But, if you believe in the dividend signaling theory, could the act of declaring dividends itself act as a signal for decreased volatility in a stock? There appears to be evidence indicating as such.

A 1989 research paper found that after a company began declaring dividends, its stock price experienced lower price movements after earnings release and price variability falls after the initiation of a dividend policy. The conclusion being that investors tend to look at the dividend policy as a signal or bellweather to the future much more than earnings itself. Certainly, this would be reflected in the greater price effects of a dividend paying stock increasing or decreasing its dividend than merely meeting earnings expectations.

There is, admittedly, self-selection in the process; dividend stocks tends to attract certain types of investors who place a relatively larger premium on the cash flow component of the stock than pure momentum traders/investors.  This raises the larger question of whether it is wise to purchase a dividend paying stock purely on the dividend increase itself rather than the underlying fundamentals. However, very few of us are purely rational beings in the market.

Not surprisingly, where dividends cease to be paid, two inter-related results arise. Without a dividend policy, earnings variance increases and earnings predictability decreases.  This is explainable in that earnings and cash flow have to be predictable to declare dividends in the first place; dividends tend to be revoked once you lose that predictable cash flow (see banks in 2008 and 2009) which effects earnings expectations as well. Without any consistency in earnings, companies are more likely to meet earnings expectations and be punished by the markets.

The conclusion seems to be that investors with low tolerance for volatility and a desire for some type of predictability, but require an exposure to equities, should gravitate towards dividend paying stocks.

In 2007, the 5 largest American financial services stocks (JP Morgan, Wells Fargo, US Bancorp, Bank of America and Citigroup) represented 14 cents of every dollar paid in dividend by the S & P 500. In 2010, these same 5 stocks are on track to pay only 2 cents of every dollar in dividend.

Many believe that the banks will not be increasing their dividend soon. All of this bad news raises the question of whether it is time to reassess how many bank stocks should be in a dividend portfolio.

Let’s start with the simple question- are the good times over for banks increasing dividends? The short answer may be yes.

S & P has stated that the historical dividend growth rates is 5.6% per annum. From the period of 1995 to 2008, the smallest percentage dividend increase by Wells Fargo was 7.6% with 7 annual double digit percentage increases. Similar results can be found for each of Wells Fargo’s counterparts.  In other words, banks displayed historically anomalous dividend increase behavior in the last 15 years which can be replicated if the same economic conditions exist.

If you assume that banks will revert back to the norm with respectable 5.6% ish annual dividend increases, it stands to reason that investors, such as myself, have no real reason to be overweight in banks especially since dividend increases for some banks are really recovery of dividend cuts of the past several years.

If you assume the opposite, that banks will eventually regain their shine, the events of 2008-2009 indicate that such rewards come with proportionate down side risk; these are not your grandma’s bank stocks and regulatory freedom gave the banks the rope to succeed and eventually hang themselves as well. Given such risk/reward, should an investor with a less than steely resolve, and a strong stomach for watching paper losses, be overweight in this industry?

Regardless of whatever assumption one believes, since diversification is the only free lunch in investing, what should be the proper weighting of banks? There is no definitive answer other than what is NOT the correct answer for most investors.

A portfolio consisting of over 50% in banks stocks would be overweight. Money could invested in industries which work counter-cyclical to bank stocks (consumer staples) or may be more stable dividend payers (utilities which is regulated on pricing supporting, and capping, earnings).

The question, which I do not have an answer to, is whether one counts bank issued preference shares in determining the percentage of bank stocks in an asset allocation. In a near worst case scenario, a bank could suspend its dividend payments but continue to pay its preference shareholders. But the yield on preference shares is partially a function of the balance sheet health of a bank. Any thoughts are appreciated.

A weighting of under 10% may probably be too low for many average investors, keeping in mind that bank stocks tend to lead both Wall Street and Main Street recoveries and, once upon a time, they were actually safe stocks. If we return to a pre 1991-2008 normal, they may soon be again.

