Let's take a look at a story I've been watching closely all year. It was one of my 10 major predictions for 2012. You might have heard of this one.
How will the election affect the stock market?
Election years fascinate me. Especially these days when so much of what drives the economy depends on policy and politics.
Stock market bottoms tend to occur during the first two years of a president's term, not the last. In fact, on average, the very bottom occurs on average just before the mid-term elections.
There's tons and tons of research out there on this long-term pattern, but it was easy enough to do my own quick and dirty study. Given the option, I'd rather gather my own data and come to my own conclusions rather than use somebody else's.
I looked at 112 years of history in the Dow and categorized each year depending on where it fell in the election cycle.
As you can see, it's a pretty interesting skew. The average annual return for the DJIA for the last 113 years is 3.41%1. So years 2 and 3 really stand out. It really pays to be out of the market during years with a midterm Congressional election and back in again once all that political drama is done with.
Presidential election years (like this current year) tend to be above average, but not tremendously so given the wide distribution of outcomes. The years before an election year are the really good ones. With a return of 6.9%, they're twice as good as average! Year 1 of presidency tends to be kinda "meh" and investors should probably just sit on the sidelines during Year 2 of the cycle, the one where we have mid-term elections.
Is this statistically sound? Probably not. This basic study looks only at performance during calendar years. But for some reason, analysts in this industry obsess about calendar year performance. It also doesn't incorporate the range of trading in a given year. That would be much more useful and tell us more about the nature of our distribution. But for time being, this will do.
If you want to play with these numbers yourself, feel free to use my data.
Since I have the data up, here are a few other things you can look at.
We'll start with a chart of average performance by month.
Everybody's seen this chart before, so there shouldn't be any surprises about which months tend to be the good ones and which tend to be losers.
Next, here's the percentage of time that the month has been positive.
Pretty cool, huh? Some of these calendar months really stand out.
Now let's look at the standard deviation of each month's performance².
Most months fall in the same basic range, but certain months tend to be quite a bit more volatile than others.
Finally, let's do a scatter plot that relates them.
This is a crude way to gauge the risk/reward of a given calendar month, but it's worth doing because returns are meaningless unless you relate them to risk. (It's remarkable how few investors actually do this.) In these types of risk/reward charts, you want to be closer to the top left corner. That means more return and less risk. Being in the lower right corner is bad. That means lower return and more risk.
As we look back through all of this data, you can see that the range of outcomes for the month of December has historically been relatively narrow -- if you're going to bank on a month, you're probably best served by banking on December. On top of that, December also exhibits the highest average monthly performance. And as an added bonus, December also exhibits the highest percentage of profitable months.
In other words: be long in December.
Conversely, this simple analysis suggests that September is a terrible month to be invested. You know what else suggests that September is terrible month to be invested? My gut. Don't laug. Your gut is an incredibly complex machine that represents the sum total of your entire life's experiences. Knowing how to properly read it and knowing when to trust it can be a valuable skill. There's not a single time of the year that my gut gets more nervous about being in the market than September and October. Now you see the data as to why.
Anyway, September is not only the worst performing month, but it also has the highest standard deviation and finishes in positive territory less often than any other month.
Can you trade strictly off analysis like this? Sure.
Should you? Probably not.
That being said, I spend a lot of time talking about the importance of context and the macro backdrop. This kind of information is incredibly useful for helping to understand context.
Why It Matters
I bring all this election cycle stuff up because I think we're getting close to a secular peak.
This is a pretty bold call. The last time we had a secular extreme it was way back in March of 2009. These are the kinds of calls that analysts only get to make once every few years. I don't think the market has topped out quite yet. I do think there's still a little juice left to be squeezed out of the current market, and the recent action has been much more constructive than you may realize.
What does this mean? Well, it means that if you're a long-term investor who's been waiting around to enter the markets, you might be well served by waiting a little longer. If you're a shorter term, directional trader, your bias should be towards the long side right now. If the market violates 1340 or worse, 1315, maybe think about changing the direction of that bias.
Generally speaking, I'm the kind of investor who's happy just waiting around until something extreme happens. Then I'll make a move, which is usually to fade the consensus. I wrote about this a few months back right here at Seeking Alpha. For the most part, I ignore investor sentiment data or just use it as background context. The exception is when investor sentiment moves to an extreme. Then it can be very useful.
In the middle of May, bullish investor sentiment fell to the lowest level since early 2009. I just about fell out of my chair with excitement and assembled a quick strategy which basically amounted to "buy the market for the next 3-6 months." So far, it's worked out reasonably well. My predictions usually don't play out in such a textbook manner.
I see no reason why the market can't challenge 1425 from here. I also see no reason why the market couldn't just say, "screw you!" and fall off the table, secular top already in place.
But if you want a medium term trade, here's what I'd do:
- Be long the market (or buy it on the next little dip).
- See if the market breaks through 1425ish and avoids making a new low below 1275.
- If it can't, you should heed the advice of the philosopher Ludacris and "get out the way."
I'm still watching that old peak of 1576. I think we could get there or reasonably close to it. But the market's kinda running out of time. I love being long in December and the turn of the year, but I think that a lot of reality is going to sink in during the first quarter of next year. All the political trickery and magical thinking will be finished. As we saw above, years 1 and 2 of the presidential cycle tend to be the worst and I think they're the worst because those are the years where pragmatism matters.
I'm not sure if I want to get long the market for the next few years as we're approaching a secular high, close to entering the unfavorable years based on the presidential election cycle, and about to begin a new era of tough policy decisions.
