For Financial Intermediary, Institutional and Consultant use only. Not for redistribution under any circumstances.

Research Insights at Schroders: – Climate change – AI within Investments – Venture capital – Private Securitized loans – The quarterly reporting conundrum – Multi-Asset and the future of portfolio construction

Investment HorizonsFall 2018

Perspectives on inescapable truths

We are pleased to off er Investment Horizons, our perennial compilation of research articles inspired by our global client engagements.

In this edition, we focus on helping you navigate the current late market cycle, that is, managing the “inescapable truths” of low prevailing yields, divergent policy, and extended valuations across the public equity markets.

But, we know that many of you are also faced with other types of non-fi nancial challenges. Social change has never been more infl uential, climate change is an unavoidable topic for us as investors and as human beings, and the volumes of publically available meta-data has increased the level of competition, and the need for greater risk-awareness, across all industries.

Looking ahead, the world is clearly changing, and we hope this edition provides some differentiated insights to help manage today’s “truths.”

Whether it be regulatory mandates, social and political risk, or technological innovations, our investment teams at Schroders will continue to look for ways to conduct business in a more effi cient manner, extract alpha in a differentiated way and be better stewards of client capital.

As always, if there is anything you would like to discuss further, please contact your local Schroders representative.

Contents

4Climate change: The forgotton physical risks – by Andrew Howard and Marc Hassler

Most climate analysis focuses on the impacts of steps to limit temperature rises, such as carbon prices or clean energy investment. Physical risks, on the other hand, have received less attention. Our analysis examines the implications in this regard for companies and investments.

13The opportunity in early stage venture capital– by Steven Yang, Michael McLean and Duncan Lamont

Venture capital provides access to investment opportunities in game-changing companies with the potential to disrupt entire industries…but only if done correctly, and early.

28Take nothing for granted: Portfolio construction in today’s great unknown environment – by Remi Olu-Pitan and Clement Yong

They say you can’t have your cake and eat it too. Yet, over the last few years, investors have enjoyed a fantastic mix of high returns, low volatility and low correlations. All of these factors have been tailwinds to portfolio construction. But going forward, crucially questioning our assumptions, and taking nothing for granted may be a much more prudent philosophy.

9Humans or robots: Who’s really driving the car? – by Mark Ainsworth

Artifi cial intelligence (AI) is an increasingly popular topic, particularly for the fi nance industry. However, it is our belief that Intelligence Augmentation (IA) is the real source of innovation in the industry.

20Investing in today’s Commercial Real Estate market: It’s all about that basis – by Michelle Russell-Dowe and Jeff Williams

Today, the US Commercial Real Estate market is characterized as a market with high property valuations. But, by going to private, middle-market areas investors can fi nd much lower cost basis, and higher total return prospects.

25Less can be more: How to improve fi nancial reporting –by David Knutson

While recent tweets have sparked the debate on the usefulness of public company’s quarterly reporting requirement, we believe the crux of the debate is more about that it’s too much information, often times it’s not the right information,and it costs a lot to produce and process.

Climate change: The forgotten physical risks

Most climate analysis focuses on the impacts of steps to limit temperature rises, such as carbon prices or clean energy investment. Physical risks, on the other hand, have received less attention. That oversight is remiss; the impacts are lower, but they are also more certain. Physical damage lags atmospheric CO2 concentrations, which have been rising for decades, and will drive greater disruption even if emissions fall now. Our analysis examines the implications for companies and investments.

Marc HasslerSustainable Investment Analyst

Andrew HowardHead of Sustainable Research

Over the last year, we have invested heavily in developing tools to help our analysts, fund managers and clients better understand the threat climate change poses. No single measure can capture the breadth of climate impacts, but combined they provide a rounded view of the challenge ahead. We have described our analysis of the investment implications of these impacts in previous research, concluding that up to 20% of the value of global listed companies could be exposed to climate risks1.

Physical costs are risingThe science linking levels of greenhouse gases (GHGs) in the atmosphere with global temperatures and, as a result, more volatile and damaging weather patterns, is clear. Temperature rises lag increases in GHG concentrations in the atmosphere by around 40 years, meaning that even if emissions stopped tomorrow, the earth’s average temperature would likely rise by a further 0.6 degrees2. Further disruption from the effects of changing weather patterns therefore looks unavoidable, meaning bigger risks to physical assets and infrastructure. Figure 1 graphically illustrates the close correlation between temperature rises and that disruption.

Figure 2 plots Munich Re’s estimate of the annual costs associated with climate damage. Both the level and uncertainty of climate costs have risen over recent decades. We believe larger impacts are very likely in the future.

1 www.schroders.com/en/lu/professional-investor/features/climate-change dashboard

2 https://earthobservatory.nasa.gov/blogs/climateqa/would-gw-stop-with-greenhouse-gases/

Quantifying the extent to which individual companies and portfolios are exposed to physical climate risks is an important element of preparing for a more challenging future.

Figure 1: Greenhouse gas emissions, temperatures and disasters are all rising together

-100

0

100

200

300

400

500

1950 1960 1970 1980 1990 2000 2010 2020

Indexed, 1950-current Average = 100

Temperature change Greenhouse gas emissionsNumber of disasters Economic damage

Source: EM-DAT, NASA, Schroders and UN FCC. Based on most recent data available in May 2018.

Figure 2: Weather-related losses caused by climate change are growingIndexed weather losses

Indexed weather losses Standard deviation of losses over previous 20 years

0

50

100

150

200

250

300

201620122008200420001996199219881984198010

20

30

40

50

60

Source: Munich Re and Schroders. Based on climate costs published in 2018.

Climate change: The forgotten physical risks4

Figure 3: Sustainability has moved up corporate agendas

2008

Top

5th

2018

Economic Geopolitical Societal Environmental

Source: World Economic Forum 2018 Global Risks report.

In that context, the limited attention investors have paid to physical risks seems remiss. Climate risks are primarily viewed through three lenses: regulation, fossil fuel exposure and clean energy growth. All are important but rely on action to combat climate change that is far from assured. Physical risks, on the other hand, are likely to be unavoidable.

Businesses recognize the threatGlobal business leaders recognize the risk climate change poses to their operations. The World Economic Forum’s annual survey of business leaders, which asks them to identify the biggest risks they face, refl ects growing concerns over environmental challenges and climate change in particular. Whereas a decade ago none of the top fi ve risks refl ected either social or environmental trends, this year’s survey clearly refl ected business leaders’ understanding of the scale of the challenges ahead. Four of the top fi ve risks are environmental (Figure 3).

Similarly, companies’ responses to the annual survey run by CDP, a charity that promotes carbon disclosure, show that there is widespread appreciation of the strategic risks physical damage poses. The most recent responses to questions on physical climate risk are summarized in Figure 4 below. Around 80% of the over 2,500 companies canvassed identifi ed risks stemming from the disruptive impacts of physical climate change, with capital-intensive industries generally most aware. Companies also recognize the urgency of the challenge, expecting risks to crystallize within four to fi ve years across all sectors.

Interestingly, companies in the energy sector – which our analysis highlights as the most exposed – show relatively limited recognition of physical climate risks. We will continue to press companies in the industry to address the effects of climate change and formulate strategies to mitigate these risks. The views of companies in most other sectors are broadly in line with our fundamental analysis.

Figure 4: Energy aside, most sectors seem well aware of climate change risks

Based on responses to a question in the most recent CDP survey where companies were asked: “Please describe your inherent risks that are driven by changes in physical climate parameters”. Gradual risks include the responses: change in mean (average) precipitation; change in mean (average) temperature; change in precipitation pattern; change in temperature extremes; and sea level rise. Disruptive risks include: change in precipitation extremes and droughts; induced changes in natural resources; other physical climate drivers; snow and ice; tropical cyclones (hurricanes and typhoons); and uncertainty of physical risks. Multiple responses are possible. Companies were also asked about the timeframe for the risks they identifi ed. We have plotted simple averages for each sector. Source: CDP and Schroders, latest data available as of June 2018.

Proportion of companies in each sector (%)

Gradual risks Disruptive risks Risk not mentioned

0 20 40 60 80 100

FinancialsUtilities

Consumer StaplesMaterialsTelecom

IndustrialsIT

Consumer DiscretionaryEnergy

Health Care

Expected timeframe of impact (years)

3.5 4.0 4.5 5.0 5.5

60

65

70

75

80

85

90

95

100Financials

UtilitiesConsumer Staples

Telecom MaterialsIT

Consumer Discretionary

Healthcare

Industrials Energy

% companies expecting impact

Climate change: The forgotten physical risks 5

Analyzing the impact on individual companiesWe have developed an objective framework to assess the valuation implications of companies’ exposures to the risks of physical damage caused by climate change. The analysis is grounded on the premise that – in theory – companies could insure themselves against such risks. We have estimated the cost of buying a 13-year insurance policy to cover climate risks and plotted it against companies’ enterprise values (the 13-year policy life refl ects our estimate of the average remaining life of a typical company’s assets).

Figure 5: Translating national risks into company exposures

Objective measures of climate damage in each country

Companies’ exposures to those climate risks

Project growth in damage as climate impacts escalate

Enterprise value adjustment for physical climate risks (%)

-5

-4

-3

-2

-1

0

Oil

& G

as

Util

ities

Basi

c Re

sour

ces

Trav

el &

Lei

sure

Real

Est

ate

Auto

mob

iles

& P

arts

Tele

com

mun

icat

ions

Cons

truc

t. &

Mat

eria

l

Chem

ical

s

Reta

il

Ind.

Goo

ds &

Ser

vice

s

Food

& B

ever

age

Med

ia

Hea

lthca

re

Pers

& H

ouse

hld

Goo

ds

Tech

nolo

gy

Estimate cost of insurance for life of companies’ assets

Indexed values of USD losses

Indexed weather lossesStandard deviation of losses over rolling 20yr periods (RHS)

0

50

100

150

200

250

300

201620122008200420001996199219881984198010

20

30

40

50

60

We examine the damage every country has suffered from the physical effects of climate change, using data from the NGO Germanwatch, which tracks the damage caused by climate change relative to the GDP of each country. We estimate regional risk levels based on constituent countries, weighted by GDP.

We translate companies’ reported geographic exposures into consistent country names, or regions if country exposures are not disclosed. By mapping country or region risks to companies’ geographic exposures, we estimate the average exposure companies currently face to climate damage.

We project the expected damage in each country using Munich Re’s historical data, which provides the basis to estimate how company damage will rise over the remaining life of their assets (typically 13 years).

Using standard insurance industry loss ratios, we estimate the annual costs companies would face each year to insure against climate damage to their operations based on the risk they face. We discount that payment stream to estimate the equivalent one-off payment covering the life of their assets, which is compared to current enterprise values.

Our methodology is illustrated in Figure 5 and explained in more detail in the appendix, along with the assumptions used. By putting a price on the cost of “neutralizing” climate damage we create an objective assessment of the impact on corporate valuations. We realize this analysis is more theoretical than practical – we don’t know of any companies that have taken out multi-decade climate insurance – but the approach provides a robust way to gauge physical risk exposures, and is commonly used to answer other investment questions1.1 The approach is similar to the “no arbitrage” approach often used to value

fi nancial instruments and risks, particularly in derivatives markets.

Source: Germanwatch, Munich Re, Schroders estimates and calculations. All based on most recent data as of June 2018. For illustrative purposes only. Does not refl ect any actual portfolio nor transaction.

Climate change: The forgotten physical risks6

In addition to measuring individual company exposures, we have also developed tools to help fund managers gauge the risks facing their portfolios.

Combining company exposures with portfolio weights provides a high level view of risk, whether intended or not.

An illustration of the scale of potential risk facing global equities using the same estimated insurance costs/enterprise value formula is shown by the effect on six major world indices in Figure 7.

Figure 7: Overall risk exposures can be gauged from the impact on equity indicesEnterprise value adjustment for physical climate risks (%)

-2.0

-1.5

-1.0

-0.5

0.0

TOPIX100

MSCIWorld

MSCIACWI

S&P500FTSE 100MSCIEmergingMarkets

Source: MSCI, Schroders. Based on most recent data available in March 2018

Applying our analysis to around 11,000 listed companies across the world provides a consistent and objective basis to assess fi rms’, sectors’ and portfolios’ exposures to physical climate risk. Figure 6 ranks companies’ exposures to physical climate risks in different sectors based on their estimated insurance costs divided by their enterprise value. Predictably, capital-intensive sectors operating in more vulnerable parts of the world face the biggest impacts, whereas those with asset-light business models are least exposed.

Figure 6: Sector exposures tend to vary according to the capital intensity of the business

Enterprise value adjustment for physical climate risks (%)

-5

-4

-3

-2

-1

0

Oil

& G

as

Util

ities

Basi

c Re

sour

ces

Trav

el &

Lei

sure

Real

Est

ate

Auto

mob

iles

& P

arts

Tele

com

mun

icat

ions

Cons

truc

t. &

Mat

eria

l

Chem

ical

s

Reta

il

Ind.

Goo

ds &

Ser

vice

s

Food

& B

ever

age

Med

ia

Hea

lthca

re

Pers

& H

ouse

hld

Goo

ds

Tech

nolo

gy

Source: Schroders. Based on most recent data available in March 2018.We have excluded fi nancial sectors from this summary given the low direct exposure of their fi xed assets understates the risk embedded in their assets or liabilities.

ConclusionThe damage infl icted by climate change through increasingly volatile weather patterns is rising quickly and is already signifi cant for many companies. Despite being far more certain than risks stemming from actions and policies to limit its eff ects, physical damage receives far less

attention from investors than analysis of mitigation eff orts. Our proprietary framework assesses companies’ potential exposures to physical climate risks, helping inform the decisions of analysts and fund managers, as well as gauging the exposures facing the portfolios they oversee.