I would suggest as a model dividend portfolio something analogous to Nurse911’s dividend portfolio in terms of sector diversification and mixture of growth/mature stocks. As you can see, his dividend portfolio has a goldilocks exposure to financials- not too hot and not too cold.

Finally, there is a geographic drifting towards bank stocks one has to be aware of. The U.S. S & P Dividend Aristocrats has only a 7% weight in financials. The Canadian counterpart has approximately 39%.  In other words, purchase a dividend ETF tracking the Canadian S & P Dividend Aristocrats will naturally give you a large exposure to bank stocks.

This gap is partially explained by the poor results of the American banks versus their Canadian counterparts and partially by the fact the Canadian stock markets are basically compromised of financial stocks and commodity stocks (which suggests buying a board based Canadian equities ETF in tandem with a Canadian dividend ETF is substantially an exercise in redundancy). Thus, an overweight position in bank stocks may be a function of geography and there should be a greater focus by Canadian investors to ensure they are not drifting into an overweight in bank stocks.

As we settle into a year of the unknown, what warning signs should dividend and income trust investors look for to determine if a dividend paying stock or income trust may be going down the wrong path. The following are 3 real life examples:

Moving away from organic growth to new business lines. Kingsway Financial Services is a mid-cap company which became successful providing insurance to higher risk drivers and truckers. At its very peak, it was paying a 7.5 cent dividend per share a quarter; a respectable dividend for a mid-cap stock. However, in many respects, its success sowed the seeds of its own destruction.

Flush with confidence, it started a rapid expansion into the United States, a market which it had little experience in, and into other business lines (it bought a hurricane insurance company at one point which stretches the concept of synergies). The company, at its peak, had 9 operating units- quite a lot for a mid-sized company. Then, the bottom fell out.

The company’s downfall was partly due to declining economic conditions and partly due to bad underwriting. Underwriting mistakes, which means you underestimated the risk of loss, often happens if a new entrant under-prices itself to gain market share or does not understand the business well.

In 2008, it began to divest of assets. So much so, it gave away the shares of one of its subsidiary to a charity to avoid taxes (this transaction is now under investigation). The stock traded in the mid-teens five years ago. It now trades for less than $2 and it pays no dividend.

Kingsway is a good example of how empire building can destroy shareholder value.  Investors want dividend growth but one has to be careful how a business is pursuing this growth. If the company begins moving away from its niche, it can be seen as a warning sign. Kingsway is but a recent example. TransCanada’s dividend cut of 1999 was the end result of a business moving out of its core competence to empire building. The silver lining for TransCanada was it had a significant economic moat in its main business line to recover from this error.

The company is over-generous in its pay-out policies. This warning sign is unfolding as we speak to Riocan REIT.  Many have wondered whether Riocan was overly generous in its distribution policies during good times. In particular, its distribution policies seemed to be premised on the assumption that profits from the sale of properties would supplement regular rental income.

This led to the unusual situation of Riocan’s occupancy rate increasing in 2008 and 2009 but increasingly distributing more than it brought in; the lesson being fluctuations in earnings, especially premised on big transactions, coupled with high payouts can be warnings signs to dividend investors.

According to Riocan’s financial statements, for fiscal 2008, it paid out 102.9% of adjusted funds from operation (adjusted funds from operations, or AFFO, is a non-GAAP measure that is a “true” measure of cash flow in a real estate concern). For fiscal 2009, it paid out 127.2% of AFFO or it took in $1.00 and paid out $1.27.2.

This situation is mitigated somewhat by the fact Riocan offers a dividend reinvestment program which allows the company to pay distributions in shares rather than cash. The result is that in 2009 it took in $1.00 and paid out $1.04.5 in cash and the remainder in shares.

One wonders how much longer this situation can be sustained. Alternatively, economic prospects brighten and the increase in cash flow puts the company at a cash flow even position. In this situation, does management push the company back to an unsustainable payout betting on a long term recovery? Cautious investors would probably say no and want to pay it safe.

