Many readers liked the Triple-Play Income Strategy I developed using StockScreen123 and presented on Seeking Alpha (click here for the most recent article). But one commenter raised a very important point: trading costs.
While this is always an important consideration, it’s especially so for income investors. If such outlays amount to, say, 2% per year, many equity investors can look beyond that using a line of reasoning to the effect that overall portfolio capital gain potential is sufficiently great to make it worthwhile. For income investors, expected annual returns, while still not precise, are a lot more visible making them more sensitive to pretty much any commission burden.
As a refresher, Triple-Play refers to three components that make up a full-fledged income strategy. These reflect three general approaches to income-stock selection.
- The Core portfolio is one that aims to achieve a yield at or near an “ideal” level determined by market conditions which, based on where we stand today, is in the neighborhood of is about 6.5%. That’s a general target, not a specific rule (as we’ll see below, this week’s list comes in lower). If reality matches theory, a portfolio of such stocks should produce capital gains or losses averaging close to zero. As it turns out, real-world performance has often been better owing to market inefficiencies.
- The Dividend Growth portfolio sacrifices some yield in order to take advantage of capital-gain potential expected to come about as a result of payouts that rise over time. Many income-seekers find it hard to embrace a strategy that suppresses yield. That’s understandable. With interest rates and yields so low now, it seems terribly painful to give up a single basis point. But growth is an important reason why anyone would ever consider stocks as opposed to bonds (which usually yield more). If interest rates rise, as seems likely going forward, investors may come to appreciate rising dividend streams in ways unlike what we’ve experienced in this generation.
- Finally, the Prudent Yield Hog portfolio is aimed at people who can handle risk and really want to max out on yield. The risks here should not, however, be brushed off as due-diligence boilerplate. They’re real and if one is not careful, substantial losses can result as companies deteriorate and dividends get cut or eliminated. I’ve found though, that risk can be controlled through a vigorous fundamental-oriented protocol I use to pre-qualify stocks before looking at the highest yields.
I’ll provide updated lists for each of these portfolios at the end of the article. But the discussion today will focus on trading costs, what you can do to reduce them and how such measures may impact overall returns. For impatient readers, here are the spoilers.
- The Core portfolio strategy can easily be modified in ways that reduce commissions.
- Modifying the Prudent Yield Hog is more challenging, but doable for those willing to put forth some extra effort.
- As to Dividend Growth, forget about it: If you can’t handle the trading costs associated with a 15-stock portfolio rebalanced once every three months, I suggest looking for a different, lower-cost, brokerage firm or a mutual fund that follows this strategy.
Each portfolio holds 15 positions. I’m in all three of them. But one could focus on a single strategy or pick two out of three. I trade at FolioInvesting.com so for me, commissions are irrelevant (a $290 annual fee allows me to trade all I want, commission free, if I’m willing to have my trades executed during one of the firm’s two-times-per day trading windows (one at 11 AM ET and the other at 2 PM ET). What I’m, doing, here, could be considered a do-it-yourself version of what Bloomberg recently described as “portfolios-to-go.”
I can’t know what your specific trading-cost burden would look like since I don’t know what firm you use nor do I know the size of your account. But you can easily make an estimate by following an approach like this:
Assume there’s $50,000 in one of these 15-stock portfolios. Every time it gets rebalanced (four times per year), about three quarters of the portfolio turns over, an assumption I base on the average I’ve seen from the backtests. Each such occasion involves two trades: selling one stock and buying another. If I turn over all stocks every quarter, that would come to 120 trades per year. If I’m at E*Trade or Ameritrade, where the per-trade commission is $9.99, that would mean I’d be paying $899.10 per quarter, or $3,596.40 year in commissions. For a $50,000 portfolio, that would be amount to 7.2%, a figure that’s downright frightening. It’s absolutely unacceptable. Even a $100,000 portfolio would involve a still-terrible trading-cost ratio of 3.6%. On average, the portfolio turns over about 75% per year, so cutting the number of trades accordingly will help a bit, but still leave us with an unacceptable burden. Going to conventional firm with lower commission rates (such as Trade King, which charges $4.95 per trade) can help.
