china; $41bn (2015) china; $178bn (2020)

DISCLOSURE APPENDIX AT THE BACK OF THIS REPORT CONTAINS IMPORTANT DISCLOSURES, ANALYST CERTIFICATIONS, LEGAL ENTITY DISCLOSURE AND THE STATUS OF NON-US ANALYSTS. US Disclosure: Credit Suisse does and seeks to do business with companies covered in its research reports. As a result, investors should be aware that the Firm may have a conflict of interest that could affect the objectivity of this report. Investors should consider this report as only a single factor in making their investment decision.

10 April 2017 Global

Equity Research Food Retail

Global online grocery The Credit Suisse Connections Series leverages our

exceptional breadth of macro and micro research to deliver

incisive cross-sector and cross-border thematic insights for

our clients.

Research Analysts

Stewart McGuire, CFA

44 20 7888 6531

[email protected]

Olga Bystrova, CFA

7 495 967 8871

[email protected]

A-Hyung Cho

82 2 3707 3735

[email protected]

Antonio Gonzalez, CFA

52 55 5283 8921

[email protected]

Stephen Ju

212 325 8662

[email protected]

Edward J. Kelly, CFA, CPA

212 325 3241

[email protected]

Andre Kukhnin, CFA

44 20 7888 0350

[email protected]

Victoria Petrova

49 69 75 38 2272

[email protected]

Grant Saligari

61 3 9280 1720

[email protected]

Tobias Stingelin, CFA

5511 3701 6301

[email protected]

Keiichi Yoneshima

81 3 4550 9740

[email protected]

Evan Zhou

852 2101 6745

[email protected]

CONNECTIONS SERIES

On the cusp of a seismic shift Online grocery adoption is increasing worldwide and threatens to upend the

economics of all food retailers, regardless of geography. We analyse the

options available to grocers and conclude that in developed countries, only

centralised, automated fulfilment can profitably meet the needs of customers.

The risks to incumbent food retailers are widespread and serious, particularly

for those companies with large supermarkets located too far from customers;

these risks increase with the size and duration of their store lease portfolios.

This report is published in conjunction with a report from our capital goods

team on Warehouse Automation and Robotics, specifically looking at the

companies with exposure to grocery supply chain automation.

■ Technology disruption is coming to grocery. The barriers to efficient

online grocery operations are falling as companies solve the main cost

problems of picking and delivery. We estimate that $37bn of online grocery

infrastructure will be spent in the US to achieve a 25% penetration rate.

■ Business model choice will separate winners from losers. We identify

~20 different permutations of picking and delivery methods within online

grocery. Very few meet the criteria of high service quality and reasonable

profitability due to the exceptionally difficult logistics and the low margins.

■ IGD projects double-digit CAGRs for online grocery in all key markets.

We expect developed regions to rapidly catch up to the UK in online

grocery. If the US had UK-levels of adoption, online grocery revenue would

be US$70bn vs. US$7bn currently.

■ Key companies to watch. We identify a number of companies that are

either leading, trialling or at risk from online grocery (often all three),

demonstrating the nascent and uncertain nature of the channel.

■ CORRECTION: In this version of the report we have corrected the RHS

axis of Figure 1, as the full values were truncated in the original publication.

Figure 1: Online grocery revenue – double-digit CAGRs to 2020

Source: IGD, Credit Suisse research

$0

$100

$200

$300

$400

$500

$600

$0bn

$10bn

$20bn

$30bn

China UK Japan US France S. Korea Germany Australia Belgium Netherlands

2015 (LHS)

2020 (LHS)

2015 Per capita (RHS)

2020 Per capita (RHS)

China; $41bn (2015)China; $178bn (2020)

https://plus.credit-suisse.com/u/_X0Qx2AN-V1Ni

10 April 2017

Global online grocery 2

Key tables and charts

Figure 2: Margin potential for centralised, automated fulfilment is materially higher than traditional grocery

Source: Credit Suisse research

Figure 3: Attitudes about buying groceries online Figure 4: Have shopped online for groceries (%)

Source: Nielsen Source: Nielsen

Figure 5: Ocado KPIs and LfLs Figure 6: Mixed palletising robot

Source: Company data, Credit Suisse research Source: Kuka

Picking location In-store Dark store Hybrid 3rd-party Centralised In-store Centralised Centralised Centralised

Picking technology Staff Staff Staff 3rd-party Staff Staff Staff Robotic Robotic

Delivery location Home Home Home Home Home C&C C&C Home Home

Delivery mechanism Driver Driver Driver 3rd-party Driver Customer Customer Driver Automated

Industry leader Tesco Tesco Peapod Instacart Ocado Leclerc n/a n/a n/a

Gross margin (avg. basket) 27% 27% 27% 27% 27% 25% 25% 27% 27%

Total direct costs -29% -26% -27% -6% -21% -22% -14% -19% -11%

Direct P&L per basket -1% 1% 1% 21% 6% 3% 11% 9% 17%

0% 20% 40% 60% 80% 100%

Global

Asia-Pacific

Europe

Africa / Middle East

Latin America

North America

Considerers Trialists Regulars Avoiders

0%

5%

10%

15%

20%

25%

30%

35%

40%

45%

China South

Korea

India Europe North

America

Africa /

Middle East

Latin

America

0%

5%

10%

15%

20%

80%

85%

90%

95%

100%

2011 2012 2013 2014 2015 2016

On time delivery or early

Order accuracy

LfLs (RHS)

10 April 2017


Executive summary “The factory [or warehouse] of the future will have only two employees, a man and a dog.

The man will be there to feed the dog. The dog will be there to keep the man from

touching the equipment.” – Warren Bennis1. For further reading into warehouse and

logistics automation, we refer you to today's publication from our capital goods team

Warehouse Automation and Robotics.

Our eight key takeaways for the global online grocery market are:

■ Online grocery growth will accelerate in all key markets. Improved mobile phone

interfaces, easier to navigate web pages, tailored features, more valuable promotions,

automated 24-hour pick-up points, delivery passes, integration with home automation

systems and same day delivery all add tremendous value to the customer. The offline

model will likely be seen as static and passé, with a declining value proposition.

■ Centralisation is vital for scale, high service quality and margin. The abundance

of business models is a result of grocers trying to move online inexpensively. Third-

party solutions (i.e. Instacart) add material costs to the customer and are likely to be

short-term, niche solutions. Using an existing store estate as a quasi-distribution

network is also a stopgap measure.

■ Automation and scale will solve the delivery problem. The need for centralisation

runs counter to delivery economics, which improve when the picking operations move

closer to customers. Potential solutions exist to reduce delivery costs, including

shrinking fulfilment centres and locating them nearer customers, stand-alone click and

collect facilities, sophisticated routing algorithms, and driverless delivery technologies.

■ Low labour costs help, but high labour costs may help more. The paradox of

labour within online grocery is that low labour costs facilitate online grocery (as we see

in China – at least for now), but high labour costs are needed to provide the incentive

to switch to an automated, scaled solution. This ultimately leads to a point where a

lower variable cost model establishes a cost advantage over the traditional store.

■ Most developed markets are amenable to online grocery. The main factors that

provide the conditions for online grocery to flourish are improving in all jurisdictions.

The capital costs of the most efficient online business models continue to fall, which in

turn expands the addressable market. Even a modest shift in market share to online

(say, 20% to 30%), is likely to have far-reaching, negative ramifications for incumbents.

■ Technology changes cannot be overstated. The ability to substitute capital for

labour is reaching a level where automation is making inroads into the most

economically challenging parts of online grocery: picking a basket of goods and

delivering it to the customer. Where grocery was traditionally a local venture, we expect

the scale benefits of technology to spread rapidly across borders.

■ Amazon is a threat to incumbents. Despite AmazonFresh's slower-than-expected

rollout, the strategic and tactical implications remain positive because grocery is: (i) the

largest consumer spend vertical; (ii) drives greater engagement from higher shopping

frequency; (iii) leverages existing fulfilment infrastructure; (iv) accelerates share gain in

adjacent categories; and (v) decreases friction on handling returns.

■ Potential for the adoption of existing marketplace models. We are starting to see

the third-party marketplace model gain traction in grocery. Amazon's arrangement with

Morrisons in the UK offers customers access to multiple credible food sources. Marks

& Spencer does not have an online food strategy, and it would be groundbreaking to

see M&S and Morrisons products listed alongside branded goods on Amazon.com.

1 Mark Fisher (1991), The Millionaire's Book of Quotations


10 April 2017


Key players within online grocery

This report is intended to show the opportunities and pitfalls of various business models

for online grocery. We are still at a very early stage in this channel shift with less than 1%

of global grocery revenues occurring online. The winners and losers from this shift are not

yet known, nor are they easy to predict because the outcome will be based largely on the

decisions that will be made by management teams over the next few years.

In this report, we have drawn together the perspectives of all our regional retail analysts.

We have identified a number of companies that are either market leaders in grocery (and

are therefore at risk), or are working on positioning for the shift to online, or both. These

include the companies listed in Figure 7. Out of all of the companies we cover, only one –

Ocado – stands out as a technological leader with no legacy assets, but this does not

necessarily make it the leader in a sector where access to capital and localised supply

chains remain significant barriers to entry.

We also refer readers to the report published from our capital goods team on Warehouse

Automation and Robotics, specifically looking at the companies with exposure to grocery

supply chain automation. The following players have the widest product offering, and

featured prominently in our discussions with industry experts: Dematic (KION), Swisslog

(KUKA), Daifuku, Bastian Solutions (private) and Intelligrated (Honeywell).

Figure 7: Country rankings2 and key companies


2 For an explanation of the 13 characteristics used in the ranking, please see Appendix A.

Country /

Region

Country

Score*

Companies to

WatchRating / TP Key companies within country / region for online grocery

US 66

Amazon

Walmart

Kroger

Ahold Delhaize

OUTPERFORM, TP $900

OUTPERFORM, TP $80

NEUTRAL, TP $30

OUTPERFORM, TP €24.50

Continues to expand grocery offerings in US and UK

Largest mass merchant globally with 19% share of the US grocery market

Rapid click and collect expansion; currently unwilling to invest in direct delivery

Multiple online operations in multiple jurisdictions

Canada 65Loblaws

Longo's

Not rated

Private

Offline leader; click and collect offered but no delivery

Toronto-based grocer with full-service online delivery

UK 63

Tesco

Ocado

Morrisons

UNDERPERFORM, TP 160p

OUTPERFORM, TP 410p

NEUTRAL, TP 230p

Sophisticated online operations, but levered to large stores and long leases

The leader in purpose-built automation for grocery

Strategic partnerships with Ocado and Amazon, plus upstream leverage

China 61

Alibaba

JD.com

Yihaodian

OUTPERFORM, TP $127

NEUTRAL, TP $32

Private

Parterning with top international supermarket brands, focus on fresh

Strong #2 to Alibaba, good strategic partnerships (Walmart, Yihaodian)

Strong procurement and fulfillment expertise

France 57 None

All major players have extensive click and collect or 'Drive' locations, which total

more than 4,000 throughout the country. France has a very high proportion of

hypermarkets, which is a strategic negative.

Germany 57 NoneHigh discounter penetration and lack of an anchor metropolitan area like London

or New York has placed German retailers relatively behind in online grocery.

Scandinavia 55 ICA Not rated Innovative 'delivery to fridge' model.

Australia 54Wesfamers

Metcash

NEUTRAL, TP A$41.92

OUTPERFORM, TP A$2.54

Heavy exposure to bricks and mortar retail but offering online grocery

Most likely wholesale distribution partner for Amazon in Australia

South Korea 54

E-MART

Homeplus

Lotte Mart

OUTPERFORM, TP W235,000

Private

NEUTRAL, TP W220,000

Two fulfillment centres and increasing penetration via same day delivery

Same-day delivery service

Growing SKU selection, planning to have 3 fulfillment centres by 2019

Russia 52 NoneMajor players are watching developments but are currently only in testing

phases.

Japan 49 None

Some private, B2B and niche players operate in key cities, but Japan is

currently lacking leadership from the large consolidated food players (such as

Ito-Yokado, Aeon and Seiyu) or well-capitalised internet-only disruptors.

Mexico 41 Walmex NEUTRAL, TP MXN39 Investing in online infrastructure, especially relative to competition

Brazil 37 GPA Food NEUTRAL, TP R$70 A pioneer of online grocery in Braz il with dedicated fulfilment capacity

* Country score is the sum of the 13 characterisics used to create the country spider charts in Part II



10 April 2017


Table of contents

On the cusp of a seismic shift 1

Key tables and charts 2

Executive summary 3

Key players within online grocery ............................................................................. 4

Part I: The grocery sector 6

A very brief history of grocery ................................................................................... 6

Business models: Centralised or Distributed? ......................................................... 7

Other economic considerations for online grocery ................................................. 14

The growing role of Amazon .................................................................................. 15

The inflection point is coming ................................................................................. 17

Part II: Country analysis 18

Where does online grocery work best? .................................................................. 18

Geographic analysis ............................................................................................... 19

Part III: Online grocery economics 39

The cost of picking ................................................................................................. 39

The cost of delivery (or collection).......................................................................... 46

Cost and margin comparisons of all business models ........................................... 49

Online grocery costs from the customers' perspective .......................................... 50

Appendix A: Multi-factor country model 52

Appendix B: Further reading / viewing 54

10 April 2017


Part I: The grocery sector

A very brief history of grocery

Step-changes in grocery distribution are not new. It wasn't until 1916 at a Piggly Wiggly in

the US that customers were able to browse groceries themselves rather than asking

someone behind a counter for them. It took another 20 years to incorporate produce,

baked goods and meat into what we now call a supermarket. The hypermarket, born in

Belgium 20 years later added general merchandise under the same roof. Format-wise,

things then remained reasonably static until the 1990s.

Webvan, the first online grocery delivery company of any size, was founded in 1996 and

went public at a valuation of $4.8bn. Five years later it was bankrupt, with the media

branding it one of the largest dot-com flops in history. The management of Ocado (a UK-

based, pure-play online grocer) cites the implosion of Webvan as a critical event that

provided the lead time to hone operations with few competitors. We also believe it soured

the appetite for online grocery within the US for a significant period of time.

As a result, consumable categories have been slower to gain traction amongst online

shoppers, with only certain categories (personal-care, beauty products online, packaged

grocery) growing well. Fresh groceries have much lower penetration. They are highest in

Asia, where 40% of Chinese, 39% of South Korean and just over 33% of Indian

respondents say they’ve purchased fresh groceries online (source: Nielsen). These

numbers drop materially in Europe (9%) and North America (9%), the traditional leaders in

online marketplaces. 64% of respondents were concerned about the freshness or quality

of products purchased online.

There are a large number of sceptics of

the online grocery model. Our research

shows that some believe online grocery

shopping is a niche service for time-

starved, cash-rich families, technology

buffs or foodies who are looking for

unusual or exotic ingredients. We also

read that consumers like going to the store

to buy groceries.

However, we believe the appeal is much

broader and the value proposition much

greater. From a customer's perspective,

there are good reasons to try online

grocery – but also, many more reasons

why, even after trying, customers switch

back to traditional stores (Figure 8).

This is particularly true when a lack of dedicated resources results in unmet expectations,

especially if there is a premium being charged for the service. So far, online grocery

underwhelms in most countries. We believe much of the blame lies in the choice of

business model of online grocery.

Done correctly, however, online grocery is of immense value to the consumer. According

to Kantar Worldpanel, households shop for the same 400 items on a weekly basis, if not

more often. If the outcome of an online shopping trip meets or exceeds an in-person shop

at similar price levels, we expect the shift to accelerate. We believe we are now

approaching that point.

Figure 8: Some pros, many cons


Positives Negatives

Shop at any time Sign-up hassle

Time savings Website navigation

Chore avoidance Selection issues

Wide range of items Substitutions

Hidden cost savings Returns

Delivery coordination

Slot availability

Premium cost

Fresh food anxiety

Lose out on "experience"

10 April 2017


Business models: Centralised or Distributed?

Online grocery business models, fulfilment models, customer experiences and economic

sustainability are inextricably linked and vastly different. When looking at the logistics

value chain from a customer's perspective, there are four discrete elements to a shopping

trip: getting to the store, picking the basket, checking out (paying and packing) and

transporting the shopping home (Figure 9). We believe that many shoppers could consider

each of these tasks a chore – i.e. best avoided, if possible.

Within these four elements, the key component, in our view, is the picking function: For an

online grocer, this is the most time intensive part of the process (and so the highest labour

cost portion) and is also the target of customers' two main anxieties:

■ Are the fresh items the ones they would have picked?

■ If an item is unavailable, what is the best substitute?

Figure 9: The online logistics value chain; outsourcing chores


The business model (and its accompanying fulfilment model) has a material effect on

customer experience, so it ultimately drives overall adoption of online grocery. The most

straightforward model (and still the most common) is to pick from a store close to the

customer. In recent years, grocers have added 'dark stores (dedicated quasi-warehouses,

efficiently merchandised, not open to the public), hybrid stores (stores with some

dedicated non-public space for certain categories) and centralised fulfilment centres

(CFCs, large, warehouse-style operations that receive good from manufacturers), but

these still form a relatively small percentage of all online baskets.

If we take the fulfilment models and incorporate the two main options for getting a

completed basket to the customer's house (home delivery and 'Click and Collect' or C&C),

we can create the matrix in Figure 10. Each cell represents a unique economic model for

the grocer and a different value proposition for the customer, and each combination can

have either a subtle or significant impact on overall economics.

Figure 10: The way baskets are picked and delivered changes the economics


Create

shopping

list

Pick itemsCheckout

/ pay

Transport

to home

Pick itemsCheckout

/ pay

Transport

to home

Traditionalcustomer functions

Pick itemsCheckout

/ pay

Online grocer functions

Travel to

store

In-store Dark store Hybrid Centralised

Home Delivery

Store Pick Up ? ? Currently in use

Non-Store Pick Up Not currently used

Unattended Collection ? Most common

Picking Methodology

Delivery

Method

10 April 2017


We worked with Kurt Salmon, an international consulting firm that specialises in retail and

consumer products, to analyse individual costs that are incurred within the UK during

online grocery fulfilment. Later in the report we explore detailed economics of the four

main picking models and the two main delivery options as well as the relative impact of the

cost of labour and capital between regions.

All online grocery is not created equal

Online grocery adoption has been slow. The negative issues highlighted in Figure 8 show

the main arguments against online grocery from a customer's perspective. The main

reason why these issues persist is cost. Online grocery is an order of magnitude more

difficult to execute than online general retail, which makes the economics for the grocers

extremely poor, and hence there are few incentives to solve the issues. Exacerbating this

issue is the fact that customers are exceptionally sensitive to substitutions and delivery

timing in grocery, which sets the bar very high.

There are four main reasons why this cost disadvantage exists:

■ Individual item values are low. Picking costs are very high as a percentage of value.

■ Orders have a large number of items. As number of items increases, achieving a

100% pick rate becomes challenging. Mistakes and substitutions introduce high-cost

return processes.

■ Temperature chain of custody. Food safety regulations require frozen and chilled

products to be handled independently, requiring much greater logistical effort.

■ Delivery timing is determined by the customer. Because delivery slots are chosen

by the customer (within a narrow window), delivery routing is inherently inefficient.

In addition to the high-cost nature of online fulfilment, cannibalisation of store-shopped

baskets is margin dilutive, reducing the incentive for grocers to move online. What we end

up with is a piecemeal approach, where tentative grocers link up with early adopters to

create a suboptimal system that grows slowly and creates a high degree of scepticism.