The lesson being watch out for a company that pays out dividends and income trust distributions generously as a means to entice new investors. It could be building a house of cards that may fall in bad times.

A change in management philosophy. Too much debt is the enemy of the dividend investor. Lenders can force dividend paying companies to slash its dividends through covenants found in loan agreements or, more subtly, force companies to slash dividends as a condition of maintaining credit facilities.

As we work our way through the recovery, a dividend investor in a high debt company may face a double whammy. A difficult re-financing market for some industries may force the company to free up cash by slashing the dividend or there may be a change in philosophy about risk tolerance levels.

Manulife Financial is perhaps one of the more (in)famous example of a change in management philosophies about risk and debt. Under a CEO that attracted a lot of press, a company which managed risk for a living began to take risks itself. In the beginning, such risk taking was rewarded by the market and earnings translated to increased dividend payments to investors. However, the risk taking lead to the company incurring  large annuity related obligations it will have to pay in the future-with the underlying assets to pay for such obligations reliant on the performance of the market (some of this exposure is unhedged).

The CEO who oversaw the creation of this situation had a much more liberal view of assuming debt than his successor and some believe that he was more interested in creating an asset management company than operating an insurance concern (see my first point above). The new CEO slashed the dividend- twice- in order to give the company financial flexibility and appears to have a very conservative view of incurring any new liabilities.

This factor bears special attention since the super star CEO who thought he was an i-banker may be giving way to more custodial type management. American banks notwithstanding, the accountants and operational people seem to be taking over from the deal making CEO’s (see Walmart, Rogers, CN).

The lesson being be careful of the company that increased dividends by taking greater than normal risks. It some point, its corporate culture may kick back in and revert back to its more conservative ways at the cost of the dividend.

________________________________________________

None of these factors alone spell trouble for a dividend paying company but they should be warning signs and signal to an alert investor to keep a close eye on the situation.

The act of conversion alone is not to blame for such a large distribution cut. Revenue and earning decreases in a down economy have only poured salt into the wound of income trust investors. As the drop dead date for the new tax regime looms, will increased income trust conversions lead to yield chasing by investors seeking alternatives?

Income trusts typically have higher yields than conventional stocks. For example, the long term adjusted funds from operations (AFFO) yield for REITS is historically at 8.1% (AFFO is considered a truer measure of a REITs cash flow). If investors, weaned on high income trust yields, decide to substitute income trusts for alternatives, will their baseline analysis be to other high yield products?

One hopes not for several reasons. As many commentators have noted, dividend growth is more attractive than dividend yield.  Dividend yield is calculated as annual dividend paid per share/price per share. Thus, high yields are a function of high dividend payments, low share price or a combination of both which means the company has little room to pay out more or the expectations of growth are limited. This may not be a large factor for someone with a short investing horizon but it is important for long term investors who can increase returns by receiving constant dividend increases.

The other issue is what are the substitutes? The natural substitute is to move to a REIT (generally exempt from the new tax regime) but REITs are not value plays (at least in Canada; on the income side, Canadian REITs are paying out 94% of AFFO meaning do not expect large distribution increases) or are facing some serious short-term issues (American REITs suffering from increased vacancies and refinancing concerns where the asset already possesses a high loan to value ratio with the value continuing to erode).

The other natural substitute, dividend paying financial stocks, also have their well publicized issues. When JPMorgan Chase announced that its loan loss provisions in the last quarter have not significantly declined, it confirmed what many suspected. The banks are a long way out of the woods yet.

This is not to suggest that pursuing a high yield substitute is, in and of itself, a wrong strategy to pursue; for example, heavily regulated utilities tend to have high yields and some margin of safety given electricity is a necessity. However, a dogmatic chasing of yields could lead to trouble. The prudent approach is to find a balance between yield and growth and to wait out a large movement of cash out of converting income trusts to other high yield products.

What then are realistic expectations of return on stocks and real estate?

