Roger Gibson (from the 2010 FPA Retreat
Excerpts from Roger Gibson's presentation at the 2010 FPA Retreat in San Antonio.
Roger Gibson's breakout sessions at the 2010 FPA Retreat was one of the most interesting presentations of the meeting because Gibson argued persuasively for the benefits of maintaining a consistent asset allocation, and then showed data that was in surprising agreement with Hunt's.
Gibson showed that the high asset correlations of 2008 were not terribly unusual; the securities markets have experienced many periods when everything seemed to move up or down in step. What WAS unusual was the magnitude of the losses, amid even more extraordinary volatility. Looking back at 2008, Gibson found that in 21% of the trading days, the S&P 500 was up or down at least 5%--and that was the CALMEST of the asset classes. Commodities also saw 21% of its trading days swing 5% or more. International stocks were up or down 5% on 33% of the 2008 trading days; for REITs, the figure was an astonishing 54% of trading days.
Gibson offered a discussion of market timing issues. What do you say if a client wants to get out of the market because the losses are too much to bear? Gibson asks whether the client thinks the S&P 500 is headed to zero and all those companies disappear. "If that's what the client expects, then it's a different conversation," Gibson told the group. "If they expect that free enterprise is gone, then you aren't talking about the portfolio any more. You're talking about a cabin in the woods, with guns, and kruegerrands buried in the floorboards."
If the client doesn't expect free enterprise to collapse, then the discussion centers around how much farther the market can drop. Gibson calculated what would happen at various times if the markets returned to their lowest decile PE ratios, which of course was not too scary back in March of last year, but has become increasingly scary as PEs have soared over the past 18 months.
Later, we were shown a slide which calculated the 10-year normalized P/E ratios (the Shiller PE) at the beginning of successive historical 10-year periods--basically putting some numbers to Hunt's recommendation that you buy into the market during times of low valuations. Gibson's slide showed the subsequent average 10-year compounded return if you bought into the market when the PE ratio was below 12 (and stocks were cheap); when it was 12 to 16 (and stocks were moderately cheap to about the historical average of 16.5); when PEs were somewhat expensive (16 to 20), and when they were well above their norms (over 20).
The results were eye-opening. If you buy into the market when stocks are relatively cheap, the average return is nearly 15% a year. Buy when the market is expensive (as it is now) and you can expect 4% returns, on average, and a negative equity risk premium. Hunt confirmed!
But... Gibson warned the audience that this information is not easy to act on. The PE10 briefly touched down to 11.7 in March of 2009, he reminded the audience. But at that time of maximum fear, how many clients are going to be willing to double down and dramatically raise their exposure to stocks? Similarly, when the market is roaring along and valuations march to ever-higher levels, how many advisors are going to be willing to tell their clients to take their money out of the market--and risk having it go up for another year before the inevitable collapse? They'd lose clients to opportunistic (and less value-oriented) advisors who are saying the markets are going to keep going up and don't miss the action. It's a good point.
Later, Gibson showed a scatterplot of all of the 10-year returns graphed against the beginning PEs for each 10-year period, and then drew a best-fit line through the various dots. There were visible exceptions, but in general, the higher the initial PE, the lower the subsequent 10-year return, and vice versa. (The best-fit line, in case you're interested, is defined by the equation: Return = -.58(PE10) + 20.07, which predicts the average return going forward given any PE level.) This graph was interesting in itself, but then Gibson pointed out the particular dots which represented the ten years ending at the end of 2008 and 2009, which were awful, but well above the best-fit line. He said that instead of talking about the terrible decade of below-zero annualized returns, we should be talking about the "lucky" decade, since, based on historical norms, the decade could have been a lot worse. "Try telling THAT to your clients," Gibson said with a chuckle.
Gibson showed that the high asset correlations of 2008 were not terribly unusual; the securities markets have experienced many periods when everything seemed to move up or down in step. What WAS unusual was the magnitude of the losses, amid even more extraordinary volatility. Looking back at 2008, Gibson found that in 21% of the trading days, the S&P 500 was up or down at least 5%--and that was the CALMEST of the asset classes. Commodities also saw 21% of its trading days swing 5% or more. International stocks were up or down 5% on 33% of the 2008 trading days; for REITs, the figure was an astonishing 54% of trading days.
Gibson offered a discussion of market timing issues. What do you say if a client wants to get out of the market because the losses are too much to bear? Gibson asks whether the client thinks the S&P 500 is headed to zero and all those companies disappear. "If that's what the client expects, then it's a different conversation," Gibson told the group. "If they expect that free enterprise is gone, then you aren't talking about the portfolio any more. You're talking about a cabin in the woods, with guns, and kruegerrands buried in the floorboards."
If the client doesn't expect free enterprise to collapse, then the discussion centers around how much farther the market can drop. Gibson calculated what would happen at various times if the markets returned to their lowest decile PE ratios, which of course was not too scary back in March of last year, but has become increasingly scary as PEs have soared over the past 18 months.
Later, we were shown a slide which calculated the 10-year normalized P/E ratios (the Shiller PE) at the beginning of successive historical 10-year periods--basically putting some numbers to Hunt's recommendation that you buy into the market during times of low valuations. Gibson's slide showed the subsequent average 10-year compounded return if you bought into the market when the PE ratio was below 12 (and stocks were cheap); when it was 12 to 16 (and stocks were moderately cheap to about the historical average of 16.5); when PEs were somewhat expensive (16 to 20), and when they were well above their norms (over 20).
The results were eye-opening. If you buy into the market when stocks are relatively cheap, the average return is nearly 15% a year. Buy when the market is expensive (as it is now) and you can expect 4% returns, on average, and a negative equity risk premium. Hunt confirmed!
But... Gibson warned the audience that this information is not easy to act on. The PE10 briefly touched down to 11.7 in March of 2009, he reminded the audience. But at that time of maximum fear, how many clients are going to be willing to double down and dramatically raise their exposure to stocks? Similarly, when the market is roaring along and valuations march to ever-higher levels, how many advisors are going to be willing to tell their clients to take their money out of the market--and risk having it go up for another year before the inevitable collapse? They'd lose clients to opportunistic (and less value-oriented) advisors who are saying the markets are going to keep going up and don't miss the action. It's a good point.
Later, Gibson showed a scatterplot of all of the 10-year returns graphed against the beginning PEs for each 10-year period, and then drew a best-fit line through the various dots. There were visible exceptions, but in general, the higher the initial PE, the lower the subsequent 10-year return, and vice versa. (The best-fit line, in case you're interested, is defined by the equation: Return = -.58(PE10) + 20.07, which predicts the average return going forward given any PE level.) This graph was interesting in itself, but then Gibson pointed out the particular dots which represented the ten years ending at the end of 2008 and 2009, which were awful, but well above the best-fit line. He said that instead of talking about the terrible decade of below-zero annualized returns, we should be talking about the "lucky" decade, since, based on historical norms, the decade could have been a lot worse. "Try telling THAT to your clients," Gibson said with a chuckle.
Trackbacks
Trackback specific URI for this entry
No Trackbacks
