Updated February 2021
INVESTING IN HIGH DIVIDEND EQUITIES
Abstract: Over long time frames the majority of equity returns have come from dividends and not from capital gains. Dividend yields grow slowly, about 1% per year and have a very low standard deviation, under 1%. Percentage dividend yields on equities can differ greatly over time with higher equity prices producing lower dividend yields. Long term, countries with high dividends have outperformed those with low dividends by over two times annually. High dividend equities are value stocks and track high book-to-market equities closely although DFA prefers the latter measure of value. Returns data for Wisdom Tree's globally diversified large and small capitalization high dividend equity portfolios is examined and compared with DFA's models.
Investing in high dividend stocks is a traditional idea in the financial advice business and was first widely promoted by Graham and Dodd back in the late 1930's. Dividends go in and out of fashion. When equity markets are struggling higher dividend paying stocks sometimes do better than the market and decline a little less when it is falling. This apparently occurs because investors reach for dividends if they can't be confident about capital gains. This was recently illustrated in 2011 where the S&P 500 went up and down during the year but closed within a fraction of a point from where the year began while companies paying dividends jumped 10.4%, far ahead of other indexes, including value and small cap indexes.
Dividends are paid from retained earnings or current earnings and the return from dividends, the dividend yield, is calculated by dividing the dividend paid by the stock price. Over the long run dividends have provided, by far, the bulk of equity investors returns. Dimson, Marsh, and Staunton et al. ("Triumph of the Optimists, 2002, Credit Suisse Annual updates, 2009-2012) found that from 1900 through 2005 the real total return from global equities averaged 5% with an average dividend yield of 4.5%, 90% of the total return from equities. As of early 2021 the S & P 500 gave a dividend yield around 1.5%. When dividends are low fixed prices are high. Dividends were by far the largest component of real very long term equity returns. Other studies for different time frames or equity asset classes show 65% or more of the total return in equities comes from dividends. The dividend discount model is a widely used approach for valuing companies based on the theory that a stock is worth the discounted sum of all its future dividend payments. Profits and earnings can be easily manipulated but dividends are mostly paid in cash and can't be faked. They are about as close to certain returns as a stock investor is going to get. As of 2019 dividends are usually taxed at the low Federal rate of 20% versus much higher taxes on interest earned from taxable bonds should investors be in higher tax brackets. For investors with income in the highest tax brackets the dividend tax rate is 23.8%, far lower than the maximum Federal income tax rate of 37%.
Dividends grow very slowly. For 1900 through 2005 Dimson et al. show that, equally weighted by country, the global equity market enjoyed no real (inflation adjusted) dividend growth. America performed better than most, achieving 1.3% annual dividend growth after inflation (5.0% nominal) and thanks to its heavy weighting dragged the annual global dividend increase up to 0.8% per year. Using the rule of 72 this suggests the time dividends take to double in inflation adjusted terms varies somewhere between 55 and 90 years. If we look at nominal dividend yields for the market they grow at about 5% annually. If buying when stock prices are high in valuation metrics and dividends are very low, as they are in 2021, investors should not count on dividend yields and income from them to increase rapidly. Dividend payouts look less appealing when interest rates and bond yields are at average levels or above them and interest rates and dividend payout percentages globally in 2021 remain very low by historical standards.
Dividend yields have varied greatly over time depending upon equity prices. S&P 500 dividend yields have been as high as 13.8% in 1933 and as low as 1.1% in early 2000. As of end 2016 the S&P 500 dividend yield was 2.05%, historically close to a 135 year low set in early 2000. In early 2021 they are now in the 1.5% range. In the first half of last century dividend yields averaged 5% and stock yields exceeded bond yields until 1959. Up until that time it was widely thought that stock investors should receive higher yields than bond investors since they were taking more risk and they did. Narratives regarding investment thinking often change over time or contradict each other. In the second half of last century the statistical relationship reversed between stocks and bonds only to reverse again in 2010 but at much lower interest rates. In 2008 the Federal Reserve pushed bond interest rates and along with massive increases in US money supply had kept rates close to that through 2021. In the second half of last century US bond yields exceeded equity dividend yields by an average of 3.9%. In early 2021 they are about the same.
Different time samples show very different dividend growth rates, yields, and dividend payout ratios. From 1871 through 1945 real dividend growth was 0.74% annually with an average yield of 5.31% and dividend payout ratio of 70.81%. From 2005 through 2015 real dividend growth was 4.97% with an average yield of 2.07% and the dividend payout ratio was a low 37.37% as capital gains became dominant in equity returns. There is no way of predicting from historical data when or how these statistics might change in the future.
