- Some contributors assert that ex dividend price adjustments are transient and are eliminated in market noise within a few hours to a few days.
- It is impossible to separate the effect of underlying market movements on the price of any specific stock from time-to-time from other things affecting the price.
- It is possible, though, to separate the effect of underlying market movements by analyzing a dividend-paying ETF that tracks a broad market index like the S&P 500.
There have been a series of articles over the past few years debating whether or not ex dividend price adjustments are lost in the noise of the market within a few hours to a few days. Some have pointed to this assertion to support a claim that dividends have a limited effect on the ongoing market price of a stock and, therefore, that the dividend is, in effect, a "bonus", because the stock price would be the same whether or not a dividend was paid. On the other hand, financial professionals tell us that ex dividend price adjustments are not transient and result in a permanent reduction in the price of a stock. Neither side can prove or disprove its position by looking at any particular stock, as it is not possible to filter out market noise and isolate price changes strictly related to the dividend... but this is not the case for a security that actually tracks the market, for example, the SPY.
The SPDR S&P 500 Trust ETF (NYSEARCA:SPY) is an exchange traded fund (ETF) that allows investors to invest in and trade the entire S&P 500 as a bundle. Through a variety of methods, it maintains its underlying constituents in virtually the exact same proportions as the index itself. Unlike the index itself, though, SPY also collects all the dividends paid by those constituents and, after deducting a small management fee (currently .09% per annum), distributes those dividends to shareholders on a quarterly basis. So, to be clear, SPY reflects all of the price movements of (subject to occasional small premiums or discounts) and all of the dividends paid by companies in the S&P 500. The only minor difference is the management fee. This creates an interesting laboratory to examine the effect of dividends on a stock price, as movements in the underlying market (defining the market as the S&P 500, in this case) can be filtered out to isolate price changes that are solely related to the dividend.
In order to analyze this, I used Google Finance to obtain daily closing prices for the S&P 500 index and SPY for 2012 and 2013. I then calculated the daily percentage price change for each, and subtracted one from the other to arrive at a daily divergence. Finally, I calculated the cumulative divergence. Here are the results:
The results show a sawtooth pattern, where the cumulative differential gradually increases over the entire quarter in anticipation of the next dividend, then returns roughly to parity every single ex dividend date. Clearly, the effect of dividends on pricing is not a matter of only hours or days. There is also no indication of a continuing cumulative enhancement to stock prices from dividends based on the return to parity each quarter. The growth in market-filtered pricing is lost, and the subsequent increase is solely related to the next dividend. As the SPY is simply a reflection of all of the constituents of the S&P 500, it is reasonable to assume that such a pattern would apply to its dividend paying constituents as well, if there was a way to filter out movements in the overall market on the price of each individual stock.