THETRADETECH DA I LY
in-depth
THE OFFICIAL NEWSPAPER OF TRADETECH 2020
As a new decade begins, Chris Hall looks at how the role
of the buy- and sell-side trader has evolved since the flash
crash of 2010, and finds that while the buy-side has levelled
up with brokers, progression has not always been smooth.
I
t was 2010 when electronic trading first
hit the headlines. Trade automation
was hardly new. Technology-driven
innovations had been accommodated in the
US and Europe under Reg NMS and MiFID
respectively. But execution algorithms
and high-frequency trading (HFT) were
still a mystery to senior executives at asset
management firms, not to mention their
institutional and retail clients.
This changed on May 6, 2010, when a
combination of factors - speed, structural
weaknesses and a looming European
sovereign debt crisis - ignited the ‘flash
crash’, a brief but alarming collapse in US
stock prices. The Dow Jones Industrial
Average plunged almost 1,000 points,
only to rapidly regain composure, leaving
regulators, traders and investors to wonder:
what on earth just happened?!
First, a poorly parameterised index futures
order from a mid-market fund manager was
identified as the trigger. Later, the finger
was pointed at Navinder Singh Sarao, the
‘hound of Hounslow’, an autistic amateur
trader who played the markets from his
bedroom.
Closer to the truth was author and ex-
bond salesman Michael Lewis, whose
2014 expose, ‘Flash Boys’, suggested the
flash crash was an accident waiting to
happen. Exchanges, regulators and brokers
had facilitated a new form of high-speed
market-making and the entry of a new breed
of market participant, allowing revenues to
blind them to the systemic risks.
By this point, a cat-and-mouse battle
between traditional buy-siders and HFT
firms was in full swing. Buy-side traders
were already treading carefully on the
major stock markets, alert to the potential
risks of interacting with counterparties
deploying sub-millisecond technology to
front-run them.
Asset managers looked to alternative
trading venues, including dark pools, but
here too they could encounter danger,
especially when venue operators, typically
brokers, were less than transparent about
the identity of other participants.
“Ten years ago, the buy-side probably
relied too heavily on their brokers. But
the flood of questionnaires from plan
sponsors and other clients in light of ‘Flash
Boys’ gave trading desks a mandate to take
more responsibility, investing in staff and
technology to take control of their order
flow,” says Chris Jackson, global head of
equity strategy at Liquidnet.
Learning curve
Buy-side traders were embarking on a
decade-long learning curve, gradually
deploying faster technology, better analytics
and more granular data, driven partly by
greater regulatory and investor scrutiny.
Often dependent on the execution services
of still-conflicted brokers, they interrogated
post-trade execution performance to
identify more accurately where they could
execute large orders safely, and where they
were under greatest threat. Sometimes,
there was a balance to be struck, and the
risk was considered worth taking.
Better data and technology have enabled
more effective interaction, says Gregg
Dalley, global head of trading at Schroders
Investment Management.
“We still don’t have a consolidated tape,
but we have much better access to quality
data, particularly since MiFID II. When
trading in systematic internalisers (SIs), we
get a lot of granular data back on our fills
which feeds into our post-trade analysis.
HFT firms are a significant part of the
market now, not just in terms of liquidity
provision, but also execution services such
as algorithms as they look to diversify their
revenue streams.”
Access to more detailed data has not only
put the buy-side on more of a level footing
with newer market participants. The
“The buy-side has had to grow up. In the
past, there was a tacit understanding that
the sell-side would supply infrastructure and
other execution-related services. Today, if
you want it, you pay for it.”
CARL JAMES, GLOBAL HEAD OF FIXED INCOME TRADING, PICTET ASSET MANAGEMENT
insights that trading desks have achieved
through deeper analysis has increased
productivity through process automation
and streamlining. Both on they buy- and
sell-side, the economic realities of lower
margins and higher regulatory costs over
the decade have fuelled innovation.
Schroders now has a global equities team
of 17 traders located in four countries,
having previously employed more in
London alone, trading multiple times the
volume, covering more instruments in more
countries and far more investment teams
and portfolio managers (PMs).
“This is all down to technology,” says
Dalley. “In the time it took to pick up a
paper ticket when I first started, time stamp
it and pick up the phone to a broker, you
can now hit a single button that optimises
the execution strategy based on thousands
of data points and back testing, as well as
route, execute and book the trade. Traders
have had to evolve and embrace technology
and the progression has been amazing.”
Dalley’s traders must have a broader
set of skills, operating and understanding
beyond their specific field of responsibility
and collaborating with technologists
and data scientists to improve execution
performance.
“The quantitative execution research
team can back up traders’ hunches with
solid evidence, independent of selection
bias,” he explains. “By matching algos
to specific stock characteristics, we can
automate more trades in small size, whilst
the human traders focus on orders with
larger ADV (average daily volume) or
complexity.”
For Neil Joseph, head of equities trading
for EMEA at JP Morgan Asset Management,
the story of the decade is a shift from
automating execution to automating a wider
range of workflows between traders, PMs,
and the sell-side: “This has helped our
EMEA team to execute 50% more orders
per trader than three years ago, whilst
reducing trading costs by 20% over the
same period,” he says.
Examples of technology-assisted
workflow innovation include the automated
generation of targeted notices of liquidity
opportunities to relevant PMs and a
mechanism for flagging block trading
signals from the sell-side.
As at Schroders, Joseph’s team work
closely with dedicated technologists to
develop, test and implement incremental
improvements, then measure their impact
Issue 1
TheTradeNews.com
19