Imagine
cows fed and milked entirely by robots. Or tomatoes that send an e-mail
when they need more water. Or a farm where all the decisions about
where to plant seeds, spray fertilizer and steer tractors are made by
software on servers on the other side of the sea.
This is what
more and more of our agriculture may come to look like in the years
ahead, as farming meets Big Data. There’s no shortage of farmers and
industry gurus who think this kind of “smart” farming could bring many
benefits. Pushing these tools on to fields, the idea goes, will boost our
ability to control this fiendishly unpredictable activity and help
farmers increase yields even while using fewer resources.
The big
question is who exactly will end up owning all this data, and who gets
to determine how it is used. On one side stand some of the largest
corporations in agriculture, who are racing to gather and put their
stamp on as much of this information as they can. Opposing them are
farmers’ groups and small open-source technology start-ups, which want
to ensure a farm’s data stays in the farmer’s control and serves the
farmer’s interests.
Who wins will determine not just who profits
from the information, but who, at the end of the day, directs life and
business on the farm.
One recent round in this battle took place in October, when Monsanto spent close to $1 billion to buy
, a data analytics firm. Last year the chemical and seed company also bought
to partner to provide farmers information and prescriptions in
near-real time. Deere has pioneered “precision farming” equipment in
recent years, equipping tractors and combines to automatically transmit
data collected from particular farms to company databases. DuPont,
meanwhile, has rolled out a service that analyzes data into “
Many
farmers are wary that these giants could use these tools to win
unprecedented levels of insight into the economics and operational
workings of their farms. The issue is not that these companies would
shower the farmers with ads, as Facebook does when it knows you’re
looking to buy sneakers. For farmers, the risks of big data seem to
pierce right to the heart of how they make a living. What would it mean,
for instance, for Monsanto to know the intricacies of their business?
Farm
advocacy groups are now scrambling to understand how — if given free
rein — these corporations could misuse the data they collect. “We’re
signing up for things without knowing what we’re giving up,” said Mark
Nelson, director of commodities at the Kansas Farm Bureau. In May, the
American Farm Bureau Federation, a national lobbying group,
published a policy brief outlining some potential risks around these data-driven farm tools.
For
farmers, the most immediate question is who owns the information these
technologies capture. Many farmers have been collecting digitized yield
data on their operations since the 1990's, when high-tech farm tools
first emerged. But that information would sit on a tractor or monitor
until the farmer manually transferred it to his computer, or handed a
USB stick to an agronomist to analyze. Now, however, smart devices can
wirelessly transfer data straight to a corporation’s servers, sometimes
without a farmer’s knowledge.
“When I start storing information up
on the Internet, I lose control of it,” said Walt Bones, who farms in
Parker, S.D., and served as state agriculture secretary.
Justin
Dikeman, a sales representative with DuPont, said farmers continue to
own whatever data they collect on their operations. A spokesman for John
Deere also said farmers own their data, and that farmers have the
opportunity to opt-out of the company’s cloud services. Monsanto did not
reply to a comment request.
Details on who owns what at which
stage of the analytics process is less clear, though. Even if a contract
guarantees that farmers own the raw data, for instance, it’s not
obvious whether farmers will be able to access that data in a
non-proprietary format. Nor is it evident how easily farmers can stop
these devices from collecting and transmitting data.
How
corporations use the information is another central concern. One worry
is the giants will harness the data to engage in price discrimination,
in which they charge some farmers more than others for the same product.
For example, details on the economic worth of a farm operation could
empower Monsanto or DuPont to calculate the exact value the farm derives
from its products. Monsanto already varies its prices by region, so
that Illinois farmers with a bumper crop might be charged more for seeds
than Texas farmers facing a drought. Bigger heaps of data would enable
these companies to price discriminate more finely, not just among
different geographic regions but between neighbors.
Another issue
is how the value of this information will be determined, and the profits
divided. The prescription services Monsanto and DuPont are offering
will draw on the vast amounts of data they amass from thousands of
individual farms. Farmers consider much of this information – such as on
soil fertility and crop yields – confidential, and most view details
about particular farming techniques as akin to personal “trade secrets.”
Even if the corporations agree not to disclose farm-specific
information, some farmers worry that the information may end up being
used against them in ways that dull their particular competitive edge.
