Editor's Note, July 30: This article has been revised since original publication as author fixed an error in formula and corrected the tables and related content.
Seeking Alpha has recently published several interesting articles on the merits and risks of dividend investing versus purely growth-oriented stocks or even broader index investing. These articles have also led to many valuable comments from several SA members, adding to the richness of the debate with interesting insights. The purpose of my article is to present another view on this topic from the perspective of a small retail investor, considering taxation and compounding principles. For the purpose of this article, I refer to three types of common stocks: Dividend stocks that don't necessarily grow their dividends but declare them consistently ("DS"), Dividend growth stocks with a stated policy and track record of increasing dividends annually ("DG") and growth companies that don't declare any dividends (and don't intend to) but are more concerned with market valuation of their equities ("GO" stocks). The acronyms for easy reference are: DS - Dividend Stagnant, DG - Dividend Growth and GO - Growth Only. Some readers may wonder why I included DS as a category. That's because there are quite a few stocks that provide dividends but don't bother to increase them for many years (perhaps trying to be "safe" in all economic conditions and to avoid the risk of cutting a higher dividend in the future). The DS category also serves a useful role in the scenario analysis. There are many companies that one can cite as examples for each category (DS, DG or GO), so I will refrain from making this article company-specific to make a different point.
One of the fundamental points of contention in this debate seems to be the principle (advocated by the growth focused or index fund advocates) that the intrinsic value of a company drops by the same amount of dividend issued. This is best shown by example. If Company A and Company B both have identical prospects (10% long-term earnings growth and same valuation growth tied to earnings growth -- for simplified analysis), and Company A gives a 3% dividend yield and Company B doesn't offer any dividends, the question for a retail investor is, which is better? One camp would say that Company A's market cap should fall by the same intrinsic value of the dividend (3%). Since both companies are growing earnings, this fall in intrinsic value can be modeled by a correspondingly slower share price appreciation. Even Warren Buffett weighed in on this matter in his 2013 letter to Berkshire investors, saying the individual investor is better off selling a small number of shares (instead of getting dividends) to fund his income needs. This way, investors can create their own "income" (basically, capital gains) every year, customized to their needs. As a result, Mr. Buffett says, not all investors are forced to get the same income per share as in a dividend paying company. Of course, a key notion in Mr. Buffett's view is that the company can continue to invest 100% of its retained earnings to generate faster growth in cash flows in the future than the investor can, thus accelerating the internal compounding better than what a dividend investor may do on his own by "external" compounding. This notion also assumes that company management will always be able to find such above-average investment opportunities for all of their free cash flow. Berkshire (NYSE:BRK.A) (NYSE:BRK.B) certainly has proven itself in this regard, but what about other companies?
In the first scenario below, I have assumed that there are no (or minimal) dividend taxes for the retail investor. This is not an unreasonable assumption considering that qualified dividends of about $42,300 (for single filer) or $84,700 (for couple filing MFJ status) were effectively Federal tax-free for 2013, if there is no other income and just standard deductions (Source: Bankrate.com's 1040 calculator). Even in Canada, known for high personal income tax rates, dividend income until C$50,000 has practically no tax, as this 2012 article reports, so the figures should be a bit higher for this year. For a dividend yield of 3%, US qualified dividends from a rather substantial portfolio of $1.4 million (for a single filer) or $2.8 million (for MFJ filers) would be essentially federal tax-free, though the dividend income recipients may pay other taxes depending on where they live. Given that most retail investors are likely to have investment assets well below these levels, we can ignore the impact of taxes in the first scenario below. Also, for the purpose of this analysis, another assumption is critical. Investors in company A are assumed to reinvest their entire dividends, so that investors in both companies don't spend any current income from either dividends or selling shares. So, in both cases, compounding is in full force.
Scenario 1: DS company vs. GO company, zero or minimal dividend tax
No Dividend Growth, Market appreciation rate reduced by Dividend Yield. $1000 initial investment, 20-year horizon.
Even in this stark scenario, the investor in DS company A fares a bit better than GO company B. I say "stark" because one of the tenets of DG investing is to stay invested only in companies that grow their dividend and not keep it stagnant. The net value is 2% higher by the end of Year 20 for the DS investor (NYSE:A) versus GO investor (NYSE:B). Note that this scenario assumes that the market "extracts" the value out of Company A that issues dividends, by reducing the appreciation rate accordingly. So, the total return is the same as it is linked to the same 10% annual earnings growth. Growth shows up as appreciation rate net of dividend yield for company A and 10% share appreciation rate for the growth-only company. Also, the DS company gives the same dividends per share ($0.30) year over year regardless of earnings growth, so the yield declines each year while the share count increases every year as dividends are reinvested. In this scenario, I assumed that the market reduces the intrinsic value of company A every time the dividend is declared and applies a correspondingly lower appreciation rate on its shares. Many DGI advocates believe that there is hardly any reduction in the market value of a company after the dividend is issued.
Using Excel's Goal Seek function for dividend tax rate, I got an effective tax rate of 8.09% at which point there is no difference between Company A and Company B results keeping the Year 20 value constant for both. For tax rates below this level, there is a marginal benefit for Dividend investors in Company A, all else being equal.