This partly explains why even in the UK, where the highest-quality offerings are available,

there are numerous impediments to a mass market shift to online grocery. Consumer

habits are hard to break, and the perception that online grocery is expensive, cumbersome

and delivers low-quality produce make it a challenge for marketers. Only 27% of the UK's

population shops online for groceries, and of that group, ~50% shop 5 or fewer times per

year. There is clearly work to be done to increase the frequency of those who already

shop online, and to increase the overall percentage of people who shop online.

Figure 11: UK frequency – mostly low Figure 12: UK penetration – modest

Source: Kantar Worldpanel, Credit Suisse research Source: Kantar Worldpanel, Credit Suisse research

0%

10%

20%

30%

40%

50%

60%

70%

1 to 12 13 to 20 21 to 30 31 to 40 41+

Frequency of Shop (per year)

?

0%

10%

20%

30%

40%

50%

60%

2012 2013 2014 2015 2016

?

10 April 2017


We note penetration looks like it is actually slowing at a time it should be picking up. Why

do fewer than 30% of consumers shop online, and of those, perhaps only 20% could be

considered "high frequency" shoppers?

The data suggest that choice of business model has a material impact on acceptance and

growth of online grocery. Currently, the vast majority of online orders in the UK are still

picked in stores (Figure 14). The consumer does not differentiate between fulfilment

models, but relies on the same set of performance indicators to determine overall value:

assortment, product availability, freshness, substitutions and delivery performance.

Figure 13: UK online market share Figure 14: UK market share by model

Source: Kantar Worldpanel, Credit Suisse research Source: Kantar Worldpanel, Credit Suisse research

The arguments for centralisation

Assortment: Ocado has the broadest assortment among all UK grocers (online or offline)

with >50k grocery SKUs (2015: >47k, 2014: >43k, 2013: 34k). Inventory only needs to be

carried in a small number of locations (for Ocado: 2), and adding SKUs is inexpensive.

Availability: We compile the service and availability statistics that are published in a UK

trade publication and supplement the data with our own availability statistics on Ocado.

The consistency shown by the centralised model has been notable; out of the last 25

weeks of data, there were only two weeks where two items were out of stock and four

weeks where there one item was out of stock; the other 17 weeks had 100% in-stock

availability, materially better than in-store availability (Figure 15).

Figure 15: Product availability; the centralised model excels

Source: The Grocer, Credit Suisse research

Tesco

42%

Asda

16%Sainsbury

17%Morrisons

5%

Waitrose

3%

Ocado

17%

Amazon

0%

In-store

73%

Dark

store

5%

CFC

22%

0%

10%

20%

30%

40%

50%

60%

70%

80%

90%

100%

Nov-16 Dec-16 Jan-17 Mar-17

Asda Morrisons

Sainsbury's Tesco

Waitrose Ocado

10 April 2017


Freshness: Centralised fulfilment centres receive goods directly from suppliers and orders

are shipped directly to customers. This shortens the supply chain and extends the lifespan

of fresh produce. A shorter supply chain coupled with FIFO inventory management also

minimises waste. A store-based fulfilment model must balance between high inventory

levels (to ensure availability), or low inventory levels (to maintain throughput and

freshness). Local demand fluctuations can have a material effect on this balance, and also

mean that customers in different regions will experience different service levels.

Substitutions: Substitutions are largely a function of inventory control, and are closely

related to availability. Centralised fulfilment has much greater availability, and so inherently

has fewer substitutions. More broadly, the customer is interested in order accuracy (which

is the metric that Ocado tracks). Accuracy is a combination of substitutions and errors.

Centralised operations have more automation and easier to maintain picking processes

that can reduce errors.

Delivery performance: We discuss the complexities of delivery in the next section, but we

believe that centralised order routing allows for more accurate delivery time estimates.

Ocado publishes KPIs with respect to delivery metrics and order accuracy. Figure 16

shows that both have been consistently high. We believe Ocado puts significant effort into

maintaining these metrics, but it is our view is that it is their business model that gives

them the opportunity to achieve these high scores. The corollary of this is an in-store

picking model, which is not inherently able to support high KPIs, even at great effort.

Figure 17 shows Ocado's revenue by year of customer acquisition. We would expect each

horizontal bar to trend downward over time; new customers cannot be added to a cohort

but existing customers leave (to competitors, moving away, or otherwise). However, each

of the bars in the diagram are quite flat, implying that any loss of customers is being

balanced by spending increases by the customers that remain.

Not only is each cohort's sales relatively flat, each subsequent cohort is adding an

increasing level of spend (i.e. the bars are getting wider). This is supported by Ocado's

organic growth, which has averaged 15% over the past five years. The high and consistent

KPIs give customers reasons to become repeat customers.

Figure 16: Ocado KPIs and LfLs Figure 17: Ocado sales by cohort

Source: Company data, Credit Suisse research Source: Ocado, Credit Suisse research

Our supposition is that the adoption of store-based fulfilment for online grocery is slowing

materially, either because companies are trying to slow down growth (due to low

profitability/cannibalisation), or customers are slowing adoption (due to poor value), or a

combination of both factors. The Kantar data in Figure 18 show that Tesco's contribution

to online growth was negative for much of 2016, while Ocado, Morrisons and Amazon – all

centralised fulfilment operations – continued growing strongly.

0%

5%

10%

15%

20%

80%

85%

90%

95%

100%

2011 2012 2013 2014 2015 2016

On time delivery or early

Order accuracy

LfLs (RHS)

2007 20142008 2009 2010 2011 2012 2013

£

10 April 2017


Figure 18: Contribution to online growth by UK retailer

Source: Kantar Worldpanel

The arguments against centralisation

While there are many and varied benefits from centralisation, there are also two main

drawbacks that need to be considered. The first issue relates to drive times. By increasing

the distance between the central facility and customers, drive times increase, thereby

adding costs – both direct costs such as fuel and labour, but also indirect costs such as

depreciation on transportation infrastructure. The second issue relates to scale. A fully-

utilised facility would maximise profitability, but the ramp-up period of a large facility can

take a long time, thereby depressing IRRs.

Both of these issues raise the question of the optimal size of a CFC, and are the main

reasons why a grocer may decide to open a dark store close to customers rather than a

CFC. Currently the capacity crossover between a dark store and a CFC is in the £150m to

£350m/year range – anything larger than £350m would be a centralised operation, while

anything smaller than £150m would be a dark store. Ocado is currently aiming at reducing

the lower bound for CFCs to allow the construction of facilities that are closer to customers

(and to reduce capex requirements).

Ocado attempts to bridge the distance from its CFCs by 'trunking' – trucking consolidated

orders via tractor-trailer trucks to a 'spoke, where orders are broken out into delivery van

sizes. Currently, just one CFC fulfills orders for the entire London area, supplemented by

eight spokes, as shown in Figure 19. The commissioning of CFC4 in 2018 should alleviate

pressure on this CFC, and also reduce the trunking requirements into London.

The added step of trunking, and the potentially longer drive times from the nearest spoke

to the furthest away delivery are the main additional costs of centralised fulfilment when

compared to store picking. We estimate this to be ~2% of sales. To the extent that online

operators are able to position fulfilment centres closer to the ultimate catchment area, this

cost will decline, although we do not expect it to disappear.

-4%

0%

4%

8%

12%

16%

May 2015 2016 2017

Tesco Asda Sainsbury Morrisons

Waitrose Ocado Amazon Online

10 April 2017


Figure 19: Ocado's current fulfilment infrastructure around London

Source: Company data, Map data ©2017 Google, Credit Suisse research

The data seem to bear out this headwind. As shown in Figure 21, trunking and delivery

costs as a percentage of sales have not fallen materially for Ocado, which is not surprising

given the lack of obvious leverage in the delivery model.

Figure 20: Ocado's London spokes Figure 21: Ocado delivery expenses

Source: Company data, Credit Suisse research Source: Company data, Credit Suisse research

However, with drivers' labour representing ~60% of current trunking and delivery costs, it

could be argued that the other costs (capital, fuel) have fallen, implying higher efficiencies

overall. These efficiencies come from higher customer densities and improved routing –

both of which should continue to improve given that Ocado's market share is still low.

We can attempt to unpick these costs by looking at the increasing level of wage rates

using national statistics and backing into the total costs for trunking and delivery using

Ocado's published costs.

Centralised Fulfillment Centre (operational)

Centralised Fulfillment Centre (under construction)

Spoke

Shaded circles’ radius = 5km

Distance from

CFC (km)

Enfield 26 24

Park Royal 32 36

West Drayton 50 39

South Ruislip 31 45

Dagenham 63 48

Kent 69 51

Surrey 67 53

Wimbledon 51 76

SpokeDrive Time

(mins)

10%

11%

12%

13%

2010 2011 2012 2013 2014 2015 2016

10 April 2017


Figure 22: UK labour rates Figure 23: Delivery cost breakdown

Source: ONS, Credit Suisse research Source: Company data, Credit Suisse research

But are these improvements any better than what we could expect from a store-pick

model? In our view, the answer is yes. We estimate that Tesco delivered £3.5bn of

groceries in 2016 using 5,500 vans, or £640k/year per van; Ocado delivered £1.2bn of

groceries per year using 1,350 vans, or £870k/year per van – a 30% delta after adjusting

for pricing differences3. The trunking costs that Ocado incur appear to be more than

compensated for by the efficiency gains it has been able to achieve from the 'last mile'. We

think this is due to three main issues:

■ The 'n+1' van problem. For each site, some number of redundant vans are required to

ensure continuity of service if there is a breakdown or a roadworthiness issue. The

more sites in operation (which are much greater in a store-based vs. centralised

model), the more redundant vans are required.

■ Routing efficiencies. In a store-based model, routes are optimised by individual van

(or potentially, by store). This reduces the ability to route efficiently, which requires

larger fleet sizes and larger catchment areas to provide optionality and flexibility.

Routing algorithms involve some of the most complex areas of study within

mathematics and computer science; for grocery, the added constraints of weight and

volume within a route plan further complicate the problem. The high level of complexity

offers myriad opportunities to squeeze out efficiencies if operations are centralised,

while a store-based model does not have these opportunities.

■ Coverage areas. Store picking allows for large coverage areas; both Tesco and

Sainsbury are able to reach the vast majority of the UK population with their online

offers, whereas Ocado reaches 70% of the population, but services only ~25% of the

area. The extended reach of store-based fulfilment creates locations where asset

utilisation is relatively low.

Taking all of these factors into consideration, it is difficult to come to a definitive conclusion

regarding the true costs of delivery for each business model. We believe that Ocado and

Tesco provide excellent case studies to analyse delivery issues, but neither should be

assumed to be a 'general case'; different implementations of their models could have

different outcomes. However, our analysis shows greater opportunities for cost savings

with wider catchment areas, and so we expect trunking and delivery costs to trend lower

for centralised operations, while store-based costs should remain relatively static. Longer-

term, step-changes in technology (for example, driverless vans, drone-bots) coupled with

tweaks to the business model have the potential to materially cut delivery costs.

3 The difference between £870k and £640k is 36%, but part of that difference is due to pricing rather than volume, which we argue

is not a function of delivery efficiency. Our analysis shows that Ocado prices at a 6% premium to Tesco. On a like-for-like volume basis, this would reduce the efficiency premium from 36% to 30%.

96%

98%

100%

102%

104%

106%

108%

110%

112%

114%

£420

£430

£440

£450

£460

£470

£480

£490

£500

£510

2010 2011 2012 2013 2014 2015 2016

Labour rates (£, LHS)

Labour rates (indexed, RHS)

35%

40%

45%

50%

55%

60%

2010 2011 2012 2013 2014 2015 2016

Driver labour

Residual (efficiency gains)

10 April 2017


Other economic considerations for online grocery

The value of data

We see the increasing role of data analytics as a way for retailers to add value to their

suppliers (and in the process, monetise the data they accumulate). Ocado has shown it

can increase supplier revenue as a percentage of overall revenue (Figure 24), a trend that

we expect to continue. The majority of Ocado's orders are now placed via a mobile device,

where screen real estate is limited. The value of being able to place specific products on a

particular customer's screen is very powerful. Figure 25 shows offers on the home screen

of an Ocado customer who had made multiple purchases of each one of those items in

previous orders.

Figure 24: Supplier income (% revs.) Figure 25: Ocado app home page

Source: Company data, Credit Suisse research Source: Credit Suisse research

Differential product pricing

As an adjunct to the rising importance of data, an important benefit of online shopping is

that every 'storefront' can be customised for each consumer. Customisation is already

happening with promotions and delivery costs, but we should expect pricing to become

individualised (or at least categorised) as well.

National pricing is a 'taboo' subject in the UK – every grocer sells the same product in the

same format at the same price, no matter whether they are in Central London or in a

remote part of the country. These issues do not exist in larger countries (France, US,

Canada, for example), and we do not believe they should persist online either. Multi-

channel grocers are competing with their own stores, so price customisation may be more

challenging, but grocers with no store infrastructure would not be subject to the same

constraints.

Same day delivery: Not just a risk to convenience operations

We expect another step-change to occur when same day delivery (SDD) becomes

mainstream. SDD will enable grocers to move away from bad trading practices, drive

customer engagement and (potentially) boost margins.

When customers changed from daily shopping at a local market to a supermarket (and

subsequently to the hypermarket), they benefited from a wider assortment, higher quality

and lower prices. The trade-off was that customers then had specific, sometimes

inconvenient, shopping missions: e.g., a weekly shop, top-up shops, food for today, food

for now, special events. These benefits were driven by new stores and significant supply

chain efficiencies, but many consumer wants and needs remained unfulfilled. Their goal is

1.0%

1.5%

2.0%

2.5%

3.0%

3.5%

2010 2011 2012 2013 2014 2015 2016

10 April 2017


to have a complete inventory of fresh groceries at home, low costs, wide variety, and low

involvement.

These goals are extremely difficult to meet within the context of the existing large

supermarket infrastructure (and are challenging to meet with current online offerings,

although they come much closer). But once online can promise SDD, consumers will be

able to shop for groceries the way they always wanted to, with no planning, no stocking up

or bulk shopping. The size of the basket becomes reasonably consistent, which helps both

consumers and online operators; 'convenience' is redefined because proximity benefits

break down as delivery lag approaches zero. Waste is reduced and customers are happy.

The retailers (and ultimately, the suppliers) can then move away from their mass-

marketing, mass-promotion and mass-merchandising models. Promotional intensity would

fall as it would not be required to encourage customers to choose a particular store.

Consistent, at-home replenishment removes the focus on bulk shopping and a truly

personalised interface reduces price sensitivity and provides an easier path to trading up

and trading across.

The reward of this model will be loyalty. Traditional online fulfilment methods, such as in-

store picking or C&C, do not lend themselves to this model because of the built-in

inefficiencies and the fact that they have to follow the same pricing/promotion structure as

the mother stores. Only centrally-fulfilled (or large dark stores) offer the scale and

efficiencies to enable widespread SDD while maintaining a broad, in-stock assortment.

Home automation: Beyond music and the weather

Home automation has evolved into what is now a home assistant, with Amazon's Echo

perhaps the most recognisable example. And although Google Home was only released in

November 2016, we note that the underlying AI-based technology in the form of Assistant

is rapidly gaining distribution across the spectrum of Google's applications as well as

Android phones. While Amazon and Google are the frontrunners in commercialising this

technology, it is difficult to ignore Apple, as it was among the first to demonstrate its

expertise in natural language processing with the release of Siri.

In 2013, Google CEO Larry Page articulated the concept of a connected, multi-screen

home, a point worth raising once again as we are on the cusp of advancement of smart

home appliances/refrigerators that should be able to determine when replenishment is

required, in addition to performing the basic task of maintaining shopping lists. This

reduction of transactional friction should help the adoption of online grocery, in our view.

What may be a troubling development for grocers is that the number of companies who

have the technology/engineering resources to develop and maintain applications within a

connected home environment will be few and far between, with either Google and/or

Amazon (or Apple) serving as the primary consumer gateway for transactions.

The growing role of Amazon

The business that Amazon started in books has spread to all kinds of goods, but it wasn't

until 2010 that it started to test the feasibility of expanding into the single-largest sector

within retail: grocery. As noted earlier, there has been a spectacular failure (Webvan), and

some small successes (Peapod in the US, Tesco and Ocado in the UK), but online

grocery remains less than 1% of all grocery sales globally.

And just as Amazon has gained share over time in other product verticals, it would not

surprise us to learn that it has similar ambitions to be a Top 5 grocer over the longer term.

Amazon grocery model still in flux

That said, handling fresh produce is vastly different from other categories. And it should be

no surprise that Amazon's full offer grocery service (AmazonFresh) has seen a slower-

than-anticipated rollout. Our expectation in 2H13 when the company started to expand out

10 April 2017


of its home base of Seattle and into Los Angeles and San Francisco, was for Fresh to be

offered in 25 additional markets by the end of 2014. (Please see our report, The Key

Product in Amazon Fresh is Not Just Produce but Rather Same Day Delivery from 17

December 2013) .

The complexity that food adds to the fulfilment process is extremely high, a summary of

which is in Figure 26.

Figure 26: Difference between food and non-food fulfilment


According to an academic paper that analysed Amazon's fulfilment processes4, Amazon

undertakes numerous other steps within fresh produce that have no parallels in non-food.

These include more advanced receiving functions, transforming cases to individual SKUs,

preparation (individual wrapping, for example), replenishment (from non-pickable locations

to pickable locations), produce-specific storage and packing, trimming, inspection, ongoing

rating and cleaning. Overall complexity is much higher than in non-food, and so training

requirements are necessarily higher, as are costs.

And just as it has already done across many other product verticals, Amazon has entered

into a number of deals that continue to demonstrate its ambivalence to source inventory as

well as first-party or third-party ownership – as long as the consumer gains access to the

best selection at the best price.

The delivery process is also not yet optimised. Amazon trucks (not third-party) transport

orders directly from the fulfilment centre in Seattle, whereas orders for customers in Los

Angeles and San Francisco are transported in bulk from fulfilment centres to local delivery

stations and then separated onto different delivery routes (similar to Ocado's trunking

operations).

Lack of current market share should be no comfort

Even after seven years, AmazonFresh remains relatively underpenetrated. It is not

available in most US cities, and overall market share is immaterial.

As we have noted previously, AmazonFresh as an offering should not be viewed just as an

encroachment into grocery, but rather a broader rollout of same day delivery. Amazon's

existing product selection within Prime Now has seen a faster rollout than grocery and

continues to expand vis-à-vis its third-party offering along with shipping discounts across a

greater array of products facilitated by Fulfilment By Amazon. Grocery is one of the last big

steps, which should aid consumer engagement and frequency.

4 V. Modi, Application of Flexible Labor and Standard Work in Fulfilment Center Produce Operations, MIT, 2015

Order fulfillment step Food Non-food

Temperature checksPotential disagreements with driver when

product arrives outside of specificationsNone

Prioritisation Required, due to freshness concerns None

Quality check In-depth, multi-faceted Cursory

Depalletising Complex; similar products look alike Straightforward

Storage Location findingHighly regulated, due to temperature

requirementsStored anywhere within FC

Expiration dates Closest date picked first No expiration dates

Order Fragile items segregated Based on routing efficiency

Quality controlMulti-faceted quality checks during

pickingCursory damage inspection

Item combinations Multiple exceptions (i.e. garlic + flowers) None

Temperature combinations Grouped, where possible None

Receiving

Picking

https://plus.credit-suisse.com/r/V2vbmK1AF-WElY95

https://plus.credit-suisse.com/r/V2vbmK1AF-WElY95

10 April 2017


The inflection point is coming

After 20 years since the first pure-play online grocery operation started, we are seeing a

business model emerge that provides excellent service, but at a high cost. We expect this

business model to develop into one that companies can use to drive costs materially

lower, while maintaining, or even increasing KPIs.

