I was looking at previous occurrences where the S&P 500 hit a new year to date low in August. Using Yahoo!, the years seem to be 1953, 1966, 1973, 1974, 1977, 1981 1982 (the decade 1973-1982 sure had a lot), 1990, 2004, and 2011. Of the years, 1982 and 2004 where the only ones where the August low was the low of the year (2 of the 3 most recent). Using Stockcharts (which does not seem to have the years prior to 1981) the year that seems to be the most similar to this year to date was 1990:
http://stockcharts.com/h-sc/ui?s=$SPX&p=D&st=1990-01-01&en=1991-01-01&id=p91451617551&a=241455009
where new lows were hit in August, September and October, although finishing the year well off the lows.
If the S&P had a new low in the August – December time frame [ 1953-2010] (17 occurrences), only 3 times did it also hit a new high – the most recent was in 2004 (when the August drop was not as pronounced as this year).
I was also curious about the yearly range of the S&P. Still using the Yahoo! database, the median yearly range [(high - low)/prior close] is about 22% (average range is closer to 25%). This year the range has been 22%, so if it makes no higher highs or lows, it will be a typical year. Furthermore, the August – December range had a median of 14% (mid 50% range is 11%-19%), so this August’s action of 16% already exceeded the median range.
Finally, I did a quick look to see where the median finish was in the range ([year close]-[year low])/([year high]-[year low]). For all years, the close is near the top of the range at 82%, but if the low was in the August – December time frame, the close was slightly above the middle of the range (56%) (still well off the lows of the year).
So we might have some more downside, but overall it would seem likely the market will finish the year higher than where it is now.
