Description

The typical store building size is ~25k square meters, moderately larger than a WMT Superstore or Costco (due to more space devoted to fresh foods) and significantly larger than Carrefour or Tesco. Koreans appreciate the broad aisles, high ceilings, and open space due to the cramped nature of other retail outlets such as traditional markets or neighborhood supermarkets.

 

The sales split is roughly:

-          55% fresh food

-          23% beverages (milk, water, alcohol, soda) and packaged foods

-          22% “general merchandise”: sundries + hardlines + softlines. Color from IR is that sundries like detergent and toothpaste are the biggest portion of this bucket.

 

As such, the large majority of Emart gross profit derives from “supermarket”-type goods.

For reference, 57% of COST’s sales come from the “supermarket” lines of packaged foods, sundries including cleaning supplies, and fresh foods. COST includes health & beauty within hardlines, where it’s the #3 contributor of 5 named categories, and also includes gas within the total sales. So on adjusted basis, COST’s “supermarket-type” revenues are in the low 60%s, and food is a little under 40%.

PSMT, the Latin American membership hypermarket, has a closer sales split to Emart – for PSMT, food (54%) + sundries including health/beauty (26%) are 80% of sales.

WMT US “supermarket” category is 67% of revenues (Grocery including sundries/packaged food is 56% + Health & Wellness of 11%).

TGT “supermarket” is 47% of revenues, with Food/pet revenues within this at 21%.

Korean meals are very dependent on fresh foods and side dishes. Thus, compared to many U.S. super/hypermarkets, Emart has a higher sales mix of fresh food compared to packaged food – 55% of Emart sales are perishables (a bit higher if you include milk) vs. fuel-adj 56% at Ingles Markets, 51% at Sprouts, 51% at Village Super Market, 42% at Albertsons, 28% at Kroger, 15% at Costco.

Whole Foods and The Fresh Market are outliers, with 65-67% of revenues being perishables. Emart is below these levels due to more sales of sundries and general merchandise.

At >90% “supermarket” revenues and >80% (mostly fresh) food/beverage revenues, and with high perishables as % of sales, I view Emart brick+mortar as mostly insulated from online pressures.

In the past, Emart’s sales mix was heavier on general merchandise, but the general merchandise sales have declined over time due to online competition; meanwhile, fresh food sales have grown nicely as Emart has solidified its reputation for “best quality at a good price” through its industry-leading direct procurement and quality control, enabled by economies of scale.

 

In addition, Emart’s private-label food sales are growing quickly due to rapid consumer embrace of two new house brands:

-          Peacock: high-end, high-quality ready-to-eat frozen/refrigerated dishes. Mostly Korean foods and soups, these dishes are well-regarded for their taste and low amounts of preservatives/excess calories. Developed at Emart’s large food R&D center.

-          No Brand: Emart’s answer to COST’s Kirkland Signature – Emart signs long-term, high-volume purchasing agreements with large foreign generics manufacturers after doing in-depth due-diligence on quality, safety, etc. No Brand products are 35-85% cheaper than equivalent branded SKUs, and average a ~65% discount based on UBS analysis of 3 items (my own in-store analysis of 7 items came up with ~69% average discount - UBS is in the right ballpark).

 

Emart is the best one-stop-shop for Koreans’ frequent grocery shopping: quality, pricing, and breadth of selection are all superior to what you can find at a neighborhood supermarket, traditional market, or convenience store.

Admittedly, general merchandise will continue to be a headwind to comps but I am optimistic about Emart due to the continued growth in private-label, with sales +10% YoY in Q1 ’16 and set to accelerate, as well as continued market-share gains in food/beverage vs. other hypermarkets based on Emart’s higher produce quality, superior/broader/cheaper private-label selection, and overall lower prices.

There is now a sizable gap in terms of sales per store, per square meter, and overall sales and gross profits between Emart and the two next-largest competitors, and I believe this gap will continue growing as the gross profit leverage on fixed costs gives Emart a larger budget to invest in pricing, sourcing, and R&D.

There is still significant runway for Emart to take market share in brick+mortar: Emart sales are 11t KRW/year while total Korean hypermarket sales are 49t KRW/year, and the less-advantaged formats of neighborhood supermarkets and traditional food markets combine for 56t KRW/year in revenues.

Emart is owned by E-Mart Inc (139480 KS, $15m daily average value traded), a publicly-traded holdco spun off from the Shinsegae Co (004170 KS) department store holdco just over 5 years ago. Today, Emart-brand hypermarkets generate slightly over 100% of parent-level OP and well over 100% of consolidated OP.

However, I believe that various non-Emart-brand parts of Emart Inc are extremely valuable, such that Emart Inc stock could appreciate by 3x through the end of 2019, to 500k KRW, even if OP from the flagship Emart brick+mortar hypermarket declines from Q1 ’16 run-rate levels and receives a conservative multiple.

Below is my summary sum-of-the-parts analysis, using my valuation of the equity slice of the non-parent subsidiaries.

Keep in mind that Emart Inc’s market cap today is 4.6t KRW, and there is 3.5t of net debt at the holdco (parent) level which contains Emart, online, and Traders.

 

Emart Inc valuation	Dec 2019	Q1 2016
Emart hypermarkets	7,526	8,614
Online	7,418	1,634
Traders discount warehouses	3,751	853
Mall equity stakes (Shinsegae Property)	1,996	711
Starbucks Korea 50% stake	1,614	1,313
Supermarkets	1,177	110
WithMe convenience store	683	88
Non-core stakes	809	647
Value of hotels, tax-adj net	-	913
Value of Samsung Life, tax-adj	-	882
Shinsegae Duty Free	-	140
Emart China	-	(30)
Gas stations/MRO	13	13
Gross value SOTP	24,986	15,888
Net debt at parent level	(3,524)	(3,466)
Net value SOTP	21,462	12,422
Per share	769	446
Upside vs. today, zero NAV discount	4.7x	2.7x
Appropriate NAV discount	35%	40%
Modeled stock Px	500	268
Upside vs. today	3.1x	1.6x
NAV discount on today's stock Px	79%	63%

 

Summary of the Emart Inc businesses and how I value them:

 

Emart hypermarkets

The legacy core segment is comping flattish this year, though volatile from month-to-month. OP in Q1 was down 1% YoY, though up significantly on a sequential basis. There have been a high number of recent new openings – store base increased by 4% last year – but newbuilds will slow to a trickle going forward.

I think Emart hypermarkets deserve a similar multiple to country-leading mature grocery-heavy hyper-and super-market chains like Carrefour, Walmart US (carved out Int’l/Mexico), DIA, and Tesco. I tried to carve out the brick-and-mortar operations only for these comps, and calculated that the run-rate TEV/NOPAT multiple is ~15x for all except Tesco, which trades at 21x due to expectations of continued OP bounceback.