The technological advances in customer interfaces, data analytics, robotics, and delivery

channels will make centralised online grocery a far superior value proposition to both

customers and grocers. In many ways, it is a mindset shift – away from costs added

(picking labour, delivery costs, IT infrastructure) to costs removed (store expenses, rent,

labour, utilities, maintenance).

Revisiting the pros and cons from Figure 8, we show the remedies for each of the

customer issues we raised. We expect that over the medium term, customers will receive

a significant increase in value without a commensurate rise in prices.

Figure 27: Technology, over time, will address all of the negatives


The end game

The complexities of grocery means that the sector is still at the very early stages of a shift

to online – perhaps five years behind general merchandise. In non-food, not only did

technology and software need to improve materially but so did the systems integration

processes, which are materially easier than those needed for food.

Figure 28 shows the gross profit contribution per basket for each of the main business

models using an average basket size for an online order (£90).

Figure 28: Centralised operations provide a clear path to materially higher margins

Source: Company data, Credit Suisse research

The reward for centralising and automating grocery fulfilment is significant: operating profit

margins are expected to be ~2x of those in traditional store-based models, and over 3x

once the next round of technology advances is implemented.

Positives Negatives Remedies

Shop at any time Sign-up hassle Use single sign on / 1-time event

Time savings Website navigation Website improvements

Chore avoidance Selection issues Website improvements / learning

Wide range of items Substitutions Operational improvements / scale

Hidden cost savings Returns Operational improvements

Delivery coordination Delivery alternatives / technology improvements

Slot availability Operational improvements / scale

Premium cost Scale / technology advancements

Fresh food anxiety Operational improvements / scale

Lose out on "experience" Virtual reality

Picking location In-store Dark store Hybrid 3rd-party Centralised In-store Centralised Centralised Centralised

Picking technology Staff Staff Staff 3rd-party Staff Staff Staff Robotic Robotic

Delivery location Home Home Home Home Home C&C C&C Home Home

Delivery mechanism Driver Driver Driver 3rd-party Driver Customer Customer Driver Automated

Industry leader Tesco Tesco Peapod Instacart Ocado Leclerc n/a n/a n/a

Gross margin (avg. basket) 27% 27% 27% 27% 27% 25% 25% 27% 27%

Total direct costs -29% -26% -27% -6% -21% -22% -14% -19% -11%

Direct P&L per basket -1% 1% 1% 21% 6% 3% 11% 9% 17%

10 April 2017


Part II: Country analysis

Where does online grocery work best?

Global socioeconomic and demographic trends will contribute to rising demand for

convenience amongst an increasingly urban and ageing populace: According to the UN

60% of the world population is expected to live in urban areas by 2030; by 2050, the

number of persons over the age of 60 will triple, from 606m today to nearly 2bn. Both

trends will contribute to rising consumer demand for convenience offerings by retailers.

Dispelling the population density myth

One of the most common pushbacks we hear regarding online grocery is that it requires

very high population densities to work. London has become the epicenter of that argument

– it has a big, rich population that seems well-suited to online grocery.

Figure 29: Population densities and wealth of selected countries

Source: UN, World Bank, Credit Suisse research

But there are more cities similar to London than one might think. Canada is a good

example of just how misleading country statistics can be. With the world's second-largest

land mass and a modest population of 36m, Canada seems like a poor candidate for a

service that needs high population densities.

However, the region around Toronto encompasses 630km2 (a mere 0.007% of the

country) but it has a population of 2.7m (7.5% of the country).

Country Area (km2) Population (m) Pop. per km

2GDP per capita

South Korea 97,000 50.6 522 27.2

Netherlands 34,000 16.9 497 44.4

Belgium 30,000 11.3 377 40.3

Japan 365,000 127.0 348 34.5

United Kingdom 242,000 64.9 268 44.0

Germany 349,000 81.5 234 41.3

Switzerland 40,000 8.3 208 80.8

Italy 294,000 60.8 207 30.0

China 9,388,000 1371.0 146 8.0

Denmark 42,000 5.7 136 51.8

Thailand 511,000 68.1 133 5.8

Poland 306,000 38.0 124 12.6

France 548,000 64.4 118 37.6

Austria 83,000 8.6 104 43.8

Turkey 769,000 77.7 101 9.2

Spain 500,000 46.4 93 25.8

Republic of Ireland 69,000 4.6 67 61.7

Mexico 1,944,000 121.0 62 9.5

South Africa 1,213,000 54.0 45 5.8

Colombia 1,109,000 48.0 43 6.1

United States 9,147,000 321.0 35 56.2

Braz il 8,358,000 204.0 24 8.7

Sweden 407,000 9.7 24 51.1

Finland 304,000 5.5 18 42.2

Norway 365,000 5.2 14 74.4

Saudi Arabia 2,150,000 31.0 14 20.8

Russia 16,377,000 143.4 9 9.3

Canada 9,093,000 35.8 4 43.3

Australia 7,682,000 23.8 3 56.3

Total 71,816,000 3,108.2 43

10 April 2017


Toronto has a density of 4,284 people per

km2 – approximately 80% of the density of

Greater London5.

We see many more of these high-density

metropolitan areas within low-density

countries, including Australia, Sweden,

Saudi Arabia, Russia and Brazil. To be

clear, we are not trying to determine which

countries could replace grocery stores with

online grocery, just whether there are

population centres that are large enough to offer the required economies of scale.

Ocado's latest customer fulfilment centre is its smallest yet; a proof of concept of how

small a CFC can be and still reach peak efficiencies. Its capacity of £350m/year would be

fully-utilised in a catchment area of only 1.3m people, assuming a 10% market share. We

also note that different business models require different population densities. We analyse

the various operating models later in this report.

Geographic analysis

Expenditure on food varies widely across the world, but online grocery can thrive beyond

developed markets by implementing different business models.

Figure 31: Annual income spent on food (2008)

Source: Washington State University, USDA/Economic Research Service, 2008

5 Defined by the area inside the M25 ring road, which is 1,564km2 and has 8.5m people (source: ONS) for a density of 5,432

people per km2

Figure 30: Misleading densities

Source: ONS, City of Toronto, StatsCan, Credit Suisse research

Population Area Density

(m) (km2) (pop./km

2)

London 8.5 1,564 5,432

Canada 35.8 9,093,000 4

Toronto 2.7 630 4,284

Greater Toronto 4.6 1,539 2,970

10 April 2017


In order to assess the viability of online grocery within a given market, we created a multi-

dimensional model loosely-based on work first published by McKinsey & Co. to analyse

the European online grocery market. Our model measures characteristics that correlate to

the overall viability for high efficiency, centralised online grocery operations. These

characteristics can be grouped into three categories: infrastructure (such as broadband

penetration), economic viability (such as population and GDP), and latent demand (such

as prevalence of double-income households). For a complete description of our

methodology, please see Appendix A: Multi-factor country model.

Europe: Leading characteristics

Within Europe, only the UK and France can lay claim to having a meaningful share for

online within grocery, at ~6-7%. Figure 32 shows the UK's characteristics which,

unsurprisingly, are mostly favourable.

Figure 32: UK grocery characteristics; only latent demand is weak


The UK has five highly-developed online grocery operators.

0

1

2

3

4

5

6

Broadband penetration Tablet / smartphone

penetration

Online share of retailspend

Amazon penetration

Start-up / independentculture

Urban drivinginfrastructure

Metro areas > 1mMetro areas > 5m

GDP per capita

Density of supermarketspace*

Car ownership*

Prevalence of double-income households

Inclement seasonalweather

* inverse correlation

10 April 2017


Figure 33: The UK grocers all have material online operations


The presence of Ocado – an online only new entrant – had the greatest influence on how

the UK market developed. Ocado operated a home delivery model from its founding,

which drove competing offers from the major supermarkets. We believe that without

Ocado the UK model would have been much more grocer-friendly, with click and collect

dominating the online model, similar to France.

France is larger than the UK, but its city characteristics are similar. There are seven

metropolitan areas of 1m people or more, dominated by the Paris region of over 12m

people. For comparison purposes, Ocado has said that approximately 50% of its sales

(~£600m) come from within the M25 ring road – an area with a population similar to Paris.

Figure 34: France – Characteristics are even better than they appear


The biggest different between the UK and France is the method of delivery; the UK is

predominantly delivery to home, France is almost exclusively via click and collect.

Euromonitor estimates that France has 3,500 click-and-collect/drive locations, used by

more than 4m households.

Ocado*

Leading online-only grocer, currently operating 2 CFCs near capacity. A third CFC - its smallest - is

ramping up, while CFC4 - its largest - is under construction near London, with expected

commissioning in 2018

Tesco*

The UK's largest grocer and the leader in online grocer, with an estimated £3.7bn in online grocery

sales. Majority of baskets are delivered to home using in-store picking, but also operates 7 dark stores

and an extensive click and collect operation.

Sainsbury*The UK's second-largest grocer and a follower in online. Sainsbury opened its first dark store near

London late last year.

Asda

The Walmart-owned retailer is the third-largest grocer and has 16% of the UK online market. It has

had poor store performance and the worst online grocery offering according to Which? Magazine

2017 customer survey.

Morrisons*

A late entrant into online grocery, Morrisons is now the leader in low-risk, capital light solutions. It

partnered with Ocado to "white-label" a fully-integrated offering, although this was an onerous deal. It

has since renegotiated that contract after it took capacity in Ocado's 4th CFC. It has also partnered

with Amazon to supply it with a full fresh offering, and is going to allow Amazon to pick from its stores

in situations where customers are far away from fulfillment centres.

Marks &

Spencer*

High-end grocery with no online solution. We expect some form of supply agreement to an existing

online player rather than a home-built solution. Both Ocado and Amazon are possibilities, although a

deal with Ocado would be complicated by its supply agreement with Waitrose.

*covered by Credit Suisse equity research

0

1

2

3

4

5

6


penetration


Amazon penetration




GDP per capita


Car ownership*




10 April 2017


Germany is more decentralised, which reduces the economic appeal for online grocery.

Figure 35: Germany – Technology infrastructure and latent demand issues


Also, as the home market for the two dominant discount operators, Aldi and Lidl, there is

an inherent bias towards low-cost staples that would be an impediment in any rollout of

online grocery. However, the potential for a disruptor such as AmazonFresh to launch may

change this significantly, and force some of the more traditional grocers to increase their

online focus.

We also see potential for growth within northern Europe given the high internet

penetration. Key population centres around Stockholm (2.3m), Copenhagen (2.0m) and

Oslo (1.5m) could each support dedicated online facilities, even though overall population

figures are modest.

Figure 36: Scandinavia's lack of density is a barrier to adoption


Asia: Vast differences between countries

China is the world leader by total sales

China's retail e-commerce market grew by 36% in 2016 and accounted for ~15% of total

retail sales (source: National Bureau of Statistics of China) – a higher proportion of than

0

1

2

3

4

5

6


penetration


Amazon penetration


Urban driving

infrastructure


GDP per capita


Car ownership*




0

1

2

3

4

5

6


penetration


Amazon penetration

Start-up / independent

culture

Urban driving

infrastructure


GDP per capita

Density of supermarket

space*

Car ownership*




10 April 2017


that of the US, Japan and many other developed countries. Apparel and consumer

electronics products have been well-penetrated, and online shopper growth is maturing in

top tier cities.

Chinese consumers spend ~30% on food and beverage – much greater than developed

markets (US: 7%, Germany: 11%). Spending patterns are also very different: according to

Nielsen, Chinese consumers like to shop frequently for food, with a focus on fresh.

Online grocery has already reached high levels within China due to lower supermarket

density than developed markets, a relatively fragmented food retail market, low car

ownership, fast moving lifestyles and poor offline retail infrastructure. These characteristics

create latent demand for e-commerce grocery shopping.

According to CNNIC, China's internet penetration reached 53% (64% of whom shop

online), with top tier cities near developed countries' level. Smartphone/smart device

ownership continues to grow strongly. In China, the main online players are Alibaba and

JD (rather than Amazon).

Euromonitor estimates China's offline grocery retailing sales to reach Rmb4.9 trillion by

2020, representing ~3% five-year CAGR. In our view, online sales of groceries is likely to

take an increasing share of the total market, from Rmb245 billion (US$39bn) in 2015,

5.4% of grocery sales in total, to ~Rmb 1.3 trillion (US$178 bn) in 2020, or ~20% of the

market – a 40% CAGR over 2015-2020.

China's large population and high density in metropolitan area provides opportunities to

develop online delivery of food retail in urban areas (13 cities with greater than 10m

people). Nielsen's white paper on Fresh E-Commerce Development (October 2015)

reported that out of 1,600 interviewees living in seven megacities (>10m people), >63%

shopped for milk and dairy products online, >50% had purchased fruit, and 38% had

purchased meat, vegetables or seafood with higher basket sizes than offline shopping.

China's e-commerce giants Alibaba and JD.com are highly interested in online grocery

given China's population and customers' shopping habits. Both companies are

experiencing growth slowdown in their core categories, and see grocery as their next

opportunity. In addition, it is a very fragmented market, with various food specialty e-

retailers emerging rapidly.

Figure 37: China, unsurprisingly, leads in terms of population density


0

1

2

3

4

5

6


penetration


Alibaba penetration




GDP per capita


Car ownership*




10 April 2017


The main online grocery retailers are Alibaba's Tmall Supermarket and JD Supermarket.

Both offer a similar range of products and compete head-to-head in terms of price

(including promotion and user subsidies), product variety, quality and delivery service.

In July 2015, Alibaba announced the launch of an Rmb 1bn online grocery promotional

campaign, and teamed up with Cainiao (Not Covered) to offer the same-day delivery

services to Beijing residents. The service has now been expanded into more cities. On 14

July 2016, Tmall Supermarket announced the 'Double two billion plan' to provide Rmb 2

billion subsidies for consumers and will invest Rmb 2 billion into supply chain infrastructure

for service improvement.

The sheer scale of operations in China has led Alibaba to seek out multiple partners to

assist in sourcing, supply chain management, logistics and delivery. Tmall works closely

with western players such as Germany's Metro Group and Spain's DIA to bring more

premium and imported goods to Chinese consumers. Alibaba invested in and partnered

with Yiguo.com (Not Covered) to operate Tmall's fresh food section because of its strong

cold-chain logistics capability and focus on premier brands and quality fresh produce. In

addition, Alibaba invests actively in supply chain and logistics partners like warehouse

operation service provider ALOG and last-mile delivery service provider Ewinshine,

helping to strengthen the overall supply chain efficiency and ensure delivery quality. These

partnerships allow Alibaba to execute its strategy of offering same-day or next-day delivery

services while continuing to expand within China.

Figure 38: China online grocery led by Tmall and JD Supermarket


We think JD is the best-positioned online grocer in China. It is the strongest competitor to

Alibaba, and moved aggressively last year by forming a strategic alliance with offline

leader Yonghui supermarket (and acquiring a 10% stake). Yonghui is the country's fifth-

largest supermarket chain and has an excellent reputation for its fresh food business (45%

of total sales). JD also acquired the top online supermarket platform Yihaodian and signed

Tmall

Supermarket*

Partnering with top international supermarket brands for high-quality imported products and cooperate

with Yiguo.com to operate fresh produce. Leverage Cainiao (and partners) to manage warehousing

and fulfill orders. Avg. order size is >Rmb100 and free shipping from Rmb 88.

JD

Supermarket*

Allied with Walmart, Yihaodian for better product variety, and China's largest crowdsourcing logistics

platform Dada. Invested in Yonghui, China's top 5 local supermarket. Free delivery from Rmb99, and

offer "within 2 hour delivery" for Rmb 99.

Yihaodian

8-years of experience in online grocery now offering more than 8 million SKUs, well-equipped with in-

house fulfillment and strong procurement capability. Free delivery from Rmb68 for all major cities.

China's leading e-commerce company focusing on grocery. Strong in Eastern China (in particular

Greater Shanghai) due to its convenient logistics and high-quality groceries. However, due to the

ownership acquisition by Ping'An and later Wal-Mart, as well as the departure of founders, the

company has experienced slower for expansion in recent years. In late 2016, Wal-Mart sold YHD to

JD.com for Rmb40 billion.

Feiniu.com

(Sun Art*)

Launched in 2014. Large offer size. Average order is Rmb160-170 with monthly GMV at

Rmb250m. Delivery across China through its offline stores and 3rd-party couriers.

Benlai LifeSmall, specialty online retailer that offers ~5k SKUs of fresh produce. Has built out full cold-chain and

large warehouses and tier 1 cities and provides speedy delivery in selected regions.

Womai.comBacked by the state-owned grain group COFCO, Womai focused offer on premium imported food,

wine, oil and grain, etc. Geographically focused on tier 1 cities and nearby regions.

Others

Womai.com (a wholly-owned subsidiary of China National Cereals, Oils and Foodstuffs Corporation –

China's state-owned food processing holding company); SF Express' SFBest (a leading express

couriers' online marketplace selling food products); Fruitday (fruit specialty retailer), TooToo Organic

Farm (organic and high quality food retailer); Benlai Life (fresh produce e-commerce start-up), Fields

(Shanghai's premium online grocery store), Missfresh.cn.

Vanguard

(e-Wanjia)

Officially launched in June 2015, focusing on fresh food of more than 800 kinds. Vanguard aims to

develop the online platform by leveraging its robust offline resources and advantages in procurement

and logistics, and then broaden the geographic coverage and customer base.


10 April 2017


a cooperation agreement with Wal-Mart to improve procurement and warehousing

management. JD can leverage JD Daojia and the crowdsourcing platform Dada for more

efficient last-mile delivery in the country.

Japan has significant potential, but no integrated offerings yet

Japanese consumers tend to shop at small, local stores and convenience stores – a 'little

and often' strategy that is typically more difficult to supplant with online grocery, where

delivery costs are high (unlike China). One of the main themes in Japanese grocery

recently has been the consolidation of convenience chains and pressure on supermarket

operators to close unprofitable stores – a situation that is a tailwind to moving online.

Against that backdrop, it may be somewhat surprising to see that online grocery is well-

developed and large – the Yano Research Institute forecasts the domestic online food

market was ¥3,377bn in FY15 (US$30bn), and should continue to expand in the mid-

single digits.

The service is dominated by a paid membership co-operative called the Japanese

Consumers’ Co-operative Union (JCCU), which takes mail, internet and catalog orders.

JCCU has both legacy and internet order system. As of 2015, the JCCU had a dominant

41% market share in online grocery, compared to shopping websites at 35% and food

makers’ direct sales at 17%. We expect sales to increasingly migrate to online

supermarkets and shopping websites.

Figure 39: Online grocery by channel Figure 40: Online grocery trends

Source: Yano Research Institute, Credit Suisse research Source: Yano Research Institute (estimates), Credit Suisse research

Rakuten, the top domestic e-commerce player, and Yahoo Japan’s subsidiary ASKUL are

revamping their distribution networks in a bid to boost sales of daily sundries, and

ultimately fresh food. Amazon Japan has yet to enter the field, but it has a very good

logistics network already in place. Seven & I offers click-and-collect at all of its 15,000 7-

Eleven convenience stores, although the lack of in-store space makes high-volume, full

shopping baskets challenging.

