Does Chowder Rule Dividend Growth Stocks?

Summary

• The Chowder Rule compares the dividend yield and trailing 5 year dividend growth rate.
• Two yield categories of dividend growth stocks are analyzed.
• The Chowder Rule appears to have a very strong correlation to returns on one but not the other.

Doe the Chowder rule enhance or detract from dividend growth stock returns? This article will serve up some food for thought.

Chowder Rule Defined

What is the Chowder rule? Chowder is the handle of a Seeking Alpha contributor. Chowder writes sensible articles on the practice of dividend growth investing. After refining the list of dividend growth stocks down to a short-list of high-quality names, Chowder examines the trailing dividend growth rate and dividend yield to determine if the mix of value and growth warrants a timely buy.

Chowder's readership has turned this principle the famous Chowder Rule:

1. If dividend yield is 3 percent or higher - the sum of the dividend yield and the trailing 5 year dividend growth rate must be at least 12
2. If the dividend yield is less than 3 percent - the sum of the dividend yield and the trailing 5 year dividend growth rate must be at least 15
3. There is a separate set of rules for utility stocks

My first impression is that the logic is sound. This reminds me of Peter Lynch and his use of the price-to-earnings ratio with the projected growth rate (PEG ratio). We should demand more growth if a stock has less value and demand more value if a stock has lower growth. The two should complement each other for the total package.

But does a sensible system translate into a tangible benefit? Will this result in higher total return? Let's find out.

Dividend Growth Chowder Test

The test will only consider dividend growth stocks with a trailing 10-year annual dividend increase.

This is not a buy and hold strategy. Why not? Because then the test would be meaningless from a statistical point of view. We want to know if stocks which meet the Chowder Rule provides additional return while fulfilling the rules requirements. If you wanted to know the average annual return of value stocks, your portfolio must hold value stocks. The same is true of growth stocks or dividend stocks or any other factor you can think of. We need to see the correlation between return and the Chowder Rule. If a stock fails to meet the Chowder Rule, it is put in the non-Chowder-approved portfolio.

The test starts in January 2000 and ends in January 2017.

Chowder Rule and Dividend Yield 3% Minimum

Dividend yield is a minimum of 3% and the combination of the trailing 5 year dividend growth ratio plus the dividend yield is at least 12%. This is a 'Chowder Approved' portfolio.

Dividend yield is a minimum of 3% and the combination of the trailing 5 year dividend growth ratio plus the dividend yield is less than 12%. This is the Non Chowder Approved portfolio.

Overall, we can say that the Chowder Rule in this instance does lead to higher risk-adjusted returns. We see an annual outperformance of 2.39%. Sharpe and Sortino ratios compare return to volatility and if we use these metrics we can say that the extra upside that comes from the Chowder Rule is worth the additional volatility.

Yet the one feature that disturbs me a little is the Chowder Crash in 2015. What happened there?

A closer examination shows that the oil sector (GICS 10) is largely to blame. This major portfolio drawdown could be avoided by removing the oil sector or MLPs in general. But many would cry foul saying you cannot remove entire sectors in retrospect to make the backtest look better. Maybe you would have avoided oil and maybe you would have bought it.

An Added Industry Performance Rule

A better rule which makes sense if you want to cut down on risk is to remove any industry, which over the trailing 6 months, has under-performed the market by 10% or more. If we apply this rule and re-run the test, the results are more stable.

Minimum Yield 3% - Chowder Approved Portfolio

And this is the effect of that same rule in dividend stocks with a minimum 3% dividend yield where the Chowder number is less than 12. Some tickers in this portfolio is Exxon (XOM), Chevron (CVX), International Business Machine (IBM), Altria (MO), Qualcomm (QCOM) and Occidental Petroleum (OXY).

Minimum Yield 3% - Non Chowder Approved Portfolio

Some stocks on this list are AT&T (T), Verizon (VZ) and McDonald's (MCD).

What this shows me is that the Chowder rule can definitely enhance returns, but as Chowder himself has pointed out this is one of the final considerations before buying. You first want to ensure that you are buying into a high-quality company. I am sure that Chowders filtering of a high-quality company would improve these results even further.

