Why Dividends Are Irrelevant in Explaining Investment Returns

There's a common misconception that investing in dividend stocks guarantees higher investment returns. This belief is often rooted in the regular income dividends provide, their historical link with financially robust companies, and the psychological appeal of receiving cash payments. The idea is further reinforced by financial frameworks such as the dividend discount model, which values a stock based on its future dividend payments.

Building on this prevalent but misguided belief about dividends, this post examines their actual impact on investment returns through the lens of two key academic contributions: the Miller-Modigliani Dividend Irrelevance Theory and the Fama-French Multifactor Models. These seminal studies, particularly the work of Fama-French, present compelling evidence that dividends are not a determinant in explaining investment returns.

Miller and Modigliani's Dividend Irrelevance Theory

To understand Miller and Modigliani's Dividend Irrelevance Theory, you first have to understand how capital is returned to shareholders in the stock market.

Putting aside the less common methods for returning capital to shareholders, such as special dividends, return of capital, and liquidation, the two primary channels in which capital is returned to shareholders are dividends and share buybacks. These are described further below:

• Dividends: Dividends are optional cash payments that are distributed to shareholders out of a company's profits. Simply by owning a share of stock in a dividend-paying company before the "ex-dividend date," you can receive income without needing to sell your shares. For example, if you own 100 shares in a company that announces a dividend of $1.00 per share, you would receive $100 pre-tax (100 shares × $1.00) directly into your brokerage account.
• Share Buybacks: Share buybacks reduce the number of outstanding shares on the market. Although they do not result in cash appearing in your brokerage account, they do effectively increase the share of company profits that accrue to each remaining shareholder.

The concept of dividend irrelevance was first put forth in a seminal 1961 paper by Merton Miller and Franco Modigliani titled "Dividend Policy, Growth, and the Valuation of Shares." In their work, Miller and Modigliani argue that, absent factors such as taxes and trading costs, dividends should not impact an investor's decision-making. They famously stated:

"Before frictions like trading costs and taxes, investors should be indifferent between $1 in the form of a dividend, which causes the stock price to drop by $1, and $1 received by selling some shares."

— Miller and Modigliani

Miller and Modigliani contend that corporations have various avenues for employing their capital. They can invest in new projects, fund research and development, or engage in mergers and acquisitions. If none of these routes offer a reasonable return on investment, the capital is then returned to shareholders, primarily in the form of dividends or share buybacks.

According to PWL Capital, from 1973 to 1976, the average return on capital to U.S. shareholders, through both dividends and share buybacks, was about 4.4% of the total market capitalization.

Further, research from S&P Global has shown that since 1997, the aggregate amount of buybacks exceed the cash dividends paid by U.S. firms, as shown in the exhibit below:

Specifically, S&P Global claims the following:

"Since 1997, the total amount of buybacks has exceeded the cash dividends paid by U.S. firms (see Exhibit 1). The proportion of dividend-paying companies decreased to 43% in 2018 from 78% in 1980, while the proportion of companies with share buybacks increased to 53% from 28% during the same time period."

— S&P Global

Over the years, dividends were initially the dominant form of returning capital to shareholders, but share buybacks have increasingly become the preferred method. This parity between dividends and share buybacks further strengthens Miller and Modigliani's theory that dividends are irrelevant, as a company that mainly uses dividends to return capital to shareholders is no better or worse than a company that mainly utilizes share buybacks.

Hypothetical Scenario: Dividend vs. No Dividend

Let's consider two hypothetical companies, both with an initial book value of $10 per share. Company 1 distributes a $1 dividend, while Company 2 chooses not to pay a dividend. Both companies are otherwise identical in terms of size, expected future profits, and rate of reinvestment. According to valuation theory, their market prices should reflect their respective book values and the discounted value of their future profits.

If you own 5,000 shares of Company 1, you would receive $5,000 in dividends (5,000 × $1). After the dividend, your shares would have a value of $45,000 (5,000 × $9). In total, your portfolio would consist of $45,000 in Company 1 stock and $5,000 in cash, summing up to $50,000.

