Financial theorists often talk about maximizing shareholder value when discussing dividends. They argue that because of the double taxation of dividends, shareholder value is maximized by retaining earnings and avoiding the second taxation. The Modigliani-Miller theorem goes one step further and states that even if dividends are not taxed, a company's dividend policy is irrelevant to the value of the company. The underlying assumption is that an efficient market is available to the shareholder to offer a reasonable price for the shares. A shareholder can simply sell when necessary. They would have to pay capital gains taxes, but these are often much less than the tax rates on dividends. With taxes minimized, value is maximized.
The problem with this approach is it completely ignores the value a dividend policy has for both the shareholder and the company in general. First and foremost, dividends provide much needed route to reap the returns of stock ownership. There are only two ways for shareholders to obtain returns from their stocks: market price and dividends. It seems imprudent to depend completely on one and consider the other worthless. When entering an investment, I'd like to have as many exits available as possible. So, although there may be added expenses associated with dividends, there is also added value in the form of an additional mechanism of obtaining returns.
Due to its lower volatility, dividends may be a superior mechanism for obtaining returns over capital gains. This is largely due to managers managing distributions. In good times, they raise dividends slower than earnings growth, and use the excess capital to supplement distributions in difficult times. The result is a much more steady and reliable income stream than if one were to rely solely on the market for returns.
This phenomenon is evident in the dividend history of the S&P 500 over the past 50 years. Since 1960, the standard deviation of annual price changes in the S&P 500 has been 16.62% while the annual changes in dividend payments of the underlying stocks has had a standard deviation of only 6.31%. Furthermore, the S&P 500 has had 13 down years in the period while dividend payments have dropped in only 7 of the years. The largest annual drop for the S&P 500 was 38.49% in 2008 while the associated drop in dividends was only 20.78%. This lower volatility of dividend streams makes them quite compelling as a source of predictable returns for shareholders. Data available here.
Another argument often levied against dividend paying companies is that the capital returned to investors could be better used to grow future earnings. Some go as far as avoiding dividend stocks completely. They argue that paying a dividend is an admission that the company has few growth opportunities. Who would want to invest in a company with poor growth prospects? But, evidence suggests that not only is this argument false, the exact opposite appears to be the case. According to an article by Robert Arnott and Clifford Asness published in the Financial Analysts Journal, higher dividend payouts are actually associated with higher earnings growth rates:
The historical evidence strongly suggests that expected future earnings growth is fastest when current payout ratios are high and slowest when payout ratios are low. This relationship is not subsumed by other factors, such as simple mean reversion in earnings. Our evidence thus contradicts the views of many who believe that substantial reinvestment of retained earnings will fuel faster future earnings growth.
One possible explanation for this could be how companies grow. Companies with low (or no) dividend payouts often accumulate large amounts of cash which is then used for acquisitions. It is the growth through acquisition model. Companies that payout a significant portion of their earnings in dividends don't have that luxury. They must be innovative to grow. The focus of the managers must be on developing internally rather than acquiring innovation through acquisition. The result is better for shareholders in two ways: they get higher current returns in the form of dividends and higher future returns due to higher growth rates.
The actual cost to an investor of a dividend policy varies greatly due to the multitude of tax situations investors face. For example, in a Roth IRA account, where there are no taxes on investment returns, both dividends and capital gains are equally exempt. This contrasts with someone in the highest tax bracket in the US. in 2012, their capital gains are only taxed at 15% while their dividends are taxed at 35%. Dividends can be quite expensive for an investor in this situation.
Although there is often a cost associated with a company's dividend policy, the benefits of not being completely dependent upon market price for returns has a value. Due to the plethora of tax situations, and the continued development of future tax policies, each investor should look at their own situation and determine whether the benefits of a dividend policy outweigh the tax costs. For many, they will find that the additional cost is worth added flexibility and value that dividends provide.
Disclosure: I have no positions in any stocks mentioned, and no plans to initiate any positions within the next 72 hours.