Monday, 31 December 2012

Estimating Future Dividend Income


A long hard look at a few numbers lead me to conclude that I would need $80,000 in dividends per year when I turn 55 in 2034. This income will be in addition to my DB pension. The hard part is determining how much I need to invest in dividend growth stocks to reach my objective.

Future dividend income heavily relies on the dividend growth rate. A change of 1% in the average dividend growth rate can make a huge difference over a 20 or 30 year time span.

To put words into numbers, I ran a simulation for 3 different dividend growth rates the Kraft Food Group might have over the next few decades (a while back, I ran the same type of simulation for NSC). 

Inputs into my simulation:

- 100 shares @ $45.60 ($4559.95)
- $2.00 dividend in 2013 (4.39% initial yield)
- All dividends are re-invested
- Fractional shares are permitted
- The stock price increases at the same rate as the average dividend growth rate

The numbers in 22 years:

Div Growth Rate
4%
5%
6%
# Shares
230
227
223
Dividends/year
$1,050
$1,264
$1,519

There’s only a few hundred dollars between each of the annual dividends, but look at the difference in terms of percentages.

($1264 - $1050) / $1050 = 20%

            ($1519 - $1264) / $1264 = 20%

The numbers in 30 years:

Div Growth Rate
4%
5%
6%
# Shares
317
310
304
Dividends/year
$1,975
$2552
$3290

            ($2552 - $1975) / $1975 = 29%

            ($3290 - $2552) / $2552 = 29%

I myself would be satisfied with an average dividend growth rate of 5% over the next 20 to 30 years. This would give me a total long-term return of 9.39% (4.39% + 5%) as well as a great yearly dividend.

I run such simulations for all the stocks I own. Based on my current portfolio, I estimate yearly dividends at $43,000 by the time I’m 55. I try and pick reasonable dividend growth rates, but a crystal ball I do not have.

How do you estimate future dividend income?



Friday, 21 December 2012

A Look at UNP


Over the past several weeks, I’ve posted my analysis of several railroad stocks (CNI, NSC and CSX). I’m completing the series with UNP.

Average Dividend Yield Analysis*

UNP
Year
High Price
Low Price
1Q
div.
2Q
div.
3Q
div.
4Q
div.
Annual Dividend
High Yield
Low Yield
2002
$32.58
$26.50
$0.20
$0.20
$0.20
$0.20
$0.80
3.02%
2.46%
2003
$34.75
$25.45
$0.23
$0.23
$0.23
$0.23
$0.92
3.61%
2.65%
2004
$34.78
$27.40
$0.15
$0.15
$0.15
$0.15
$0.60
2.19%
1.73%
2005
$40.63
$29.09
$0.15
$0.15
$0.15
$0.15
$0.60
2.06%
1.48%
2006
$48.75
$38.81
$0.15
$0.15
$0.15
$0.15
$0.60
1.55%
1.23%
2007
$68.78
$44.79
$0.15
$0.175
$0.175
$0.175
$0.675
1.51%
0.98%
2008
$85.80
$41.84
$0.22
$0.22
$0.22
$0.27
$0.93
2.22%
1.08%
2009
$66.73
$33.28
$0.27
$0.27
$0.27
$0.27
$1.08
3.25%
1.62%
2010
$95.78
$60.41
$0.27
$0.27
$0.33
$0.33
$1.20
1.99%
1.25%
2011
$107.89
$77.73
$0.38
$0.38
$0.475
$0.475
$1.71
2.20%
1.58%
2012


$0.60
$0.60
$0.60
$0.60
$2.40


2013


$0.69
$0.69
$0.69
$0.69
$2.76


Stock prices and dividends are per calendar year.
11y ave
2.36%
1.61%
Dividends are recorded in the quarter they were paid.
5y ave
2.23%
1.30%







3y ave
2.48%
1.49%







5y
$123.64
$211.60







3y
$111.42
$185.81






Super Cheap
$85.05


The table shows the 3 year average dividend yield to be 2.48% ((3.25% + 1.99% + 2.20%)/3). I use the numbers in the “High Yield” column for obvious reasons.

Using the current annual dividend of $2.76, one would have to pay $111.42 for a 2.48% yield.

Buying the stock at its low in 2009 would have earned you an initial yield of 3.25%. Using the current annual dividend, that corresponds to a stock price of $85.05 today (see “super cheap” price in the table above).

Graham Price

UNP’s 2009, 2010 and 2011 EPS were $3.75, $5.53 and $6.72. 3Y Ave EPS = $5.33.
UNP’s BV is $41.49.
Graham Price = SQRT (3Y Ave EPS * BV * 22.5) = $70.56

Cyclically Adjusted Price to Earnings Ratio

UNP’s 10Y Ave EPS = $3.46
CAPE = Stock Price / 10Y Ave EPS = $125.00 / $3.46 = 36
A CAPE below 20 is good.

Conclusion

UNP increased EPS from $2.53 in 2002 to $6.72 in 2011. That’s an increase of 2.7 times over 10 years. Not stellar, but not bad either. For the same period, CSX’s EPS increased 4.5 times, NSC’s EPS increased 4.6 times and CNI’s EPS increased 2.8 times.

Based on my three metrics, the stock price of UNP is too expensive. Moreover, UNP failed to increase EPS at the same rate as CSX and NSC. For the time being, I’ll concentrate on CSX and NSC.


*My method is not perfect as the stocks low price may have occurred before the annual dividend increase.