Thanks to two catastrophic bear markets over the past dozen years, the strategy of buying stocks and holding them forever has fallen into disfavor. But that doesn't mean a buy-and-hold investing strategy is all bad; it just needs some tweaking. So in that spirit, allow us to introduce a variation that we think makes a lot of sense. Call it "buy and hold and collect and grow," or BHCG for short.

The strategy is simple: You buy stocks that regularly boost their dividends and hold for the long haul. By doing so, you hitch a ride with cash-rich businesses that generate higher revenues, profits and cash flow year after year. The best of these companies are committed to boosting their dividends by double-digit percentages in all economic and market cycles.

BHCG stocks don't necessarily pay superhigh dividends, as does AT&T (T, news), with its lavish 5.9% yield. But current income isn't the point. Rather, the idea is to target companies whose share prices rise steadily along with their dividend streams. If the strategy works as expected, you earn a handsome yield based on the price of your initial purchase.

Some experienced investors believe BHCG offers the best mixture of safety and opportunity. One of them is Tom Cameron, who entered the investment business in 1953. Now head of Dividend Growth Advisors, in Ridgeland, S.C., and a co-manager of the Dividend Growth Trust Rising Dividend Growth (ICRDX) fund, Cameron requires any stock he buys to have annualized dividend growth of at least 10% over the previous 10 years. When he talks with company bosses, Cameron says, he makes sure big dividend increases are "part of the culture." He immediately boots out any company that cuts its dividend. In 2008, Cameron, 84, dumped all of his shares of Bank of America (BAC, news) after the company halved its quarterly payout. B of A shares traded at $28 at the time. They now fetch less than $7.30.

To see how dividend growth and share-price appreciation work in tandem, study McDonald's (MCD, news). In 2001, Mickey D's paid out 23 cents a share in dividends on a stock that averaged about $30 a share, for a yield of 0.8%. In 2002, McDonald's raised its annual dividend to 24 cents. Then the company's fortunes improved, and McDonald's decided to give more generously to its shareholders. By 2006, the rate was $1 a share. After a 15% boost in November 2011, the Golden Arches now pays at a rate of $2.80 a year. That's a 1,100% increase since 2001, or 28% annualized.

Dividend growth magic

Order some McDonald's stock today and you'll pay $99 a share. That puts the current yield below 3%, which isn't life-altering if you depend on dividends to supplement your income. But, remember, BHCG is a long-term plan, and the real story is the progression of the yield over time. If you paid $30 a share for McDonald's in 2001, your current yield on that original purchase works out to a lusty 9.3%.

And what does the future hold for McDonald's stock? Market volatility has been off the charts lately, and traders sometimes ignore company fundamentals. But over a decade or more, the market won't let a stock languish if a company relentlessly raises its cash payout. Over the past five years, McDonald's has generated growth in earnings and free cash flow (the source of money to pay dividends) of 15% to 20% a year. It's hard to imagine a company as large as McDonald's generating such rapid growth indefinitely. But low-double-digit growth should be within reach as McDonald's continues to benefit from growing consumer spending around the globe, especially in emerging markets.

Still, past performance is not always a harbinger of future results. A company can have a robust dividend-growth record but come with flaws that make it a risky investment. The most significant omen: when free cash flow doesn't support an expansive dividend policy. Rather than bottom-line earnings, free cash flow -- earnings plus depreciation and other noncash charges, minus the capital expenditures needed to maintain a business -- is the source of the money a company can use to disburse dividends. If a company promises more dividends than it generates in free cash, it needs to borrow money, starve the business or gradually liquidate itself.

Meridian Bioscience (VIVO, news), which makes test kits for diagnosing diseases, is an example of a company that is no longer able to cover its payouts. Meridian has a history of big dividend hikes -- 21% annualized over the past 10 years. But it finally froze its payout rate in 2011. This once-great stock has been awful since 2008, and its prospects are gloomy with the dividend engine in the repair shop.

More from Kiplinger's Personal Finance magazine: