US Median House Prices Fall

[KrazyHorse]

F.e., the difference in monthly carrying costs on a 30-year loan versus an interest-only loan isn't as large as you might think

About 18% at an APR of 6%

[DanS]

About right. I have it as a 16.6% reduction.

Now consider that the reduction in interest rates from a high of 8.52% (May 2000) to a low of 5.23% (June 2003) decreases the monthly carrying cost by 28.5% on a 30-year loan. Given that interest rates apply to all mortgages, we can surmise that by far the bigger impact on prices is the interest rates.

The average mortgage rate since 1970 has been 9.30%. The 5.23% rate for June 2003 was the lowest recorded rate since 1970.

[Adam Smith]

Originally posted by KrazyHorse
As I mentioned earlier, the fact that people are (were) driving housing increases by the growth of longer-term and creative financing options implies that there was more going on than simply long-term adjustment to low rates.

If prices were simply climbing slowly to their equilibrium at the new, low interest rates then there would be no need for people to finance housing on longer amortization terms (or 0 amortization or negative amortization). Therefore we may conclude that housing prices continued to climb above their equilibrium level.

Why?

It has to do with capital gains from owning a house.

Housing is a capital good in that its services are consumed over time. The effective cost of owning a capital good in any year t is

v(t) = (r + d)P(t) – dP(t)/dt

where
v(t) is the effective cost of owning the capital good (what it costs you to own the asset today, considering all costs over the lifetime of the asset)
r = interest rate (opportunity cost of capital)
d = deterioration rate (keeping the house in same state of repair)
P(t) = price of asset (demographics, supply, demand and macro stuff is in here, but interest rate is not, in interest of simplicity)
dP(t)/dt = change in price of asset over time (appreciation or depreciation of asset price as determined by actual market conditions)
(Source: Nicholson, Microeconomic Theory, 9th ed, p. 512)

Most people forget about this last term, but its pretty important. If the house is appreciating in value, then you get more money back when you sell it. You will essentially be getting an extra return on your initial investment, which lowers the effective price of owning the house now. There are several important implications. On the demand side, the capital gain allows the asset price (or interest rate) to increase while still maintaining the same annual payment. On the supply side, it allows lenders to make more risky loans (e.g., interest only) since there is a greater chance that lenders will get their money back if asset prices are rising. The capital gain term also explains why housing is cheaper than it appears in booming areas such as California, and more expensive than it appears in depressed areas such as Detroit.

[KrazyHorse]

I understand the fact that it makes more sense to spend more in a market where housing prices will continue to appreciate. But the outlook for the long-term appreciation rate didn't change from 2000 to 2003 in most of the major markets. So that last term shouldn't have changed, and people shouldn't have been so much more willing to take on higher housing debt loads.

Sometimes there are very good reasons that last term changes. If a gold mine is discovered near a small town then the outlook for appreciation goes up, so housing prices go up. But San Francisco didn't suddenly change...

[KrazyHorse]

My thought is that it's quite simply a bit of a bubble. It's not based on good sense or any classical micro theory. Just people wanting to believe that a 10% increase one year means 10% increases every year from then on, world without end, amen.

[Colonâ„¢]

Originally posted by KrazyHorse


That is wrong, wrong, wrong.

The reason a decrease in interest rates adds value to a home is that, given the same monthly payment it is possible to pay off a larger debt in the same amount of time (since less of the payment goes to interest on the principle.

Which is a narrow perspective that doesn't consider the impact aggregate demand may have on house prices. Lower interest rates lowers the treshold for borrowing. This increases money supply, aggregate demand, and eventually the price level. The corollaly is an increase of money supply growth leading to a continuous increase in aggregate demand and an increase of the inflation rate. (The fact the inflation in question is taking place in an assets-class doesn't alter the dynamic).

Again, the implication of your stance is that all what interest changes do is temporarily affect the rate of price changes, after which they'd fall back to a rate set by another factor (which would be?). There wouldn't need to be any changes in Fed policy.

An accomdative monetary policy is not a policy that continuously lowers interest rates, it's a policy that keeps rates below their natural level, which is where inflation is kept at a stable rate. The lower the level of interest rates (and not the bigger the change) the more powerful the monetary stimulus.

