On April 16, 2018, I published an article titled "How Far Could Magna International Fall?" In that article, I warned investors that even though Magna International (MGA) was a great company the stock price has historically been subject to deep cyclical drawdowns. Here is the table of Magna's historical drawdown data I shared in the original article:
I noted that MGA stock was susceptible to drawdowns in the 45% to 75% range, but that the stock tended to recover fairly quickly, and that made it an attractive investment near the bottom of the cycles. In April, when I wrote that article, I thought we were reasonably close to a top in Magna's stock price. So, I suggested it would be prudent for MGA shareholders to rotate out of the stock and into something more defensive like Invesco S&P 500 Low Volatility ETF (SPLV).
I received a surprising amount of pushback in the comment section from readers. Sometimes, if I'm writing about popular stocks like Boeing (BA) or Micron (MU), I expect that sort of pushback, but for whatever reason, the MGA comments caught me by surprise. And at first, the stock price rallied after the article was published.
But after that initial rise, it didn't take long for MGA to drop. It is currently down over 25% from its peak and here is how it has done relative to SPLV since my original article was published:
We now have over a 20% spread that has opened up between SPLV and MGA. What is even better is that in the 7 months that have passed, SPLV has had a positive return and performed significantly better than cash would have. This is precisely what the strategy of getting more defensive was trying to accomplish. Let's review that strategy.
Long-only Rotational Strategy
For most investments, I try to let the strategy and the process do the heavy lifting. I spend as much time refining and developing strategies and processes as I do analyzing individual stocks. This is because if I can simply avoid psychological mistakes and be disciplined with the process and the research, I know that I can produce superior market returns as long as my bets aren't too concentrated and I get enough opportunities. Charlie Munger's philosophy of mistake avoidance is so ingrained into my psyche at this point, the vast majority of my time is spent, not coming up with 'winning' investments, but simply trying to avoid below-average investments. And if one can avoid below-average investments, they will, by definition, produce above average results.
So, when developing this cyclical strategy, instead of asking the more common question of "How much could this stock rise?" I asked, "How far could this stock fall?" When I combined that question with a rejection of the popular assumption that volatility equals risk, I was able to develop a fairly unique strategy for evaluating stocks that had a history of high cyclicality. By getting defensive and rotating out of cyclical stocks near the top of the cycle, I realized it would create a high probability to purchase more shares, without paying any more money, near the bottom of the cycle. As long as the company was above average quality, I worked under the assumption that the business would likely recover from the downturn and so would the stock price.
In just seven short months since my original MGA article, we see how the strategy can work. If someone owned 100 shares of MGA back on April 16 and sold those shares to buy SPLV, they could now sell SPLV and, without adding any additional money, buy 121 shares of MGA now. Whatever future stock appreciation MGA might experience from this point forward, whether it be 10 years, 20 years or 30 years into the future. The shareholder who rotated out and then back in would be 21% richer as long as the stock eventually recovered its previous highs.
One can, of course, be as aggressive as they wish with the strategy. But as we can see from the results already this year, the strategy can work. Now let's take a look at some previous downturns and see what sort of spreads MGA has produced in the past relative SPLV or SPY. That, combined with our historical drawdown data, might give us an idea of when to rotate back into MGA for maximum gain.
SPLV wasn't in existence until 2011, but as we can see, MGA suffered a drawdown that year which produced a spread of about 40% between the two investments around December of 2011.
Again, if we start in January of 2015, we have some opportunities of a 35-40% spread opening up between the two investments.
Before 2011, I don't have SPLV to use as a reference, so I'll use SPY, but I expect SPLV to fall less than SPY during the next downturn, so SPY is a conservative choice to use as comparison.
This chart is from the Great Recession and starts in January of 2008. The important thing to remember here is that when we get really big drops, especially below -50%, that the spread between the stocks effectively gets amplified. So if SPY drops 50% and MGA drops 75%, we might think that we would gain 25% more shares rotating back into MGA from SPY. Actually, we would gain 100% more shares. Because if both stocks were trading at $100 in January of 2008, and SPY dropped to $50, but MGA dropped to $25, you could now double your original number of shares by rotating back into MGA (even though at the time of your rotation back in you would have half of your original capital).
This dynamic creates a situation where there are big rewards for being patient before rotating into the target stock during a downcycle, and in my view, they usually outweigh the potential opportunity cost one might suffer by missing the stock entirely.
Before we move on, let's take a look at the downturn that started in 1998 because most of the similarities I see in today's market, at least with regard to cyclicals, are with that downturn.
1998 is a good example of why getting defensive instead of going to cash can pay off when using a rotational strategy. Whereas in 2009 we could have doubled our shares of MGA even while SPY dropped 50%, in 2000 we could have could have tripled our shares because SPY was up 50% while MGA was down 50%. And I actually think we are probably closer to something similar to 1998 right now than we are 2008 because a lot of cyclicals have already dropped significantly while the SPY just made a new all-time high.
Please note that I haven't started any of these charts at the exact cyclical peaks of MGA because I know it's unrealistic to think we can pick exact tops. I just started them at the beginning of the year near the downturn in question. If we go back to the table in the introduction, we can see that Magna has a tendency to fall 40-50%, sometimes even when there isn't a recession, but, if the stars align in just the wrong way, it can suffer extremely deep drops. This leads me to believe that we ought to at least take a chance on the stock after it has fallen 40% from its peak price. That would have gotten us into every significant downturn MGA stock has suffered except the one in 1994. A 40% drawdown would produce a buy price of about $40.60, so that will be my first buy price.