Emart has its own “OP bounceback” potential. Short-term, in Q2 ’15 and the beginning of Q3 ’15, there was hugely depressed sales and OP due to the MERS virus scare in Korea. Thus I use seasonally-adjusted run-rate Q1 ’16 NOPAT for the hypermarkets and apply a 15x multiple.

Long-term, there is a good chance the government will loosen or remove harsh operating-hour and mandatory-closing-day restrictions that were implemented in phases during 2012-2014.

First, the government slapped on operating-hour restrictions on all hypermarkets in 2012 in an attempt to protect neighborhood supermarkets and traditional markets. Stores that used to be open from 8am-midnight are now mostly open from 10am-10pm.

Second, in 2012 the government also forced all hypermarkets to close their doors two days per month. Initially, many hypermarket managers closed during weekdays, which have lower traffic and less sales volume than weekends. However, the government mandated weekend closings that were essentially phased in from 2012-2014.

Both these additional regulations caused significant OP and same-store-sales pullback, with comps of -4% in 2012, -5% in 2013, and -2% in 2014. OP was down more over this period due to operating leverage.

While some of the comp headwind was from a decline in general merchandise sales due to online competition in these lines, management thinks a majority of the headwind came directly from the loss of 2 high-volume days/month plus the reduction in the morning and nighttime shopping hours. (Koreans are historically relatively heavy nighttime shoppers.)

During this time and also since, Emart has continued to take market share from its smaller, less-profitable hypermarket rivals Lotte Mart (owned by Lotte Shopping, 023530 KS) and Homeplus (formerly owned by Tesco, now owned by MBK Partners private equity).

While these rivals are smaller and thus lack the procurement network, purchasing power, private-label R&D capabilities, and brand name of Emart, they are still material in size and so Emart brick+mortar can grow or at least keep sales flat for several more years simply by taking share from these two declining players. For context, Lotte Mart’s brick+mortar sales are a little over half of Emart+Traders, and currently this segment of Lotte Shopping has negative EBIT; EBITDA is down 20% YoY and at or slightly below the maintenance CapEx level.

The other main competitor, Homeplus, is doing even worse – EBITDA was down 11% YoY in FY Feb ‘15 then took another steep ~70% drop in FY Feb ‘16 (full numbers not out yet – this is based on parent-only financial statements). Recently, both competitors seem to be focusing on profit margins rather than market share or revenue, which bodes well for Emart’s short-term sales and profitability. Homeplus is larger than Lotte Mart, and only ~20% smaller than Emart in terms of top-line revenue.

For my Dec ’19 OP estimates, I do have hypermarket OP falling from a seasonally-adj run-rate of 758b KRW in Q1 to 662b KRW in FY 2019.

The main cause for my relative pessimism is the continued rapid growth of online retail in Korea, which grew in the mid/high teens in 2015 but is slowing down a bit as the three “social commerce” online players run low on cash. Nonetheless, online cannibalization will maintain pressure on Emart’s general merchandise sales. The continued growth of convenience stores will pressure packaged foods and beverages. In addition, the continued rapid growth of Emart’s own leading online-grocery offering will cannibalize offline grocery to an extent.

On the other hand, Emart is enjoying recent success with opening store-within-stores for electronics and home furnishings, and expanding the food courts with new concepts like Vietnamese food.

There are still a lot of inferior competitors with large grocery sales – traditional markets, Homeplus, neighborhood grocery stores, and Lotte Mart – so Emart brick+mortar will take market share within older shoppers age 40+. From my own experience, as well as several acquaintances’, Emart’s produce/meat quality and private-label portfolio are vastly superior to these competitors and should increasingly drive preference toward Emart as the private-label portfolio rapidly adds SKUs and continues lowering prices as private-label volumes increase.

Gross and operating margins will also enjoy a tailwind due to the increased private label penetration as a percentage of sales. Today, private-label is still a relatively low ~13% of total sales – COST is 25% and rising - and I think Emart can follow the recent WMT playbook of driving materially increased same-customer loyalty through private-brand innovation, price, quality, and SKU increases.

-          WMT is currently increasing private-label SKUs worldwide. According to the WMT Chief Merchandising Officer: “customers are asking more and more for more private brand. It’s a key loyalty driver and those customers who participate in private brands make more trips and spend more…it is building a more-engaged relationship and it’s a deeper penetration of how they shop at WMT and why they shop at WMT…private brands are increasing both in terms of SKU count and new items. We believe winning with private brands comes down to 3 things: price, innovation, and quality. And we’re spending a lot of effort to invest in product development, sourcing, technology, and new talent to be able to build this business.” (Mar ’16)

o   Emart is following this exact same strategy as well. Partly due to strengthening their private brand lineup over the last year, WMT was able to accelerate ex-fuel ex-online US comps from flattish a year ago to up ~2.5% YoY in Q1.

o   If I understood the transcript correctly, WMT says that the loyal, private-brand centric customers spend a whopping 50% of their WMT spend on the private labels. Clearly, consumers in other wealthy countries see a compelling value in mid/high-quality private brands and from speaking to acquaintances, I believe Korean consumers do as well.

o   “No Brand,” the game-changing 65%+ cheaper quality private brand, was launched in 2015 and will have several hundred SKUs by year-end.

 

In addition, Emart has huge latent real-estate value, owning 94% of its land sites and 89% of its buildings at the Emart+Traders segments. This includes 30-40 prime redevelopment sites - out of 151 total stores and 142 owned land parcels - worth several trillion KRW in aggregate, e.g. a large super-prime store connected to a subway station in the booming neighborhood of Haeundae in Busan.

All these factors plus the optionality on the inevitable loosening of the operating hours/days regulations should protect from further Emart multiple compression, and thus I leave my NOPAT multiple at 15x for the hypermarkets in 2019 (as I assume that the regulations will likely be lifted in ~2020, during the term of the next Korean president).

Online

Over time online should become the second real plank in the Emart story. Today, online is growing quickly at ~27% annually, but still losing money. The business model is in transition from a structurally negative-margin “Instacart model” to structurally positive-margin “Ocado model.”

In the “Instacart model,” Emart’s own workers and contracted deliverers pick items in the store, store them on-site, and make deliveries in small batches in a small truck. Most of the time, there is so much order backlog that same-day delivery is not available. In addition, the SKU availability is quite small and there is no real-time inventory sync.

As land is expensive in Korea, the “Instacart model” in-store storage facilities are small and thus quite tight and inefficient. Labor costs, especially around minimum wage, are rising very quickly in Korea these days and so picking items from the shelves is very expensive.

In the U.S., Instacart texts you through the app to ask about substitutions when the specific ordered SKU is out of stock at the grocery store. In Korea, one of the world’s “busiest” cultures, this would be unacceptable and thus the current from-store delivery model is constrained to a subset of only the most basic SKUs that are almost always available in every store.