Climate change: The forgotten physical risks 7

AppendixPhysical risk exposure is calculated by combining country-level risk measures with companies’ reported geographic footprints:

– Companies report the amount of their assets in different locations. Depending on the level of granularity provided, we map these locations on to a standardized list of countries or regions. Where companies do not disclose their geographic locations, we assume all of their assets are located in the company’s domestic market.

– Separately, we calculate the current level of damage climate changes causes in each country using the costs/GDP ratios calculated by Germanwatch, an NGO (non-governmental organization), based on Munich Re data. These ratios are averaged over 20 years, limiting the sensitivity of the analysis to annual fl uctuations. We also calculate regional exposures by weighting country costs according to the GDP of the country in question.

– We calculate the expected annual damage to each company’s assets using national or regional average cost ratios, companies’ reported tangible assets and the geographic distribution of those assets.

– We extrapolate the multi-decade trend in climate damage (the global climate damage/GDP ratio has grown by 4.1% annually since 1980) to project expected damage up to 2030, a period which refl ects the approximate remaining life of the average company’s tangible assets.

– We estimate the costs of insuring against this expected damage using global average insurance industry loss ratios (around 0.6).

– We discount the future cost of the modelled insurance premia using 10-year US Treasury bonds to approximate risk free rates.

– We compare the present value of the modelled insurance premia to companies’ current enterprise values to gauge the impact on fi rm valuations.

– As with all modeling, such outputs are theoretical in nature and must be considered as no more than an approximate representation of performance, not as indicative of how it would have performed in the past, or will perform in the future. Simulated data are the result of statistical modeling, with the benefi t of hindsight, based on a number of assumptions and there are a number of material limitations on the retrospective reconstruction of any performance results from performance records. In addition, climate-related outcomes are extremely diffi cult to model, and there can be no assurance of how future climate patterns will prevail. There can also be no assurance that this performance could actually have been achieved using tools and data available at the time. No representation is made that the particular combination of investments would have been selected at the commencement date, held for the period shown, or the performance achieved. This data is provided to you for information purposes only as of the dates of this material and should not be relied on to predict possible future performance.

Climate change: The forgotten physical risks8

Humans or robots: Who’s really driving the car?

Artifi cial intelligence (AI) is an increasingly popular topic, particularly for the fi nance industry, but despite recent advancements many applications of this extraordinarily expansive area of computer science are still very much in their infancy. This is particularly the case when attempting to apply AI to Asset Management. It is our belief that Intelligence Augmentation (IA) is the real source of help to the industry, which uses AI to help humans make decisions, rather than making the decisions for them. Here, we explore the ways in which human led data science can be used in a sustainable and repeatable way to assist the fundamental asset management industry.

IntroductionAI is an increasingly popular topic. Coverage of recent autonomous driving road tests across the world has only fostered the excitement and the imagination of what is to come. From the consumers’ perspective, these and other AI advancements are an advert of how technology might make the future look. Within the investment industry, naturally these innovations have created intrigue as to whether such systems will eventually replace human portfolio managers. ‘Robo-advisers’ already exist, providing basic fi nancial advice based on algorithms that are calculated from the questions that are being asked. Computing power continues to grow and a vast library of historical market information is at our disposal from which to attempt to predict future outcomes with greater accuracy. Is AI-driven asset management that far off?

For all the hype, we believe that some reservation is warranted before we give our fi duciary responsibilities (and our capital) to the robots.

The engine behind AI is “machine learning”, the use of statistical algorithms and techniques autonomously to learn, and systematically “improve,” outcomes for a given task without any explicit programming or additional human-led insight. One of the most common commercial applications of machine learning is “predictive analytics.” AI is a heterogeneous computer science and there are a number of different types of predictive analytics systems, but it is possible to compare the use of IA and AI in driving to how it could also be applied to investing.

License and registration, please Conceptually driving a car and investing have some similarities – both involve getting from A to B as quickly and easily as possible without taking too many risks. In the case of investing, ‘B’ (the destination) can be capital preservation or retirement planning.

In the fi eld of autonomous driving, the type of AI used is called an Autonomous Driving Platform (ADP). While the ADP – the AI component – is technically what simulates the human, that is “drives” the driverless car, it is in fact the various external factors (Component 01 in Figure 1 on the next page) being instantaneously and continuously collected which deliver the critical component in the driving system. These external inputs are fed into the ADP (Component 02) which, along with a pre-loaded guidance database, produce a synthetic driving behavior similar to that of a human driver, such as adherence to traffi c signals, accelerating or deceleration, and responding to obstacles in or near the path of the car.

This is referred to as a Perception-Action Cycle. What makes the ADP unique is that all of the information is stored and constantly re-evaluated in order to better refi ne the external output. That is, the system “learns” how to be a better driver with each mile driven.

What determines the quality of learning is the quality of the input. When it comes to obtaining optimal AI outputs, we identify fi ve keys computation parameters: 1) a constant environment where the rules are fi xed and don’t change 2) the data is digital, quantifi able 3) abundant data (this could vary by industry) 4) low uncertainty, and 5) clear objectives. In our view, these elements are necessary for AI to succeed.

Mark AinsworthHead of Data Insights,Investments

Humans or robots: Who’s really driving the car? 9

Shor

t ran

gera

dars

02

03

01Autonomous

driving platform

AI-based function inautonomous vehicles

In-vehicle data collection &

communicationsystems

Control policy makes decisions

Memorydatabase

Input

OutputAction data

Shor

t-ra

nge

LiD

ARs

Trifo

cal

cam

eras

Long-range

LiDARs

Electronically

scanning

radars

GPS

antennas

Dedicated short

range communications

antennas

Supe

rco

mpu

ters

Data Communications

Systems

Automated drive • Object detection • Voice recognition• Speech recognition • Gesture control • Eye tracking • Driver monitoring •Virtual Assistance • Mapping Systems • Safety systems

AI computing systems like IBM’s Watson, and Deep Blue, and Google’s AlphaGo have been highly publicized gaming successes in their own rights, and in this domain AI has performed well. Some hedge-funds also employ algorithmic AI trading strategies which seek to ‘predict’ patterns in short-term market trends and exploit these trends through thousands of trade executions.

However, what makes these systems so powerful in areas like gaming and problem solving is that they are, for the most part, bound within a closed informational universe. Games have rules (a constant environment), little to no uncertainty around these rules, a clearly defi ned objective, and there is a calculable (albeit overwhelming) amount of quantifi able moves, countermoves, and so forth that can be drawn upon. But what happens when the rules change, or the context of the inputs change?

While AI platforms make for great headlines, the reality is many produce incorrect outputs and almost all of them require some additional human intervention – coding and algorithm amendments – in order to function properly. If we refer back to those fi ve parameters, most AI systems (especially investment-driven programs) would check ‘no’ for adherence to the entirety of them.

IA on the other hand, has been around for about the same time as AI, has a much more ubiquitous track record and can be observed in virtually all areas of our daily lives.

IA is simply the enhancement of intelligence through technological means.1 Going back to our automobile example, cars have had features that augment the human’s abilities for many decades and include everyday features like mirrors, temperature warning lights or speedometers that provide timely information about the state of the car. These are all designed to help the human drivers of the vehicles make better decisions about how and where to drive. Modern cars are full of much more sophisticated features too: seatbelt sensors, parking sensors, blind spot collision alerts and of course, satellite navigation (or sat-nav) are modern IA examples that are meant to help make us better drivers.

What’s in an acronym: How AI and IA work together As investors, IA is a much more relevant area of science than AI on its own as it enables us to extract insights that few others can even identify – even with the data being there in plain sight. This has tremendous advantages when it comes to fundamental investing. Note that many of the driving examples above allow the driver to ‘see’ things that are otherwise hidden or hard to perceive – (is there a car in my blind spot?, is there traffi c on the route of my journey?) You can drive your car without this information, but you will drive it to your destination more quickly and safely if you have fi lled in these blind spots.

1 For purposes of this paper we have excluded physical IA like medical prosthetics, etc.

Perception-Action Cycle example for autonomus driving models

Case Study #1: Electronic gambling: Seeing the potholes several miles ahead

Similar to the ADP mechanism, one way future investment insights can be developed is through geo-spatial data combined with predictive analytics. In October 2017, the UK Government announced a consultation on changing the regulation regarding the maximum stake available on fi xed odds betting terminals. Fixed odds betting terminals are electronic gambling machines which are normally situated in betting shops. At the time the maximum stake on a single bet was £100, but the consultation was reviewing proposals to lower the betting to either £50 or £2. UK listed gambling companies such as William Hill stated their belief that they would have to close shops if the

maximum stake was reduced as it would affect their ability to generate revenue at individual stores. William Hill estimated that 70% of their net revenue came from fi xed odds betting terminals around this point in time.

To help determine the potential impact, the Schroders Data Insights Unit (DIU) and European Equities investment team at Schroders developed a model using data science techniques to approximate the number of shop closures that William Hill might face if the regulation for fi xed odds betting terminals was changed to allow a maximum stake of £2. Using location data for all high street betting shops mapped with population

Humans or robots: Who’s really driving the car?10

Any fund manager considering their investments has access to many useful pieces of information about a company – its fi nancial state, its revenues, the stated plans of its management. But there are other important things that investors do not currently have access to through traditional channels. Examples of questions that traditional information doesn’t answer include:

− What do consumers really think about the brand? − How are consumers’ opinions being infl uenced by the

company’s strategic initiatives or by a scandal that might have occurred?

− What demographic group does it currently most appeal to, and is its growth going to be achieved by focusing on them or expanding to other groups?

− Are its retail outlets located in the areas with those types of consumers within driving distance?

Consumer brand companies will be able to answer those questions, as it is central to how they plan to achieve their corporate strategic goals. Such companies have whole departments busily mapping the country, analyzing data

on local populations, running surveys and analyzing loyalty card data to have a clear and timely view. However most investors only get to see fragments of this information in company reports and earnings guidance, rather than a complete picture of that company and their sector.

But if the datasets that can fi ll these blind spots are publically available, as they are sold by research agencies, published by governments and circulated by industry bodies, very little is proprietary to the asset manager itself. Why then is this a blind spot? As previously mentioned in our case study, the issue is that the datasets are far too big and too messy for an investment professional to cope with. Turning the data into the insight to fi ll the blind spot requires deep expertise and specialism in statistical methods, in data engineering, and in mathematical modeling. Imagine knowing that you can see the GPS traces of millions of phones when really you just want to know if the highway has a traffi c jam. You need someone to create an IA interface for you, something that boils down all that data into something designed purely to augment your intelligence, to help you make a better decision. This is what Schroders DIU seeks to do.

data, Schroders were able to model different outcomes if there was a change in legislation to determine how many shops would survive relative to the local population.

The eventual estimate by our data scientists was the closure of 929 William Hill shops if the £2 maximum stake was imposed. As a result, the European Equities investment team decided not to invest in the stock at that time as it appeared that the market had already

priced in that either no legislation change would be recommended or guidance from the consultation would not be implemented and the maximum stake would remain £100.

After the consultation released its recommendation for the maximum stake to be changed to £2, the UK government announced that it would legislate to this effect and the maximum stake on a fi xed odds betting terminal would be reduced. Having completed their own analysis, William Hill issued guidance that they would need to close approximately 900 shops.

In this instance, we were able to utilize a variety of data science techniques to create insights that might not be feasible by using traditional research analysis while the investment decision remained with the investment team.

Case Study #2: The Athleisure Trend: Threading the needle on guidance and price sensitivity

Brand perception and trend analysis are signifi cant factor for businesses, and investors. In 2017, after Footasylum (a UK athleisure retailer) announced poor earnings in Q2 2017, the market had become nervous that the whole athleisure trend was beginning to wain and that this was a trend that would affect all athleisure companies – including other retailers in the sector. Schroders UK investment desk wanted to investigate the contagion risk of declining consumer interest in the athleisure segment as a whole, but also the brands that comprise these retailer consignment inventories (such as Adidas, Converse, Nike, Puma, Reebok) as the investment desk held a position in one of Footasylum’s competitors. Schroders DIU had identifi ed and sourced data that provides an indication of consumer sentiment towards

companies and brands, however the data set is large and unwieldy and it would be impossible to gain insights from it without the use of data science techniques.

Schroders DIU was able to provide consumer sentiment reports showing that there had been no drop in purchase consideration for the athleisure brands by UK consumers and no drop in sentiment towards the brands that many of these retailers sell. This confi rmed the investor’s faith in the fundamentals of the company they held and that the wider sector was not in trouble. This directly prevented any precautionary reduction in the holding. Upon the earnings announcement for Q3 2017, company results for this holding did not disappoint, and the stock of our company to rose 14% on the day.

Humans or robots: Who’s really driving the car? 11

Returning to the fi ve conditions for AI tools to succeed, we believe that long-term investing will remain a human task, unlike some self-driving initiatives that may achieve more imminent success.

1. Constant environment: This isn’t the case for investing. Markets are always changing, with constant innovations in market structure, in regulations, and the behavior of the market participants.

2. Abundant data: This can be applicable for short-term trading, but it would be diffi cult to apply this to fundamental investing. Good quality data from company accounts only exists for the past few decades, so on a two-year horizon there are only 10-20 distinct data points per company. This is not nearly enough for any algorithm to tease out any complex relationships. The Gobal Financial Crisis of 2007-08 played out more like the 1930s depression than any of the more recent recessions, and there are few data sets that go back as far as that. Even if these data sets were accessible, there is doubt that they would provide any discernible insights given how many other things are completely different after 80 years of social, fi nancial and technology developments.

