Talking Charts – Sentiment Rules
Tuesday, September 14th, 2010This post first appeared in the September edition of: Cirios Trends: In Search of Real Estate Opportunities
One of the chief challenges in trying to be experts in home price trends is that you are constantly challenged. So, one of the most common questions we receive is “how can you predict what direction housing prices will go?” The short answer, that we cannot, is the simplest and most accurate. However, there are certain home price metrics that are predictive — that is, can help forecast the direction of prices. One of those, which we discuss often, is Inventory. Rising supply often portends lower prices, as buyers begin to gain the upper hand in negotiations. Another metric, one we developed here at Cirios, we call “The Overbid Indicator.”
This metric tries to approximate buyer sentiment, which provides insight into how many buyers there are relative to sellers and how excited those buyers are to buy. We assign a value of 1 to any sale above the list price, a value 0.5 to a sale exactly list, and a value of 0 to any sale below list. These values are arbitrary (you could just as easily use 2, 1.5 and 1, etc) but when the ratio is looked at over time, it can show what portion of sales occur above list. The more sales above list, the more scarce well-priced properties are, and the more listing agents are counting on bidding wars indicative of periods when demand overwhelms supply. We have found that this metric is highly predictive, particularly at cusps, where sharp trend shifts can push home prices in one direction or another.
One final point on this metric is that tends to be highly seasonal. That is, during the spring and summer buyers are by nature more excited, and typically more inclined to make offers above list. The opposite is true during the winter, especially around the holidays. The next iteration of this metric will attempt to factor in these seasonal aspects for a smoother, more realistic representation of
sentiment.

It wouldn’t be an issue of Cirios Trends without some mention of East Palo Alto, our poster child for a pocket of affordable housing smashed in between some of the most expensive real estate in the country. Looking at the overbid indicator above, demand overwhelmed supply during the boom, shown by a rising red line. The indicator peaked in early 2005, a full year before prices peaked in 2006. On the other side, sentiment, as measured by this metric, bottomed in early 2008, also around 12 months prior to a trend shift. Of course, two data points are far from conclusive in saying that the Overbid Indicator is a 12-month leading indicator of home prices, but it certainly can help warn market participants that the winds of change are afoot. Currently, despite a leveling of prices, demand remains strong in East Palo Alto, and has not slumped nearly as deeply as other markets (see next page).
———————————————————————————————————————-

Much like East Palo Alto on the previous page, the Overbid Indicator for South San Francisco peaked well before home prices did. And, similarly, sentiment bottomed out well before prices. One distinct difference between to the two cities is recent activity. Recall that East Palo Alto has seen sentiment remain steady, while looking above you can see that the Overbid Indicator started falling in South San Francisco at the beginning of 2010. One likely reason for this is that East Palo Alto is a more foreclosure-driven market, and bank owned inventory has been slow to seep out into the market this year. This has kept inventory levels low, while all cash investors have been eagerly looking for low priced homes.
South San Francisco, on the other hand, is driven by short sales and regular sales, which have seen a steady increase in the past 6-8 months. While it would be premature to predict a dramatic drop in South San Francisco home prices, we are cautious, given that buyers in that area have turned more cautious as well.
———————————————————————————————————————-

As we often discuss, home price trends are becoming increasingly local, as structural factors affecting prices begin to be supplanted by fundamentals. This is a positive sign, not necessarily for prices themselves, but rather for restoring health to markets, and eventually providing support for sustainable, renewed appreciation. In this graph, we compare two distinct zip codes within Redwood City and look at how important it is to understand divisions within cities. 94063 is located on the east side of El Camino Real, consisting of generally lower income neighborhoods. 94061 on the other hand, is made up of quiet, tree lined streets and abuts Atherton, one of the wealthiest communities in the country. Unsurprisingly, the effects of the housing boom and bust were felt more strongly in 94063, where peak to trough declines nearly hit 50%, while 94061 experienced declines of “just” 34.4%. Looking at this chart, one could glean that if you had purchased two identical homes in 94063 and 94061 in our base year of 1998, you would have been only slightly better off on the 94063 home.
This shows that while it is typically “safer” to buy in a more established neighborhood, you are still exposed to price declines. Meanwhile, buying in the less established, more gritty area certainly poses risks, but there is more potential upside as demographic changes can bring more wealthy buyers into the area.
———————————————————————————————————————-

Anyone who tries to paint the entire picture of an area with a single home price metric is missing the point, and probably has an agenda. We saw on the previous page that over the past decade or so, buyers in 94063 did almost as well as 94061. The graph above paints a different picture. Or rather, helps glean how a buyer may fare if they were to buy now in either area. Back in 1998, homes in 94061 sold at around a $50/sqft premium to those in 94063. This spread held through the dotcom boom, but as the housing bubble began to inflate in 2002, loose lending standards evened the playing field, values converged and homes in the two zip codes sold at a similar price per square foot. (It is important to note, however, that during this time, the average size of a sold home in 94063 was 1133 sqft vs. 1551 sqft in 94061. All else being equal, smaller homes sell at a higher price per square foot than larger ones, so these relative values should be taken with a grain of salt). As prices collapsed, the spread between these two areas has grown, now standing at well over $100/sqft, the widest in recent memory.
This can be viewed in two ways. First, prices in 94063 have crashed further, as high foreclosure rates in the area reflect a homeowner population that could not afford bubble prices. Second, and more interesting, we would argue that on a relative basis, homes in 94063 are selling at a steep discount to those in 94061. If that spread reverts to the mean, then buying in 94063 now could in fact be a better investment than buying in 94061.