STOCKS

Jeremy Siegel notes that total nominal return (return before inflation) for stocks from the period of 1802-2006 was 8.3%. What we witnessed from 1985-2006, where nominal return was 12.4%, was a historical anomaly. To use an often-quoted phrase of 2010, the “new normal” should be a reversion to historical mean.

However, one never uses the past to predict the near future. To realistically predict future expectation of growth, one uses the Gordon Equation which is (dividend yield + dividend growth rate). The S & P 500 dividend yield was 2.0% in 2009. It has been predicted that the S & P 500 dividend growth rate will be 6.1% in 2010 (according to Business Week, the historical dividend growth rate is 5.56%) . In other words, the “experts” predict expected equity return of 8.1% which falls within the historical range.

Practically speaking, and let’s assume high single digit growth does come true, subtracting inflation and transaction costs, a realistic real return for most retail investors should be in the range of 5-7% if one tracked a broad based index; aim for this sweet spot and one has a reasonable expectation of return without subjecting oneself to above average risk.

Any investment that is advertised to yield greater than this  range requires a greater risk tolerance on the investor. Thus, anyone pitching an investor a product returning 10-15% this year is not lying but is implicitly asking the investor to undertake a large amount of risk. The question is not whether the 10-15% return is realistic but whether one has the stomach to suffer loss in the event the manager reaches too far to attempt to produce such return.

Remember one of the key rules of investing: never invest any money you cannot afford to lose.

REAL ESTATE

Admittedly, since all real estate is local, the following is a broad based review and not an indication of local conditions.

Appreciation

To quote William Bernstein:

“..the best data on house price suggests that after taking inflation into account, the answer [to how much a house appreciates] is slim to none. These data focuses on historical data from three nations.  Real house prices (TMW- in other words after inflation return) in the United States did not rise at all between 1890-1990…Thus, at most you will receive a 3 per cent real (1 per cent price increase plus 2 percent net ‘dividend’) return on your home …”

The study Bernstein cites on real returns from 1890-1990 is a 2006 piece by Robert Shiller. Shiller, as many of you know, was one of the more famous economists who called the housing bubble before it burst.

Income  from real estate

The standard way to measure return on real estate is to determine the capitalization rate or cap rate. A cap rate measures expected asset level return. It is calculated by: annual net operating income/cost.  For example, a rental property nets $10,000 of rental income on a $100,000 purchase. Thus, the cap rate is 10%. There are downsides to using cap rate to determine return but it is generally viewed as a standardized measure of income production from real estate.

It is difficult to determine cap rates for smaller rental properties or non-institutional real estate investors. Collection of data would be difficult given the hundreds of thousands of real estate investors who would have to give honest feedback to someone.  However, data is much more readily available for institutional based commercial real estate. To this end, the National Council of Real Estate Investment Fiduciaries states that historical commercial real estate cap rates is approximately 7.6%.

Cap rates for most retail real estate investors may be lower than 7.6%. Most commercial leases work on what is known as a triple net basis. In plain English, the tenant has to pay for all of its proportionate costs of the leased premise including taxes, operating costs/maintenance and insurance. Most commercial leases  grant the landlord the right to adjust these costs retroactively if their estimates were off.

Smaller real estate investors, especially operating in rent control regulatory regimes, are granted no such right to full recovery (not to mention the opportunity costs of tending to the property which commercial landlords hire managers to carry out). Accordingly, small scale real estate investing may have a cap rate lower than 7.6% analogous to stock return being eroded by transaction costs.

If we assume that there is some slippage between institutional and non-institutional real estate investors, one is left with a cap rate within the range of 5-7%.

WHAT DOES THIS MEAN TO YOU?

The above suggests that “magic beans” promising returns of double digits over the medium to long term is accompanied with higher than average risk tolerance. If one cannot stomach such risk, then Jack- he of a long investing horizon and no assets to lose- may be better off buying the beans instead.

