March Seasonal Market Patterns

Here’s some perspective on the historical view of the markets in the month of March.  This sort of data is certainly no holy grail, but it provides perspective on what can be a very seasonally sensitive world at times (via the Stock Trader’s Almanac):

In this next chart, DJIA’s 1-Year Seasonal Pattern for years 1950-2012 has been plotted (green line). Post-election years from 1950-2009 (black line) as well as 2013 year-to-date (blue line) are also presented. In all years, DJIA is on average unchanged in February. In post-election years, Februarys on average have declined just about 1.5%. As of today’s close, DJIA is down 0.55% for February, a little worse than all Februarys since 1950, but still faring much better than the typical post-election year February.

And some specific notes on March:

  • Recent record: S&P 19 up, 10 down, average gain of 1.3%, fourth best.
  • Rather turbulent in recent years with wild fluctuations and large gains and losses.

Cullen Roche

Mr. Roche is the Founder of Orcam Financial Group, LLC. Orcam is a financial services firm offering research, private advisory, institutional consulting and educational services.

More Posts - Website

Follow Me:


  1. I personally don’t find this seasonal data very useful. It strikes me more as data mining than anything else. Sorry.

  2. Seasonal adjustments have their roots in the fallacious “real bills” doctrine:

    “In the original federal reserve act of 1913 “It was anticipated that credit extended by the Federal Reserve Banks to commercial banks would rise & fall with seasonal & longer term variations in business activity”

    And: “From the beginning, the Federal Reserve was reasonably successful in accommodating the seasonal swings in the demand for currency—in the terminology of the act, providing for “an elastic currency”.”

    The FOMC is tasked to provide yearly seasonal adjustments as business activity waxes (the FRBNY’s “trading desk” injects reserves) & wanes (mops them up). Reserve requirements are based on the reserve ratio associated with transaction accounts 30 days prior. Bankers are legally obligated to meet this restriction based upon commercial bank credit creation during the holidays.

    Note: RRs reflect the actual monetary base as excess reserves & an expansion of the currency component are both contractionary. I.e. lending by the CBs may not be “binding” however, draining legal reserves is still gives the FRBNY’s trading desk: “leverage over money & bond markets” (V.P. Fed). Feb 27 2007 is a case in point.

    The problem is the FOMC doesn’t recognize that the theory & mechanics are the same for seasonal mal-adjustments & the “real Bill’s” arguments.

    The Fed screwed up during the holidays. They always do. This year is worse than others (as prices at the pump tell you). Efforts to manage “expectations” have failed. When will the Fed be “back-on-track”? – probably, once again, thru sheer luck (or when both inflation & the economy simply “stall-out”).

  3. Roc’s in MVt are cummulative figures. What’s seasonal is also part of money flows roc’s for real-output & inflation. That’s why both gas prices stock prices have jumped. The roc in the proxy for real-output is increasing whereas during most years it falls at this time. That’s why stock prices will continue to climb.

  4. 3rd column is the roc in the proxy for real-output. The 4th column is the
    roc in the proxy for inflation:

    2012-07 100163 0.08 0.54 83810 16353
    2012-08 103760 0.11 0.55 85416 18344
    2012-09 107290 0.13 0.63 87106 20184
    2012-10 106424 0.07 0.60 88800 17624
    2012-11 109790 0.10 0.55 90595 19195
    2012-12 110684 0.15 0.50 92249 18435
    2013-01 117329 0.17 0.59 94063 23266
    2013-02 117554 0.17 0.62 95895 21659
    2013-03 117554 0.21 0.55 97771 19782
    2013-04 117554 0.17 0.53 99516 18038
    2013-05 117554 0.13 0.54 101213 16340
    2013-06 117554 0.10 0.51 102923 14630
    2013-07 117554 0.10 0.43 104583 12971
    2013-08 117554 0.07 0.28 106050 11504

    See that roc’s in RR (the 24 month proxy for inflation) peak in Feb. So commodity prices peak in Feb. But the roc in the proxy for real-output (10 month roc) keeps climbing – so stock prices will continue upward. This is only as we hold things constant. Given additional POMOs by the Central Bank stocks aren’t going to stop rising anytime soon.

    And the BOG has room to wriggle as the current run up in MVt is transitory. I.e., the roc in MVt (proxy for inflation) & thus commodity prices, will now fall & finally bottom in August. Can you szy zinger?

  5. To me the most important line in this seasonal chart is the blue line showing that this year in 2013 we have gone up way faster than normal in the first two months. In order to “revert to the mean” we need to either go sideways for a while or else have a healthy correction of 5% or more. That would bring us much closer to the typical seasonal pattern for a post-election year. Thanks, I find this helpful.

  6. I’m a bit confused as to what ’0′ on the vertical axis represents. Looking at the historical lines it would just appear to be historical starting (Jan 1) averages, but if that’s the case why is current year starting at ~2?