[gdcox]

The analysis uses the average ratio of median income to median prices between the mid-70s and 2001 (2.8 ‘income multiplier’) to calculate the size of the current bubble.

Whilst I am no apologist for or denigrator of the theory of the bubble (it exists, is big and ugly and I would hang all brokers who sold 100% plus sub-prime loan to value mortgages in recent years!), there are in my view two notes of caution that needs to be applied regarding this bubble size work.

1) The period chosen (mid 70s to 2001) includes a first half when mortgage interest rates were abnormally high (for a variety of reasons). The abnormally high mortgage rates then (in nominal and real terms) had the effect of constraining the income multiplier because mortgage demand was unnaturally limited and hence prices were at a lower level . So I think the ‘normal’ multiplier is larger than 2.8 and hence the scale of the bubble today is smaller than the authors say.
2) As a financial economist by background, I think the methodology of using static data from the past is suspect; especially in isolation. Much better to make a current dynamic comparison with the rental markets. As house prices fall this and possibly next year, there will come a point at which it is cheaper for people to buy on a mortgage than to rent and another point at which the rental yield on repossessed properties becomes irresistibly attractive for landlords. These two points may coincide in time of not. But either way, the bubble will have been eliminated by definition when landlord and renters are incentivized and act to buy en masse. I suspect that point is well above the level suggested in the article because rents have probably risen a more than have median incomes: eg I don’t think California will go down 60% (recent level to bottom as the article suggests) even if as seems likely foreclosure sales keep on rising.

I sadly do not have time to do the analysis suggested above, but I suspect that the factors above mean that the bottom is above the non-bubble average national, state or city price level stated in the article; though I am conscious that in some micro areas there will be a huge undershoot of prices simply because of the shear scale of new subdivisions put into the market by new build well away from centres of employment (eg DR Horton’s 50% fire sale) .

Have I missed anything?

Graham Cox

[gdcox]

The analysis uses the average ratio of median income to median prices between the mid-70s and 2001 (2.8 ‘income multiplier’) to calculate the size of the current bubble.

Whilst I am no apologist for or denigrator of the theory of the bubble (it exists, is big and ugly and I would hang all brokers who sold 100% plus sub-prime loan to value mortgages in recent years!), there are in my view two notes of caution that needs to be applied regarding this bubble size work.

1) The period chosen (mid 70s to 2001) includes a first half when mortgage interest rates were abnormally high (for a variety of reasons). The abnormally high mortgage rates then (in nominal and real terms) had the effect of constraining the income multiplier because mortgage demand was unnaturally limited and hence prices were at a lower level . So I think the ‘normal’ multiplier is larger than 2.8 and hence the scale of the bubble today is smaller than the authors say.
2) As a financial economist by background, I think the methodology of using static data from the past is suspect; especially in isolation. Much better to make a current dynamic comparison with the rental markets. As house prices fall this and possibly next year, there will come a point at which it is cheaper for people to buy on a mortgage than to rent and another point at which the rental yield on repossessed properties becomes irresistibly attractive for landlords. These two points may coincide in time of not. But either way, the bubble will have been eliminated by definition when landlord and renters are incentivized and act to buy en masse. I suspect that point is well above the level suggested in the article because rents have probably risen a more than have median incomes: eg I don’t think California will go down 60% (recent level to bottom as the article suggests) even if as seems likely foreclosure sales keep on rising.

I sadly do not have time to do the analysis suggested above, but I suspect that the factors above mean that the bottom is above the non-bubble average national, state or city price level stated in the article; though I am conscious that in some micro areas there will be a huge undershoot of prices simply because of the shear scale of new subdivisions put into the market by new build well away from centres of employment (eg DR Horton’s 50% fire sale) .

Have I missed anything?

Graham Cox

[gdcox]

The analysis uses the average ratio of median income to median prices between the mid-70s and 2001 (2.8 ‘income multiplier’) to calculate the size of the current bubble.

Whilst I am no apologist for or denigrator of the theory of the bubble (it exists, is big and ugly and I would hang all brokers who sold 100% plus sub-prime loan to value mortgages in recent years!), there are in my view two notes of caution that needs to be applied regarding this bubble size work.

1) The period chosen (mid 70s to 2001) includes a first half when mortgage interest rates were abnormally high (for a variety of reasons). The abnormally high mortgage rates then (in nominal and real terms) had the effect of constraining the income multiplier because mortgage demand was unnaturally limited and hence prices were at a lower level . So I think the ‘normal’ multiplier is larger than 2.8 and hence the scale of the bubble today is smaller than the authors say.
2) As a financial economist by background, I think the methodology of using static data from the past is suspect; especially in isolation. Much better to make a current dynamic comparison with the rental markets. As house prices fall this and possibly next year, there will come a point at which it is cheaper for people to buy on a mortgage than to rent and another point at which the rental yield on repossessed properties becomes irresistibly attractive for landlords. These two points may coincide in time of not. But either way, the bubble will have been eliminated by definition when landlord and renters are incentivized and act to buy en masse. I suspect that point is well above the level suggested in the article because rents have probably risen a more than have median incomes: eg I don’t think California will go down 60% (recent level to bottom as the article suggests) even if as seems likely foreclosure sales keep on rising.