I could simply dump the screen- and rank-based models and go to a regular mutual fund (z-z-z-z-z) or pick stocks the traditional way, one at a time. The latter doesn’t sound so bad. It’s what many investors do. I will tell you, however, that after having been an investment professional for more than 30 years with 18 of them as a professional securities analyst and over a dozen years of intensive experience with numerical models, I have to confess that as much as it hurts my ego to say it, the latter wins hands down. That’s so for me, it’s so for most people I know, and it’s so for more pros than many realize. That’s why I go the extra mile when it comes to controlling trading costs and invest through a firm that makes it feasible for me to assume zero.
That said, I have to acknowledge that most readers pay commissions on a per-trade basis, and may not be inclined to change to a lower-cost brokerage firm. So it’s necessary to consider the two main ways an investor can cut costs: (1) trade less frequently by stretching the holding periods; or, (2) shrink the number of positions in the portfolio. We’ll consider both approaches.
To set an overall context, imagine a $7.95-per-trade commission rate (based on Fidelity brokerage; note that you can trade at Schwab for $8.95 and at Vanguard or Scottrade for $7.00). I’m also going to assume assets average $75,000 over a year. That starts us off with an assumption that trading costs eat up a too-high 3.8% of the portfolio. So let’s see what we can do about that.
Stretching the Holding Period
Longer holding periods can help a lot. Trading every six months instead of once per quarter would cut the hypothetical commission burden from 3.8% to 1.9%. Trading once per year brings costs down to about 1% of average assets.
Testing and real-world experience so far with the Triple Play as designed (15-stock portfolios rebalanced four times per year) suggests the returns could be high enough to absorb such trading costs. The question, however, is whether a longer holding period compromises investment performance.
Most of my fundamental models are tested with an assumed four-week holding period, and I’ve been using the same time frame for live implementations. Income investing is different. Here, I like to assume holding periods of three months. Most dividend-paying companies follow quarterly payment schedules, but there is no uniformity as to when, during the course of a quarter, each company makes its payment. So if I hold a portfolio of yield-oriented stocks for less than three months, there’s a chance I’ll sell some before actually getting a dividend payment. That’s a very big deal since dividend payments comprise a large portion of expected return.
I should also mention that I don’t actually buy stocks intending specifically to hold for only four weeks or three months. What I actually do every three months for income stocks is re-run the model. At that time, if the stock is still there, I continue to hold. If it’s no longer on the list, I sell. Naturally, I also initiate positions in stocks that are new to the list. If a situation is reasonably stable, it is more likely to pass my model time after time. So while some fans of long-term investing might think in terms of a five-year holding period, I would instead think in terms of three-month holding periods which just happened to produce the same answer (i.e., that the stock made it onto my list) twenty times in a row. Put another way, I don’t do buy-and-hold. I do buy-and-review.
The question, here is whether we can get away with stretching the review period for income stocks from three months to six months or even a year (the need to assure collection of dividend payments means the review period should be a multiple of three months).
I went back to StockScreen123 and re-did the backtests looking at six-month and one-year review periods. I did that for the full ten-year trial period (3/31/01 – 3/31/11). I also carved out a smaller period that didn’t have any spectacular market ups or downs (12/31/03 – 12/31/07). Finally, I looked at a period dominated by the crash (12/31/06 – 12/31/07) in order to see how longer periods impact the model’s ability to cope with crisis.
Let’s now examine the differences in annualized performance for the Core, Prudent Yield Hog, and Dividend Growth portfolios when holding periods are reviewed beyond three months. All of the tables that follow reflect changes in share price performance only. Assume no systemic changes in average yields.
Table 1 () suggests we could reasonably stretch the Core-portfolio review/rebalance periods from three months to six when we look at the full test period, and the challenging 2007-08 period.
It seems strange that the more-or-less “normal” period, the one that runs from 2003-07, should show such a big difference. On further examination, however, I saw that the oddities were concentrated in mid- to late-2007, which overlaps the beginning of the financial crisis. Some REITs took especially bad spills.
Test results must always be interpreted in the context of common sense. On that basis, I would suggest you could accept six months or even 12 month rebalancing intervals so long as you pay attention to your portfolios more closely. If any stocks act oddly and look to be moving meaningfully out of synch with the others, break ranks with the model and sell them before the next review date. If you effectively monitor, it looks like the longer holding periods will have little systemic influence on your results. They could be a bit better. They could be a bit less. Absent special cases such as the beginning of the financial crisis, the differences do not appear to be systematic.