On the other hand, if you own 5,000 shares of Company 2, which did not pay a dividend, the value of your stock would stand at $50,000 (5,000 × $10). You have the option to "create" your own dividend by selling shares, but you are not compelled to take a distribution, as you would be with Company 1.

Whether the market is bullish or bearish, the value of a company diminishes by the exact amount of the dividend paid. In a downturn, receiving a dividend is the same as selling shares. Thus, Miller and Modigliani's thesis holds strong: dividends are essentially irrelevant in a frictionless world, and the emphasis should instead be on a company's overall profitability and investment strategies.

Limitations in Miller and Modigliani's Dividend Irrelevance Theory

Although the Dividend Irrelevance Theory by Miller and Modigliani has been influential in shaping the academic and practical understanding of how dividends affect stock valuation, its assumptions often fall short of capturing real-world financial complexities.

Factors such as different tax treatments for dividends and capital gains, transaction costs, and even liquidity constraints can significantly impact an investor's returns. Additionally, dividends can serve as signaling mechanisms due to informational asymmetries between managers and investors and can help mitigate agency conflicts (the conflicts arising between shareholders and company managers over the control and use of corporate resources).

While the limitations of Miller and Modigliani's Dividend Irrelevance Theory raise questions about its full applicability in real-world markets, they do not negate its core principle that dividends do not impact a firm's value in an ideal, frictionless market. These limitations serve as qualifiers to the theory's range of applicability and can often be minimized in a well-managed diversified portfolio. As such, the theory remains a foundational framework for understanding the role of dividends in investment returns, rather than an absolute rule.

Fama and French on Dividend Irrelevance

The following section will discuss Fama and French's multifactor models and further research on dividends in a diversified portfolio, to further highlight that dividends may not be as pivotal in determining stock returns as many investors think.

The central argument here lies in understanding that the five key factors outlined below account for the majority of stock returns in a diversified portfolio. Notably, dividends are not among these factors.

Five Factors That Explain Market Returns

To comprehend why dividends are irrelevant to investment returns, it's important to understand the key factors that drive returns in a diversified stock portfolio.

Five key factors collectively account for roughly 95% of returns in a diversified stock portfolio. These factors have their roots in the Capital Asset Pricing Model (CAPM) introduced in the 1960s, which initially explained about 70% of portfolio returns based solely on market risk (beta).

The model was later refined by Fama and French who introduced their Three Factor Model in 1992. By including additional factors of value and size alongside market risk, this new model could explain approximately 90% of returns.

Later, in light of further evidence showing that even the Three Factor Model overlooked variations in average returns related to profitability and investment, Fama and French introduced an updated Five Factor Model in 2015. This refined model accounts for around 95% of returns.

The remaining variance in returns across all three models is commonly attributed to alpha, representing unexplained factors such as momentum.

These five factors are described further below:

• Market Risk (Beta): Originating from the CAPM, this factor measures a stock's volatility compared to the overall market and is fundamental to understanding market risk and expected returns.
• Value (HML): Introduced in Fama and French's 1992 model, this factor suggests that stocks with high book-to-market ratios generally outperform growth stocks and are likely undervalued.
• Size (SMB): Also added in the 1992 model, the size factor suggests that small-cap stocks tend to outperform large-cap stocks, although they come with greater risk.
• Profitability (RMW): Introduced in the 2015 Five Factor Model, RMW (Robust Minus Weak) indicates that companies with higher operating profitability typically outperform those with lower profitability.
• Investment (CMA): Also part of the 2015 model, CMA (Conservative Minus Aggressive) shows that firms with lower asset growth generally yield better returns than those investing more aggressively.

These five factors provide investors with a robust framework to comprehend the key drivers of stock returns. As evident, dividends are not included among these factors and thus do not contribute to explaining the majority of investment returns in a diversified stock portfolio.

Related: How to Calculate and Interpret the Fama and French and Carhart Multifactor Models.

Fama and French Research on Dividend Irrelevance

While it might seem surprising that dividends do not play a role in explaining investment returns, especially given prior historical data and statistics suggesting otherwise, the reality is that dividend growth stocks often exhibit higher exposure to value, profitability, and investment factors. These are the true drivers of their performance, not their ability to consistently pay and grow dividends over extended periods of time.

