It isn't hard to be bullish on housing. Prices relative to income are back close to normal. Interest rates are low. Population growth and household formation are up. New construction is almost nonexistent. Mostly hated, the future of homebuilders like Lennar (NYSE: LEN) and Toll Brothers (NYSE: TOL) is much brighter than most assume. It's ugly today, but the stars are aligned for a strong rebound over the coming decade.

All still true. But the housing market may have just taken a step back.

Real estate data firm CoreLogic revealed this week that the National Association of Realtors may have inadvertently overstated a widely followed measure of housing sales by as much as 20% over the past four years. NAR estimates that housing sales came in at 4.9 million last year. CoreLogic counted just 3.3 million.

What this could mean for the housing industry is subtle but important. Housing sales data is used to calculate months of supply. This is done by taking current number of homes on the market and dividing by monthly sales. It looks like this:

Source: Census Bureau.

The red line, six months, in this chart denotes the long-term average, but it's more significant than that. Six months of housing supply is typically the magic number that determines whether home prices are rising or falling. Below six months' supply, and prices are usually rising; above it, and down they go. This isn't a mysterious force; it's the strong hand of supply and demand.

The question now is whether the recent months-of-supply data is bogus. If CoreLogic is right and housing sales data has been inflated by 20%, then months of supply is higher than we think. With the current reported figure at 8.5 months, do the math: Reality might be closer to 10-11 months' supply.

What's that mean? It doesn't change current housing prices, but it says a lot about where future prices could be headed. And it ain't pretty. The higher the months of supply, the further prices need to drop before the market finds equilibrium.

How much lower is anyone's guess. I ran a slew of housing numbers in January and came to the conclusion that nationwide prices still needed to fall before hitting a happy balance -- probably another 10%-15% from current levels. If CoreLogic is correct, bump that up an additional 5% or so. This isn't an extreme forecast. Housing expert and Yale economist Robert Shiller said yesterday that "there's a substantial risk of home prices falling another 15%, 20% or 25% more."

Interestingly, homes nationwide actually look sensibly priced based on price-to-income ratios. An average income can buy a pretty average house today. Valuations aren't the cause of declines anymore. Inventory alone is the sole driver and could push prices below what makes sense on a price-to-income basis. If you're in the market to buy a home, the next year could be very kind to you. If you're trying to sell, this is an utterly terrible time, and there's a good chance that whatever you can fetch will be below anything resembling fairness. Welcome to bear markets.                                                                                                                

Asked about a housing recovery, JPMorgan Chase (NYSE: JPM) CEO Jamie Dimon recently sounded quite optimistic: "When people are working, when there are more jobs, more households forming and people go back to buying cars, they're going to want their apartments and homes. And that's when you'll start to see a recovery in home prices."

He's right -- but it's only half the story. The biggest driver of home prices over the next year or two won't be whether people can afford them. It'll be about supply. And right now, there's a lot of it.