What strikes me about the data is how unrealistic investors became about expectations of return over the last 10-15 years. A real return of 4-6% in the stock market or in real estate investing is nothing to sneeze at (consider that most non-managerial salaried employees receive raises of 3-5% in good times which is really a 1-2% raise after inflation; this tends to suggest prudent investing yield greater year over year return than human capital). Perhaps there is nothing wrong with the stock market or real estate market. There may be, instead, something wrong with our expectations.

It is interesting to note that nominal return on stocks and real estate are similar; if you add in the 1% appreciation plus the 7.6% cap rate, you end up at 8.6% nominal real estate return vs. 8.3% nominal return on stocks.

What it suggests is that, in a vacuum, opportunity costs between the two assets classes is not material different. However, what Bernstein overlooks is that real estate appreciation, small as it is on a global basis, is tax-free in many jurisdictions as it pertains to the principal residence.

There may be a greater tax effect on stock sales depending on what jurisdiction you live in (some jurisdictions outside North America have no capital gains tax on stock sales).  Stock investing is a “passive” compared to real estate investing. This raises the question of whether the government tax policy on real estate is a recognition that sweat equity has been contributed which should be rewarded with better tax incentives.

To go back to my initial point, the above are average expectations of return based on historical data. There are obviously local effects and greater or lesser than average returns in short-term transactions. But again, the question is not whether one wants higher returns- we all do- but  does one have the risk tolerance to pursue such returns?

Dividend payout ratio begins to decrease. Given that dividend payout ratio is determined by dividends paid per share over earnings per shares, the ratio tends to move downward if earnings per share increase. However, the one thing to keep in mind is that a decreasing dividend payout ratio does NOT mean a company will necessarily increase its dividend.

Most companies have ideal dividend payout ratios. In most cases, a company will not declare a dividend increase unless the ratios come back into their optimal range. For example, most financial institutions have an ideal dividend payout ratio of 35%-50%. TD Bank Financial Group’s dividend payout ratio was 70.3% as of the 12 months ending October 31, 2009 as opposed to 49% in the prior 12 month period.

In other words, it is unrealistic for dividend shareholders to simply ask for a dividend increase as soon as the company begins to return to profitability especially in the battered financial services industry.

Earnings must be sustainable. Given that dividends are “sticky”, companies are loathe to increase them unless they know that the medium to long term financial prospects of the company are bright. Many companies were able to turn profits by cutting costs; over time, an over-reliance on this strategy is not a recipe to build a good business. RBC has already indicated that trading profits from 2009 are simply not sustainable over the long term; in other words, it was a blip caused by how badly the market had crashed.

The end result is that attention must be paid to the quantity of the earnings going up as well as the quality of the earnings in order to support continual increases in dividends over time. Typically, this is created by organic growth over a wide variety of product lines (hence, most tech companies are not dividend payers; they are typically one-trick ponies in terms of product lines). A good example of organic earnings growth company would be Johnson and Johnson (with a very healthy dividend payout ratio of 41%).

Cash on hand. On a simplistic basis, a board of directors must decide whether cash either goes back into the business or back to the shareholders. If the company has relatively little cash on hand, logic dictates that the money should go back to the business. If the company has a lot of cash on hand and earnings begin to go back up, it will have greater opportunity to increase dividends.

However, what is important to note is that increasing cash on hand must be balanced by what the cash should be earmarked for. If there are more pressing needs than returning monies to shareholders, looking simply at cash on hand is not a good metric for anticipating dividend increases.

For example, let’s take a look at RioCan REIT. It has a cash on hand of approximately $215 million as of September 30, 2009; a not insubstantial sum given real estate use up a lot of cash. However, in its financials for the period ending September 30, 2009, it reports it is paying 34.5 cents to each unit-holder but it is only making 27 cents a unit from adjusted funds from operations (a non GAAP measure for real estate companies). In other words, it is paying out more than it is making.

