Sometime this fall, an international standards body will sign off on a tracking technology that will launch one of the most massive purchases of a new electronics product by business in history. It is not too dramatic to suggest that this technology will lead major companies like
Procter & Gamble
to buy hundreds of billions more semiconductors per year than they obtain today and lead to an entirely new relationship between people and the things they buy.
This technology, called radio frequency identification, or RFID for short, has been around for a couple of decades in a variety of crude forms. But with the advent of a global standard, the prospect for radically lower-cost devices and increasing demands for a streamlined worldwide supply chain have led a vanguard of key Fortune 500 manufacturing and retailing companies to push RFID to the front of their spending plans for the next three to 10 years.
"In terms of chips sold, it will dwarf any other semiconductor application out there now," said Bill Allen, a spokesman for
An Internet of Things
At its most basic, RFID simply enables businesses to identify and track their assets wirelessly. Slap a 50-cent signal-emitting transponder onto a case of Coke or a pallet of Cover Girl makeup, and a manufacturer can remotely monitor its progress from the manufacturing plant through the warehouse to the trucker to the distributor to the supermarket warehouse, with diminished threat of human error. Slap a 5-cent transmitter on the can of Coke itself, and the retailer knows immediately, via a reader and antenna on a "smart shelf," when Aisle 7A needs replenishing.
Unlike a bar code, which identifies an aluminum cylinder filled with brown fizz simply as a can of Coca-Cola and must be read one-by-one via a line-of-sight wand, an RFID transmitter gives each can its own reprogrammable digital identity and can be read wirelessly at a 10,000-item-per-second clip. This "electronic product code," or EPC, allows a retailer to know that a can -- or, more important, something dangerous or perishable such as a drug or a package of fish -- began life on the third conveyer belt of Building B of the manufacturer's Sarasota, Fla., plant on June 14.
That may sound prosaic, but it radically changes the relationship between giant manufacturing concerns and their products. Something like 15% of everything a large company makes is essentially unaccounted for at any given time -- as much as $2 billion worth of missing stuff for a company the size of P&G. Not lost. Not stolen. Just sitting around somewhere beyond the reach of inventory takers. As a result, manufacturers estimate they put about 15% more products in their supply chain at every level than they or their retailers really need, just in case they run out. Any technology that reduces that number to 10%, or 5% or even less, will lead to tens of billions of dollars in annual savings.
At a more ethereal level, however, RFID broadly enables an Internet of things rather than computers, drawing civilization closer to the day when computing devices connect every object to every other object in pervasive and ultimately unthinkable ways. "We like to imagine a world where there are billions of small computers not on desktops, but embedded in the real world," says Ravi Pappu, co-founder of Boston-based computer engineering firm ThingMagic, which advises several RFID manufacturers.
Wal-Mart Pressures RFID Chipmakers
The prospect of continuous monitoring of things disturbs privacy advocates, but consumers have already proved themselves willing to trade some confidentiality for convenience. An example is the one-click service at
, in which shoppers allow the Web site to harbor their credit card information in exchange for the ease and speed of single-click purchasing. In a fully RFID-powered world, a smart-card carrying consumer could conceivably roll her shopping cart directly out of a supermarket past a reader that would tote up her purchases, charge her credit line, deduct the items from inventory and hand her a receipt -- and then disable the tracking devices.
If you think that's a far-off concept, consider that
just last week said it plans to require its top 100 suppliers to put RFID tags on shipping crates and pallets by January 2005 -- a move widely seen to be pivotal in the broad adoption of the technology. Kevin Ashton, executive director of the Auto-ID Center at the Massachusetts Institute of Technology, told reporters that Wal-Mart's adoption of electronic product code standards "should give companies confidence that the day of a single, interoperable RFID system is close at hand."
The sticking point today is the price of RFID devices, which vary in range, frequency and power from passive reflectors that transmit a few inches to ones that are battery-powered and transmit more than 100 yards. Currently the cheapest models run about 30 cents. Wal-Mart wants them to cost 5 cents in order to make a business case for its vision of a fully connected supply chain.
The differential explains why this is one of the rare technologies that is being pushed harder by the end users than by the electronics manufacturers. Texas Instruments and
, who dominate RFID device manufacturing today, want to make increasingly elaborate, expensive and high-margin chips, not obnoxiously cheap commodity chips. But if they don't come through fast enough, the Wal-Marts and P&Gs have threatened to cut them out of the deal and order the devices through small start-ups they partially control.