“If
you inadvertently teach Monsanto what it is that makes you a better
farmer than your neighbor, it can sell that information to your
neighbor,” said John McGuire, an agriculture technology consultant who
runs Simplified Technology Services and developed geospatial tools for
Monsanto in the late-1990s. And if the corporation gathers enough
information, “it opens the door for Monsanto to say, ‘We know how to
farm in your area better than you do,’” he said.
There are also no
clear guidelines on how this information will be used within commodity
markets. Real-time data is highly valuable to investors and financial
traders, who bet billions of dollars in wheat, soybean and corn futures.
In a market where the slightest informational edge makes the difference
between huge profits and even bigger losses, corporations that gather
big data will have a ready customer base if they choose to sell their
knowledge. Or they could just use it to speculate themselves.
“If
this real time yield data goes into the cloud and a lot of market
investors get into it, there is potential for market distortion,” said
Kyle Cline, policy advisor for national government relations at the
Indiana Farm Bureau. “It could destabilize markets, make them more
volatile,” he said.
John Deere has
stated
it will not share data with anyone it believes will use it to influence
or gain an advantage in commodity markets. Monsanto, DuPont and other
firms have not, however, issued similar public statements.
Some
farmers and smaller manufacturers also worry that data analytics will
give conglomerates like Monsanto and DuPont more power to compel farmers
to buy other lines of products. Monsanto, for example, has proven
highly adept at leveraging its wide suite of products to support one
another.
How Monsanto used its dominance in one business (genetic
traits) to benefit others (seeds, fertilizer) was the focus of a
three-year antitrust investigation by the Justice Department. (DOJ
closed the probe last November without taking any action).
In
recent years, Monsanto, DuPont and John Deere have also expanded into
selling farmers a variety of financial services and insurance. John
Deere, for one,
acknowledges that its financial division may consult data from a farmer’s machinery, if the farmer permits.
Other
private corporations are also competing for a share of the big data
pie. Established equipment manufacturers like AgCo and Case IH have been
expanding their data analytics services, and some high-tech upstarts
are also joining the game. The Climate Corporation, the weather data and
insurance company Monsanto bought in October, for example, was founded
by a former Google employee.
Open-source groups attempting to provide farmers with some similar technologies include
ISOBlue, a project based at Purdue University, which teaches farmers how to capture and independently store their own data.
FarmLogs,
a Michigan-based company backed by Silicon Valley money, sells software
and data analytics that let farmers fully control the information
collected. “We’re pushing back against the monopoly on information” that
some existing vendors create, said FarmLogs founder Jesse Vollmar, who
grew up on a farm.
What is not clear is whether these smaller
open-source companies will be able to keep up with the established
giants over the long run.
“Monsanto has its fingers awful deep
within our industries,” McGuire said. “Its expansion [into data
analytics] should scare a lot of people.”
To be sure, much depends
on how widely farmers adopt the privately-designed “decision-support”
services. Monsanto has estimated this to be a
$20 billion market,
but there is no proof yet whether the company will be able to process
these reams of data into profitable farming and business strategies.
Whether Monsanto’s bet will pay off is “tough to validate,” said Paul
Massoud, an analyst with Stifel Nicolaus.
Some experts question
whether relying on prescription-based services is in farmers’ best
interest at all. “I don’t see farmers themselves crunching numbers, so
[I] doubt they’ll be learning anything more about how to farm well,”
said Bill Freese, expert on agricultural biotech and science policy
analyst with Center for Food Safety. “Monsanto’s scheme does not really
represent farmers embracing data analytics, but Monsanto embracing it to
better sell the seeds it wants to sell with a pseudo-scientific
rationale.”
A new group called the Grower Information Services
Cooperative thinks the best way farmers can protect their interests
during the transition to big data is to organize. Formed in west Texas
last December, GISC is pushing a model where farmers would store their
data in a repository through the co-op, and companies would pay the
group a fee to access it. The system would give farmers technical as
well as legal ownership, and provide a way for them to share in its
monetary value, said Mark Cox, controller and communications director
for GISC.
“Growers need to be proactive in how their information
is managed,” Cox said. “Otherwise all that economic power will
consolidate to these corporations and the grower will be at even more of
a disadvantage. We don’t want the grower to become a tenant on his own
farm.” GISC began accepting members this month, and is meeting with farm
bureaus around the country to publicize its mission.
Soil sensors
and seed planting algorithms may be a game-changer. Whether farmers
fully reap the fruits of that harvest, though, will depend not on
technologies but on the legal technicalities that bind their use.
By