We see the characteristics of the Japanese market generally favourable to online, with the

notable exception of smartphone penetration, which is very low for such a developed

economy.

Shopping

websites

JCCU

Bio food

retailers

Supermarkets

Convenience

stores Food

manufacturers

direct

0

500

1,000

1,500

2,000

2,500

3,000

3,500

4,000

4,500

2012 2013 2014 2015 2016 2017e2018e2019e2020e

JPY

(bill

ions)

10 April 2017


Figure 41: Latent demand drivers are weak but economic viability is strong


Overall, Japan is still searching for the best model, which is made incrementally more

difficult without the large consolidated food players (such as Ito-Yokado, Aeon and Seiyu)

to drive the market forward. The entry of AmazonFresh would be a material event, and

would hasten the shift from third-party services to the supermarkets/manufacturers .

Figure 42: Japanese online grocery evolving rapidly


Korea: High online growth; Grocery penetration material

Korea already has high e-commerce penetration with 22% share of retail spend in 2016,

one of the highest globally. However, we still expect this channel to register a 16% 2-year

CAGR (2016-18e) and remain the fastest growing, boosted by the intense competition and

changing consumer behavior.

0

1

2

3

4

5

6


penetration


Amazon penetration


culture



GDP per capita


space*

Car ownership*




Rakuten*

Rakuten currently delivers fresh foods to homes via Rakuten Mart. Rakuten’s presence in e-

commerce is based on the marketplace model, and has recently acquired Kenko.com (healthy food)

and Soukai Drug (prescriptions and health food). We think their distribution networks are likely to be

merged and improved to a point where fresh food deliveries are possible.

LOHACO

ASKUL (a consolidated subsidiary of Yahoo Japan) operates LOHACO, a B2C e-commerce

Website. Leveraging its B2B logistics network, LOHACO has offered its own designated delivery time

service in some areas of Tokyo and Osaka since mid-2016. LOHACO’s main focus is currently on

daily sundries, and it could make an entry into the food sector, in our view.

Radishbo-ya

Radishbo-ya (an NTT docomo subsidiary) provides mail/online order services for vegetables, meats,

seafood, processed foods and daily sundries. Instead of aiming to replace supermarkets, the company

offers services for regular home delivery and one-time deliveries of specific products.

Oisix Oisix (3182, NR) also offers regular scheduled deliveries of specialty cooking ingredients.


10 April 2017


Figure 43: Online grocery nearing 10% Figure 44: Online food is fast growing

Source: KOSIS, Credit Suisse estimates Source: KOSIS

Retail has been challenging in recent years due to an intense competitive landscape in the

online channel, which has resulted in on-going cannibalisation of the offline channel. That

said, online was mainly focused on non-food categories but interest has shifted to the

grocery category, into which most retailers try to expand.

As such, we see the biggest growth opportunity in online grocery retail. Among different

product categories, the food category is still the least penetrated with only 9% of grocery

shopping online vs. aggregate penetration of 22% across different categories. Also, food is

the largest consumer spend category in Korea, and involves more frequent shopping and

has stronger loyalty. As technology improves and consumer behaviour changes, we

expect online grocery to continue to grow faster than other categories. The infrastructure

required to support online grocery is well developed, in particular broadband penetration

and tablet/smartphone ownership, which supports the shopping experience for grocery.

We believe the worst in the competition among online players, especially in FMCG, is

behind us. Consequently, we see significant interest and competition, not only amongst

Korea's grocery operators but also amongst pure online players.

Figure 45: Online infrastructure and consumer trends are dominant


We see significant interest and competition amongst Korea's grocery operators.

0%

10%

20%

30%

40%

Total Home

appliances

Cosmetics Apparel

and

fashion

Food,

beverage,

agricultural

Other -10%

-5%

0%

5%

10%

15%

20%

25%

30%

35%

-

1,000

2,000

3,000

4,000

5,000

6,000

7,000

8,000

9,000

2009 2010 2011 2012 2013 2014 2015 2016

(W bn) Online grocery size Online grocery growth

Online channel growth

0

1

2

3

4

5

6


penetration


Amazon penetration


culture

Urban driving

infrastructure


GDP per capita


space*

Car ownership*




10 April 2017


Figure 46: South Korean online grocery evolving rapidly

Source: Company data, Credit Suisse estimates

Online grocery remains a high barrier to entry business due to the sizeable investment

required for refrigeration logistics and need for a strong 'retail brand' to support product

quality perception.

Amongst online players, we believe E-Mart will remain the most compelling player thanks

to its leadership and strong brand equity in food retailing and popular private labels such

as Peacock and No Brand. E-Mart's online business was 6% of parent revenue in FY16

and we believe the company can increase that to 10% by 2018, in part due to the new

online distribution centers it started operating since 2014. Scale and cost efficiency should

continue to improve with these online centers. As per the official guidance from the

company, it aims to double the revenue to W1.5tn by 2023 with six online distribution

centers in total, led by grocery. The repurchase rate, which is close to 90%, suggests that

E-Mart can maintain its market leadership in this channel.

Online Grocery in Russia: waiting for the customer

Currently, the Russian market ranks well against its global peers in terms of e-commerce

share (3% of sales in 2016 or RUB 920bn), and mobile technologies usage (on par with

that of Germany). However, the Russian economy is slowly coming out of recession and

the average customer’s real income is only now stabilising. While we see no structural

barriers to online grocery development, online grocery growth will need some changes in

consumption patterns as well as yet-to-be-developed infrastructure.

E-Mart*E-Mart has two fulfillment centres and continues to increase strength in fresh food category through

same-day delivery system. Average number of daily order is around 30,000 for two centres.

HomeplusHomeplus has same-day delivery service (if ordered before 4pm). Also, the company hires "Pickers"

to select and purchase fresh foods that are ordered online.

Lotte Mart*Lotte Mart targets to increase the number of SKUs and invest in online fulfillment to enhance its

delivery service. The company targets to have 3 centres by 2019.

CJ OS*CJ O Shopping has opened a premium store at online CJ Mall to overcome the negative perception

on online fresh food category.

Ticket

Monster

TicketMonster started to sell fresh foods through its direct purchase stores since last year. It offers

same-day delivery service from 7am.

WeMakePrice

WeMakePrice started the direct delivery service for fresh food November 2016. It has established

21,800㎡-sized cold storage and it also provides "next-day delivery service" for the orders made

before 10pm.

Coupang

Coupang cooperates with Nonghyup (Korea National Agricutural Coop). It delivers (within the same

day or the next day) products that are sourced from Nonghyup. Coupang offers 10~20% price

discount as well (compared to Nonghyup price).


10 April 2017


Figure 47: Some positive attributes, but not ready for rollout yet


The majority of sales come from hypermarkets, which have a greater than average basket

size. The major hypermarket and supermarket chains, (Lenta, Magnit, X5, Dixy and O’key)

are watching developments in online food and are ready to act beyond their current testing

phases if they see any substantial uptick in demand (or an aggressive move by their

competition). We see the Russian market as having good potential for online grocery

growth in the upcoming years, which should be triggered by the following:

■ Upwards dynamics of total real income;

■ Strong marketing efforts of major food-retail players; and

■ Changing consumer behaviour patterns toward time-saving online purchases.

Australia: Positioned for growth

The Australian market has many of the characteristics for rapid online food growth and

online adoption is generally high in other retail categories. To date, inadequate scale and

lack of patient capital have been barriers to the development of large scale local food

online. The entry of Amazon to the Australian market (expected in 2017) would likely

cause a material acceleration in the online food channel.

The online segment of food retail was ~A$3.5bn in the year to 30 June 2016 (3% of food

retail). We estimate Woolworths online food sales at A$1.2bn and Coles at A$1.0bn,

together accounting for three-quarters of the online retail food market in Australia. The

relatively low share of online food retail to date can be explained by the large market share

of the incumbent traditional grocers.

0

1

2

3

4

5

6


penetration

Online share of retail

spend

Amazon penetration


culture

Urban driving

infrastructure


GDP per capita


space*

Car ownership*

Prevalence of double-

income households

Inclement seasonal

weather


10 April 2017


Figure 48: Australian online grocery dominated by Woolworths and Coles


Woolworths and Coles offer most of their respective ranges online, with the exception of

in-store manufactured items such as cut delicatessen and hot food. Coles typically

charges higher prices online than in store, whereas Woolworths generally has the same

pricing in store and on line.

With high broadband and smartphone penetration and high online spend, Australia ticks

most of the boxes in terms of customer propensity to shop online.

Figure 49: Infrastructure and large population centres support online grocery


Urbanisation is high, with three cities – Sydney (4.3 million), Melbourne (4.1 million) and

Brisbane (2.1 million), having the requisite population level to support a large scale

specialist grocery online competitor. High labour cost (including base wages and punitive

retail penalty rates) improves the economics of a centralised picking model. Discounter

penetration is modest and, with only Aldi operating in the hard discount segment of the

market, discounter competition is at the low end of comparable markets. Latent demand

would appear reasonably high, but without Amazon in the Australian market, capturing that

latent demand has been left to the major incumbent retailers, which have little incentive to

compete strongly in online.

Woolworths*

Offers ~22,000 skus online – pantry and fresh. Combination of fee per delivery and subscription. Flat

$11 delivery fee on orders up to $150. Free delivery on orders $300 and over. Delivery saver for

$89/year offers free delivery on all orders over $100

Coles*

Offer ~19,000 skus online – pantry and fresh. Prices online are typically higher than in store.

Minimum $50 order. Delivery fee $4-$18 per order. Free delivery on orders over $100 when paid

for using Coles Mastercard.

Harris FarmFree delivery on first three orders and minimum $80 spend. Delivery fee $15 on orders less than $80

and free on orders $150 and above. Available in the Sydney region.

Kogan PantryLaunched in January 2015. Stocks 600 grocery products. Fixed shipping cost $9.99 anywhere in

Australia via Australia Post.

Aussie

Farmers

Direct

Founded in 2005. Provides delivery of fruit & vegetables, meat and some groceries. Does not charge

an explicit delivery fee. Franchises cost between $50k and $110k. Franchisees receive a margin on

goods sold.

Grocery RunPantry, health & beauty, cleaning and households goods items. Flat rate $9.99 per delivery. Delivery

is through Australia Post 1-2 business days

Grocery

ButlerShopping service


0

1

2

3

4

5

6


penetration


Amazon penetration


Urban driving

infrastructure


GDP per capita


Car ownership*




10 April 2017


The Melbourne and Sydney retail food markets would be ~A$4bn each, on the assumption

that online delivery captured 20% of retail food sales. The relevance of that statistic is that

it implies that the Australian urban markets are sufficiently large under even moderate

penetration assumptions, to support a centralised online infrastructure.

Figure 50: Attractive opportunities within the largest Australian cities


The online scenario with the most impact would likely involve the entry of Amazon. Without

Amazon, the emergence of a domestic large scale online competitor outside of

Woolworths and Coles is less likely, in our view. Amazon does not presently have a

distribution infrastructure in Australia, although has been reported widely in the press as

likely to launch in Australia in 2017.

Supermarket gross margin in Australia appears to leave room for an efficient large scale

online competitor to slip potentially under the umbrella of pricing provided by the

incumbents. Ambient grocery gross margin is c. 18% and fresh is c. 30%. Pick from store

models operated by Woolworths and Coles have a fully end-to-end cost of online fulfilment

of ~18% of basket costs; a centralised model at scale would be likely to provide a similar

or lower cost of doing business due to the substitution of capital for relatively high cost

labour. The profitability of incumbent store based online models would be likely to

deteriorate.

Whilst the gross margin structure of the Australian industry is not excessive, there are

sufficient alternative sources of grocery supply for Amazon that would create some

downward pressure on pricing. Consequently it is likely that an Amazon entry to food

would cause some downward pressure on gross margin. Improved services would be

likely to increase the proportion of grocery sales online and increase online related capital

expenditure by incumbents.

Metcash's position is an interesting one.

As the only large scale food wholesaler, it

has the scale in buying and warehousing,

which would position it well to partner with

Amazon. Metcash has some capability in

unit pick solutions due to its supply to the

convenience sector.

There is a potential scenario whereby

Metcash can be a third-party merchant on

Amazon (being the only large scale national grocery supplier) and provide a readymade

A$0.0

A$0.5

A$1.0

A$1.5

A$2.0

A$2.5

A$3.0

A$3.5

A$4.0

A$4.5

0.0 0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0 4.5 5.0

Population (m)

SydneyMelbourne

Brisbane

Perth

Adelaide

Hobart

Deliv

ery

mark

et

(bn)

at 2

0%

share

of

reta

il fo

od

Figure 51: Valuation sensitivity


Woolworths Wesfarmers Metcash

-50bp gross margin -7% -5% -9%*

-100bps sales revenue -4% -2% -1%

A$100m annual capex -4% -5% n/a

* 25bp reduction

10 April 2017


scale buying platform for Amazon, much the way Morrisons has done so for Amazon in the

UK.

In terms of assessing earnings and valuation sensitivity, we propose a scenario whereby

Amazon entered one of the larger population centres (Sydney, Melbourne or Brisbane)

and grew to a market share of 10% in the chosen market over five years; that would be

half of the probable delivery market. That would be a circa 10bp per annum drag on sales

growth from that market. Expansion to other markets would see that drag multiplied.

Increasing the service cost of incumbents would be more material. A 10ppt increase in

service cost across 5% of sales would have a 50bp impact on gross margin. Every 50bp of

gross margin has 5% impacts on the valuation of Woolworths and Coles. The more subtle,

but perhaps a change with more impact could result from a reduction in supplier

promotional support as a result of additional channel choice. It would not be realistic for a

gross margin impact to be borne entirely at the wholesale level by Metcash – a 25bp

reduction would have a 9% impact on gross margin. An A$100mn per annum increase in

capital expenditure has 3% impact on valuation of Woolworths and Coles.

The Americas: Laggards, but expect rapid growth

The US and Canada are behind Europe in terms of penetration of online grocery, but only

because competition had not forced unprofitable growth – as we reach the inflection point

in technology, we expect US (and possibly Canada) to expand rapidly.

Figure 52: The US is ripe for online grocery growth


The US already has the most diverse set of online grocery models in operation. The battle

for online is most intense around New York City and down through the Philadelphia to

Washington, DC corridor.

There are at least 30 metropolitan areas of 2m people or more, totalling over 150m

people. Each of these regions could be serviced by an online-only model, making the US

the most lucrative market globally.

Beyond the issue of density, other factors, such as truck weight limits, parking and traffic

flows actually favour large urban/suburban centres found in the US compared to European

countries.

0

1

2

3

4

5

6


penetration


Amazon penetration




GDP per capita


Car ownership*




10 April 2017


Figure 53: Online penetration across the US, with significant concentration within the densest areas


10 April 2017


Figure 54: Major players are evaluating; new entrants are expanding


Peapod: All models under one roof

Peapod is the largest online grocery retail platform in the US, with 2015 sales of ~€590m,

covering ~17% of US population (>50m people). It was founded in 1989 and grew using a

combination of its own facilities and agreements with grocery retailers, including Ahold's

Stop & Shop and Giant Food banners.

Strategically, Peapod straddles the various

online grocery business models. Where

Ahold Delhaize has no stores, Peapod

relies on centralised distribution (i.e.

Chicago). In other areas, it uses a small

"ware-room" attached to a store, which

allows it to fulfil orders from a dedicated

area while using some of the store's

infrastructure. It also operates an in-store

picking model in some areas.

It offers both home delivery (~66% of

sales) and click and collect in the areas it

serves, which include its existing store

footprint on the eastern seaboard plus

Chicago and the Midwest. Peapod is co-

branded with the Ahold US store banners.

The use of various models has been driven by a desire to reduce the costs of growth,

especially capital dedicated to online operations. As Ahold Delhaize pointed out at its

recent capital markets day, the capital cost of its online operations are lower than "best-in-

class" automated facilities (which we believe was referring to Ocado).

One of the reasons we prefer the centralised operating model for online grocery is the

opportunity for continuous improvement. There is little the sector can do to materially

improve the level of productivity for in-store picking. Ocado (and other centralised facilities)

on the other hand continues to reduce capex per unit of capacity and increase all-in

productivity. Ahold Delhaize has stated that Peapod generates EBITDA margins of 3-5%

in 'dense' areas, but there is no other disclosure.

Walmart*

Leading grocer in the US, and spending heavily in e-commerce. Walmart bought Jet, a general

merchandise e-commerce company for $3.3bn in 2016 to accelerate its expantion into online. In

grocery, Walmart is focused on C&C, and has rapidly expanded online grocery to more than 600

pickup locations across 100 US markets (compared to just 150 locations across 20 US markets last

year). WMT has tested delivery in select cities.

Kroger*

The largest traditional grocer in the US with ~11% market share. Has tested online home delivery but

no plans to expand further. KR has added 420 ClickList locations last year and now offers online

ordering in over 640 stores.

Ahold

Delhaize*

Ahold Delhaize's Peapod division offers mid-west and east coast coverage for online home delivery.

Operations are a mixture of centralised (but not automated) facilities, hybrid stores with dedicated,

non-customer picking areas, and some store picking.

Instacart

The most advanced third-party service that operates in 36 markets in 25 states. Localised labour

enables rapid order turnaround, but at high cost. Partnerships include Whole Foods Market, Publix,

Harris Teeter and Costco. The business model has shifted from 100% consumer paid to shared retail

margin due to customer pushback on high prices.

Fresh Direct

Fresh-focused, high-end grocery with a concentrated footprint around New York City. The company

raised $189m in 2016 and is planning expansion to Washington, DC. Operations are centralised in

Long Island City, with a second facility under construction in the Bronx.


Figure 55: Peapod warehouse

Source: Company presentation

10 April 2017


Figure 56: Peapod has reasonably good operating metrics

Source: Ahold Delhaize company presentation, Credit Suisse research

Figure 57: Eastern seaboard Peapod coverage


Canada: Excellent infrastructure and harsh winters

In many ways, Canada has the best characteristics for online grocery of any of the

countries we have analysed. Although population is low, it is clustered around five major

centres. Grocery is highly concentrated, which is appealing to a disrupter like Amazon. For

much of the country, driving to a grocery store from January to March (i.e, the middle of

winter) is particularly difficult – an extra incentive for consumers to outsource the process.

Figure 58 shows the characteristics for online grocery for Canada. As noted earlier, the

Greater Toronto Area encompasses 4.6m, with a density of ~55% that of London. The

makeup of that density is positive, with easy-to-navigate high rise towers along the main

arteries and single family dwellings further out.

Peapod Ah.nl "Best in Class" Ocado

All in productivity - 135 150-160 200+

Item completeness - 99.6% ~99% 99%+

Driver timeliness - 99.0% ~95% ~95%

Capex - ~€10m >€100m ~€120m

Capacity (orders per week) - 35k - 200k

Product range 12,000 - - 50,000+

Delivery slots 2 hour - - 1 hour

Boston

New York City

Philadelphia

Harrisburg

Baltimore

Washington DC

10 April 2017


Figure 58: Canada has many characteristics that suit online grocery growth


Mexican ecommerce remains still largely underpenetrated

Mexico has long been a market where structural barriers have limited ecommerce growth.

Among these barriers, we highlight low credit card penetration (c.4% consumer credit to

GDP; source: IMF, Mexico Central Bank), and low fixed broadband penetration (c.47%;

source: IFT). However, we think these barriers will be increasingly less relevant in the

coming years, as the ecommerce market has found alternatives to capture strong growth:

offline payment methods have largely proliferated (accounting for c.22% of total

ecommerce payments), and we expect smartphone penetration to increase from c.40% as

of 2015, to 66% by 2018E.