In addition, RioCan has been issuing more units for acquisitions and general operating funds; the latest being a $10 million raise in November. Thus, while there appears to be cash on hand, it should be committed towards obligations, particularly its shortfall on monies paid out to that taken in, other than increasing its distribution. In this light, a distribution hike should be viewed skeptically.

_________________________________________

The above factors are some items to consider. Put them together and an ideal candidate to increase dividends tends to have the following characteristics:

1. The Company remains in its ideal dividend payout ratio zone (most companies will tell you want that zone is in its financials);
2. The Company is growing organically and not relying on some blip to increase earnings that cannot be repeated over time; and
3. There are no pressing needs for its cash on hand that would take priority over returning money to shareholders.

What is not up for debate is that Manulife, assuming a full uptake of its offering, now has what it seeks. Its Minimum Continuing Capital and Surplus Requirement (“MCCSR”), an insurer’s equivalent of a bank’s capital ratios, is at an estimated 256 which is well above its peers at 215 (anything over 150 is acceptable). In plain English, Manulife has more money socked away than the regulators require in case its underwriting was poor and it has to pay out too many policies. Manulife merely is following the banks who have also raised their capital ratios to higher than required levels to protect against another credit crisis.

While no doubt giving assurances to the public at large, are fortress levels of capital (which I will use as short-hand for high capital ratios and/or MCCSR) good for dividend investors with large portfolios in financial services companies?

The first issue is how a financial services company arrives at fortress levels of capital. If, devoid of any ideas, a company simply hordes earnings created by organic growth, this is not a dividend, or shareholder, friendly management strategy. But it can be addressed by a shareholder revolt to replace management with one that will return money to shareholders through dividend increases or acquisitions.

However, an average dividend investor is harmed if a company builds fortress levels of capital through multiple share issuances. The foremost issue is shareholders are being diluted: for example, Manulife has diluted its shareholder by more than 10% in less than a year. The more troublesome issue is that the proceeds of a share issuance are not being used to grow the company, thereby increasing the chances of a corresponding dividend increase, but to build up capital levels.

There is a strong argument that building up capital levels should be lauded by dividend investors.  In an unfriendly business climate, it may be prudent to simply retain what you have; increasing dividends on blind faith in the future, murky as it is these days, is a danger sign if one wants to invest in dividend growers. Certainly, recent history shows financial services companies should be more cautious in a highly-integrated and leveraged capital market environment.

I would not suggest that a company simply increase dividends dogmatically  without watching its financial risk management. In fact, it is hard to argue that creating a fortress levels of capital is bad for dividend investors if one’s choice is either to protect the dividend or to decrease it (although in Manulife’s case, it slashed the dividend and diluated the shareholders twice; they better be issuing damn good holiday cards this year).

But the larger issue may not be the downside risk protection in creating fortress levels of capital but what happens when it is no longer needed. In 1998, the Federal Reserve Bank of New York published a paper examining the trend of bank holding companies decreasin their capital levels after building them up during the late 80′s and early 90′s.

In the abstract, this trend could be seen as alarming but, at the time at least, the author found some comfort in the fact the capital ratios were decreasing because of increasing shareholder payout. Sounds good but here’s the problem. While dividends increased 5%, share repurchases (aka share buybacks) increased 70%. In fact, while dividends increased modestly, share repurchases are often the dominant form of shareholder payout in the largest 25 bank holding companies.

Share repurchases are problematic for several reasons. While the offer is typically higher than market price, it can be seen as a means to, over time, reduce the aggregate dividend paid through reducing share capital. Share repurchase programs can also be started and ended relatively quickly. In other words, it does not signal as well to the market that the company believes its future growth prospects are healthy as compared to a large dividend increase; the stickness of dividends tend to mean that once a dividend is increased, the company is committed to paying it for long periods of time.  A share repurchase is a one-trick pony which appeals more to Wall Street than buy and hold dividend investors.

Finally, as many others have often pointed out, share repurchases show the company to be poor investors- it is buying  high and selling, via issuing stocks, low. It is, in the relative scheme of things, an unimaginative way to step down from fortress levels of capital.

