If I recall correctly, early in this thread, or perhaps another thread, Jax talked about the long term strategy was to have 20 or so stocks in his portfolio. I've seen enough studies in the past to convince me that you can be very well diversified at 20 stocks. Plus, as an individual, 20 stocks is a reasonable number of companies that you can follow, read annual reports, follow earnings, etc... with out getting overwhelmed. Stocks will rise and fall with the market, but its highly unlikely 10 of his companies are going to go "belly up" over night. He may have 1 or 2, but that's 5-10% of his portfolio. If he's paying attention, he can get out before things go from bad to worse, most companies don't go bell up over night.
Another note, is that he's targetting companies that have a strong history of not only paying dividends, but increasing their dividend annually. Its generally a safe assumption that if a company has a history of paying dividends regularly and increases their dividend every year, they are going to be on stable ground. I've seen companies decrease their dividend, or suspend it. They generally take a pretty big hit when that news is announced, but they don't get driven into the ground. I'll assume that is a pretty big red flag in this type of investment, which perhaps means you sell that stock and find another to replace it. The goal is capital accumulation and to turn that capital accumulation into a steady stream of dividends that can support, if not completely fund your retirement.
I'll use Coke (KO) as an example of a company that has a strong history of providing dividends and I believe (correct me if I'm wrong) a company that has regularly raised their dividends) Lets say you retired in the late 90s w/ 5% of your 1,000,000 portfolio invested in Coke. Coke hit 40ish in the late 90s, which would mean you had roughly 1,250 shares. B/w the tech bubble, housing bubble, poor economy, etc.. Coke is just now (adjusted for a 2-1 split a few years ago) at the same level it was in the late 90s (40ish a share). In 1995, Coke was paying $.22 a share for coke per quarter in cash dividend. Today, they are paying $0.56 (split adjusted) per share of Coke.
In 1997, you're 1250 shares of coke were netting you $700 a year in dividends. Today, your 2500 shares of coke (assuming you haven't made any additional purchaes in coke the share total is to adjust for the 2-1 split) you're making $2,800 a year. So while the per share price of Coke hasn't been anything stellar in the last 2 decades from start to finish, you've seen a 300% increase in your cash dividend on a year basis.
When you have a dividend history like Coke, which is just an example of the kind of consistancy that Jax is trying to find, may not be the yield he is hoping to generate, the company can take a decade or two wading thru the mud, going nowwhere in particular, even taking a big drop in price from its high, while you sit back and accumulate an increasing dividend on a yearly basis. If all you are concerned about is the quartley, monthly or yearly dividend, you worry more about the consistancy that is devliered and less about the short term prospects of the company.