Dividend-Payers Versus Nonpayers
Conventional wisdom has long held that dividend-paying stocks tend to outperform stocks that do not pay dividends.
The Forecasts’ research shows that nonpayers have outperformed by a wide margin since 1990.
On the surface, we have the makings of a conflict. But it turns out both sides are correct.
Nonpayers win in recent years
The average dividend-paying stock in our Quadrix® research universe delivered a 14.1% return in rolling 12-month periods since 1990, representing nearly 20 years of returns. In contrast, the average stock that does not pay a dividend returned 19.3%. To dissect those returns, we broke the data set into four periods, starting with the five-year period starting in January 1990. Dividend-payers lagged by a substantial margin in three of the four periods.
We aren’t the only ones who noticed this phenomenon. A Charles Schwab ($15; SCHW) study of stock returns from 1990 through 2008 found similar underperformance for dividend-paying stocks. Two factors that contributed to this trend are:
• The market bubble of the late 1990s, during which investors flocked to technology stocks and others that did not pay dividends. Buying tends to beget buying, and the returns of non-dividend-payers in 1990s were astronomical.
• The real-estate and financial meltdown of 2007 and 2008, which hit the dividend-rich financial sector the hardest.
In addition to arithmetic average returns, we also calculated geometric average returns, which reflect the change in wealth over time and do a better job of accounting for risk. The spread between the returns of dividend-payers and nonpayers is smaller as measured by geometric returns, but it still favors nonpayers, which delivered a geometric average return of 15.5%, versus 12.5% for dividend stocks.
Dividends hold long-term edge
When you look back further than 20 years, the picture changes. According to researcher Kenneth French, from 1928 through 2009, stocks that do not pay dividends delivered annualized returns of 8.1%, versus 8.9% for the 30% of U.S. stocks with the lowest yields and 10.9% for the 30% with the highest yields.
To put that in perspective, $1,000 invested in high-yield stocks at the start of 1927 would have grown to $4.7 million by the end of 2009, more than eight times the value of a portfolio containing nonpayers. Other studies of market returns starting in the 1960s and 1970s illustrate similar trends. Before 1990, dividend-paying stocks did provide superior returns.
Unfortunately, we cannot predict what the future holds. Have we entered a new reality in which non-dividend-paying stocks will outperform dividend-payers over the next 20 years? That’s a tough call. But the answer to that question is less important than the lessons we can glean from a review of history. Here are four points to remember:
• Income stocks less volatile. Dividend-payers have delivered superior risk-adjusted returns. While nonpayers in our Quadrix universe outperformed dividend-payers by an average of more than five percentage points since 1990, those returns were far more volatile. The information ratio measures return divided by standard deviation to simulate return per degree of risk. Dividend-paying stocks managed an information ratio of 0.77, versus 0.63 for nonpayers.
• Highest yields not the best. Among dividend-paying stocks, the absolute level of the yield has had little bearing on performance in the last 20 years. In rolling 12-month periods since 1990, S&P 500 stocks yielding more than 0% but less than 2% averaged 11.8% returns. Stocks yielding 2% to 4% averaged 11.7% returns, and those yielding more than 4% averaged 11.8% returns.
• Low payout ratios yield high returns. A Credit Suisse study found that among dividend-paying stocks with similar yields, those with the lowest payout ratios tend to outperform. This makes intuitive sense, since a low payout ratio bespeaks a company with sufficient earnings power to grow its dividend over time while still investing in its business.
• Dividend growers outperform. A Ned Davis Research study found that from January 1972 through December 2009, S&P 500 stocks that grew their dividends returned an annualized 9.3%, versus 8.6% for all dividend payers in the index, 7.1% for stocks that did not change their dividends, and a negative return of 1.3% for stocks that cut or eliminated the payout.
We considered the above points when selecting the stocks in the table below, all of which offer dividend growth, payout ratios below the average for stocks in their yield category, and solid Quadrix Overall scores, which suggest strong fundamentals. Three are reviewed below.
IBM’s ($130; IBM) dividend has climbed at an annualized rate of 22% over the last three years. The company pays out just 25% of its profits in dividends, well below the 47% average for stocks yielding from 2% to 3%. In addition to the dividend, IBM also spends aggressively to buy back its own stock, trimming the share count by more than 20% over the last five years. IBM’s balanced approach extends beyond cash deployment to an operating strategy broad in both product mix and geographic presence. The hardware and services giant has delivered consistent double-digit growth in per-share profits over the last five years, along the way increasing its returns on equity and investment. Wall Street projects June-quarter earnings of $2.58 per share, up 11%. IBM has topped the consensus profit estimate in each of the last four quarters. Slated to report quarterly results July 19, IBM is a Focus List Buy and a Long-Term Buy.
Despite their generally strong balance sheets, less than 15% of the technology stocks in our Quadrix research universe pay dividends. Tracking down a decent yield is even more difficult — the average semiconductor stock yields 0.7%. Enter Intel ($21; INTC) and its juicy 3.0% yield. Intel began paying a dividend in 1992, and the payout has nearly doubled over the last five years. The strength of the semiconductor giant’s balance sheet — and its recent dividend-growth history — suggest the payout will continue to rise in the future.
In the June quarter, Intel earned $0.51 per share excluding a $1.45 billion fine from the European Commission, up 183% from the year-ago quarter and $0.08 above the consensus. Revenue rose 34% to $10.77 billion. The midpoint of management’s sales guidance for the September quarter is $11.6 billion, topping the consensus of $10.92 billion. Intel is a Focus List Buy and a Long-Term Buy.
In one form or another, Rogers Communications ($35; RCI) has existed since 1920, first as a movie-theater operator and now as a leading wireless carrier and cable-TV provider in Canada. Not until 2000 did the stock start paying a dividend, but the current yield is a generous 3.3%, well above the broadcasting group’s average of 0.9%. In the last few years, management has increasingly used the quarterly distribution — growing ninefold since the start of 2007 — as a vehicle to enhance shareholder returns. Rogers’ dividend has risen more than 20% over the last year, yet the payout still represents just 43% of trailing 12-month earnings, well below the average of 72% for stocks yielding from 3% to 4%. Earning our top Quadrix Overall score of 100, Rogers is a Focus List Buy and a Long-Term Buy.