I have read a few articles recently on whether or not automatically re-investing dividends is the right thing to do. There have been well-written articles on both sides of the argument that have covered the pros and cons of reinvesting vs. pooling dividends. See http://seekingalpha.com/article/1350861-dividends-should-i-reinvest-or-not and http://seekingalpha.com/article/1290731-the-wisdom-of-not-reinvesting-dividends for a couple of those articles. For this article, I will touch on the advantages of each option and then focus on when each option is more appropriate.
There is a time and a place for each of these strategies. It is important to use the pros of each of these strategies to your advantage, while minimizing the effects of the cons as much as possible. The advantages of reinvestment include faster compounding, ability to invest small amounts of capital, dollar-cost-averaging, and peace of mind. I include peace of mind, because if you decide to go with the re-investment approach, you do not need to actively monitor the stock market. After you make your initial decision, all you have to do is sit back and let the company you chose work (though I do recommend checking in from time to time to ensure the fundamentals are still intact). This can save time and lower stress, as you do not have to worry about the day-to-day fluctuations of Mr. Market.
Because of these advantages, I highly recommend reinvesting dividends for new investors and people with smaller portfolios. When buying smaller amounts of stock, transaction costs can really eat into total performance. In my personal accounts, I try to keep total transaction costs to .05% or less of each transaction. What this means is that if it costs me $7 a trade, I will not make a trade for less than $1400. Fortunately there are now sites, such as Sharebuilder.com, which have an automatic investing option for $4 or offer free trades for new account holders. Not only was I able to initiate positions in a few companies with solid track records of paying and increasing dividends, but I was also able to put the dividends they paid back to work right away as opposing to waiting until they built up to a high enough level to meet my investing threshold.
The primary advantage of pooling dividends is higher control over allocation - you have the ability to shift money away from a companies that may be getting overvalued to companies that provide the best potential returns at that point in time. This ability is often highly underrated. If your average portfolio dividend yield is 3%, and you blindly reinvest this 3% regardless of valuation, you may be locking yourself into lower returns for this 3%. If instead you pool the 3% and deploy it all into the company that has the best opportunity to generate returns at that time, you make sure every dollar in portfolio is working to its highest potential.
One argument against pooling dividends is that if you think a company has gotten overvalued, then you should just sell it entirely and reinvest the proceeds into the best possible opportunity. I dislike this approach for a couple reasons. First, it assumes I will be able to accurately predict the direction of the stock market in the short term. Just because McCormick (NYSE:MKC) is trading at 23x earnings today doesn't mean it won't be trading at 25x earnings tomorrow. If the fundamentals are there, as there are with MKC, which has increased earnings and dividends every year since 2006, I don't want to sell the stock. Second, my goal as investor is to own highly successful companies with a proven track record and to build my stake in those companies over time. Two holdings in my retirement account are Johnson & Johnson (NYSE:JNJ) and General Mills (NYSE:GIS). I do not intend to sell them just because they have both had significant price appreciation so far this year, because I believe that regardless of their valuation, their ability to generate above average returns in any market condition will make them good investments in the long run. However if I feel there is not a significant margin of safety at the current level, I am quite comfortable pooling the dividends and waiting for a better opportunity.
Personal circumstances are by far the most important factor in your decision whether or not to reinvest dividends. You need to find the investment style you are most comfortable with, and then choose the option that fits best with your style. The good news is that you don't have to only pick one.