Can January’s US equity market performance tell us something about rest of year returns? We ran the data, but first here are some background points on this idea:
- There’s an old Wall Street adage that the stock market’s performance during the first 5 trading days or month of a year can predict whether the entire year will turn out well or not. “So goes January, so goes the year”, the saying goes.
- One reason given for this belief: positive money flows that stem from Americans making 401K/IRA contributions in the New Year before they max out by Tax Day in April. The better the early-year flows, the more investor/affluent consumer confidence in the year ahead.
- Some call this the “January Effect”, although that actually is something completely different: the tendency of US small cap stocks to rally in the first month of the year. Tax loss selling in December abates in January, and in illiquid small caps that can make for positive price action.
To see if January’s returns do foreshadow the year as a whole, we looked at the US stock market from 1980 (when 401K/IRA contributions grew more common) through last year to see if this theory checks out. Here’s what we analyzed and found:
- We calculated the S&P 500’s price returns for the 1) first day of the year 2) first 5-days (aka week) of trading and 3) entire month of January for every year back to 1980. We then compared these returns to the S&P’s annual price returns.
- The correlation between the returns of the first day of trading and entire year is 0.18, or an r-squared of just 3.2 percent.
- The correlation between the returns of the first 5 days of trading and entire year is 0.25, or an r-squared of 6.3 pct (better than 1-day, but still low).
- The correlation between the returns of January and year as a whole is 0.41, or an r-squared of 16.8 pct (actually relatively high for a simplistic one-variable model).
Takeaway: not only does the correlation progressively rise from the first day, to first week, to first month of trading relative to a whole year of returns, but a 17 percent r-squared between the performance of January and an entire year is greater than one may have thought likely. With this relatively strong correlation in mind, we dug deeper into the data to see how often the S&P’s returns during these periods are positive or negative, and how that corresponds to its annual price returns:
First Trading Session of January (average return of +0.2 pct):
- Negative 51 pct of the time (down 0.9 pct on average), but still ends the year higher 76 pct of the time (annual return +9.7 pct on average).
- Positive 49 pct of the time (up 1.4 pct on average), and higher in 75 pct of these years (annual return +10.9 pct on average).
Takeaway: there’s essentially a 50/50 chance of the S&P ending in positive or negative territory on the first day of trading in a new year, but it still finishes the year higher most of the time (+75 pct) either way. This year, it was up 0.6 pct on the first day.
First 5 Days of January (+0.3 pct on average):
- Negative 37 pct of the time (down 2.4 pct on average), but still ends the year higher 73 pct of the time (up 5.6 pct on average).
- Positive 63 pct of the time (up 1.8 pct on average), and higher in 77 pct of these years (up 13.0 pct on average).
Takeaway: the S&P is up the first 5 days of trading during most years, and generates over double the annual return of years versus when it was negative during the first week of trading.
Entire Month of January (+1.1 pct on average):
- Negative 39 pct of the time (down 3.7 pct on average), but still ends the year higher 63 pct of the time (up 2.2 pct on average).
- Positive 61 pct of the time (up 4.1 pct on average), and higher in 84 pct of these years (up 15.5 pct on average).
Takeaway: the S&P is usually positive during January (+60 pct of the time), and generates a much better return during these years versus when the index is negative in the first month of the year (average annual return of +15.5 pct versus +2.2 pct).
Additionally, if you’re wondering how January fared during 2000 and 2008 (notably poor performing years for US equities), here’s the data:
- 2000 (down 10.1 pct): first day (-0.95 pct), first week (-1.89 pct), first month (-5.09 pct)
- 2008 (down 38.5 pct): first day (-1.44 pct), first week (-5.32 pct), first month (-6.12 pct)
Bottom line: we understand why investors take January’s return as indicative of the rest of the year’s performance, especially with the experience of 2000 and 2008 as those years started off terribly. Of course, this concept is not bulletproof, but history shows a positive performance for the S&P in January tends to lead to much stronger returns for the year overall.