3. Low uncertainty: This is not applicable for investing. Financial markets are volatile and unpredictable with prices driven by investors operating on multiple different time horizons and thrown by market impacts, irrational biases and bubbles. Complex networks of algorithms can trigger fl ash crashes, adding to market volatility.1

1 http://uk.businessinsider.com/what-actually-caused-2010-fl ash-crash-2016-1?r=US&IR=T

4. Clear objective: For investing as a general concept this is clear, but for individual funds with different objectives, there’s no single moment in time when any given investment has succeeded or not. Due to fl uctuating prices, securities have a constantly changing price which may or may not eventually yield a profi t. Investment objectives are dependent on the audience: for example a pension fund with the majority of members close to retirement is unlikely to want the volatility of a Small Cap Equities fund, while the same fund may be very appealing to those with a larger risk appetite.

5. The information is digital and in the system: Good investors synthesize all the information available to them relevant to their investment that they can use, (including, but not limited to: research reports, their understanding of market forces, human psychology of management, regulators and politicians, the mood of the market). However the nuances of that information, particularly the more qualitative aspects of it would be diffi cult to digitize.

ConclusionWe believe the biggest technological opportunity for achieving better investment outcomes for our clients is for us, as data scientists, to augment the intelligence of our fund managers, closing their blind spots and allowing them to see further, more clearly and more reliably. The benefi ts are particularly signifi cant when other investment managers lack the ability to close these blind spots, whether through lack of technology or initiative or scale or structure, as this then provides an ‘information edge’. And there are ample opportunities to use ‘AI’ techniques such as Machine Learning to refi ne and sustain this information edge.

It is still up to the individual manager whether to buy or sell a company because there are always a multitude of factors to consider. But think how good it feels to exit the highway a mile before congestion and take a traffi c-free detour while everyone else sits in their cars because they didn’t know the jam was there. Conversely, an even more well-informed driver, with specifi c familiarity of a newly opened route or upcoming incident might elect not to blindly follow the guidance that his or her ‘big data’ GPS system instructs them to take – in some instances a rather dubious detour based solely on an unattended algorithm. This, in our view, is where AI and IA fi nd their optimal sweet spot.

Humans or robots: Who’s really driving the car?12

The opportunity in early stage venture capital

Venture capital has fi nanced some of the biggest successes in technology of our timeVenture capital has a rich history of delivering both strong returns for investors and helping early stage companies realize their potential.

Previous venture capital (or VC) success stories include the fi rst US public company to be worth more than $1 trillion, Apple, and other household names such as Microsoft, Amazon, Facebook and Google, each worth more than $500 billion. VC has turned entire industries on their heads. The music, newspaper, and media industries have all been disrupted in this way. Furthermore, venture capital-backed healthcare companies have developed medicines that are used to defeat previously incurable diseases, while others search for a cure for cancer. It would not be an overstatement to say that venture capital has been behind many life changing developments.

From an investor’s perspective, venture capital provides access to the growth stories of tomorrow.1 It is also a signifi cant part of the investment landscape – the venture capital/growth sector makes up around a third of annual private equity fundraising volumes.

However, it has also been described as a game of “high risk poker”.2 Venture capital, if done incorrectly, can be a way to lose a signifi cant amount of money and in the past, some

1 In contrast, public markets are increasingly capturing a larger, more mature snapshot of the corporate sector as companies elect to stay private for longer. For more information, see What is the point of the equity market? Schroders, April 2018

2 Michael Moritz, Sequoia Capital, https://www.bloomberg.com/news/articles/2015-10-17/sequoia-s-michael-moritz-q-a-venture-capital-is-high-risk-poker-

investors have burned their fi ngers. Memories of crashing technology stocks during the Dotcom bust also haunt some investors. Invest with bad funds and you can lose most or all of your capital!

This is how certain misconceptions have sprung up around venture capital. However, it is important to note that losing money often results not from venture capital investing itself, but from relatively amateur mistakes that we think skilled investors can easily avoid.

Venture capital provides access to investment opportunities in game-changing companies with the potential to disrupt entire industries…but only if done correctly. In our experience 20% of companies drive 80% of returns, so success hinges on accessing the best opportunities. However, rather than being stacked against you, a well thought-through approach to portfolio construction, in our view, can ensure the odds are stacked in your favor.

Venture capital – the basics

Venture capital is early fi nancing for start-ups that are taking an idea and trying to grow it into a large scale business. The technology and healthcare companies which are typically backed by venture capital have the potential to change entire industries and have a profound effect on the way we live our lives.

Typically, at the time of initial investment, a company receiving venture capital backing only has a few employees and co-founders and potentially a beta version (early prototype) of their product or service.

Venture capital can allow these companies to increase in scale in an effi cient manner, normally over a period of several years. It can take them from this very early stage to the point where they can be generating millions, or even billions, of dollars in revenue.

The opportunity in early stage venture capital

Michael McLeanInvestmentDirector, Schroder Adveq

Duncan Lamont,CFAHead of Research and Analytics, Schroders

Steven Yang, CFAHead of Global Venture Investments, Schroder Adveq

13

Misconception 1: Venture capital is cyclical and now is a bad time to investSome investors look at the high valuations of publicly listed technology stocks to mean that the sector is frothy and that this is a bad time to invest in venture capital, given its technology focus.

Figure 1: Technology valuations have been soaringForward price/earnings multiple*

*Note: shows the median forward PE multiple for Apple, Facebook, Google, Amazon, Microsoft, and Intel. Source: Thomson Reuters

However, while investor concerns are understandable, they are also misplaced. First, the nature of early stage venture capital means that entry valuations are driven by strength of idea, stage of development, management team, probability of commercial success, and other factors, all of which are completely unrelated to public markets.

Secondly, venture capital and private equity more generally are long-term investments. Money is committed upfront but not drawn down for several years, with divestments typically not for several more. This phasing of drawdowns and realizations over time, limits the risk of an investor losing out by making a commitment at what turns out to be a bad time.

Thirdly, the fact that venture capital investments are likely to be held for up to 10 years before an exit is sought means that today’s public market valuations are almost an irrelevance.

To give an extreme example, even many investors who allocated to venture capital during the peak of Dotcom hysteria, in 1999, managed to escape relatively unscathed. The internal rate or return (‘IRR’) of the (pooled) average venture capital fund was barely negative, at -0.9% a year, while top quartile funds delivered positive returns (+3.2%). Bottom quartile managers understandably fared far worse, losing -11.4% a year.3 It may surprise some readers to learn that investing at the peak of the biggest technology bubble in history did not turn out too painful for investors in venture capital, so long as the worst performers were avoided. Funds that were raised in the run up to the fi nancial crisis fared even better. The (pooled) average from the 2006 vintage year was 9% a year, with top quartile funds delivering 13% a year. Investors who shied away because of fears over a potential equity market downturn would have lost out. 3 Source: Cambridge Associates, Schroder Adveq. Rankings and performance refl ect

past performance which is no guarantee of future results.

In addition, it is important to distinguish between today’s well-known technology champions and the technology sector or venture capital universe more generally. In the US, historically the largest market for venture capital,4 an average of around 1,000 companies a year receive fi rst round fi nancing from venture capital. Even in the depths of the fi nancial crisis, start-up creation, the lifeblood of venture capital, remained at a healthy level. Furthermore, as described later, on a global level, the pool of venture capital prospects has been increasing at a rapid pace. So there is no shortage of opportunities.

Misconception 2: It is impossible to gain access to the best venture capital funds and venture capital-backed companiesLike many of the misconceptions highlighted in this paper, this has mutated from a valid starting point. It is an uncontroversial fact that access to the best funds, and companies, is essential for success in venture capital. The difference between winners and losers is especially marked. About 20-30% of venture capital-backed companies go out of business but around 20% end up worth over $250 million. Those 20% drive more than 80% of returns. To have the best chance of success, it is important to invest in funds which can access and add value at the most promising prospects.

The importance of fund selection shines through in the performance fi gures. Over the 2003-13 vintage years (more recent years excluded as it typically takes fi ve years for a venture capital portfolio to develop), top quartile venture capital funds outperformed the technology-focused NASDAQ index of public companies by 2.5% a year and outperformed bottom quartile funds by 14.5% a year, on average. On a compound basis over the 10-12 year life of a venture capital fund, the difference is even starker.

Figure 2: The importance of access to top fundsInterquartile range for venture capital fund IRR by vintage year

Source: Cambridge Associates, Schroder Adveq

While the need to invest in top performing managers with strong reputations is widely recognized, the challenge is accessing them. Many restrict who they partner with. They normally prefer to work with stable and reliable Limited Partners (LPs) that have been in the business for decades, through multiple investment cycles.There is sound reasoning behind this approach. The effort involved in taking an early stage company and helping it grow should

4 1997-2017 average = 1,020. This is not unduly infl uenced by the Dotcom boom: 2004-17 average = 1,001.

0

5

10

15

20

25

30

Aug-

13

Dec

-13

Apr-

14

Aug-

14

Dec

-14

Apr-

15

Aug-

15

Dec

-15

Apr-

16

Aug-

16

Dec

-16

Apr-

17

Aug-

17

Dec

-17

Apr-

18

Aug-

18Top tech companies

-10

-5

0

5

10

15

20

25

30

2003

2004

2005

2006

2007

2008

2009

2010

2011

2012

2013

14 The opportunity in early stage venture capital

not be underestimated. Managers need to put a lot of time and energy into each of their investments. General Partners (‘GPs’) tend to take a board of director’s role with each and dedicate a signifi cant amount of time advising and helping companies grow. The best GPs add the most value and therefore tend to attract the best entrepreneurs.

But it is a misconception that success in venture capital is restricted to seasoned investors. Even novice investors can get access to top funds, so long as they invest through the right partner. By piggy-backing on a more experienced LP’s credentials, through a co-investment or fund of funds investment, it is possible to get a seat at the top table. Without that, investors risk missing out on the top venture capital-backed companies and therefore missing out on a signifi cant percentage of return potential in the venture capital market.

Misconception 3: Venture capital is high risk“Most VC-backed companies fail, so it is too risky to invest in venture capital. Look what happened during the dot-com bubble.”

This is a common opinion of venture capital but in many respects we feel that it is also wrong. As highlighted earlier, it is true that a notable percentage of venture capital-backed companies fail to survive. However, others deliver returns of several hundred or even thousand percent. The individual

Figure 3: Venture capital has generally been less risky than the broader equity marketsAnnualized standard deviation of returns

17% 16% 17%19%

8% 10%

0%

5%

10%

15%

20%

MSC

IW

orld

MSC

IU

SA

NAS

DAQ

Russ

ell 2

000

Gro

wth

Cam

brid

ge G

loba

lVe

ntur

e Ca

pita

l

Cam

brid

ge G

loba

lPr

ivat

e Eq

uity

**

Maximum drawdown: Q1 2008-Q4 2017

-43% -40%

-31%-34%

-18%

-32%

-60%

-40%

-20%

0%

MSC

IW

orld

MSC

IU

SA

NAS

DAQ

Russ

ell 2

000

Gro

wth

Cam

brid

ge G

loba

lVe

ntur

e Ca

pita

l

Cam

brid

ge G

loba

lPr

ivat

e Eq

uity

**

** Includes buyout and growth strategies. All analysis is based on quarterly returns. Source: Bloomberg, Cambridge Associates and Schroder Adveq. For illustrative purposes only. Funds and indices shown do not refl ect any recommendation to buy or sell any security. Other funds would have achieved different results.

nature of each investment can drive signifi cant diversifi cation benefi ts at the portfolio level. Venture capital funds normally include around 30 companies, spread across multiple sub-sectors of technology. This broad company and sector diversifi cation helps mitigate risk, thus lowering volatility and drawdown. Even more so, venture capital fund of funds portfolios are typically diversifi ed across several hundred underlying companies, thus lowering risk further.

One consequence is that, although individual venture capital investments are higher risk than public equity markets, venture capital portfolios are not. In fact, over the past 10 years, which includes the fi nancial crisis, venture capital has been less volatile than public markets. It also suffered a more limited drawdown during the crisis.

One challenge to this analysis would be that venture capital investments are not always fully revalued on a regular basis, so this may be understating true risk.

To help rebut this, using simulation analysis we have quantitatively analyzed venture capital risk using realized entry and exit valuations for companies which Schroder Adveq has invested in (see Figure below for more information). As they refl ect actual transaction prices they are free from any valuation concerns. This analysis confi rms that, although individual venture capital investments are high risk, risk declines rapidly as portfolio size increases. This is indicative of very low correlations between individual

Our simulation analysis, explained

We have analyzed the return volatility of thousands of synthetically created portfolios of thousands of underlying venture capital investments of differing sizes. Figure 4 shows a measure of return variability across these portfolios. A high fi gure on the vertical-axis means that there is a wide spread of returns between portfolios of a given number of investments. The chart shows how this varies as the number of portfolio companies increases.

Figure 4: Venture capital risk declines rapidly as the number of investments in a portfolio increases

Standard deviation of total value to paid-in (TVPI) across 10,000 portfolios

Source: Schroder Adveq, 2018. Refl ects equal-weighted portfolio simulations of TVPI. Please refer to the back of this report for important information on simulations.

0.0

1.0

2.0

3.0

4.0

5.0

0 5 10 15 20 25 30Number of companies in portfolio

15The opportunity in early stage venture capital

investments, driving signifi cant diversifi cation benefi ts. As a consequence, contrary to popular opinion, portfolios of venture capital investments are far less risky than might be thought.

How to invest in venture capital successfully: The fi ve secrets of successVenture capital isn’t as risky as investors think, it is possible to get access to top funds/companies, and timing markets isn’t key. With all that in mind, how do you build a successful venture capital portfolio?

1. A global approach is criticalSilicon Valley in the US, the largest innovation hub in terms of companies created globally, is probably the most well-known market for venture capital. However, the industry is much more global in nature, with 10 innovation hubs spread across the US, Europe and Asia.5

Although the US has historically been the largest market for venture capital and has continued to expand, the main driver of the industry’s growth over the past decade has, in fact, been elsewhere (Figure 5). Asia, specifi cally China, has seen the largest growth. Europe and the rest of the world have also expanded in scale.