Chowder Rule and Dividend Yield Under 3%

The next set of tests looks at the Chowder formula for stocks which have a dividend yield of less than 3%. The Chowder Number needs to be at least 15 if we are to buy. We will run with and without the industry performance rule that will remove lagging industries.

Yield Less 3% - Chowder Approved Portfolio

Yield Less 3% - Chowder Approved Portfolio Remove Lagging Industries

Yield Less 3% - Chowder Non Approved Portfolio

Yield Less 3% - Chowder Non Approved Portfolio Remove Lagging Industries

I am having difficulty finding any additional benefit from the Chowder Rule when analyzing lower yielding companies. To my eye, it seems to make little difference whether the 5 year dividend growth rate is high or low in these companies with a small dividend yield.

Yield On Cost

A final consideration is to look at the yield on cost of the 4 portfolios. These portfolios will all have the 'remove lagging industries' rule applied.

Minimum Yield 3% - Chowder Approved

Current Yield: 5.6% Yield on Cost: 97%

Minimum Yield 3% - Non Chowder Approved

Current Yield: 4.0% Yield on Cost: 31%

Minimum Yield Less Than 3% - Chowder Approved

Current Yield: 1.9% Yield on Cost: 9.4%

Minimum Yield Less Than 3% - Chowder Non Approved

Current Yield: 2% Yield on Cost: 9.4%

What I Learned From This

What am I taking home from this? The Chowder Rule seems to work very well on stocks of a higher yield. Good job Chowder!

What areas might there be room for improvement? One might be to tweak the formula similar to the PEG ratio. For instance, we could compare the price-to-dividend ratio (inverse div yield) to the trailing 5 year growth. This would allow for a single number rank that makes sense even when dividend yields are very high or very low.

Another area would be to consider other ratios besides the trailing 5 year dividend growth number. My tests show that this number has little forecasting ability for the stock or future dividends. Perhaps using the trailing 5 year sales growth or trailing 5 year earnings growth or the forward 5 year earnings growth estimate might make more sense.

What are your thoughts? Does Chowder in fact rule?

Analyst’s Disclosure: I/we have no positions in any stocks mentioned, and no plans to initiate any positions within the next 72 hours. 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.

Comments

[SDS (Seductive Dividend Stocks)]

Kurtis,
THANK YOU FOR RESEARCH.
Although I do not use Chowder rule directly I think it is good enough metric for DG beginners. IMO turnover is quite strong, and (as expected) number of companies that satisfied Chowder rule was quite large in 2008/2010 (so peak on 3rd part of your 1st picture). So 2 questions: performance if rebalance is not so often and where average investor find cash to accommodate peak for this period /I know that nobody can answer on 2nd question 8-)/.
Also in my bare eyes there are 2 periods (before and after ~2009) when Chowder-like portfolio outperformed SP500. Outperformance was lower before 2009 and stronger after it, I guess because of extra low interest rates in 2009-2016. Hence I'd not expect strong outperformance of DG stocks in bull market with more historically common interest rates.

SDS

[Dale Roberts]

Your article suggests that a 3% yield and 10 year DG history was a magic bullet that beat the SP 500 by 4% annual and 6% annual when one applies the Chowder Rule. That's massive of course

I would love to see more on this, and more support, and test of some buy and hold and add scenarios of the 3% plus threshold.

Once again an index that has been tracking higher yield large and mid cap stocks from 2000 did not find this level of alpha (though it did find generous alpha over total underlying universe). At the shop, we recently created an international portfolio based on these high yield MSCI indices. I believe the universe for US MSCI is the total investable US stock markets, representing about 2500 companies.

Could there (have been) that much additional alpha found by way of slightly more generous yield 3% vs MSCI index parameter of minimum 30% above market average, combined with a 10 year DG history of your study vs min 5 year DG history of MSCI indices?

Perhaps the lesser cap universe of MSCI becomes an anchor, or we could look at it that the SP 500 inclusion becomes an advantage in the high yield landscape.

Hope this all makes sense. But net net, comparing your results to that of the MSCI high yield index. With thanks ...

