Sell in May and Walk Away

Sell in May and Walk Away

Investment wisdom which is also suitable to a crocheted wall-hanging won’t generally make you rich: Markets are far too sophisticated and far too beholden to the vagaries of the news cycle to kowtow to a six-word ditty. But over the years, and all indications are that this year, the popular adage that an amateur investor should “sell in May and walk away” begins to look as though it just might mark a crucial exception.

This article explores the provenance of the saying, its conformity to long-term and recent historical patterns of return, and then points its focus forward to the issue of why this year, more even than most, it might not be such a bad idea to take your grandfather’s broker at his word. The entire article should take about fifteen minutes to read. Our next article will explore just what it might mean, this year in particular, to “walk away.”


Thanks to Google we can quickly run down a few solid leads on the origins of this time-honored maxim, the most credible of which asserts that the saying isn’t just old; it’s British. It goes all the way back to the time when the British were the ones making money in financial markets, instead of the ones tying dynamite to themselves and lighting the fuse and then anxiously googling “what is dynamite” to try and find out what they’d just done.

Apparently the original saying (which the JLI has to admit sounds distinctly more British) was, “Sell in May and walk away; come back on St Leger’s Day.” This director’s-cut version is far less helpful to us now — for reasons that are at once obvious and less so — because it happens that St Leger’s Day is a horse race (um … not a day, apparently), and occurs in early to mid-September. Obviously anyone who didn’t know this would be left at sea by the second half of the quip, but I put it to you that this isn’t even the best reason why the second half has fallen off. The best reason being the stock market crash of October 1929.

Even if this is your second financial article ever, you probably know that October is a notoriously dangerous time for equities markets. Indeed from a statistical perspective, the month wouldn’t actually need a second data point after ’29 to stand as the worst of the twelve — and there have, in fact, been several other pretty rotten Octobers in the meantime. Black Monday (in 1987) happened in October. Friday the Thirteenth (the 1989 market correction, not the movie) happened in October. The Asian Flu of 1997 happened in October. Even the financial meltdown of 2008, which torpedoed the already sputtering Presidential campaign of John McCain, happened in October.

Like all things financial, some of this is down to irrational psychology. As soon as things start to look dicey in pumpkin-spice month, leveraged speculators begin to wonder if they’re the ones destined to look stupid for having failed to learn history’s lesson. So they start pressing the sell button at a far earlier sign of trouble than they might in, say, February. But because they’re leveraged, they have to take whatever they can get for their assets, which drives a self-perpetuating wedge between the bid- and ask prices of the assets they hold, at least until the market’s demand side can catch up — which then doesn’t happen, as more would-be bargain buyers see the gathering gloom, ascribe it to a historically shitty month, and either sit the whole thing out or become sellers themselves.

What then of St Leger’s Day? Easy fix, since nobody left in the world’s trading pits has any idea what that is, anyway: Just drop the second half of the saying, and in its place advocate an informal closing of the “go away” window on 1 November.

An amateur investor like me, who knew nothing else, would buy the Dow 30 or S&P SPDR on 1 November every year, and sell it on 1 May every year. Six months in and six months “out.” And here’s the thing about that: It shouldn’t work. And yet it somehow does.

Equities markets over time have done predictably better between November and May than they have between May and November. And the effect for the small-cap stocks, where the bulk of the action is for mutual funds and ETFs, is considerably larger.

In considering just why this shouldn’t work as a strategy, things quickly get a bit abstruse for most of us lunchpail types, but the relevant dynamics can be summed up in the single word arbitrage. You may even have heard this term, and wondered exactly what it’s all about. In its broadest, most basic terms, arbitrage refers to the existence of artificial money-making opportunities for reasons of poorly behaving markets. If, for example, you could buy a Sony Playstation 3 for $200 on the day before a given Thanksgiving a few years ago, and sell it for $500 two weeks later, that’s arbitrage: You haven’t actually done anything to “earn” those extra $300. They happen because the Sony Playstation 3 was incorrectly priced by the company, and at that price there simply weren’t enough Playstation 3’s to meet the demand.

For a large enough market with good enough information, these wedges of opportunity quickly disappear for no more complex reason than that they are inherently self-cannibalizing. The act of wanting the Playstation 3 to take advantage of reselling it, will itself bid up the acquisition price, higher and higher until it meets the market’s willingness to pay and the wedge of resale profitability disappears. This is how markets work: If there isn’t enough of something, the prevailing price rises until enough buyers are culled from the queue — and enough new sellers are enticed in by the chance to make a profit — that the volume of sellers and the volume of buyers are brought back into balance. Indeed it’s where the word equilibrium comes from.