This shift has been mirrored in the make-up of the fund industry, with large number of Asian venture capital managers in particular having been established in the past decade.

Figure 5: A global opportunity set for venture capital Number of fi rst round fi nancings by region

Source: Venture Source, 2017

Different regions and hubs have different strengths. Silicon Valley is such a successful hub for both technology and healthcare because of the concentration of talent in the region that continues to be replenished by strong universities in the region. However, elsewhere in the US, Los Angeles, Boston, and New York are all key locations. In Asia, the top innovation hubs are Beijing, Shanghai, and Tokyo. China has become a technology leader in certain areas, such as electric vehicles, autonomous vehicles, gaming, and, to an emerging extent, biotechnology. Part of the driving force behind China’s advancement as a technology leader is its 750 million mobile internet users, which is more than double the

5 Silicon Valley, Los Angeles, Boston, New York, Beijing, Shanghai, Tokyo, Berlin, London, and Paris.

entire US population. This is a large market to both test and sell products into. India will also become an important region in the future due to its large population, which is increasingly adopting technology. In Europe, you have Berlin, London, and Paris as the major technology hubs.

A narrow focus on only one region would miss potentially game-changing opportunities in other areas of the world. Globalization and the resultant dispersion of technology talent around the world makes this an even more pressing issue. This is why a global approach to venture capital investing is so crucial.

2. Focus on topics of the futureThe two main areas to concentrate on within venture capital are technology and healthcare. The top venture capital managers will be investing in the next-generation themes but these are constantly changing and certain topics come and go.

The rapid, persistent increase in computing power predicted by Moore’s Law6 over 50 years ago has been the driving force behind the quick adoption and acceleration of technology innovation. Today we have 50 times more computing power in our mobile phones than in the Deep Blue supercomputer that beat a world chess champion for the fi rst time in 1997. They are also 120 million times faster than the supercomputers that sent a man to the moon in 1969. Ten years ago, the top venture capital managers were focused on what were then emerging topics, such as social networking, cloud computing and Software-as-a-Service. However, those themes have mostly been played out. Today, venture capital managers are now focused on some of the following topics: autonomous vehicles, artifi cial intelligence (AI), gene editing, curing cancer, electric vehicles, blockchain and digital health. Two of the more important enabling technologies which are a current focus are gene editing technology and AI/machine learning. Gene editing technology has the potential to drive advancements in curing diseases that currently do not have an effective cure. Drugs to cure various types of cancer are one application as are orphan diseases and central nervous system disorders. In recent years, venture capital played a crucial role in curing diseases such as Hepatitis C, retinal dystrophy, Leukemia, and other diseases. This has all been possible from breakthroughs in bio technology, such as new drug creation and gene editing.

AI and machine learning (ML) are projected to have a profound impact on the rate of technological progress. Growth in processing power (measured in calculations per second) has been exponential and by 2050, it is predicted that consumers will be able to buy a device with the computational power of all mankind for the price of a refrigerator today. This date is known as the technological singularity.

AI and ML are primed to disrupt multiple industries, including: fi nance, legal, government, manufacturing, retail, agriculture, energy, transportation, insurance, and customer service. For example, AI has been one of the driving forces behind the advancement of autonomous vehicles we have seen in the last few years.

6 The prediction that the number of transistors on a computer chip will double every two years, leading to a doubling in computer processing power every two years.

0

1,000

2,000

3,000

4,000

5,000

2004

2005

2006

2007

2008

2009

2010

2011

2012

2013

2014

2015

2016

2017

USA China Europe Rest of world

16 The opportunity in early stage venture capital

Several automotive companies have acquired venture capital-backed companies to accelerate the technology development. Drones and next-generation robotics are already changing manufacturing and distribution across the supply chain. In healthcare, ML algorithms already shorten time to market for multiple drug development programs and help to detect health anomalies across many diseases more effectively than doctors Internet-of-Things (IoT) devices have proliferated in recent years and AI is allowing new ways for these devices to interact with each other and provide advanced analytics for both the energy and retail sectors.

This is an exciting time to invest in technology and venture capital is the driving force behind the technology that is empowering exponential AI growth.

3. Maximize “portfolio quality density”

The statistical distribution around the top 20% of companies driving about 80% of total venture capital market returns has remained the same for several decades.7 If you don’t have a concentrated portfolio of those top 20% companies then your venture capital returns will get diluted by the losers in the portfolio. This is why quality density matters in venture capital.

As mentioned earlier, it is crucial to focus on established, typically access-restricted GPs when investing in early stage venture capital. The top quartile GPs tend to have the highest quality deal fl ow and attract the top entrepreneurs, which in turn provide the strongest investment returns. Characteristics include strong brand value, with a repeatable track record of exiting several game changing companies and evidence of value they have added at their portfolio companies.

In addition to building a core portfolio of the top access-restricted GPs, it is also important to selectively add new

7 Source: Schroder Adveq study of venture portfolio companies going back to 1994

emerging GPs, who have the potential to be a top 20 ranked venture capital fi rm in the coming years. These tend to have previous experience at the top venture capital funds or were previously successful entrepreneurs/operators at top technology companies.

4. Do not get over-excited and be cautious about frothy parts of the market

Although we argued earlier that the structure of private equity funds means that venture capital investors need not be intimidated by high valuations in the publicly listed technology sector, that does not mean they should be blind to dynamics in the private equity market.

Our research8 has found a statistically signifi cant inverse relationship between fundraising volumes and future performance. This intuitively makes sense. High levels of fundraising leads to increased dry powder, tougher competition for deals, higher prices being paid and lower subsequent returns. The reverse is also true with slack periods for fundraising foreshadowing an environment more conducive to performance. We have quantifi ed and formalized this relationship in the Schroder Adveq Fund Raising Indicator (‘FRI’).9 It evaluates fundraising cycles and normalizes the magnitude of cyclical over- or under-fundraising in a way that permits comparison across regions and sectors. In this way, it can be used to identify areas of the market that are better placed to deliver returns and those which may be best avoided.The Schroder Adveq FRI currently suggests that the late stage venture capital market is at risk of overheating due to large sums of money having been raised in recent years (Figure 6) at a rate of increase well above the long term trend. Valuations of late-stage venture capital investments have also been soaring – in the US, the median late-stage company is worth six times more than it was fi ve years ago.

8 Where should you invest in private equity today?, Schroder Adveq, September 20189 See footnote 8

$ billion

0

20

40

60

80

100

120

140

160

180

18E16141210080604020098

$ billion

0

20

40

60

80

100

18E16141210080604020098

Fund raising (infl ation-adjusted) Long-term trend 80% confi dence interval for long-term trend

Late stage/growth capital includes 50% of SoftBank Vision Fund and includes investment activity from non-traditional sources of capital (e.g. corporate investors). Source: Preqin, Pitchbook, Zero2IPO, Schroder Adveq 2018

Figure 6: An overabundance of late stage venture/growth capital has been raised...

..but early stage venture capital fundraising is in a healthy state

17The opportunity in early stage venture capital

Why has this occurred?For one, venture capital-backed companies have been staying private longer with the median time to a liquidity event (IPO, or sale to another fi nancial or corporate investor) being roughly 10 years. This compares with roughly fi ve to six years in the past. Many of these technology and technology-enabled companies are generating strong growth and achieving revenue of more than $1 billion, but are deciding to forego the public markets. This has led to a proliferation of so-called “unicorns”, venture capital-backed companies valued at more than $1 billion. There are currently 258 unicorn companies globally, up from only a handful seven years ago.

This creates demand for larger sums of late stage fi nancing than in the past. Previously the public market would have provided much of this, but latterly it has been the late stage venture capital market. Investors have also fl ocked to the sector, attracted by the prospect of capturing some of this growth. Given the size of these companies, “pre-IPO” fi nancing rounds, a focus for late stage venture capital, have increased considerably. Both the demand and supply-side have played a part in the rapid growth of the market.

In contrast, capital fl ows for early stage investments have remained relatively stable and median pre-money valuations across the US, Europe, and Asia have increased only modestly. The Schroder Adveq FRI confi rms that early stage venture capital is better positioned to deliver strong returns than the more crowded late stage market (Figure 7).

Figure 7: Late stage venture capital is at risk of overheating but the early stage market is better positioned

Number of fi rst round fi nancings by region

0

100

200

300

400

500

Early stage venturecapital 2000

Late stage/growth capital

Early stageventure capital

Risk of overheating

Long term trend

Source: Schroder Adveq, 2018

In addition, while the growth of the late stage market creates challenges for investors in that space, it is a positive for the early stage market. It creates a larger pool of capital for early stage investors to sell into as an exit, in addition to the IPO or merger and acquisition routes, or for less dilutive follow-on fi nancing.

The long-term nature of private equity investing means that market timing is less of an issue than in public markets. However, portfolios can be improved by tilting new commitments towards less crowded parts of the market where competition is weaker. It is easy to get excited about headline grabbing companies but a more disciplined

approach is likely to be more successful. The early stage venture capital market beats the late stage market hands-down on this front at present.

5. Mix primaries, secondaries and co-investments

Investing in the top venture capital managers is a necessary step to building a high performing venture capital portfolio. However, further enhancements can also be made.

Given the signifi cant dispersion in returns, we think that a portfolio approach spread across multiple GPs can help mitigate risk and give more certainty over returns. The specialized nature of many venture capital GPs also means that a portfolio diversifi ed in this way can better cover different industries, topics and regions without compromising portfolio quality density.

Investments in secondaries (funds which buy stakes in existing funds from LPs) and co-investments (where an LP invests in specifi c deals alongside a GP rather than via a fund structure), both also have the potential to add considerable value.

Secondary private equity transactions typically take place at a discount to the net asset value (‘NAV’) of the stake in question – this is the price the selling investor has to pay for exiting an illiquid investment early. The average venture capital secondary discount has been in the 17-25% range over the last fi ve years (Figure 8), much greater than elsewhere. The compelling discounts in venture capital secondaries help to mitigate the j-curve effect10 that naturally occurs in venture capital investing.

Figure 8: Venture capital secondaries offer a bigger discount than elsewhereSecondary pricing, % of net asset value

70%

80%

90%

100%

2013 2014 2015 2016 2017

Venture Buyout Real estate All strategies

Source: Source: Greenhill, December 2017

Co-investments serve a different but equally valuable function—they can enhance portfolio quality density (described earlier). By “doubling down” on the portfolio winners of the top venture capital managers, it is possible to get more dollars to work in the future game-changing companies in technology and healthcare. This strategy allows for increased concentration within the top 20% of venture capital-backed winners, thereby maximizing return potential.

10 The process by which it can take several years for a private equity fund to start to generate positive returns.

18 The opportunity in early stage venture capital

ConclusionWe think venture capital is an attractive asset class due to the ability to get early access to potential market leaders of tomorrow, as well as generating strong returns for investors at comparatively low risk. It is possible to get exposure to both the best venture capital funds and top venture capital-backed companies, thus providing access to top quartile returns. Venture capital is also not as cyclical as most investors think, providing consistent strong returns through market cycles. Within the venture capital market, it is currently best to invest in the early stage part of the market.

Investors should concentrate on investing in established and typically highly access restricted fund managers. Preference should be given to fund managers that invest in themes of the future and have access to high quality deal fl ow and top entrepreneurs. It is important to take a global approach to venture capital investing as more game changing companies are being built outside of Silicon Valley, such as in Asia and Europe. Finally, we believe a successful venture capital portfolio should consist of a mix of primary fund commitments, direct co-investments, and secondaries.

19The opportunity in early stage venture capital

Investing in today’s Commerical Real Estate market: It’s all about that basis

Jeffrey Williams, CFAFund Manager,Securitized Credit

Today, the US Commercial Real Estate (CRE) market is characterized as a market with high property valuations. With this challenge in mind, we believe there are attractive opportunities to lend in areas of the US CRE market, where capital provision is ineffi cient. Lending allows for attractive return prospects without compromising on credit or capital structure. As well, lending in less effi cient markets does not require reliance on fi nancial leverage to generate return. Investment strategies incorporating US CRE bridge loans, particularly loans that are “middle market”, can provide investors with attractive return prospects, and lending on properties with a “reset” cost basis off ers a high level of downside risk mitigation.

,

Location. Location. LocationThe upswing in CRE prices has garnered a lot of media attention. In particular, the ”Big Six” cities (New York, San Francisco, Los Angeles, Boston, Chicago, and Washington DC), larger “trophy” properties, and multi-family apartments have each benefi tted tremendously from aggressive capital provision that is driving fi nancing rates down and driving valuations up. The Big Six cities have seen particularly aggressive capital provision, driving up values and increasing the property and loan cost basis. At the same time, property values in other major cities such as Raleigh, Atlanta, and Columbus have not increased as quickly, even though fundamentals, such as occupancy, rent growth, and supply/demand dynamics in those markets can be as strong (if not stronger) than the fundamentals seen in the Big Six cities. This is an extremely important point, as we illustrate in Figure 1. When seeking to purchase larger properties in the Big Six cities today, or when lending on them today, you will likely have a much higher cost basis. But an elevated cost basis is not a condition of all major cities. By focusing on smaller loans, and major cities outside of the Big Six, we believe investors can fi nd attractive risk/return combinations. Commercial real estate markets are governed by local factors. Beyond fi nancing costs, an important component of a commercial real estate investment is the cost basis in the property. The cost basis in a commercial real estate investment is the total cost to acquire the property divided by the total square footage of the property, basically the price per square foot. The cost basis for a loan secured by a commercial real estate property, or to a lender, is the loan amount divided by the total square footage of the property.

Figure 1: Commercial Real Estate comparisons: How smaller big cities compare to the Big Six

Source: Schroders, CoStar, March 2018. Not intended to be a comprehensive listing

of markets.

