Big Data From Talking Turbines Signal Productivity Boom
At the Top Crop farm in Dwight, Illinois, 200 turbines rise from a sea of corn and soybeans, their blades gently turning day and night to spin up wind energy. They talk to one another, unheard by the human ear, seeking to keep pace with their neighbors’ output.
If one falls behind, sensors reach out to an office about 850 miles away in Schenectady, New York, where General Electric Co. (GE)’s remote operations center, using data from 19,000 windmills, finds the most efficient way to help. Intelligent monitoring of the machines has helped GE fix faults, limit snags and pre-empt thousands of failures.
Machines increasingly communicate among themselves and with people. Mobile devices allow round-the-clock interconnectivity. Computers crunch terabytes of data. Such innovations have convinced economists from GE’s Marco Annunziata to Erik Brynjolfsson of the Massachusetts Institute of Technology that the stage is set for a wave of productivity gains to rival the 10-year Internet boom that began in 1995.
“I’m quite optimistic,” said Brynjolfsson, a professor at MIT’s Sloan School of Management in Cambridge. “When I look at the technologies coming along, I project some big improvements from productivity.”
Stronger productivity would be good news for the U.S. and the global economy. It would allow faster growth without generating higher inflation. Companies could pay their workers more while still enjoying rising earnings. Higher profits would be a tonic for the stock market. Rising tax revenue would make it easier for the U.S. government to reduce its budget deficit.
So far, the projected surge in efficiency is more forecast than fact. Employee output per hour among non-farm businesses rose 0.3 percent in the 12 months ended in June, a tenth of the average 3 percent gain from 1995 to 2004, according to the Labor Department in Washington.
The meager advance buttresses the arguments of such productivity pessimists as Robert Gordon, professor at Northwestern University in Evanston, Illinois, who contends that innovation is faltering and that the U.S. is in for a prolonged period of near-stagnation.
“Pessimists may be paying too little attention to the strength of the underlying economic and social forces that generate innovation in the modern world,” Bernanke said. “The number of trained scientists and engineers is increasing rapidly, as are the resources for research being provided by universities, governments, and the private sector.”
Two former Fed economists who worked with former central bank Chairman Alan Greenspan in spotting the Internet boom now say they see a chance of a second wave of productivity gains.
The U.S. may be in “a pause period between the PC era and what comes next, with hand-held devices, massive connectivity and big data,” said Daniel Sichel, now a professor of economics at Wellesley College in Massachusetts. Such a second stage in innovation could lift the long-run trend growth rate of non-farm business productivity to about 2.5 percent.
“We are still near the starting line of the second wave,” said fellow researcher Stephen Oliner, now at the American Enterprise Institute in Washington. “Linking together existing technologies to create a new product takes a lot of work and a lot of trial and error to figure out how it can be done most efficiently.”
History suggests that productivity gains driven by general-purpose technologies can come in waves, University of Chicago professor Chad Syverson wrote in a paper earlier this year. That was the case with electrification around the turn of the last century, when a decade-long acceleration in productivity was followed by a multiyear slowdown, then a subsequent speedup.
“There are always lags between the introduction of new technologies and productivity gains,” Brynjolfsson said. “In my papers, we found that companies that installed big, new enterprise information systems didn’t get the full benefits for five to seven years.”
Bill Murphy, chief technology officer of the Blackstone Group LP (BX) in New York, the world’s largest private-equity firm, agreed that “there’s still tons of productivity to be gained out there.”
The benefits may be harder to come by than they were in the late 1990s and early 2000s, when computing power was doubling every year. Murphy compared such huge advances to being able to take a pill to lose weight.
Now, productivity-enhancing technology is more like buying a treadmill -- a tool that only works if you change your behavior and actually use it. “Just buying a treadmill doesn’t mean you’re going to lose weight,” he said. “It’s not going to do it just by itself.”
Companies have reasons to make the effort.
“There’s $14.4 trillion of value at stake in the next decade as we connect more things that drive productivity, that drive efficiency in the supply chain, that drive an expanded customer experience and revenue,” said Robert Lloyd, head of development and sales at San Jose, California-based Cisco Systems Inc. (CSCO), the world’s biggest maker of networking equipment.
That would be equivalent to a 2 percent increase in global output and a 20 percent advance in corporate profits, he said in an Aug. 22 interview.
Progress in five major areas -- social networks, mobile devices, cloud computing to make data available to those devices, big-data analytics to glean more useful information, and more effective security -- are working together to drive productivity, said Brian Lillie, chief information officer of data center operator Equinix Inc. (EQIX) of Redwood City, California.
“My sales guys have access to data constantly,” he said. “My engineers have access to technical data all the time. I can be coaching baseball and doing work on my phone. I’m not saying whether that’s good or not, but I know I’m a hell of a lot more productive than I used to be.”
He said he measures productivity at the most basic level by revenue per employee. On that basis, the company is keeping up: it has increased revenue between 5 percent and 7 percent more than headcount in recent years, according to Lillie.
Michael Feroli, chief U.S. economist for JPMorgan Chase & Co. in New York, remains unconvinced by such anecdotes. He puts the trend growth rate of productivity among non-farm businesses at 1.5 percent, half the level of the 1995 to 2004 period.
Companies aren’t making the investments needed to boost efficiency more, the former Fed researcher said. Adjusted for inflation, the stock of high-technology equipment and software in the economy rose at annual rate of 3.1 percent from 2009 to 2011, down from an average increase of 8.5 percent over the 25 years through 2008, according to Feroli. He reckons that growth in such high-technology capital stock is still running below 4 percent.
While investment dwindled during the recession, the economy now is primed for a pickup in capital spending as the domestic and global outlook improves, said Annunziata, chief economist at Fairfield, Connecticut-based GE. Companies are leaner, profitable and cash-rich, and the incentive to invest will grow as technological breakthroughs begin to pay off more.
“This is a wave of innovation that will boost productivity by much more than the Internet revolution,” he said, describing it as the Industrial Internet. “We are seeing the turning point in the adoption and speed of adoption of these technologies. Companies like ourselves are investing more and more.”
U.S. productivity gains could return to 3 percent in the initial push in the next five to 10 years, reaching as high as 3.5 percent as the business-to-business innovations gain traction over the following decade, Annunziata predicts. He also said this next phase will play out for a longer time, lifting economic growth, and “that’s something the U.S. clearly needs.”
To contact the reporters on this story: Shobhana Chandra in Washington at firstname.lastname@example.org; Rich Miller in Washington at email@example.com; Peter Burrows in San Francisco at firstname.lastname@example.org