20 ★ Wednesday4 March 2020

Ewan Kirk

Markets Insight

AngloGold Ashanti,Barrick Gold nda
Newmont ere in demand as the Federalw
Reserve’s earlier than expected cut of its
main policy rate by a half a percentage
point helped pull gold prices higher.
Sectors seen as inflation hedges, such
as utilities and real estate, also turned
higher with mobile phone tower owners
American Tower nda Crown Castle
among the best performers.
But the worsening outlook for net
interest margins from lower rates
weighed on the banks.JPMorgan Chase
andCiti ere among those to retreat.w
ChipmakerAMD ose on the back of ar
Piper Sandler upgrade to “overweight”.
It forecast continued market share
gains fromIntel, saying its chips were
offering the same or worse performance
at higher prices in both the desktop and
server end markets.
Tesla limbed after JMP Securitiesc
turned positive with the broker arguing
that sustainable growth over the next
four to five years justified an enterprise
valuation of 5 times 2021 revenue.
Target aded after the retailer gavef
more cautious 2020 profit guidance.
Stealth BioTherapeutics oubled afterd
US regulators said its lead drug to treat a
rare paediatric heart condition was
eligible for priority review.Bryce Elder

Wall Street Eurozone London

Qiagen ed the Stoxx Europe 600 gainersl

afterThermo Fisher greed at the seconda

time of asking to buy the Dutch

diagnostic equipment maker.

US-based Thermo agreed to pay

$10.1bn, a 23 per cent premium to

Qiagen’s closing valuation on Monday.

Lufthansa ed a bounce among travell

stocks as the US Federal Reserve’s rate

cut encouraged bargain-hunting.

ButThyssenkrupp eakened onw

worries that Covid-19 would stall its debt

reduction plan with NordLB analysts

saying improved dividends looked more

unlikely from the steelmaker in spite of

the recent sale of its elevator division.

Spanish drugmakerPharma Mar

surged after saying it was investigating

whether one of its compounds was an

effective treatment for the coronavirus.

Clariant ose afterr Sabic f Saudio

Arabia, the Swiss speciality chemicals

group’s biggest shareholder, raised its

stake from 22 per cent to 31.5 per cent,

just below the level where it would trigger

a mandatory takeover offer.

AB InBev rifted lower after Bank ofd

America turned cautious on the brewer.

Macroeconomic headwinds and

coronavirus disruption made progress

unlikely against challenging year-on-year

comparatives, it said.Bryce Elder

British Airways ownerIAG as the bestw

performing blue-chip stock.

HSBC, repeating “buy” advice on

airlines including IAG and Ryanair, said it

did not foresee “existential doubt” for the

sector under base-case coronavirus

scenarios.

HSBC,Barclays,Standard Chartered

andLloyds ll slipped as investors priceda

to near-certainty a round of central bank

actions to follow the US Federal Reserve’s

half percentage point interest rate cut.

Wm Morrison ed the supermarketsl

lower after industry market share data

showed it was the laggard among the big

four in February.

4 Imprint, the US-focused promotional

products maker, led the FTSE 250 gainers

after delivering in-line full-year results

and saying rading had been strong.t

EngineerRotork allied after matchingr

expectations with its full-year results as

improved margins counteracted a small

shortfall in sales.

Hunting ell after UBS took the oilfieldf

equipment maker off its “buy” list, citing

lower US shale exploration budgets.

Huntsworth, the healthcare public

relations agency, jumped to its best level

in more than a year after agreeing to a

£400m takeover by private equity group

Clayton, Dubilier & Rice.Bryce Elder

3 Wall Street jolted by a double dose of
policy announcements
3 Lingering coronavirus concerns send
US Treasury yields to fresh record lows
3 Haven asset gold rallies above $1,
an ounce

Markets were buffeted by back-to-back
policy statements yesterday, prompting a
retreat and then spike in US stocks.
Wall Street opened lower after G
officials stopped short of announcing
concrete action to address the economic
impact of the coronavirus outbreak.
Following a conference call yesterday,
the group of nations issued a statement
stating a “commitment to use all
appropriate policy tools to achieve
strong, sustainable growth and safeguard
against downside risks”.
But an hour later, the US Federal
Reserve caught markets off-guard by
cutting its main policy rate by half a
percentage point, its largest cut since
2008, citing “evolving risks” from the
spread of the contagion.
The Fed move came “as a big surprise”,
said Neil Birrell, chief investment officer
at Premier Miton. “Cuts were already
discounted but not so much so soon.”
The large-cap S&P 500 index shot up
after the Fed announcement, only to
retreat 30 minutes later as investors
weighed the impact of the move.
“Interest rate cuts cannot make
factories produce more if they are unable

to source vital inputs from virus-hit

countries,” said Jennifer McKeown, head

of global economics at Capital Economics.