Yet, unlike Instacart which essentially charges a markup to (non-Whole Foods) grocery food prices to cover the high labor and delivery costs, Emart and the two main competitors keep prices roughly the same online as in-store. All three players are losing money at online grocery due to this, and so they all strictly ration the delivery capacity and control the number of pickers and deliverers in order to prevent losses from exploding.

Thus, while all 3 players promise “same-day” store-based delivery, it is not actually available - and availability is uncertain until you check - for the majority of shopping attempts. In addition, the on-site order storage at most hypermarkets has now reached capacity, and it would be folly to invest CapEx to expand the in-store storage coolers at a negative ROI.

Emart built the first online grocery distribution center (DC) in Seoul, and completed it in the second half of 2014. After taking over a year to work out kinks, learn how best to automate picking and delivery, and ramp volumes, Emart completed a second, much larger and more automated, center in Gimpo near Seoul.

I call this the “Ocado model” as it is similar to Ocado in the UK, except in a more-favorable market environment where the majority of the population lives in an ultra-dense setting in a mid/high-rise apartment with a security office on the ground floor of each building that can receive parcels or a grocery box and store it for several hours.

In addition, Ocado must compete with attractive click-and-collect offerings from brick & mortar rivals, while in Korea click-and-collect is not attractive due to parking constraints, a much lower automotive mode-share for commuting, a lower auto ownership percentage, and a stronger cultural preference for “do-it-for-me” service models in general (high preference to receive delivery vs. click-and-collect in any vertical, not just grocery).

Emart management also says that Emart’s 2^nd DC is more-automated than Ocado’s DCs (which makes sense as it’s newer and more expensive).

All the Emart hypermarket advantages are also being brought to bear in the online segment: lower-cost sourcing, fresher produce, better private-label portfolio, and some brand synergies with former parent Shinsegae (you can order Emart groceries through the Shinsegae app).

From discussing with friends and colleagues, I feel that Emart online should take rapid market share in the Korean grocery space, with consistent availability of same-day delivery a near-term catalyst. According to Emart tests, conversion rates go through the roof when a same-day slot is guaranteed available.

With the much higher daily delivery capacity at acceptably positive margins of the DC-based Ocado model compared to the Instacart model, Emart offers same-day delivery slots to customers shopping in areas nearby the DCs if they order before 3pm. Eventually they hope to push the potential order time a bit later, to 4-5pm as economics of density (orders per hour per square kilometer) improve.

Due to the limitations of the Instacart model, only a small subset of the total 70k Emart SKUs (at the large stores; 40k at smaller ones) are available from store-based ordering. As more DCs are constructed, online shoppers can once again enjoy the full in-store experience – the 2^nd DC has ~35k SKUs and will ramp up to ~50k shortly.

In the long term, online grocery delivery should be a “winner-take-most” market due to:

1) delivery economics of density - the more orders/km2 you have, the lower your average fulfillment cost because the average fulfillment time and distance per delivery goes down.

2) Virtuous cycle of revenue growth à leverage on fixed costs at the automated DCs à increased gross profit à budget to use for innovation in private-brand, pricing investments, speeding up delivery, tech R&D, etc.

3) Korea is very much an “app-ified” society, and app users compare prices significantly less than web users. Emart should be the go-to grocery app and can fulfill all the grocery needs of Korean families, ordering multiple times a week (orders above ~$30 have free delivery).

4) Increased overall volumes (brick + mortar and online combined) will give Emart even more sourcing power and even more brand power.

Emart’s brand power is real; in a recent UBS survey Emart received by far the best quality scores out of any Korean retailer in the survey. Emart’s marks for app quality and pricing were bad though – I believe these are outdated perceptions based on Emart’s former app UI/UX and higher online pricing from 2012-2014. In 2015 and 2016, Emart made huge investments in these areas as capacity increased and variable costs were removed from the business. As time passes, consumers will come to appreciate this.

Additionally, the synergies with former parent co Shinsegae are significant. The SSG (Shinsegae) app is a Top 25 shopping app in Korea, and SSG.com one of the largest web shopping portals. Shinsegae digital properties host not only the legacy department store’s wares but also Emart and Traders digital experiences.

Emart’s separate apps also rank well above other hypermarkets’, with both the O2O app and the online grocery app ranking around #20 in Shopping category downloads. By comparison, Lotte Mart has 2 apps ranked in the 40s and one in the ~60s, while Homeplus’ sole app is ranked ~#40 and laggard GS Supermarket is ranked ~#80. Recent subscale upstarts like Hansalim have basically failed, with app downloads trailing off sharply post the February launch to currently ranked ~#250 in Shopping.

The agglomerated SSG app and website allow grocery shoppers to experience a “one-stop shop” where high-quality branded luxury goods, cosmetics, apparel, electronics, homeware/bedding, etc. can be purchased from Shinsegae within the same environment as Emart groceries and sundries.

Many consumers today order products like bottled water or packaged foods along with household items like pillows or doormats from online eBay-type marketplaces like Gmarket, 11^th St, Auction or quasi-marketplaces like Coupang, Tmon, and Wemap. Some of these users will switch to SSG’s consolidated digital platform - including Emart for grocery items - for a better-curated, more-trustworthy, faster-delivery, more-branded experience. (Think about how Tmall has far outgrown Taobao for the last several years.)

The fact that SSG can also offer delivery of a wide variety of high-quality fresh and frozen food and perishables, a key advantage over the marketplace/quasi-marketplace competitors, should allow the SSG app to become a true one-stop retail platform with a large, sticky userbase giving SSG a high wallet share, similar to Tmall’s status as a high-quality, branded one-stop shop in China or Amazon’s similar status in the U.S. and Europe (and both have now added fresh groceries in several cities).

As online takes more and more wallet share in Korea and basket sizes (items purchased per day) grow, consumers will comparison shop less and less on a per-item basis, and will be more and more loyal to one or two digital properties they trust to provide the best overall price, selection, delivery speed, reviews, and quality. Users will compare digital stores over time, but on more of a basket basis rather than item-by-item. (Especially since Google, and thus Google Shopping, market share in Korea is low and Naver Shopping is not as good or as popular.) SSG as well as Emart will be secular beneficiaries of this trend toward quality digital superstores.

Emart’s public guidance for online as of March 2016 was 5t KRW in sales in 2023. This timeline was accelerated to 2020 last month at a UBS conference. Emart believes that, with a total of 6 DCs by 2020, it could service a good majority of the Korean population (probably 2 more in the Seoul area plus one in Busan and one in Daegu).

Q1 runrate online sales were almost 0.8t KRW, so the new guidance represents a massive 53% sales growth CAGR. Given the expected acceleration and hypergrowth, valuation is not easy; also, until the acceleration actually happens, the market will be skeptical. Today I value online at 2.1x sales – Ocado trades for 1.4x sales today and is growing a lot slower than Emart online’s current growth rate, and back when Ocado’s growth profile looked somewhat better in mid-2015 it traded at 2.5x sales.

