Dividend Growth Machine
08th Oct. 2017
This article is provided for informational purposes only. No consideration of company fundamentals has
been made on any stocks mentioned. The “portfolio” created does not take any personal circumstances
into consideration. recommends that investors do further research before considering any of the
stocks mentioned in this article. The best way to use the information provided in this article is as a stock
screener that provides a platform for further research.
Dividend Growth Investing is a strategy that focuses on companies which regularly raise dividends. In this
article we try to construct a dividend growth screen of New Zealand stocks purely from using historical
quantitative data.
Goal:
The goal of this exercise is to design a diversified screen using
quantitative data of New Zealand stocks that are growing their
dividend. We believe that dividend growth is important because:
• It is a tangible sign of financial strength for a company if
they can send more money each year back to
shareholders.
• Dividend growth is often correlated with top line revenue growth.
• As dividends increase, yield on cost for investors also increases.
• Dividend increases are often accompanied by share price appreciation.
Mathematically this is simple, a company worth $10 that pays a $1 dividend
trades on a dividend yield of 10%. If this company increases its dividend to $2
per share and the market still values the company on a 10% yield then the
share price has appreciated to $20.
• Investors often overlook dividend growth in favour of short term yield. This is
can be dangerous as stocks with high yields often have high yields for a reason.
For simplicity we will create a 20-stock screen. The screen will have clear buying and selling criteria.
Assumptions:
This is important for any stock research but particularly for quantitative. We
recommend that before doing any stock research investors examine the assumptions
or biases that they have. Assumptions will be different from investor to investor and
there is no right or wrong way to go about it. For this screen we have the following
biases:
• All information should be able to be created using tools that are available for
free to any investor who has an internet connection.
• Companies that pay dividends should be profitable.
• Companies that are growing dividends should not be making dividend cuts. A dividend growth stock by any reasonable person’s definition does not have
dividend cuts.
• Likewise, dividend growth companies should be growing their dividends.
• Dividend growth is momentum based, with the
companies that have grown their dividends fastest
historically being more likely to continue.
• Dividend growth should be driven by underlying
revenue growth.
• A company with little track record of paying a
dividend is more likely to cut it in the future.
• Special dividends are irregular and are therefore ignored. In this context we
view special dividends as bonuses.
• We do not make investments for tax reasons and we therefore ignore
imputation credits. Any tax benefits are considered a bonus.
• Dividend growth companies can be found across different industries and
company size. We do not exclude or include based on company size or
industry.
• Dividend yield, price to earnings and other earnings valuations are of
secondary importance to dividend growth.
Step 1, Creating an Initial List:
Head to the Google finance stock screener (http://finance.google.com). This is a basic but free to use stock screener that you can use to filter selected stock by quantitative data.
According to Google finance there are 336 stocks that are listed on the New Zealand
Stock Exchange. The first thing that we must do is filter out all those that do not pay a
dividend.
This immediately filters out 238 instruments and we are left with 98.
One of our assumptions is that dividend growth companies should be profitable. The
basis behind this bias is that it is difficult for a company to grow or even sustain a
dividend if they remain unprofitable.
This assumption proves relatively accurate as only 6 further companies are filtered out.
We are left with 92 companies. This gives us our initial list.
Step 2, Cleaning up the List:
While the Google Finance screener is very good, it is still a free screener and it does
have data impurities. Our job now is to go through one by one and tidy up this data.
Instruments that we are looking to delete are:
• Delisted stocks such as Nuplex (NPX) and Hellaby (HBY)
• Exchange Traded Funds
• Restricted stock such as Fonterra.
After this clean up our list is reduced further to 85 NZX companies.
Step 3, Adding our Assumptions:
This is the real grunt work of the job and can take hours (the good news is that we did
it for you). The NZX (www.nzx.com) website contains the last 3 years of dividend data
for each stock. The next step is to add dividend payments on a cent per share for each
company for each of the last three years, ignoring special dividends and imputation
credits. Companies that do not meet the below criteria are excluded:
• Any company that cut its dividend over any of the prior three years is
automatically excluded. Our assumption here is obvious, any company that
cuts its dividend cannot be considered a dividend growth stock. This excluded
companies such as Hallensteins Glassons (HGL), Sky TV (SKT) and Steel & Tube
(STU). What we subjectively found is that many of the companies operating in
struggling sectors like retail and construction were excluded.
• Those that have not paid dividends for each of the last three years are
excluded. This immediately excluded companies such as New Zealand King
Salmon (NZK) and New Zealand Refining (NZR). Our assumption here is that
companies that do not have a long history of paying dividends are less likely to
continue paying dividends the future.
