When reading a post with a keyboard, you can type the keys , and . to move backwards and forward between posts! You can also press 'return' to read through posts one at a time. There's freedom in Shrewd'm!
- Manlobbi
Halls of Shrewd'm / US Policy
No. of Recommendations: 5
No. of Recommendations: 20
I did like this bit : )
By the end of June, Berkshire held a whopping $234.618 billion in Treasury bills. That’s more than the U.S. Federal Reserve's Treasury bill holdings
Jim
No. of Recommendations: 2
What do members here think T-Bills, the broad stock market index and Berkshire's equity portfolio will return over the next 5-10 years?
In nominal terms I would guess something like 4.5% for T-Bills (the current yield), 3.5% for the S&P 500 (the trailing twelve months earnings yield) and 5.5% for Berkshire Hathaway's equity portfolio (S&P 500 plus 2 percentage points). If this is more or less correct, then a very high allocation to T-Bills makes sense.
If I'm allocating based on my estimated returns for the next 30 years instead of 5-10 years, then I would up the estimated S&P 500 return to maybe 8% (historical return of 10%, minus 2 percentage points for reversion to the mean, trailing P/E of 17). In that case my allocation to T-Bills would be quite low.
What is your thinking? Even if we don't want to give numbers out loud, we are still making bets in our portfolio allocations. I hope we all guess right, or what's right for us.
No. of Recommendations: 3
I've posted this before. It cites a Kelly analysis regarding asset allocation between stocks and T-Bills.
"Some here might like enjoy this paper on the Kelly criterion as applied to stocks.
https://sites.math.washington.edu/~morrow/336_18/2...As you know, the Kelly ratio is the fraction of one's money that when bet, or the fraction of one's portfolio that when invested, mathematically gives the highest rate of return over time. The result is found at the bottom of page 8. According to this analysis, for a portfolio of stocks and T-Bills the optimal fraction is
f = (expected return of the stock market - expected return of T-Bills)/(standard deviation of stock market)^2
The year-to-year standard deviation of the return of stocks is about 0.2, so the result is
(expected return of stocks - expected return of T-Bills) ---> optimal fraction of portfolio in stocks
4 ---> 100%
3 ---> 75%
2 ---> 50%
1 ---> 25%
0 ---> 0%
For example, if your portfolio consists of T-Bills and an S&P 500 index fund, and if you think that the S&P 500 will return 3% over some period of years and that T-Bills will also return 3%, then you want to invest 0% in the S&P 500 and 100% in T-Bills. If your portfolio consists of T-Bills and BRK/B, and if you think that BRK/B will return 6% over some period of years and that T-Bills will return 3%, then you want to invest 25% in T-Bills and 75% in BRK/B (actually slightly more in BRK/B since the standard deviation of BRK/B is slightly less than the standard deviation of the S&P 500).
This analysis contains some simplifying assumptions, but qualitatively it looks reasonable."
No. of Recommendations: 0
By the end of June, Berkshire held a whopping $234.618 billion in Treasury bills. That’s more than the U.S. Federal Reserve's Treasury bill holdings
It's fun to consider, but it's an almost meaningless point. Because the Fed has been rolling off assets for a year plus now, and because "bills" only have terms up to a year, it's natural that those roll off the balance sheet the quickest. And sure enough, the Federal Reserve holds more than 3.7 TRILLION (as of Oct 23) of "notes" and "bonds" which have terms between one year and ten years for notes and between ten years and 30 years for bonds. But those are just terms used, all of them, bills, notes, and bonds are treasury debt, and all are backed by the same thing (full faith and credit of the USA) just with varying initial duration.
No. of Recommendations: 6
In nominal terms I would guess something like 4.5% for T-Bills (the current yield), 3.5% for the S&P 500 (the trailing twelve months earnings yield) and 5.5% for Berkshire Hathaway's equity portfolio (S&P 500 plus 2 percentage points). If this is more or less correct, then a very high allocation to T-Bills makes sense.
I don't know. What I do know is that we haven't had a "natural" recession for 15 years or so. Despite those who think that government+federal reserve have "cured" recession forever and ever, the odds are high that there will one sometime during the next 5 to 10 years. And when there is a recession, interest rates are dropped, especially very short-term interest rates like those on T-bills. So therefore I don't think there's a chance that T-bills will yield 4.5% for 5 or 10 years. Heck, the 26-week T-bill is ALREADY below 4.5% (and has been since the middle of September), and the 52-week T-bill has been below 4.5% for a few months now (IIRC, since August).
Now, as far as Berkshire returns go, I think it nearly all depends on the SEQUENCE of S&P500 returns. If the S&P500 muddles along, with single digit returns for all 5 to 10 years, then perhaps yes, it might sit on all that cash for quite a long time. But if the sequence has at least one LARGE drop (like 50% or close to it), then opportunities will surely arise such that deployment of large chunks of capital will happen. I don't know what those opportunities might be, but when any big problems and panic arise, opportunities suddenly appear like magic. For example a crisis of faith in banks can lead to some sweet preferred convertible deals. Of perhaps a crisis in commercial real estate could someday provide some sweet deals? And heck, even if many run of the mill good performing companies get caught in a widespread downdraft, and they will, then good investment opportunities are sure to follow for those who can move quickly. And those who own those ultra short-term T-bills are the ones who indeed can move fast. You don't see Berkshire buying any treasury notes (1-10 years) or treasury bonds (more than 10 years), do you?
