See "A Model of Mortgage Default" by Campbell and Cocco, November 2010,
http://kuznets.fas.harvard.edu/~campbell/papers/mortdefault24112010.pdf
for a discussion of mortgage defaults and IOs [Interest Only Mortgages] with a different perspective.
IOs are used by borrowing constrained borrowers because of lower required monthly payments. IOs do not amortize principal and have higher remaining balances later in their life than amortizing mortgages.
IOs have a lower default rate in their early years (lower required payments), but higher default rates in later years when there is negative equity because they have a higher remaining balance.
In a bubble, there are many speculators (whether they cause or follow bubbles is an open debate). Speculators try to use as much leverage as they can to buy assets. Speculators would tend toward IOs because they can borrow more. In currency and commodity markets, they would use futures and options.
As the economy tanked and home prices started to decline, IO users were the most borrowing constrained, which means the most income constrained, and they had the most negative equity because of the lack of amortization with higher remaining balances. Their borrowing constraint with high negative equity caused higher defaults. Plus, if they own multiple homes, as likely for speculators, a negative income shock to a single borrower will cause defaults on several homes and IOs. There is a multiplier default effect.
IO use was coincidental to the collapse of the housing market and defaults, not the cause of the bubble or subsequent collapse.
In speculative markets, speculators look for means to use as little of their own equity and as much leverage as they can. IOs were used to borrow more to purchase more homes with more leverage, but if they were not available, speculators would look for other ways to maximize their use of equity and borrow as much as possible.
The use of IOs just shows there were speculators in the housing market. It does not show causation for the speculation or the defaults. If you eliminate or restrict IOs, speculators will find some other means to increase their ability to speculate in a hot housing market.
Past IO use in a housing market showed there were speculators in that market. Eliminate IOs and you will not stop speculation in a hot housing market. Does speculation cause bubbles (Did speculation cause house prices to rise and then drop)? Valid arguments exist for yes and no.
Showing posts with label U.S. Housing Market. Show all posts
Showing posts with label U.S. Housing Market. Show all posts
Friday, December 10, 2010
IO Mortgages And The Housing Bubble
A comment I posted on "The means to speculate" on The Economist Free Exchange blog:
Sunday, November 7, 2010
How Much Extra Is A Homeowner Willing To Pay For Future Home Price Appreciation?
Since homeowners in normal economic environments expect house prices to appreciate, how much extra are buyers willing to pay for that future appreciation?
With reasonable assumptions, in a heated housing market, homeowners are willing to pay about 23 percent more for a home to benefit from future price increases.
Option pricing theory allows us to make a reasonable guess about the amount in the purchase price that is attributable to future price appreciation.
For most of the period from World War II until about 2006, homebuyers expected to sell their houses for more than the original purchase price.
Almost everything else we buy and use is either thrown away, given away or sold for less than the purchase price. Homes are an exception to a general rule for things we use.
According to the National Association of Realtors, http://www.realtor.org/press_room/news_releases/2010/11/survey, the average length of homeownership is eight years.
The value of the expected price appreciation of a house can be modeled by looking at the cost of buying an eight-year call option on a home. The owner of the call option will benefit from any appreciation in the home's price without having to purchase the home. Any purchaser of a home also benefits from price appreciation. The purchase price of a home therefore includes the cost of an embedded call option.
One can use the well-known Black-Scholes option pricing model to compute a price accurate enough for the purposes of this post (it is not the only mathematical way to compute the price of an option but it is simple, relatively easy to use, and many free online Black-Scholes option pricing calculators are available).
Using 20 percent as a reasonable amount for the standard deviation input of the price change in the Black-Scholes model in a heated housing market, the value computed for an eight-year call option of a house is about 23 percent of the fundamental home price.
A homebuyer would have to pay about $369,000 for a home that is fundamentally worth $300,000 because of the extra benefit of price appreciation conferred on homeownership.
The negative side of paying for expected appreciation is that when there is no market expectation of home price appreciation, the price will drop by the amount of the embedded option for future price increases. A call option with no expectation of future price increases will approach zero value.
A home that is worth $369,000 in a heated housing market, such as leading up to the recent financial crisis, will drop back to close to the $300,000 fundamental value for almost a 20 percent decline.
In using option pricing, I am assuming that any benefit to the homeowner from the elimination of the need to pay a landlord any rent payment is completely offset by taxes and maintenance costs.
Residential home price to rent rations are typically in the range of 15 to 20, which makes the yearly rent about 5 to 6 percent of the price of a home. The elimination of rent represents the equivalent of a dividend to a homeowner.
Homeowners also have costs that offset the elimination of rent benefit.
I assumed home structures represent about 75 percent of the purchase price with land the other 25 percent and that home structures need repair, maintenance and replacement and have an average structure life of about 20-25 years. Over 20-25 years, these depreciation costs are about 4-5 percent (1/25-1/20) of the structural costs or about 3-4 percent of the total purchase price (structure price is ¾ of the total price). Real estate taxes and other non-rental costs such as heat, water, etc. are easily around the 2 percent range. The total cost of home ownership is about equal to the rental benefit of 5-6 percent. It is a reasonable assumption without precise data to view costs as completely offsetting benefits.
