| Tesla production facilities in Fremont, California. © 2014 Airbus Defence and Space. |
The stock price of Tesla Motors, a company that has had its share of ups and downs, looked pretty uncertain two years ago. In the fall of 2012, CEO Elon Musk had revised his earnings forecast downward. A hedge fund manager who prefers to remain anonymous had invested in the company but had become skeptical that Musk was going to meet his much ballyhooed target of delivering 5000 of Tesla’s luxury electric Model S sedans by the end of the year. So the fund manager turned to a form of nonpublic information that is perfectly legal and not quite as invasive as it sounds: spying by satellite.
Tom Diamond, the president and one of three co-founders of Remote Sensing Metrics in Chicago, the firm that contracted a satellite for the hedge fund, produced some data that convinced his client that Tesla’s stock was not about to soar. He did that by sending a satellite to hover over the Tesla production facilities in Fremont, California, where the monitor picked up an image of the outdoor lot for new cars. Diamond says his client was working on the assumption that if the electric carmaker was to meet its production goal, there would be about 500 cars on the lot in mid-October 2012.
“The satellite imagery showed only 92 cars,” says Diamond. He isn’t sure if the hedge fund manager shorted the stock or unloaded his long positions, but Musk later announced that Tesla would be able to deliver about 2,500 Model S sedans for the year, and the client made money from betting against Tesla. (The stock, though still volatile, has since soared from around $33 at the end of 2012 to $140 at the start of 2014 to $240 as of March 19.)
Welcome to the next generation of hedge fund research. RS Metrics and other such research firms, which include its London partner Geospatial Insight and the energy research firm Genscape, Inc. in Louisville, Kentucky, contract government or privately-owned satellites to furnish their investment clients with data they’ve extrapolated from physical evidence of production or customer traffic. The satellites survey mines and crops to measure the output, retail parking lots to see how much business they’re getting, oil drilling to monitor production and increasingly, power usage and other activity that might indicate the economic health of a country or an industry. How to use the data is up to the investor.
What the satellites don’t do is probe what’s going on inside buildings, but it turns out that exterior activity can reveal quite a bit about a company or even a mine. A hedge fund manager might use that kind of data to help forecast stock or commodity prices or an overall economic outlook. Although Diamond keeps his hedge fund client names confidential, he says they’ve included quantitative managers, who are always interested in raw data, as well as equity managers who want to check out specific companies and macro managers looking for economic insight.
In the past year the number of satellites available from government agencies and private companies has doubled, from around 70 to over 140. Some of the satellites are operated by government agencies such as the National Aeronautics and Space Administration and the European Space Agency. But there is a growing number of private operators, including Digitalglobe, Airbus Defence and Space and newer entrants such as PlanetLabs, Urthecast and Skybox.
| A Target parking lot. © 2014 DigitalGlobe |
RS Metrics routinely provides satellite data to hedge funds and other financial clients about the traffic patterns at 24 big box retailers, including Best Buy, Target and WalMart. The reports are designed to compare the number of cars in the parking lot over time to let investors know if business is slowing down or picking up. The satellites have also been able to survey competing retailers, so that a portfolio manager can see whether Whole Foods is getting more customer traffic than Trader Joe’s, for example, or the traffic patterns at Dollar General versus Family Dollar and Panera Bread versus Starbuck’s.
At a time when emerging markets are a tough strategy—the Hedge Fund Research Emerging Markets Total Index declined 0.66 percent in the first two months of 2014—satellite surveillance might be able to help fund managers scope out where the best opportunities are, says Dan Schnurr, the chief technology officer and one of three co-founders at two-year-old Geospatial Insight.
Schnurr, a surveyor by profession, has used satellite intelligence for 20 years to make maps for infrastructure projects and survey land for oil drilling potential, but last year his firm began producing similar reports for hedge funds and other asset managers. He has provided data from satellite surveillance of crude oil, copper, wheat and palm oil production in emerging and frontier markets.
“If you send an analyst to investigate a mine in an emerging market, he’ll see what the officials want him to see,” says Schnurr. “Satellite imagery, on the other hand, gives you a peek under the hood.”
A hedge fund manager hired Geospatial to survey an oil palm plantation in Indonesia, for example. There was already existing earth satellite imagery, and Schnurr was able to compare that with the surveillance he contracted to show how the growth area had expanded over time. The imagery also allowed his firm to assess the age and health of the existing plants and even the environmental impact.
And in a twist on all of the allegedly fraudulent Chinese companies that Carson Block’s Muddy Waters Research uncovered, RS Metrics discovered just the opposite for a hedge fund manager in China. The fund manager came to Diamond’s firm thinking he might have a loser on his hands. He’d invested in a large Chinese coal producer, then been hit with a report from a Chinese research firm claiming the company’s production numbers were fraudulent and that it was producing almost no coal.
“We used historical imagery and saw stockpiles growing and shrinking,” says Mike Gantcher, managing director and head of sales at RS Metrics. The imagery, he says, provided evidence that production quantities had fluctuated over a period of several years, but the company was very much in operation. Furthermore, the satellite monitored the employee parking lot and found just over 200 cars. “We determined that 200 people don’t show up to work if the company isn’t producing,” says Gantcher.