I always use two mostly predetermined entry points for cyclical stocks. As I noted above, the first one I would make around $40.60. Please keep in mind that even though I give exact prices for entry points, that in the end, everything I do in my analysis is an estimate. Sometimes if I'm going on vacation or something and I'm not going to be near a computer for several days, I'll put a market order in for stock that is getting close to my buy price, but most of the time I watch the price as it gets close to my buy price and just use my best judgment on when to pull the trigger. If the stock is falling like a rock, I don't feel that I have to get in front of it midday at my exact price. Conversely, if it feels like it is making a bottom near my buy price, I might get in early. That's what happened with Cummins (CMI). I had written an article like this one and set a buy price of $126, but the stock felt like it made a bottom before that so I bought at $131.
That said, I think producing a non-fuzzy number is important to help keep my focus and not let emotions get too much in the way of making decisions. This is especially important for my second entry point. I use a second potential entry point for a couple of reasons. Sometimes, like with Magna, there are two possible historical scenarios, one a medium price drop, and one, a deep price drop. Since I don't know for sure what sort of drop the stock will experience, dividing my purchases up helps get into more stocks when we have corrections like in 2015-16, rather than missing out waiting for a deep dive. But being prepared to make one more purchase keeps me from panicking if the downturn turns into something worse, and it helps me make more rational decisions because I did the analysis ahead of time when I was clear-headed and wasn't fearful the world was coming to an end. So, I use two entry points for a mixture of practical and psychological reasons.
For the second entry point, if we look at the historical data we see two drops 80% or deeper. The first one looks to be a post-IPO drop that coincided with the 1987 crash. It's unlikely we will face that sort of crash again. The second deep dive was during the Great Recession when auto manufacturers were going bankrupt and financing for a new car was difficult for many consumers to get. So, on one hand, I don't think we should expect that even a deep drop will be as deep as those two were. On the other hand, we've had fairly long and extended cycle when it comes to autos. People are taking out longer loans and paying more money for new cars than before the last cycle. The Federal Reserve is raising rates and I see plenty of used cars on the market. I think lots of younger people are forgoing buying a car because of services like Uber (UBER) and Lyft (LYFT). These trends of extremely long-term loans, ride-share services, and low-interest rates may produce, if not a super-cycle, perhaps at least a bigger cycle than we had in the past.
I can remember around the year 2000, my wife and I bought our first car as a couple, and the interest rate for a used car on a four-year loan was 13-14%. I bought a similar car of similar age last year and the rate was about 4%. If interest rates ever truly normalize, or even if the Fed just overshoots a little bit, the effect it will have on new car purchases will be greatly magnified, especially given the 6-year loans people are taking out nowadays.
These are mostly anecdotal observations, but they help to potentially provide a downturn rationale about why just because the next downturn probably won't be another Great Recession, doesn't necessarily mean MGA stock can't drop 60% or 70%. So, based on the trends I'm seeing, I think my second entry point should be at 70% off the highs, and I think I should weight that entry higher than my first one.
I consider MGA a high-quality business, and for high-quality businesses, I consider a full position a 4% portfolio weighting. Much like my approach with Micron stock that I wrote about last month, I plan to weight the first purchase at 1% portfolio value, and the second one at 3%. My decision to do this has more to do with how I view the height of current cycle than it does past history (historic probability would suggest a 50/50 split of 2% at $40.60, and 2% at $20.30 would be reasonable). So, one could adjust their approach how they see fit if they think the current cycle wasn't as steep as I think it has been, but I think the chances of a deeper dive in the stock price are fairly substantial.
When I write an article like this one where I sort of pre-announce my intentions to buy a cyclical stock, it should be noted that usually, I haven't done a full analysis yet. Before I actually buy a cyclical stock, I put the stock through several impairment tests that are designed to test whether or not there is something different in this cycle from previous cycles that might prevent the stock price from fully recovering its previous highs in a timely manner. Usually, if I write an article on a stock it is because I have run through at least a couple of those tests, and on the surface, the business looks good.
For example, I know that Magna International has been publicly traded for over 25 years and has previously gone through two major downturns. That's one of my tests. I also know that it has a history of fast recoveries from those downturns and that it has an A- credit rating from S&P. All these things serve to let me know, at least on the surface, MGA looks like a good cyclical candidate. But if eventually I examine the business more closely and I don't like management that much, or they are facing a threat to their business model in the near future, I might lower my weighting for the stock, lower the price at which I would be willing to buy, or decide it's just not for me. Usually, I try to at least look hard enough before I write an article that I'm fairly certain it's good enough to buy, but I always reserve the right to change my mind if I find something I really don't like after a closer examination.
Operating under the assumption that after a closer examination everything checks out for Magna International, I'll be a buyer with a 1% weighting at about $40.60, and a 3% portfolio weighting at $20.30, should the price fall that far.
Disclosure: I am/we are long CMI.
I wrote this article myself, and it expresses my own opinions. I am not receiving compensation for it (other than from Seeking Alpha). I have no business relationship with any company whose stock is mentioned in this article.