That should not be a surprise. The companies whose shares are most likely to wind up in the Core portfolio tend to be quite stable. Even if they’d fall out of the model in the course of a three month period, they’re still likely to be close to the borderline and not be significantly better or worse than those that take their places, and may just as easily come back when the model is rebalanced next time around.
Table 2 () suggests a more-or-less similar outcome for the Prudent Yield Hog except that here, you’d really need to be extra vigilant about major crises and rely on your own judgment to recognize them as they unfold and bail out of stocks before a six- or 12-month holding period has run its course.
For companies to get into the Prudent Yield Hog model, the stock price will likely have already digested some worst-case fears (thus producing the high yields). We’d expect some quick gains if events unfold in such a way as to make the market less fearful. That scenario would be fine. So, too, would inertia: no change in fundamentals, resulting in continuing steady company and stock performance and us continuing to collect our dividends. The problem would be noteworthy deterioration. We’ve seen from basic model testing we can live with having the model eliminate such stocks once every three months. But as the holding period is stretched, it becomes more important to rely on yourself to notice something going badly wrong and act.
Finally we have Table 3 (), which examines the Dividend Growth model.
The first thing I note is that all differences are negative, as opposed to Tables 1 and 2, where we saw mixtures between negative and positive. This, right away, suggests something more systemic. Also, considering that it will be hard for many investors to get trading costs much, if at all below 1% per year with a 15-stock portfolio, and considering that the stocks in this portfolio are likely to yield in the neighborhood of 3%, we definitely can’t afford any of the return diminutions we see in the Table.
Here, again, the outcomes we see are related to the nature of the stocks. We’re dealing, for the most part, with garden-variety business companies. Perhaps they’re a bit more stable than some stocks, but much less stable relative to the sort of firms we see in the Core portfolio and even the Prudent Yield Hog portfolio (crisis aside). We can’t wait six months or a year to rebalance this kind of portfolio. Information changes too quickly. (Indeed, I’d prefer to rebalance this portfolio every four weeks but don’t do so because, as noted, we must be in each stock long enough to collect a quarterly dividend.)
In sum, if you want to stretch your holding period, you can do so with the Core portfolio or the Prudent Yield Hog so long as you increase your personal vigilance as time passes. You’re not looking to jump on every little event. You’re looking for significant aberrations. If you’re comfortable with that sort of monitoring, feel free to cut your trading costs by holding Core and Prudent Yield Hog stocks for up to a year before referring again to the model. If you’re want to pursue a Dividend Growth strategy, you’ll have to think seriously about your brokerage arrangement and place cost considerations ahead of whatever other comfort factors you may have with your present firm. If you don’t want to make any such changes, consider looking for a mutual fund that emphasizes dividend growth.
Holding 5 Positions Instead of 15
The other way to reduce trading costs is to cut the number of positions you will hold in the portfolio.
This is not something to be done lightly. There are those who preach the virtues of focus portfolios, placing all or most of you money in a few “best ideas.” This sounds great on paper. But it often underestimates how challenging real-world securities analysis can considering we’re mere mortals who, regardless of how much research we do, cannot see into the future. Consider, for example, the track record of Robert Hagstrom, one of the more well-known Focus pundits who wrote many books about Warren Buffett including one entitled The Warren Buffett Portfolio: Mastering the Power of the Focus Investment Strategy. His words sound great, but the mutual fund he manages based on those principles, Legg Mason Capital Management Growth Fund (LMGTX), is pedestrian at best, as can be seen from a track record that often reveals high risk and average to below-average returns culminating, as of this writing, in a one-star (lowest) Morningstar rank. Indeed, Morningstar’s current analyst report makes reference to Hagstrom’s ongoing increase in the number of portfolio positions as well as his movements away from Buffett’s emphasis on consistent businesses.
Reducing the number of positions in a portfolio is a matter to be handled with great care, as opposed to simply following the pronouncements of an author or blogger everyone in the world thinks knows how to invest like Warren Buffett.
Interestingly, Table 4 () suggests the kinds of companies we’re most likely to see in the Core portfolio may be most well-suited to focus portfolios. (The table shows changes in share price performance. Assume no systemic changes in average yields.)