Of course financial markets are the biggest, and have the best information, of any markets on the planet. Nowhere on earth does information travel with greater speed and wider reach, than in the pricing of equities. Which should be enough on its own to render the old saw flat-out wrong. If whole swathes of us insist on irrationally selling our equities every May, for no other reason than the arbitrary word written across the top of our wall calendars, and then buy everything back each November for the same absence of a good reason, there should be an enormous arbitrage opportunity just sitting out there in plain sight for someone willing to do the exact opposite. Few businesses actually suffer from structurally predictable erosion in their earnings just because the calendar winked over from April 30, and while some businesses do experience increased profitability in November and December, that effect is usually baked in. Point being: In markets as well-defined and closely watched as global equities, the act of following a multi-hundred-years-old rhyming couplet for our guidance should end up making the other guy rich, not us.

Notice how I keep saying ‘should.’

From 1900 to 2015, and mind you that’s two near-retirement-dead-guys-from-heart-attacks back to back, investors who owned equities from 1 May to 1 November earned a median annualized return of 2.4%, and an average of 1.4%. Investors who did the exact opposite, and only owned from 1 November to 1 May, earned a median of 4.2% and an average of 5.2%. Only 59% of the time were the May-November investors up at all, while the November-May investors won positive returns for themselves at a 70% clip.  (Source: Wise-owl Independent Research.)

I don’t want to get too technical here, but note that the spread between the average and median is also reversed for the two halves of the calendar. The median is higher than the average from May to November, indicating that the average is being dragged down by outliers to the red-ink side (we might even have predicted this much ourselves). But for the November-May period the opposite is true: the outliers that skew the average away from the median are to the winning side. That’s an enormous difference in its own right. And, again, it just shouldn’t be so — and yet somehow is.

Over a ten-year period from 2006-2015, the May-November investor who salted away $1,000 at the beginning would have … exactly that much, almost to the penny, at the end. The November-May investor would have almost $1600.

Explanations (?)

Aside from the empty psychology of not wanting to be the dumb guy who ignored an old saying and lost money for it, there are a few reasonably plausible explanations that don’t actually require whole swathes of the economy to experience lower earnings reports during the summer and early fall, or whole swathes of the global investment class to come down with the same pandemic case of stupidity every May Day. My personal favorite theory is that many of the world’s middle class educators don’t enjoy twelve-month salaries, and instead must shift their own financial resources every year in order to spend their summers without a paycheck. These folks are biased heavily in favor of selling any stocks or other financial instruments they might own in May, in order to bank the proceeds for day-to-day expenses until their salaries start up again in the fall. And before you say it, no: educators aren’t a huge chunk of the world’s investment pool. But they don’t actually have to be. As long as there’s nobody pushing things the other way — no similarly sized, similarly middle class, similarly handcuffed cohort who needs to figure out how to cover its expenses in November — the educators planning to eat their profits for picnic lunch would be enough to start the cycle every year. After which the cycle perpetuates its own history, thanks to its own history.

Quo Vadis, 2018?

It remains the JLI’s position (as it has been for an embarrassingly long time now) that equities are dangerously overbought. Numerous methods of tracking the issue point to the same conclusion, but perhaps the easiest to scale and comprehend is the “Relative Strength Index,” or RSI. This nerdy little stat operates on a time-constant scale of zero to 100, and any time it exceeds 70, a pullback in equities is more likely than continued growth. At time of writing (midnight EST on 26 January 2018), the RSI for the Dow Jones 30 Industrials Average is 86.76 — a historically high figure in its own right, and even more striking when one appreciates that it’s more than halfway from the usual sell-prediction value of 70, to the maximum possible score of 100. The RSI for the S&P 500 is only slightly less inflated at 84.

And then there’s the six-hundred pound black swan in the room.

Few political opinion leaders expect the GOP to retain control of both houses of congress through the upcoming midterms, and in the meantime the President is evidencing at least some indications that he may soon speak to Robert Mueller and his team at the OIC, under oath. Wall Street hasn’t historically favored one party over the other — indeed equities were up handsomely under both President Clinton and President Obama — but old paradigms have taken more than their share of body blows since the 2016 election upset, too. At this point, to this columnist at least, it seems clear enough to say more-or-less axiomatically that big market movers in New York have loved what they’ve been getting from Trump, McConnell and Ryan over the past year.