Figure 59: Infrastructure is the main impediment for online grocery in Mexico


Credit Suisse economics team expects Mexico to have one of EM’s highest growth rates

in ecommerce sales over the next 4 years, at 23% CAGR – outpacing the overall growth

rate of Latin America (c.18%), and specifically, that of Brazil (12%). Total online sales

(including grocery + other categories, but excluding travel) reached ~US$6bn in 2015 and

are expected to at least double by 2019. Online grocery sales at this point are de minimis

(see report on Mexican Ecommerce sector here).

0

1

2

3

4

5

6


penetration


Amazon penetration




GDP per capita


Car ownership*




0

1

2

3

4

5

6


penetration


Amazon penetration


Urban driving

infrastructure


GDP per capita


Car ownership*




https://doc.research-and-analytics.csfb.com/docView?source_id=eqgl&document_id=1054125581&format=PDF&language=ENG&serialid=5l74YkjWPazn7e0Faj%2b3lOBmJTf%2ftmi2kM9bqiL5kZs%3d

10 April 2017


Today the largest player in online grocery retailing is bricks-and-mortar leader, Walmart

México y Centroamérica (70% owned by Wal-Mart Stores, Inc.). For Walmart, online sales

account for 0.9% of total sales (~US$250mn annual sales, as of 2016). Walmart is

achieving growth rates of c.27% y/y, with 156 click-and-collect locations across Mexico

(out of c.2,291 stores in the country), and a delivery business model, which has been

particularly strong at high-income supermarkets, called Superama (responsible for c. 7%

of total sales). We do highlight though, that c.36% of Walmart online sales in Mexico

continue to be driven by general merchandise items (as opposed to pure grocery sales).

Walmart is now spending ~US$100m in annual capex on ecommerce (c.13% of total

capex); below the levels of some US retailers (30-35% of total capex allocated to

ecommerce), but significantly outpacing local competition. In absolute terms, Walmex is

investing over 10x the absolute dollar amount that its nearest competitor, Soriana, is

investing (among brick-and-mortar players), even when the actual sales gap on the

‘physical’ world reaches ‘only’ c.3.5x.

We think this investment gap will result in a larger sales gap in online than the one seen in

physical sales. The question then becomes whether pure ecommerce players can be a

challenger as well. Rappi and Cornershop have become two up-and-coming pure

ecommerce players that are growing in the Mexican market, albeit sales figures for these

two players are not publically available. Amazon entered Mexico in mid-2015 and has so

far not launched AmazonFresh.

Brazil: Many positive characteristics, but household economics the key headwind

The Brazilian online grocery market was estimated at ~US$330m (R$1.1 billion) in 2016

according to Euromonitor, representing only around 3% of total internet retailing in the

country. Lower income and lower urbanisation rates are part of the reason for the modest

size of the local market, but significant back office and logistics' bottlenecks and lower

internet penetration are also key factors.

However, we believe that the structural growth outlook for online grocery retail remains

good, and the gradual clearing of some of the hurdles coupled with higher investment in

marketing and infrastructure are expected to help the segment to fulfill its potential over

time.

Figure 60: Brazil's population and density are the main positive factors


Although the market is small, it has been growing during one of Brazil's worst-ever

recessions. According to Euromonitor, the online grocery market grew at a CAGR of 16%

between 2013 and 2016 and 18% in 2016 (Figure 61), a year in which overall retail sales

0

1

2

3

4

5

6


penetration

Online share of retail

spend

Amazon penetration


culture

Urban driving

infrastructure


GDP per capita


space*

Car ownership*

Prevalence of double-

income households

Inclement seasonal

weather


10 April 2017


declined by ~6% YoY and GDP declined 3.6% YoY. More interesting is that online grocery

grew faster than non-food online for the first time.

With the exception of Companhia Brasileira de Distribuicao (CBD), there are no listed

players with large online grocery operations (CBD's online grocery business generated

over R$200m in sales in 2016, for ~20% share). Neither Carrefour (the largest food retailer

in Brazil) nor Walmart (the third-largest) is selling groceries online. There are some

regional players such as Mambo, Sonda and Zona Sul operating in the segment, but are

still very small.

Figure 61: Online food and non-food growth

Source: Euromonitor, Credit Suisse research

Per capita grocery spend remains low, but there is a large overall addressable market;

R$316bn (US$81bn) in gross revenues according to ABRAS in 2015. There are large,

densely populated metropolitan regions where sizeable mid-to-high end income classes

live that are fully connected and able to pay for convenience. The statistics we have so far

are encouraging: average online basket size is ~ 6x higher than an in-store basket.

Consumers expectations are as high as in other regions, where timeliness of deliveries

and high quality fresh are minimum requirements.

The challenges in logistics are vast, especially in a country with the limitations of Brazil.

CBD has spent a number of years and sizeable investments to develop its own know-how

and remains a pioneer in the segment. Until late 2016 CBD was delivering to customers in

the city of São Paulo directly from three stores, limiting the company's growth potential,

although it still exceeded double-digits.

It is our understanding that the company did not heavily invest in marketing at that time, as

faster growth would have jeopardized the quality of the service. The situation changed in

November 2016 when the company opened a 25,000m2 dedicated distribution centre for

online grocery along with 20 stores that are able to deliver a reduced number of SKUs in

up to 4 hours. There should be 40 stores to process "Express" deliveries by year end

2017, according to the company. We believe that the company's growth rates will

accelerate and that this will drive overall industry growth and bring competitors such as

Carrefour into the mix. Currently we see some local players using the store distribution

model but their operations are small in nature.

As a departure from other regions, online grocery complexity has seen the emergence of

'verticals'' in Brazil. Over the last few years, several specialised websites have emerged,

such as those looking to sell beer (emporio.com, cervejastore.com.br), wine

(redvinhos.com.br, winebrands.com.br), dry goods (homerefill.com) and other categories,

which require specific logistics' structure. This is more feasible in a low labour cost

environment, but this was also the case in the US when online shopping was starting to

emerge (and were ultimately consolidated).

0%

5%

10%

15%

20%

25%

15,000

20,000

25,000

30,000

35,000

40,000

2012 2013 2014 2015 2016

Food and DrinkInternet Retailing

(R$m)

Internet Retailing(R$m)

E-CommerceGrowth (RHS)

Food and DrinkGrowth (RHS)

10 April 2017


Part III: Online grocery economics

The cost of picking

At its most basic, picking 'technology' is somebody other than the customer walking

around a store, shopping just as a customer would. It is inherently more efficient than a

customer because there is no browsing – but that's where the basic efficiencies end. The

hunt for cost savings has pushed this simple process into a wide variety of business

models.

In this section, we explore the economics of the four main picking models.

1a. In-store picking (own labour): No way to run an online business

The most widely-used methodology to pick orders is to use a retailer's own staff to pick

orders in their own stores. This is logical, obvious and enticing; the assets are there (and

could be viewed as a sunk cost); the staff, initially at least, can add picking tasks to their

regular activities. Orders may be incremental – acquired from a competitor, so even a high

cost order could be EBIT accretive.

At some point, however, online order quantities reach a point where they add to store

operating costs (i.e. the cost of getting goods to the store, getting them on the shelves and

the store costs themselves). Looking specifically at the UK grocers, we do not believe they

had any choice during the period of rapid growth over the last five years. Using stores to

fulfill orders allowed them to test the acceptance of online grocery and enabled a rapid and

wide rollout with only minimal incremental costs. They had to respond to the threat of

Ocado and show that they were adapting to a new channel.

But as online grocery grew, cracks in the model appeared. Grocers using the store-pick

method are finding it logistically challenging, expensive, margin negative and no longer a

competitive advantage.

We illustrate these issues in Figure 62, which breaks out the main costs of fulfilling an

average-sized UK online grocery order of £90, with home delivery.

Figure 62: The costs of in-store picking with home delivery (£90 basket)

Source: Kurt Salmon, Credit Suisse research

Our analysis shows that in-store picking with home delivery is loss-making on a direct cost

basis (i.e. even when indirect costs such as rent are excluded)6.

As noted above, part of the appeal of store picking centres on the fact that the

merchandise is already there, in the store, just waiting to be picked. It's almost like it

6 An alternative way of approaching profitability is to ignore many of the variable costs that are associated with regular store

operations and just focus on the incremental costs of picking and delivery. However, the starting point for this exercise would be an operating margin of perhaps ~4% (£3.60 on a £90 basket). Picking and delivery costs would then be paid from that margin, which is perhaps even starker evidence of losses. Note that this assumes the basket is not incremental.

of basket of costs

Distribution to store Variable -3.6% 12.6% (£ 3.24)

Merchandising Variable -1.5% 5.2% (£ 1.35)

Stock loss / shrinkage Variable -0.8% 2.6% (£ 0.68)

Payment charges Variable -0.6% 2.1% (£ 0.54)

Picking Variable -9.5% 33.1% (£ 8.54)

Delivery logistics Fixed per delivery -9.2% 32.1% (£ 8.27)

Digital marketing Mix -0.7% 2.6% (£ 0.66)

Web IT Mix -0.3% 1.0% (£ 0.27)

Customer services / CRM Largely variable -2.5% 8.7% (£ 2.25)

Total direct costs -28.7% 100.0% (£ 25.79)

Cost percentageCostType of costProcess step

3 key factors:

78% of direct

costs

10 April 2017


appeared on the shelves for free. But it didn't – and that is the first of our three main costs

that store picking must absorb – the 'Distribution to store' line item7.

The other two large costs are picking and delivery, which include a number of direct but

hidden costs, such as team management and the IT costs associated with each order.

Delivery costs include fuel, maintenance and depreciation on vehicles, warehouse labour

and driver costs.

The most difficult costs to allocate are the IT costs. They are 'central' costs, but are driven

purely by the online offer (e.g., photographs, product information databases, web page

design, order entry, customer relationship management) Some of these costs are fixed,

and are therefore scalable, while some are variable.

At this point, our analysis suggests grocers are losing money for each delivery they make,

yet we have still not accounted for the rent and associated building costs (e.g., rates,

electricity, security). We are also assuming that this basket is an incremental sale; if the

basket in question had been shopped in person rather than online then the cost of

cannibalisation (i.e. the profit forgone) also needs to be added to the operating loss from

the online basket.

Perhaps the most deceptive issue with in-store picking is that there is a theoretical limit to

picking efficiency, which is necessarily low. This means that scale has very little value as

volumes grow. The layout of a supermarket is designed to be inefficient: grocers want

customers to go to the back corner to get milk, the other corner for meat, and then snacks

near the check-outs. The idea is to get customers to spend as much time as possible

browsing around the aisles (also known as 'dwell time'), which is counterproductive for in-

store pickers. And there is very little technology or automation that can help; wearable

scanners and systems can optimise picking routes and provide some level of assistance,

but they are not groundbreaking.

We are also concerned about using in-store picking as a transitional strategy to something

more sophisticated. The online team, IT staff and management will be focused on avoiding

out of stocks, substitution management and product-supply planning. The substitution

logistics process is complex because substitute items may require segregation from

ordered items so the customer can agree which ones should be kept. There is an added

process where the rejected substitute comes back to the store and is reconciled to the

original invoice, which is important from a customer service/wastage perspective but adds

little value.

Finally, we note the negative impact on store operations. Significant store-based fulfilment

makes inventory management more challenging, requires additional assortment efforts

and impairs customers' shopping experiences as they try to dodge store staff picking

online baskets.

1b. In-store picking (third-party): Somebody has to pay

Third-party picking is an extension of the 'gig economy'. The private company Instacart,

based in California, is the leader in third-party fulfilment as demonstrated by its rapid

expansion, the supportive store economics and its capital-raising ability (it raised $220m in

2014 at a valuation of $2bn and $400m in 2017 at a valuation of $3.4bn).

Instacart was co-founded by a former Amazon employee and launched in 2012. The

model uses an 'aggregated' storefront that allows customers to shop for groceries online

and then employs local labour to pick and deliver orders from a variety of shops. Its asset-

light model supports high growth – it already operates in 36 markets in 25 states across

the US and unlike Uber's regulatory burdens, is relatively unencumbered from growing

internationally.

7 We use data from Kurt Salmon consultants for distribution costs, which are broadly similar to Tesco's disclosure at its most recent

capital markets day. Tesco identified Net Distribution Costs in its cost breakdown, which we estimate to be ~£1.45bn on £48.4bn of sales, or 3%.

10 April 2017


The two key differentiating factors of thirdd-party fulfilment compared to most other grocery

delivery services are speed and cost. Instacart can deliver within an hour because of its

local labour and extensive store network. The use of local labour also keeps costs down –

there is no requirement for temperature controlled vans; shoppers use their own vehicle.

However, Instacart has been the most expensive model for customers by far because it

charges in full for the service, unlike in-house operations.

Instacart's original model charged delivery fees to customers, service charges to retailers

and marked up individual items. The Wall Street Journal reported that on a $69 order, the

company’s profit was $1.40 (a 2.0% margin).

The Wall Street Journal looked at the markups across a number of items and found that

prices ranged from a (rare) discount of 15% to a premium of 26%. Delivery charges are

based on timing, order size and whether the customer has an annual delivery pass

($149/year, which makes deliveries free on orders >$35).

On a random selection of products Instacart's item pricing was 30% higher than an

average of Peapod and FreshDirect. Once the lower delivery fees were incorporated, the

premium fell to 21% - still in the realm of a 'luxury' service, in our view.

Figure 63: The Instacart model is fully-costed … and unsustainable, in our view

Source: WSJ.com, Credit Suisse research

This price discrepancy led to key changes within the Instacart model. The company now

states clearly whether prices on Instacart are the same as local in-store prices, and

customers can see the different pricing policies for each of Instacart's partners. In most

cases, the retailer sets the prices for their products on Instacart; less often Instacart sets

the price.

This means that retailers are now sharing their margin, although the details of that revenue

share are not disclosed. In an interview with the New York Times, the CEO said “Instacart

was adding so much more volume and new dollars to the store that it made sense for

them to partner with us” – which appears to be exactly the same philosophy the UK

grocers had when online grocery started. The appeal of the incremental basket means that

the margin on that basket starts at 25% to 30% - its gross margin. But if the prices are the

Peapod FreshDirect Average InstacartInstacart premium

vs. Avg.

Bananas (6) 2.34 1.99 2.17 2.34 8%

Broccoli 2.39 3.49 2.94 4.49 53%

POM pomegranate juice (48 oz.) 9.99 11.99 10.99 15.79 44%

Fage Total 0% Greek yogurt (17.6 oz.) 3.99 3.99 3.99 5.49 38%

Part-skim mozzarella cheese (8 oz.) 3.00 4.59 3.80 5.69 50%

Organic fat-free milk ( 1/2 gal.) 4.99 5.69 5.34 8.58 61%

Cage-free large brown eggs (12) 3.79 3.79 3.79 4.49 18%

Coca-Cola (2l) 1.89 1.99 1.94 2.49 28%

Dunkin’ Donuts Original Blend coffee (12 oz.) 8.69 9.99 9.34 10.29 10%

Tropicana orange juice, no pulp (59 oz.) 3.00 4.39 3.70 4.89 32%

Cheerios (8.9 oz.) 3.59 3.99 3.79 5.09 34%

Kellogg’s Special K cereal (18 oz.) 3.89 6.29 5.09 7.69 51%

Nutella (13 oz.) 4.99 4.29 4.64 5.59 20%

Thomas’ English Muffins (12 oz.) 4.29 4.19 4.24 2.49 -41%

Item total 60.83 70.66 65.75 85.40 30%

DELIVERY FEE 9.95 5.99 7.97 5.00*

TAX AND OTHER 1.05 1.18 1.12 0.22

Total 71.83 77.83 74.83 90.62 21%

* average of 1-hour and 2-hour delivery costs

10 April 2017


same as in-store and a customer can get their goods within an hour, the risk of

cannibalisation is extremely high. We see a sharing of margin as another way of fully-

costing the picking and delivery service and paying it from store profits.

From the retailer's perspective, Instacart still offers a relatively low-cost and rapid way to

get online, but at the expense of giving up control of the customer relationship (and

perhaps more importantly, their data). For mid-sized chains, Instacart may be the initial

'lowest-risk' path to online fulfilment. Once the customer base has been confirmed, offering

an integrated online solution may be the next logical step.

Ultimately, we do not believe this is will grow much beyond a niche solution. The cost to

the customer is high, which structurally limits the addressable market; any reduction in

costs must necessarily come from store subsidies. Once more efficient solutions became

broadly available we expect Instacart's market share to shrink unless it is able to monetise

the data it gathers to offset lower margins.

2. Dark stores: For better and for worse

Dark stores are dedicated online picking and shipping facilities, but look and feel like a

store within the logistics network. They receive goods from distribution centres and

merchandise them much more efficiently than a store, with the layout and facilities more

akin to a warehouse than an actual store. Dark stores would typically carry a full or even

extended supermarket product range, which minimises substitutions.

At least some level of automated goods handling is likely, with the most sophisticated dark

stores employing significant amounts of technology to improve picking rates, minimise

errors and lower costs. For example, dark stores often use refrigerated chambers rather

than refrigerated cabinets.

Figure 64: Inside Tesco's brightly lit, latest generation 'dark' store

Source: Tesco

Although dark stores perform better than the in-store pick model, there are three issues

that have a negative impact on profitability. First, because dark stores are still 'stores'

within the network, distribution costs and the associated double-handling of goods remains

an issue.

10 April 2017


Figure 65: Swisslog zone-pick system with conveyors

Source: Swisslog

The second issue is rent; the in-store model picks baskets from existing stores 'rent free'.

Although it is only an accounting change, once there is a dedicated operation for online

orders, the full occupancy cost needs to be charged back onto orders.

Figure 66: Waitrose dark store; picking with visually enhanced RFID scanners

Source: Waitrose

Third, a dark store is still sub-scale. Dark stores do not enjoy the efficiencies that come

from more consistent inventory turns, and automation equipment and other operational

improvements are not as prevalent as they are in a centralised fulfilment operation.

Figure 67: Wearable scanners – Yes Figure 68: Palletising robot – No

Source: Motorola Source: Kuka

We estimate that Tesco's latest generation dark store in Erith has a yearly throughput of

up to £150m, compared to Ocado's smallest CFC, which has a design capacity of £350m.

10 April 2017


Ultimately, dark stores and CFCs perform the same function at different scale.

3. Hybrid stores: Better, but not optimal

Peapod is the global leader in using hybrid stores. By having associated 'ware-rooms',

Peapod leverages the distribution function and general store costs, but is able to

segregate online picking from regular store operations. However, due to the low

throughput there is little potential for automation, and range is limited unless orders can be

supplemented with lower velocity products from the attached store. While preferable to

dedicated in-store picking, we do not view hybrid stores as having enough incremental

benefits to be a long-term solution.

Can hybrid stores solve space issues?

Because there is an abundance of oversized stores within grocery retail, even the modest

benefits of hybrid stores will result in some being re-tasked as online fulfilment centres.

For example Tesco turned a store near Dudley into a hybrid operation; store space fell

from 108,000 ft2 to 66,500 ft

2, with the recovered space being converted into dot-com

fulfilment and a C&C site.

However, converting a site is complicated and expensive. Ingress and egress is different

for the high volume of large trucks and vans compared to a store designed for customer

parking and individual customer flow. Store locations may be sub-optimal, requiring

extended drive times for the catchment area compared to choosing a greenfield site. Other

critical factors such as roof height and the additional capital costs of retrofitting an existing

store layout are also impediments to repurposing existing stores.

