As big a bogeyman-in-the-closet as ARM resets are claimed to be by doom-scenario believers, this is the ONLY research I have found that examines the magnitude of probable foreclosures.
It doesn't take a genius to quote how many billions or trillions of mortgage dollars will reset in '06 or '07. Big deal. The only relevent issue is how much of that will come back in foreclosures, and that question is never, ever answered supportably by bubbleheads.
I realize many here operate on the assumption that most people are essentially blundering idiots, but lending institutions have become very good at forecasting default rates, and the conclusions in that report tend to support that fact.
If anyone knows of any other bonafide research into ARM reset defaults, let's see it.
You make good points Dr. Who. I would only add that lending institutions have become very good at forecasting default rates during a prolonged period of falling interest rates, low inflation, rising property values and high employment. If current interest and inflation conditions persist, we should expect many of their estimates to be proven wrong with time.
Dr Who said: "The only (sic) relevent issue is how much of that will come back in foreclosures"
Dr Who makes a good point about backing up arguments and claims (even with the condescending tone.) However, the claim quoted above is simply incorrect. One of the risks of ARMs is affordability, and in a rising interest rate environment some ARM'd homeowners may be forced to sell their properties. This will add to inventory numbers and can lead to downward pressure in pricing, regardless of foreclosure rates. Sometimes it's not enough to have research to support your claims - sometimes you need to have a claim worth supporting.
Dr Who makes a good point about backing up arguments and claims (even with the condescending tone.) However, the claim quoted above is simply incorrect.
is simply inaccurate. (Couldn't resist. Maybe dead-on for tone, though.)[The devil has my keyboard, he will pass.]
The link provides a mountain (for me) of data and hypotheticals that cover many other scenarios/possibilities than foreclosure rates, yes? It is readable (even for me!) and persuasive --covering 60%(!!) of all mortgages in this country.
A mighty find. Thanks again to DoctorWho.
Hmmm Im not clear on your point. I think perhaps you misread what I wrote. Alas, I wasnt talking about the Cagan research in my comment. I was referring specifically to the foreclosure relevance statement that DoctorWho made. Thus I quoted it specifically, to avoid any confusion. I assume that since you believe my comment was inaccurate that you may really believe that foreclosures are the only relevant issue. If that is the case, Id like to hear why you believe that.
"I realize many here operate on the assumption that most people are essentially blundering idiots, but lending institutions have become very good at forecasting default rates, and the conclusions in that report tend to support that fact."
With the advent of the MBS market, no one is all that worried about loan quality, believing that packaging all of these risky loans into one security minimizes the risk to a tolerable, predictible level through diversification.
Those risk averse lenders are more likely to lend if they believe they can sell the loan to someone else. They have no incentive to check the credit expansion, as they do not suffer if any individual loan fails.
With rising property prices, lenders aren't concerned with loan quality, because the collateral on the loan went up in value.
They wouldn't worry about real losses if they think they could sell the property and recover their money.
How else could people in So Cal get loans greater than 500,000 when they only make 50-60,000?
On the way up, this makes everyone look like geniuses--borrowers, brokers, and lenders, and reinforces itself. Liquidity begets more liquidity, and reduces defaults--for a time.
The problem--systems with positive feedback become inherently unstable.
In a market, that is the point when "everyone" knows something is true (ie. "Real Estate always goes up" or "We can always refinance or sell").
Everyone in the business is banking on increasing property prices to bail them out--not only borrowers, but lenders who made a risky loan in the belief that property prices would rise, providing them with a margin of safety. They become indirect land speculators with the borrower.
Ample liquidity encourages borrowing greater and greater amounts. Yet, the price rises in the underlying collateral eventually reach a point when new supply comes on the market, and demand is reduced.
How will resets translate into defaults?
Higher mortgage costs => less consumer discretionary spending
Less consumer spending => slower economic growth
slower economic growth => flat or lower wages (which are already subpar) AND
Slower growth => greater unemployment (3 and 4 reinforce each other for a time)
greater defaults => decreased volume of home mortgages
lower volume of home loans => fewer sales of homes
lower volume of home loans => lower home prices GO BACK TO 7.
The idea that bubble proponents could have data on this issue is silly. There hasn't been any historical precedent for such lax lending in modern times.
