US dividend tax rates are poised for significant changes for the second time in just ten years. Without Congressional action, the tax rate on dividends will rise to the investor’s marginal income tax rate at the end of this year. This means that the rate an investor pays for a qualified dividend could nearly triple from 15% to 43.4% (including the 3.8% new surcharge on investment income) at the highest tax bracket. This could lead to a decline in the after-tax dividend yields of stocks overnight (e.g., if one considers a stock with a $2.00 annual dividend on a $40 stock, the after-tax yield on that security is due to fall from 4.25% ($1.700/$40) to 2.83% ($1.132/$40) at the highest tax brackets).
Alternatively, after-tax yields would not depreciate as severely if the price of dividend stocks becomes depressed by the market’s anticipation of a tax hike. For example, consider a dividend stock valued on the basis of an after-tax dividend continuing in perpetuity. Using a stock with a $5.00 pre-tax annual dividend ($4.25 after-tax) and using a discount rate of 4.0%, the dividend stock would be valued at $106.25 using a dividend discount valuation approach. However, if the tax rate rises to 43.4%, the after-tax dividend becomes $2.83 and the stock would be valued at $70.75 (implying a share price decline of 33.4%).
Unfortunately, it is difficult to predict a reliable price impact in the market or on individual stocks, because investors’ tax situations differ. The aforementioned example only examines one investor who pays the highest, full marginal tax rate. Fortunately, there is empirical data available that illustrates how the market behaved when the dividend tax was moving in the other direction—in the first half of 2003.
2003: Resurgence in Dividends
By late 2002, dividend tax cuts had been publicly proposed and discussed in various venues. However, major milestones in the discussion were President Bush’s formal proposal for the elimination of the dividend tax (among other items) on January 7th, 2003 and the tax cut’s final passage into law (albeit in a different form than conceptualized in January) on May 28th, 2003.
How did these events affect the stock market and dividend stocks in particular? In the first six months of 2003, the S&P 500 rose 10.8%, but the sectors that would be most affected by the change—the ones with yields that were greater than that of the aggregate index (Consumer Staples, Energy, Financials, Telecommunications Services and Utilities) averaged a lower return (7.1%) than that of the lower yielding sectors (11.5%). However, examining valuations may be more appropriate because the higher yielding sectors could have had poor operating performance over the period. Looking at price-to-earnings (P/E) ratios, the “dividend” sectors (excluding the Telecommunications Services sector because its P/E ratio was negative at the start of 2003) experienced a 22.5% increase in their P/E valuation compared to -7.0% for the total S&P 500 index. Furthermore, even though dividend and non-dividend stocks have comparable returns over a 20-year horizon (with an average monthly, equal-weighted return of 0.82% versus 0.84%), dividend stocks’ P/E ratio rose 2.8% in the first six months of 2003—greater than the 1.0% decline for the aggregate index (the P/E ratio of non-payers was negative at the start of 2003, which makes a direct comparison difficult).
Given that the market showed some indication of an increased interest in dividend stocks in early 2003, did companies respond with dividend initiations or increases in dividend payouts? Though it is difficult to generalize the reason for the trend, there was a major resurgence in dividend payments in 2003. Over the course of the year, the number of companies paying a trailing twelve-month dividend rose from 353 to 364 (growth of 3.1%) in the S&P 500. While this growth rate seems trivial, the change in the dividend tax rate may have had a longer-term impact on companies paying dividends. The 20-year trough in the number of dividend stocks in the S&P 500 occurred in Q3 2002—a period that closely coincides with the time that the tax cut was first being discussed. In the years that followed, the number of payers steadily increased to 391 in May of 2008 (when the number dipped again during the financial crisis). Similarly, the number of companies increasing their year-over-year DPS payout troughed in 2002, but then steadily increased for a number of years until plateauing near 20-year highs in 2006 and 2007. Of course, there are many factors that drive market trends. Among other influences, the aftershock of the dot-com bubble bursting may have caused investors in the mid-2000s to place a higher value on mature investments paying dividends over young companies promising growth.
2012: External Factors Have Muted the Threat of a Tax Hike
Is the market now showing a reversal of the trends experienced in 2003? Not particularly. As previously reported in “Dividend Quarterly”, the P/E valuations of dividend stocks have actually increased relative to non-dividend stocks. The trailing price to earnings ratio (P/E) for non-dividend stocks (15.6) led the P/E of dividend stocks (13.5) by only 216 basis points at the end of Q2 2012, which is substantially lower than the 20-year monthly median of 1,343 basis points. Excluding months when non-dividend stocks had a negative P/E ratio, the premium at the end of Q2 marked the fourth smallest in the past twenty years (on a monthly basis). However, valuations in dividend stocks could be elevated because interest rates are historically low and investors are interested in higher-yielding securities.
While it is difficult to isolate changes in valuation due to fear of dividend tax increases from changes in valuation due to investors’ surging demand for higher yields, the long-term impacts that followed the signing of the dividend tax cut in 2003 may indicate how companies will respond to higher dividend taxes. The response could include a reduction in the number of dividend payers or a decline in the number of companies increasing DPS year-over-year, which is especially plausible since both metrics are at multi-year highs. The proportion of dividend-payers in the index broke the 80% threshold for the first time in the 2000’s in Q1 2012, and Q2 2012 had the largest number of companies increasing TTM dividends per share in at least 20 years (296) in the S&P 500.
At least one company has publicly commented on how an increase in the tax would affect the dividend decision. Darren King, CEO of Laboratory Corp. of America, said on a July 19th earnings call that a rise in the potential marginal tax rate on dividends “…really takes away from the attractiveness of a dividend. So some of it [the decision to pay a dividend] is going to depend on what are – what is Congress and the Executive Branch going to do…”.