The end of 2013 is quickly approaching. We certainly don’t want to jinx investors with two weeks of market action left, but considering the S&P 500 is up 27.3% with dividends and 24.7% ex-dividends, we feel there’s a strong chance the market will hold on and finish up over 20% for the year. For much of this year, especially the back half, we’ve heard numerous rumblings about “bubbles” and the return of excessive exuberance. Surely we won’t solve any debates here. We thought a simple examination of market performance in the two years after an up 20% year might give some color on what we can expect going forward and help frame expectations.
Over the past 68 years, the S&P has closed up 20% or more on a price basis 18 times, or approximately 26% of the time. Four of those occurrences clustered together during the late-1990s hyper-bull market run. All in all, 20% up moves in a year aren’t in the rare “black swan” category, but they’re still nothing to sneeze at. Here is a list of prior years with 20%+ performance:
|
YEAR
|
PERFORMANCE
|
|
|
2009
|
|
23.45%
|
|
2003
|
|
26.38%
|
|
1998
|
|
26.67%
|
|
1997
|
|
31.01%
|
|
1996
|
|
20.26%
|
|
1995
|
|
34.11%
|
|
1991
|
|
26.31%
|
|
1989
|
|
27.25%
|
|
1985
|
|
26.33%
|
|
1980
|
|
25.77%
|
|
1975
|
|
31.55%
|
|
1967
|
|
20.09%
|
|
1961
|
|
23.13%
|
|
1958
|
|
38.06%
|
|
1955
|
|
26.40%
|
|
1954
|
|
45.02%
|
|
1950
|
|
21.68%
|
|
1945
|
|
30.72%
|
The years above represent a wide range of market and economic environments. Every decade from the 1940s forward is represented, though secular bull market decades tend to produce the most occurrences, as one might expect.
What do the performance metrics look like in the next year following one of these big up move years? Reasonably decent, actually. In most cases, there is a continuation of performance, though the gains tend to be more muted than the prior years gains. Out of the eighteen 20%+ yearly moves, the following year produced positive performance 14 times, with 11 of those years producing performance that exceeded the average 8.47% yearly gain achieved from 1945 onward. However, in 17 of the 18 occurrences, performance the following year fell below performance during the 20%+ year. Performance acceleration is rare, therefore. Looking at the summary statistics for this group of follow-on years, we see performance that comes in slightly higher than the long-term market average, with less volatility:
|
Performance in Year Following Up 20% Year
|
Overall Performance Metrics, 68 Years
|
|||||||
|
Count
|
18
|
|
|
|
Count
|
68
|
|
|
|
Average
|
9.95%
|
|
|
|
Average
|
8.47%
|
|
|
|
Median
|
10.89%
|
|
|
|
Median
|
10.62%
|
|
|
|
Max
|
31.01%
|
|
|
|
Max
|
45.02%
|
|
|
|
Min
|
-11.87%
|
|
|
|
Min
|
-38.49%
|
|
|
|
StDev
|
13.39%
|
|
|
|
St Dev
|
16.68%
|
|
|
Now, let’s fast-forward two years. The situation changes, especially when looking at the summary performance statistics for the second year after a big up move. The second year following a 20%+ year has been positive 11 times out of 18, lower than observed in the set of data pertaining to the year-after. The performance in year two following a big up move has been lower on average as well, though it should be noted that the difference in means test between the values in the following year and the values two years later didn’t produce statistical significance at the 5% level.
|
Performance In Year 2 Following 20%+ Move
|
||||
|
Count
|
18
|
|
|
|
|
Average
|
6.30%
|
|
|
|
|
Median
|
2.81%
|
|
|
|
|
Max
|
31.01%
|
|
|
|
|
Min
|
-14.31%
|
|
|
|
|
St Dev
|
0.141283
|
|
|
|
Taking all into consideration, there were only 7 occurrences out of 18 in which there were two consecutive positive years following an up 20% year, with three of those occurrences occurring consecutively in the late 1990s. Also of note, the worst overall year in the data sets for years following a 20%+ move was -14.3%. Granted, this is 5th from the bottom in terms of overall negative performance. Still, for the doomsday crew crowing that this year’s returns are a harbinger for historically devastating declines immediately ahead, the historical record doesn’t support this notion.
Does valuation play any role in follow-up performance? We went back and applied the year-end CAPE P/Es to the years experiencing 20%+ returns to see if valuation levels correlated whatsoever with performance in the following year, the thought being that higher valuations might contribute to worse performance the next year. There was correlation (0.375), but very little to shout from the rooftops about. More interesting is that the correlation value was positive, meaning that the general statistical relationship leans towards a “higher the valuation at the end of a 20%+ year, the higher the performance the subsequent year” narrative. Of course, the low correlation and R-squared values indicate there’s a lot of variance in that relationship.
So, what are the general takeaways? What can we guesstimate as far as what 2014 and 2015 might look like after such a strong 2013? A strong 2013 doesn’t mean that markets have followed Icarus’ lead and flown too close to the sun. There’s a strong probability that the markets will generate solid, positive performance in 2014. It’s very unreasonable to expect another “out of this world” year, however. Historically, the year after has been in line with broader market performance averages. Chances are, however, that choppiness (and its friend frustration) will return by 2015 at some point. In the past, market performance has caught a mild hangover by year two. This dovetails with other models, such as the volatility model we wrote about several weeks ago predicting a pickup in volatility in mid to late 2014. Most encouraging: no 20%+ year in the post WWII era has ever represented the final year of a secular bull market, or the beginning of a devastatingly negative series of market outcomes. Again, the historical record works against those expecting a major market collapse in the next year or two.
