The EconomistFebruary 8th 2020 Business 55
2 Tretikov, a vice-president of its artificial-
intelligence (ai) business. But none of
these executives appears likely to succeed
Mr Nadella unless he sticks around for a
few years more.
Besides more men, Big Tech corner of-
fices can expect more geeks. Whereas Ms
Rometty and Mr Benioff came from sales,
Mr Krishna (who oversees ibm’s cloud and
ai business) and Mr Taylor (who worked as
Facebook’s technology chief before he
joined Salesforce) are engineers by trade.
Mr Williams, too, has an engineering de-
gree. Mr Scott is a computer scientist. All
are also seasoned managers by now.
Each new ceo will face distinct chal-
lenges. Mr Krishna has to complete ibm’s
pivot from conventional computing, such
as mainframes and information-technol-
ogy services, to ai and the cloud. Mr Taylor
would have to integrate Salesforce’s recent
acquisitions. Mr Williams’s main task
would be to grow Apple’s services business
as it sells fewer iPhones. Whoever takes
over at Microsoft must ensure that Mr Na-
della’s remarkable reinvention of the soft-
ware-maker as a big-data and cloud-com-
puting behemoth stays on course.
If Mr Benioff goes, only three of Ameri-
ca’s ten biggest tech firms will be run by
their creators: Amazon, Dell and Facebook.
Mr Bezos and Michael Dell are in their 50s.
Mr Zuckerberg is a stripling 35. Neither is
going anywhere; Mr Dell tried retirement
once before and it did not agree with him.
They could nevertheless take some point-
ers from tech’s new chiefs, who tend to be
less abrasive and more politic than flam-
boyant founders. Those are useful traits at
a time of techlash from politicians and
calls for cuddlier behaviour from woke
consumers and employees. 7
Bartleby I am Number 0.6
A
close friendof Bartleby’s just got
the news that their department was
shedding 2.6 workers. At first sight, the
concept of 0.6 of a worker sounds pretty
odd. But workers who are freelance, on
temporary contracts, or in part-time
employment register in the headcount as
less than a whole number.
Being classed as 0.6 of a worker seems
dehumanising. Few people want to be
thought of as just a number, let alone a
fraction. In “The Prisoner”, a cult British
television series from the 1960s, the hero,
played by Patrick McGoohan, resigns
from his job as a secret agent only to be
abducted and taken to a village. He is
only referred to as “Number 6” and his
frequent escape attempts are frustrated.
Although he insists that “I am not a
number, I am a free man”, the audience
never learns his name. The programme
has a very 1960s vibe—it focuses on the
individual’s efforts to assert himself in
the face of a repressive, conformist soci-
ety. At one point, the title character de-
clares: “I will not be pushed, filed,
stamped, indexed, briefed, debriefed or
numbered. My life is my own.”
These days many workers would
sympathise. They feel pushed, filed,
indexed and numbered. When they apply
for a job, they may be assessed by artifi-
cial intelligence, which parses résumés
for key words without which an appli-
cant’s odds of an interview lengthen.
Based on works like “Evidence-Based
Recruiting” by Atta Tarki, who claims
that scores in general-mental-ability
tests have a strong 65% correlation with
job performance, firms may ask candi-
dates to take an intelligence test.
When they get a job, employees find
the indexing and numbering continues.
Workers at warehouses have to pick a
certain number of items per hour; those
at call-centres are assessed by software
that monitors their hourly number of
calls, and the amount of time spent on
each one. Fall behind the target and you
may feel unable to take a break. When their
task is completed, employees are often
rated again, this time by the customers.
Manufacturing workers have long faced
these kind of numerical targets, as well as
the need to clock in and out of work. The
big change is that similar metrics and
rating systems are spreading to more and
more parts of the economy. Academics get
rated by students; nurses may be judged
on a “behaviourally anchored rating scale”
which assesses how much empathy they
showed to patients.
Ratings are at the heart of the gig econ-
omy, where workers are connected with
employers and customers via the internet.
Just as TripAdvisor ratings allow holiday-
makers to assess hotels, Uber drivers get a
score out of five. The same goes for ratings
on services like TaskRabbit (for odd jobs)
and Etsy (for arts-and-crafts sellers).
Such systems are understandable in
parts of the economy where output is
difficult to measure precisely. But they
can be arbitrary. People might give an
Uber driver a poor rating because they
are in a bad mood or because they en-
countered unexpected traffic disruption
(the drivers themselves also rate custom-
ers, which is meant to discourage abuse).
The result can be increased insecurity
for gig-economy workers. Their income
is uncertain when they are at the mercy
of the assessment system. Even a tiny fall
in their rating—of, say, 0.6—can harm
their job prospects. A detailed study* of
65 gig-economy workers found that they
relished their independence but it came
with a host of personal, social and eco-
nomic anxieties.
Even full-time workers may find
themselves dependent on their score in
one category or another. Businesses want
to avoid accusations of hiring biases on
grounds of gender or ethnicity; using
“objective” rating systems can protect
them from discrimination lawsuits. And
employees need to be concerned about
how they are rated.
Gianpiero Petriglieri of the insead
business school says that, since firms no
longer offer jobs for life, everyone is an
independent worker whether they like it
or not. The key passage in your cv may
not be the universities you attended, but
your rating in categories like teamwork,
innovation and adaptability.
Heaven forbid, the system even ex-
tends to journalists. Some publications
reward writers based on the number of
clicks their articles attract. Find out more
in next week’s Bartleby column: “How
the Kim Kardashian diet can boost your
iqandjobprospects”.
How modern workers are at the mercy of ratings
.............................................................
* “Thriving in the gig economy ”, by Gianpiero
Petriglieri, Susan Ashford and Amy Wrzesniewski,
Harvard Business Review, March-April 2018