Please Note: Blog posts are not selected, edited or screened by Seeking Alpha editors.

My Portfolios Construction Principles (Jan 21, 2012)

The purpose of my real portfolio is create good total return but I concern more about income flow that rise in average faster to compare with real inflation than about capital gains. I rely on history / that shows that income stream due to dividends and grow of dividends provides ~90% of total return.

Hence, I invest in companies that pay dividends (distributions) and
a) dividends growth with time (recent DGR>0) at least during last 5 years and yield is above 1% (see footnote 1)
b) dividends are steady (recent DCR = 0, although long term DGR is preferably positive) at least during last 10 years and yield is above 8%
c) dividends (paid at least during last 5 years) fluctuate with time (so annual DCRs can be positive and negative, although long term DGR is preferably positive) and yield is above 10%,
d) dividends were cut (but not omitted or suspended) recently but I expect recovery and yield is above 3%.

As a small investor I almost always ignore market cap and trading volume (except if they extremely low, e.g. only 1000 shares are traded per day or that I call nano-cap) because my buy/sell activities cannot disturb a stock price. Usually I buy whole chunk of shares I want to own at once, sometimes I sell 33% or 50% of my holding if stock price strongly appreciate (see footnote 3).

I use modification of equal-weighted portfolio (see footnote 2). I think my portfolio should be widely diversified and I monitor about 1000 companies that I split into 4 groups:
a) Medium confidence in ability to pay dividends in next 10+ years - I invest 1X real $ in such company when I consider that stock price is right,
b) High confidence in ability to pay dividends in next 10+ years - I invest 2X real $ in such company when I consider that stock price is right,
c) Low confidence in ability to pay dividends in next 10+ years - I invest 0.5X real $ in such company
d) Zero confidence in ability to pay dividends in next 10+ years - no real money are invested for long term in these companies (see below about short term).
The "status" of company can be changed and it can be moved between these groups.

I buy a company only when I consider that it stock price is right for given financial conditions of company and status of economy.
My real portfolio is spreaded between regular taxable account and IRA account. I collect high yield stocks in IRA and low yield stocks in regular account. I use "buy and monitor" approach for my real portfolio.
I "purchase" stocks from groups 1-3 in my artificial portfolio when I do not have enough cash and if I consider that company is belong the group 4. I use "buy and forget" approach for my artificial portfolio.
My real portfolio is diversified between different industries and countries.

I usually long for my investment (5-10 years) for groups 1-3 companies there "equal value at risk" approach - i.e. invest the same amount of money in each company from the group is applied. For these 3 groups I essentially use quasi equal value at risk with coefficients from 0.5X to 2X where X is "money on the table".

I use short-term dividend capture approach for group 4 stocks and hold very few (couple) of them in IRA account for less than 1 year (added 11/4/2013 - I use this in 2012 when I expected significant special dividends, I almost phased out this in 2013).

I constructed and maintain my portfolio with following goals in mind:
a) insure most of total return via dividends and their growth;
b) perform as good as market in upturns;
c) underperform market during strong (>10%) inflation;
d) outperform the market during downturns;
e) underperform market during bubbles with non-dividend paying stocks (I expect to see bioscience and nanotechnology bubbles in XXI century);
f) overperform market during bubbles with dividend paying stocks (I expect to see a bubble with such companies /drug??/ in XXI century).

Portfolio expenses is the concern, so I try to minimize them - trade rarely, use free information as much as possible, consider taxes (select tragitional or IRA account to minimize taxes), etc.

As I mentioned I also maintain the artificial portfolio with purpose to learn as much as possible about dividend stocks. I do not care about diversification of my artificial portfolio and about companies included in this portfolio (so until they disappear or move in real portfolio they stay in this artificial one).


1. Yield should be higher if you don't have much time before retirement.

2. I'm not sure that this is optimal case.
It seems interesting to compare performance of Div. Champions like portfolio (positive DGR for more than N years) as follows

a) cap-weighted
b) equal weighted
c) dividend $$$ weighted (how much $$$ company distributes in dividends, not dividend per share) /see footnote 4/

with quarterly, semiannual and annual rebalances and dividends investments as

i) DRIP-like in the same company
ii) re-invest in accordance with weights a) b) c) above between different companies
iii) keep for new dividend champions (how?).

It seems also useful to compare such indexes with S&P500.
What do you think about such project?

3. See "A Fool and free money" at

4. /Added 21 Aug 2012/ SA members Selling Theta and Counterpoint discussed "income weighting" portfolio for DG investors (in and in ) I touched early in this post (see also footnote 2 above). Honestly, I dislike the idea (see my comments to these articles). I prefer "equal value at risk" approach described in this post. Accordingly an investor spends the same amount of money (ignoring inflation) in each company in portfolio.