Turning Around The Jubak Picks Portfolio: It Takes Time (Frustrating Amounts Of Time) But By Late 2016 The Recovery Was In Place: 2017 Total Return 13.13% Despite 27.2% Cash

Turning around a portfolio can take a frustratingly long time. The Jubak’s Picks portfolio has had a performance problem since 2014. But I think I’ve finally put a recovery in place.

The total return on the portfolio was 13.1% in 2017. That still badly lagged the 21.7% total return on the Standard & Poor’s 500. But it’s promising–considering that the portfolio finished the year 27.2% in cash. (Which itself was better than the 37% cash at the end of 2016.)

And it’s quite an improvement from the 2.5 loss of 2016 (when the S&P 500 returned 12%) or the 3.0% loss of 2015 (when the S&P 500 returned 1.25%).

I’ve updated the portfolio performance buttons to show the 483% total return on the portfolio since its inception in May 1997 and the 13.1% total return in 2017.

For the last two weeks I’ve been cranking through the performance figures for this portfolio trying to identify the problem (or more precisely to see if my take on the problem is the same as at the end of 2016 when I started to implement a solution), to see where I am in implementing that solution, and to see if that solution still makes as much sense at the beginning of 2018 as it did near the end of 2016.

I’ve reached some conclusions on the general and specific issues in turning around a portfolio (any portfolio) and this portfolio in particular.

First, one of the reasons that it can take so long to turn around a portfolio is that it can take a while to identify the problem. For example, in 2014 the 4.91% total return on the Jubak Picks Portfolio lagged the 13.5% total return on the S&P 500. But that left open the question of whether this was just a one-year failure of stock picking on my part of a more troubling and deeper problem. After the loss of 3% in 2015 it was more likely that there was a problem in the portfolio–but the S&P 500 itself only returned 1.25% that year. It wasn’t until the 2.5% loss in 2016 that it became unarguably clear even to me (I can be dense sometimes) that there was a portfolio problem.

Second, admitting that a portfolio has a problem is crucial but it’s not the solution. And coming up with a solution can take some time. Fortunately, I run a number of different portfolios with different strategies and comparing those with the performance of Jubak Picks helped identify the problem and formulate a solution. In 2016, for example, when I lost money on the Jubak Picks Portfolio (2.5% remember) my 50 Stocks portfolio killed the index (total return 12%) with a 21.5% total return. Same in 2017 when that portfolio managed to beat even the 21.7% total return for the S&P 500 that year with a total return of 28.2%. So what were the differences between the two portfolios that made a difference? The 50 Stocks Portfolio is always fully invested–no 37% cash balance as with the Jubak Picks Portfolio in 2016. And I don’t make any attempt to time the market with that portfolio–it’s a set of long-term picks and short of a bear market it rides the dips.

Third, stating the problem in a way that doesn’t amount to kicking yourself repeatedly in the ass isn’t all that easy–but it is important if you really want to solve the problem. So, for example, deciding that I’ve lost whatever skills I had as a stock picker doesn’t really get me anywhere. But noting that I’d abandoned a key part of the credo behind Jubak Picks does. I’ve always said that the portfolio will frequently be contrarian because 1) the market is frequently wrong in the short term) and 2) because being contrarian results in bigger gains (when you’re correct.) But the credo also notes that sometimes it’s important to just go with the flow and take what the market gives you. The point in this great bull rally that stretches back to 2009 wasn’t to overanalyze it but to run with it. Instead in 2015 or so I decided that the question was “Is this bull market rally right or wrong?” And I concluded, completely wrongly, that I was smarter than the market move and that the rally was wrong. Which is how you wind up at 37% cash in 2016, 33% cash in 2015. But holding too much cash wasn’t my only problem. Convinced as I was that the rally was wrong–I can prove to you, if you’re interested why the market could not have possibly rallied in 2014-2016–but since the rally was wrong, I would not buy the leaders in the rally. Instead I went looking for special situations and contrarian plays. Sometimes they worked out–like the buy of Precision Castparts before Warren Buffett decided to buy the company–but sometimes they didn’t–like the buy of Hain Celestial Group (HAIN) just before that promising buyout candidate discovered major accounting problems that pushed any potential takeover way, way out into the future. The problem with deciding that you’re going to build a portfolio totally on special situations and contrarian plays is that it is really, really hard work. (There are investors who can do this and I admire their smarts and work ethic.) By deciding that I was going to avoid all the rally leaders, I not only passed up some juicy profits but I traded general market risk (that the rally would fail) for stock specific risk (that a company would develop a problem that I hadn’t foreseen.) The end result was that I added risk to the portfolio while limiting market-tracking profits.

Fourth, turning around a troubled portfolio takes more time than you might imagine because not only do you have to sell the clunkers, but you have to replace them with good picks. Otherwise you wind up just building cash on the sidelines and that doesn’t do a thing for performance. I date the turnaround in this portfolio to October 26, 2015. That’s when I bought shares in Alibaba (BABA), exactly the kind of stock that I had avoided when I was convinced that the rally was wrong. Alibaba didn’t go straight to stock heaven–in fact I finished 2015 with a slight loss in that position. But the buy was important for the turn in my thinking that it represented.

Fifth, there’s always a good chance of a relapse. Despite buying Alibaba in October 2015, I wasn’t done with my conclusions that the rally was wrong. In early 2016 I actually used an ETF to go short the S&P 500 so convinced was I that the rally was on its last legs. I also, in July, went short emerging markets. This was actually not as bad for the portfolio in the long run, although plenty painful in the short run, as refusing to buy stocks participating in the rally had been. I kept hold of those stocks–largely because since I was hedged with my S&P 500 short I didn’t need to sell the Aibabas in my portfolio. Reversing those buys was actually fairly easy–one of the advantages of an ETF is that you can sell a whole lot of stuff with just one “sell”–and as the portfolio and I moved into 2017 I added rally stocks such as Nvidia (NVDA) and Nektar (NKTR). My progress in reducing the cash position in this portfolio hasn’t been as rapid as I’d like: I still finished the year with a tad over 27% cash (but that’s better than the 37% cash in 2016.)

Sixth, and maybe this is the most crucial point, it’s important to realize that no fix is permanent. I was really, really early in calling for rally’s demise back in 2014 and 2015 (and being really, really early, let’s be honest, is the same as being wrong). But the rally is three or four years older than it was in those years. It is close enough to an end so that while I’m not in the “end of the rally” prediction business, I am compelled to recognize that the market is showing some of the characteristics of a late stage in a big rally with, most prominently, a significant uptick in volatility.  (Just look at the market action today, March 19, if you like an example.) That means that while I’d still like to go with the market flow when I can, doing so requires more caution than it did in 2016 and 2017.

Continuing the turn around in the performance of the Jubak Picks Portfolio isn’t going to be a trivial task in 2018–especially with that 27% cash position. But that is my goal for the year.