Dimson et al working with data from Ken French of DFA for 1927 though 2010 found an average yield premium for high dividend stocks of 4.4% per year for 21 countries. They also found that high dividend stocks had lower standard deviations for those 21 countries with a standard deviation of 22.6% for high yield equities, 24.0% for lower yield equities, and 33.9% for zero yield equities. The Sharpe ratio, (return – risk free rate of return/standard deviation) of the high dividend strategy (0.42), was almost twice that of a lower dividend yield strategy and almost treble that of the zero yield strategy. Dimson et al. concluded that, "It is therefore hard to explain the superior performance of yield-tilt strategies in terms of risk, at least as conventionally defined." Dimson et al also ranked 19 countries for dividend yield from 1900 through 2011 and found large differences between yield quintiles with high-yielding countries outperforming low yielding countries by an appreciable margin. An investment of one dollar in the lowest yielding countries at the start of 1900 would have given an annualized return of 5.5% versus 13.4% for the top quintile of high yielding countries. During the 1970's when inflation and interest rates rose from the 3% range to over 10% high dividend equities outperformed the total stock market by 3.49% annually.
In 2016 both dividends and stock buybacks were funded by current earnings and rising corporate debt as companies borrowed at very low interest rates. In 2015 S&P 500 companies spent 106% of their net income on buybacks and dividend payouts and very high stock buyback levels have continued through up to 2021. This is beneficial to shareholders and reflects top management's desire to boost stock price and increase the value of their large personal grants of options and restricted stock. Insiders in America's corporations have been consistently selling their personal stock grants and exercising their stock options at or near all-time record levels since the 2008 financial crisis. Although only 20% or so of US stocks pay dividends they account for 77% of total US market capitalization since high dividend paying equities are concentrated in very large cap equities and indexes are market cap weighted.
Strong longer-term equity returns are far less likely if investors buy when dividends are low then when they are high. The best time to buy equities has been when dividend yields have been at or above 6%. If investors bought equities when dividends were high in the early 1930's, mid-1950's, or early 1980's they did very well over the next 10 to 20 years. From the early 1980's on dividends declined from high levels to record lows in the 2000's.
Dividends fluctuate and can be cut but their volatility as measured by standard deviation is very low compared to stocks. S&P reports a standard deviation of 15.46% for the S&P 500 for 1982 through 2012 while dividends on the S&P 500 for that period had a very low standard deviation of 0.47%. Over the last 30 years dividend payouts as a percentage of earnings dropped from around 70% to 30% in the S&P from equity price growth. DFA (March 2013) cites data showing that the propensity of firms to pay dividends declined globally from 71% of firms in 1991 to 61% of firms in 2012. During the 2008 credit crisis, the Wall Street Journal (11-7-2008) reported that 36 companies in the S&P 500 cut or suspended their dividends. From November 2007 through November 2009 dividend payouts from the S&P 500 declined 23% primarily due to big declines in former high dividend paying financial stocks. By mid-2010 dividend payouts recovered, set a new growth record over 15% for 2012, and have risen to record levels in 2021. In the five years through 2007 there were 298 companies which raised dividends and only 12 reductions or suspensions. Dividend suspensions or cuts are often seen as a sign of company weakness though in 2008 some relatively strong companies did so to increase cash cushions. Twenty S&P 500 utility stocks initiated or raised dividends in 2008. Since 1955 there has been an average of 15 dividend increases for each dividend decrease or suspension.
DFA's models aim to increase equity portfolio returns through tilting portfolios towards higher yielding slices of the world's equity markets, namely value stocks and small cap stocks, and incorporate profitability and momentum screens. DFA views dividends as a component of value but prefers book-to-market as a measure of value. Ken French of DFA analyzed returns on US stocks by dividend yield from 1927 through 2010 (Credit Suisse 2011 Global Investment Returns Yearbook). High yield equities returned 11.2% annualized, medium yield 10.3%, low yield 9.1% and zero yield 8.4%. This clearly illustrates the total return advantage produced by higher dividend equities. Data provided by DFA for the 1928 through 2011 period show that the top quintile of value stocks as defined by book-to-market value gave an annualized return of 12.1%, above that of high yield equities. DFA's view is that book-to-market is a better strategy for capturing total return in value stocks than dividends but both approaches capture returns in value and overlap. When small cap high dividend payers are taken out and the small cap premium is removed DFA's large cap book to market model underperformed high dividend stocks for 1927 through 2010 producing an annualized total return of 10.5% for US large cap value versus 11.2% for the top 30% of high dividend stocks which are mostly large caps. DFA's targeted small cap value index returned 13.2%. High dividend small cap growth data is not available for 1927 through 2010 so a direct comparison between high dividend and book-to-market models cannot be made for that period.