Even during the 20's when interest only loans were last popular (aka. balloon mortgages), property was only 2-3 times annual wages. Now it is what--6-8 times annual earnings?
The key is timing. Who knows what the lead or lag would be for these factors to take effect? I'd venture to say we starting to slide down the slo
Mortgage applications should be plotted against existing home sales and new home "completions" (not sales), suitably lagged. The home building industry has moved from build-to-inventory to build-to-order model. As such, most new home sales are not completed units, rather units that have not been even started. (Thirty years back most homes sold were actually completed.) So, the homebuyer today actually makes a mortgage application only when the builder tells him that his/her home is ready or will be ready shortly. Completions have actually shot up recently. They are up 22% yoy. Builders are rushing to complete projects for which they have orders and close the sales before the market tanks enough to change buyers' mind. More evidence for this hypothesis is the fact that current sales of completed units have actually declined yoy.
Watch for the purchase application in the next few months.
About that paper by Cagan. First off, he has done a terrific job. I went through it a couple of months nback when it first came out and wrote a short note for our clients. Cagan does a thorough job but drawas the wrong conclusions. Look at the table in page 32. The $110 billion losses is for first mortgages alone. He does not take into account the fact that second mortgages (piggy back etc) will be wiped out. Some forty percent of those mortgages had a second mortgage associated with it at origination itself! Those losses will be another few tens of billions They will be entirely wiped out) in a foreclosure. Second, Cagan's analysis on page 32 is only for ARMs. Sure, fixed-rate borrowers dont have payment shock, but that does not mean they wont default. The current low foreclosure rate and even lower mortgage losses is thanks to the strong housing market. When homes it on the market, both will go up. In short, we could be seeing $200 billion in losses, maybe more (Cagan, nor I, have taken into account loans made in 2003 and earlier, a substantial proportion of whcih have minimal or negative equity). Total home mortgage outstanding is about 9 trillion. So we are talking about a loss of about 2% on the entire portfolio. 2% increase in charge-off over normal levels would make mortgages junk bonds. Last time I checked mortgages were not priced like junk bonds.
A static analysis of losses usually will ignore all kinds of positive feedback effects and underestimate the final loss. That is what happened in Japan. I have been a bank lending officer and believe me things can go much worse than the worse case scenario analysis done ex ante.
The "people are not fools" argument is just a debating point. If todays lenders are not fools then yesterday's most certianly were--they left profit opportunities go abegging. Time will tell who were fools and who were not.
I was referring specifically to the foreclosure relevance statement that DoctorWho made. Thus I quoted it specifically, to avoid any confusion. [Only partially successful it appears.]
And that DocWho statement, was:
The only relevent issue [for DocWho anyway] is how much of that [the ~$1T reset] will come back in foreclosures, and that question is never, ever answered supportably by bubbleheads. [You know that's us.]
So I thought your remark:
Dr Who makes a good point about backing up arguments and claims (even with the condescending tone.) However, the claim quoted above is simply incorrect.
backed up with this argument:
One of the risks of ARMs is affordability, and in a rising interest rate environment some ARM'd homeowners may be forced to sell their properties. This will add to inventory numbers and can lead to downward pressure in pricing, regardless of foreclosure rates. [ ie being forced to sell is not identical with being forced into foreclosure ] Sometimes it's not enough to have research to support your claims - sometimes you need to have a claim worth supporting.
[Well, if you want to be a fussbudgit (even at my discomfort of appearing to support tone-deaf DocWho, I am easily roped into a discussion of articulating what is an incorrect claim and what is not.
was a refutation (" simply incorrect") of DocWho's assessment of the article, an article that contains many scenarios of price pressure/shifts and is largely concerned to minimize the "horror" of the ~$1T reset of mortgages in the coming year. Partially successful, I thought.
Hence I thought your statement was inaccurate, (ok, your style forced me to "simply inaccurate") and that DocWho's claim was a major component of the Cagan article/thesis, a claim that lacked a little precision --perhaps in an effort to provoke comment.
Market uptrends are usually characterized by higher highs and higher lows.
When the 4 week moving average takes out the March 2005 low, this would be added evidence of a trend reversal.
Those Fannie reports are good stuff.
They mis-titled that graph as being "1990 to Present" which it obviously is not.
