Forecasting this market with the Fed doing what they’re doing is really a very inexact chore. However, there are some historic parallels that can be drawn.
One of them is to see what’s happened to the market in years after such a strong year as we’ve just had. Again, though, the point of caution here is that in the past, when $SPX rose by 29% (or whatever it’s going to post for this year – currently at 29.2%), it was not being driven by the Fed, but by more natural market factors. Hence, the reaction to those factors in previous years could be completely different that the market’s reaction to more (or less) Fed impetus next year.
But, for the record, here is some data regarding past performances after strong years:
1) Since 1950, there have been 17 years when $SPX rose at least 20%. In the following year, there were 14 years with gains and 3 with losses. The losing years were 1952, 1981, and 1990.
On average, the next year was up 11.2%, but during the 1990's, four years (1995-1998) produced a second year of +20% of more, except for 1998, which was up “only” 19.5%. These were the only back-to-back years of +20% or more, except for 1954-1955. I really don’t know what was happening back then, except for the fact that in 1954 the market finally had made back the losses since 1929 and was trading at new all-time highs.
2) another statistic that may have some meaning is what happens after two consecutive years of +10% gains or more. We are in that situation now, since the market was up 13.4% in 2012 and, of course, is up much more than that in 2013. This statistic is far less bullish: since 1950, there have been 14 occurrences of back-to-back years of +10% of more. The following year has been up only 8 times, while it’s been down or flat 6 times.
The most recent occurrence was 2011. After back-to-back gains of 23.5% and 12.5% in 2009 and 2010, $SPX was absolutely flat (0%) in 2011.
Only four times out of those 14 has the following year been up 10% or more. Once was 1952, and the other three were in that sequence from 1995-1999.
3) finally we can combine the two sections of data: when there have been back-to-back years of 10% gains or more, and the second year is up more than 20%. That has happened only six times since 1950. Four were gains, and two were losses. There was a tiny gain in 1956, losses in 1981 and 1990, and then three large gains in 1997 through 1999. On average, the gain was 10.6 % the next year.
So, from these figures, it seems somewhat likely that 2014 will be an up year, but it will likely be a modest gain – perhaps 10% or less. That could change, of course, if the Fed (as I suspect) really doesn’t back off its accelerator because the economy (I suspect) really isn’t improving.
Regardless, it seems likely that there will be more substantial corrections, which would also imply that volatility will be higher in 2014.
None of these are particularly shocking revelations, but things could become shocking if this Goldilocks environment changes – and there is always a chance of that. In fact, that chance is probably larger than normal right now, merely because of the fact that there hasn’t been anything that has really bothered this market since the fall of 2011 – and especially since the fall of 2012.
What could go wrong? I would think that the major source of problems could be the Fed itself. If they send off negative vibes, or for some reason, if traders realize that what the Fed is doing is really no lasting guarantee of higher stock prices, then some severe problems could arise. These might be accompanied by lackluster economic numbers at the national level – perhaps in unemployment and housing – although so far any news in either direction seems to be interpreted positively by investors.
Our main clues will continue to be the four main areas of technical analysis that we follow: breadth, equity-only put-call ratios, volatility, and the chart of $SPX itself. Severe overbought conditions in 2013 didn’t result in a breakdown of the chart of $SPX. So, if there is one thing that traders should have learned in 2013 is that unless the chart of $SPX breaks down through support – upsetting the higher high, higher low patterns, there is really no reason to take on a bearish attitude.
Source: The Option Strategist Newsletter: Volume 22, No. 24. Read all the articles and get the complete trading recommendations by subscribing to The Option Strategist now.
© 2023 The Option Strategist | McMillan Analysis Corporation