4. Centralised fulfilment: The only path to online profitability

A centralised business model removes traditional expenses (rent, labour) and adds

assortment breadth, improves order accuracy, fresher food and scale. We view the model

as inherently superior to one that piggy-backs on existing store infrastructure.

Because a CFC operates as a warehouse, a picking operation and a distribution centre all

in one, the 'distribution to store' line item cost (shown in Figure 62) is eliminated. The scale

of the operation allows for infrastructure solutions such as independently-routed,

computer-controlled customer totes, automated inventory systems, goods-to-man

workstations, optimised picking methodologies by product line and other strategies.

A key technology used in both Ocado CFCs and Tesco's dark stores are 'goods-to-man'

(GTM) workstations. Bar-coded customer totes (bins) move around the facility without

human interaction on conveyor systems controlled by computer. Some inventory items are

also managed in the same way – via computer-controlled conveyor systems. GTM

technology synchronises inventory totes and customer totes and routes them to a manned

picking station.

GTM enables very fast picking because the picker doesn’t have to move – they take goods

from the inventory tote on one conveyor and put them into a customer tote on the adjacent

conveyor. GTM technology is ideal for hard-to-pick items, such as low-volume goods

where a picker would have to walk a long way to reach the inventory. This compares to

fast-moving items such as bananas or baked beans, which would normally be kept on

traditional shelving given its much lower capital cost.

While these technologies create efficiencies over store-based picking, the long-term

benefit of industrial automation is the ability to incorporate new technology as it improves.

This is evident in Ocado's new CFC, which uses radically different processes in a 'hive'

format to pick orders. This technology will bring GTM capabilities for all ambient and

chilled goods, leading to a material increase in picking rates. We expect a step-change in

productivity.

10 April 2017


It's not just about replacing people

The most obvious advantage of a hive system is to reduce the labour costs associated

with picking a basket. However, there are a number of other benefits beyond mere cost

savings:

■ Concurrent picking. Whereas an order within a semi-automated facility may take a

number of hours to wind its way through the conveyor systems, in a hive system

picking 50 items takes the same time as picking one item. An order that can be picked

and at the dock door within 10 minutes allows for same-day delivery, potentially <1hr,

depending on the location of the facility.

■ Dynamic order processing / buffering. Existing facilities are generally one-way; an

order starts the fulfilment process and moves through completion. Latency between

completion and shipping is minimal. For orders placed in advance in a hive structure,

orders can be picked and placed back in the hive, available for shipping later. This

optimises the picking utilisation and also allows for order changes until just before

shipping.

■ Density. By creating a 3-dimensional storage system, the footprint of the facility can be

materially reduced, thereby allowing it to be positioned closer to customers and

reducing overall costs.

■ Redundancy. Independently-targeted bots and multiple-pick stations allow control

systems to bypass any areas with mechanical failures.

■ Future-proofing. The pick stations have been designed to incorporate robotic picking,

further enhancing efficiency without altering the overall structure. We would expect

heavier 'first in the bag' items like bottled water or laundry detergent to be the first

categories earmarked for robotic picking.

■ Modular and scalable. Allows facilities to grow as demand grows. This improves

companies' IRR and ROCE metrics.

Ocado designed its hive system specifically for food, but other three-dimensional storage

and retrieval systems have been implemented before (Figure 69 – note that the AutoStore

bots cover two cells, whereas Ocado's bots in Figure 70 are one-celled, allowing for much

higher throughput). Ocado has stated that it expects CFC4 (still under construction) to

achieve a units per hour (UPH) target of "over 200" compared to the 160 it achieved in

2015. At that rate, we estimate that 2016's CFC costs would have been ~6.2% of

revenues rather than 7.7% and added £17m of additional margin. We also believe the

target is undemanding given that GTM workstations currently exceed 600 UPHs.

Figure 69: Asda AutoStore Installation Figure 70: Ocado's 'hive'

Source: Asda, Swisslog Source: Company data

10 April 2017


More importantly, this solution is scalable. The smallest hive installation at CFC3 in

Andover is expected to have 1,000 robots. As long as the control systems (e.g., software

logic, communications) are able to scale, adding more hardware to the grid is relatively

straightforward, leading to higher throughput and more SKU choice. The three-

dimensional, very dense nature of the system means that the real estate footprint required

for any given throughput is materially smaller – one of the key drivers in determining CFC

locations.

Lastly, because so much of the basket is now using GTM technology, we expect robots to

replace stationary workers, which is exactly what the CEO of Ocado said at this year's

results presentation:

"Every human touch point in [CFC3] is designed to one day be replaced

by a robotic solution" – Tim Steiner, CEO of Ocado

For further reading into warehouse and logistics automation, we refer you to today's

publication from our capital goods team Warehouse Automation and Robotics.

The cost of delivery (or collection)

Delivery costs are either the most expensive or second-most expensive part of the

fulfilment process, depending on the size of the basket (because picking is a variable cost,

while delivery is almost entirely a fixed cost per order). Online basket sizes remain high

(for now), and many are shopped as substitutes for specific missions (e.g., parties,

holidays, ). The more mainstream online grocery becomes, the more downward pressure

on basket sizes. The impact of this can be seen in Figure 71, using a £90 basket size.

Figure 71: Basket size sensitivity relative to delivery costs


The costs of delivery are straightforward: labour, fuel and maintenance / depreciation on

the vehicles. The primary difference when comparing food and non-food delivery costs is

that food customers choose their own time slots, which causes a number of knock-on

effects:

■ Longer routes;

■ Erratic routes; and

■ Customer service pressure, requiring time buffers.

Furthermore, temperature controls, product returns, rejected substitutions and absent

customers all add complexity to the delivery process, almost none of which apply to non-

food. Scale and high drop densities help counteract some of these issues, but costs are

relatively linear, and defy easy solutions.

Ocado has been able to gain some operational leverage by increasing the drops per van

(rather than increasing drops per hour). In addition to higher customer densities,

companies can improve efficiencies by providing incentives for less popular time slots (or

disincentives for more popular ones), widening delivery windows, and adding capacity on

key days (such as Sunday). Using a combination of these initiatives Ocado has been able

to improve drops per van per week from 133 in 2010 to 176 in 2016 (+32%), with a long-

term target of 190 (Figure 72).

Our cost estimates are based on between 3 and 4 drops per hour, with the bulk of that

cost being the van driver's labour, based on UK rates. Drops per hour can vary on a city-

Basket size £60 £70 £80 £90 £100 £110

Gross margin per basket 17.00£ 19.50£ 22.00£ 24.50£ 27.00£ 29.50£

Delivery cost (£ 8.27) (£ 8.27) (£ 8.27) (£ 8.27) (£ 8.27) (£ 8.27)

Percentage of gross margin 49% 42% 38% 34% 31% 28%


10 April 2017


by-city basis due to road infrastructure and general traffic conditions, while labour rates

can vary widely on a country-by-country basis.

Given the high cost of delivery, there is significant momentum for C&C amongst grocers.

C&C is still additive to costs, but not nearly as additive as home delivery. It can be a

compelling offer: it is more convenient than doing an in-person shop and the pickup point

effectively becomes a hypermarket. If collecting from a store, customers can add items to

their basket upon collection and it eliminates the inconvenience of being home at a certain

time as is required for home delivery. Non-store collection points can be anywhere, as we

see in Figure 73, which shows an unattended, refrigerated collection site for Waitrose.

Amazon is set to launch a purpose-built C&C site in Seattle for a service it calls

AmazonFresh Pickup. This would be the first model that uses centralised picked with a

C&C delivery model, and should offer extremely compelling economics given its capital

light nature and lower operating expenses.

Figure 72: Ocado's drops / van / week Figure 73: Unattended C&C

Source: Company data Source: Credit Suisse research

For retailers, C&C is a way of eliminating the logistically difficult and expensive 'last mile'

while leveraging existing online store fronts, back-end logistics expertise and broad

networks. Compared to the loss of 1.4% per basket in our example, a C&C order

generates 3.3% of positive contribution (again, on a £90 basket size).

However, C&C does not solve all of the problems nor does it eliminate all of the costs that

are borne by home delivery. Orders still have to be picked by retailers' labour, baskets are

still handled as far as the collection point, and sizeable IT costs are still incurred. C&C

then introduces some additional costs, especially when the pickup point is within a store.

Additional warehousing adds labour, storage and transaction processing costs and

throughput is constrained by individual stores' warehouse and staffing levels. Until now,

stores have not been designed with C&C in mind. We also make no allowances for

returns.

Improving delivery economics

All of the incumbents are looking to improve their economics, which over the near term,

will likely be a combination of revenue increases, cost reductions or shared assets. Longer

term, the replacement of driver labour is the most obvious target to save costs.

Fees: A question of value

The easiest method is to just charge more. Assuming that volumes stay reasonably

constant, higher fees will immediately improve the economics.

Tesco increased its minimum spend for online orders for home delivery from £25 to £40

and added a £3 surcharge on top of the existing delivery charge for any order under the

£40 limit (those charges range from £1 to £6 depending on time of day). Tesco's online

growth slowed rapidly from the time it instituted the change, as we showed in Figure 18.

Other retailers moved in the same direction. Asda increased prices in January of last year

120

130

140

150

160

170

180

190

200

2010 2011 2012 2013 2014 2015 2016 IPOTarget

NewTarget

10 April 2017


and Waitrose added a £2 fee for orders under £30 on C&C orders. We expect more

tweaks throughout 2017 as retailers grapple with the changing economics of grocery.

Dynamic delivery pricing is more

sophisticated than a simple increase in

price. Unlike picking, the value of delivery

to the customer can be estimated because

retailers can determine the effort required

to get from the customer's house to the

nearest grocery store.

Adding this knowledge to planned van

routes, the time and day of delivery, store

location data and other factors provides an

opportunity to dynamically price delivery

slots. Data analytics can predict whether

this is an incremental or a cannibalised

basket. Pricing can be set so the customer

gets good value, but does not get a free

ride (i.e. you get what you pay for), which

is especially valuable for hard-to-reach

locations.

We already see this in plain vanilla

delivery services such as the Trader Joe's

in midtown Manhattan (Figure 74). This

delivery program is for own-picked baskets

(i.e. you go to the store and shop, but then

the store takes over delivery). As an aside,

the cost of this service appears extremely high (between $6.95 and $17.95 plus tip,

depending on location) when compared to the delivery fees charged for online elsewhere.

Shared delivery

There is potential for a delivery van that shares suppliers and drives around the same

neighbourhoods on a consistent basis. This materially increases drop density – the key

issue in current online delivery economics. We see potential for existing delivery services

(like UPS), taxi services (including models such as Uber or Lyft), or meal delivery services

(such as Deliveroo or Just Eat) to participate in a milkman-like model.

A local delivery van operator would open up online grocery to lower-margin channels, such

as discount grocers (e.g., Aldi, Lidl, DIA). There would be less flexibility on delivery slots

and less choice due to the discounters' limited assortment, but there would be materially

lower basket prices. The ability to pre-package items or groups of items would further

reduce costs.

Meal kits or recipe boxes are often paired with home delivery – partially from necessity

(meal kit producers are 'asset light' with no store presence), but also because they see this

as part of their value proposition. We believe that the delivery element is not a competitive

advantage, and that the consolidation of delivery models could improve the economics for

this segment as well.

Direct-to-fridge delivery

In Sweden, ICA has partnered with PostNord (not listed) and Glue (not listed) to offer ICA

Infridge, a service that delivers groceries directly to customers' refrigerators. Glue has

smart locking technology that allows authorised delivery personnel to enter homes,

notifying homeowners when their door is unlocked and re-locked. This model improves

routing efficiency, eliminates any delivery coordination issues and increases asset

utilisation. There is a modest technological hurdle in retrofitting door locks, but we think the

trickier hurdle is getting customers comfortable with strangers being in their homes.

Figure 74: Manhattan delivery costs


10 April 2017


Driverless vans, drones and robots

Ocado estimates that labour is 60% of its overall delivery cost. Those costs are

approximately 11.5% of revenues, so being able to eliminate drivers would add almost

700bps of margin – easily the single-biggest line item within the cost structure.

Driverless vehicle technology continues to evolve; it is just a matter of time before the legal

and technological issues are resolved. However, deliveries without any human interaction

introduce more challenges beyond the basic logistics: how do you actually get the

groceries, and what if there is a problem (missing items, substitutions, returns) with any of

them?

We see solutions to these problems, including perhaps a local delivery person that does

the 'last 10 metres' rather than the 'last mile'. For addresses where access is easy,

secured compartments can be created so that customers can go to the vehicle and

remove their groceries. While tweaks to the business model may be required, we see

driverless vehicles as a key enabling technology that should move online grocery below

the cost curve of in-person shopping.

Interestingly, driverless delivery would breathe new life into the store-pick model; by

eliminating up to 600bps of cost, nearly all models would become profitable. At that point,

customer adoption would be more dependent on the relative quality of the service and the

relative cost of operating stores versus centralised facilities.

Cost and margin comparisons of all business models

Figure 75: Overall cost comparisons of various picking and delivery methodologies


Picking methodology In-store Dark store Hybrid 3rd-party Centralised In-store Centralised

Delivery location: Home Home Home Home Home C&C C&C

Industry leader Tesco Tesco Peapod Instacart Ocado Leclerc Amazon

Gross margin: £90 basket 27% 27% 27% 27% 27% 25% 25%

Costs:

Distribution to store -4% 0% 0% 0% 4% 0% 4%

Merchandising -2% 1% 1% 0% 1% 0% 1%

Stock loss / shrinkage -1% 0% 0% 0% 0% 0% 0%

Payment charges -1% 0% 0% 0% 0% 0% 0%

Picking -9% 2% 2% 9% 5% 0% 5%

Delivery logistics -9% -1% -1% 9% -2% 7% 5%

Digital marketing -1% 0% 0% 1% 0% 0% 0%

Web IT 0% 0% 0% 0% 0% 0% 0%

Customer services / CRM -3% 0% 0% 3% 0% 0% 0%

Savings vs. Store Pick - 2% 2% 22% 8% 7% 14%

Total direct costs -29% -26% -27% -6% -21% -22% -14%

Direct P&L per basket -1% 1% 1% 21% 6% 3% 11%

Cost savings relative to in-store picking + home delivery

10 A

pril 2

01

7

Glo

ba

l on

line g

roc

ery

5

0

Online grocery costs from the customers' perspective

Figure 76: Delivery economics across retailers; comparisons are not straightforward. All currencies are local.


AmazonFresh* Amazon Pantry* AmazonFresh* Amazon Pantry* Ocado* Tesco* Sainsbury*

US US UK UK UK UK UK

Cost of

membership

Amazon Prime Membership

+

$14.99/m

Amazon Prime: $99/yr,

$10.99/m

Amazon Prime Membership

+

£6.99/m

Amazon Prime:

£79/yr, £7.99/mFree Free Free

Cost of

DeliveryFree: Orders > $50 $5.99 per Prime Pantry box Free: Orders > £40

Per Order:

£2.99 for first box

99p for each add'l box

£2.99 - £6.99

Free: orders> £75

Discounts available <9am

and >9pm)

Eco-friendly slots

highlighted.

£4 for all orders under £40

£1 - £7. Discounts slots

(Tue, Wed, Thursday)

Free: Mon-Thu on orders >

£100

Minimum

spend$50 None £40 None £40

None; £40 minimum for free

delivery using Delivery Pass

£25; £40 minimum for free

delivery using Delivery Pass

Delivery passes - - - -

Anytime

£109.99 pa, £49.99/6m,

£10.99/m

Tues, Wed & Thurs £69.99

pa, £34.99/6m, £6.99/m

Anytime

£60/yr, £6/m

Tue, Wed & Thurs £30/yr,

3/m

Anytime

£60/yr, £35/6m, £20/3m

Tue, Wed & Thurs

30/yr, £18/6m, £10/3m

Woolworths* Coles* Carrefour* Auchan Grocery Gateway Instacart Walmart*

Australia Australia France France Canada US Mexico

Cost of

membership- Free

Free registration for home

deliveryFree Optional, $149/yr Free

Cost of

Delivery

$11 for standard delivery

$9 orders $150-200

$6 orders 200-250

$3 for orders $250-300

free: > $300

$18 for premium slots

$10 for average slots

$4 for discount slots

$8 orders € 50-100

€ 5 for orders €100-150

free: > € 150

Variable:

$9.99 per delivery, may be

higher for certain slots.

Slots are 90-minute

windows.

Orders>$35

$3.99 - 2 hrs delivery

$5.99 - 1 hr delivery

Orders $10-35

$7.99 - 2 hrs delivery

$9.99 - 1 hr delivery

Home delivery:

Shipping cost 30 MXN

(covers 5Km away from

closest stores) each

additional KM outside this

area will charge 5 MXN

Minimum

spend$50 $50 € 50

€120 for delivery; No

Minimum purchase for store

pickup

$45 $10 n/a

Delivery passes

$89 /yr, $60/6m, $40/3m

free deliveries on orders

>$100

- - - -$149pa; free deliveries on

orders >$35-

* covered by (or parent company is covered by) Credit Suisse research

10 A

pril 2

01

7

Glo

ba

l on

line

gro

cery

5

1

Figure 77: Delivery economics (cont.)


Peapod* Fresh Direct Kroger* Walmart* ah.nl* Lenta* O'key*

US US US US The Netherlands Russia Russia

Cost of

membershipFree Free Free Free Free None None

Cost of

Delivery

$6.95: Orders $60 to $100

$9.95: Orders >$100

Variable savings rates from

$2 to $4 depending on time

$5.99 to $9.99

Eco-friendly slots

highlighted.

Standard Delivery: $10.95

Priority fee: addl $5

Pickup: $4.95

Variable ($7-10) depends

on the Weight and location

of delivery

Free Store pickup

From €3.91 and € 12.95

depending on time of

delivery (6 days per week

between 7am and

10.30pm); Free with the

purchase of certain products

(which change). C&C: €1.50

<30kg: RUB249; 30-40kg:

RUB299; 40-60kg:

RUB399; 60-80kg: RUB499

(over 90% of orders are

<30kg)

Free for orders >

RUB2700, RUB249 for

orders below.

Minimum

spend$60 $30

Grocery: $50

Floral /event planning: $25

$30-50 depending on

location

€70 unless "free delivery"

items purchased; no C&C

minimum.

RUB500 None

Delivery passes

$59/6m

Free for orders>$100

$6.95 credit for orders $60

to $100

$119/yr, $69/6m - -

Not offered, but €1

reduction in fees if total

spend exceeds €100

None None

Alibaba* JD* Rakuten* Lohaco (Askul)* Omni7 (Seven & I)* Aeon X5*

China China Japan Japan Japan Japan Russia

Cost of

membershipFree Free Free Free Free Free Free

Cost of

Delivery

RMB20 for order < RMB50

and weight within 5KG;

RMB5 for order between

RMB50 -RMB88 and weight

within 5KG; RMB1 for every

extra 1KG above 5KG

Rmb6 delivery fee for order

below minimum spend +

Rmb1 for every 1kg above

the bar; Rmb99 for premium

service (2-hour delivery);

Rmb6 to select more

specific time slots

Free delivery above ¥5,000.