I sadly do not have time to do the analysis suggested above, but I suspect that the factors above mean that the bottom is above the non-bubble average national, state or city price level stated in the article; though I am conscious that in some micro areas there will be a huge undershoot of prices simply because of the shear scale of new subdivisions put into the market by new build well away from centres of employment (eg DR Horton’s 50% fire sale) .

Have I missed anything?

Graham Cox

[gdcox]

The analysis uses the average ratio of median income to median prices between the mid-70s and 2001 (2.8 ‘income multiplier’) to calculate the size of the current bubble.

Whilst I am no apologist for or denigrator of the theory of the bubble (it exists, is big and ugly and I would hang all brokers who sold 100% plus sub-prime loan to value mortgages in recent years!), there are in my view two notes of caution that needs to be applied regarding this bubble size work.

1) The period chosen (mid 70s to 2001) includes a first half when mortgage interest rates were abnormally high (for a variety of reasons). The abnormally high mortgage rates then (in nominal and real terms) had the effect of constraining the income multiplier because mortgage demand was unnaturally limited and hence prices were at a lower level . So I think the ‘normal’ multiplier is larger than 2.8 and hence the scale of the bubble today is smaller than the authors say.
2) As a financial economist by background, I think the methodology of using static data from the past is suspect; especially in isolation. Much better to make a current dynamic comparison with the rental markets. As house prices fall this and possibly next year, there will come a point at which it is cheaper for people to buy on a mortgage than to rent and another point at which the rental yield on repossessed properties becomes irresistibly attractive for landlords. These two points may coincide in time of not. But either way, the bubble will have been eliminated by definition when landlord and renters are incentivized and act to buy en masse. I suspect that point is well above the level suggested in the article because rents have probably risen a more than have median incomes: eg I don’t think California will go down 60% (recent level to bottom as the article suggests) even if as seems likely foreclosure sales keep on rising.

I sadly do not have time to do the analysis suggested above, but I suspect that the factors above mean that the bottom is above the non-bubble average national, state or city price level stated in the article; though I am conscious that in some micro areas there will be a huge undershoot of prices simply because of the shear scale of new subdivisions put into the market by new build well away from centres of employment (eg DR Horton’s 50% fire sale) .

Have I missed anything?

Graham Cox

[gdcox]

The analysis uses the average ratio of median income to median prices between the mid-70s and 2001 (2.8 ‘income multiplier’) to calculate the size of the current bubble.

Whilst I am no apologist for or denigrator of the theory of the bubble (it exists, is big and ugly and I would hang all brokers who sold 100% plus sub-prime loan to value mortgages in recent years!), there are in my view two notes of caution that needs to be applied regarding this bubble size work.

1) The period chosen (mid 70s to 2001) includes a first half when mortgage interest rates were abnormally high (for a variety of reasons). The abnormally high mortgage rates then (in nominal and real terms) had the effect of constraining the income multiplier because mortgage demand was unnaturally limited and hence prices were at a lower level . So I think the ‘normal’ multiplier is larger than 2.8 and hence the scale of the bubble today is smaller than the authors say.
2) As a financial economist by background, I think the methodology of using static data from the past is suspect; especially in isolation. Much better to make a current dynamic comparison with the rental markets. As house prices fall this and possibly next year, there will come a point at which it is cheaper for people to buy on a mortgage than to rent and another point at which the rental yield on repossessed properties becomes irresistibly attractive for landlords. These two points may coincide in time of not. But either way, the bubble will have been eliminated by definition when landlord and renters are incentivized and act to buy en masse. I suspect that point is well above the level suggested in the article because rents have probably risen a more than have median incomes: eg I don’t think California will go down 60% (recent level to bottom as the article suggests) even if as seems likely foreclosure sales keep on rising.

I sadly do not have time to do the analysis suggested above, but I suspect that the factors above mean that the bottom is above the non-bubble average national, state or city price level stated in the article; though I am conscious that in some micro areas there will be a huge undershoot of prices simply because of the shear scale of new subdivisions put into the market by new build well away from centres of employment (eg DR Horton’s 50% fire sale) .

Have I missed anything?

Graham Cox

[gdcox]

Rich,

So the MLS data only need to be corrected by adding REOs though auctions?? (still learning the terminology!)

Graham

[gdcox]

Rich,

So the MLS data only need to be corrected by adding REOs though auctions?? (still learning the terminology!)

Graham

[gdcox]

Rich,

So the MLS data only need to be corrected by adding REOs though auctions?? (still learning the terminology!)

Graham

[gdcox]

Rich,

So the MLS data only need to be corrected by adding REOs though auctions?? (still learning the terminology!)

Graham

[gdcox]

Rich,

So the MLS data only need to be corrected by adding REOs though auctions?? (still learning the terminology!)

Graham

[gdcox]

Graham

Nice post.

Pity the securitization was not all made up of this stuff.

[gdcox]

Graham

Nice post.

Pity the securitization was not all made up of this stuff.

[gdcox]

Graham

Nice post.

Pity the securitization was not all made up of this stuff.

[gdcox]

Graham

Nice post.

Pity the securitization was not all made up of this stuff.

[gdcox]

Graham

Nice post.

Pity the securitization was not all made up of this stuff.

[Author]

Posts