The dividend yield on the FTSE All Share index over the last 50 years has been slowly declining. The chart below shows the dividend yield over time together with a regression line. The trend is falling even if at a fairly show rate. In the 1960s 5% dividends were the norm. By 2000 this was less than 4% and in recent years not much over 3% has been typical.
So what does this mean for the average investor? Well clearly they are getting less income on their stock market investments. I would also argue that their future returns will be less than historical averages. If we use dividend yields to value stocks then stock market prices are 50-80% more expensive than their 1960s levels. Some investors prefer to use Price Earning ratios to value companies arguing that these days companies retain more of their profits for future growth. Maybe. But I would argue that earnings are easier to manipulate than dividends which actually have to be paid out to shareholders. Companies are also more reluctant to reduce a dividend, so cyclical stocks with varying earnings will tend to have more stable dividends making them easier to value fairly.
In the US the story is even more grim. Typical yields of 5.5% in the early 1950s are less than 2% today. For me the US stock market is grossly overvalued:
Will dividend yields improve in the future? Who knows. If they did then stock prices would need to adjust to compensate for this. What could cause this? Recent stock market returns have been low and if they continue to underperform other asset classes then people might start to lose interest investing their money in stocks. Simple supply and demand would cause stock prices to revert back to prices that would yield dividends closer to historical averages. Admittedly this is a fairly extreme example but in valuing stocks I like to use this as an adjustment factor.
The Adjustment Factor
As a long term investor I use a 20 year time frame to discount this factor. I use the 5.5% dividend level which is what was typical in the 1950s-1960s in the US and UK. True it's a bit extreme but I want to build a strong margin of safety in my valuation model.
Adjustment Factor = (((3 year average dividend/price)/5.5%)^(1/20))-1
So for a stock with a price of 500p and the last 3 years of dividends of 20p, 25p and 30p we would get the following:
Adjustment Factor = ((((20+25+30/3)/500)/5.5%)^(1/20))-1
This can then be used to value the stock based:
Expected Annual Return = Dividend Yield + Growth Factor + Adjustment Factor
Why is this important? Well let's consider choosing between two shares, A and B:
Share A, Dividend 2%, Growth 6.5%, Total 8.5%
Share B, Dividend 4%, Growth 1.5%, Total 5.5%
Which is better? Share A appears to be better as the combined dividend and growth factor give a higher expected return. To buy $10 of income in share A costs $500, for $10 in share B costs $250. Using this logic share C with a dividend of 0.1% and growth of 10% would be better than both share A and share B. But this share costs 40 times as much as share B to get the same initial income. Given a high enough growth rate infinity could be a fair price to pay for a share.
By using an adjustment factor we attempt to eliminate this problem. Therefore:
Share A, Dividend 2%, Growth 6.5%, Adjustment -4.9%, Total 3.6%
Share B, Dividend 4%, Growth 1.5%, Adjustment -1.6%, Total 3.9%
Share C, Dividend 0.1%, Growth 10%, Adjustment -18.2%, Total -8.2%
As you can see none of the shares are great investments. Share C is an overpriced piece of junk with a negative expected return. B is a bit better than the growth share, share A.
This helps us to compare growth and value or income shares more fairly. Admittedly the adjustment factor favours income stocks as opposed to growth stocks. As there is more uncertainty and therefore more risk in growth stocks I don't find this to be a particularly negative problem with the method.