By Konrad Putzier
In 2013, Barbara Zweig listed her penthouse co-op at the Pierre on Fifth Avenue with Sotheby’s International Realty for $125 million.
In December of that year, the asking price was chopped to $95 million. As of February 2015, it is asking a mere $63 million, now listed by Brown Harris Stevens.
In a two-year stretch during which the average co-cop sales price per square foot on the Upper East Side grew by 8.5 percent, according to REBNY, this particular co-op’s price fell by almost 50 percent.
In other words: the original asking price was completely ludicrous.
Zweig’s co-op is an extreme example of the liberty many brokers and sellers take when pricing a property.
While mortgage appraisals follow strict guidelines, there is no definitive science behind residential sales pricing.
Often enough, the asking price depends not on a careful analysis of neighborhood comparables, but on a broker’s gut, or on what a seller would ideally like to get.
This means asking prices can be way out of sync with the market, which creates costs.
In the case of Zweig’s penthouse, the seller assumes the cost of sitting on an unsold apartment for two years — not to mention the brokerage’s lost time and expenses on staging. Moreover, a price chop can send the wrong message to prospective buyers, making it even harder to sell.
But increasingly, New York’s residential real estate industry is waking up to the need for more accurate pricing. Last year, Compass’ co-founder Ori Allon told Real Estate Weekly that the company is working on an algorithm to price properties — although he has since declined to go into detail.
Perhaps the most interesting initiative comes from Camilo Galvis and his startup AssetCast.
Galvis has developed a program to price real estate assets, based on ten years of research and quantitative analysis.
He claims that, with the help of AssetCast, sellers can increase their revenue by three to seven percent by picking the optimal sales price.
The company was born out of research at Columbia University. Galvis helped two business school professors, Omar Besbes and Costis Maglaras, research a 2011 paper on optimal pricing, which developed a “revenue maximizing dynamic pricing policy”. The method became the foundation of AssetCast.
Maglaras is the startup’s head of analytics, and the firm claims to have since tested its method on $1.1 billion worth of inventory.
Galvis explained that AssetCast goes beyond traditional comps in pricing. Instead of simply comparing a two-bedroom on the Upper East Side to other two-bedrooms, the program will compare more precise characteristics such as the number of bathrooms.
Moreover, it will adjust the optimal sales price for an apartment as other units in the building sell — hence the “dynamic” in the description.
Landlords, brokers or financial firms looking to use AssetCast will get a software interface individually configured to an asset, which they can then manage themselves.
The software comes with a price, but Galvis argues this will be more than offset by the benefits of finding a revenue-maximizing sales price.
Galvis’ background seems ideally suited for his role leading a tech startup.
The Colombia native started out in IBM’s technology department before joining the real estate development company Fortune International in Florida in 2001 — which soon put him at the heart of the U.S. real estate and financial crisis.
Galvis claims that seeing how dramatically financial firms mis-priced real estate assets in the run-up to the crash gave him the inspiration for AssetCast.
Several years and a long research stint at Columbia University later, Galvis has moved to New York in the hopes of reforming the country’s biggest real estate market.
For now, his goal is to grow his company and get the industry to pay more attention to pricing. “Hopefully, we can start this conversation and get people thinking about it,” he said.
* AssetCast is a venture between Camilo Galvis, Costis Maglaras, AssetCast Head of Analytics and David and Lyn Silfen Professor of Business in the Decision, Risk and Operations Division at the Graduate School of Business at Columbia University; and Soulaymane Kachani, AssetCast Head of Research and Professor at Columbia University’s Department of Industrial Engineering and Operations Research and Vice Dean of Columbia’s School of Engineering and Applied Science.