How to pick up insulation to downturns – it’s all about that basisA low basis in a property not only enables equity owners to enjoy greater margins when rents are rising, but it also offers a level of protection in the event property values are declining. From a lending perspective, a low basis is a competitive advantage as it provides the borrower with a cushion to charge lower rents than what would be possible for properties with a higher basis. The lower the rent, the larger the tenant pool, which can help to maintain occupancy, preserve cash fl ow, and provides cushion to enable debt service coverage in a downturn.

Michelle Russell-DoweHead of SecuritizedCredit

0

100

200

300

400

500

600

700

800

900Price per square foot

1982

Q1

1983

Q2

1984

Q3

1985

Q4

1987

Q1

1988

Q2

1989

Q3

1990

Q4

1992

Q1

1993

Q2

1994

Q3

1995

Q4

1997

Q1

1998

Q2

1999

Q3

2000

Q4

2002

Q1

2003

Q2

2004

Q3

2005

Q4

2007

Q1

2008

Q2

2009

Q3

2010

Q4

2012

Q1

2013

Q2

2014

Q3

2015

Q4

2017

Q1

San Francisco New York Los Angeles Raleigh Atlanta Columbus

20 Investing in today’s Commerical Real Estate market: It’s all about that basis

One way to get access to a property with a low cost basis is by lending to borrowers whose properties have recently seen a basis reset, such as with bridge loans, which are short-term (3-5 year) loans that often fi nance an acquisition of a property that is underperforming its local market.

Consider the pre-crisis commercial mortgage-backed securities (CMBS) market. Pre-2008, many commercial properties were fi nanced with a high degree of leverage. As over-leveraged properties that were originated before the crisis reached their maturity date, it was not uncommon for the borrower to default. Over-leveraged properties typically have a high degree of debt service which strains cash fl ow and limits the borrower’s ability to put capital into the property. Often, when borrowers are not re-investing capital into a property, the property becomes less desirable and occupancy falls, resulting in a default under its existing loan. A bridge loan would facilitate a purchase of the property, at a substantially lower price, allowing the new borrower to invest in the property and move the property toward market performance/stabilization. Figure 2 illustrates what we consider to be a typical bridge loan. The previous owner purchased the property in 2007 for $15.5mm or $152/sf and obtained a CMBS loan for $10.7mm or $105/sf. The previous owner did not have the resources to retain tenants in the property and the occupancy rate fell substantially below the overall market rate. At the loan maturity, the borrower defaulted and a new borrower purchased the property for $7.5mm or $74/sf, lowering the cost basis in the property substantially. A senior bridge loan fi nanced the new purchase at $55/sf, nearly 52% less than the prior loan basis.

Figure 2: A typical bridge loan example: The power of the reset

Source: Schroders. For illustrative purposes only. Other loans would have varying terms.

This bridge loan example demonstrates the potential power of resetting the property cost basis. In this local market, competing properties trade at $140/sf area. This property at $74/sf can offer highly competitive rental rates and still offer an attractive return on equity. The fl exibility to compete on rents becomes extremely valuable during a downturn in real estate markets. This, therefore, is a defensive way to play the real estate markets at a local level.

Figure 3: Many regions offer bridge loan PSF levels that are remarkably lower than GFC era – West Palm Beach Offi ce Properties example

Source: Trepp, CoStar, Schroders through May 2018. For illustration only. Loan values and prices are subject to change, and can vary loan to loan.

When looking to the CMBS market for data around the importance of the cost basis, Figure 3 presents a time series sample of the cost basis for CMBS loan originations in a single city, in this instance, we’ve selected West Palm Beach, Florida. The orange line represents the historical median price-per-square-foot. It is easy to see the more than 30% decline experienced during the Global Financial Crisis (GFC). The blue diamonds, on the other hand, represent the actual loan PSF for individual properties in the CMBS market. Notably, by lending, the investor has picked up a comfortable cushion to stressed conditions, even to the GFC. Most of the blue diamonds in Figure 3 (the loan per square foot) remain lower than the median price per square foot during the GFC. As well, looking at smaller loans, and a current investment the price PSF and loan PSF are even lower than the median. This represents a healthy margin of safety.

For further illustration, we have indicated the cost basis and the loan basis for a bridge loan in the same city. While the city as a whole has average loan balances of $140/sf and average valuations of $220/sf, this particular bridge loan has a $55/sf loan basis to facilitate an acquisition of a property with a $74/sf basis. This bridge loan creates an attractive proposition with an inherent defensive posture in the event of a near-term market downturn.

The lower PSF in this city and the margin of safety to GFC are typically seen in the major cities outside of the six largest cities as shown in Figure 4. It is, indeed, all about that basis.

0

50

100

150

200

250

May-06 May-08 May-10 May-12 May-14 May-16 May-18

Loan per SF Price PSF

$PSF

Bridge Loan Example

Previous CMBS Loan New Loan

Purchase Price $15,500,000 $152/sf $7,500,000 $74/sf

Total Debt $10,700,000 $105/sf $5,625,000 $55/sf

Cash Equity $1,875,000

21Investing in today’s Commerical Real Estate market: It’s all about that basis

Figure 4: Cost basis have risen, and can vary dramatically, within the Big Six: New York example

Source: CoStar, Intex, Schroders as of June 2018. For illustration only.

Performance history bears out the benefi t of lower cost basis and lower loan-to-value

Having a cushion against valuation changes are critical in today’s market. In our view, this cost basis metric, is a way of assessing cushion, and is as important as the loan-to-value (LTV) ratio when assessing the equity or credit risk in commercial real estate. History, as well, bears out the benefi t of protections like a lower cost basis and a lower LTV ratio.

One of the obvious ways to mitigate credit risk in the commercial real estate loan market is to lend at lower LTV ratios. Life insurance companies have been long-time participants in the commercial real estate loan market and have enjoyed excellent credit performance on their loan portfolios due to low LTV ratios. Their performance is demonstrated through extremely low delinquency rates, even during and following the GFC (Figure 5). This is in contrast to the CMBS conduit market which has seen LTV ratios fl uctuate with cycle expansions and produce corresponding elevated historical delinquencies.

Figure 5: Historical loan delinquency rates

Source: Mortgage Bankers Association as of December 2017.

Figure 6 presents historical quarterly average Life Company loan-to-values alongside Moody’s concluded loan-to-values for conduit CMBS securitizations. While Moody’s uses its proprietary valuation model to conclude its loan-to-value

ratios; actual deal level weighted average LTVs do not accurately capture total deal leverage given the presence of pari passu notes. The Moody’s data accurately refl ects the trend in total leverage across new issue conduit CMBS deals.

Figure 6: Historical Life Company LTV and CMBS Moody’s LTV

Source: ACLI and Moody’s December 2017

The benefi ts of private lending: How middle-market loans can offer downside protection and income opportunities

A key takeaway from the Insurance industry’s CRE experience is that private commercial real estate loans provide a tremendous advantage – customization. Private loan market can offer investors the opportunity to customize a portfolio of loans with conservative loan-to-value ratios and with attractive return potential, and with protection from potential declines in commercial real estate values.

In our view, loans with a principal balance between $2mm and $25mm (i.e., the “middle market”) secured by properties in major cities outside of the Big Six, present compelling opportunities. Most CRE debt funds focus on loans with balances between $25mm and $100mm due to the ability to scale those investments. The reality is that it takes the same amount of time to underwrite a $5mm loan as it does a $50mm loan.

As such, smaller loans have traditionally been the domain of the community banks or lenders reliant on CMBS funding. Currently, due to regulation, regional and community banks are near maximum exposure to real estate loans. The CMBS market, however, has not fi lled in the gap. The decrease in available capital for smaller loans has created ineffi ciency in middle market lending on CRE.

As a result, for moving down in loan size, compensation for senior loans with comparable LTV ratios can be nearly 1.50% higher. The loans in the middle market category have a considerably lower basis, which provides additional margin of safety against market declines. The compensation may even be higher in mezzanine loan space, where nearly 2% higher coupons can be earned, and here the loan-to-value ratio is often more conservative than for larger mezzanine loans.

LTV%

0

20

40

60

80

100

120

140

55.0

60.0

65.0

70.0

75.0

80.0

1987

1988

1989

1991

1992

1994

1995

1996

1998

1999

2000

2002

2003

2004

2006

2007

2009

2010

2011

2013

2014

2015

2017

Life Co. Loan LTV Conduit CMBS Moody's LTV (RHS)

0

100

200

300

400

500

600

700

Jun-04 Jun-06 Jun-08 Jul-10 Jul-12 Jul-14 Jul-16 Jul-18

Loan per SF Price PSF

101 Ave of Americas 215 Park

Avenue

215 Park Avenue

101 Ave of Americas

$PSF

0.00%

1.00%

2.00%

3.00%

4.00%

5.00%

6.00%

7.00%

8.00%

9.00%

10.00%

4Q99

4Q01

4Q03

4Q05

4Q07

2Q08

4Q08

2Q09

4Q09

2Q10

4Q10

2Q11

4Q11

2Q12

4Q12

2Q13

4Q13

2Q14

4Q14

2Q15

4Q15

2Q16

4Q16

2Q17

4Q17

CMBS Life Companies Fannie Mae Freddie Mac Banks & Thrifts

22 Investing in today’s Commerical Real Estate market: It’s all about that basis

Figure 7: Historical commercial real estate prices

Source: CoStar, Schroders as of March 2018. Loan values and prices are subject to change over time. Yields shown are subject to change over time and do not offer any guarantee of future results.

The advantages of customization are not only evident in loan pricing and leverage or cost basis, but also in the added benefi t of controlling the servicing of the loan. Private CRE loans have had stronger credit performance and lower realized delinquency rates, when compared to securitized CRE loans. One reason for that performance difference is that direct CRE lenders could more adeptly address borrower issues by modifying or extending loans that encountered trouble during the fi nancial crisis, this fl exibility is more limited once loans are owned within a securitization vehicle. The ability to customize solutions when loan performance doesn’t go as anticipated is an added benefi t to private lending that offers the potential to reward a loan owner with better outcomes during times of distress and results in a preferred experience for the borrower.

Why now?Historically low levels of interest rates, low yields, and reduced expected returns have sent global investors in search of more return, yield, or income. We see this risk-seeking disposition across investors, as many have large over-weights to credit or to equities. The CMBS market is another area where risk seeking behavior can be seen. The CMBS credit curve, or the yield spread difference between securities rated AAA and those rated BBB- (triple-B minus), is at the lowest level seen post GFC. Trying to earn more income when the credit curve is fl at forces an investor to go further down the capital stack, when they are least compensated to do so. We think now is the point when investors should be rotating into investments with more stability or a greater margin of safety given compensation for taking on greater credit risk is lower.

Figure 8: Historical CMBS spreads

Source: JP Morgan, Schroders, through March 2018.

0

200

400

600

800

1000

1200

1400

1600

1800

0

100

200

300

400

500

600

700

800

900

1000Bps Bps

2005 2006 2008 2009 2010 2012 2013 2014 2016 2017

BBB- Pricing (LHS) AAA to BBB- Basis (RHS)

Attributes

Middle Market Bridge Loan

Large Market Bridge Loan

Middle Market B-Note

Middle Market

Mezzanine

Large Market

Mezzanine

Loan Size

$2-$25 million $25+ million $1-$10

million$2-$20 million $20+ million

Loan-to-Value 55%-65% 55%-65% 60%-75% 60%-75% 65%-80%

Loan per square foot

$135-$160 $300-$350 $145-$185 $145-$185 $350-$435

Yield 6.25% 4.75% 9%-10% 8%-10% 6%-8%

Interest Rate Duration

None None None None None

SourceRegional,

Community Banks

Money Center/Regional

Banks, Debt Funds

Private Investors

Regional, Community

Banks

Money Center/

Regional Banks, Debt

Funds

23Investing in today’s Commerical Real Estate market: It’s all about that basis

ConclusionWe think private commercial real estate loans are an attractive alternative where investors can earn an attractive return. By exploiting lending ineffi ciencies in the market, such as loan size, an investor can fi nd higher return and more protection from a downturn in the economy and/or real estate market values. Risk/return profi les can be enhanced by targeting less crowded middle market real estate loans, where, by virtue of lower competition property valuations are more reasonable and a lower loan PSF is available. Credit risk can potentially be lowered by lending on properties with a lower price PSF. Therefore, it’s all about that basis.

It is important to highlight that accessing these investment opportunities is not easy; it requires sourcing and servicing

ability. The major opportunity exists in middle market bridge loans with a “reset” cost basis, and servicing is an important component in this segment. Lending less than $25 million per property in cities like Jacksonville, Dallas, or Charleston, SC, to name a few, reduces competition and can create a higher return, while providing downside protection through a lower cost basis. Smaller CRE loans have not been targeted by the massive amount of global QE capital. But for those with the ability to properly source them, investing in a well-diversifi ed portfolio of these loans can be an attractive way to earn a stable return profi le over a 3-5 year period, eff ectively addressing a challenging investment environment by capitalizing on ineffi ciency.

24 Investing in today’s Commerical Real Estate market: It’s all about that basis

Less can be more: How to improve fi nancial reporting

While recent tweets have sparked the debate on the usefulness of public company’s adherence to the Security Exchange Act of 1934’s quarterly reporting requirement, we believe the crux of the debate focuses too much on the frequency, and not enough on the actual problems. That is, it’s too much information, often times it’s not the right information, and it costs a lot to produce and process. This article refl ects one Credit Analyst’s view on the subject.

David Knutson, CFAHead of Credit Research,Americas

I will never forget the explanation of the law of diminishing marginal utility from a college professor. “The fi rst beer tastes and makes you feel the best.” In economic terms, the high initial utility of a good or service declines toward zero marginal utility as its available supply increases. This economic law, in many ways, is connected to the ongoing debate between quarterly versus semi-annual fi nancial reporting and guidance.

