Dividends can be controversial.  Are they tax-efficient?  Not as good as compounding capital gains over a long period, and it will be worse when the Bush tax cuts expire.  There is no tax on buying back shares, but individuals get taxed on dividend payments.

Are they the best way to tilt value portfolios?  My guess is no.  There are many factors that drive the calculation of value, and dividends are one of them.  A multifactor model including dividends will probably beat a dividend yield only model.  It will definitely allow for a more diverse portfolio, rather than being just utilities, financials, LPs, etc.

Do dividend-yield tilted portfolios always do better than the indexes?  No, they don’t always do better.  Take the current period as an example.  These two notes from Bespoke are dated, but still instructive.  The total returns off of stocks with above average dividend yields has been poor recently.  Part of that is the current trouble in financials.  Part of it is the financial stress that is leading to cuts in dividends (again, mainly at financials).

Dividend paying stocks tend to lag when bond yields rise, also.  I remember an absolute yield manager who floundered in the early-to-mid ’90s when rates rose dramatically and bonds proved to be greater competition for the previously relatively high-yielding stocks.  They had a great time in the ’80s as yields fell but 1994 proved to be their undoing.

That said, dividends are an important part of total returns, probably one-third of all the money a diversified portfolio earns.  Also, on average, companies that pay dividends also tend to do better in the long run than companies that don’t pay dividends.  Why?

Dividends have a signaling effect.  They teach management teams a number of salutary things:

• Equity capital has a cash cost.
• Be prudent risk takers, because we want to raise the dividend if possible, and avoid lowering it, except as a last resort.
• Focus on free cash flow generation.  Be wary of projects that promise amazing returns, but will require continual investment.
• Be efficient at using capital generated from free cash flow.  The dividend forces management teams to do only the most productive capital projects.  Increasing the dividend is alternative use of capital that must be considered.
• Dividends keep management team honest in ways that buybacks don’t.  Buybacks can quietly be suspended, but in the American context, a dividend is a commitment.

Now, if you are going to use dividend yields as a part of your strategy, you need to pay attention to two things:

• Payout ratios, and
• Growth of the dividend is more important than its size

Is the company earning the dividend?  Do they have enough left over to pay for capital expenditures for maintenance and growth?   Be careful with companies that have high dividends.  My belief is that companies with middling dividends tend to offer value, but the really high dividends portend trouble.  High dividends tend to be cut during periods of financial stress, as we are seeing today.  This article on newspaper stock yields does not convince me.  I have been a bear on the industry for the last ten years.  You can’t maintain high dividends in a industry with significant competition from new entrants (Internet destroying ad revenue, classified ad revenue, and sales revenue).

REITs have decent dividend yields, but the companies with the best total returns had low dividend yields, but they grew them more rapidly.  In general, growing dividend yields where payout ratios are not deteriorating are usually good stocks to own.  Think of it this way, the dividend yield plus its growth rate will approximate the total return of the stock in the long run (for dividend paying stocks).

Two more notes before I end.  FIrst, special dividends usually not a good idea; they signal reduced prospects for the company to deploy capital productively; better to do a dutch tender and buy back shares.  When Microsoft did their special dividend four years ago, I made the following comment at RealMoney:

		David Merkel
		Note From Fed Chairman: Don’t Worry, Be Happy
		7/21/04 12:46 PM ET
Alan Greenspan completed his testimony slightly after noon today. The Q&A went quicker than usual. No real news from the affair; Dr. Greenspan tells us that inflation is not a problem, growth is not a problem, there is no systemic risk, the carry trade is reduced, a measured pace of tightening won’t hurt anyone, etc.Very optimistic; I just don’t go for the Panglossian thesis that everything can be fine after holding the fed funds target so low for so long. Bubbles develop when credit is too easy.And as an aside, I’d like to toss out a dissenting question on Microsoft (MSFT:Nasdaq). I know that the software business is not capital intensive, but if Microsoft disgorges a large amount of its cash, doesn’t it imply that they don’t see a lot of profitable opportunities to invest in it? Buying back $30 billion of Microsoft stock is a statement that they see no better opportunities (that the government will allow them to do), than to concentrate on current organic opportunities. It implies that additional organic growth opportunities are limited, no? No positions in stocks mentioned

TSCM quoted me in two articles at the time of the special dividend.  I was ambivalent about the buyback, and Microsoft stock has done nothing since then.

I also wrote this article to talk about the value of excess cash flow to management teams.  My view continues to be that excellent management teams should be given free rein to add value, while poor management teams should pay out excess cash to shareholders.

Also, there is a rule in the reinsurance business: buy back shares when the price-to-book ratio is under 1.3; issue special dividends when the price-to-book is higher, and you have slack capital.  But be careful.  Slack capital can be valuable.  I remember Montpelier’s special dividend before the 2005 hurricanes.  Ill-advised in hindsight.  The stock was a disaster, and is the only time in my career that I have flipped from long to short on a stock, post-Katrina.

Finally, I don’t look for dividends.  It is a factor in my models, but not a big one.  That said, 20 of 36 of the stocks in my portfolio pay dividends, and I receive a 2% yield or so on the portfolio as a whole.  I would rather focus on free cash flow, but dividends follow along behind free cash flow.

Bringing this back to the present, be wary.  High dividend yields, particularly on financial stocks, may be cut.  Analyze the payout ratios on stocks you own.  In general, dividends are good, but analyze the situation to determine the sustainability of the dividend.