In 2019, online should be solidly profitable, though with a low OP margin at that time due to continued growth OpEx and costs related to the ramp-up of DCs. I think Emart online can do 2.6t KRW in sales in 2019, a 43% CAGR. In my model, OP margin on this is 4.7% or 124b KRW, based on IR color on the long-term cost structure with net costs of the Ocado model ~9% of revenues below the Instacart model.

I then apply an admittedly very-rich 2019 multiple of 2.8x sales or 79x trailing OP, though I believe this is justified since my model has OP growing 3x YoY at that time, and sales growing rapidly at 40%+, with runway left to take share in an addressable market of ~100t KRW annual “grocery” sales (my estimate, which Emart management agrees with).

My estimated 2019 market share of Korean grocery for Emart online is 2.6%, which should see further rapid expansion in the 2020s due to winner-take-all scale benefits such as high order density per neighborhood leading to lowest cost-to-serve, as well as the opening of dedicated online DCs in Korea’s smaller cities leading to the most-efficient ROI on marketing.

 

Traders discount warehouses

Emart launched Traders, a non-membership Costco-type warehouse, in late 2010. At 35% of sales, Traders’ perishable sales mix is much lower than Emart but higher than Costco. Only 3500 SKUs are present at Traders, and like Costco and Sam’s most are packaged in bulk. About half the SKUs rotate seasonally. Prices are much, much lower than elsewhere in Korea on a per-unit basis and there is a Costco-like “treasure-hunting” aspect with many unique, often imported products.

Currently there are 10 Traders locations, but due to extremely good same-store sales performance  - +13% so far this year on top of +8% for 2015 and +7% for 2014 – Emart Inc plans to materially accelerate the buildout of the Traders store footprint.

As Traders sales per ft2 are materially above Emart’s, despite lower GPM% the EBIT ROI is significantly higher than a newbuild Emart store. In addition, the screamingly-high comps have led to rapid OP margin expansion, to 2.7% in Q1 ’16 vs. 1.8% a year ago despite growing the store count by one store (over 10% footprint growth) during that time.

As Traders continues to scale, Emart plans to open more overseas procurement and quality offices, and Traders will also accrue more and more purchasing power and brand reputation. In addition, Traders benefits from Emart’s market power in terms of purchasing Korean produce, meat, packaged foods, CPGs like toiletries, etc. Traders also has a thriving private-label business where profitability will benefit as the store footprint, and thus private-brand volume, grows.

I give Traders a high 39x run-rate NOPAT multiple on today’s run-rate NOPAT, due to the rapid same-store OP growth with even more margin expansion guided for the short term, plus the opportunity to double store count over the next few years and continue growing rapidly thereafter, with EBITDA ROICs in the mid-teens for new stores. At Dec ’19, my model gives Traders 29x trailing NOPAT, similar to today’s NOPAT multiples for mature but growing comps like COST (30x), Walmex (29x), and PSMT (~27x).

 

Mall equity stakes

Emart, through their 90% subsidiary Shinsegae Property, owns a 51% stake in a mall project in an eastern suburb of Seoul, Hanam. Shinsegae Property’s partner in this venture is Taubman (TCO), who has made extensive remarks about the project, Starfield Hanam, in their earnings calls and conferences. The mall will open in two months, ahead of schedule and under-budget. It’s expected that, after ~2 years of post-open ramp-up, the levered AFFO ROI will be roughly 7.5%, steadily growing after that.

I anticipate that Shinsegae Property and Taubman will announce a 2^nd mall project in the coming months, after the (likely) successful opening of Starfied Hanam. Starfield Hanam is expected to open at least 95% leased and 80% occupied and “hopefully” above these levels (according to TCO), with a significant mix of luxury tenants. According to Bloomberg, a TSLA showroom will likely be among the tenants at Starfield Hanam.

While far from an optimal use of cash – at today’s stock price levels, a share buyback would be the most accretive use, or else accelerating online and Traders CapEx even more – recycling FCF through building malls in partnership with a good lessee/operator in TCO is surely preferable to paying down net debt. This is because Emart is actually materially underlevered and still has huge dry powder to increase net debt if it so desires.

Currently, Emart Inc’s short-term bonds trade at a 1.3% YTM, while 4-year fixed-rate debt yields 1.5% on the market, significantly lower than the interest expense paid out on today’s income statement. Net debt is a conservative 22% of my estimated gross SOTP value.

The bonds that Emart Inc issues from time to time are unsecured and non-amortizing. The only true covenants are a 4x max liabilities:equity ratio (0.9x today) and a limitation on disposing more than 50% of the assets on the balance sheet.

In addition, Emart plans to open Emarts and Traders at the Taubman malls. The Emart and Traders stores are expected to act like U.S. anchors back in the 1970s and 1980s and drive a significant amount of traffic to the overall mall. (This anchor effect is working as expected at a few other non-owned malls, e.g. Kintex in Ilsan.) This will allow Emart Inc to have an equity stake, at cost, in the value-creation that their anchors drive, in place of getting discounted rent like U.S. anchors did in leases that began in the 1960s-80s.

The plan to amass a collection of 3+ mall stakes over the years and then unlock value somehow – by selling these to a mall company, or maybe injecting into a pureplay property company or REIT structure – has been floated by Emart management. As the value-unlocking event is likely not within the horizon of my model, I value the mall ventures today at cost, and in Dec ’19 at 1.35x cost.

 

Starbucks Korea stake

Mr. Chung Yong-jin, the scion of the Shinsegae Lee/Chung family, is currently the de-facto CEO of Emart Inc. (His official title is vice-chairman.) In his late 20s, not long after returning from college in the U.S., he negotiated a deal for a 50/50 JV with Starbucks to form Starbucks Korea. You may say “that was a no-brainer” but remember, at that time (1997-98) Korea was in the depths of the IMF crisis so capital was scarce and consumers had closed their wallets. In addition, Starbucks’ market cap then was around ~$3b; it wasn’t nearly the global household name it is today.

The Starbucks operating agreement is favorable to Starbucks Korea. Starbucks Korea pays Starbucks HQ 5% of sales as a royalty. In return, Starbucks Korea has a perpetual brand and technology license, as well as the ability to procure ingredients and supplies from Starbucks HQ at favorable prices. Starbucks has similar agreements with over 20 licensees worldwide, at an average 6% rate, so Starbucks Korea is slightly advantaged. In addition, as an early mover in the Korean market Starbucks captured “coffee shop” and “3^rd place” mindshare among Koreans aged 20-45.

Thus, Starbucks Korea is growing extremely quickly – it grew revenue at 25% and EBITDA at 23% in 2015 while generating slight cash, on top of 28% revenue and EBITDA growth in 2014 supported by moderate cash burn from fast store openings.

