I have written before (here, here, and here) about the dividend yield model and its uncanny reliability since 2009. Since the Fall of 2009, the S&P 500 has traded a levels producing dividend yields of between 1.8% and 2.2% with a mid-point at 2.0%. These dividend yield levels (based on 12 month trailing dividends) have been useful because the 1.8% level has been a "sell" signal (in early summer 2011) and the 2.2% level has been a "buy" signal (in late Fall 2012).
I generally consider the mid-point (2.0%) to be fair value and based on extrapolations from March dividend yield, current dividend yield should be $36.70 producing a fair value of $1835. I consider the 2.2% level to be a very strong support which has not been breached in four and a half years. Based on current dividends, that support level is $1668. We should be reasonably confident that, barring extraordinary economic developments, the current correction will not take us below that level.
We have been in a period of rapidly increasing dividends and all of these levels should increase (at the same rate as the rate of dividend increase) on a month to month basis. By July, the support level should be over $1700 and it will continue to rise at roughly 1% per month (the same rate at which the dividends have been increasing).
Why The Model Works - The model works because, as yield increases, dividend oriented investors become yield hungry and find stocks attractive in comparison with fixed income investments. Given the fact that dividend increases have been reliable since 2010, investors stand to reap not only current yield but increasing yield on original cost as dividends are increased. When the yield on the entire index reaches 2.2%, there are many stocks with yields over 3%. A three percent current yield - combined with the promise of annual increases - looks very good compared to what is currently available in fixed income. Dividends also continue to enjoy favorable tax treatment. Many dividend paying companies also engage in share repurchases and the silver lining of a price decline is that more shares can be repurchased for the same amount of money. This tends to create a tailwind for earnings per share. Investors should note that the model works only with the S&P 500 (an index dominated by large dividend paying stocks) and does not really provide useful information about "momentum" stocks or indices which include a higher percentage of stocks which do not pay dividends.
The Trend in Dividend Increases - We have had a robust trend of dividend increases with double digit percentage increases for three years. Recently, the increase percentage has been showing signs of decline retreating to the 11-12% level (the basis for my 1 percent per month assumption). I think we may get two or three more years of double digit percentage increases but, in the long run, the increase rate will almost inevitably decline. In the next several years, a number of large banks will likely increase dividends as capital has been rebuilt. In addition, Google (NASDAQ:GOOG) is weighted heavily in the index and it pays no dividend. If it follows the example of Apple (NASDAQ:AAPL) and starts sending out checks, it will have a big effect on the aggregate number. But, in the long run, dividends can't increase much faster than nominal earnings per share and nominal earnings per share can't increase much faster than nominal GDP. With nominal GDP slugging along and increasing at between 3 and 5% per year, it is unlikely that nominal dividends will continue to increase at double digit percentage rates. I see at least four possible scenarios playing out; here is how they will affect the model's operation.
1. Steady As She Goes - We continue to slug along with low inflation, low growth and low interest rates. Corporations eke out per share earnings increases through cost cutting, share repurchases and acquisitions leveraged with low interest debt. The model continues to operate and the market follows a trendline of growing at the rate of dividend increases. We get another year or two of double digit increases and then drift down to a 6 or 7 percent level. This may lead investors to demand a higher initial yield but, with continuing low interest rates, 2% growing at 6 or 7 percent per year will not look bad.
2. Sudden High Inflation - In the very near future, inflationary trends emerge and force the Fed to increase interest rates. The only thing that I can imagine would do this would be a spike in oil prices due to a blow up in the Middle East. Given our increased domestic production, a good deal of the money would stay here and get put into increased drilling. Canada would prosper and the Canadian dollar would soar creating a nice tailwind for the US economy and corporate earnings. However, the combined effect of fewer dollars in consumers' pockets, a dollar which is appreciating against most other currencies, and uncertainty would put pressure on the economy. We might have the nasty combination of higher interest rates and lower dividends or a lower increase in dividends. Investors might demand higher dividend yields at the very time that dividends themselves were declining. No big surprise here; $200 a barrel oil would tank the market. The model helps explain why and to what extent.
3. Growth Developing Into Inflation - In this scenario growth accelerates over the next few years and, after a time, inflation rears its ugly head leading to interest rate increases. In this scenario, it is very plausible that the market would set a mid-point yield level of 2.5% (or higher) as interest rates increase. This, of course, would translate into lower stock prices. However, in this scenario, rates would increase only after a period of economic growth. Thus, dividends might be considerably higher by the time rates go up. For example, after 4 years, dividends on the index might be at the $55 level so that a 2.5% yield would produce a price of $2200. Not exactly a great return if it takes 4 years to get there, but it is very likely that the higher dividends might take us to $2400 or $2500 before we pulled back due to higher interest rates.
4. Recession - An actual recession could give the market a double whammy. Dividend levels could be reduced or at least threatened so that investor confidence is undermined. This might lead investors to demand higher yields (as they did in early 2009). Higher yields and lower dividends would lead to much lower prices. Again, it should be no big surprise that a recession would be bad for the market. In our current circumstances, I think it could be a worse than "normal" development because of the Fed's inability to lower rates from current levels, the danger that the dollar would shoot up due to international uncertainties, and the extreme skittishness of investors when confronted with the possibility of dividend cuts. One of my favorite stocks - Cedar Fair (NYSE:FUN) - was taken out and shot when it cut its dividend a few years ago. We could see some very ugly market reactions if dividend cuts became common.
Conclusion - The market can pull back some more from here but I will be very, very surprised if it goes below 1688 and that threshold will increase each month by about 1%. It doesn't take a math whiz to figure out that by early next year, the support level will be about where we are now with fair value getting into the 2000 range. I don't see any economic data pointing to any of the above scenarios except for number 1. If you are nervous about the second scenario, you should consider being overweight in oil stocks. I lightened up a little earlier this year and I am now making up a list of stocks I will try to grab on declines.
Disclosure: I am long AAPL, GOOG, FUN. I wrote this article myself, and it expresses my own opinions. I am not receiving compensation for it (other than from Seeking Alpha). I have no business relationship with any company whose stock is mentioned in this article.