| Popis: |
The accumulation of properties by Spanish banks during the crisis of the first decade of the 21st century has definitely changed the housing market. An optimal house price valuation is useful to determine the bank’s actual financial situation. Furthermore, properties valued according to the market can be sold in a shorter span of time and at a better price. Using a sample of 24,781 properties and a simulation exercise, we aim to identify the decision criteria that Spanish banking used to decide which properties were going to be sold and at what price. The results of the comparison among four methods used to value real estate—artificial neural networks, semi log regressions, a combined model by means of weighted least squares regression, and quantile regressions—and the actual situation suggest that banking aimed to maximize the reversal of impairment losses, although this would mean capital losses, selling less properties, and decreasing their revenues. Therefore, the actual combined result was very detrimental to banking and, consequently, to the Spanish society because of its banking bailout. |