• We also excluded companies that had not grown their dividends at all. This
knocked out many companies such as Contact Energy (CEN), NZX (NZX) and Sky
City (SKC) whose dividends had remained constant.
After removing companies that did not have three years of dividend payment history,
had cut their dividend or had not growth their dividend we were left with 37 stocks.
• Finally, we excluded companies that had not increased their revenues each
year for the previous 3 years. Our assumption here is that dividend growth is
driven by top line revenue growth. This removed many of the best yield
generators on the NZX such as Air New Zealand (AIR), Genesis Energy (GNE)
and Mercury Energy (MCY).
Step 4, Dividend Growth:
We were left with 26 stocks each of which was profitable, had paid a dividend in each
of the last three years, had grown their dividend for each of the last three years and
who had not cut their dividend in the last three years. For the final step we averaged
out each company’s dividend growth as a percentage over the last three years. This
was then placed in order. The company with the highest average dividend growth rate
was at the top and the lowest was at the bottom. Since our goal was to find the top 20
stocks by rank we removed the seven stocks at the bottom of the growth list. These
unlucky companies were ARV, PFI, GTK, KPG, VCT and ARG.
The List:
Dividend Yield
We believe that we have produced a satisfactory list for dividend growth investors.
The average dividend yield across the screen is equal to 3.50%. We believe that this is
satisfactory considering the current low interest rate environment.
Average Dividend Growth
Across the 20 stocks the dividend growth rate ranged from 7% to 51%. The average
dividend growth rate was 18.22%. While it involves a significant amount of crystal ball
gazing, if this rate was maintained over a 10-year period the total yield on cost would
be a staggering 18.66% for the investor. This shows the power of dividend growth.
chart on the previous page shows a 3.50% dividend yield growing at 18.22% per year
versus a 6% yield growing at 5% per year.
Valuation
The average Price to Earnings ratio for the selected stocks was 19.28x on a trailing 12-
month basis. There is not much that can be read into this. Considering our goal is to
create a satisfactory dividend growth screen we would only want to relook at our
methodology if the average PE was extreme in any direction. For example, if it was
below 8 or above 40.
Market Capitalisation
One of our assumptions was that dividend growth stocks can come in any size or shape.
The smallest stock in the portfolio was Solution Dynamics Limited which was a tiny
$28.5 million. At the other end of scale was Auckland Airport and Fisher & Paykel who
both sport market capitalisations of over $7 billion. The average market capitalisation
across the screen was $1.78 billion. The pie chart below shows the allocation of the
screen by market cap:
The screen was incredibly diverse. Containing everything from technology to logistics. The only sector noticeable by its absence was utilities which had zero representation. The largest weighting went towards retirement villages which had 3 stocks; Summerset, Metlifecare and Ryman Healthcare. With only 3 companies we do not consider this an over-weighting. The table on the left has the final count but the
pie chart below visualises the diversification quite nicely.
How to Buy:
In this instance we believe that satisfactory results will be achieved through fixed
fractional allocation. There could be arguments to weight the screen towards higher
yielding or growth stocks but for the robust purposes of this article we will assume an
equal allocation of 5% per stock. 100% across 20 stocks.
When to Sell:
This would be an easy to manage screen and would probably only need to be looked
at once per year. We would sell if any of our dividend growth assumptions are breached:
• The company cuts, suspends or cancels its dividend. Or does not grow its dividend.
• If annual revenue declines.
• The company stops reporting a profit.
• Every 6 or 12 months the list can be revised. We want to be in the companies that are growing their dividends the fastest. Any companies that are not ranked within the top 20 can be sold.
• This rebalancing will ensure that the investor is always invested in the stocks
and industries with the fastest growing dividends.
Conclusion:
Any screen selection method has its limitations. Ignoring qualitative fundamental
information is the obvious limitation with any quantitative screen. This screen ignores any change of management, new product, company disaster or any number of qualitative factors. On the flip side, the objective nature of the screen removes any investor emotion, interpretation or biases from the decision-making process. The screen therefore, is only as good as its assumptions. An investor with a different definition of dividend growth will have different assumptions and get different results.
We believe that the goal to build a dividend growth “portfolio” on the NZX through quantitative data has been met. The screen above is diversified across market capitalisation and industry. Investors that are interested in dividend growth investing may consider using a quantitative construction method that we have outlined in this article as the basis behind their portfolio.