Login
Select your application

Buy and Sell Signals in the Housing Market

FRIDAY, APRIL 08
A Crash Course in the Housing Market
Many people study the stock market by analyzing the patterns and direction of stock market prices to assess good times to buy and sell. This approach is usually referred to as technical analysis.

A benefit and a challenge with this approach is that there are so many people studying the same prices with the same general approach that their actions can be self-fulfilling for periods of time and then fall apart when enough of them lose confidence or take another tack. This may explain why stock prices can rise for a long time and then suddenly crash.

The market for single family homes is completely different because you cannot place bets on houses and rapidly change course in milliseconds like you can with the stock market. Even if most people agree that prices will rise or fall it is a long, cumbersome and expensive process to buy or sell houses. Therefore, a change in market conditions or a new market consensus cannot get into home prices rapidly. Instead direction changes develop over a period of months or even years.

For example, 6 weeks into the Covid pandemic, and a month after housing related stocks have swooned 30% or more, houses are still closing within 5% of their pre-pandemic price levels. There may be a consensus that home prices will decline, but it just cannot happen very fast. Buyers and sellers that are closing this week negotiated the price and signed contracts 6 weeks ago. While some buyers may back out of deals and lenders may refuse to lend, it is far less likely that a seller who holds a 5% or 10% deposit and expects to close for a contract price this week would agree to change the price on the eve of closing. We therefore may see fewer closings but we will not see home prices swoon 30% within a matter of weeks.

The tortoise-like nature of home price changes creates an interesting opportunity for investors. The slow pace lets investors watch as markets lumber along for months. Investors can hop on the ride up or down with a fairly high degree of confidence that current trend will continue for a while. There are no guarantees of course, but if you can measure the trend accurately you can reasonably rely on it continuing for far longer than a stock market trend. Stock markets change directions in milliseconds. Housing markets change directions every 5 or so years.

There are no free lunches in life, and I am not going to claim there is a free lunch for home investors. The ability to buy into appreciating markets and foresee a sustained decline comes with important strings. In order to benefit, first, you need to actually be able to accurately determine the market direction and then must execute a trading strategy much more difficult that logging on to your online broker and issuing buy and sell orders.

The Edge that Can be Gained
To a degree it has always been possible to benefit from the predictability of home prices by staying in close touch with a local market and gathering stats that have been compiled for decades such as the National Association of Realtors median sales prices of existing homes and more recently by looking at the price indexes of a company I co-founded over 25 years ago, Case Shiller Weiss.

WA has now made it even easier by producing house specific price indexes on 90 million homes which we also apply based on the principles described above to produce 90 million price forecasts each month.

WA converts these 90 million indexes into a very useful statistic, "Percent of Houses Rising". Defined each month for over 1000 US markets, this is simply the percent of houses in each market that are rising each month at an annualized rate of 2% or more.

Percent of Houses Rising at various geographic levels and market segmentations is one of the most effective leading indicators in WA's house specific forecasts.

This metric is also very predictive of turns in appreciation at the metro level.

When zoomed in on major metros, Percent of Houses Rising appears as below.
Sell Signals
When zoomed out the metric shows how the crash unfolded from 2004 to 2008.
Metros went into decline across a wide range of time periods and there was a signal that lasted several months.
Color-coding is 100% rising (green) to a middle percent rising (yellow ) to a full on crash with zero rising (red).
Note that not rising does not necessarily mean falling but in the above case nearly every house fell in value.
Below are graphs of example metro price indexes and percent rising indexes. You can see when the percent rising indexes went into decline vs the price indexes and the sell signals they provide.
First circle indicates a precipitous drop and unless percent falling is holding very steady I would have said "high risk". The market surprises by percent rising going back up and actual price index flattens out. Only begins dropping 3 years later. A highly unusual market.

Upon second percent rising drop I would have repeated "high risk" even louder. Would need to see high profit transactions to think risk is warranted.
Blue percent rising index shows a very obvious top coming. At the same level shows an obvious recovery coming.
Not a very early warning on the way down but a strong buy signal on the way up. Would be even briefer without index revision shown here.
Brief but screaming warning. Note what has been happening recently.
A bit of a fake out but it did not go down so not as weird as Denver.
Buy Signals
WA tracks over 1,000 markets when counting single family and condo markets separately. In order to illustrate the effectiveness of using the Percent Declining as a market signal for many secondary markets, below are several tables that include all such markets. Percent declining is the percent of all houses declining in value each month by an annualized rate of more than 2%.

Since these are percent down metrics, a zero means no houses are falling in value, so they are color-coded green. Markets with a high percent of declining houses are color coded red.
The graphic below begins in 2006 so you can see how the crash looks using the percent falling metric. You can see about 250 of the 1,000 markets. The markets have been sorted so that the markets with the fewest declining houses are at the top.

The first 50 markets actually never had any consistent value declines during the meltdown. As you go from top of the graphic to the bottom an increasing portion of the time the markets were declining during the meltdown.
Looking at the top 50 rows shows markets that were not just largely spared during the meltdown. They had very few houses ever decline in value since 2006. A sample of these indexes are graphed below.

The thick black line shows the median of these markets. Median decline was only 11%. Then most rise gently. These could be interesting markets to look at if they are large enough and have sufficient houses needing attention.
Below is the price index and percent falling for Trenton NJ. This example was chosen because it lies roughly in the middle of the severity of decline during the meltdown.


The period circled seems to indicate a buy signal.
Weiss Analytics provides the next generation of home price indexing, forecasting, and
analytics and is the only provider of house-specific repeat sales indexes in the US.
Founded by national housing index expert Allan N. Weiss, co-founder of Case
Shiller Weiss, Weiss Analytics combines leading industry experience,
proprietary analytics, and state-of-the-art big data computing power
to deliver revolutionary products with unprecedented resolution. Our
unique tools power better decision-making and help mitigate the financial
risk of home ownership and investments. Weiss Analytics is an independent
and trusted information source for home buyers and sellers, real estate
professionals, financial institutions, and hedge funds.

"Don't be fooled by averages. You're buying a house, not the market."