Dale

[Dale Roberts]

Thanks Kurtis, I appreciate all of the work and replies. I think we are on the same page now that I understand the universe. I would kindly suggest first defining the universe in each article. For my mind, in the land of dividend growth investing the number of years of dividend growth history would be the most or of the most importance. We know the recent history of the Dividend Aristocrats, ha.

I will post my question or suggestion in the next post.

Dale

[Dale Roberts]

Hi Kurtis 3% yield with 5 yr DG history has alpha (over sp 500) of 10% is from your article... Alpha 10.39%.

3% yield insisting on 10 year DG is from your above comment. "If you demanded that those stocks also be 10 year streak DG stocks, you actually lose 0.5% annual performance compared to the large basket of S&P 500 stocks with a min 3% yield." (10.39%-.50%).

All said, I understand what your data was showing. It's throwing up some contractions I think. And to clarify the universe is sp 500?

Dale

[Kurtis Hemmerling]

Aaahhh... I wonder if we are talking about the same thing.

The stocks tested in the article all had a 10 year dividend growth streak as a minimum. That was to build the stock universe.

The 5 year trailing dividend growth rate was the factor tested on the 10 year DG stocks.

[Kurtis Hemmerling]

One reason might be that I didn't mean to use the S&P 500 as a benchmark. Although it is automatically included I see on my test. I should have used the RSP Guggenheim S&P 500 equal-weight ETF as a measuring tool. Annualized return over the same period is 8.52%. \

By this measure, S&P 500 stocks with a dividend yield min 3% (and held equal weight) would have an average 3.5% annual out-performance over the period. If you demanded that those stocks also be 10 year streak DG stocks, you actually lose 0.5% annual performance compared to the large basket of S&P 500 stocks with a min 3% yield.

The majority of the out-performance comes from equal-weighting. So the size effect plays a massive role here. Value, as determined by dividend yield, plays a role.

The goal for this article was to test the Chowder Rule against itself - not against a common benchmark. If I meant to do that - I should've, would've and could've spent more time showing what an equal-weight, appropriate dividend yield basket of stocks would look like - with and without any dividend growth requirement.

[Dale Roberts]

Thanks Kurtis, so from article a 3% yield with 5 yr dg history has alpha of 10%, but go to 10 dg history and it's down to 3% alpha. The difference is non sp500 universe?

Dale

[Kurtis Hemmerling]

Sorry Dale. I am a little daft at times. I don't follow.
What two things are you comparing? Where does the 5 year dg come from?

[David Van Knapp]

Kurtis, Your universe was the SP500, right? All stocks in your portfolios were in the 500 at the same times?
Dave

[Dale Roberts]

While the near 20 year numbers are impressive, we see the "Chowder Rule" seems to find trouble or volatility in certain periods.

The charts (Chowder and non Chowder) seem to suggest though that one could simply insist on a 3% plus yield and destroy the SP500 with 10-12% alpha. That seems a little over promising. That does not mesh with the a look at the MSCI high yield indices that asks for higher yield and a 5 yr dividend growth history. The alpha is about 1% annual. And that slight alpha might come by way of mid cap inclusion, not the higher yield plus DG history.

Your numbers don't seem to mesh with MSCI that has been tracking a similar (but not exact) approach? It's just so far off? Kurtis what do you think is causing this?

Dale

[Dale Roberts]

Interesting work, thanks Kurtis. I certainly was one who asked for a look at dividend growth rates, but did not request a look at the Chowder rule.

If you have time I'd certainly still like to see a look at dividend growth rate bands. Perhaps with a 1.5% yield / plus inclusion parameter. Being selfish I'd like to see that 10 year plus dividend growth history as well.

And perhaps for SP 500 members and non members.

With thanks, Dale

[Chowder]

The Chowder Rule is useless if you are looking for a strategy that indicates forward looking performance guarantees. I didn't come up with it to determine who would be an out-performer and who wouldn't be.

Most people don't know how or why I came up with that criteria.

The criteria was based on my personal long-term rate of return I want from our investments. I'm not into trying to "maximize" returns. I have no way of knowing how to manage for that.

I did want to own companies that paid dividends, and I did want the initial dividend to be 50% higher than what an S&P 500 Index might pay. The S&P 500 is generally around 2% so a 3% yield or better is what I shot for.

My objective was to try and determine a way where most of that 8% annualized return over the long-term would be buoyed by the dividend which is much more reliable to count on than share price, especially since the dividend is part of the total return formula.

At the time of my back testing it did appear that a combination of a 3% yield and 9% dividend growth or better showed better than 8% annualized rates of return over the long term. The higher the yield, the lower the dividend growth rate needed to be. The lower the yield, the higher the dividend growth needed to be.

All we need to achieve our long-term objectives is an annualized rate of return of 8%. Those who have different objectives this is probably of no use to them and I wasn't the one promoting it, only the one implementing it.

It's much ado about nothing actually and most should probably move on and find ways to achieve the objectives meaningful to them, provided they know what their expectations are as opposed to chasing the market.

[Kurtis Hemmerling]

Thank you Chowder for commenting. I hear what you are saying. I wrote this because many investors were requesting it and I finally relented. If further testing is requested I will remove your name from it.

[Chowder]

Kurtis, I have no problems with what was written, I only wanted to clarify the motivation behind the criteria. If people want to look into it as part of their DD routine, I think it's important to understand that those numbers are based on what I need when it's time to reach the distribution point.

Since I only need an 8% annualized return, I think I can do that with less risk taken if I can get close to half of that return from yield and dividend growth.

If the companies chosen using the Chowder Rule ended up showing a very high rate of return over the time frame in which the back tests were done, it was coincidental and not predicted.

It wasn't a leading indicator of forward returns, it was more of an insurance policy to create a more reliable and predictable income stream. This is a different context from what others perceive.

You may continue using my name without pause (I am not offended) because there may be others who want a little more clarification on what the intent was behind the numbers.

Like anything else used as an indicator, it has strengths and weaknesses, and since I am the one that used the data, I'm aware of what some of those strengths and weaknesses are.

I thought your article was well written.

[diamondjimbob]

Kurtis,

I'm curious about one of the items you mentioned. You said that you included IBM as an example company in the sample portfolio "Minimum Yield 3%". Whether it was Chowder Approved or not isn't my problem. You are including IBM as a company with a 3% dividend. That certainly is true these days, but it wasn't true in January 2000 where your sample starts. In January 2000 IBM was paying a quarterly dividend of $0.12 per share. The price of a share of IBM at the beginning of January 2000 was about $112.00. That results in a yield of less than .5%. It certainly would not have qualified as a stock with a minimum 3% yield at the time. I didn't look at any of the other companies.

Is there a problem with the companies you are including in the data, or am I missing something?

[Kurtis Hemmerling]

I use Portfolio123 which uses CapitalIQ as a data source. It is point in time.

This means that when I start on Jan 2000, it looks at companies (even ones that are now delisted) which had a dividend yield and trailing 5 year dividend growth rate at that exact point in time. It only uses available data which was known at that time. For instance, if an earnings item was wrong and later revised, the database would give you the erroneous figure and then adjust only on the date it was changed.

[frommi]

I can`t really reproduce these results on p123, can you give a clue what your rules outside of the chowder rule are (buy/sell)? Position sizing, ranking etc.?

[Kurtis Hemmerling]

This work is done in the screening backtest area. No ranking, just the two Chowder rules in the rule section and the 10 year div grow universe. Yield min 3. Yield + div grow minimum 12.

[frommi]

Ok thank you, but with weekly rebalancing into equal weight without slippage or transaction costs this looks very unachievable. Can you change the rebalance to yearly to give a more realistic picture?

"Slippage is only applied when a stock enters or exits screen. It is not applied to rebalance to equal weight."

[Kurtis Hemmerling]

I understand that this isn't trade ready. This wasn't a portfolio construction test, as in how you personally choose to trade a dividend growth system. This was a factor test to determine the correlation between the Chowder Rule and total return. Designing and testing a model requires lots of different analysis and tools and a whole lot more time. But hopefully the Chowder Factor test will give you some ideas for your own portfolio construction.

If you have Portfolio123 you should be able to test it a hundred different ways and find something that works for you.