To use the dating analogy, it is the morning after that may scare the dividend investor more than the night before. The arguments for building fortress levels of capital in this climate for is reasonable and pursasive to most rational dividend investors. However, no one knows how dividend paying companies with fortress levels of capital will act once the fortress is disassembled to a mere trench levels of capital. Recent history  indicates that it will not be as friendly as a buy and hold dividend investor wants it to be. Unfortunately, this is a story that will not be unfolding for some time.

As to whether Manulife has become a value play, I am of the agreement that Manulife needs to be watched carefully. A new CEO who cannot seem to stay on message (if you are going to dilute the shareholders do it once), has not seemed to master investor relations and with large legacy issues is juxtaposed with market share in the critical Asian market and a strong brand domestically. As usual, conduct your due diligence and proceed accordingly.

Such as?

1. Comparing dividend yield vs. 10 year treasury yields revels market appetite for equity risk. Since 1958, the S & P  dividend yields have only exceeded 10 year treasury yields twice: once in 1958 and once in 2008. As I noted before, it seems strange that money would rush into a riskier investment (stocks) that produced less yield than a guaranteed instrument unless you believed capital appreciation would more than make up the difference of a low dividend yield vs. bond yield. You end up in this particular situation because the market is confident of capital appreciation going forward and that the risk premium in investing in equities is manageable. Conversely, decreased appetite for risk should result in a decrease in the spread between treasury yields and dividend yields (as witnessed in 2008 when the market had no risk tolerance).
2. Large gaps between treasury yields vs. dividend yields can be read as a danger sign. Large gaps between 10 year treasury yields and S & P dividend yields occurred in the early 1980′s, explained by the rise in interest rates. They occur again during the tech boom, caused by the rapid rise in stock prices. In the latter instance, as history attests, this large gap reflected an unsustainable run up in equity prices not justified by any fundamentals; the lesson being the larger the gap, the larger the danger of a correction. In the former instance, the guaranteed yield on government issued debt would impede capital to public markets.
3. In zero to low interest environments, dividend yields exceeding government bond yields tend to produce rallies… for a while. Japan has had a de facto zero interest rate environment for the better part of two decades. In 1998, 2003 and 2005, dividend yields moved higher than government bond yields triggering a market rally each times; in essence, the market said “things can’t be that bad that we are investing in essentially 0% government paper.” However, the 4th time this happened in 2007, the rally did not occur. Perhaps the issue is that you can only stimulate the economy for so long with low to zero percent interest rates before one realizes that there are real structural issues with the economy and the market rallies, short as they are, mask a more fundamental issue (a scary possibility for the American economy).
4. Dividend yield and growth is arguably the best indicator of future market returns. The esteemed William Bernstein recently wrote about  Professor Myron J. Gordon who predicts long term stock market growth through a simple formula. Over long period of time, average equity growth = dividend yield + historical dividend growth. Historic dividend growth is 4.3%. The rationale behind the formula is elegantly simple. Smart companies generally do not increase dividends unless they are profitable. Profitability increases stock price. Thus, working backwards, stock price growth is related to dividend growth. If many companies continue to hold the line on dividend increases or increase them below the historical rate, the market could be in for a rough ride.
5. Dividend yields can be a demographic indicator. There appears to be an inverse correlation between S & P dividend yields and mutual fund in-flows. As the baby-boomers begun to invest for retirement, we witnessed a focus on capital appreciation over cash in hand and a decline in dividend yields. But, as noted above, in an aging population like Japan’s, you have a flight back to safety with a reversion back to the mean despite multiple rallies. It would appear that a long march upwards for dividend yields may not be a positive development for the market as a whole.

What does this all mean? Dividend yields can be predictive of market movements ahead. Since they are proxies for equity risk, albeit an imperfect one, a snap shot of yields can give you insight into the market’s risk tolerance.

The dividend yield of the S & P 500 stood at 2.28% as of September 30 of this year. Is this too high or low? Only time will tell.

