The ongoing financial crash was caused by overleveraging at all levels of society, from Wall Street to Main Street to the slums. The initial cause, however, was the popping of the subprime mortgage bubble.

At their bubblicious peak, American homes were theoretically worth $24 trillion. The amount of wealth that has evaporated in the popping of the American real estate bubble so far appears to be in the $5 trillion range, to pick a very round number. The blogger Dr. Housing Bubble recently estimated the loss to be $4.68 trillion using Case-Shiller data. Another source estimates $6 trillion. And we haven't necessarily hit bottom yet. So the wealth loss already amounts to a quarter to a third of US GDP—greatly magnified, and spread around the world, by the simultaneous metastasizing of poorly-understood financial derivatives.

I've long argued that the massive ideological and demographic trends in our society toward "diversity" played an underappreciated role in the disaster. Now we're now getting very close to finding the smoking gun that proves my "Diversity Recession" theory.

Many readers have expressed doubts that minorities could possibly play a large enough role in the mortgage market to matter. Actually, they do. In fact, we can now see that, more than anything else, the Housing Bubble was a Hispanic Housing Bubble. Mortgage dollars flowing to Hispanics for home purchases increasing almost eightfold from 1999 to 2006!

 As I've  argued, the Bush Administration wanted to turn Hispanics into Republican voters by making them homeowners through easy credit. George W. Bush and Karl Rove don't deserve all the blame, however. Their lax mortgage policies were largely a continuation of trends to boost minority and low income mortgage access that were well under way in the Clinton Administration—as I pointed out in my June Takimag.com article on "The Diversity Recession." These lax mortgage policies also had the secondary effect of encouraging residential real estate speculation—"flipping"—by minorities and non-minorities alike.

The federal government doesn't make it easy for citizens to find information on mortgage defaults and foreclosures by race. But it does collect a huge amount of information by race on mortgages handed out, in order to encourage lending to minorities by threatening lawsuits against financial institutions accused of discriminating against them.

A very helpful reader sent me a link to the federal Home Mortgage Disclosure Act website: The HMDA National Aggregate Report,

I've chosen to look at conventional home purchase mortgages originated in 2006, the peak of the Bubble, the year of the worst "toxic waste" mortgages.

Unfortunately, I couldn't figure out how to break out subprime dollars, which is where most of the unexpected defaults occurred. But looking at total dollars loaned on the purchase of homes, prime and subprime aggregated together, is revealing enough.

I looked at total mortgage dollars originated for home purchases in 2006. It appears the minority share of overall mortgage dollars was slightly larger (35%) than their share of the population (about 33% in 2006). This is due to higher average mortgage sizes for minorities ($188,000) than for whites ($183,000).

This may seem counterintuitive—until you stop and think about it. Minorities tend to be concentrated in metropolitan areas with expensive land prices. Rural areas with very cheap land are almost all white.

Further, America's largest and most expensive state, California, the epicenter of the housing bubble and thus the global financial earthquake due to subprime defaults, is now majority minority (with non-Hispanic whites making up only 43 percent of the financially tarnished Golden State's population in 2005).

For home purchase mortgages originated in 2006, Asians averaged $255,000, Hispanics $183,000, non-Hispanic whites $183,000, and blacks $153,000.

Compared to 1999 (the first year in the federal database), which was before the Housing Bubble, it's striking to note how much more mortgage money has flowed to Hispanics. The growth in mortgage dollars for home purchases by Hispanics grew 691 percent from 1999 to 2006! Hispanics originated only $21 billion in purchase mortgages in 1999 v. $163 billion in 2006.

Not surprisingly, four heavily Hispanic states—California, Florida, Arizona, and Nevada—account for 50 percent of the mortgage defaults in America in 2007, and, due to California's ridiculous home prices, no doubt an even larger share of defaulted dollars.

Blacks also received far more mortgage dollars in that seven-year stretch from 1999 to 2006, up 397 percent from $17 billion to $84 billion. Both Hispanics and blacks participated heavily in the subprime market, with two to three times higher percentages of their mortgages being subprime than among whites. So much of this breakneck expansion in borrowing among Hispanics and blacks must have been due to subprimes, which is where the financial collapse started.

Despite rapid immigration, Asians were up only 218 percent in total new mortgage dollars from 1999 to 2006, from $24 billion to 77 billion. We know they mostly stuck to prime mortgages, at about the same rate as whites.

Total minority purchase mortgages taken out in 2006 were $360 billion, compared to $678 billion for non-Hispanic whites. So, minorities were slightly over-represented in purchase mortgage dollars relative to their share of the population.

Unfortunately, changes in reporting methodology from 1999 to 2006 make comparison difficult for non-Hispanic whites. (They weren't broken out separately from "Whites" in 1999, so the 1999 figures may or may not include some Hispanics. In contrast, non-Hispanic whites are identified separately in 2006.)

It's not all that important a methodological problem, though, because Hispanic borrowing wasn't huge in 1999. So, my estimate for non-Hispanic whites is that mortgage dollars flowing to them increased about 100 percent over those seven years.

The picture in refinancing of existing mortgages in 2006 is quite similar, with minorities getting 33 percent of home refinancing dollars originated in 2006. Interestingly, the average minority refinancing was bigger ($218,000) than the average non-Hispanic white refinancing ($188,000). Refinancing dollars flowing to Hispanics increased more than seven-fold, while whites were up somewhat more than double.

So the ethnic change wasn't quite as extreme as in home purchase mortgages, but they weren't very different. The total value of refinancing and purchase mortgages were fairly similar in size in 2006. So minorities accounted for about 34 percent of purchase and refinancing of mortgages in 2006.

I couldn't find usable numbers in the database on subprime dollars alone, although a more assiduous researcher may well be able to tease out the facts. But if minorities in 2006 accounted for 35% of all mortgages (see above), they would have accounted for a higher share of subprime dollars mortgages. Defaults so far have been concentrated in subprime adjustable rate mortgages. They accounted for 6% of mortgages and 39% of defaults.

Therefore, it's likely that it will turn out that the majority of unexpected default dollars, above normal trend lines, in 2007 were from defaults by minorities.

The conventional wisdom that emerged from the crisis of the Great Depression dominated American ideology until almost 1980. Similarly, the reigning ideas that congeal in the next few weeks about the causes of this crash will determine the course of politics for decades to come. Right now, the elite consensus (as in the 1930s) is that the free market failed. The truth, to which we blinded ourselves in an orgy of political correctness, is that the America of 2008 doesn't have the human capital to justify the valuations of wealth it thought it had.