Scenario 2: DGR company vs. GO company, no tax on dividends
Here, we assume that the DG company increases dividends by 7% each year (presumably with the intent to maintain a decent dividend yield as share prices appreciate due to underlying earnings growth). I chose 7% because that's the DGR at which the yield remains constant as the share price appreciates each year. The DG investor fares a bit better than the DS investor in the previous case and also, the GO investor, but what's an interesting conclusion here is that the end result (Year 20 ending balance) does not change much compared to Scenario 1. This is because of the critical assumption made that every time a dividend is declared, the market is reducing that year's share appreciation rate corresponding to the dividend yield for that year. If the market behaved like this, then it would not matter much if the company grew dividends by 0% (DS company) or 10% (strong DG company). If the theoretical notion that market growth is hampered by dividends (due to reduction in intrinsic value) is true, then the contention of 'dividends don't matter' advocates is right. However, there have been many articles on SA citing actual market experiences stating otherwise. The market appears to reward companies that consistently grow dividends because they send a strong signal that business is growing and the future appears strong. Only if we grant the fundamental premise of the 'dividends does not see it way makes it a case of theoretically correct, but practically wrong!
Scenario 3: DGR company vs. GO company, heavy tax on dividends
7% Dividend Growth, Growth Rate Reduced by Dividend Yield, 35% tax rate on dividends. $1000 initial investment, 20-year horizon.
Here, the scenario is the same as above but I assumed that the tax rate for dividends are the highest at marginal income (35% rate). This is way above the current 15% tax rate on qualified dividends and even above the 23.8% effective rate that high income wage earners pay on dividends since the 2013 budget deal. In addition, the scenario still assumes that the market reduces the appreciation rate of company A by the same yield annually as a "penalty" for issuing dividends.
Due to the heavy 35% tax rate on dividends, the dividend investor fares worse than the growth investor. If one applies the current 15% qualified dividend tax rate, this scenario is more favorable for the DG investor, as expected, but the fact remains that taxes reduce the end result to make it inferior to the growth-only investor. This is shown in this table below:
Again, the operative assumption here is that the market appreciation rate for Company A is reduced each year by that year's corresponding 'gross yield'. In other words, the market doesn't care what the investor pays in dividend taxes and automatically reduces the intrinsic value by the gross dividend issued by the company. The investor is assumed to reinvest after provisioning for taxes. Naturally, this is sub-optimal, in this case, neither the investor nor the company issuing dividends (much less wanting to grow them each year) would want to bother at all with dividends. Though this is a theoretical scenario, what this tells me is that despite the tax on dividends, the fact that companies pay and investors look forward to them is because it sends a strong signal that the intrinsic value is strong and perhaps even growing. This is entirely opposite to the notion of 'dividends don't matter' crowd because in all the scenarios above, we have granted the key concession that share appreciation rate is reduced by a corresponding amount in each year the dividend is issued.
In all scenarios above, I have not considered many other variables and market behavioral effects that may influence the share price growth of both companies. These include risk-free interest rates, company-specific growth opportunities for reinvestment of retained earnings, debt-equity ratios and other company-specific factors. A main assumption is that the 10% total return each year matches the 10% intrinsic earnings growth rate for both companies. Further, the scenarios assume that all earnings are equally converted into free cash flow by both companies, and their managements are equally adept. A risk factor against company B is whether investors will be able to reinvest all their earnings profitably to achieve consistent earnings growth in the second period (Years 11-20) as they may have been able to do in the first period (Years 1-10), because the attractiveness of reinvestment opportunities declines as the company matures and grows in size. The same factor applies to company A as well, but presumably, the consistent dividend (scenario 1) and increases (in scenario 2) would be a tacit acknowledgment by company A that they are trying to optimize between future earnings growth and current shareholder return. Higher interest rates may also reduce the valuation of dividend stocks more than growth stocks, presumably because income-oriented investors would prefer lower risk instruments for income generation, whenever available, compared to stocks, so the dividend yield must go up to maintain sufficient risk premium over interest rates. A counterpoint to this is that since dividends are reinvested at lower valuations -- as long as the earnings growth story of the company is intact -- the long-term value should be better than in the hypothetical scenarios above. Note that the above scenarios consider a predictable annual appreciation of stocks with no declining price periods, so the dividends are always reinvested under appreciating share price conditions.
Dividends are only part of the total return, but can be an important part. Even while granting the key tenet of the "growth only" proponents that the intrinsic value of a company is reduced by the same magnitude of the dividend, it appears that the dividend company delivers more value to the investor in case of zero or low effective tax environment on dividends. Should the tax scenario change towards higher taxes on dividends, then the growth-oriented investor fares better as long as the main operative assumption of 'dividends reducing intrinsic value' holds good (which may not as per many DGI articles). This is why this remains the biggest point of contention between the two sides. The fact that companies declare dividends and grow them consistently (despite the fact they are taxed) shows that there is merit to the DGI argument that dividends don't impact the intrinsic value or growth potential of companies in practice. This is the real world evidence that runs counter to the scenarios above.
Disclosure: The author has no positions in any stocks mentioned, and no plans to initiate any positions within the next 72 hours. The author wrote this article themselves, and it expresses their own opinions. The author is not receiving compensation for it (other than from Seeking Alpha). The author has no business relationship with any company whose stock is mentioned in this article.