This debate has been going on for years and has generally focused on the cost versus benefi t to equity investors. The fi rst precedent of corporate reporting dates back to 1903, with US Steel having voluntarily published its annual report. However fi nancial information from public companies was not required until 1911 when Kansas passed the fi rst Blue Sky law. Over time most states had similar statutes, but after the Wall Street Crash and following the Great Depression, the federal government took reporting authority away from the States and issued the Securities Exchange Act of 1934 requiring standardized periodic fi nancial information on public companies to help investors assess and price the cost of debt and equity capital.

Public companies are required to submit Form 10-Q (“Q”) for each of the fi rst three quarters of a company’s fi scal year and an annual report on Form 10-K (“K”) at the end of their fourth quarter. If a major event occurs during the quarter, companies must complete a Form 8K or “current report”. Most investors react to the initial quarterly press release published on an 8K a few weeks after quarter end. A few weeks after the press release, the Q arrives in the form of an unaudited phone-book-sized document fi lled with legal jargon and accounting boilerplate.

These are phone books, by the way.

The recurrring costs associated with public reporting requirements such as legal, fi nancial reporting, investor relations and audit-related fees are considerable, especially for smaller companies. This is at times thought to be one of the factors contributing to the decision to switch from a listed company to a privately held company. Over the last decade, the number of public companies has more than halved from 7,600 to approximately 3,500 today.1

1 Jay R. Ritter, Warrington College of Business Administration, University of Floriday; University of Chicago Center for Research in Security Prices

1

4

8

12

16

20

2 3 4 5 6 7

X

M’MU

MY

O

Initial utility

Mar

gina

l util

ity

Diminishing MU curve

Zero MU

25Less can be more: How to improve fi nancial reporting

This has especially impacted the high yield market where more than a third of new issuers are from private companies. While costs will vary widely depending upon the size and complexity of a company, average costs associated with being a public company is estimated to run more than $1,500,0002 annually. Adding to preparation costs is the cost associated with the quarterly consumption of the Qs.

There are approximately 14,000 investment manager research staffers in the US.3 If these researchers spend an hour studying fi nancial reports of a couple dozen companies four times a year, it adds up to over 1,000,000 hours spent studying historical information. While costs associated with the preparation and analysis of quarterly reports are signifi cant, unnecessary quarter-end trading and the corresponding transaction costs are quite possibly the biggest cost associated with frequent fi nancial information.

Although identifying unnecessary bond transactions costs specifi cally tied to quarterly reporting would be diffi cult, it is important to note overall trading costs are substantial. A 2015 study from the University of Southern California Marshall School of Business that analyzed the costs of trading bonds, known as the bid-ask spread, estimated total transaction costs for the 12 months prior to March 2015 of $26 billion.4 This is especially signifi cant when considered relative to the overall level of Treasury yields and credit spreads. Beyond the most liquid series of the largest issuers, buying and selling transacting costs for corporate credit can reach 5-10% or more of the credit spread, which signifi cantly reduces total return to the investor.

While these consumption and transaction costs are born by the issuer, analysts, and investors, the enterprise cost has become the most scrutinized. Business leaders are worried that companies are too concerned with short-term profi ts instead of focusing on investing in their workers, research, or operations. While research and development spending is rising in the US,5 the argument is that it could be higher or increasing at a faster pace if there were fewer public assessments. Management could have the fl exibility to increase strategic investments, which might reduce profi ts temporarily, but could pay off handsomely over the long term if such capital expenditures were properly administered. There is evidence from a McKinsey study that a longer-term perspective indeed increases revenue versus a short-term focus. According to the study, from 2001 to 2014, the revenue of long-term-oriented fi rms cumulatively grew, on average, 47% more than the revenue of short-term-oriented fi rms, and with less volatility. Similarly, the earnings of the long-term-oriented fi rms grew 36% more than those of other fi rms over this period, and their economic profi t was 81% higher.6

2 https://www.strategyand.pwc.com/media/fi le/Strategyand_Considering-an-IPO.pdf3 http://www.pionline.com/article/20170301/INTERACTIVE/170309988/number-of-

investment-management-industry-analysts-declines-in-2016 4 Harris, Lawrence, Transaction Costs, Trade Throughs, and Riskless Principal

Trading in Corporate Bond Markets (October 22, 2015). Available at SSRN: https://ssrn.com/abstract=2661801 or http://dx.doi.org/10.2139/ssrn.2661801

5 https://data.worldbank.org/indicator/GB.XPD.RSDV.GD.ZS?end=2015&locations=XU&start=2001

6 Barton, D., J. Manyika, T. Koller, R. Palter, J. Godsall and J. Zoffer. “Measuring the impact of short-termism.” Feb. 2017, McKinsey Global Institute.

The real debate particularly when it comes to enterprise cost is not about quarterly or semi-annual reporting, it is about guidance. Companies manage analyst expectations by selectively guiding the market up or down prior to earning reports. In addition to formal and informal forward looking information, management teams spend a lot of time massaging earnings using both accounting choices, known as earnings management, and operating discretion such as the timing of research and development expenses, accruals and impairments. Although public companies are not required to provide information about how they are doing during the quarter, about 25%-30% have offi cial guidance policies.7 All of this is done, in many instances, to deliver results that superfi cially beat analyst expectations.

It wasn’t always this way. The guidance business took off during the latter half of the 1990s, after a Congressional Act was passed protecting companies from liability if their projections were not realized.8 At the time, executives generally believed that guidance would result in higher valuations from lower volatility and improved liquidity from more frequent information or transparency, particularly in in cyclical sectors.

However, studies have found that short-term guidance does not meaningfully affect investment, valuation or volatility.9 It doesn’t matter if information is meaningful or noise, when it provokes transactions it is good for traders but costly to asset owners. Also much like costs related to Qs, public company executives have noted that the burden of providing and managing to guidance reduces long-term investment and is a motivating factor for public companies to de-register and become private.10

It is likely that these (and other) academic fi ndings led to the UKs Financial Conduct Authority having removed its quarterly reporting requirement in 2014.11 But with that, more than 90% of UK-listed companies continue to report on a quarterly basis as many are also listed in the US and have to report quarterly anyway.

Today, there are growing concerns that managing to a particular sales, margin or earnings per share could result in an ineffi cient allocation of investment, which in turn ultimately impairs the long-term prospects of the fi rm. While the payoff from strategic plans is diffi cult to predict annually or even quarter to quarter, in the last two decades, fi rms have become very good at playing the guidance game. Very large, complex, global companies have a remarkable ability to land earnings to within a penny per share of guided expectations. For example, a year ago for the 3Q17 quarter, not one of the biggest US fi nancial institutions earned less than the guided estimates. In fact, over the last two years,

7  Analysis of guidance policies performed by KKS Advisors and Prof. George Serafeim of Harvard Business School using FactSet Guidance data.

8 The Private Litigation Reform Act of 1995 9 Call, A.C., S. Chen, A. Esplin and B. Miao. “Long-term earnings guidance:

Implications for managerial and investor short-termism.” May 2016. http://www.hbs.edu/faculty/conferences/2016-imo/Documents/LTMF_May%2022%202016.pdf

10 Jamie Dimon and Warren Buffett WSJ oped “Short-Termism is Harming the Economy”

11 For disclosure purposes, in 2016, Schroders, a UK-based company, elected to end its quarterly earnings. However, this article is intended only to further the discussion on this topic, and refl ects the views and opinions of the author only, and does not necessarily refl ect the views of Schroders.

Less can be more: How to improve fi nancial reporting26

the largest banks in the US with billion and trillion dollar balance sheets managed to meet or exceed earnings per share (EPS) expectations 94% of the time!

From the perspective of a buy-side research analyst, three changes could potentially have meaningful benefi ts:

1. Investors, industry groups and regulators should urge management to curtail guidance

2. Replace 10-Q reporting with streamlined, periodic reporting of key performance indicators

3. Reform regulatory reporting so that semi-annual and annual fi lings become more meaningful to investors

Much like the law of diminishing marginal utility, we think frequent ad-hoc guidance should be replaced with periodic news releases on Form 8K. The information could be a mix of lightly audit fi nancials with an explanation of material changes in revenues and earnings that occurred since the last annual or semiannual fi ling. These key performance indicator updates, provided every six weeks, would refl ect how the market functions today while being less expensive and time consuming. As this reporting replaces guidance, it would hopefully reduce the incentive for period-end fi nancial statement manipulation.

ConclusionRecently there have been calls for the SEC to study if changing from quarterly reporting to a six-month system would save money and allow for greater investment fl exibility. While this is a good thing, we would caution that simply moving from quarterly to semi-annual reporting would be less than ideal, as studies from the UK suggest that it creates ineffi ciencies as investors seek, and trade, on alternative sources of information. Instead, we believe investors would benefi t from Ks and a detailed semi-

annual report with an updated format based on investment principles, versus an archaic prescriptive approach. This can be accomplished by providing industries and issuers greater fl exibility to customize information that is more valuable to providers of debt or equity capital.

Greater long-term transparency, less short-term noise. Now that’s something this credit analyst would raise a beer to.

Less can be more: How to improve fi nancial reporting 27

Take nothing for granted: Portfolio construction in today’s great unknown environment

Take nothing for granted:Portfolio construction in today’s great unknown environment

They say you can’t have your cake and eat it too. Yet, over the last few years, investors have enjoyed a fantastic mix of high returns, low volatility and low correlations. All of these factors have been tailwinds to portfolio construction. There are uncertainties facing each of these key inputs and it is time to ask the ‘what-ifs’ and plan the ‘to-dos’ in light of these new circumstances. We off er a framework for thinking about portfolio construction going forward, crucially questioning our assumptions, and taking nothing for granted.

The portfolio construction challengeClassic portfolio theory suggests that more returns can be gained if more risk is taken. The curve that captures this is the effi cient frontier. A ‘good’ effi cient frontier is one which is upward sloping and steep, refl ecting a good trade-off between returns and risk.However, levels and slopes of effi cient frontiers can change. Over the last six decades, the realized effi cient frontier of a two asset combination between equities and bonds, using the decade’s realized return, volatility and correlation, have yielded very different effi cient frontiers. The effi cient frontier of the 1950’s can probably be deemed as the ‘classical’ effi cient frontier, which is steep and upward sloping. Contrast this with the effi cient frontier of the 1980’s;

even though the slope of that curve was much fl atter than the 50’s, the level of returns from equities and bonds was signifi cantly higher. This was obviously a great decade for returns.Now consider the effi cient frontiers of the 1970’s and 2000’s, the former of which were characterized by high infl ation while the latter period was marred by the Global Financial Crisis (GFC). The effi cient frontier of the 70’s appears as a dot relative to other decades’ effi cient frontiers due to the narrow range of returns. In that decade, equities returned 5.5% while bonds returned 7.3%, therefore there was very little opportunity to harness returns from just equities and bonds. In the 2000’s, equities loss 1% while bonds returned 6%. Investors were rewarded to be predominantly invested in bonds during that decade.

0

5

10

15

20

25

0 2 4 6 8 10 12 14 16 18

Return, %

Risk, %1950s 1960s 1970s 1980s 1990s 2000s

456789

10

5.688 5.69 5.692 5.694 5.696 5.698 5.7 5.702

Return, %

Risk, %

456789

10

4.3 4.4 4.5 4.6 4.7 4.8 4.9 5 5.1 5.2

Return, %

Risk, %

Clement Yong, CFAAnalyst, Multi-Asset

Remi Olu-Pitan, CFAFund Manager, Multi-Asset

Olu-Pitan, CFAanager,sset

Figure 1: Effi cient frontiers are not static1970s

2000s

Source: Schroders, Bloomberg, Thomson Reuters Datastream, Shiller for the respective periods stated. Effi cient frontiers were generated using realized historical returns, volatility and correlation for equities and bonds for each decade. Results shown are for illustrative purposes only.

28

Take nothing for granted: Portfolio construction in today’s great unknown environment

In recent years, the effi cient frontier has benefi ted from the decline in volatility, below average correlations and above average returns. This has been due to low macroeconomic volatility, further magnifi ed by the unconventional measures adopted by central banks globally. GDP growth has been extraordinarily stable while infl ation has steadied at the 2% mark.

Financial market volatility has fallen in tandem with macroeconomic volatility, aiding fi nancial assets. Equity returns have been stellar for a majority of the last 30 years while bond yields have been on a clear downward trend. Current volatility is signifi cantly low, lying at the bottom of its long-term history. Finally, the negative correlation between equities and bonds has meant that investors have found diversifi cation easily.

The tailwinds that have delivered the ideal effi cient frontier through a combination of high returns and less risk are fading. As the economic cycle matures and central banks move from quantitative easing to tightening, the suppression of fi nancial market volatility will come to an end and correlations are likely to rise. Furthermore, protectionist rhetoric on international trade could well bring back more macro volatility to the table. Finally, while we can’t say for sure whether and when infl ation will structurally move higher, it is worth remembering that infl ation can rise a lot quicker than many expect, like the experience in the 60’s. These signposts are worrying, turning tailwinds into headwinds, and we expect future returns to be lower than average.