Now, say that we have two different situations: one in which the interest rate on a 30 year fixed mortgage is at 8% for this year, then at 5% for the next 9 years, the other in which the interest rate is at 8% for the next 9 years and then goes to 5% during the 10th year. In both cases we look at the value of a given home now and in ten years time. From a purely micro perspective on interest rates (in other words assuming that income and population growth are equivalent in both cases).

At time t = 0, the market conditions are equivalent. It's the same home in the same market, so the price in both cases is X.

At time t = 10 years, the market conditions are again equivalent. There has been the same wage and population growth and interest rates are the same going forward from here. The house is again the same, so the price in both cases is Y. The fact that one market experienced low interest rates for a longer time doesn't mean ****. They both experienced the same price increase in the same amount of time, so the growth rate in both cases was the same. If I'm wrong then you have to tell me what it is in case 1 which causes a difference from case 2.

This means that interest rates provide a one time explanation for price increases. Continued low interest rates does not, by itself, explain continued high growth rates for housing prices.

Interest rates continuously set below equilibrum level (and as you know, the fed funds rate is not set by the market so this is possible), would lead to a continuous increase of money supply, which would indeed lead to a continuous increase of prices.

[Colonâ„¢]

Originally posted by KrazyHorse
My thought is that it's quite simply a bit of a bubble. It's not based on good sense or any classical micro theory. Just people wanting to believe that a 10% increase one year means 10% increases every year from then on, world without end, amen.

So you believe that inflation is an exogenous factor?

[Provost Harrison]

Originally posted by KrazyHorse
To first order, the price of a house can be described as

P(t) = f(I(t))*Q(t)

P(t) is the price
Q(t) is the hypothetical price in a given market at a set interest rate
I(t) is the interest rate
f(I) is a correction factor based on the interest rate. Low I = high f

Typical physicist. Reality is far more complicated...decreasing overall house prices (especially if proportionally) may cause panic making more people sell increasing supply...or a large increase in interest rate may increase the number of repossessions and thus increase the supply of properties on the market...and then there are external factors...it is impossible to treat this kind of scenario like this.

[Adam Smith]

Originally posted by KrazyHorse
My thought is that it's quite simply a bit of a bubble. It's not based on good sense or any classical micro theory. Just people wanting to believe that a 10% increase one year means 10% increases every year from then on, world without end, amen.

And when the market shows that those expectations are not realized, as it has in the last year or so, the last term changes.

[Kidlicious]

Originally posted by Colonâ„¢
Lower interest rates lowers the treshold for borrowing. This increases money supply, aggregate demand, and eventually the price level. The corollaly is an increase of money supply growth leading to a continuous increase in aggregate demand and an increase of the inflation rate. (The fact the inflation in question is taking place in an assets-class doesn't alter the dynamic).

Good discussion. This is what everyone is trying to tell you KH.

[KrazyHorse]

Interest rates continuously set below equilibrum level (and as you know, the fed funds rate is not set by the market so this is possible), would lead to a continuous increase of money supply, which would indeed lead to a continuous increase of prices.

i.e. inflation. And if a drop in the fed funds rate causes additional inflation of 1% per annum this somehow explains an additional 5% per annum increase in housing?

The simple fact is that people were spending more of their incomes on housing, as a percent (or equivalently taking on longer amortization periods). A combination of inflation and lower interest rates does not predict that.

[KrazyHorse]

Originally posted by Adam Smith
And when the market shows that those expectations are not realized, as it has in the last year or so, the last term changes.

That's all I'm saying...

I just don't understand why it happened so strongly in the US...

[Kidlicious]

The CPI measures prices for consumer goods not capital assets. A 1% increase in the CPI may very well have nothing to with a 5% increase in housing prices.

The fact that you are missing is that the money supply can increase quite rapidly when interest rates remain unchanged. That money has to go somewhere, and lately it's been going into purchases of houses.

[rah]

I was reading in the paper the other day that the percentage of their gross income paid for housing is dramatically increasing. There's a much higher percentage in the 30s range (wish I had a link)

So I did a quick rough calculation and noticed that for our family it's around 10%

How is it for the rest of you?

[Dauphin]

For most people that I know in my age group, its around 25%-30% of take home pay spent on rent. That includes people sharing. I don't know what size mortgages others have, but it would probably be a bit more than that.