Seeking Alpha
View as an RSS Feed

Kurtis Hemmerling  

View Kurtis Hemmerling's Comments BY TICKER:

Latest comments  |  Highest rated
  • The Hidden Risk Of High Dividend Growth Stocks [View article]
    Thank you for your comment David. I respect your right to protest with silence. In turn I ask that you respect my decision to disseminate my research in the format that I believe best.

    I will report on my findings in my next article.
    Feb 18, 2015. 11:00 AM | Likes Like |Link to Comment
  • The Hidden Risk Of High Dividend Growth Stocks [View article]
    Dave,
    I am still investigating and will highlight my findings (and subsequent tests) in the next article.

    But there is another anomaly that no doubt led to turnover - which in turn led to lower income - as 'higher growth lower yielding' stocks were replaced. The sell rule looked at the indicated or announced dividend compared to what it was 52 weeks ago and if that number did not increase - the stock was sold. This was to give a timely sell if a stock announced a cut, instead of waiting around and collecting the lower dividends for a year and then selling because realized dividends dropped.

    However, you run into problems when a company delays the dividend announcement. If there is more than 52 weeks between dividend announcements - the indicated dividend did not grow between the 52 week period and thus the stock is sold.

    To correct this, I will ask P123 to add in a buffer of a couple weeks so stocks are not sold if the announcement is late.
    Feb 18, 2015. 08:32 AM | Likes Like |Link to Comment
  • The Hidden Risk Of High Dividend Growth Stocks [View article]
    There are differences in how you can compute dividend growth. For instance, this study did not count special dividends. But it did include ADRs. Compustat has a very good database. I tried to verify the ADR data with various other sources, but the other data providers were often spotty (missing years of dividend data) and did not always properly adjust for splits.

    The default format for calculating dividends per share in P123 is as follows:
    "Returns the sum (or count) of the dividends per share that were paid in the period specified. By default it returns the sum of REGULAR dividends using PAY DATES."

    I can adjust any of these parameters as needed (use ex-div date or the announcement date as well as using TTM instead of the fiscal year). Differences will come when the record date for a dividend occurs with-in the company's fiscal year but the pay date is months later and is included in the following fiscal year when calculating dividend growth.

    Other factors are exchange rates. Dividend growth is calculated based on USD per share. The dividend may have appeared to grow in one currency but after exchange rates it declined. Or vice versa. A company may grow its dividend 20% annually but if the inflation rate is 50% and you convert this back into USD income, it would not be considered a dividend growth stock in my books.

    Thus, there may be turnover with ADR's due to the above reasons. If the ADR experienced a capital loss, your re-investment will show an income reduction.

    But I will run the tests again excluding ADRs with the inclusion of a market cap minimum. I will include some of the suggestions in the comment section.

    I have also asked P123 how they calculate exchange rates and other factors pertaining to ADRs.
    Feb 18, 2015. 05:02 AM | Likes Like |Link to Comment
  • The Hidden Risk Of High Dividend Growth Stocks [View article]
    Thank you Craig for your comments.

    This article was not an attack on DGI or investors who use it to build income. I personally believe very strongly in dividends for many reasons of which I will not go into here. DGI stocks make sense.

    The article was to show factor testing. Fama and French tested various factors across the entire global universe of stocks and found a premium (or additional return over the market) with a low price-to-book factor. The same is generally true with smaller stocks and stocks with the higher 12 month trailing returns (minus the month you are buying - so 11 months). But do you go out and buy a low p/b stock and expect to make money? Hardly. As a factor, it is something to consider with all the other aspects of due diligence. Or value in general.

    The purpose of this article was to discover if the 'high dividend growth' factor led to higher future income streams. My limited testing showed that it did not. This was not meant to be a portfolio for trading but a factor test. What investors take away from this will vary. Some may not rely on trailing dividend growth rates to project forward income. Others will look for additional factors to combine with trailing div growth to create a model. And others, such as myself, will keep researching as to why this occurs and what the meaning of it is.

    To me, this is an interesting puzzle I want to explore. But I am not for or against anything in particular. I just want to see where the evidence leads me to understand what drives the stock price and income.
    Feb 17, 2015. 06:49 AM | 1 Like Like |Link to Comment
  • The Hidden Risk Of High Dividend Growth Stocks [View article]
    The portfolio turnover was very low... 10% - 15% or so and only when there was a dividend cut.
    Feb 17, 2015. 06:35 AM | 2 Likes Like |Link to Comment
  • The Hidden Risk Of High Dividend Growth Stocks [View article]
    No, this was not the case. The only rebalance point was if the holding was more than 50% over- or under-weight. The 4 week re-balance only affected the factor test for total return but not the 100 stock simulation.
    Feb 17, 2015. 06:33 AM | 1 Like Like |Link to Comment
  • The Hidden Risk Of High Dividend Growth Stocks [View article]
    I wanted to clarify in case it wasn't clear that all stocks tested, even the high yielding ones, were dividend growth stocks with 9 trailing years of growth and with an indicated dividend greater than it was a year ago.
    Feb 16, 2015. 04:59 PM | Likes Like |Link to Comment
  • The Hidden Risk Of High Dividend Growth Stocks [View article]
    Thank you for your comments. The article was not attacking dividend growth stocks nor was it suggesting a model to trade. The goal was to test whether a higher dividend growth number provided any additional benefit or whether the stock merely needed to grow the dividend even by a fraction of a percent. I believe that many DGI investors do very well for the reason that they do not rely merely on high dividend growth fir stock selection.

    The first chart is simply to illustrate how some investors view the relationship between yield and growth. I could have used 20 percent yield and 2 percent growth vs 0.5 percent yield and 100 percent growth as it was just to highlight the assumption of dividend growth persistence.