Starbucks parent is growing top-line by ~11% and EBITDA by 13-14% constant-currency, and trades for 17-18x trailing EBITDA. I believe Starbucks Korea will still be growing a bit faster than this in 2019, so I give it a 19.5x trailing EBITDA multiple in my model. For today’s value, I give Starbucks Korea 25x trailing EBITDA, which is arguably conservative.

 

Supermarkets

Today, Emart owns a neighborhood supermarket chain that’s running at slightly below OP breakeven. Its economics have been hurt by the operating hours/days restrictions, and it is also suffering some share loss to convenience stores.

However, as Emart’s private-label SKU portfolio is expanding from several hundred to several thousand from 2015 to 2017, Emart will be able to successfully launch a super-cheap “Aldi clone” that sells a highly-curated assortment of around a thousand mid-quality private label goods at very low prices, centered on its No Brand brand. Emart could then mix in some select high-end, high-quality, mid-priced SKUs from its Peacock range. (Aldi U.S. has ~1400 SKUs per store.)

The Aldi-clone concept will be named No Brand Supermarket and has been rumored in the local press since last year. I expect Emart to open the first few locations in the next few months.

I see a need for such a supermarket chain in Korea, where the shift to convenience stores is partly driven by the high prices and lack of compelling SKUs in the neighborhood supermarkets, as well as the lack of brands, quality control, and pleasant shopping environment in the traditional markets. (Convenient, more-walkable locations are also a factor in the share shift to convenience stores.)

With enough compelling No Brand SKUs priced at a wide ~50% discount to supermarkets’ SKUs, and even cheaper compared to convenience stores, I think an “Aldi clone” Emart supermarket could draw substantial shopper traffic even from outside a 2-minute walking radius, thus effectively competing against the convenience stores that are currently taking share.

Korean cities are dense enough that drawing shoppers from a 7-minute walking radius often generates sufficient store traffic to earn good ROICs at a neighborhood supermarket. The pitch vs. a convenience store would be “walk 2 minutes further, save 50%, and get better food freshness and selection.”

Assuming the Emart team can execute on the rollout, SKU creation/selection, and brand-building process, the Supermarket subsidiaries’ value can expand through incremental CapEx as well as a multiple expansion from my 1.0x book today to 1.25x book in 2019.

 

WithMe convenience stores

WithMe is an upstart convenience store franchise that offers those franchisee operator/owners with mid-to-high levels of sales significantly higher gross margins and take-home pay (dividends) compared to one of the other 4 major Korean convenience store chains.

The WithMe business model is based mostly on a fixed monthly franchisee fee per-store; in addition, Emart earns a low markup on selling all the inventory into the stores. The other 4 Korean convenience store chains have historically charged a relatively fixed percentage on total store sales. Thus, for outperforming stores with good locations or good franchisees with high levels of traffic and volume, it makes economic sense to move to WithMe when the existing franchise contract with another player expires (the contracts generally last 5 years).

Another reason franchisees should prefer WithMe over the other four chains is the much stronger product assortment offered by Emart, enabled by synergies with the hypermarket/online procurement operations. In addition, Emart’s private-label brands such as the hit ultra-cheap No Brand and gourmet ready-to-eat Peacock - as well as others such as E Everyday Value – create end-consumer preference, which drives traffic, revenue and better profitability and ROIC for the franchisee. This effect should accelerate as the number of private-label SKUs jumps rapidly over the next few quarters.

Talks with WithMe franchisees in Korea have confirmed these presumptions and due to these advantages, WithMe has doubled store count from 617 locations a year ago to 1207 as of Q1 (vs. over 30,000 convenience stores nationwide).

While WithMe is losing money today due to being subscale and the store-rollout expenses, I expect WithMe to generate nice profitability in the future, especially as the private-label % of sales continues to increase through the introduction of more No Brand and Peacock SKUs and as franchisees who opened WithMe stores over the last year and a half tell their friends and family about the superiority of the WithMe model to alternative convenience store chains.

Today, WithMe has only ~1/8 the stores of the 3 large convenience store players in Korea, but Emart has devoted a large budget for branding, product creation, new-franchisee training, etc. that befits a much larger-scale chain than WithMe’s current 1,207-store base. Emart management is confident in the WithMe store-opening pipeline for the next several months. I expect this segment to go from a drain on consolidated OP to OP-breakeven sometime in 2017, due to expected positive operating leverage.

The largest player in the space, BGF Retail, is expected to earn 212b KRW of OP (despite significant growth expenses) this year on a store base of almost 10,000. As detailed in the “risks” section, the entire convenience store space is growing quickly on both same-store and store-count levels. There is ample runway for WithMe to increase its store base by several times over the next 3.5 years.

 

Non-core stakes

Emart Inc owns non-core stakes in several “Shinsegae children” subsidiaries. The largest of these is Shinsegae Food (031440 KS), with a publicly-traded stock valuing Emart’s stake at ~350b KRW. Shinsegae Food manufactures a portion of the Peacock SKUs - though the product creation is done in-house at Emart – as well as several other high-quality processed-food products. Shinsegae Food also recently purchased all 105 Smoothie King locations in Korea, and they say they will open many locations within Emart food courts.

The Shinsegae children are all publicly-traded, thus easy to value, but I don’t believe they will be disposed of unless Emart stock trades at an even higher discount to NAV than the current ~63% discount. Management views the relationships between Emart and the Shinsegae children as strategic, and anyway the market value is relatively small in the scheme of things (around 400b KRW total).

Emart Inc also owns a 3.3% stake in Costco’s Korean opco, a relic of a time when Costco wanted to enter Korea in the mid-‘90s but Emart (then Shinsegae) controlled many of the sites that Costco sought. As part of a long-term lease deal, Emart was able to extract a small amount of equity in the Costco Korean opco, which based on Korean annual filings I believe is worth a little north of 100b KRW.

It has been publicly reported that 3 Costco leases expire in 2018, at which point it’s expected these extremely high sales per ft2 stores will be converted into Traders locations. After this happens, the likelihood of Emart disposing its Costco Korea holding also increases significantly.

Emart also owns various miscellaneous non-core holdings that generate zero or negative earnings, such as a new Emart store in Vietnam, a medium-sized lakeside golf resort, a new home shopping channel/online property that is growing rapidly, and a digital payments company. I give these miscellaneous holdings less than 100b KRW of value today, but there is substantial optionality in a few of them (notably the home-shopping channel, Shinsegae TV Shopping).

 

Hotels

Emart owns and operates two extremely valuable Westin Chosun hotels through its subsidiary Shinsegae Chosun Hotel. These hotels generate minimal OP, and zero to slightly negative net income today. However, the property sites themselves are extremely valuable as they are considered “trophy properties” in Korea.