Nothing, mark this, nothing that is likely to happen politically over the next ten months, will resonate with such persons as good news. Not least because the situation can’t actually get much better for The Street than it has been. The easy money on predicting how subsequent political change will be interpreted by financial speculators is all to the down side. Oh, and I haven’t even mentioned what’s going on in Korea right now, and how the passing of the upcoming Olympics might remove whatever flimsy speed-brakes remain on what is quickly becoming a terrifying situation there.

Let’s be clear: Neither a bloated RSI, nor a fractious military standoff with Pyongyang, nor even an increasingly ominous-for-Trump picture of the OIC’s ongoing investigation, are themselves automatic selloff triggers for you and me. Throughout financial history the biggest bull markets have always exhibited a tendency to outlast their own fundamentals, not least because the amateur investor’s information set lags the true picture of where things are, often by several steps in the information food-chain and at least one big shift in psychology. And this is doubly true of bubbles — which the present-day equities markets could well be.

By their nature, bubbles tend paradoxically to be less volatile than their mature-bull-market counterparts. That is, until they explode. If the reason a given stock or entire stock market is going up is because it’s going up, that dynamic will automatically withstand almost any impetus to short-term correction.

Look at what’s recently happened with Bitcoin: A valueless asset, predicted by all available experts to crash spectacularly, just continued to inflate in value anyway — and this went on for months, if not years. People need to make money, and the chance to easily make it is a tempter that cuts both ways. If one knows about a rising market, and doesn’t capitalize, the potential for accompanying self-hatred is just too much for too many persons to bear. It’s not that no one wants to miss the party; it’s that at least some of us couldn’t even live with ourselves, if we missed the party. The psychological stakes are almost as high when it comes to missing out on a gain as they are when it comes to actually losing. As a result of which the assets in question will just continue to go up and up and up until the divergence from fundamental value can be ignored no longer, and all the volatility that should have been experienced all along is back-loaded as a single, gob-smacking correction.

And for the jet-lagged investor this is exceedingly good news.

Recall from our debut column that the principle behind what I’m calling “jet-lagged investing” is, in a nutshell, to set tight stop-sell orders on rising ETFs, ratchet those stop-sell orders with every gain in the assets to which they are tied, and then not worry if the stops were triggered at a transitory local minimum — because the asset is being sold at a positive return, anyway.

This has proven a sound strategy for me through any number of different market characteristics, but it is never more potent as a chance to have one’s cake and eat it too than it is during bubbles — precisely because bubbles don’t fluctuate and thus they don’t trip the stop too soon. A long position can be held, even taken on fresh from scratch, provided the investor scrupulously raises his stops penny for penny with the market’s increases. When the stop trips and he cashes out, he is guaranteed to walk away from the bubble in the black. (By the way, this assumes that the crash isn’t so precipitous and so severe that the stop trips at a significantly higher price than the actual value at which the sell order executes, but these days the markets have built-in holds after decreases of a certain amount, in large part to reduce the possibility of this happening. It’s a risk, but not a big enough one to scare a person away from making money.)

This strategy serves the JLI right now, and continues to occupy the JLI’s highest rung of one-size-fits-allc counsel to readers, family and friends. To anyone who might ask about the twin vulnerabilities of an overbought market and a looming calendar-driven selloff, I would encourage staying long — perhaps even initiating some new long positions as long as they aren’t too wacky. As we get closer and closer to the first of May, adjust the stop-sell thresholds to tighter and tighter windows for downside movement, until even slight hiccups in asset price will eventually cause the stops to trigger at a tidy profit, ideally sometime in the second half of April. Then avoid long positions on high-volatility assets until after the election.

With the order formats available to rank-and-file players like us, the assortment of available strategies need not begin and end with selling now or selling on 1 May. What we can do instead, is jet-lagged investing. We can maintain our long positions through the remainder of this (irrational-?) bull market, but with stop thresholds that are progressively less and less risk-tolerant, until the stops eventually trigger. And then ….

And then?

…That’s the subject of our next article.


Happy hunting.

Dave O’Gorman
The Jet-Lagged Investor
Phnom Penh, Cambodia

Article uploaded on 26 January, 2018
12:00PM Indochina time,
12:00AM Eastern Standard Time

Leave a Reply

Your email address will not be published. Required fields are marked *