This point was validated by Fama and French in their seminal 1993 paper, "Common Risk Factors in the Returns on Stocks and Bonds." In this research, they showed that portfolios sorted by dividend yield did not have three-factor alphas that were statistically different from zero. This indicates that dividends do not offer any unique information about expected returns once market risk, company size, and relative price are taken into account.

To put it simply, the returns of a portfolio based on dividends can be fully explained by its sensitivity to market beta, company size, relative price, profitability, and investment. Dividends, therefore, do not add any additional predictive power for expected returns that aren't already captured in the five-factor asset pricing model.

In a subsequent 1998 study titled "Stock Returns, Dividend Yields, and Taxes," Fama and French examined dividend yield as a potential value factor. When testing dividend yield alongside price-to-book (P/B) and price-to-earnings (P/E) ratios, they found that dividend yield delivered the smallest value premium.

This finding has direct relevance for dividend investors: if dividend-paying stocks are primarily value stocks, then the higher returns often associated with dividend payers may simply be a manifestation of the value factor. In such cases, directly targeting value stocks would offer a more efficient method for capturing the value premium.

In closing, while portfolios built on metrics such as dividend yield or dividend growth may include stocks with strong profitability, investment potential, and value characteristics, dividends themselves prove to be an inefficient vehicle for capturing these attributes.

Real-World Performance of Dividend Investing Strategies

Beyond academic frameworks, it's useful to examine how dividend-focused investment strategies actually fare in the real-world market. One reliable data source for this purpose is the SPIVA Canada Year-End 2022 Scorecard, which provides interesting insights into the performance of dividend-paying stocks.

The Scorecard reveals that the S&P/TSX Canadian Dividend Aristocrats Index, comprised of companies with a history of increasing their dividends, generated an average annual return of 7.5% over a 10-year period ending in 2022. Interestingly, not a single one of the 67 Canadian dividend-focused mutual funds that were available at the start of this 10-year span managed to outperform this index; they averaged an annual return of just 6.1%.

Complicating this picture further is the fact that both Canadian and U.S. Dividend Aristocrat Indices have periodically outperformed broader market indices. It's worth noting, however, that this outperformance is largely attributed to these indices being more exposed to value, profitability, and conservative investment factors, which have been shown to be significant in explaining differences in portfolio returns.

Yet, not all dividend-paying companies are necessarily exposed to these key factors, and conversely, many companies that are well-aligned with these factors do not pay dividends.

Thus, a singular focus on companies with growing dividends could inadvertently expose your portfolio to these key performance-driving factors. But such exposure is far from consistent and may inadvertently lead to a less diversified portfolio.

By narrowing your investment options to only those companies that pay dividends, you run the risk of missing out on other potentially profitable investments that do not fall within the dividend-paying category. Moreover, given that dividends have been shown to be unrelated to broader market returns, the strategy of focusing solely on dividend-paying stocks appears to be an unreliable path for achieving long-term investment success.

Therefore, while it may seem tempting to associate dividends with market-beating returns, the evidence suggests that this strategy is not a reliable one. Furthermore, this approach could compromise both the diversification and long-term stability of your investment portfolio.

The Bottom Line

Based on a thorough examination of theories, research studies, and real-world market data, it's clear that dividends shouldn't be the main focus for investors aiming for long-term value. From the Dividend Irrelevance Theory by Miller and Modigliani to the multifactor models by Fama and French, the evidence shows that dividends don't play a key role in determining overall investment returns.

Instead, the primary drivers of returns in a well-diversified stock portfolio are market risk, company size, value, profitability, and investment factors. This point is further emphasized by real-world data like the SPIVA Canada Year-End 2022 Scorecard, which shows that focusing solely on dividends is not a reliable strategy for consistently beating the market.

Putting too much emphasis on dividend-paying or dividend-growing stocks can unintentionally compromise portfolio diversification and may miss out on the true factors that drive investment value. The idea that dividends are essential for achieving superior investment performance is not supported by current academic and market evidence.

Investors would be well-advised to consider this body of evidence as a guiding framework for more strategic investment decisions, instead of relying on dividends as a metric for long-term investment success.