[via Yahoo] An article in Morningstar.com quotes the Oracle of Omaha, Warren Buffett, who said "In business, I look for economic castles protected by unbreachable 'moats", and points to Buffett's philosophy of wider moats being far more important than the deepest of moats.
Many investors easily surmise the first point about moats--that a competitive advantage is required--but they miss the importance of the second, which is above-average returns. If the business doesn't throw off attractive returns (making the castle a worthwhile target for competitors), then who cares if the business has a competitive advantage? Autos and airlines are two businesses with very high barriers to entry, but few new competitors are trying to crash the party in a race for single-digit ROEs or bankruptcy.
Over the years, Buffett has frequently referred to his desire to widen the moats of his companies, but it's rarer to see him refer to a moat's depth. In his 1995 letter to shareholders, he notes his pleasure in the widening of auto insurance subsidiary Geico's moat, thanks to rock-bottom operating costs falling further and improved underwriting. A year later, in his 1996 letter to shareholders, he notes that "The economies of scale we enjoy should allow us to maintain or even widen the protective moat surrounding our economic castle." In a 2001 memo to employees, he writes, "What should you be doing in running your business? Just what you always do: Widen the moat, build enduring competitive advantage, delight your customers, and relentlessly fight costs."
Thus, assuming the competitive advantage is of sufficient depth, it can be more productive to widen the moat than to deepen it. To purposefully strain the analogy, it doesn't really matter if the moat is 10 feet deep or 100. Knights aren't going to capture the castle by shoveling dirt into the moat and walking over the refilled trench; they're going to use a ladder or a drawbridge.