[Mudville9]

Kurtis, how often is the CRI (Chowder Rule Index) rebalanced? Monthly? Quarterly? Thanks for the insight. Great article.

[Kurtis Hemmerling]

Every week to ensure proper placement. But keep in mind that it only changes in and out of Chowder approval based on dividend yield and 5 year trailing div growth. Not high turnover factors.

[Dividend Pro]

Critical takeaway:

Minimum Yield 3% - Chowder Approved

Current Yield: 5.6% Yield on Cost: 97%
YOC = about 17x increase
----------------------...
Minimum Yield Less Than 3% - Chowder Approved

Current Yield: 1.9% Yield on Cost: 9.4%

YOC = 10x smaller increase than for 3+% portfolio.
Current yield = 3x smaller than 3+% portfolio.
------------------------

High yield wins.
Total return 5x greater for 3+% portfolio.

Where are all the dividend-hating commenters hiding now?

[Dividend Pro]

Very good work! Useful info, and refreshingly unique article.

[billinsd]

I have personally met Chowdah.
He is very humble,very bright and Im sure he would tell you his rule is one tool in his work chest.
Thankfully he got me into UTES and it has paid off well for me.
While Chowdah is the only person from S/A I have met in person,others here have helped me greatly.
Mr Fish' work is not to be ignored,his lists are a bible to dividend growth investing.
Bob Wells,Chuck Carnevale,Dave Crosetti,Brad,Eric Landis,and a host of others.
Forgive me if I did not name everyone.

[Cashflow Curator]

"A better rule which makes sense if you want to cut down on risk is to remove any industry, which over the trailing 6 months, has under-performed the market by 10% or more" So would that be the "Chowder + Momentum Crackers" rule? Or perhaps the "Chowder, hold the potatoes" rule? :)

[David Crosetti]

A better place to find that commitment would be in the work of David Fish with his Dividend Champions, Contenders, and Challengers.

[Kurtis Hemmerling]

Thank you for your comment.

The base universe of this test would be equivalent to Dividend Contenders plus Champions using a point-in-time database to remove survivorship bias.

[garymc98]

If you want to identify companies whose boards and executives are devoted to dividend payment and increases, I can think of a couple of ways. One is to go around asking them and saying "You sure? You promise?" Another is to look at their past performance, as with one of the factors in this rule, and depend on it's predictive capability. Even though we're constantly advised "Past performance is no guarantee..," what else is reasonably practicable?

[richjoy403]

>>>Even though we're constantly advised "Past performance is no guarantee..," what else is reasonably practicable?"<<<
__________________

gary -- Dividend growth does not occur from the wishes and intentions of a well-meaning BoD.

I suggest you consider the root source of dividends...earnings growth and FCF in relation to the dividend payout ratio are the most reliable indicators a dividend-payer is likely to continue.

Rich-unck:12hrs

[richjoy403]

gary -- I intended that would read "I suggest you SUPPLEMENT dividend history with the root...

[Kurtis Hemmerling]

Thank you garymc98 for your insight.

The idea was not so much as to what the board of directors intend, but to find a factor that has some forecasting ability. If it turns out that a sentiment survey on the board of directors has a correlation to high total return 10 years down the road and a whopping big yield on cost - then I am sure we will see that kind of survey done. But as it is, we are trying to look for estimates and guesses and backward looking factors that consistently results in higher total return. Not an easy feat and only half science based.

[Hungry for Knowledge]

Should clarify : growth rate of yield, stated as a percentage, i.e. 10% growth rate, etc

[Hungry for Knowledge]

The Gordon Growth Model (GGM) is the official name of this metric. Yield+growth of yield= anticipated growth of share price

[respond2you]

Thanks Hungry!

[BoomBoom99]

"Yield+growth of yield= anticipated growth of share price"

Incorrect. You're confusing yield with dividend. Yield is annual dividend/stock price. Dividends grow, yield does not.

[respond2you]

Hm. Thanks BoomBoom. The Dividend Drill defines the simple form as :
Return = Yield + (DGR + Share Shrink)

so i guess it's not the same as the GGM.

[respond2you]

Is there any difference between the Chowder Rule and using the Dividend Drill (http://bit.ly/2jQBCA5) with a floor on yield?