-4.0%

1.0%

6.0%

11.0%

16.0%

21.0%

1949

1952

1955

1957

1960

1962

1965

1968

1970

1973

1975

1978

1980

1983

1986

1988

1991

1993

1996

1999

2001

2004

2006

2009

2011

2014

2017

0.0%

5.0%

10.0%

15.0%

20.0%

25.0%

Equities BondsMin Median Max Current

-4.0%-2.0%0.0%2.0%4.0%6.0%8.0%10.0%12.0%14.0%16.0%

1949

1952

1955

1957

1960

1962

1965

1968

1970

1973

1975

1978

1980

1983

1986

1988

1991

1993

1996

1999

2001

2004

2006

2009

2011

2014

2017

-0.8-0.6-0.4-0.2

00.20.40.60.8

1949

1952

1955

1957

1960

1962

1965

1968

1970

1973

1975

1978

1980

1983

1986

1988

1991

1993

1996

1999

2001

2004

2006

2009

2011

2014

2017

Rolling correl Correl (LT avg)

Figure 2: Lower macro volatility = lower fi nancial market

US real GDP (rolling 5y vol)

Volatility

US CPI

Correlation

Source: Schroders, Bloomberg, Thomson Reuters Datastream. Data as at 31 August 2018. Rolling charts displayed are rolling fi ve-year charts. Ranges are based on data from December 1949, with a fi ve-year lookback. Data shown refl ect past results which offer no guarantee of future results.

Returns High

Volatility Low Correlations Low

Returns ?

Volatility ? Correlations ?

29

Figure 3: Future returns are expected to be lower than average

Source: Schroders, Bloomberg, Thomson Reuters Datastream. Future returns are based on Schroders economic team’s 30-year forecast. Data as of August 31, 2018. Historical data based on the last 30 years.

Tailwinds to headwindsThe ‘current’ effi cient frontier refl ects the desirable mix of higher returns, low volatility and correlations, which investors have enjoyed in recent years. A minor shift in returns, volatility and correlations will have a meaningful impact on the portfolio construction results. As these variables change, the risk that the portfolio construction yields disappointing results will rise, as we depict in Figure 4. If we assume a lower expected return, say 5.3%,1 we see a material shift in the effi cient frontier downwards and to the right (‘lower returns’). Next, if we assume equity volatility moves back to its historical average (~15%), the effi cient frontier shifts even lower. Finally, we adjusted the assumption of correlation from mildly negative to positive, further exacerbating the downward trend.

Figure 4: The portfolio construction challenge

Source: Schroders. Effi cient frontier consists of US equities and US government bonds. Current: Expected equity returns 7%, volatility 10%, correlation -0.3. Lower returns: expected equity returns 5.3%. Lower returns, higher volatility: expected returns 5.3%, volatility 15%. Lower returns, higher volatility and correlations: expected returns 5.3%, volatility 15%, correlations 0.3. Expected returns and volatility for bonds are 3.9% and 7% respectively throughout. For illustrative purposes only. Does not refl ect any actual portfolio.

The fi nal result is unnerving, especially considering how far it is from the ‘current’ effi cient frontier. This is precisely what concerns us as the ‘new’ effi cient frontier is telling us that returns are going to be much harder to achieve and requires us to take on more risk in order to achieve returns.

1  Schroders Economics team forecast

Because volatility is expected to increase, investors have tobe prepared to be exposed to more volatility even if they stick to their current asset allocation. Furthermore, the heightened relationship between equities and bonds means that we have to deal with a ‘fl atter’ effi cient frontier. Risk is no longer rewarded as handsomely as before.

A framework to improve the odds of an optimal outcome

Many institutions still need to achieve a reasonable return, around 7% on average. We interviewed a large number of investors about how much return they expect to generate going forward.2 More than 40% of global investors expect more than 7% return while this proportion is higher in North America, at 54%. This is fairly refl ective of an infl ation +4% target.

Figure 5: Investors still expect a high return on average

Source: Schroders. Schroders’ 2017 Institutional Investor Study. See footnote 2 for details.

To improve the odds of meeting the required 7% return, or equivalent infl ation +4% return, investors will need to embrace the riskier areas that offer value. Given where starting valuations of most traditional assets are, we expect returns to be harder to come by. Hence investors should consider some of the following approaches:

– Increase allocation to laggards such as emerging market (EM) assets

– Increase allocation to active managers – Adding leverage

Emerging market assetsEmerging market assets generally offer a premium over developed markets. While this premium has not always been consistent, we do expect return prospects of emerging markets from both the equity and debt side to be higher than their developed market counterparts (left hand chart of Figure 6). While emerging markets offer value, valuations can diverge from fair value for a considerable length of time. In fact, emerging markets have been trading at a large discount as compared to the US (right hand chart of Figure 6) for about a decade but yet, US equities outperformance has been well-documented.

2 500 institutional respondents: 115 in North America, 200 in Europe, 150 in Asia and 35 in Latin America. Respondents were sourced from 15 different countries.

0.0%1.0%2.0%3.0%4.0%5.0%6.0%7.0%8.0%

Global equities Global bonds

Historical Future (Schroders forecast)

Historical vs. Future returns

3.03.54.04.55.05.56.06.57.07.5

4 6 8 10 12 14 16

Return, %

Risk, %Current Lower returnsLower returns, higher volatility Lower returns, higher volatility and correlations

0%

5%

10%

15%

20%

25%

30%

35%

40%

45%

0% or less 1-4% 5-6% 7-9% 10% +

Global North America

Investor’s expected returns of equity

Take nothing for granted: Portfolio construction in today’s great unknown environment30

Emerging market assets, however, are also much more volatile especially when currency valuations are taken into account too. Investors therefore have to embrace more volatility and possibly adopt a longer-term view to unlock the emerging market premium.

Active managersInvesting in bottom-up, unconstrained active managers allows investors to tap into alpha potential. This approach is especially useful in certain environments, which we will discuss later on in the paper. However, alpha is never a guarantee and the alpha delivered by active managers can vary over time. Figure 7 shows that the proportion of US equity managers outperforming their benchmarks can be inconsistent, or ‘lumpy’ over time. Investors also need to accept more idiosyncratic risk when investing in active managers, i.e. not all the risk is refl ected in the macro environment and the risk may actually be style, or even manager specifi c.

Figure 7: Percentage of US equity managers outperforming their benchmarks

Source: Schroders, Morningstar, Data as of March 2017. Rolling 3 years shown.

LeverageLeverage can help enhance returns as it magnifi es portfolio positions. Consider a hypothetical leveraged strategy which identifi es the allocation between equities and bonds that yields the highest risk-adjusted returns on the effi cient

frontier.3 The portfolio is then scaled to a certain target risk, say 8%, using leverage.

The identifi cation of the ‘right’ portfolio is crucial and will heavily depend on the correct correlation assumption. In Figure 8, we show what happens to the realized (or ex-post) volatility of this leveraged strategy if the correlation is stretched from -0.3 (current correlation between equities and bonds) to +0.5. In essence, this measures how much excess risk is taken as correlation rises. The strategy takes on too much leverage and therefore excess risk is taken.

Contrast this with a diversifi ed strategy, which is fully invested but not leveraged. We have basically assumed our diversifi ed strategy consists of the equity/bond split that creates a portfolio that has 8% risk while assuming current correlations.4 When the correlation between equities and bonds start rising, as expected, the realized volatility of the portfolio also rises, but at a lower rate of change as a leveraged portfolio. Leverage therefore amplifi es the issue with rising correlations and can lead to excess risk taking more quickly.

Figure 8: Leverage can lead to excess risk taking

Source: Schroders, Thomson Reuters Datastream, Bloomberg, Data as of August 31, 2018. For illustration only. Does not refl ect any actual portfolio.

3 In portfolio theory, this is referred to as the tangency portfolio. 4 Clearly, this portfolio will have a worse starting risk/return trade-off than the

tangency portfolio to achieve the 8% risk budget without leverage.

Figure 6: Emerging markets offer a premium, but valuations can diverge for a long time

012345678910

Equities BondsDeveloped Emerging

0.00%10.00%20.00%30.00%40.00%50.00%60.00%70.00%80.00%

12/1/1992

5/1/1994

10/1/1995

3/1/1997

8/1/1998

1/1/2000

6/1/2001

11/1/2002

4/1/2004

9/1/2005

2/1/2007

7/1/2008

12/1/2009

5/1/2011

10/1/2012

3/1/2014

8/1/2015

1/1/2017

7.0%7.5%8.0%8.5%9.0%9.5%

10.0%10.5%11.0%11.5%12.0%

-0.3

-0.2

5

-0.2

-0.1

5

-0.1

-0.0

5 0

0.05 0.

1

0.15 0.

2

0.25 0.

3

0.35 0.

4

0.45 0.

5

Volatility

CorrelationLeveraged volatility Unleveraged volatility

Increase in risk dueto rising correlation

0

1

2

3

4

5

6

1/1/

1995

12/1

/199

5

11/1

/199

6

10/1

/199

7

9/1/

1998

8/1/

1999

7/1/

2000

6/1/

2001

5/1/

2002

4/1/

2003

3/1/

2004

2/1/

2005

1/1/

2006

12/1

/200

6

11/1

/200

7

10/1

/200

8

9/1/

2009

8/1/

2010

7/1/

2011

6/1/

2012

5/1/

2013

4/1/

2014

3/1/

2015

2/1/

2016

1/1/

2017

12/1

/201

7

US equities Emerging equities

Source: Schroders economic team, Bloomberg, Thomson Reuters Datastream. Data as of August 31, 2018.

Expected returns EM vs. DM Price to book

Take nothing for granted: Portfolio construction in today’s great unknown environment 31

The challenge with all of the aforementioned approaches, emerging markets, active management, leverage, is that it involves taking on more risk. However, not all investors can take on more risk such as underfunded plans. Likewise, not all investors have a long time horizon and therefore the path of returns matter. A volatile sequence of returns can have an adverse impact on the desired outcome. For participants of Defi ned Contribution plans, high returns in the early years when cash fl ows are usually small followed by low returns in later years when cash fl ows are larger can be problematic. In contrast, retirees and endowment funds in the net withdrawal phase are adversely impacted by lower near term returns versus stronger returns in the later years.

Narrowing the range of outcomesTaking on more risk to generate higher returns is thus not necessarily the answer for all investors. To increase the probability of a positive outcome, investors need to stay invested but favor strategies that help to narrow the range of outcomes, provide a smoother path of returns and letting the benefi ts of compounded returns come through.

Some strategies which we think allow investors to do so are: – Private assets – Minimum volatility equity – Multi-asset strategies

Private assetsMany plans have diversifi ed into alternative asset classes in recent years with the intention of eking out return sources that are less correlated to equities. Not all alternatives are equal, however, and an objective assessment of risk premium is crucial. In the US for instance, private equity (buyouts only in our example) has generally outperformed public markets but this spread has narrowed in recent years. Going forward, we expect this spread to return to its historical norms and offer an important premium over public markets.

Private debt markets tend to offer wider net spreads to liquid markets due to lower default rates and higher recovery rates. Spreads do vary in the private debt world so investors have to ensure that they are getting a decent trade-off for the risk that they are taking.

We expect private assets to outperform. However, investing in private assets require a signifi cant amount of resources, and manager selection within the asset class is crucial. Private assets would also require a long-term horizon for investors to unlock their illiquidity premium.

Figure 9: Consider non-traditional alternative assets

US large cap is S&P 500 index; US small cap is Russell 2000 index; public real estate is MSCI USA REIT index; private equity are PreqIn indices; private real estate is NCREIF Fund Index - Open End Diversifi ed Core Equity Fund Index. All fi gures are total returns and private asset returns are net of fees. Sources: FTSE Russell, MSCI, NCREIF, Preqin, and Thomson Reuters Datastream. Data as of 12/31/2017. Note: IG and HY corporate bond credit loss rates incorporate default losses (default rates adjusted for recovery rates) and price changes arising from changes in credit quality (net downgrade losses). Investors in private debt will not generally experience downgrade losses (or upgrade gains) as the credit spread component typically remains unchanged unless there is an impairment (high risk of default). Consequently, private debt loss rates above only refl ect default losses. Figures are shown for illustrative purposes only and may not be refl ective of credit spreads or default experience on any individual investment or portfolio. Source: Bank of America Merrill Lynch, Callan Associates, CBRE, De Montfort University, NEOS, Preqin and Schroders

Minimum volatility equityMinimum or low volatility equity has been a popular equity style among investors. The style prescribes that there is a low volatility anomaly among stocks, whereby low volatility stocks tend outperform high volatility ones. Furthermore, the style is attractive as investors stay invested in equities while bringing down the risk of that allocation.

However, the style is very much predicated on using volatility/standard deviation as a sole measure of outperformance. We do not have a view on which equity style will outperform, but we do believe that using a single, backward looking measure in isolation is suboptimal. Such an approach ignores fundamental variables such as valuations, quality and interest rate sensitivity. A more holistic approach to investing in equities, which combines a wider range of risk and valuation measures, is a better approach.

11.29.7

17.6

8.306.3

8.5

16.5

5.30

02468

101214161820

Since 31/12/00 Since 31/12/07 Since 31/12/12 Expected

Private equity buyout US large caps

Private assets typically outperform, net of fees (%)

0.81.9 1.7

1.0

4.14.3

6.0

0123456789

IG c

orpo

rate

bon

ds

Seni

or in

fra

debt

Seni

or r

eal e

stat

e de

bt

HY

corp

orat

e bo

nds

Juni

or in

fra

debt

Mid

-mar

ket d

irec

tle

ndin

g

SME

Loan

s

Net credit spreads (%) = higher spreads - lower default rates

Take nothing for granted: Portfolio construction in today’s great unknown environment32

Figure 11: ....which means an increase in probability of underperformance

0.0%

2.0%

4.0%

6.0%

8.0%

10.0%

12.0%

Lowest volnow

Mid vol now Highest volnow

Returns

Source: Schroders, Thomson Reuters Datastream, Bloomberg, data as of August 31, 2018. Five-year volatility lookback, Five-year subsequent returns. Data shown refl ect past results which offer no guarantee of future results.

Source: Schroders, Thomson Reuters Datastream, Bloomberg, data as of August 31, 2018. Five-year volatility lookback, Five-year subsequent returns. Data shown refl ect past results which offer no guarantee of future results.