Charges range from ¥648 to

¥1,080


Charges range from ¥350 to

¥500

Free delivery above ¥3,240

(¥1,500 for seven net

shopping). Delivery charges

¥324


Delivery charges ¥324Variable

Minimum

spend

Free delivery for order

above RMB88 and weight

within 20KG

Free delivery above RMB99

and under 10KGNone None None None RUB2000

Delivery passes -

Rmb149/yr. - includes 60

free delivery vouchers, free

return and exchange, cash

back and other services.

n/a n/a n/a n/a None

* covered by (or parent company is covered by) Credit Suisse research

10 April 2017


Appendix A: Multi-factor country model We have created a country framework with 13 independent factors that we believe

correlate to online grocery adoption and profitability. We have used this model to compare

the key European countries along with the US, Canada, Japan, Australia, South Korea

and China.

Figure 78: Model characteristics for the UK


Our work is based on a model used in a Coca-Cola Retailing Research Council report,

which engaged McKinsey to analyse online grocery in Europe. Broadly speaking, the

characteristics fit within the same four broad categories. However, our model is materially

different and results in different conclusions.

We have eliminated a number of backward-looking metrics that we believe do not help

predict the rate of online adoption. For instance, "Grocery basket size" and "Share of

stock-up baskets" were used to provide insight in the economics of online grocery; our

view is that metrics like basket size change with the introduction of online grocery, either

due to incentives provided or a change in behaviour. For example, the pricing and

promotion schemes employed by Ocado are used to provide incentives to move towards

bigger baskets.

We also do not include "Cost of labour" as a material factor in online grocery adoption.

Labour could be viewed as inversely correlated to online grocery growth (i.e. cheap labour

helps lower picking and last-mile delivery costs), but this is primarily true for an in-store

picking model. When centralised picking is used, higher labour costs actually benefit

online grocery because technology can be used to automate processes and tasks, thereby

widening the labour cost differential to traditional grocery.

The effect of discounter penetration on online grocery is unclear. They do not have a

significant presence online (hence, reducing the pressure to move online), but we believe

that online gives traditional grocers a point of differentiation, thereby increasing the

potential for a move online. Awareness of very low prices may be a detractor.

Urban factors are critical in online grocery, but overall population density is misleading, as

we discuss within the report. For example, over 600m people in China are in rural regions,

wholly unsuitable for online grocery. Our focus is on the number and size of metropolitan

areas, which we have now incorporated into the model.

We have included weather as a factor. For example the average daily high in Montreal

during the month of January is -4°C, the average low is -14°C and it receives an average

0

1

2

3

4

5

6


penetration


Amazon penetration




GDP per capita


Car ownership*




10 April 2017


snowfall of 50cm during the month – conditions that we believe are conducive to online

grocery for the 4.0m people in the Greater Montreal area. Countries that have excellent

weather score a 2 (the highest score), while those that have seasonally poor weather such

as Canada would score a 6 (the lowest score). Our scoring is based on temperature and

precipitation. It does not include hurricane/tornado risk.

Other factors that we have included are the road infrastructure within the key metropolitan

areas (which correlates to service levels and costs) and car ownership, which is inversely

correlated to online shopping (for home delivery), but positively correlated for C&C..

We also note that traffic patterns, dwelling mix, and transportation regulations all have an

impact on delivery costs and demand.

10 April 2017


Appendix B: Further reading / viewing

■ YouTube - YASKAWA Motoman robot depalletizing random-cases

YouTube video, Industrial Perception, 2014

■ Robotics in Logistics; A DPDHL perspective on implications

DHL Customer Solutions and Innovation

■ Internet of Things in Logistics

DHL Trend Research and Cisco Consulting Services

■ Shaping the Future of Online Grocery

Coca-Cola Retailing Research Council, Europe by McKinsey & Co.

■ Ocado Warehouse Distribution Robot

Case study, Tharsus

■ Computer scientists find new shortcuts for the infamous traveling salesman problem

Wired, January 2013

■ How Retailers Should Think About Online Versus In-Store Pricing

Rafi Mohammed, Harvard Business Review, January 2017

■ Introducing "Handle"

YouTube video, Boston Dynamics, February 2017

■ Infridge delivery by ICA, PostNord and Glue

YouTube video, April 2016

■ How to win in online grocery: Advice from a pioneer

McKinsey & Company, December 20164

https://www.youtube.com/watch?v=RJd8WgDT4vI

http://www.dhl.com/content/dam/downloads/g0/about_us/logistics_insights/dhl_trendreport_robotics.pdf

http://www.dhl.com/content/dam/Local_Images/g0/New_aboutus/innovation/DHLTrendReport_Internet_of_things.pdf

https://prod-wp.pub.coke.com/wp-content/uploads/sites/24/2015/03/shaping_the_future_online_grocery_english.pdf

http://www.tharsus.co.uk/case-study/ocado-smart-platform-robot/

https://www.wired.com/2013/01/traveling-salesman-problem/

https://www.wired.com/2013/01/traveling-salesman-problem/

https://hbr.org/2017/01/how-retailers-should-think-about-online-versus-in-store-pricing?utm_campaign=News%20and%20Whitepapers&utm_content=47290180&utm_medium=social&utm_source=linkedin

https://www.youtube.com/watch?v=-7xvqQeoA8c

https://www.youtube.com/watch?v=lN0-XueUD9Q

https://www.youtube.com/watch?v=lN0-XueUD9Q

http://www.mckinsey.com/industries/retail/our-insights/how-to-win-in-online-grocery-advice-from-a-pioneer#0

http://www.mckinsey.com/industries/retail/our-insights/how-to-win-in-online-grocery-advice-from-a-pioneer#0

10 April 2017


Unless otherwise indicated, all prices are taken at the close of the trading session of the pricing date quoted. Companies Mentioned (Price as of 05-Apr-2017) AEON (8267.T, ¥1,639) ASKUL (2678.T, ¥2,992) Ahold Delhaize (AD.AS, €19.56) Alibaba Group Holding Limited (BABA.N, $107.44) Alphabet (GOOGL.OQ, $848.91) Amazon com Inc. (AMZN.OQ, $909.28) Apple Inc (AAPL.OQ, $144.02) CJ Logistics (000120.KS, W161,000) COFCO Meat Hldg (1610.HK, HK$1.91) Carrefour (CARR.PA, €21.62) Chedraui (CHDRAUIB.MX, MXN40.05) Companhia Brasileira de Distribuicao (PCAR4.SA, R$60.03) Costco Wholesale Corporation (COST.OQ, $167.0) Coupang (Unlisted) DIA (DIDA.MC, €5.426) Daifuku (6383.T, ¥2,707) Dixy Group of Companies (DIXY.MM, Rbl249.0) E-MART Co. Ltd (139480.KS, W212,500) Honeywell International Inc. (HON.N, $123.57) ICA Gruppen (ICAA.ST, Skr303.9) Ito-Yokado Co., Ltd. (Unlisted) J Sainsbury (SBRY.L, 256.3p) JD.com (JD.OQ, $31.53) Just Eat (JE.L, 564.0p) Kion Group (KGX.DE, €61.6) Kroger Co. (KR.N, $29.31) Kuka (KU2G.DE, €101.25) Lenta Ltd (LNTAq.L, $6.9) Loblaw Companies Limited (L.TO, C$71.2) Lotte Shopping (023530.KS, W220,000) Magnit (MGNTq.L, $38.8) Marks & Spencer (MKS.L, 334.8p) Metcash (MTS.AX, A$2.31) Metro DE (MEOG.DE, €29.25) NTT DoCoMo (9437.T, ¥2,632) Ocado Plc (OCDO.L, 246.7p) Oisix (3182.T, ¥2,235) Okey (OKEYq.L, $2.15) Publix Super Mkt (PUSH.PK, $14.75) Rakuten (4755.T, ¥1,142) Soriana (SORIANAB.MX, MXN43.98) Tesco (TSCO.L, 184.25p) United Parcel Service Inc. (UPS.N, $106.74) Wal-Mart Stores, Inc. (WMT.N, $71.65) Walmex (WALMEX.MX, MXN43.12) Wesfarmers (WES.AX, A$44.26) Whole Foods Market (WFM.OQ, $30.25) Wm Morrison (MRW.L, 232.3p) Woolworths (WOW.AX, A$26.77) X5 Retail Group (PJPq.L, $34.14)

Disclosure Appendix

Analyst Certification Stewart McGuire, CFA, A-Hyung Cho, Antonio Gonzalez, CFA, Stephen Ju, Edward J. Kelly, CFA, CPA, Andre Kukhnin, CFA, Grant Saligari, Tobias Stingelin, CFA, Evan Zhou, Victoria Petrova, Olga Bystrova, CFA and Keiichi Yoneshima each certify, with respect to the companies or securities that the individual analyzes, that (1) the views expressed in this report accurately reflect his or her personal views about all of the subject companies and securities and (2) no part of his or her compensation was, is or will be directly or indirectly related to the specific recommendations or views expressed in this report.

10 April 2017


3-Year Price and Rating History for CJ Logistics (000120.KS)

000120.KS Closing Price Target Price

Date (W) (W) Rating

05-Sep-16 202,500 270,000 O *

* Asterisk signifies initiation or assumption of coverage.

O U T PERFO RM

3-Year Price and Rating History for E-MART Co. Ltd (139480.KS)


Date (W) (W) Rating

24-Jul-14 229,000 250,000 N

14-May-15 252,500 R

15-May-15 252,500 250,000 N

04-Aug-15 245,000 R

25-Sep-15 226,000 250,000 N

16-Nov-15 216,000 260,000 O

10-May-16 186,000 248,000

10-Aug-16 166,500 217,000

07-Nov-16 160,500 228,000

03-Feb-17 204,000 235,000


N EU T RA L

REST RICT ED

O U T PERFO RM

3-Year Price and Rating History for Lotte Shopping (023530.KS)


Date (W) (W) Rating

12-May-14 304,000 350,000 N

10-Nov-14 284,500 320,000

05-Feb-15 248,500 300,000

24-Feb-15 247,000 R

09-Jun-15 237,000 300,000 N

10-Aug-15 204,500 250,000

09-Nov-15 219,000 240,000

05-Feb-16 237,000 210,000

09-May-16 235,500 206,000 U

08-Aug-16 194,500 190,000

02-Feb-17 226,000 220,000 N


N EU T RA L

REST RICT ED

U N D ERPERFO RM

The analyst(s) responsible for preparing this research report received Compensation that is based upon various factors including Credit Suisse's total revenues, a portion of which are generated by Credit Suisse's investment banking activities

As of December 10, 2012 Analysts’ stock rating are defined as follows: Outperform (O) : The stock’s total return is expected to outperform the relevant benchmark* over the next 12 months. Neutral (N) : The stock’s total return is expected to be in line with the relevant benchmark* over the next 12 months. Underperform (U) : The stock’s total return is expected to underperform the relevant benchmark* over the next 12 months. *Relevant benchmark by region: As of 10th December 2012, Japanese ratings are based on a stock’s total return relative to the analyst's coverage universe which consists of all companies covered by the analyst within the relevant sector, with Outperforms representing the most attractive, Neutrals the less attractive, a nd Underperforms the least attractive investment opportunities. As of 2nd October 2012, U.S. and Canadian as well as European ra tings are based on a stock’s total return relative to the analyst's coverage universe which consists of all companies covered by the analyst within the relevant sector, with Outperforms representing the most attractive, Neutrals the less attractive, and Underperforms the least attractive investment opportunities. For Latin American and non-Japan Asia stocks, ratings are based on a stock’s total return relative to the average total return of the relevant country or regional benchmark; prior to 2nd October 2012 U.S. and Canadian ratings were based on (1) a stock’s absolute total return potential to its current share price and (2) the relative attractiv eness of a stock’s total return potential within

10 April 2017


an analyst’s coverage universe. For Australian and New Zealand s tocks, the expected total return (ETR) calculation includes 12-month rolling dividend yield. An Outperform rating is assigned where an ETR is greater than or equal to 7.5%; Underperform where an ETR less than or equal to 5%. A Neutral may be assigned where the ETR is between -5% and 15%. The overlapping rating range allows analysts to assign a rating that puts ETR in the context of associated risks. Prior to 18 May 2015, ETR ranges for Outperform and Underperform ratings did not overlap with Neutral thresho lds between 15% and 7.5%, which was in operation from 7 July 2011. Restricted (R) : In certain circumstances, Credit Suisse policy and/or applicable law and regulations preclude certain types of communications, including an investment recommendation, during the course of Credit Suisse's engagement in an investment banking transaction and in certain other circumstances. Not Rated (NR) : Credit Suisse Equity Research does not have an investment rating or view on the stock or any other securities related to the company at this time. Not Covered (NC) : Credit Suisse Equity Research does not provide ongoing coverage of the company or offer an investment rating or investment view on the equity security of the company or related products.

Volatility Indicator [V] : A stock is defined as volatile if the stock price has moved up or down by 20% or more in a month in at least 8 of the past 24 months or the analyst expects significant volatility going forward.

Analysts’ sector weightings are distinct from analysts’ stock ratings and are based on the analyst’s expectations for the fundamentals and/or valuation of the sector* relative to the group’s historic fundamentals and/or valuation: Overweight : The analyst’s expectation for the sector’s fundamentals and/or valuation is favorable over the next 12 months. Market Weight : The analyst’s expectation for the sector’s fundamentals and/or valuation is neutral over the next 12 months. Underweight : The analyst’s expectation for the sector’s fundamentals and/or valuation is cautious over the next 12 months. *An analyst’s coverage sector consists of all companies covered by the analyst within the relevant sector. An analyst may cover multiple sectors.

Credit Suisse's distribution of stock ratings (and banking clients) is:

Global Ratings Distribution

Rating Versus universe (%) Of which banking clients (%) Outperform/Buy* 45% (64% banking clients) Neutral/Hold* 39% (61% banking clients) Underperform/Sell* 14% (53% banking clients) Restricted 2% *For purposes of the NYSE and FINRA ratings distribution disclosure requirements, our stock ratings of Outperform, Neutral, and Underperform most closely correspond to Buy, Hold, and Sell, respectively; however, the meanings are not the same, as our stock ratings are determined on a relative basis. (Please refer to definitions above.) An investor's decision to buy or sell a security should be based on investment objectives, current holdin gs, and other individual factors.

Important Global Disclosures Credit Suisse’s research reports are made available to clients through our proprietary research portal on CS PLUS. Credit Suisse research products may also be made available through third-party vendors or alternate electronic means as a convenience. Certain research products are only made available through CS PLUS. The services provided by Credit Suisse’s analysts to clients may depend on a specific client’s preferences regarding the frequency and manner of receiving communications, the client’s risk profile and investment, the size and scope of the overall client relationship with the Firm, as well as legal and regulatory constraints. To access all of Credit Suisse’s research that you are entitled to receive in the most timely manner, please contact your sales representative or go to https://plus.credit-suisse.com . Credit Suisse’s policy is to update research reports as it deems appropriate, based on developments with the subject company, the sector or the market that may have a material impact on the research views or opinions stated herein. Credit Suisse's policy is only to publish investment research that is impartial, independent, clear, fair and not misleading. For more detail please refer to Credit Suisse's Policies for Managing Conflicts of Interest in connection with Investment Research: https://www.credit-suisse.com/sites/disclaimers-ib/en/managing-conflicts.html . Credit Suisse does not provide any tax advice. Any statement herein regarding any US federal tax is not intended or written to be used, and cannot be used, by any taxpayer for the purposes of avoiding any penalties. See the Companies Mentioned section for full company names The subject company (SBRY.L, DIXY.MM, UPS.N, GOOGL.OQ, AAPL.OQ, MKS.L, 000120.KS, HON.N, 023530.KS, 139480.KS, 4755.T, AMZN.OQ, BABA.N, CARR.PA, CHDRAUIB.MX, JD.OQ, LNTAq.L, MGNTq.L, PCAR4.SA, PJPq.L, SORIANAB.MX, TSCO.L, WALMEX.MX, WES.AX, WMT.N, WOW.AX) currently is, or was during the 12-month period preceding the date of distribution of this report, a client of Credit Suisse. Credit Suisse provided investment banking services to the subject company (GOOGL.OQ, AAPL.OQ, BABA.N, CARR.PA, PJPq.L, WALMEX.MX, WMT.N) within the past 12 months. Credit Suisse provided non-investment banking services to the subject company (SBRY.L, GOOGL.OQ, AAPL.OQ, HON.N, 4755.T, BABA.N, CARR.PA, PJPq.L, TSCO.L, WALMEX.MX, WMT.N) within the past 12 months Credit Suisse has managed or co-managed a public offering of securities for the subject company (GOOGL.OQ) within the past 12 months. Credit Suisse has received investment banking related compensation from the subject company (GOOGL.OQ, AAPL.OQ, BABA.N, CARR.PA, PJPq.L, WALMEX.MX, WMT.N) within the past 12 months Credit Suisse expects to receive or intends to seek investment banking related compensation from the subject company (SBRY.L, DIXY.MM, UPS.N, GOOGL.OQ, AAPL.OQ, MKS.L, 8267.T, 000120.KS, COST.OQ, HON.N, 023530.KS, 139480.KS, 4755.T, AD.AS, AMZN.OQ, BABA.N, CARR.PA, CHDRAUIB.MX, JD.OQ, LNTAq.L, MGNTq.L, PCAR4.SA, PJPq.L, SORIANAB.MX, TSCO.L, WALMEX.MX, WES.AX, WMT.N, WOW.AX) within the next 3 months. Credit Suisse has received compensation for products and services other than investment banking services from the subject company (SBRY.L, GOOGL.OQ, AAPL.OQ, HON.N, 4755.T, BABA.N, CARR.PA, PJPq.L, TSCO.L, WALMEX.MX, WMT.N) within the past 12 months As of the date of this report, Credit Suisse makes a market in the following subject companies (AAPL.OQ).

https://plus.credit-suisse.com/

https://www.credit-suisse.com/sites/disclaimers-ib/en/managing-conflicts.html

https://www.credit-suisse.com/sites/disclaimers-ib/en/managing-conflicts.html

10 April 2017


A member of the Credit Suisse Group is party to an agreement with, or may have provided services set out in sections A and B of Annex I of Directive 2014/65/EU of the European Parliament and Council ("MiFID Services") to, the subject issuer (DIXY.MM, JE.L, UPS.N, GOOGL.OQ, 8267.T, 000120.KS, 023530.KS, 139480.KS, 4755.T, AMZN.OQ, BABA.N, CARR.PA, CHDRAUIB.MX, JD.OQ, LNTAq.L, MRW.L, MTS.AX, OCDO.L, PCAR4.SA, PJPq.L, SORIANAB.MX, TSCO.L, WALMEX.MX, WES.AX, WMT.N, WOW.AX) within the past 12 months. As of the end of the preceding month, Credit Suisse beneficially own 1% or more of a class of common equity securities of (MRW.L, PCAR4.SA). Credit Suisse beneficially holds >0.5% long position of the total issued share capital of the subject company (GOOGL.OQ, 000120.KS, 023530.KS, 139480.KS). Credit Suisse has a material conflict of interest with the subject company (PCAR4.SA) . Credit Suisse or its controlled entities, controlling entities, or entities under common control hold directly or indirectly a relevant participation in the capital stock of the subject company/companies.[PCAR4]. For purposes of this report, a relevant participation means a participation of 5% or more in a type or class of shares of the capital stock of a company. Credit Suisse or one of its affiliates is acting as an intermediary in a public offering of securities issued by Brasil Foods or referenced in assets or receivables of Companhia Brasileira de Distribuicao.