I haven't a clue when we'll see this secular peak. But by the end of 2013 I think it will be clear. And I think that the midterm year of 2014 will be the worst year since 2008. It could even be an all-time banner year for badness. Make sure you have a sturdy desk to hide under.
Until then, if you want to know when we hit the final top, it'll be the next time that the Street is overwhelmingly bullish. If the market breaks on through to a brand new high and the data is lookin' good and investors are feelin' great it will almost certainly mean that the top is here. Picking tops is an exercise in futility, but with this kind of contextual analysis I think you can get close enough.
I could really see that happening this December or January. I could see political cowardice in the face of the fiscal cliff and the application of more band-aids, and that'd probably make the market happy. I could see extension of existing policy, preservation of stimulus, and I could see it all coincide with additional last-gasp QE3-4-5 efforts from the Fed. It'd be just enough to stoke investor confidence.
But we all know that none of these efforts resolve the structural, secular problems. Those will work themselves out in a different way.
Other things to watch
There are, of course, two kickers to all of this.
- Negative real rates. Investors don't have a choice right now. They are literally forced to buy stocks and the only thing that will stop them is an acute fear of collapse, when the fear outweighs the greed. I think this means that the decline will begin slowly and accelerate towards an environment where investors are too afraid to own stocks at any price.
- Massive corporate earnings margins from a super-friendly environment for corporate America. This makes equity valuations seem very attractive. I can't argue with this data. According to certain metrics, stocks are not expensive right now.
If you're going to pay attention to only two things, these are the things you need to watch. Neither one of these factors will serve as the catalyst for the next bear market. But both will act as gasoline near an open flame, transforming a normal correction into something truly incendiary and secular.
Interest rates aren't going to spike any time soon. I mean, they could. Some people have been arguing for years that all this stimulus and money-printing is going to lead to massive inflation and higher interest rates. But it hasn't happened yet. And I'd still assign it a fairly low probability. If the transmission mechanism changes and the velocity of money spikes, then sure, rates could follow. Politics could play a role here -- for example, we could enact policy that greatly incentivized banks to lower standards and make loans again -- but I don't think it will.
Instead, the main risk here is deflation. Deflation would eliminate the negative real rate condition. In a deflationary environment, a 2-year T-Bill yielding 0.25% looks really, really attractive. This is why the Japanese keep gobbling up JGBs. Something to think about while we complain or marvel at these peculiarly low interest rates.
Policy could also "resolve" that second condition. Despite our macroeconomic problems, the stock market is still a pretty great place to be if corporations are knocking it out of the park. But large corporations probably won't keep posting record earnings margins forever. In fact, they certainly won't. Profit margins don't keep expanding endlessly. By definition, they can't. They move in big, unavoidable cycles. It's a side effect of capitalism. (Or the quasi-capitalism the U.S. employs.)
Since the thought of raising taxes on the middle class is abject heresy and raising taxes on the wealthy won't move the needle enough, perhaps increased corporate taxation could be on the agenda for the 2014 election? Not that special interest would permit our Bought Congress to let that happen. But who knows. Never forget that in a Democracy, citizens are the ones with the vote. I know it seems unfathomable for the current system to change for special interest groups to relax their grip on Congress. But the path of history is cyclical, not linear.
Maybe the populist revolution we all thought was going to happen back in 2008 doesn't take place until the 2014 midterm elections? Maybe that's when all the tea partiers are kicked out and replaced with a new breed of radical populist Democrat? Maybe that's when Occupy (Wherever) finally evolves into something with real focus and meaning with clever, manipulative individuals at the helm instead of aimless pseudo-revolutionaries? Don't forget about the psycho-social sub-strata in the U.S. right now. There's plenty of fear and anger just beneath the surface.
In any case, none of these things will be the catalyst for the next bear cycle. The catalyst will be an unforeseen acute crisis or a recession. Once the feedback loop gets going, it'll be ugly for a while. And all the pieces are in place for that feedback loop to happen.
During crises, people start throwing around the term "perfect storm" with reckless abandon. After the analysis we've gone through today, you should be able to see how all of these factors -- factors that would indeed generate a "perfect storm" -- are linked. Honestly, I'd be very surprised if we didn't see a "perfect storm" materialize sometime in the next 3-4 years. There's a good chance it'll be worse than the last "perfect storm," too.
Consider: at some point in the next couple of years we are highly likely to have an environment where the economy is slow, the long-term structural challenges appear too great to overcome, investors are increasingly more afraid, policy gets unfavorable towards corporations and their margins, deflation becomes a distinctly tangible threat, cyclical election patterns start working against us, and earning a risk-free 0.15% on your money looks pretty darn attractive.
It's funny, when you step back, the one thing that's clear about the market is that it moves in big cycles. Perhaps this is the reason why. While the market is moving in one direction, all the dominos are slowly being lined up in another. And then at some inevitable point, all the new dominos are in place and all that's needed is a simple flick of the finger.
If you look closely, pretty much all the pieces are in place for the next secular bear market.
We're just waiting for that flick.
(Click here if you want that DJIA data set)
1 I like using the Dow because it includes dividends and inflation automatically in its divisor. No surprise, over the long run, what's left over is pretty close to GDP. Remember your proxy equation for really long-term investment returns: market performance = GDP + Inflation + Dividends.
2 Ideally, I'd rather look at the daily annualized volatility within each month than just the standard deviation of each entire month. But then my data set balloons from 1351 data points to around 41,000. Still, my guess is that the data will suggest a similar conclusion. September will still be one of the more volatile months while December one of the least volatile.
Additional disclosure: see cognitiveconcord.com/legal-notice for additional disclosure.