First, let’s state the obvious. In order to be a long-term dividend payer, a company has to be able to sustain cash flow necessary to pay dividends quarter over quarter. The implication being that dividend paying stocks do not necessarily have to be profitable all the time but rather they cannot be losing money over the long-term lest they jeopardize the dividend payment and investor confidence.

However, we appear to be investing in a world characterized by increasing loss reporting. A University of Chicago study on the relationship between dividends and earnings found that the fraction of losses reported by non-dividend payers in the U.S. grew from 28% in 1974-1979 to 52% by 2000-2005. Dividend payers followed the same general trend of increasing fraction of losses reported but in a far lower percentage than non-dividend payers with losses compromising of 3.5% of dividend payers in 1974-1979 to 11% in 2000-2005 (but this percentage of losses has been more or less consistent since 1985-1989). In other words, dividend paying stocks are less likely to report losses.

The losses reported by dividend payers also tend to more palatable than from non-dividend paying stocks. The authors found that over half of losses reported by dividend payers were due to one-time items versus approximately 25% of all losses derived from special items by non-dividend payers.  The authors conclude from this finding that losses tend to be better (for lack of a better term) than non-dividend payers since they are not attributed generally to operational inefficiencies.

The issue with this conclusion is that there is nothing barring companies from repeatedly reporting “one time” items quarter after quarter to hide operational inefficiencies. However, the large differentiation between loss reporting between dividend and non-dividend payers does tend to indicate that if this is occurring it is not in sufficient scale.

The flip side of the analysis is the author’s also found that dividend paying stocks have greater “earnings persistence” than non dividend earnings; the implication being that a dividend paying company showing a profit one quarter are more likely to show profits subsequently.  Yet, caution should be exercised for companies who rely on share buybacks. Their earnings persistence tends to be less than its counterparts who rely on steady dividend payments and are more likely to report losses (although still in fewer instances than non-dividend paying firms).

The study is interesting but should be read with two caveats. The sample size (1974-2006 studying NYSE, AMEX and NASDAQ) is relatively small and it excludes utilities and financial firms (which, given both are traditionally dividend payers, may distort the data somewhat).

If you are interested in finding the best dividend stocks, I would reiterate what other bloggers have written. The most reliable dividend data is found in the company’s financials themselves and understand what makes a good dividend stock. Start there and move outwards towards an industry comparison. While the study is interesting, please remember that Enron also paid dividends so don’t judge a book purely by its dividend paying cover.

The implication being that an overlap exists between broad based equity indexes and dividend paying stocks. But, just how large is the overlap?

The U.S. is considered one of the two countries (Japan being the other) where dividend concentration is less pronounced. However, if one purchased the Vanguard Large Cap ETF (VV) and the Vanguard High Dividend Yield ETF (VYM), of the top 50 holdings of each, there are 24 stock in common. This includes a staggering 19 of the top 20 holdings in VYM by weight.

Similarly, if one purchased iShare’s flagship Cdn. Large Cap 60 Index Fund (XIU) and iShare Cdn. Dividend Index Fund (XDV), one would find 15 of the 30 stocks compromising XDV are also found in XIU. This figure has more context if one considers that the overlap constitutes over 60% of XDV by weight.

The result is that one is not diversifying by investing in a large cap ETF and a dividend based ETF. Instead, an investor has merely created a redundancy in their portfolio and failed to mitigate against downside risk.

The issue becomes magnified if that same investor commits mutual fund-itis and begins purchasing niche ETFs which overlap the broader equity based ETFs. The ETF pairings to watch for in particular would be purchasing a broad based equity ETF and then a financial services industry ETF and/or preference share ETFs. Given the latter are typically issued by the same companies that pay dividends, if they can’t make a preference share dividends, which typically ranks in priority to common share dividends, they are not paying any class of shareholders a dividend.

The lesson being, and as a preview to a post next week, choice and selection are often not the best thing for the investor.