It is, though, here:
http://www.fanniemae.com/media/pdf/berson/weekly/2006/011706.pdf
It will be interesting to see if reset foreclosures track anywhere near what these people concluded:
http://www.firstamres.com/pdf/MPR_White_Paper_FINAL.pdf
Thank you for that excellent link Doctorwho, (regardless of their fairly optimistic outlook).
No problem. They are optimistic, aren't they?
As big a bogeyman-in-the-closet as ARM resets are claimed to be by doom-scenario believers, this is the ONLY research I have found that examines the magnitude of probable foreclosures.
It doesn't take a genius to quote how many billions or trillions of mortgage dollars will reset in '06 or '07. Big deal. The only relevent issue is how much of that will come back in foreclosures, and that question is never, ever answered supportably by bubbleheads.
I realize many here operate on the assumption that most people are essentially blundering idiots, but lending institutions have become very good at forecasting default rates, and the conclusions in that report tend to support that fact.
If anyone knows of any other bonafide research into ARM reset defaults, let's see it.
You make good points Dr. Who. I would only add that lending institutions have become very good at forecasting default rates during a prolonged period of falling interest rates, low inflation, rising property values and high employment. If current interest and inflation conditions persist, we should expect many of their estimates to be proven wrong with time.
Dr Who said: "The only (sic) relevent issue is how much of that will come back in foreclosures"
Dr Who makes a good point about backing up arguments and claims (even with the condescending tone.) However, the claim quoted above is simply incorrect. One of the risks of ARMs is affordability, and in a rising interest rate environment some ARM'd homeowners may be forced to sell their properties. This will add to inventory numbers and can lead to downward pressure in pricing, regardless of foreclosure rates. Sometimes it's not enough to have research to support your claims - sometimes you need to have a claim worth supporting.
JS's remark
Dr Who makes a good point about backing up arguments and claims (even with the condescending tone.) However, the claim quoted above is simply incorrect.
is simply inaccurate. (Couldn't resist. Maybe dead-on for tone, though.)[The devil has my keyboard, he will pass.]
The link provides a mountain (for me) of data and hypotheticals that cover many other scenarios/possibilities than foreclosure rates, yes? It is readable (even for me!) and persuasive --covering 60%(!!) of all mortgages in this country.
A mighty find. Thanks again to DoctorWho.
calmo,
Hmmm Im not clear on your point. I think perhaps you misread what I wrote. Alas, I wasnt talking about the Cagan research in my comment. I was referring specifically to the foreclosure relevance statement that DoctorWho made. Thus I quoted it specifically, to avoid any confusion. I assume that since you believe my comment was inaccurate that you may really believe that foreclosures are the only relevant issue. If that is the case, Id like to hear why you believe that.
"I realize many here operate on the assumption that most people are essentially blundering idiots, but lending institutions have become very good at forecasting default rates, and the conclusions in that report tend to support that fact."
With the advent of the MBS market, no one is all that worried about loan quality, believing that packaging all of these risky loans into one security minimizes the risk to a tolerable, predictible level through diversification.
Those risk averse lenders are more likely to lend if they believe they can sell the loan to someone else. They have no incentive to check the credit expansion, as they do not suffer if any individual loan fails.
With rising property prices, lenders aren't concerned with loan quality, because the collateral on the loan went up in value.
They wouldn't worry about real losses if they think they could sell the property and recover their money.
How else could people in So Cal get loans greater than 500,000 when they only make 50-60,000?
On the way up, this makes everyone look like geniuses--borrowers, brokers, and lenders, and reinforces itself. Liquidity begets more liquidity, and reduces defaults--for a time.
The problem--systems with positive feedback become inherently unstable.
In a market, that is the point when "everyone" knows something is true (ie. "Real Estate always goes up" or "We can always refinance or sell").
Everyone in the business is banking on increasing property prices to bail them out--not only borrowers, but lenders who made a risky loan in the belief that property prices would rise, providing them with a margin of safety. They become indirect land speculators with the borrower.
Ample liquidity encourages borrowing greater and greater amounts. Yet, the price rises in the underlying collateral eventually reach a point when new supply comes on the market, and demand is reduced.
How will resets translate into defaults?
lower volume of home loans => lower home prices GO BACK TO 7.