By midday, the S&P 500 was down 1.

per cent while the tech-heavy Nasdaq

Composite fell 1.3 per cent.

European stocks, which had been up

more than 3 per cent, gave up some gains

after the Fed announcement. The Stoxx

Europe 600 closed 1.4 per cent igher.h

Asian markets were also volatile with

early optimism stemming from a Monday

rebound in US stocks fading. Tokyo’s

Topix fell 1.4 per cent, having been up 1.

per cent, while the CSI 300 index of

Shanghai- and Shenzhen-listed stocks

rose 0.5 per cent higher, retreating from a

1.3 per cent rise earlier in the session.

Reflecting the cautious mood, haven

assets rallied. Gold climbed more than 3

per cent to breach $1,600 an ounce while

the yield on the 10-year US Treasury fell 6

basis points to hit a fresh record low of

1.0199 per cent.

Global oil benchmark Brent crude,

which had recently fallen below $50 a

barrel, rose 0.8 per cent to $52.31 a barrel.

Ray Douglas

What you need to know

Treasuries rally after Fed rate cut fails to allay virus anxiety

-year Treasury yield ()

Source: Bloomberg











Mar Mar





The day in the markets

Markets update

US Eurozone Japan UK China Brazil
Stocks S&P 500 Eurofirst 300 Nikkei 225 FTSE100 Shanghai Comp Bovespa
Level 3073.25 1487.26 21082.73 6718.20 2992.90 107101.
% change on day -0.55 1.23 -1.22 0.95 0.74 0.
Currency $ index (DXY) $ per € Yen per $ $ per £ Rmb per $ Real per $
Level 97.660 1.118 107.555 1.282 6.981 4.
% change on day 0.308 0.449 -0.343 0.549 0.332 -0.
Govt. bonds 10-year Treasury 10-year Bund 10-year JGB 10-year Gilt 10-year bond 10-year bond
Yield 1.043 -0.627 -0.116 0.387 2.787 6.
Basis point change on day -4.800 0.000 2.490 -1.700 2.400 -9.
World index, CommodsFTSE All-World Oil - Brent Oil - WTI Gold Silver Metals (LMEX)
Level 348.05 52.31 47.43 1599.65 16.92 2639.
% change on day -0.16 -0.68 -0.08 -0.63 -1.57 1.
Yesterday's close apart from: Currencies = 16:00 GMT; S&P, Bovespa, All World, Oil = 17:00 GMT; Gold, Silver = London pm fix. Bond data supplied by Tullett Prebon.

Main equity markets

S&P 500 index Eurofirst 300 index FTSE 100 index

| |||||| ||||||||| ||||

Jan 2020 Mar

2880

3040

3200

3360

3520

| |||||||||||||||||||

Jan 2020 Mar

1440

1520

1600

1680

1760

| ||||| |||||||| ||||||

Jan 2020 Mar

6400

7040

7680

Biggest movers
% US Eurozone UK

Ups

Newmont 6.

Trane 6.

Lennar 4.

Welltower 4.

Vulcan Materials (holding ) 4.

Lufthansa 8.

Lindt 5.

Airbus 4.

Novo Nordisk 4.

Telenor 3.

Int Consolidated Airlines S.a. 7.
M&g 4.
Just Eat Takeaway.com N.v. 4.
Legal & General 4.
Evraz 4.
%

Downs

Hanesbrands -6.

Charles Schwab (the) -5.

Svb Fin -5.

Comerica -5.

Raymond James Fin -4.

Prices taken at 17:00 GMT

Seadrill -6.

Thyssenkrupp -4.

Unicredit -4.

Natixis -3.

Societe Generale -3.

Based on the constituents of the FTSE Eurofirst 300 Eurozone

Barclays -2.

Hargreaves Lansdown -2.

Morrison (wm) Supermarkets -2.

Standard Chartered -2.

Hsbc Holdings -1.

All data provided by Morningstar unless otherwise noted.

H

ardlyadaygoesbywithout

investors being told that

artificial intelligence will

revolutionise investment

management. After all, AI

is being hailed as a way to enhance

image recognition, healthcare, movie

recommendations, fake news and even

thehumbletoothbrush.

