by Calculated Risk on 7/24/2007 05:03:00 PM

Later this week I'll post an update on my 2007 housing forecast. I've been waiting for the foreclosure data for Q2, and the existing and new home sales data for June, scheduled to be released tomorrow and Thursday respectively.

On Friday I pulled out the old Supply and Demand drawings to compare the housing market to an efficient market. In this post I'd like to discuss two recent shifts in demand for housing, and how I expect them to unwind.

First, here are the new and existing home sales, since 1969, normalized by the number of owner occupied units (OOU).

Click on graph for larger image.

For the recent housing boom (in sales, not price), I marked three periods on the graph. There may be some disagreement on the dates and the causes of the boom for each period, but a simply explanation is:

Period 1: This was mostly due to fundamentals of real wage growth, employment growth and demographics.

Period 2: This was primarily due to an interest rate shock (lower rates) that moved renters to home ownership.

Period 3: This was primarily due to speculation, especially home buyers using excessive leverage for speculation.

NOTE: The following models of demand shifts assume an efficient market and no shifts in supply.

Period 2: Interest Rate Shock

This diagram depicts how I'd expect an interest rate shock to impact housing demand. After interest rates decrease sharply, there would be a temporary increase in demand - perhaps for a couple of years - as renters migrate to home ownership.

According to the Census Bureau, the number of American households renting decreased by 1.4 million from 2001 to early 2004. These households probably migrated to home ownership because the "rent or buy" decision favored buying due to lower interest rates.

This increase in demand was temporary, and according to the Census Bureau, the migration from renting to buying ended by early 2004.

Looking at a Supply and Demand diagram, the interest rate shock temporarily shifted demand from D0 to D1.

This moved the quantity demanded from Q0 to Q1, and the price from P0 to P1.

When the demand shifts back (above model of temporary demand shift), the quantity demanded falls back to Q0 - but housing suffers from sticky prices, so price only declines slowly to P0.

However, we can look at the graph of actual sales (first graph), and we can see that sales didn't decline in 2004 and 2005; instead sales increased.

Period 3: Excessive Leverage as Speculation

Speculation frequently chases appreciation, and the earlier price increases, based at least somewhat on fundamentals and an interest rate shock, probably spurred many buyers to only considered their monthly costs when buying a home (during period 3). Many of these buyers used excessive leverage, speculating that the price would continue to increase into the future.

This type of leveraged activity pulls demand from future periods as shown in this diagram. The rampant speculation (with innovative mortgage products) pushed demand from D1 to D2, with associated increases in price and the quantity demanded. However, when the speculation ends, demand will eventually fall back to D3; below the level of demand (D1) when the speculation started.

These models are just a guide, and are intended for efficient markets. But this suggests to me that sales, especially of existing homes, will eventually decline to below the levels of 1998 to 2001.

Flippers and Supply Shifts

Some people may be thinking about the impact of investors (or flippers) on the demand curve. Note: This type of speculation was probably only rampant in the coastal regions. Instead of viewing investor activity as a demand shift, it might be better way to view this type of speculation is as storage - or a supply shift; when the investor buys, they remove the asset from the supply. This means that investor speculation shifted the supply curve (not shown) to the left during the period of speculative activity. When the speculator sells, the supply curve shifts to the right, as the stored units come back on the market. So the news is bad for housing: not only is the demand curve shifting left, but the supply curve is shifting right (especially in some coastal regions).

Credit: template for diagram was from Wikipedia.