The most valuable hotel, the Westin Chosun Busan, is located on the corner of Haeundae Beach in Busan. The Haeundae area is widely considered the most prime and expensive in Busan, both due to the popular beach as well as the permissive development rules that allow construction of supertall skyscrapers. There are currently multiple supertalls under development in short walking distance from the hotel.

In Haeundae, car traffic congestion is now becoming a major issue; the Westin Chosun Busan is nestled between two public parks and has a wide 4-lane exclusive access road. It is one of only five buildings in Haeundae with direct beach access (plus the under-development Haeundae LCT the Sharp) and the others are all on the opposite end of the beach, which has worse traffic and is further from central Busan, highway access, and other amenities.

The other hotel, the Westin Chosun Seoul, is located steps from Seoul Plaza near City Hall Station (Lines 1 and 2) in downtown Seoul. The valuation of this hotel is extremely difficult to calculate due to uncertainty around how much buildable square footage would be permitted in the event of redevelopment. (It would likely depend on the mayor at the time of redevelopment.) Nonetheless, the site itself is inarguably super-prime.

The hotel is located a 1-2 blocks from major corporate headquarters such as Lotte Group (Lotte Korean HQ), Shinhan Bank, Shinhan Card, OCI, and Hanjin Group, as well as government buildings like the Bank of Korea and Seoul City Hall. It is also within short walking distance of other large headquarters such as CJ Korea Express, Samsung Fire & Marine, SK Telecom, Samsung Securities, Samsung Card, Namyang Dairy, and Dongkuk Steel. All these are within a ~500m radius of the hotel.

With a relatively large land base compared to other area buildings (including some existing skyscrapers), the potential for a super-prime skyscraper clearly exists, and it’s just a matter of whether the Seoul government would approve such a development.

Based on discussions with a Korean commercial real estate broker who discussed comps - though there are limited comps with such A++ locations - I believe the pretax value of these hotels is well in excess of 1t KRW based on potential auction value. At 1.35t KRW gross fair value, the net proceeds to equity after paying disposal tax on the capital gain would be slightly under 1t KRW.

 

Samsung Life

Emart inherited significant Samsung Life stock when it was spun off from Shinsegae 5 years ago. Samsung Life IPO’d in May 2015 and at that time Emart decreased its stake by 20%. As there are no synergies with Emart and management has little interest in the life insurance business, I expect Emart to continue selling the Samsung Life stake over time. The pretax value of the Samsung Life stock today is 1.1t KRW.

 

Shinsegae Duty Free

Shinsegae Duty Free is worth relatively little and has generated significant negative OP, especially since signing an uneconomically expensive lease at Busan Gimhae Airport. That lease ended in Q2 ’16 and I expect OP to go back up to zero soon, and management expects positive OP in the coming months after some long-planned tenant upgrades.

I strongly believe the long-term plan for Shinsegae Duty Free is a disposal to the Shinsegae group’s Shinsegae Department Store subsidiary. Shinsegae Department Store recently was awarded a coveted in-city duty-free operator license and will build a duty free store located in its iconic flagship department store in central Seoul.

Thus, Emart’s subscale “Shinsegae Duty Free” business fits much better with the Shinsegae group’s business model, which already deals with many luxury brands through its department store segment. I expect some kind of disposal or asset swap to unlock roughly 0.14t KRW of value from this segment, or a relatively low 0.4x sales ex Gimhae Airport stores (comp Shilla trades for 1.0x), and only a small premium over the inventory value.

 

Emart China

Similar to Shinsegae Duty Free, Emart China’s negative OP has been a thorn in the side of Emart Inc’s consolidated income statement for the past several years. Again, these losses are caused by uneconomic long-term leases. Emart has been paying lease-breakage penalties to exit some of these leases early, and waiting until others expire. Management expects Emart China losses to go from a material 35b KRW in 2015 – it was even higher previously – to a minimal level in 2017 before all losses totally end the next year as Emart completely exits China.

 

Net debt

I have modeled the cashflow walk based on the following assumptions:

-          Maintenance CapEx for each business line similar to disclosed in the Korean quarterly filings, which are slightly higher estimates than given to the sellside.

o   Notably, even using the maintenance CapEx numbers from the filings – which I believe include the cost of building out some store-in-stores, admittedly to keep comps stable/growing slowly – maintenance CapEx at the hypermarkets is still significantly lower than D&A, supporting the use of a 15x NOPAT multiple as this would be closer to 13x maint FCF.

-          Interest expense going down moderately over time, given the much cheaper market YTMs on Emart’s bonds compared to the existing effective interest rate on the run-rate income statement.

-          Disposal of both hotels, Samsung Life, and Samsung Duty Free within the model period. Even if not all these disposals happen, the debt load will still be very manageable when looked at with an interest coverage perspective or based on the covenants.

o   As of Q1, the parent entity’s EBITDA interest coverage is 11x. As stated earlier, Emart is extremely far from the 4x liabilities:equity covenant, and even disposing of the hotels + Samsung Life + Samsung Duty Free would only be shedding 14% of today’s assets (and a lower % of 2018 total B/S assets) vs. the max restriction of disposing 50% of total assets.

-          Growth CapEx of roughly 100b KRW per new store (Emart or Traders), and 350b KRW per average new online DC.

o   In addition, I earmark some CapEx for online hubs or “spoke centers” that will be built out to improve delivery efficiency and variable economics.

-          I use the all-in Korean corporate income tax rate of 24.2%.

 

 

Downside protection

-          High alignment with upper management: a high % of their net worth is in Emart stock.

o   Controlling family – son Chung Yong-Jin is de facto CEO, and his mother is the chairwoman - owns 28% directly, not through multiple trees of holdcos like many other Korean families.

o   Also unlike other Korean chaebols, only 2 family members (mother and son) have stakes after a recent stock swap between the Mr. Chung and his sister.

-          Low multiple on today’s earnings. Trades at 9x seasonally-adj run-rate Q1 parent earnings, and ~9.5x consolidated.

o   (Parent = Emart, Traders, small amount of dividends from subsidiaries, and NI-negative Online. Consolidated multiple higher due to higher interest expense and negative OP at several non-parent segments including Duty Free and China.)

-          Huge NAV discount today – 63% NAV discount on my calculations. And the operating cos are not melting ice cubes in aggregate.

-          Emart Inc owns land worth well over the enterprise value, in addition to buildings on that land worth trillions more KRW depending on the buyer.

