It takes a lot of courage to be a bull on homebuilder stocks these days. They exist, for sure. And they aren't on mind-bending drugs. In fact, they see the world much as the housing stock bears do. You won't find any uplifting messages in their reports on the economy and housing.

    What differentiate this small breed from the bears are their views on value. They largely argue that even if the economy goes into a double dip; even if home prices fall after rising modestly in the spring; even if inventory of homes continues to rise, some of these stocks are just plain cheap.

    That's because homebuilder stocks have been brutally beaten as of late. The ISE Homebuilder Index is down 80.24% over the past five years vs. a loss of 11% for the S&P 500. The group got a reprieve during the massive rally that ended earlier this year but have been slammed since May, when the homebuyer tax credit ended, falling 23% in the past few months vs 6.5% for the S&P.

    "Home sales data supports our positive homebuilder thesis," writes Michael R. Widner, homebuilding research analyst for Stifel Nicolaus. He upgraded KB Home to buy from hold in mid-July.

    Positive is a term that many view as generous. In this world, positive usually means not nearly as bad as everyone was expecting. Put the Paxil back on the shelf.

    Take existing home sales. They were down 5.1% in June to an annual pace of 5.37 million, better than the median 5.1 million pace predicted by economists in a Bloomberg survey. And new home sales in June were 330,000 - the second worst in the 47-year history of the data series - but better than the 310,000 consensus. Housing share prices jumped briefly on the "positive" news. "This shows how bearish sentiment was going into the market," says Joshua Steiner, managing director, Hedgeye Risk Management, and a vocal housing bear.

    Go back a month, and the news worsens. In May, new home sales plunged to 267,000, the worst on record; but the industry had braced for the sharp drop because they knew that the tax credit for homebuyers, which expired April 30, had artificially boosted spring sales. Summer would be payback time. Some investors had factored that in.

    UBS analyst David Goldberg sent out a report last week urging investors to buy on the dip during the current earnings season. Like anybody else who has studied the real estate market, he sees many bumps in the road but he expects the recovery to resume later in the year. One source of optimism for Goldberg: He predicts that the big write-offs are well behind homebuilders and that they will try to avoid discounting prices, even if that means fewer sales. Goldberg's report sounds more optimistic than the homebuilders themselves: the homebuilder confidence index registered 14 in July, the lowest in more than a year.

    The bears doubt that prices can hold. "Home prices are still too expensive, relative to historic metrics," says Barry Ritholtz, CEO and Directory of Equity Research for Fusion IQ. Unemployment is high and wages haven't budged in a decade, Ritholtz explains in a blog post last month. Prices need to come down. Prices are off as much as one-third from the peak, but at Hedgeye Risk Management, Josh Steiner says big price decline still lie ahead -- anywhere from "single digits to as much as 20%."

    The inventory overhang in the existing home sector poses big competition to new homes, even though that sliver of the market has a record low backlog of 210,000 homes.

    Ritholtz focuses on two metrics to explain the problem: One, historically, mortgage debt represented 40% of home value. Today it's 62%. That's too much debt relative to the value of housing, he says. Two, homebuilders built "about five million excess homes" relative to the fundamentals that typically govern household formation. That means going forward, the pace at which new formation absorbs excess housing is likely to be slow.

    And then there are the foreclosures, possibly as high as 5 million units, as well as something called shadow inventory -- homes that owners would like to sell but would at higher prices. At a minimum that's likely to be in the 2 million range and could even be twice that, says Ritholtz.