February 27, 2012
– Comments (14) |
RELATED TICKERS: TLT
Investors should expect no more than a 2% (real) return per annum from U.S. stocks over the next ten years. Find out why in the latest issue of The Real Returns Report.
This is not the first piece on this either... I think it was a year ago or more that there was this WSJ piece that says investors should expect a net-net-net (return after inflation/fees/taxes) return of 2% per year. In the article, some famous investors like Bogle (of Vanguard fame) would gladly take a 2.5% annual return.
Thanks for your comment, SkepticalOx, and thanks for the link to the article, which is very interesting (I'm a fan of Jason Zweig, who is value-oriented and appropriately skeptical.) I was not familiar with his thought experiment -- it's a powerful idea.
In this video, which dates from last October, Bogle explains how he gets to a 7% (nominal) return. In fact, Bogle also uses the Shiller P/E, but he's not consistent when it comes to assessing its impact on returns.
For example, he says:
So those two reasons, a long-term focus and the right earnings number and reliance on history, I go with Shiller. And by his numbers, I think he has about a 17 times P/E as the norm, and it's around 21 times today. I wouldn't regard this data, being data, as a major imbalance, but it would suggest the market is probably more or less properly valued, correctly valued, or maybe somewhat overvalued.
I wouldn't qualify a market valued at a P/E10 of 21 versus a historical average of 17 (it's actually 16.4) as "more or less correctly valued, or maybe somewhat overvalued." Instead, I'd say it is at least somewhat overvalued.
Furthermore, in his return estimate "build-up" he only attributes a negative one percentage point impact from the change in the multiple. He seems to have pulled that number out of his hat! If you assume the multiple goes from 21 to 17 over a ten-year period, that equates to an annual penalty of 2.1 percentage points. In fact, his 7% total return estimate assumes no impact associated with the eanings multiple whatsoever.
That aside, Bogle is certainly one of the most sensible people out there, even if he has a hard time admitting that investors should always evaluate the impact of valuations on the merits of a "buy-and-hold" strategy.
I'm up 23% this year. I guess I should take the rest of the decade off...
I think PE10 is limited in that it ignores non mean-reverting, structural changes in the economy. Which can be significant.
Some people think market cap to GDP is a useful, mean-reverting number to value the US market. But look at Hong Kong, which might have had market cap to GDP of 50% back in 1800. Now it's 1200% or whatever. Is it going to revert back? No, there have been structural changes in their economy.
Not that I would buy the market at this price - or any price. I'm a long-short value investor. And cash looks more attractive than it did a few months ago.
Thanks for your comment, MegaShort.
That's not a limitation of the P/E10 -- the same criticism applies any time you try to compare any multiple value to its historical average.
I agree, though, that a mean-reversion framework won't work well with a non-stationary process. However, breaking down stock returns this way allows you to think more precisely about what type of impact a structural change might have.
Two potential shifts that come to mind immediately are:
- A dramatic change in corporate dividend policies. That could happen as a result of rising company cash balances, limited opportunities for growth investment and rising demand for income return from retiring baby-boomers.
- Speaking of which, demographic change could also have a significant impact on price-to-earnings multiples.
On that note, Barclays Capital's newly released 2011 Equity Gilt Study (one of the few really useful examples of sell-side research) suggests that demographic change will lower equity returns by one percentage point over the next ten years. An extra reason to be cautious when putting money at risk in the stock market.
Buffett is fond of the market value-to-GDP ratio. I don't think it's particularly useful because it doesn't account for changes in equitization of the economy.
Sounds about right. Check Crestmont Research's data for the last 100 years. After taxes and inflation, probably can squeek about 3.2% out of any given 20 year period. After fee's and friction charges 2 - 2.5% is probably average for diversified portfolio with 1% management costs. And that is assuming you don't do anything rash and emotional and reinvest the dividends.
There are bound to be people that beat that by a hefty margin, but an equal number will finish below that target.
No investment strategy beats earning a paycheck for as long as somebody is willing to pay for your productivity. You have to invest to keep your purchasing power on savings ahead of inflation and to prepare for when nobody is willing to pay you or you are done working. Taxes are inevitable. Minimize friction. Compounded annually that 2 - 2.5% actually is real money.
My fund is in the green by around 57% since initiation in March 2010. I think a performance of at least 40% CAGR should be the goal of a decent hedge fund ...
Ten Best Hedge Funds of All Time
#8,9 some that managed to "generate >40% CAGR over longer time periods" (I think).
Paul Tudor Jones
Daniel S. Loeb
I think a performance of at least 40% CAGR should be the goal of a decent hedge fund ...
You can't do that for too long, unless your aim is to own the total stock of global financial assets.
and good old William Ackman, of course ...
Well obviously if you're going to pursue a 40% CAGR strategy, you would start with $1 in capital. Starting with a lot of money wouldn't be sporting.
Ha ha ha
That and you want to have at least a few counterparties left to trade with. ;-)