The dividend growth investing (DGI) portfolio is the largest component of my active portfolio. DGI is considered an active strategy where the active part comes from stock selection rather than short term trading. As a matter of fact the intended holding period is normally very long, or even for everem> as Warren Buffet is known to prefer. I believe DGI offers the individual investor the best odds for achieving superior returns – both absolute and risk-adjusted, benchmarked to an investible index ETF such as SPY, although there may be some foreign stocks worthy of consideration. A statement like that needs to be justified of course, and “best odds” is by no means a guarantee. In this post I’ll describe why I believe DGI has this potential and the way I choose to practice it. Not everyone can copy what I do, two of the largest roadblocks being position size and skill/willingness to use options to enhance returns.
Diversification is necessary to remove idiosyncratic single stock risk so that at the portfolio level only (mostly) market risk remain. This paper does a good job explaining the research despite the unfortunate title. I care more about the standard deviation (SD) than the variance (as indicated by r-squared). SD speaks to the portfolio value whose downside volatility I want to minimize. I don’t care to have a portfolio with a high r-squared fit to the market because that will also remove extra sources of return (see discussion on factors below). At any rate, it takes only dozens not hundreds of stocks to reap the vast majority of the benefits of diversification. Note also the dates cited in the article were from the end of 1990’s. If anything, the correlation between stocks have increased due to the growing popularity of indexing. In a “perfectly efficient” market – which will probably never materialize – every stock will have the same risk-adjusted, expected return (source). In a market where all participant are indexers – which is even less likely to materialize – every stock will move in synchrony and provide identical returns. Both gedankenexperiments point to less number of stocks needed to approximate the diversification of the entire market as efficiency increases.
My own DGI portfolio is a work in progress. It contains about 30 stocks right now with an eventual goal somewhere around 50 stocks. The chart below shows the current sector weighting vs. S&P 500 market weighting that is more of a guide than a target. There is a concentration in consumer staples which are the classic stable, unglamorous, capital un-intensive compounding machines. The S&P site doesn’t yet include the newly independent REIT sector. It’s not an issue as I don’t have any REITs anyway (I have them in my passive portfolio). The portfolio yield is 3%, respectably above the 2% yield for SPY but not outrageously high. Indeed current emphasis is on the “growth” part of DGI and I consciously avoid the classic dividend stocks with high yield and low growth such as telecoms. I also make a point to put the high growth, low yield stocks like Disney ($DIS) and Nike ($NKE) in taxable and high yield stocks in tax-advantaged.
Dividends and Share Buybacks
The much venerated Economist magazine recently ran a series called “Six big ideas”, the first of which was information asymmetry and signaling. My very rough summary goes like this: there is information asymmetry in markets such as in used cars where the seller knows the true condition of the car but the buyer doesn’t; or the job market where the applicant knows his true abilities but the potential employer doesn’t. The solution is for the seller to provide an additional “signal” to the buyer as an indication of quality. In this perspective, job applicants use their degrees as “signals” of their dedication and talent, which is contrary to what education is normally thought of: a means to benefit society to make workers more productive. OK fine, I recall saying to myself as I was reading this, until I came to this “Aha” moment:
Signalling explains all kinds of behaviour. Firms pay dividends to their shareholders, who must pay income tax on the payouts. Surely it would be better if they retained their earnings, boosting their share prices, and thus delivering their shareholders lightly taxed capital gains? Signalling solves the mystery: paying a dividend is a sign of strength, showing that a firm feels no need to hoard cash.
There is plenty of evidence that an above average dividend yield (better yet a dividend grower, but not too high a yield which could indicate a recent stock price crash or an unsustainable dividend) correlates with above average returns.
Rather than a choice of either “total return” or DGI as some would have you believe, above evidence suggests DGI is a path to superior total returns. Now I understand that some may have a preference for share buybacks over dividends. The former does have advantages for taxable holdings. That’s why I prefer companies that do both. Share buybacks tend to be more volatile and will be cut first in tough times as companies defend dividends. To be more effective, share buybacks need to be counter-cyclical, tough to do but a sure sign of a management with capital allocation chops. Together dividends and buybacks make up shareholder yield which is a terrific indicator for the quality of a business. One of the main gripes I have about index funds is that I want to own more of the companies with high shareholder yield but the index fund takes that shareholder yield and distribute among the whole universe of companies. By owning the stocks directly, I don’t need to do anything in the case of buybacks; with dividends, DRIP, or let the dividends accumulate to buy another dividend paying stock are both possible. The latter is my preference since I want to have control over entry points and keep stocks in round lots.