For other important disclosures concerning companies featured in this report, including price charts, please visit the website at https://rave.credit-suisse.com/disclosures or call +1 (877) 291-2683. For date and time of production, dissemination and history of recommendation for the subject company(ies) featured in this report, disseminated within the past 12 months, please refer to the link: https://rave.credit-suisse.com/disclosures/view/report?i=290826&v=-kh79bhqxmbd3npy5i3naxy6n .

Important Regional Disclosures Singapore recipients should contact Credit Suisse AG, Singapore Branch for any matters arising from this research report. The analyst(s) involved in the preparation of this report may participate in events hosted by the subject company, including site visits. Credit Suisse does not accept or permit analysts to accept payment or reimbursement for travel expenses associated with these events. Restrictions on certain Canadian securities are indicated by the following abbreviations: NVS--Non-Voting shares; RVS--Restricted Voting Shares; SVS--Subordinate Voting Shares. Individuals receiving this report from a Canadian investment dealer that is not affiliated with Credit Suisse should be advised that this report may not contain regulatory disclosures the non-affiliated Canadian investment dealer would be required to make if this were its own report. For Credit Suisse Securities (Canada), Inc.'s policies and procedures regarding the dissemination of equity research, please visit https://www.credit-suisse.com/sites/disclaimers-ib/en/canada-research-policy.html. Credit Suisse Securities (Europe) Limited (Credit Suisse) acts as broker to (MRW.L). The following disclosed European company/ies have estimates that comply with IFRS: (SBRY.L, MKS.L, AD.AS, CARR.PA, MGNTq.L, MRW.L, PJPq.L, TSCO.L). As of the end of the preceding month, the subject company (CARR.PA) beneficially owned 5% or more of the total issued share capital of Credit Suisse Group. Credit Suisse has acted as lead manager or syndicate member in a public offering of securities for the subject company (GOOGL.OQ, AAPL.OQ, BABA.N, LNTAq.L, WALMEX.MX, WMT.N) within the past 3 years. Principal is not guaranteed in the case of equities because equity prices are variable. Commission is the commission rate or the amount agreed with a customer when setting up an account or at any time after that. I, Tobias Stingelin, CFA, certify that (1) The views expressed in this report solely and exclusively reflect my personal opinions and have been prepared independently, including with respect to Banco de Investimentos Credit Suisse (Brasil) S.A. or its affiliates ("Credit Suisse"). (2) Part of my compensation is based on various factors, including the total revenues of Credit Suisse, but no part of my compensation has been, is, or will be related to the specific recommendations or views expressed in this report. In addition, Credit Suisse declares that: Credit Suisse has provided, and/or may in the future provide investment banking, brokerage, asset management, commercial banking and other financial services to the subject company/companies or its affiliates, for which they have received or may receive customary fees and commissions, and which constituted or may constitute relevant financial or commercial interests in relation to the subject company/companies or the subject securities. This research report is authored by: Credit Suisse Securities (Japan) Limited .................................................................................................................................... Keiichi Yoneshima Credit Suisse (Hong Kong) Limited .......................................................................................................................................................... Evan Zhou Casa de Bolsa Credit Suisse (Mexico), S.A ........................................................................................................................ Antonio Gonzalez, CFA Credit Suisse Securities (USA) LLC ......................................................................................................... Stephen Ju ; Edward J. Kelly, CFA, CPA Credit Suisse Securities (Europe) Limited, Seoul Branch ................................................................................................................ A-Hyung Cho Bank Credit Suisse (Moscow) .................................................................................................................................................... Olga Bystrova, CFA Banco de Investments Credit Suisse (Brasil) SA or its affiliates. ....................................................................................... Tobias Stingelin, CFA Credit Suisse International ................................................................................................................. Stewart McGuire, CFA ; Andre Kukhnin, CFA Credit Suisse Equities (Australia) Limited .......................................................................................................................................... Grant Saligari Credit Suisse Securities (Europe) Limited ....................................................................................................................................... Victoria Petrova To the extent this is a report authored in whole or in part by a non-U.S. analyst and is made available in the U.S., the following are important disclosures regarding any non-U.S. analyst contributors: The non-U.S. research analysts listed below (if any) are not registered/qualified as research analysts with FINRA. The non-U.S. research analysts listed below may not be associated persons of CSSU and therefore may not be subject to the FINRA 2241 and NYSE Rule 472 restrictions on communications with a subject company, public appearances and trading securities held by a research analyst account. Credit Suisse Securities (Japan) Limited .................................................................................................................................... Keiichi Yoneshima Credit Suisse (Hong Kong) Limited .......................................................................................................................................................... Evan Zhou Casa de Bolsa Credit Suisse (Mexico), S.A ........................................................................................................................ Antonio Gonzalez, CFA Credit Suisse Securities (Europe) Limited, Seoul Branch ................................................................................................................ A-Hyung Cho Bank Credit Suisse (Moscow) .................................................................................................................................................... Olga Bystrova, CFA

https://rave.credit-suisse.com/disclosures


https://rave.credit-suisse.com/disclosures/view/report?i=290826&v=-kh79bhqxmbd3npy5i3naxy6n

https://rave.credit-suisse.com/disclosures/view/report?i=290826&v=-kh79bhqxmbd3npy5i3naxy6n

10 April 2017


Banco de Investments Credit Suisse (Brasil) SA or its affiliates. ....................................................................................... Tobias Stingelin, CFA Credit Suisse International ................................................................................................................. Stewart McGuire, CFA ; Andre Kukhnin, CFA Credit Suisse Equities (Australia) Limited .......................................................................................................................................... Grant Saligari Credit Suisse Securities (Europe) Limited ....................................................................................................................................... Victoria Petrova

Important disclosures regarding companies or other issuers that are the subject of this report are available on Credit Suisse’s disclosure website at https://rave.credit-suisse.com/disclosures or by calling +1 (877) 291-2683.


10 April 2017


This report is produced by subsidiaries and affiliates of Credit Suisse operating under its Global Markets Division. For more information on our structure, please use the following link: https://www.credit-suisse.com/who-we-are This report may contain material that is not directed to, or intended for distribution to or use by, any person or entity who is a citizen or resident of or located in any locality, state, country or other jurisdiction where such distribution, publication, availability or use would be contrary to law or regulation or which would subject Credit Suisse or its affiliates ("CS") to any registration or licensing requirement within such jurisdiction. All material presented in this report, unless specifically indicated otherwise, is under copyright to CS. None of the material, nor its content, nor any copy of it, may be altered in any way, transmitted to, copied or distributed to any other party, without the prior express written permission of CS. All trademarks, service marks and logos used in this report are trademarks or service marks or registered trademarks or service marks of CS or its affiliates.The information, tools and material presented in this report are provided to you for information purposes only and are not to be used or considered as an offer or the solicitation of an offer to sell or to buy or subscribe for securities or other financial instruments. CS may not have taken any steps to ensure that the securities referred to in this report are suitable for any particular investor. CS will not treat recipients of this report as its customers by virtue of their receiving this report. The investments and services contained or referred to in this report may not be suitable for you and it is recommended that you consult an independent investment advisor if you are in doubt about such investments or investment services. Nothing in this report constitutes investment, legal, accounting or tax advice, or a representation that any investment or strategy is suitable or appropriate to your individual circumstances, or otherwise constitutes a personal recommendation to you. CS does not advise on the tax consequences of investments and you are advised to contact an independent tax adviser. Please note in particular that the bases and levels of taxation may change. Information and opinions presented in this report have been obtained or derived from sources believed by CS to be reliable, but CS makes no representation as to their accuracy or completeness. CS accepts no liability for loss arising from the use of the material presented in this report, except that this exclusion of liability does not apply to the extent that such liability arises under specific statutes or regulations applicable to CS. This report is not to be relied upon in substitution for the exercise of independent judgment. CS may have issued, and may in the future issue, other communications that are inconsistent with, and reach different conclusions from, the information presented in this report. Those communications reflect the different assumptions, views and analytical methods of the analysts who prepared them and CS is under no obligation to ensure that such other communications are brought to the attention of any recipient of this report. Some investments referred to in this report will be offered solely by a single entity and in the case of some investments solely by CS, or an associate of CS or CS may be the only market maker in such investments. Past performance should not be taken as an indication or guarantee of future performance, and no representation or warranty, express or implied, is made regarding future performance. Information, opinions and estimates contained in this report reflect a judgment at its original date of publication by CS and are subject to change without notice. The price, value of and income from any of the securities or financial instruments mentioned in this report can fall as well as rise. The value of securities and financial instruments is subject to exchange rate fluctuation that may have a positive or adverse effect on the price or income of such securities or financial instruments. Investors in securities such as ADR's, the values of which are influenced by currency volatility, effectively assume this risk. Structured securities are complex instruments, typically involve a high degree of risk and are intended for sale only to sophisticated investors who are capable of understanding and assuming the risks involved. The market value of any structured security may be affected by changes in economic, financial and political factors (including, but not limited to, spot and forward interest and exchange rates), time to maturity, market conditions and volatility, and the credit quality of any issuer or reference issuer. Any investor interested in purchasing a structured product should conduct their own investigation and analysis of the product and consult with their own professional advisers as to the risks involved in making such a purchase. Some investments discussed in this report may have a high level of volatility. High volatility investments may experience sudden and large falls in their value causing losses when that investment is realised. Those losses may equal your original investment. Indeed, in the case of some investments the potential losses may exceed the amount of initial investment and, in such circumstances, you may be required to pay more money to support those losses. Income yields from investments may fluctuate and, in consequence, initial capital paid to make the investment may be used as part of that income yield. Some investments may not be readily realisable and it may be difficult to sell or realise those investments, similarly it may prove difficult for you to obtain reliable information about the value, or risks, to which such an investment is exposed. This report may provide the addresses of, or contain hyperlinks to, websites. Except to the extent to which the report refers to website material of CS, CS has not reviewed any such site and takes no responsibility for the content contained therein. Such address or hyperlink (including addresses or hyperlinks to CS's own website material) is provided solely for your convenience and information and the content of any such website does not in any way form part of this document. Accessing such website or following such link through this report or CS's website shall be at your own risk.

This report is issued and distributed in European Union (except Switzerland): by Credit Suisse Securities (Europe) Limited, One Cabot Square, London E14 4QJ, England, which is authorised by the Prudential Regulation Authority and regulated by the Financial Conduct Authority and the Prudential Regulation Authority. Germany: Credit Suisse Securities (Europe) Limited Niederlassung Frankfurt am Main regulated by the Bundesanstalt fuer Finanzdienstleistungsaufsicht ("BaFin"). United States and Canada: Credit Suisse Securities (USA) LLC; Switzerland: Credit Suisse AG; Brazil: Banco de Investimentos Credit Suisse (Brasil) S.A or its affiliates; Mexico: Banco Credit Suisse (México), S.A. (transactions related to the securities mentioned in this report will only be effected in compliance with applicable regulation); Japan: by Credit Suisse Securities (Japan) Limited, Financial Instruments Firm, Director-General of Kanto Local Finance Bureau ( Kinsho) No. 66, a member of Japan Securities Dealers Association, The Financial Futures Association of Japan, Japan Investment Advisers Association, Type II Financial Instruments Firms Association; Hong Kong: Credit Suisse (Hong Kong) Limited; Australia: Credit Suisse Equities (Australia) Limited; Thailand: Credit Suisse Securities (Thailand) Limited, regulated by the Office of the Securities and Exchange Commission, Thailand, having registered address at 990 Abdulrahim Place, 27th Floor, Unit 2701, Rama IV Road, Silom, Bangrak, Bangkok10500, Thailand, Tel. +66 2614 6000; Malaysia: Credit Suisse Securities (Malaysia) Sdn Bhd; Singapore: Credit Suisse AG, Singapore Branch; India: Credit Suisse Securities (India) Private Limited (CIN no.U67120MH1996PTC104392) regulated by the Securities and Exchange Board of India as Research Analyst (registration no. INH 000001030) and as Stock Broker (registration no. INB230970637; INF230970637; INB010970631; INF010970631), having registered address at 9th Floor, Ceejay House, Dr.A.B. Road, Worli, Mumbai - 18, India, T- +91-22 6777 3777; South Korea: Credit Suisse Securities (Europe) Limited,

Seoul Branch; Taiwan: Credit Suisse AG Taipei Securities Branch; Indonesia: PT Credit Suisse Securities Indonesia; Philippines:Credit Suisse Securities (Philippines ) Inc., and elsewhere in the world by the relevant authorised affiliate of the above. Additional Regional Disclaimers Hong Kong: Credit Suisse (Hong Kong) Limited ("CSHK") is licensed and regulated by the Securities and Futures Commission of Hong Kong under the laws of Hong Kong, which differ from Australian laws. CSHKL does not hold an Australian financial services licence (AFSL) and is exempt from the requirement to hold an AFSL under the Corporations Act 2001 (the Act) under Class Order 03/1103 published by the ASIC in respect of financial services provided to Australian wholesale clients (within the meaning of section 761G of the Act). Research on Taiwanese securities produced by Credit Suisse AG, Taipei Securities Branch has been prepared by a registered Senior Business Person. Australia (to the extent services are offered in Australia): Credit Suisse Securities (Europe) Limited (“CSSEL”) and Credit Suisse International (“CSI”) are authorised by the Prudential Regulation Authority and regulated by the Financial Conduct Authority (“FCA”) and the Prudential Regulation Authority under UK laws, which differ from Australian Laws. CSSEL and CSI do not hold an Australian Financial Services Licence (“AFSL”) and are exempt from the requirement to hold an AFSL under the Corporations Act (Cth) 2001 (“Corporations Act”) under Class Order 03/1099 published by the Australian Securities and Investments Commission (“ASIC”), in respect of the financial services provided to Australian wholesale clients (within the meaning of section 761G of the Corporations Act). This material is not for distribution to retail clients and is directed exclusively at Credit Suisse's professional clients and eligible counterparties as defined by the FCA, and wholesale clients as defined under section 761G of the Corporations Act. Credit Suisse (Hong Kong) Limited (“CSHK”) is licensed and regulated by the Securities and Futures Commission of Hong Kong under the laws of Hong Kong, which differ from Australian laws. CSHKL does not hold an AFSL and is exempt from the requirement to hold an AFSL under the Corporations Act under Class Order 03/1103 published by the ASIC in respect of financial services provided to Australian wholesale clients (within the meaning of section 761G of the Corporations Act). Credit Suisse Securities (USA) LLC (CSSU) and Credit Suisse Asset Management LLC (CSAM LLC) are licensed and regulated by the Securities Exchange Commission of the United States under the laws of the United States, which differ from Australian laws. CSSU and CSAM LLC do not hold an AFSL and is exempt from the requirement to hold an AFSL under the Corporations Act under Class Order 03/1100 published by the ASIC in respect of financial services provided to Australian wholesale clients (within the meaning of section 761G of the Corporations Act). Malaysia: Research provided to residents of Malaysia is authorised by the Head of Research for Credit Suisse Securities (Malaysia) Sdn Bhd, to whom they should direct any queries on +603 2723 2020. Singapore: This report has been prepared and issued for distribution in Singapore to institutional investors, accredited investors and expert investors (each as defined under the Financial Advisers Regulations) only, and is also distributed by Credit Suisse AG, Singapore Branch to overseas investors (as defined under the Financial Advisers Regulations). Credit Suisse AG, Singapore Branch may distribute reports produced by its foreign entities or affiliates pursuant to an arrangement under Regulation 32C of the Financial Advisers Regulations. Singapore recipients should contact Credit Suisse AG, Singapore Branch at +65-6212-2000 for matters arising from, or in connection with, this report. By virtue of your status as an institutional investor, accredited investor, expert investor or overseas investor, Credit Suisse AG, Singapore Branch is exempted from complying with certain compliance requirements under the Financial Advisers Act, Chapter 110 of Singapore (the “FAA”), the Financial Advisers Regulations and the relevant Notices and Guidelines issued thereunder, in respect of any financial advisory service which Credit Suisse AG, Singapore Branch may provide to you. UAE: This information is being distributed by Credit Suisse AG (DIFC Branch), duly licensed and regulated by the Dubai Financial Services Authority (“DFSA”). Related financial services or products are only made available to Professional Clients or Market Counterparties, as defined by the DFSA, and are not intended for any other persons. Credit Suisse AG (DIFC Branch) is located on Level 9 East, The Gate Building, DIFC, Dubai, United Arab Emirates. EU: This report has been produced by subsidiaries and affiliates of Credit Suisse operating under its Global Markets Division In jurisdictions where CS is not already registered or licensed to trade in securities, transactions will only be effected in accordance with applicable securities legislation, which will vary from jurisdiction to jurisdiction and may require that the trade be made in accordance with applicable exemptions from registration or licensing requirements. Non-US customers wishing to effect a transaction should contact a CS entity in their local jurisdiction unless governing law permits otherwise. US customers wishing to effect a transaction should do so only by contacting a representative at Credit Suisse Securities (USA) LLC in the US. Please note that this research was originally prepared and issued by CS for distribution to their market professional and institutional investor customers. Recipients who are not market professional or institutional investor customers of CS should seek the advice of their independent financial advisor prior to taking any investment decision based on this report or for any necessary explanation of its contents. This research may relate to investments or services of a person outside of the UK or to other matters which are not authorised by the Prudential Regulation Authority and regulated by the Financial Conduct Authority and the Prudential Regulation Authority or in respect of which the protections of the Prudential Regulation Authority and Financial Conduct Authority for private customers and/or the UK compensation scheme may not be available, and further details as to where this may be the case are available upon request in respect of this report. CS may provide various services to US municipal entities or obligated persons ("municipalities"), including suggesting individual transactions or trades and entering into such transactions. Any services CS provides to municipalities are not viewed as "advice" within the meaning of Section 975 of the Dodd-Frank Wall Street Reform and Consumer Protection Act. CS is providing any such services and related information solely on an arm's length basis and not as an advisor or fiduciary to the municipality. In connection with the provision of the any such services, there is no agreement, direct or indirect, between any municipality (including the officials,management, employees or agents thereof) and CS for CS to provide advice to the municipality. Municipalities should consult with their financial, accounting and legal advisors regarding any such services provided by CS. In addition, CS is not acting for direct or indirect compensation to solicit the municipality on behalf of an unaffiliated broker, dealer, municipal securities dealer, municipal advisor, or investment adviser for the purpose of obtaining or retaining an engagement by the municipality for or in connection with Municipal Financial Products, the issuance of municipal securities, or of an investment adviser to provide investment advisory services to or on behalf of the municipality. If this report is being distributed by a financial institution other than Credit Suisse AG, or its affiliates, that financial institution is solely responsible for distribution. Clients of that institution should contact that institution to effect a transaction in the securities mentioned in this report or require further information. This report does not constitute investment advice by Credit Suisse to the clients of the distributing financial institution, and neither Credit Suisse AG, its affiliates, and their respective officers, directors and employees accept any liability whatsoever for any direct or consequential loss arising from their use of this report or its content. Principal is not guaranteed. Commission is the commission rate or the amount agreed with a customer when setting up an account or at any time after that. Copyright © 2017 CREDIT SUISSE AG and/or its affiliates. All rights reserved.

Investment principal on bonds can be eroded depending on sale price or market price. In addition, there are bonds on which investment principal can be eroded due to changes in redemption amounts. Care is required when investing in such instruments. When you purchase non-listed Japanese fixed income securities (Japanese government bonds, Japanese municipal bonds, Japanese government guaranteed bonds, Japanese corporate bonds) from CS as a seller, you will be requested to pay the purchase price only.

https://www.credit-suisse.com/who-we-are

china; $41bn (2015) china; $178bn (2020)

Documents