-          Tesco sold Homeplus to MBK partners in October 2015 for 9.0t KRW TEV (based on Tesco’s Powerpoint). This is a huge multiple compared to Homeplus EBIT of 483b KRW in FY Feb ’15. Based on Korean filings, Homeplus had 414b KRW of D&A for FY Feb ’15. So EBITDA was ~0.9t KRW and the EV/EBITDA multiple was 10x.

o   However, filings since have suggested that Homeplus’ EBITDA has declined precipitously over the past year, with the main entity’s EBITDA declining ~70% YoY in FY Feb ’16 (other entities are yet to report, though). Given the deal was agreed-upon in Sept ’15, the true run-rate EBITDA multiple was higher than 10x.

o   Emart’s parent (Emart + Traders + online) EBITDA was 1.1t KRW at Q1 run-rate. 10x would be 11t KRW and doesn’t even include the near-5t KRW of value I ascribe to the other pieces of the company. And Emart’s same-store sales are much better than Homeplus, with online a larger % of the mix as well.

o   The biggest reason MBK Partners was willing to pay such a rich valuation for Homeplus: Many Homeplus stores are ripe for redevelopment into large apartment or office towers (just like Emart). Homeplus owns ~64% of its own stores. Emart owns 94% of the sites of its stores.

o   Similarly, Lotte Shopping trades for 10x trailing EBITDA. (Their EBITDA is mostly from department stores, which are at even more of a threat from online than hypermarkets, though they also derive material EBITDA from Lotte Mart.)

 

Biggest risks

-          Emart online is more cannibalistic to Emart brick+mortar hypermarket than I expect.

 

o   According to Emart, a large majority of Emart online shoppers are not located in an Emart catchment zone – e.g. the large, wealthy Seoul Gangnam area only has one very small Emart - and these consumers cannot easily go to Emart stores, but many have a strong preference for Emart’s house brands and produce/meat quality. However, as Emart online expands, there may be more cannibalization than Emart brick + mortar can offset through sheer market-share gains.

 

-          Convenience-store market share gains vs. hypermarkets may be larger and last longer than I expect.

 

o   Convenience stores are taking share in the Korean retail space, especially the grocery segment. Korean analysts see this trend as driven by a rapid increase in “1-person households.” Traditionally, Koreans in their 20s and 30s lived with their parents until marrying. However, increasingly young Koreans are moving out of their parents’ apartments before getting married and living on their own in studios or apartments. Usually these complexes have a convenience store on the ground floor, or at least one nearby. This drives young Koreans to make both impulse and daily-meal food purchases at the convenience stores.

 

§  Convenience store prices on a per-unit basis are higher than hypermarkets like Emart, and much higher than discount warehouses like Traders. However, convenience store packaging is smaller and thus excess food does not have to be stored or wasted.

 

§  Convenience stores are walk-in/walk-out and usually located a short walking distance from offices, residences, or transit hubs. Hypermarket traffic is mostly car-oriented. The popular narrative is that many Koreans in their 20s and 30s are now living auto-free or auto-light, and thus hypermarkets will continue to lose share.

 

·         However, the data implies that as this cohort ages, they do end up acquiring automobiles. E.g. within the Seoul city limits, active personal-use cars steadily increased from 2.29m in Jan ’11 to 2.46m in May ’16 despite Seoul city population actually declining during that time period (due to death of elderly and outflow to suburbs and the new Sejong City).

 

§  In addition, convenience stores often have a lack of freshness, a lack of produce, a lack of health-oriented SKUs (organics, natural foods) and a low number of SKUs in general. But despite these relative weaknesses, traffic is growing nicely. According to the Korean gov’t agency MOTIE, cumulative convenience store traffic has been growing at 6-7% annually over the last 3 years, and this trend is sustaining with BGF Retail seeing ex-cigarette comps of +8% in Q1 ’16 (+10% in FY ’15) and GS Retail seeing +6-7% comps the last few months.

 

§  Due to the good same-store comps, good store-level economics, and high number of recently retired 50- and 60-somethings looking for a second career (as a convenience store franchisee), every convenience store chain is opening new stores quickly. So in addition to same-store sales growth of ~7% right now, the total convenience store base is also increasing by a whopping 14% YoY as of Q1 vs. store-open tenor of ~7% from 2012-2014. The massive secular growth, along with some large cigarette tax increases, has led gross sales at convenience stores to skyrocket from 8.7t KRW in 2011 to ~16.5t KRW in 2015.

 

o   Emart does have a natural hedge on this trend through its WithMe 100%-owned subsidiary. If WithMe can become one of the top 3 Korean convenience store brands in the long term, it would offset some or all of the negative impact of the convenience store format’s grocery share gains. I would highlight that the top 2 Korean convenience store chains have TEVs of 6t KRW and 5t KRW, not too far from Emart’s TEV.

 

§  I believe WithMe has a good chance to become a Top 3 chain in 6-7 years due to its superior private-label SKUs and its franchise terms which are designed specifically to attract the best owner/operators that generate outsized revenues per store (since the franchisee net margin on a WithMe high-revenue store is significantly higher than the franchisee margin with other chains).

 

o   My view is that young people will shift some of their food spend from convenience store purchases to online ordering as same-day delivery capacity ramps up. In addition, there is still a loyal core of families, with the main heads of household aged upper-30s and above, as well as a sizable minority of younger Koreans who have decided to settle down and marry.

 

§  Of my Korean acquaintances in their 20-30s, inevitably those married or with kids shop at hypermarkets (brick+mortar and/or online) while those not yet married buy most of their dairy and beverages at convenience stores and most of their food at restaurants or convenience stores.

 

o   Starting in ~4 years, the number of Koreans in their 20s will begin declining rapidly, as there was a steep drop in the Korean birth rate starting in ~2001. There was also a large drop in the Korean birth rate from 1981-1985, so the number of Koreans in their early 30s is just now starting to decline quickly. In general, as society ages, 1-person households will turn into 2-3-person households and there will be some preference shift back to hypermarkets, as doing “family shopping” for meals/meal ingredients for husband+wife+child or even boyfriend+girlfriend is a superior experience at hypermarkets/online hypermarkets compared to convenience stores.

 

o   There is some limitation on the store-openings runway from the 3 large existing players that currently control almost 90% of locations (BGF Retail, GS Retail, and 7-11). The government only allows one location per brand per 250m radius. This means that a lot of the secular-growth benefit will actually accrue to newcomers who are able to open stores near high-performing Big 3 stores and more-easily open new stores near existing transit hubs. There are currently only two bona fide “late movers” in this space, Ministop and WithMe.

 

 

o   The “shift to single-person households” is not happening all of a sudden. It’s quite a slow trend that started in ~2000 when 16% of households were single-person. As of 2015, 27% of households are single-person. The latest Korean gov’t estimate is that this ratio will increase to 34% in 2035 - the shift to single-person households will actually get less rapid going forward, with the curve on the graph starting to flatten this year, due to the birthrate declines that happened decades ago.

§  Even with the same “shift to single-person-household” tailwind, BGF Retail historically (pre-2015) had +3-4% comps. My view is that the Big 3 convenience store chains will see a deceleration back to those lower comp levels at some point, as same-day online delivery of both grocery and general merchandise begins to take more share.

 

-          The Korean government may come out strongly against loosening the operating hours/days regulations.