Equal Weight and Inactivity
Supposedly a popular (long only?) hedge fund construction is the “50/50”: 50 stocks with 50% in the top 10. I don’t have the time to do the research to have that level of conviction so my default construction is equal weight as much as my desire for keeping round lots would allow. It is well known the equal weight S&P500 ($RSP) outperforms the cap-weighted $SPY, 10.88% vs. 8.75% annualized since May’03, despite higher expense ratio and turnover: 0.4%, 22% turnover vs. 0.09%, 2.77% turnover. Factor analysis (PortfolioVisulizer) shows $RSP having a market factor of 1.09, both size and value factors of 0.12, and a statistically insignificant alpha of 0.4%. Proponents of fundamental indexing would argued that anything but cap-weight avoids overweighting the recently-up-a-lot stuff.
On the impact of trading activity or lack thereof on performance, one stunning example of LONG TERM holding is the Voya Corporate Leaders trust fund (LEXCX, hat tip: Sure Dividend), a fund that started in 1935 with 30 blue chip stocks at equal number of shares (not exactly equal weight nor cap weight). It didn’t buy new stocks since inception and holdings only changed due to spin-offs or mergers. The result speaks for itself:
Source: Voya Investment Management
45 years (the data for the first 40 years is not available), no new purchases, the original 30 stocks turned into 22 and it handily beat both the Dow and S&P. In addition, a preponderance of academic studies suggest trading activity negatively affect mutual fund returns, mostly due the impact of trading cost (commission and slippage). My main take away is: weight quality stocks more or less equal (for sure not cap weight) and hold’em for a long, long time.
The MSCI paper Foundations of Factor Investing is a good intro to the topic. It defines a factor as “any characteristic relating a group of securities that is important in explaining their returns and risk”. Further, “the market can be viewed as the first and most important equity factor”. The paper went on to highlight value, size, momentum, low vol, yield and quality as the most important factors among the 300 or so that have been identified. Factors are important because they are sources of additional returns above that of the market. For some factors, such as value and size, the additional returns come with higher volatility so it can be argued that there is no advantage in risk-adjusted returns. But the low vol factor is certainly a slap in the face for CAMP. Academics refer to the factors that don’t fit their cherished theory as “behavioral” or “mis-pricing”. Also note that momentum which is one of the most robust and pervasive factors is totally contrary to the teachings of EMH. Still, you’ll see plenty of people in chat rooms who continue to regard CAMP and EMH the cutting-edge of portfolio theory. They are almost as bad as the academics who at first insisted it impossible to beat the market, then having found the factors, and then factors they had to call “behavioral” or “mis-pricing”, now use multi-factor analysis to RETROACTIVELY deny the existence of skill.
So much for the rant, the real point I’m trying to make is the typical DGI portfolio is likely positively loaded with value, dividend, quality and low vol factors; and slightly negative on size. If we take the new Nifty Fifty as representative of a DGI portfolio and compare them with the top stocks in $USMV, you’ll see plenty of overlap. Recent popularity of low vol and yield strategies probably means some future returns have been pulled forward, but to me it’s a confirmation of their long run advantage over the market.
One reason I keep stocks in round lots is to be able to write options for income generation. I keep things basic: sell covered calls or cash/margin secured puts on stocks I’d like to own. Expirations are typically 3-6 weeks out when time decay is the greatest. My goal is to have the options expire worthless so I normally stay with Delta <0.2. The low commissions at IB ($0.7 per contract, no base fee) make it worthwhile even with a couple contracts at around 1% premium. Further more, I use technical analysis to select entry points: sell calls when upside momentum is nearly exhausted or puts conversely. Being selective limits the number of trades I’m in at any given time. So far that hasn’t been an issue as I’m targeting only a couple percent of additional gains annually unlike some aggressive premium sellers targeting 20% per year. Compounded over years though I expect this to be a significant source of alpha. It’s very reassuring to know that I can generate a steady income from the portfolio when planning for early retirement. For monthly reporting purposes, gains from option trades are not included in the returns for DGI.
In summary, each of the above considerations has potential to enhance returns with varying probability of delivering. I’m under no illusion that any level of performance is guaranteed but I’m also content in having maximized my chances. Time will tell what return I’ll get in the end. By then it would be too late to change course – such is the investor’s condition.