In the long-run however, (say 20 years from now), I cannot see anything else but increases in US government spending on defense. The world would still be a hostile place, and countries need to spend on defense simply to keep up with pace of technology. This means that if you buy today, you can still experience dividend cuts five years down the road, although chances are your distributions would recover within 10 years or so. This cyclical nature of dividend booms and busts makes relying on defense dividends riskier than relying on distributions from Coca-Cola (KO).
Surprisingly, some of the best times to acquire defense companies was right after the cold war ended in 1989. Other attractive times to buy defense companies occurred right after the official end of Vietnam war in 1975.
To summarize, the major bearish factor behind defense companies is the US Government spending situation in the next few years, and the need to reduce the budget deficit and curtail the growth of national debt. This could cause stagnating or decreasing military budgets. Since the major wars the US fought over the past decade are pretty much over, this could potentially affect profitability in defense companies.
The major bullish factors include cheap valuations, and the potential for a few of those companies to repurchase massive amounts of stock at these low valuations. If I were to invest in the sector, I would focus on the companies that reduce a large portion of shares outstanding. If net income can be at least maintained, the reduction in shares would lead to increases in earnings per share, which could also result in dividend growth.
Full Disclosure: Long KO