there was little change in the share of securitized mortgages during any of the boom years from the mid-1990s to the height of the boom. The share of these pools was 57% in 1995 when rent inflation began to rise, it peaked at 62% by 2002 before the steepest moves in home prices, and then declined back to 59% at the end of 2005 when housing starts and home prices peaked.
Within this group, there was a shift to private pools, much of which were subprime. But, as we can see in the graph, there was a gradual shift from Ginnie Mae to private pools from about 1990 to 2003.
…After 2003, the GSE’s began to decline as a portion of the market also. It was during this period that private pools shot from about 10% to about 20% of the market, until the private pool market collapsed in 2007. This period was not associated with a rise in homeownership, and included the last period of sharply rising prices followed by two years of flat prices.
What I find idiosyncratic about the chart is that it is based (I think) on total mortgage debt outstanding. Also, he charts the share of mortgages, rather than total amounts. Both of those factors tend to make the chart tamp down changes in dollar mortgage flows.
One point is that the issuance of mortgages by agencies was affected by loan limits interacting with higher house prices. My guess is that the substitution of private mortgages for agency mortgages took place in locations where house prices were rising faster than the loan limits adjusted.
Yet another point is that a lot of lending was in the form of cash-out refinances (people using their homes as ATMs). I may be wrong, but I don’t think that FHA was in that business.
Another chart shows the increase in mortgage debt by income class. Kevin writes,
The proportion of mortgage debt held by the bottom 80% of households declined during this period [2004 to 2007].
What I would want to see is the behavior of the ratio of debt to equity by income class. Suppose that everybody is using their homes as ATMs. If a rich guy with a million dollar home raises refinances his $400,000 mortgage for $500,000 and a poor guy with a $100,000 home refinances his $90,000 mortgage for $100,000, then most of the new mortgage debt goes to the rich guy. But it’s the poor guy whose equity is disappearing.
There may be something to the price effect you mention.
You are right about the effects of my graph, but i think the lack of expected associations between subprime and low income debt is still surprising.
I think FHA does refinancing, but in any case if the growth in private mortgages was due to high income households refinancing, that’s different than the standard story.
It does look like the middle quintile had slightly more growth of leverage than other quintiles, but there is a significant amount of homeownership in the lower two quintiles that is very high in equity.
Here are two posts with graphs. Let me know if there are other graphs you would like to see. I might have them.
http://idiosyncraticwhisk.blogspot.com/2015/02/housing-tax-policy-series-part-11-low.html
http://idiosyncraticwhisk.blogspot.com/2015/02/housing-tax-policy-series-part-12-low.html
Here are some other numbers:
In mid 2004, there were about $7.5 trillion of outstanding mortgages. By mid 2007, that grew to $11 trillion. Private securitizations rose from about $800 billion to $2.3 trillion over that time. Growth of about $1.5 trillion – nearly half of all net mortgage growth.
In absolute numbers, from the Survey of Consumer Finances:
1st #: % of households with mortgage on primary residence
2nd #: size of median mortgage
Bottom 20% of households, by income:
2004
16% $37,000
2007
15% $40,000
2nd Quintile
2004
30% $54,000
2007
30% $51,000
Middle Quintile
2004
52% $77,000
2007
51% $89,000
4th Quintile
2004
66% $97,000
2007
70% $115,000
top quintile is similar. It’s not even a question of scale. There simply was no net activity, either in refinancing or new buyers for the bottom 40% of households, by income, even though homeownership rates were nearly 50% for those quintiles even before the boom.
I have accepted the great housing bubble was still most a private market failure with the government leading too many horses to drink too much water. So the government entities were not the worst loan issuers but everybody simply over consumed on houses. The Bush years were about couples overworking themselves to higher monthly payment and an early grave.
Bill McBride at CalculatedRisk calculated Mortgage Equity Withdrawal (MEW) as a percentage of income averaged about 7% between 2002 and 2006. During the boom years of the 1980s MEW averaged about 3%. Currently MEW is around NEGATIVE 2% – people are paying into their mortgages and not taking out.
How much does MEW distort the macro understanding of the economy? For 7% MEW is not sustainable – it is as if homeowners went “Chainsaw Al’ against their own homes – sucking the equity dry and leaving the debt to default in the hands of some sucker investors.
http://www.calculatedriskblog.com/2015/06/mortgage-equity-withdrawal-more.html
Dan W., these things don’t happen in a vacuum. Normally currency grows at 8-10% during an expansion, but from 2002 to 2008 currency growth fell from 8% all the way to below 1%. Average equity proportions were pretty level until the liquidity crisis phase, and the Fed was counteracting any expansion from mortgage growth. NGDP growth was low compared to post-WW II experience. None of the income quintiles show large changes in loan to value ratios until the bust. There was generally a typical behavior. High prices and some increase in first time buyers meant there were probably more new mortgages that had higher LTV combined with aging mortgages which were generally gathering equity as prices rose.