Psychologists often point out that people are torn between two minds, the rational and the instinctual. This makes the job of an economist a difficult one — market behavior would be easier to explain if everyone were a bit more reasoned in their judgments and actions.
For a case in point, consider the following study conducted by a team of psychologists at Cornell University. Participants were asked to estimate how much they would be willing to spend at a hypothetical restaurant, either called “Studio 17” or “Studio 97.” The researchers found that participants were willing to pay significantly more ($32.84 vs. $24.58, on average) at a restaurant named Studio 97. In other words, the higher number in the restaurant name prompted people to think of larger dollar amounts when contemplating their willingness to spend.
New research shows that this phenomenon — known as an “anchoring bias” — can even influence property appraisals. Scientists at Yildiz Technical University in Turkey examined historical property data from Istanbul’s first cadastral survey, conducted in 1875. What started as an effort to understand the key determinants of home values quickly turned into a quest to explain peculiarities in nineteenth century Turkish home appraisals.
This allowed the researchers to predict property appraisals from the matrix of data collected by the civil servants over 140 years ago. Using 315 real estate sites from three different neighborhoods in Istanbul, the researchers tested which variables (e.g., number of rooms, construction material, location, etc.) were most important in predicting property appraisals.
They found that all of the usual suspects (location, number of rooms, size, etc.) were influential drivers of property values. Curiously, however, they found that the home number was also predictive of property appraisals. According to their estimates, a two-fold increase in house number (e.g., 50 vs. 100) increased the appraised value by 10–25%.
The researchers searched for logical explanations. One theory, for instance, was that higher property numbers were associated with homes built in more valuable areas of the city. Upon further inspection, however, the researchers found no evidence in support of this association. They write, “Had real properties with high door numbers been located in the valuable parts of the city, or built earlier, those buildings would be distant from buildings with low door numbers. However, the street maps […] show no such pattern.”