Portfolio Protocol, Part 2
The protocol for the Workshop Portfolio is progressing. Today we examine the questions of overlapping stocks, rebalancing dollars invested within and among strategies, and the timing of the initial purchases. A good general guideline for deciding between competing rationales for any decision is Keep It Simple.
By Ann Coleman (TMF AnnC)
December 8, 2000
Our Workshop Portfolio protocol is taking shape nicely. It's not cast in stone yet, but we are developing a series of rules that should help guide us through the next year without resorting to emotional, ad hoc trading decisions.
I must say, it has astounded me to see just how many details need to be specified! This community exercise has proven to be even more useful and educational than I expected.
In yesterday's article we discussed the strategies that we will probably be using, our choice of brokers (Brown & Co.), and the timing of portfolio trades. Today, we will talk about overlapping stocks, rebalancing, and when, exactly, we should start.
One of the more interesting questions ("interesting" as in that old curse "May you live in interesting times") that we've discussed has been the issue of overlaps.
Because many of our stock screens start with the top-rated Value Line stocks and most include some kind of total return criteria, it's not unusual for one stock to pop up on more than one screen. We've tried to pick strategies that don't have much overlap, but we know it will happen. And with five strategies (we don't expect any overlaps with the Foolish Four!) the odds are good that it will happen frequently.
So what do you do? Do you buy the same stock twice? Buy it only in the longer-term strategy? Skip the duplicate stock in one strategy and substitute another stock from farther down the list? Buy both stocks when the strategies are for the same holding period but skip one if the strategies are of different holding periods?
See what I mean? A very interesting problem.
Although one can make a good case for any of the above arguments, the community ultimately decided on the KISS method -- Keep It Simple, Stupid. We keep it simple and follow the screens. If we end up holding iTaste (Ticker: YUMM) in three different strategies, so be it. We are going to trust the screens. After all these are the same screens that were "all" picking Qualcomm <% if gsSubBrand = "aolsnapshot" then Response.Write("(Nasdaq: QCOM)") else Response.Write("(Nasdaq: QCOM)") end if %> and JDS Uniphase <% if gsSubBrand = "aolsnapshot" then Response.Write("(Nasdaq: JDSU)") else Response.Write("(Nasdaq: JDSU)") end if %> early in 1999.
The next thorny issue was that of rebalancing. When and how do we equalize the cash among the three stocks of a single strategy or among all the strategies in a portfolio? With the Foolish Four, we simply sell the stocks that are not staying on for a second year and divide the proceeds between the replacement stocks. If one stock ends up way down or way up, we adjust that holding only if the trade size is large enough that the commission costs will be under 2%.
Again, the large number of strategies we are following tends to complicate things, and the relatively small size of the holdings dictates that we not worry too much about rebalancing just for the sake of holding equal dollar amounts of each stock.
In light of that and in keeping with KISS, it was decided to rebalance between screens once per year. In other words, we will attempt to start off each year with an equal dollar amount invested in each stock. When trading the non-annual screens, we will simply split the proceeds from any sales among the new purchases in that strategy. That would mean that we would not rebalance between any strategies until the end of the year even if they trade on the same day. Stocks that are being kept on will be allowed to "run."
But what about stocks that run wild? Last year Qualcomm and JDS Uniphase appreciated so fast that they rapidly came to dominate every strategy they were in. I don't think anyone considered that a problem at the time, but eventually, they began to fall back. That experience reinforced the results from backtests which indicated that it is best to trim the high flyers. The rule we decided on was to rebalance when any holding becomes more than 50% of its strategy. That seems sensible to me.
One other minor point regarding rebalancing (which illustrates just how thorough the community has been): Cash left over from trades will go into a pool that all strategies draw from. No strategy will "own" its leftover cash. This is the most efficient way keep as much cash working in the market as possible.
Which brings us to the final major discussion point: when we should actually start the portfolio. If you thought that this was a foregone conclusion (I seem to remember mentioning January 2), well, there's good reason to think again.
The CAPRS screen, which is new this year, relies on a "52-week total return" criterion. That criterion is only updated by Value Line once a month based on end-of-month data. Thus, the screen rankings that we run for CAPRS on December 29 will be based on data that is three weeks old.
The solution is to buy CAPRS on January 8. That will ensure that the data is fresh both this time and in subsequent quarters. But what about the other strategies? Should we stagger them as suggested by Todd Beaird in Staggered Strategies, Staggering Returns? That's an attractive idea. Todd makes a persuasive case, and that article is an excellent review of diversification in general as well.
If we were planning to add additional funds from time to time, I would definitely opt for staggered start dates. In this case we aren't planning to add cash, and the screen that appears to be most sensitive to starting at a bad time, the PEG, is running as a monthly, which effectively staggers its "start" date.
I don't think this issue has been fully resolved by the community. My preference would be to have at least a little bit of variety, perhaps starting all of the annuals on January 2, then adding another portfolio each week starting with CAPRS on January 8, then PEG, and finally Plowback. That would mean we would not be trading three strategies on July 1 and all six (including the Foolish Four) next January.
It's not that we find the trading onerous. But staggering the start dates give us a bit of protection from the dreaded "bad week" syndrome, and, frankly, keeps things interesting. Then again, there is a lot of appeal to keeping it simple and just starting everything on January 8. What do you think?
Some other interesting questions that have come up in the discussion:
The portfolio is too small for the number of strategies and too large to be attractive to beginning investors. Both criticisms are true. I regard those "problems" as excellent teaching opportunities. Both are also excellent examples of why no one should mimic this portfolio.
Another concern/suggestion was that we run a taxable portfolio as well. We can't do that as a "real-money" portfolio, but certainly it is feasible as a paper portfolio. It will need a protocol as well, and I expect it might differ from this one. For example, in a taxable account it might make sense to let winners run indefinitely, or at least for a full year (as long as they keep turning up in the strategy rankings). Selling off a big winner just to balance a strategy might not be so attractive if it triggers a short-term capital gain. Or a taxable strategy may want to adopt a "hold 'til drop" rule where stocks are only sold when they drop out of the top 10 stocks on a screen rather than the top 3. That would reduce trading costs and increase the likelihood of stocks being held for a full year. It can get rather annoying to sell a stock one month, then buy it back the next just because it slipped one level in the rankings.
Lots of interesting questions! But these can be worked out after we get the real-money, IRA portfolio set up since there will be no difference between the portfolios until the first sales take place.
Got a beef? Got a question? I urge you to join the discussion on the Foolish Workshop message board. See you there.
Fool on and prosper!
