Friday, February 26, 2010

Up and Down Wall Street: Rob Arnott "After Lost Decade, It's Still Tough to Find Returns "

I became familiar with Mr. Arnott's work when he was editor of the Financial Analysts Journal. He is one of the best on expected returns.
From Barron's:

Low yields, rich valuations point to continued paltry returns from stocks and bonds.

AFTER THE JUST-ENDED LOST DECADE of making nothing from equities, you'd think investors ought to be entitled to count on high returns from the next 10 years. Just as buying high at the peak ensures punk future gains, the opposite ought to follow after a long period of weak returns.

Don't count on it, say some insightful investment pros. Investors can't expect much more than 2%-3% in excess of inflation over the next 10 years from balanced portfolios. And fixed-income, which helped cushion the negative returns from stocks in the past decade, may be a loser in the coming decade.

In the 10 years ended last Dec. 31, the Standard & Poor's 500 index lost 0.95% a year. The taxable bond benchmark, the Barclays Capital (formerly Lehman) Aggregate Index, returned 6.33% per year.

But even with that fixed-income shock absorber, the traditional 60%/40% mix of stocks and bonds utilized by institutional investors such as pension funds and endowments, produced negative real returns for the decade (that is, less than the rate of inflation.) That's something that's never happened, writes Rob Arnott, the head of Research Affiliates, Newport Beach, Calif., consultants, in his recent missive to clients.

Actually, Arnott observes, it wasn't a Lost Decade for those who eschewed these conventional approaches. For instance, simply avoiding the capitalization weighting of the S&P 500 would have returned 5.38% per annum in the past decade, or nearly 6.5 percentage points more than the cap-weighted measure as dictated by S&P.

Other ways investors could have earned robust returns would have been to diversify into other asset classes, including real-estate investment trusts, Treasury Inflation Protected Securities and emerging-market bonds.

The MSCI US REIT index returned 10.43% in the 10 years to Dec. 31, while the Barclays TIPS index returned 7.69%. The big winner of the decade was the JP Morgan EMBI, which showed emerging-market debt returning 10.94% a year over the span. Junk bonds didn't do too shabbily either, returning 6.83% a year, according to the Merrill High-Yield index.

These asset classes produced sterling returns in large part because they started out the past decade with high yields, which were the main drivers of returns. But now, emerging-market bonds, REITs and TIPS provide about half the yields they did on Y2K, Arnott observes: 6.3% now from emerging-market bonds vs, 12.4% then; 4.6% on REITs vs 9.0% and 2.0% on TIPS vs. 4.2% then. Junk yields at the turn of the century were a relatively low 11.1% because of the frenzy for tech and telecom credits, but they have fallen into single digits, at 9.1% currently.

Future returns, by Arnott's reckoning, consist of three basic components: current yield, growth in income and changes in valuation. With yields on these asset classes relatively low, "the fat pitch of diversification in risk premiums beyond mainstream stocks and bonds is largely gone."

As for conventional stocks and bonds (as represented by the S&P 500 and the Barclays Aggregate), Arnott doesn't see the new decade being much better than the last.

The dividend yield on the S&P is up to 2.11% as of the end of 2009, low by historical standards but nearly twice where it stood near the peak of tech bubble on Dec. 31, 1999, at 1.13%. Add in annual growth in dividends of 1.2% (the average from 1900 to 2009), and the prospective return from equities is an unexciting 3.31% annually, absent changes in valuations.

On the latter score, the market's price-earnings ratio (as calculated by Yale's Robert Shiller, based on trailing 10-year earnings), is back to the 20s, a 25% premium to the historical average, after a quick foray below that average for the very long run (1871-2009), Arnott continues.

Only in the 1990s did stocks produce strong real returns when P/Es started in the 20-22-times range. For the 10 years after the 1992-95 period, stocks returned 10.6% while inflation averaged 2.4% as multiples moved even higher during the tech boom. Ah, yes, those were the days.

But during other periods with high starting P/E multiples, the future real returns were distinctly mediocre.

In the 1960s, equities returned 6.3% vs. inflation of 4.7%. In the 10 years beginning 1936-37, stocks returns 6.1% but inflation was 3.7%. And starting with the 1928-30 peak, stocks lost 1.6% a year, a slightly positive real return against deflation of 1.9% annually....MORE