The idea that bubble proponents could have data on this issue is silly. There hasn't been any historical precedent for such lax lending in modern times.
Even during the 20's when interest only loans were last popular (aka. balloon mortgages), property was only 2-3 times annual wages. Now it is what--6-8 times annual earnings?
The key is timing. Who knows what the lead or lag would be for these factors to take effect? I'd venture to say we starting to slide down the slo
Rob come back! Did you slide down the slo.....pe?
Thank you-great post
CR,
Mortgage applications should be plotted against existing home sales and new home "completions" (not sales), suitably lagged. The home building industry has moved from build-to-inventory to build-to-order model. As such, most new home sales are not completed units, rather units that have not been even started. (Thirty years back most homes sold were actually completed.) So, the homebuyer today actually makes a mortgage application only when the builder tells him that his/her home is ready or will be ready shortly. Completions have actually shot up recently. They are up 22% yoy. Builders are rushing to complete projects for which they have orders and close the sales before the market tanks enough to change buyers' mind. More evidence for this hypothesis is the fact that current sales of completed units have actually declined yoy.
Watch for the purchase application in the next few months.
About that paper by Cagan. First off, he has done a terrific job. I went through it a couple of months nback when it first came out and wrote a short note for our clients. Cagan does a thorough job but drawas the wrong conclusions. Look at the table in page 32. The $110 billion losses is for first mortgages alone. He does not take into account the fact that second mortgages (piggy back etc) will be wiped out. Some forty percent of those mortgages had a second mortgage associated with it at origination itself! Those losses will be another few tens of billions They will be entirely wiped out) in a foreclosure. Second, Cagan's analysis on page 32 is only for ARMs. Sure, fixed-rate borrowers dont have payment shock, but that does not mean they wont default. The current low foreclosure rate and even lower mortgage losses is thanks to the strong housing market. When homes it on the market, both will go up. In short, we could be seeing $200 billion in losses, maybe more (Cagan, nor I, have taken into account loans made in 2003 and earlier, a substantial proportion of whcih have minimal or negative equity). Total home mortgage outstanding is about 9 trillion. So we are talking about a loss of about 2% on the entire portfolio. 2% increase in charge-off over normal levels would make mortgages junk bonds. Last time I checked mortgages were not priced like junk bonds.
A static analysis of losses usually will ignore all kinds of positive feedback effects and underestimate the final loss. That is what happened in Japan. I have been a bank lending officer and believe me things can go much worse than the worse case scenario analysis done ex ante.
The "people are not fools" argument is just a debating point. If todays lenders are not fools then yesterday's most certianly were--they left profit opportunities go abegging. Time will tell who were fools and who were not.
Back to JS for a moment:
I was referring specifically to the foreclosure relevance statement that DoctorWho made. Thus I quoted it specifically, to avoid any confusion. [Only partially successful it appears.]
And that DocWho statement, was:
The only relevent issue [for DocWho anyway] is how much of that [the ~$1T reset] will come back in foreclosures, and that question is never, ever answered supportably by bubbleheads. [You know that's us.]
So I thought your remark:
Dr Who makes a good point about backing up arguments and claims (even with the condescending tone.) However, the claim quoted above is simply incorrect.
backed up with this argument:
One of the risks of ARMs is affordability, and in a rising interest rate environment some ARM'd homeowners may be forced to sell their properties. This will add to inventory numbers and can lead to downward pressure in pricing, regardless of foreclosure rates. [ ie being forced to sell is not identical with being forced into foreclosure ] Sometimes it's not enough to have research to support your claims - sometimes you need to have a claim worth supporting.
[Well, if you want to be a fussbudgit (even at my discomfort of appearing to support tone-deaf DocWho, I am easily roped into a discussion of articulating what is an incorrect claim and what is not.
was a refutation (" simply incorrect") of DocWho's assessment of the article, an article that contains many scenarios of price pressure/shifts and is largely concerned to minimize the "horror" of the ~$1T reset of mortgages in the coming year. Partially successful, I thought.
Hence I thought your statement was inaccurate, (ok, your style forced me to "simply inaccurate") and that DocWho's claim was a major component of the Cagan article/thesis, a claim that lacked a little precision --perhaps in an effort to provoke comment.
"perhaps in an effort to provoke comment"
Wot, me?