Jeremy Siegel (Stocks for the Long Run, 4th edition, 2008) argues that there are conceptual problems with using book-to-market as a value criterion because it does not correct for changes in the market value of assets, capitalize research and development expenses, or capture the value of intellectual property which is increasingly important for today's companies. He references unpublished work done by his research team showing that from 1987 through 2006 book-to-market underperformed dividend yields, p/e ratios, or cash flows in explaining returns. Siegel is a principal in Wisdom Tree. Wisdom Tree initiated most of its high dividend index offerings in 2006 and 2007 and offers over 30 high dividend funds covering US, international developed and emerging markets. Tilting portfolios towards dividends is sometimes referred to as fundamental weighting and as of 2019 dividend yields across their high dividend funds vary from around 3% to 5%.
We looked at the performance of Wisdom Tree's DHS US large cap high dividend index from inception in June 2006 through Fall 2012 and compared it to DFA's large cap US value index, DFLVX which is based on a book-to-market metric. Over that period DHS declined about 8.0% versus 2.9% for DFLVX but dividends in DHS were about 2% to 2.5% higher annually than DFLVX for the six year period. That added about 15% to total return and put DHS slightly ahead of DFLVX in total return. During the 2008-2009 bear market DHS declined about 66% to DFLVX's 63%. From 2009 through 2014 DHS gave a price return of 83.2% to 95.8% for DFLVX but dividends from DHS were higher and total returns were close to equal.
Wisdom Tree's US small cap high dividend index, DES, declined about 8% less from inception in June 2006 through Fall 2012 compared to DFSVX, DFA's small cap value fund. In addition, DES's dividends were about 2.5% to 3% annually above DFSVX and added another 15% to 18% to total return. During the 2008-2009 bear market DES declined about 66% to DFSVX's 69%. From 2009 through 2014 DES gave a return of 89.7% to DFSVX's 83.7% along with higher dividends.
Wisdom Tree's large cap international developed markets high dividend index, DTH, declined about 4% more from inception in June 2006 through Fall 2012 than DFA's large cap international value index, DFIVX, but its higher dividends produced a slightly higher total return. During the 2008-2009 bear market, both funds declined about 65%. From 2009 through 2015 DTH returned 9.8% to DFIVX's 11.9%.
Wisdom Tree's international small cap index, DLS, gave a flat return from inception in June 2006 through Fall 2012 compared to a decline around 16% for DFA's international small cap value fund, DISVX, and it provided higher dividends. During the 2008-2009 bear market both funds declined about 64%. From 2009 through 2014 DLS returned 29.4% to DISVX's 27.7% along with higher dividends.
Wisdom Tree's high dividend emerging market index, DGS, produced a return about 35% higher than DFA's emerging value index, DFEVX, since inception in October 2007 through Fall 2012 and offered higher dividends. During the 2008-2009 bear market DGS declined about 57% to DFEVX's 66%. From 2009 through 2015 DGS gave a return of 9.3% to a loss of 10.8% for DFEVX.
The recent returns data cited here for DFA versus Wisdom Tree models of value are far too short a time sample from which to draw any statistical conclusions. They suggest that total returns from high dividend stocks are fairly close to those of high book-to-market stocks but there are exceptions in the data. US large cap equity data back to 1927 from Ken French at DFA suggest that large cap high dividend stocks are likely to return about the same as high book-to-market stocks and are another avenue by which the value premium can be captured. High dividend stocks offer a little more in current income and a little less in long term capital gains but still return well above common stocks. Differences between the two approaches in capturing the value premium are not large and both approaches have about the same volatility and drawdowns. When most charts for comparable DFA funds and Wisdom Tree funds are overlapped on each other daily, weekly and monthly share price changes usually track closely over time. This also suggests both are capturing a similar market dimension, value.
We have many retired clients who require income from their portfolios and have been frustrated by very low interest rates on bonds since 2008 and historically low dividend yields on common stocks. An investor who wishes to augment income from their account through high dividend equities has many options. High dividend equity indexes like those from Wisdom Tree, utility indexes, Reits, and MLP's in oil and gas pipelines offer good yields in 2021 relative to other equity asset classes but Reits and MLP's have relatively high price volatility.
It is essential for investors to keep in mind that high dividend equities are equities and rise and fall with the broader equity market. High dividend equities are not identical to bonds nor do they provide the relatively low volatility of bond indexes. Roughly, historical data indicate equities are about 6x to 7x more volatile than fixed income with fixed income fluctuating about 10% to 12% over the 1995 to 2020 period. That said, a high dividend global equity portfolio can be structured in 2021 to yield about 3.0% to 6%. Equity dividend taxation rates are at 20% or 23.8% for those in the highest marginal tax brackets as of 2020. Dividend paying stocks were taxed at income tax rates until 2003, well above capital gains rates. As of early 2021 high dividend equity dividend yields are about 0.5% to 1% above those available in short investment grade corporate bonds and close to those available in intermediate investment grade corporate bonds but all yields are at or near record lows.