For those familiar with options, I used the 8-year future value (using the T-note rate) of the current price as the exercise or strike price since the call option owner has the benefit of investing the full fundamental home price in Treasuries until expiration of the call option.
House buyers increase the amount they are willing to pay for a home due to the value of future price appreciation. Option pricing is one way to model the extra cost to a purchaser of buying a home when there is a market expectation of future price appreciation. The computed option value also represents that amount of price decline when the market stops expecting house prices to increase.
Also, see my February 23, 2009 post, "Determinates Of Home Prices."
With reasonable assumptions, in a heated housing market, homeowners are willing to pay about 23 percent more for a home to benefit from future price increases.
Option pricing theory allows us to make a reasonable guess about the amount in the purchase price that is attributable to future price appreciation.
For most of the period from World War II until about 2006, homebuyers expected to sell their houses for more than the original purchase price.
Almost everything else we buy and use is either thrown away, given away or sold for less than the purchase price. Homes are an exception to a general rule for things we use.
According to the National Association of Realtors, http://www.realtor.org/press_room/news_releases/2010/11/survey, the average length of homeownership is eight years.
The value of the expected price appreciation of a house can be modeled by looking at the cost of buying an eight-year call option on a home. The owner of the call option will benefit from any appreciation in the home's price without having to purchase the home. Any purchaser of a home also benefits from price appreciation. The purchase price of a home therefore includes the cost of an embedded call option.
One can use the well-known Black-Scholes option pricing model to compute a price accurate enough for the purposes of this post (it is not the only mathematical way to compute the price of an option but it is simple, relatively easy to use, and many free online Black-Scholes option pricing calculators are available).
Using 20 percent as a reasonable amount for the standard deviation input of the price change in the Black-Scholes model in a heated housing market, the value computed for an eight-year call option of a house is about 23 percent of the fundamental home price.
A homebuyer would have to pay about $369,000 for a home that is fundamentally worth $300,000 because of the extra benefit of price appreciation conferred on homeownership.
The negative side of paying for expected appreciation is that when there is no market expectation of home price appreciation, the price will drop by the amount of the embedded option for future price increases. A call option with no expectation of future price increases will approach zero value.
A home that is worth $369,000 in a heated housing market, such as leading up to the recent financial crisis, will drop back to close to the $300,000 fundamental value for almost a 20 percent decline.
In using option pricing, I am assuming that any benefit to the homeowner from the elimination of the need to pay a landlord any rent payment is completely offset by taxes and maintenance costs.
Residential home price to rent rations are typically in the range of 15 to 20, which makes the yearly rent about 5 to 6 percent of the price of a home. The elimination of rent represents the equivalent of a dividend to a homeowner.
Homeowners also have costs that offset the elimination of rent benefit.
I assumed home structures represent about 75 percent of the purchase price with land the other 25 percent and that home structures need repair, maintenance and replacement and have an average structure life of about 20-25 years. Over 20-25 years, these depreciation costs are about 4-5 percent (1/25-1/20) of the structural costs or about 3-4 percent of the total purchase price (structure price is ¾ of the total price). Real estate taxes and other non-rental costs such as heat, water, etc. are easily around the 2 percent range. The total cost of home ownership is about equal to the rental benefit of 5-6 percent. It is a reasonable assumption without precise data to view costs as completely offsetting benefits.
For those familiar with options, I used the 8-year future value (using the T-note rate) of the current price as the exercise or strike price since the call option owner has the benefit of investing the full fundamental home price in Treasuries until expiration of the call option.
House buyers increase the amount they are willing to pay for a home due to the value of future price appreciation. Option pricing is one way to model the extra cost to a purchaser of buying a home when there is a market expectation of future price appreciation. The computed option value also represents that amount of price decline when the market stops expecting house prices to increase.
Also, see my February 23, 2009 post, "Determinates Of Home Prices."
Monday, October 18, 2010
Is The Foreclosure Rate Unrelated To The Housing Bubble And Close To Long Term Trend
Part of a comment I posted on "The Breakdown of the U.S. Mortgage Market" on Rortybomb:
Chart from Rortybomb.
Additionally and interestingly, the Mortgage Bankers Association chart shows an upward trend in mortgage foreclosures from 1979 to 2002. Defaults appear to be about 5 times greater in 2002 than in 1979.
The foreclosure drop from 2002 to 2006 can be attributed to the housing bubble price appreciation, which allowed homeowners to avoid foreclosure by selling their homes at a price that paid off the mortgage and other second lien debt.
If the home bubble had not occurred we might have seen a foreclosure rate of about 2 to 2.5 percent now, based on an eyeball estimate of the upward slope of the chart.
The high rate of current foreclosures may only be slightly higher than the expected trend line and the bubble allowed avoidance of trend line foreclosures from 2002 to 2006.
The current foreclosure rate may seem high because the housing bubble artificially allowed the foreclosure rate to deviate from the trend line and suppress it, but it may be close to trend now.
The economic and social factors that raised the foreclosure rate from 1979 to 2002 probably are an important force in today’s foreclosures and the collapse of the housing bubble and the higher than norm unemployment maybe less of a cause than common consensus believes.
Subscribe to:
Posts (Atom)