This actually makes sense if you look at the kinds of stocks that typically make the list. The variety is very narrow, with communications and pipeline firms often dominating. Not only do they tend to be stable, there’s often little difference between firms in the same industry so for better or worse, we don’t often gain much by adding more names. And actually, more often than not, we benefit by narrowing since I always pick from the top based on the ranking system used by the model, meaning I’m aiming for the best set of fundamentals. There was one bad showing, the annual rebalancing during the crisis period. Presumably, though, as you stretch holding periods, you’ll be more vigilant about problems that suggest selling before expiration of your chosen holding interval.
Table 5 () shows the impact of reduced portfolio size on the Prudent Yield Hog portfolio.
This is a bit dicey. It looks like longer holding periods for 15-stock portfolios may be a better way to reduce commissions. But if you do want to reduce the number of positions in addition to or in lieu of longer holding periods, be especially vigilant about the impact of bad economic environments. We see, in the 12/31/06 – 12/31/08 period, how much more pain a Prudent Yield Hog investor can feel when reducing diversification. Interestingly, though, during normal market periods, it might be tolerable.
Also, as you look at the performance changes (usually performance penalties) from less diversification, bear in mind that in this model, after pre-qualifying companies based on fundamentals, we sort the remaining candidates based on yield and pick from the top. Accordingly, the average yield for a five-stock Prudent Yield hog portfolio can be expected to be higher, often significantly higher, than that of a 15 portfolio. In the list below, for example, the average yield of the full 15 stocks is 10.00%. The average yield of the top five is 12.22%.
But again, and I can’t say this often enough, if you’re going to pursue a commission-reduced version of the Prudent Yield Hog, you really have to be on your toes when it comes to watching for problems!
Table 6 () shows that when it comes to the Dividend Growth strategy, it doesn’t even pay to be on your toes. It’s better to simply forget the whole thing. (As with the Core portfolio, we’re looking at changes in share price performance only.)
Here is where we’re looking at a broader variety of regular businesses, the kinds of situations that make real life so difficult for so-called focused investors. (Now we’re getting into, or closer to, the kinds of companies Hagstrom may have been looking at for his mutual fund.)
Many commentators (including me at times) extol the merits of modest yielding stocks likely to benefit, over time, from dividend streams that grow briskly. This really is a good strategy. But as you can see, it can also be an expensive one to implement successfully. You’d need a large asset base and/or a very-low-cost brokerage operation lest the drag of trading costs and/or inadequate diversification more than offset the benefits of dividend growth.
Wrapping It Up
This should give you a good idea of how you, as an income investor, can pursue interesting yield-oriented strategies while protecting your account from being decimated by trading costs. For the Core and Prudent Yield Hog strategies, you can extend review/rebalancing periods and/or cut the number of positions down as far as five so long as you increase your vigilance and get rid of any company/stock that looks like it’s coming under unusual pressure, especially with Prudent Yield Hog selections. As to Dividend Growth, you may need to consider switching to a really-cut-rate brokerage firm, or looking instead for an appropriate mutual fund.
So now, the curtain call: the stocks that currently appear in the Triple Play models. (As noted above, if you need more of a refresher regarding how these models work, you can find one by clicking here.)
Table 7 () lists the current selections for the Core portfolio. The top five are shown in green and are the ones that should be considered by those looking to cut trading costs by reducing portfolio size.
Table 7 – Core Portfolio
Table 8 () lists the stocks that currently make the Prudent Yield Hog listing. Here, too, those wishing to cut the number of positions to reduce trading costs should focus on the top five stocks, shown in green.
Table 8 – Prudent Yield Hog Portfolio
Finally, for those who can handle the trading costs associated with a 15-stock portfolio rebalanced every four weeks, Table 9 () shows the current listing for the Dividend Growth model. As discussed above, I would not recommend focusing on only the top five, or any other subset of this listing, nor would I recommend stretching the rebalancing periods.
Table 9 – Dividend Growth Portfolio
Disclosure: I am long ANH, BLK, CL, CLMT, CMO, CODI, COP, DRI, DX, ENP, PAA, PDLI, PG, RTN, SPH, TWGP, VLCCF, WAC, WSO.
Disclaimer: The positions I own reflect my own re-balancing schedule and, hence, do not entirely match the lists presented here.