolios in this next chart show Zulauf's idea exactly as he stated it. The long term result is interesting, for so much in QQQ they have much lower volatilities than I would have guessed. Looking year by year, portfolios 1 and 2 lagged by a lot in 2016, outperformed by a lot in 2022 (by going down much less) and 2023. Portfolio 3 as possibly a more realistic implementation was pretty close to VBAIX most of the time, outperforming by a lot in 2022.
Digging in a little more on Portfolio 3, I used a min volatility ETF because of the extra beta that the portfolio gets from 20% into QQQ but by removing the fixed income volatility, we can have 70% in equities with less volatility than VBAIX. Between QQQ and YCS, I'm not sure that is a small slice to carry but I think Zulauf's idea helps the portfolio, at least looking backward, especially in the the last four years where YCS was up 22%, 29%, 28% and 23% respectively. If we stop the back test at the end of 2020, Portfolio 3 compounded at 9.64%, 50 basis points better than VBAIX but the standard deviation was only eight basis points lower. Still valid but inferior versus the longer term result that benefited from the recent shocking decline in the yen.
The risk here is that if YCS can go up twenty whatever percent four years in a row, it could go down that much too. There is a 2x long yen ETF with symbol YCL. Subbing that in to get the effect of four bad years in a row...well this result is surprising. With YCL, Portfolio 3 would have compounded at 8.91% versus 8.32% for VBAIX you see above and the standard deviation of the YCL version of Portfolio 3 was 9.74%. YCL was down in those four years, and while it benefitted a little from tracking error, the best year of the previous four was a drop 17.75%.
The annual returns for both YCS and YCL for the previous four years are huge for currencies. The yen could go on a long run of strength (not a prediction, merely outlining the risk) but that might not hold Portfolio 3 back that much based on the YCL example.
I am also intrigued by the idea that YCL and YCS, FXY is 1x long yen, appear to have no correlation to equities and a very low correlation to VBAIX which is a proxy for a 60/40 portfolio.
Portfolio 3 with YCS has some merit so let's compare it how to the type of portfolio we usually build for blogging purposes to see what that looks like.
The results are remarkably similar. Portfolio 3 sort of blends 1 and 2 together and you can see a slightly higher return with a slightly higher standard deviation.
Today's post doesn't solve anything, it starts something, it starts looking at whether currency can be used as a diversifier. At this point, we don't have an answer but I don't think it's an obvious "no."
The information, analysis and opinions expressed herein reflect our judgment and opinions as of the date of writing and are subject to change at any time without notice. They are not intended to constitute legal, tax, securities or investment advice or a recommended course of action in any given situation.
No comments:
Post a Comment