Alien Technology: Start-Up at the Leading Edge
Idle threat? Maybe not. In January,
got the ball rolling by placing an order for 500 million RFID tags from tiny, privately held
of Morgan Hill, Calif. -- the first major commercial deal for the technology. Alien, which boasts
among its investors, is out in front because it developed and holds patent rights to a manufacturing process called Fluidic Self Assembly. The company says the process allows it to package tiny integrated circuits, termed NanoBlocks, into electronic product tags at a rate of two million per hour -- about 200 times faster than the 10,000-per-hour rate of manufacturing conventional ICs. Not only does that reduce the cost of the tags (the Gillette order is estimated at 10 cents per chip), but it will permit the annual production of trillions of tags in the expected future.
Louis Sirico, founder of the Washington, D.C., consulting firm RFID Wizards, recommends investors step back from the big picture and just look at the valuable uses the technology already provides today. At an automobile parts assembly plant, RFID transmitters are embedded in the foam that ultimately will go in the seat of a car. The transmitter identifies one type of foam as a $25 asset, and a denser but identical-looking piece of foam as a $45 asset. Once the foam is covered with nylon or leather, the transmitter makes it easier for the carmaker to ensure that the correct foam is inside the seat assembly; and to direct it right instead of left on a conveyer belt for special handling if necessary -- labor-intensive actions prone to human error today. "The ability to quickly identify components within a finished good without destroying the finished good is a powerful concept that doesn't exist without RFID; it's a new capability that saves money and time," Sirico said.
Unfortunately, opportunities for pure-play investments in RFID through public companies seem scant. Sirico, who advises clients who wish to implement systems, offers this analysis:
Texas Instruments makes good equipment in the lower-frequency ranges and recently sold its 200 millionth transponder, but has not adopted the new global standard and could get shunted aside. Moreover, RFID is only part of a sensors division that produces just 12% of the company's $8.8 billion in sales.
The Intermec division of
( UNA) has a promising new line of patented RFID products.
The Dutch electronics giant Philips is a major maker of equipment and has made numerous investments in private companies.
The Sensormatic division of
is a major player, partnering for well-regarded RFID devices with ThingMagic.
and wireless-device manufacturer
( SBL) each have separate and joint RFID solutions.
A small Canadian company,
, also makes well-regarded readers.
The safest bet for public equity investors may be on the big manufacturers and retailers who expect to use the technology to wring significant expense savings from their infrastructure. Mark Roberti, publisher of
, notes that a typical, well-run Wal-Mart might be out of stock 5% of the time. If RFID networks can reduce that to 4%, the 1% additional sales will add $2.5 billion to their revenues per year. "That's just Wal-Mart and does not include labor savings or improved efficiencies," he says. "The potential is mind-boggling."
This retail utopia is still several years away. Procter & Gamble, one of the most enthusiastic promoters of the technology, will start testing the new standard in the fall but won't put transmitters on cases and pallets until 2004 or 2005. Putting electronic product codes on individual bottles of Tide detergent, boxes of Cover Girl lipstick or tubes of Crest toothpaste -- about 23 billion items per year in all for P&G -- would follow that. But it is coming. "Our out-of-stock levels are much higher than we'd like," said Jeannie Tharrington, a P&G spokeswoman. "We think this technology is viable and has tremendous potential to make our supply chain more customer-centered, and we are doing everything we can to get it up and running."
publisher, said he believes the industry is about where the Internet was in 1994. "But we think it'll be bigger than the Internet," he said. "All the Web did was connect computers to computers. That's not as big as connecting things to computers."
If the business transpires as expected, one thing is certain: Those piles of chips and miles of wireless beams will produce mountains of data for companies to analyze. This summer, I will look at the burgeoning industry of business intelligence software and how it will potentially intersect with electronic product codes. In the meantime, please send your ideas on ways to invest in this trend to me at
email@example.com, and I'll publish them in a future column.
Jon D. Markman is senior investment strategist and portfolio manager at Pinnacle Investment Advisors. While he cannot provide personalized investment advice or recommendations, he welcomes column critiques and comments at
firstname.lastname@example.org. At the time of publication, neither he nor his fund held positions in the equities mentioned in this column, but positions can change at any time.