US equities

We can show the danger of using volatility as a sole measure for future performance in the next two charts. In this analysis, we look at subsequent returns following low/medium/high volatility periods in a few markets. The fi rst clear conclusion is that subsequent returns are lower following a period of low volatility. Furthermore, the average returns for US and emerging equities actually tend to be higher following periods of mid to high volatility. The pattern is less clear cut for developed equities but it is safe to say that low volatility in the present can pose a material danger to returns in the future.

Lower average returns directly translates to a higher likelihood of future underperformance. In fact, the probability of underperformance for US equities is around 65% when volatility is high and 100% for both developed and emerging equities. We’ve assumed that the performance target here is 7%, but clearly the target is irrelevant as the pattern will be exactly the same even if the actual probability differs. This suggests some vulnerabilities for minimum volatility equity styles.

Multi-Asset strategiesMulti-Asset strategies generally aim to deliver a smoother path of returns by diversifying away from traditional asset classes. Nevertheless, a few of these strategies have taken this to the extreme by targeting a low to zero correlation to equities and/or aim to deliver a tail risk hedge. This has come at the expense of returns as there is no such thing as a free lunch. Multi-Asset strategies, in our view, should narrow the range of outcomes to ensure a favorable path of returns, while not completely sacrifi cing returns.

We illustrate this in Figure 12, which shows the distribution of annualized returns in a simulation for a few strategies. Equities, as expected, offers the highest median return, but at a much bigger range too. Bonds are safer and has a narrower range of outcomes but at much lower returns. A 60/40 has historically offered an attractive range of outcomes but at some expense of returns. Multi-Asset tries to balance the returns vs range of outcomes dilemma and can therefore play a crucial role in portfolios.

DM ex US equities EM equities

-6.0%

-4.0%

-2.0%

0.0%

2.0%

4.0%

6.0%

8.0%

10.0%

Lowest volnow

Mid vol now Highest volnow

Returns

-8.0%-6.0%-4.0%-2.0%0.0%2.0%4.0%6.0%8.0%

10.0%12.0%

Lowest volnow

Mid vol now Highest volnow

Returns

Figure 10: Lower volatility may imply lower returns….

Prob of underperformance

0.0%

10.0%

20.0%

30.0%

40.0%

50.0%

60.0%

70.0%

Lowest volnow

Mid volnow

Highest volnow

US equities DM ex US equities EM equities

Prob of underperformance

0.0%

20.0%

40.0%

60.0%

80.0%

100.0%

120.0%

Lowest volnow

Mid volnow

Highest volnow

Prof of underperformance

0.0%

20.0%

40.0%

60.0%

80.0%

100.0%

120.0%

Lowest volnow

Mid vol now Highest volnow

Take nothing for granted: Portfolio construction in today’s great unknown environment 33

Figure 12: Narrowing the range of outcomes

Source: Schroders. Monte Carlo simulation for each strategy, 30-year annualized returns depicted on vertical axis. The hypothetical results shown must be considered as no more than an approximate representation of performance, not as indicative of how it would have performed in the past or future. It is the result of statistical modeling, with the benefi t of hindsight, based on a number of assumptions and there are a number of material limitations on the retrospective reconstruction of any performance results from performance records. Monte Carlo performance simulations have inherent limitations, including modeling risk and probability (or tail) risk; that is, simulation results may not account for highly unlikely positive or negative outcomes which can occur in live portfolios. This data is provided to you for information purposes only as at today’s date and should not be relied on to predict possible future performance. There can be no guarantee that these or any simulated results will occur, generate a positive return or protect against loss of principal.

We believe investors should favor multi-asset strategies that are fl exible and adaptable to different market environments. Strategies that target low to zero correlation are anything but fl exible. For example, correlations are not static and subject to change. As a rule of thumb, high correlations are usually associated with low returns. However, there can be ‘good’ types of high correlation, such as a broad bull market rally. When we looked at correlations calculated on a cross-asset basis,5 we found little difference between the average monthly returns of assets in high vs low correlation environments. We actually found that the average monthly returns during months of high cross-asset correlations to be higher than months of low cross-asset correlations, rather contrary to the rule of thumb. This represents very conducive periods in markets where performance is broad and strong across assets.

We also analyzed bottom-up correlation, which in this case refers to the correlation between constituent stocks of the S&P 500. This is essentially a measure of how narrow or broad the performance in the market is. We can immediately see that the difference in the average monthly returns between high vs low bottom-up correlations is much starker. Furthermore, the returns when bottom-up correlation is high tends to be much lower than when bottom-up correlation is low. This is because when correlations between stocks start rising, it is likely that signifi cant macro factors have fi ltered through the index and to individual stocks, which is usually a bad environment for risk assets.

5  We have used US equities, emerging equities, US government bonds and US credit in the universe of assets.

Figure 13: Correlation perspectives

Source: Schroders, Thomson Reuters Datastream, Bloomberg, Data as of August 31, 2018.

Using these two distinct views of correlations allows us to discern what kind of environment we are in. We introduce the concept of the correlation ‘clock’, created from plotting cross-asset and bottom-up correlations on two separate axes. We can then relate the different combinations to different parts of the cycle:

– High cross-asset and bottom-up correlation (top right) – This environment is typically when the economy is

in slowdown. Macro issues have already impacted asset classes on a high level and are now beginning to impact underlying stocks too. An example of this was 2007, which was obviously the period directly before the GFC.

– Low cross-asset correlation, high bottom-up correlation (bottom right) – In this scenario, the economy is now in recession.

Macro factors that have impacted markets are still present and are affecting underlying stocks but recovery in some assets are on the way. The bursting of the Dotcom bubble is a good historical example.

– Low cross-asset and bottom-up correlation (bottom left) – This scenario tends to be a great environment for

investment as investors can add value through asset allocation and security selection. The most recent example of this would be 2017, which saw a strong bull market across risk assets. Stock pickers would fi nd it easy to identify the winners from the losers.

– High cross-asset correlation, low bottom-up correlation (top left) – This scenario is usually a late cycle expansion phase.

Performance has been strong across asset classes but there is still breadth in the underlying markets. Pure macro investors may struggle to identify winning asset classes but stock pickers are still able to sift through winning stocks. An example of this period was the year leading up to the peak of the Dotcom bubble.

0.0%

1.0%

2.0%

3.0%

4.0%

5.0%

6.0%

7.0%

8.0%

9.0%

Equities Bonds 60/40 Multi-Asset

Bottom 10% Average Top 10%

-1.00%

-0.50%

0.00%

0.50%

1.00%

1.50%

2.00%

High cross-assetcorrelation

(47% of history)

Low cross-assetcorrelation

(53% of history)

High bottom-upcorrelation

(41% of history)

Low bottom-upcorrelation

(59% of history)

Returns (cross asset average) Returns (S&P 500)

Top-down correlation Bottom-up correlation

Take nothing for granted: Portfolio construction in today’s great unknown environment34

ConclusionInvestors have enjoyed a great environment for portfolio construction over the last few years. However, tailwinds are turning into headwinds and going forward, portfolio construction will be more diffi cult. Now is an opportune time to review what can be done to navigate the future environment.

If investors have the capacity and time horizon to embrace higher risk, we believe that there are certain areas that are attractive. Emerging market assets can off er a premium over developed markets, while active managers can boost returns through alpha. Leverage can also further boost returns. All of these approaches have its idiosyncrasies that investors need to embrace.

If investors are unable to take on more risk, then one should think about narrowing the range of potential outcomes and smooth the path of returns. Private assets can off er a ‘true’ alternative exposure but requires large amount of resources. Equity styles such as minimum volatility can also be useful, but may be susceptible to ignoring fundamental factors. Multi-Asset strategies should be fl exible in navigating future environments. Investors need to focus on multi-asset strategies which are truly fl exible in their approach as well as balance the trade-off between returns and range of outcomes well.

Figure 14: The correlation ‘clock’

-1.00%

-0.80%

-0.60%

-0.40%

-0.20%

0.00%

-0.80%-0.60%-0.40%-0.20%0.00%

Returns (cross asset average)Returns (S&P 500)

0.00%

0.50%

1.00%

1.50%

2.00%

0.00%

0.50%

1.00%

1.50%

Bottom-up correlation

Cross-asset correlation

− Expansion phase

− Alpha matters

− ~38% of history

− Recovery phase

− Risk on

− ~21% of history

− Recession phase

− Asset allocation matters

− ~31% of history

− Slowdown phase

− Risk-off

− ~9% of history

Source: Schroders, Thomson Reuters Datastream, Bloomberg, Data as of August 31, 2018.

Multi-Asset strategies that focus on being adaptable to different regimes are able to take full advantage of bottom-up correlation. For example, in the current environment of rising cross-asset correlation but low bottom-up correlation, active/fl exible investors will

benefi t from the alpha of their active holdings while passive investors will have to bear the full brunt of rising cross-asset correlations. Having different perspectives in assessing correlations can only be complemented if investors are active.

Take nothing for granted: Portfolio construction in today’s great unknown environment

A word about simulated results

Any simulated results mentioned must be considered as no more than an approximate representation of the portfolio’s performance, not as indicative of how it would have performed in the past or will perform in the future. It is the result of statistical modeling, with the benefi t of hindsight, based on a number of assumptions and there are a number of material limitations on the retrospective reconstruction of any performance results from performance records. For example, it may not take into account any dealing costs or liquidity issues which would have affected the strategy’s performance. In addition, gross returns would be lower if applicable management fees and expenses were factored in to the calculation. There can be no assurance that this performance could actually have been achieved using tools and data available at the time. No representation is made that the particular combination of investments would have been selected at the commencement date, held for the period shown, or the performance achieved. This data is provided to you for information purposes only as of the dates of this material and should not be relied on to predict possible future performance.

35

Important information: The views and opinions contained herein are those of the cited authors, and do not necessarily represent Schroder Investment Management North America Inc.’s (SIMNA Inc.) house view. These views and opinions are subject to change. All investments, domestic and foreign, involve risks including the risk of possible loss of principal. The market value of the portfolio may decline as a result of a number of factors, including adverse economic and market conditions, prospects of stocks in the portfolio, changing interest rates, and real or perceived adverse competitive industry conditions. Investing overseas involves special risks including among others, risks related to political or economic instability, foreign currency (such as exchange, valuation, and fl uctuation) risk, market entry or exit restrictions, illiquidity and taxation. Emerging markets pose greater risks than investments in developed markets. Products with high turnover may experience high transaction costs. Sectors/regions/asset classes mentioned are for illustrative purposes only and should not be viewed as a recommendation to buy/sell. Simulated and backtested results in general must be considered as no more than an approximate representation of the portfolio’s performance, not as indicative of how it would have performed in the past. It is the result of statistical modelling, with the benefi t of hindsight, based on a number of assumptions and there are a number of material limitations on the retrospective reconstruction of any performance results from performance records. For example, it may not take into account any dealing costs or liquidity issues which would have affected the strategy’s performance. This data should not be relied on to predict possible future performance. This newsletter is intended to be for information purposes only and it is not intended as promotional material in any respect. The material is not intended as an offer or solicitation for the purchase or sale of any fi nancial instrument mentioned in this commentary. The material is not intended to provide, and should not be relied on for accounting, legal or tax advice, or investment recommendations. Information herein has been obtained from sources we believe to be reliable but Schroder Investment Management North America Inc. does not warrant its completeness or accuracy. No responsibility can be accepted for errors of facts obtained from third parties. Reliance should not be placed on the views and information in the document when taking individual investment and / or strategic decisions. Past performance is no guarantee of future results. The opinions stated in this document include some forecasted views. We believe that we are basing our expectations and beliefs on reasonable assumptions within the bounds of what we currently know. However, there is no guarantee that any forecasts or opinions will be realized. This document does not constitute an offer to sell or any solicitation of any offer to buy securities or any other instrument described in this document.

SIMNA Inc. is registered as an investment adviser with the US SEC and as a Portfolio Manager with the securities regulatory authorities in Alberta, British Columbia, Manitoba, Nova Scotia, Ontario, Quebec and Saskatchewan. It provides asset management products and services to clients in the United States and Canada. Schroder Fund Advisors LLC (SFA) markets certain investment vehicles for which SIMNA Inc. is an investment adviser. SFA is a wholly-owned subsidiary of SIMNA Inc. and is registered as a limited purpose broker-dealer with the Financial Industry Regulatory Authority and as an Exempt Market Dealer with the securities regulatory authorities in Alberta, British Columbia, Manitoba, New Brunswick, Nova Scotia, Ontario, Quebec, Saskatchewan, Newfoundland and Labrador. This document does not purport to provide investment advice and the information contained in this material is for informational purposes and not to engage in a trading activities. It does not purport to describe the business or affairs of any issuer and is not being provided for delivery to or review by any prospective purchaser so as to assist the prospective purchaser to make an investment decision in respect of securities being sold in a distribution. Schroder Adveq Management US Inc. (Schroder Adveq) is registered as an investment adviser with the SEC. It provides asset management products and services to clients in the United States and Canada. Schroder Fund Advisors LLC (“SFA”) markets certain investment vehicles for which Schroder Adveq US is an investment adviser. SFA is an affi liate of Schroder Adveq US and is registered as a limited purpose broker-dealer with the Financial Industry Regulatory Authority. Schroder Adveq, SIMNA Inc. and SFA each are indirect, wholly-owned subsidiaries of Schroders plc, a UK public company with shares listed on the London Stock Exchange. Further information about Schroders can be found at www.schroders.com/us or www.schroders.com/ca. Schroder Investment Management North America Inc., 7 Bryant Park, New York, NY, 10018-3706, (212) 641-3800.

INVSTHOR09

@SchrodersUS

schroders.com/usschroders.com/ca

Schroder Investment Management North America Inc.7 Bryant Park, New York, NY 10018-3706(212) 641-3800