    I will write a follow up soon. I don't want to add too many rules when testing single factors, as these are not models. But if you have a simple rule you would like tested (e.g. marketcap above 1 billion; price above 5 bucks; yield above 2 percent), I will include it in my next round of tests. Please write out some very simple guidelines in the comment section. But it should allow for at least 50 stocks to pass at any point in time for the factor sorting.

    The portfolios had very low turnover as I only had one sell rule and that was if the dividend did not grow. 2008 was a bad year which increased overall turnover and average annual turnover was around 10 - 15 percent depending on the factor tested. The 100 stock simulations did not sell when yield dropped, only when dividends were cut...same with the div growth portfolio. But the replacement stock would be a higher yielding name. So the momentum argument is invalid in this instance. The 4 week rebalance was only for the factor test for total return.
    Feb 16, 2015. 04:47 PM | Likes Like |Link to Comment
  • Income Growth Vs. Dividend Growth: Apples And Oranges [View article]
    Sorry you missed the point of this article. It is a rebuttal of another article. The other article compares bond funds and debt instruments - which utilize leverage and call options - and then compares this to non-leveraged dividend growth investing.

    This article is not about comparing SDY to SDYL 2x. And 2x the leverage does not equal 2x the return over time. Just look at 3x leveraged funds and please read my previous comments as to why this is so.
    Feb 15, 2015. 09:09 AM | Likes Like |Link to Comment
  • Income Growth Vs. Dividend Growth: Apples And Oranges [View article]
    There isn't a problem but if you want to question the validity of the data providers it is best you do that elsewhere. Or sign up for free trials and compare it extensively. I already have.

    On May 30th there was one transaction of 300 shares. The adjusted price from Yahoo puts it at 21.79. Portfolio123 at 21.71. Ycharts at 21.96 (you need to select the total return chart for SDYL to get this). The current price is 54.83 with Yahoo and P123 while Ycharts has it at 54.83. You can calculate the % return from this.

    If you have further questions - please contact the data vendors directly.
    Feb 15, 2015. 12:31 AM | Likes Like |Link to Comment
  • Income Growth Vs. Dividend Growth: Apples And Oranges [View article]
    Bert, you seem to have questions as to the validity of Computstat/CIQ data. I encourage you to take that up directly with them. The big data providers (Reuters, Compustat, Bats Exchanges...) will have small differences at times due to restrictions for what counts as the opening print.

    I have tested various sources including those listed here. I prefer Computstat data. You are welcome to run various tests yourself if you have doubts and think Yahoo data is best. But I don't wish to dilute the message of this article or spend excessive time by going back and forth on data providers.

    The data produced the same result (with in 1%) on the tested sources despite small differences.
    Feb 14, 2015. 01:09 PM | Likes Like |Link to Comment
  • Income Growth Vs. Dividend Growth: Apples And Oranges [View article]
    Interesting. So if the market price was $100 per share and you sold call options with a strike price at $105 per share - someone would exercise their rights and pay a 5% premium rather than just purchasing shares on the open market?

    I guess it is possible but this has not been my experience. I would be grateful if it did happen though and would immediately re-purchase shares on the open market at a discount. Thanks for sharing.
    Feb 14, 2015. 12:56 AM | Likes Like |Link to Comment
  • Income Growth Vs. Dividend Growth: Apples And Oranges [View article]
    I compared the data from 3 different sources. There are slight differences in all 3 data sources but it is very slight. P123 uses compustat data which is considered the gold standard. I also used Ycharts total return chart and it was within 1% of P123 data since May 2012. Not too sure what you are looking for or trying to compare.

    If you don't use the adjusted price you will need to keep track of shares and purchase more when reinvesting dividends. But so much simpler just to use adjusted number.
    Feb 12, 2015. 02:30 PM | Likes Like |Link to Comment
  • Income Growth Vs. Dividend Growth: Apples And Oranges [View article]
    It is a 5 year period. The annual return is the compound annual growth rate (CAGR) and is not the arithmetic average return.
    Feb 12, 2015. 12:20 AM | Likes Like |Link to Comment
  • Income Growth Vs. Dividend Growth: Apples And Oranges [View article]
    Yes. In my P123 simulation, in the "%long" box I entered 200. Thus, the chart you see 29.7% annualized return is already 2x leveraged. Perhaps I didn't make it clear that the chart itself already includes leverage.

    If you want to do this by hand, use adjusted numbers which will already include dividends. Start with $38.44 for the historical starting price and use $78.24 for the end price. This will represent total return for the non-leveraged test. Over 5 years that is a 15.27% compound annual rate of return. It is a simple 103.54% rate of return. If you double this you get 207%.

    However, you cannot simply double the numbers and arrive at what a leveraged fund will give you. In a bull run it may be more than 2x and in a bear market it will be less than 2x. Why? It depends when you compound. Look at 2x daily leveraged funds vs. 2x monthly leveraged funds.

    If you invest $100K and use $100K of leverage or 2x the price move. But if prices double and you still only have $100K of leverage, this isn't going to give you twice the price. But when do you reset the leverage back to optimum levels? Daily, weekly, monthly? This will change your results beyond leveraging the entire 5 year period by 2x.

    As well, leveraging can have different results due to compounding. If successive months in a row are up (assuming 2x monthly leverage reset), your return is not just 2x the fund. It will be more. Consider if price went from $1 to $1.50 in a month. Next month is goes from $1.50 to $2.25. 50% up each month. Total return is 125% up. The leveraged fund when from $1 to $2. And $2 to $4. They received a 300% return which is more than double 125%.

    I compared the SDYL with my P123 method for available history and it is very close.
    Feb 12, 2015. 12:11 AM | Likes Like |Link to Comment
COMMENTS STATS
573 Comments
487 Likes