While perusing a recent note from Goldman assessing the odds of an equity market pullback predicated on near-term concerns about rising virus cases and what the bank called “extremely stretched” positioning, I ran across a couple of factoids that, while not surprising (and perhaps not even notable), I nevertheless wanted to highlight.
As a quick aside, what is and isn’t “notable” is subjective. In a world awash with data and against a backdrop characterized by a 24-7, nonstop news cycle, discerning what data points to highlight and what stories to cover is often more difficult than crafting the analysis itself. For example, how many election-related lawsuits deserve exhaustive media treatment? The easy answer is that all election-related lawsuits deserve dedicated coverage, especially those that involve the Supreme Court. But on the other hand, litigation related to the vote is almost all noise and no signal. What service have I really done readers by informing them that another lawsuit involving a conspiracy theory that nobody (including those promoting it) understands, was tossed aside by the courts?
The point is that while I can never know with certainty what readers will find valuable or consider “notable,” I do make an effort, and my experience over the past four years indicates that the vast majority of market participants aren’t apprised of most data points I choose to highlight for one simple reason: Gathering data is time consuming. I’m blessed with virtually limitless spare time. Most people aren’t.
Anyway, steering this back to where I began, consider that on Goldman’s estimates, total allocations to stocks are in the 97th%ile going back three decades.
Stocks account for nearly half of total assets among households, pension funds, foreign investors, and mutual funds. Between them, those four investor cohorts own 90% of the corporate equity market.
What about bonds? Well, as you can imagine, the allocations are low — very low. The aggregate allocation to bonds is just 22% on Goldman’s estimates, which ranks in just the 7th%ile since 1990. For households, the allocation to debt is just 18%, in the 5th%ile.
Cash, meanwhile, is just 13% of total allocations, but 16% of household assets. That latter figure sticks out. “US households represent the single largest owner of the US equity market, at 35%, and currently have cash allocations close to the 30-year average despite cash yields being substantially below average,” Goldman’s David Kostin wrote. (You’re looking at the percentile ranking for cash held by households denoted in green below.)
You can make of that what you will, and there are all manner of things you’ll be inclined to take into account including, but not limited to, elevated savings rates amid the pandemic and, relatedly, the panicked flight to cash which is still working its way off the proverbial “sidelines.”
But it’s interesting to contextualize the above using one of Ray Dalio’s answers from an “Ask Me Anything” event he convened on Reddit Tuesday.
“As for stocks, they compete with bonds. With bond interest rates where they are, bonds are trading at roughly 75x earnings,” Dalio wrote, to an inquisitive netizen.
“With the amount of money out there, and cash being such a bad alternative, there’s no good reason that stocks couldn’t trade at 50x earnings,” Ray went on to say, adding that “just like you might have never expected bond yields to be at or slightly above 0%, you might not be comfortable with these kinds of multiples for stocks [but] all investments compete with each other and where would you prefer?”
Ray Dalio On ‘Floods Of Money,’ Bitcoin, China, And Social Safety Nets
As a fiduciary and someone formerly deeply involved in the bond market, it is always curious when the risk guys and stock jockeys trash bonds. The purpose of bonds in a portfolio is usually three fold. To provide income, to reduce risk overall, and to provide diversification of assets. Many make it an all or nothing proposition. It isn’t. If you think bonds are rich you can reduce your allocation- same for other asset classes- whether it is stocks, cash, real estate, gold, bitcoin, collectibles what have you (although I admit I don’t have much use for Bitcoin as an asset class). But you don’t need to sell all of it- maybe the ideal portfolio is no longer 60/40 but that does not mean it is 100/0. And bonds come in many flavors as well. Has anyone looked at convertible bonds lately- that market has been a moonshot- although I would grant that its correlation to stocks – particularly tech and small to mid cap is high.
Hear, hear! I’ve been a bond buyer since the late 1960s and now in my middle seventies I still hold a portfolio that is, through various avenues — securities, CEFs, ETFs and mutual funds — roughly 65% fixed income. Staying mostly in bonds through my whole investing life I’ve done roughly as well as Buffett on a much smaller scale. No trash talk here. My first academic promotion and my tenure was built on research into to what makes bond portfolio managers tick. One of my students formerly ran all the US fixed income portfolios for a major Euro financial institution. Now he runs all the money for one of our fifty states. I’m all in here.
As I read this piece it occurred to me that a quirk in our accounting rules might actually be affecting the stock ownership data H was talking about here. Over the last decade, especially, the biggest buyers of stocks were US corporations engaged in buying back their own stock. However, when they do this they don’t create an asset investment reflecting these purchases. Instead they reduce their cash balance and instead of an offsetting asset, they balance the books by reducing their equity to reflect the lost cash by putting the stock in the “treasury” and deducting it from total equity. Because this stock is not considered an asset, per se, it doesn’t show up as being held by the companies that have bought it. It would be interesting to know how much stock is in those equity treasury stock accounts in total and how that would skew the ownership data shown here if that stock was considered in the totals.
In many, but not all, cases, aren’t a big percentage of buybacks simply offsetting options and share compensation?
The question one has to ask is if the low bond yields suggest some earnings trouble in the future. So the question of the “right” multiple might have to also consider the “right” earnings power. Rates may be able to be mis-priced based on true fundamentals for a while but over some time frame (It may be longer than I believe though) they may revert to a more “normal” level so therefore an investor will then need to create vol in the multiple used. PE multiple movements can be very difficult for analysts to handle. They tend to overshoot both ways. But staying focused on the North Star of true value tends to work pretty well (of course easier said than done).
When one is faced with not having money to throw away, how can owning an asset like bonds that pay less than the rate of inflation be seriously considered as desirable? And investing in an asset called a ‘junk’ bond doesn’t attract my attention either. Not to say that finding attractive yield among stocks that is relatively without risk is easy.