No. of Recommendations: 1
In nominal terms I would guess something like 4.5% for T-Bills (the current yield), 3.5% for the S&P 500 (the trailing twelve months earnings yield) and 5.5% for Berkshire Hathaway's equity portfolio (S&P 500 plus 2 percentage points). If this is more or less correct, then a very high allocation to T-Bills makes sense.
If I'm allocating based on my estimated returns for the next 30 years instead of 5-10 years, then I would up the estimated S&P 500 return to maybe 8% (historical return of 10%, minus 2 percentage points for reversion to the mean, trailing P/E of 17). In that case my allocation to T-Bills would be quite low.
A lot depends on whether we get reversion towards more historically typical valuations. If you are assuming -2% annual as the S&P reverts to lower multiples over 30 years, I assume you think that the S&P is overvalued by about 45% (0.98^30-1) and 1.02^10-1). That sounds reasonable. But then, shouldn't you also expect that mean reversion over 5-10 years, also, taking the S&P's 3.5% earnings yield down to a total return of 1.5%? And leading to an even higher allocation to bonds?
dtb
No. of Recommendations: 1
And heck, even if many run of the mill good performing companies get caught in a widespread downdraft, and they will, then good investment opportunities are sure to follow for those who can move quickly. And those who own those ultra short-term T-bills are the ones who indeed can move fast. You don't see Berkshire buying any treasury notes (1-10 years) or treasury bonds (more than 10 years), do you?
I don't know how 10-year treasuries traded in previous downturns, say, in 2000 or 2008 or even in 2022, but it would seem to me that, logically, their prices should have held up perfectly well. In fact, with interest rates dropping across all terms, the value of a 10-year bond should have increased during a crisis. Just because you want to be able to take advantage of a swoon in stock prices, that doesn't mean you have to hold that 10-year treasury to term, especially if you are sitting on a big gain. Those long-term bonds are perfectly liquid.
Does anyone have any data on how the longer-termed bonds held up in previous crises?
dtb
No. of Recommendations: 0
"Does anyone have any data on how the longer-termed bonds held up in previous crises?"
One can calculate it from charts of interest rates, but there's probably an easier way.
No. of Recommendations: 1
"But then, shouldn't you also expect that mean reversion over 5-10 years, also, taking the S&P's 3.5% earnings yield down to a total return of 1.5%? And leading to an even higher allocation to bonds?"
If the stock market P/E falls to 17 over 10 years, that will knock more than 2 percentage points off the return. It will knock about 6.7% percentage points off the annualized return. Ouch.
No. of Recommendations: 0
(expected return of stocks - expected return of T-Bills) ---> optimal fraction of portfolio in stocks
4 ---> 100%
3 ---> 75%
2 ---> 50%
1 ---> 25%
0 ---> 0%
(10.5% - 4.5% [google AI hit]) = 6.0
So, 100% stocks.
There is little reason to estimate that the S&P500 return will be other than the long-term average. This is not Lake Wobegone.
No. of Recommendations: 8
There is little reason to estimate that the S&P500 return will be other than the long-term average.
There is a plethora of posts with data suggesting that this assumption is not to be taken at face value, at least not for the expected return for the 10 year time horizon. I can appreciate that you might discount those posts, and maybe conclude that there is always someone making the argument, and that the long term average is as good a guess as any. I can appreciate that since its a risk that we are all free to own. And frankly returns for those who took the risk have turned out well for some time.
But a counterview with plenty of reasons and data behind the logic of the position has been proposed. The sum total of which has been a lot more than "little reason" to we will be on the underside of the long term average rather than the upperside of the long term average over the next ten years. Any reasoning to think we are in for a period of achieving the long term average rather than substantial underperformance?
No. of Recommendations: 4
But a counterview with plenty of reasons and data behind the logic of the position has been proposed. The sum total of which has been a lot more than "little reason" to we will be on the underside of the long term average rather than the upperside of the long term average over the next ten years. Any reasoning to think we are in for a period of achieving the long term average rather than substantial underperformance?
Basically, the future is unknown and REALLY hard to predict.
“It is thus because it has always been thus” - Jack Vance
"Prediction is very difficult, especially about the future." — Niels Bohr
"The smarter someone is the more plausible a story they can come up with."
“the stock market has predicted nine out of the last five recessions” - Paul Samuelson
"True bear markets start slowly giving many months to get out. It does not make sense to try to sell now before any indicator is triggered..." -- Newfound Research
"I don't need to know what's going to happen next. I just need to know what I'll do in response to whatever does happen." -- Jim Rogers
Also: "The smarter someone is the more plausible a story they can come up with."
========================
Any reasoning to think we are in for a period of achieving the long term average rather than substantial underperformance?
Okay, what can you do about it? If we do have a length time of substantial underperformance, what can anybody do about it? It is what it is, and we just have to take what comes. There is NOTHING that you or I or any of us can do to take anything other than the market gives us. Doesn't matter if we don't like it. All we can do is try to shift our strategies around to try to sit out the storm. But there is no reason to run and hide now, before this period of substantial underperformance comes into existance. After all, it might NOT happen. It might be that the current OUTperformance just deflates and thing go back to normal.
So, it's easy enough to have your plans in place, ready to go when/if the storm comes. It won't take too long to sell everything and go to cash, if that's your plan. Even if you had 100 stock holdings you could sell it all in an hour or two. But, since bear markets start slowly, you don't have to sell that rapidly.