Surely it is only a matter of time

before AI allows investors to sit at home

getting rich while watching recom-

mendedmovieswithsparklingteeth?

Despite that persistent hype, reality is

different. The best way to distinguish

thetwoistoconsiderwhatAIactuallyis.

It is useful to take any statement using

AIandsubstitutetheword“statistics”.

“The UK government vows to revolu-

tionise the NHS with artificial intelli-

gence” sounds somewhat less trans-

formative when one says: “The UK gov-

ernment vows to revolutionise the NHS

with statistics.” Buying a “statistics-ena-

bled”toothbrushsoundsmoreprosaic.

So how can we expect AI to help the

investment process? The caricature is

that one takes a bunch of data (prefera-

bly newfangled “alternative” data),

throws it at some sort of “neural net”

andoutpopsavaguelydefinedfinancial

gold mine. This misconception drives

seriousstatisticianscrazy.

First of all, machine learning requires

a clear goal. That was whatGoogle’s

legendary AlphaGo programme had

when in 2016 it finallybeata human

championattheboardgameGo.

But what is the goal of finance and

investment? Higher returns over time?

A two-times levered position in the

equities markets will give you two times

the return. Is it higher risk-adjusted

returns? Adding diversifying assets like

bondstoyourportfoliogivesyouthat.

My personal story that illustrates the

pitfalls of AI came in 2010 when I

became interested in genetic algo-

rithms,whichusethepowerofselection

andbreedingto“evolve”.

I wrote a library of functions to evolve

trading systems and allowed hundreds

of thousands of artificial traders to

breed. Eventually I had a huge popula-

tion of artificial traders who were doing

what many quantitative funds have

been doing for decades — which was

preciselynotwhatIwanted.

Consequently, we have to tell the AI

system something like: “Do not find me

the returns that everybody knows

about; just the subtle unknown ones.”

But this is hard to specify. Even if one

can accurately specify the desired

returns,itleadstoasecondproblem—if

the effect is subtle, it is likely to be small

or shortlived and thus hard to exploit at

scale. Across the vast global investment

industry,onlyatinyproportionoffunds

willthereforeeverlikelybenefit.

A third problem is that an AI system

learns from the past. In this respect, it is

no different from any other systematic

ordiscretionaryinvestmentprocess.

But the roblemp with finance is that

thepastisnotagoodguidetothefuture.

To use a statistical term, finance is not

“stationary”. In most AI domains like

movie recommendations and tooth-

brushing, the “target” is stationary and

the environment does not change much

(unlessyou’vehadalotofdentalwork).

This is ot true in finance. There is an

trade-off between reacting quickly to

shifts in markets and believing that old

patternswillreassertthemselves.

While one may wish an AI system to

respond rapidly to events, this eans itm

has to build a model on a very short his-

tory, which reduces the amount of data

that the system can learn from. Tough

choiceshavetobemade.

Finally, financial data is very messy.

Although it is not entirely random, the

signal-to-noiseratiois ow.l

InfieldswhereAIhasbeensuccessful,

this is typically not the case. AlphaGo,

for example, knew exactly where the

pieces on the board were. Nobody

chooses19randommoviesforeveryone

they like and then expectsNetflix ot

comeupwithgoodsuggestions.

Itispossibletousesophisticatedtech-

niques to reduce the effect of random-

ness in finance but it makes it challeng-

ingtoapplymachinelearning.

It is not all doom and gloom. While it

is unlikely that AI will create new scala-

ble sources of returns, it is proving use-

ful in more mundane tasks. AI is very

good at cleaning data and etectingd ea-f

turesingiganticdatasets,forexample.

One technique that has gained trac-

tion is to use the same AI algorithms

usedincomputergamestocreaterealis-

ticNon-PlayerCharacters,likethemon-

sters that try to kill you. These algo-

rithms can approximate how a human

trader would act in particular circum-

stances(withoutkillingyou,ofcourse).

Oncewearethroughthetroughofdis-

illusionment,managers will find many

places where AI can generate marginal

improvements.Buttheywill ecleaningb

theirteethwithanold-fashionedbrush.

Ewan Kirk is founder and chief investment

officer of Cantab Capital Partners, now part

of GAM Systematic

Don’t believe the

hype about AI and

fund management

We have to tell the AI

system something like: ‘Do

not find me the returns

everybody knows about’

MARCH 4 2020 Section:Markets Time: 3/3/2020- 18:34 User:stephen.smith Page Name:MARKETS2, Part,Page,Edition:EUR, 20, 1