" I'm happy if I can add 100 basis points of yield above the benchmark and would be thrilled with 150 basis points "
I would be interested if you could post an analysis of why this is so.
An important building block to understanding portfolio construction is that historically US markets have returned 9-10% on an annualized basis of which some portion of that return coming from dividend yield. Lately that yield has been in the neighborhood of 2%. A reasonable aspiration for a long term result would be to achieve returns somewhere close to that 9-10% which along with a proper savings rate will be most people's best shot at having enough when they need it.
Once that building block is taken to heart, time ideally is then spent on understanding the various behavioral biases that impede success. Once a little introspection is found I would then move on to understanding the big picture in the world, how and when to look like the world (the broad index that serves as the benchmark) and when not to look like the world. Also part of this can be personal assessment about how to generally navigate market cycles in a manner that allows for sleeping at night.
Against that backdrop I worked through the above process to conclude broad diversification with narrow products (individual issues and specialty ETFs) looking to mostly, repeat term coming, smooth out the ride for clients over the course of the entire stock market cycle was right for me.
If in a overly simplistic example the market returns 10% every year with some portion from gains and some from yield then the more that comes from yield the less volatility the rest of the portfolio needs to be exposed to. In a more practical sense if up a little is the most common outcome in a given year then extra yield can allow for a slightly less volatile portfolio (so a better risk adjusted result) but still allow for what I think of as proper diversification in terms of countries, themes and holdings with varying attributes such that there are no lopsided exposures that could hurt the portfolio in an extreme manner.
In my experience there a couple of risk factors that go with too large of a yield. One is that if every stock in a portfolio yields 4% then there is a good chance that the portfolio misses all sorts of market segments so then the give up is diversification. The other risk factors, and this is more of a big picture concept is that, as an example, a 6% yield in a 1% world takes some amount of risk. The end user may or may not understand the risk but it is there.





6 comments:
Thanks for your thoughtful reply.
I am not so sure that adding yield is beneficial in general, but perhaps can be in some cases. It is important for the cost/benefit of extra dividend yield to be analyzed in the context of the goals and tax implications to the beneficiary of a portfolio. There are studies that show higher yielding portfolios can actually reduce portfolio returns and that is why your statement caught my attention.
There are studies that show higher yielding portfolios can actually reduce portfolio returns and that is why your statement caught my attention.
I'm sure there are studies that support every single type of portfolio strategy. I think of my thoughts on this as being intuitive. A little extra yield means less volatility across all holdings. If one or two holdings (out of 30-40) go up 50-100% (which is very plausible) then that is even less volatility required. A holding targeted at 2% double then you get 200 basis points for the portfolio plus the 3% yield and a huge portion of the work is done for you...
It always seems everyone wants to 100% in one direction. I am currently adding yield(15-20% of account) in a mix of high yld-10% yld, utilities-4%yld, and drug type stocks - 3% yld. It seems logical to persue this coarse, but not for all of the portfolio and perhaps not all the time. But for the next six months, seems fairly safe.
Roger, thanks for your efforts.
It is true that the market has returned 8-10% on average (9.75% as of last year) but investors can't spend that, they can only spend compounded returns, and this is the rather straightforward reason that regulated investment companies were forbidden to cite average returns in their literature (not sure if that is still true in the post-modern era of regulatory capture).
This chart at http://tinyurl.com/4x2o3ef demonstrates the impact that variability in a sequence of returns alone -- never mind taxes, inflation, fees, slippage and the rest -- can have on a portfolio; e.g., just one significant negative year or a sequence of sub-par years can devastate the actual return an investor receives.
Bottom line, achieving the market's average return of 8-10% on a compounded (and appropriately risk-adjusted) basis, actually putting that money in investors' pockets, would be a top quintile performance: Any manager achieving it YoY should get a gargantuan bonus; any manager staying close might settle for merely humungous I suppose. JMO
RW,
You do a great service by pointing out the difference between simple averages and geometric averages as shown in the link you posted.
I wonder however whether the DJIA data in the link you provided includes dividends? From the Dow Jones website where they discuss the calculations of the DJIA, they say the divisor takes into account stock splits, spinoffs, and substitions, no mention of dividends.
http://bit.ly/m69sMz
Therefore, the chart you show would seem to understate the market's long term performance by disregarding the compounding effects of reinvesting dividends. I think the number Roger cited in his post is accepted as the compounded annual growth of the stock market including dividends.
Here is a link to a recent column by Jason Zweig pointing out the shenanigans many use comparing raw market averages (like SP 500) that don't include dividends to mutual fund performance/ETF/strategy de jour etc. which does.
http://yhoo.it/l19xsd
Here's a column by Allan Roth where he discusses Zweig's column and expands on it a bit.
http://bit.ly/kU71LI
This sort of sleight of hand technique makes me wary of the results from any strategy when it is compared to a market average.
Thoughts anyone?
Dear Random Roger,
- many thanks . . . maybe you have the software to make some tests ??
Just wondering how the comparison might look :
Compare
_______
yearly performance of basic Client Portfolio
with
_____
hypothetical Test-Performance of
ONLY One Holding of Double-Long
SPY [maybe Profunds Ultra have this?] : Exit Position when under 200 day Moving Average of SPY for 5 continuous days
&
Enter Position when Above 200 day Moving Average of SPY for 5 continuous days
..........
Is it possible that such a simple strategy might work ??
best regards
Ellan
Post a Comment