 

o   Currently, from talking to people in Korea and reading news articles, the overall societal sentiment is that these laws are inconvenient for the large majority of citizens and serve mostly to protect traditional markets and non-chain subscale grocers, entities which often evade taxes (especially income tax, and also VAT on non-fresh-food) due to their uncompetitive cost structures.

 

o   Nonetheless, the political winds in Korea have changed plenty of times in the past, and thus my assumption of 15x Dec ’19 NOPAT multiple for Emart hypermarket, partly on OP bounceback potential, may be too aggressive.

 

-          Despite recently accelerating the online growth guidance, growth rates in online may disappoint and be “stuck” around the 30% level due to any number of reasons (operational issues, Lotte Mart just opened its 1^st online DC and may discount aggressively to be able to claim online is a success, inability to secure land/sites for future DCs).

 

o   Even in a scenario where online growth fails to meet my model’s CAGR, I am still confident that online will grow quickly due to the high convenience level, attractive prices, and good SKU availability.

 

o   I expect Emart to give more details around their online growth assumptions/guidance in the coming months. This detail will likely include data on the conversion-rate and revisit-rate uplift experienced in the months following the opening of the Gimpo DC. This color would increase the credibility of the >50% CAGR crazy-high growth guidance.

 

-          Emart does not sell the non-core stakes (Samsung Life, hotels, Duty Free, Shinsegae-child stakes). This means the NAV discount will not narrow, or narrow less than the model’s projection of “only” a 35% NAV discount by Dec ’19.

 

o   Emart already sold a material amount of Samsung Life stock in 2015 after the IPO. I believe Emart will continue selling Samsung Life stock as long as they are doing CapEx fast enough on Online, Traders, and Malls. Management has no special attachment to Samsung Life.

 

o   Duty Free – this makes much more sense in the hands of Shinsegae, especially since Shinsegae recently won a new duty free license for central Seoul. However, Emart says they need to wait until the uneconomic leases in Busan expire and the expected new tenants are rotated in, to be able to sell Duty Free to Shinsegae at a fair price. This is supposed to happen fairly quickly, though.

 

o   Hotels – this holding is essentially land speculation, and the land has appreciated extremely quickly over the preceding decades of Shinsegae/Emart ownership. There is a material risk that management decides to keep one or both of these hotels for several more years due to anticipated continued land appreciation, despite knowing the hotels are non-core, non-synergistic assets that generate minimal OP or cashflow.

 

o   Shinsegae children and Costco Korea – I do not expect these to be liquidated in my model, though I think in real life there is a good chance that Costco Korea (3.3% stake worth ~110b KRW) will be disposed of.

§  Emart ownership of the Shinsegae children is a way for the mother and son to exercise control of these entities. Emart also has supplier relationships with the Shinsegae children, under the rationale that any long-term value creation will partly benefit Emart itself. Thus, while I think that the children may eventually be disposed of by management if the Emart stock price falls on dire straits, otherwise the children will be valued at a large discount in an Emart SOTP valuation. Fortunately, they are a very low percentage of total EV at Emart Inc.

 

-          Emart decides to delever some instead of rolling out new formats.

o   If the traffic and sales results from the initial “No Brand” Aldi-clone grocery stores are disappointing, Emart may go back to the drawing board and take several quarters to make tweaks and improvements, adding more and better SKUs, before continuing the rollout. This would put a slight damper on my model but I believe the investment case would still be intact, especially if some of the resulting excess dry powder is used toward other businesses instead of pure delevering.

 

-          Korea is an aging society, and absent more immigration the workforce will start shrinking pretty rapidly in ~2021.

 

o   Obviously this is not good for Korean consumption-related names in general, and also brings the potential for a 1990s Japan-style real-estate bust, where land prices (which are an option on redevelopment values) drop precipitously.

 

o   However, the Korean government and Korean companies are trying to battle working-age population shrinkage by taking aggressive measures to increase the female labor force participation rate, which is still relatively low at 52.9% (vs. 50.7% six years ago). Increased female labor participation benefits Emart for multiple reasons:

§  Emart is the market leader in terms of ready-to-eat meals. Emart’s hit Peacock dishes are easily boiled or microwaved and achieve taste that rivals true Korean home-cooking.

§  Traditionally, housewives who did not work would have time to shop at multiple stalls at multiple traditional markets, combined with trips to the neighborhood grocery store for dairy, bottled water, etc.

·         Those who prefer to stay with brick+mortar should find one-stop shops like Emart hypermarkets, with ready-made fresh Korean side dishes and baked goods, appealing as an alternative to traditional markets.

§  The increased female labor participation rate will be a tailwind for online grocery-shopping demand. Emart’s incremental market share in online grocery, which I expect at >50% long-term, will be much higher than its market share of existing Korean grocery (~13% incl Traders and Supermarket).

 

o   Land price sensitivity – while I agree a real-estate crash/extreme weakness is a risk, I do not think it causes the reward:risk of Emart Inc stock to be unacceptable.

§  First, the Korean government showed in 2013 that if real estate is weak for ~three quarters, it will institute significant policy loosening of regulations like max LTV, max DTI, zoning/redevelopment, real-estate capital gains tax, etc.

§  Second, the cushion between existing Emart TEV and current real-estate value is large enough that only dramatic real estate and economic weakness could wipe it out. During such a recession, Emart brick+mortar hypermarket may well take share from other retail channels given its superior price competitiveness vs. department stores, neighborhood supermarkets, and convenience stores (similar to WMT taking share in every U.S. recession).

 

Management

 

Emart is led by a strong CEO (official role “vice-chairman”), Mr. Chung Yong-jin. As related above, Mr. Chung negotiated a favorable JV with Starbucks HQ even in the depths of a bad Korean recession. While an executive at Shinsegae, which was dominated by department stores, he realized that the hypermarket format had more growth runway and better prospective long-term ROICs, so he was instrumental in building the Emart segment from scratch within Shinsegae; the rollout started in earnest in the mid-1990s. In fact, a Naver News Search reveals that local media criticized Mr. Chung in 1999 for “ignoring” the dept store segment within Shinsegae and putting all his focus on the hypermarkets.

 

Mr. Chung owns 10% of Emart shares, and his mother (the chairwoman, although relatively inactive at age 72) owns another 18%. A few months ago, Mr. Chung completed a share swap with his sister where he traded away his 7.3% stake in Shinsegae for his sister’s 2.5% stake in Emart. 

 

Mr. Chung returned to Korea from the United States in 1994; after spending a year at Fujitsu’s Korean subsidiary he entered Shinsegae as a middle manager in 1995. He was promoted to executive in Sept 1997. From Sept 1997 through the Emart spinoff, Shinsegae stock’s total return was a ~22-bagger. During this same time, the KOSPI Index was a ~3.5-bagger and even Samsung Electronics was “only” a 17-bagger (both total return).

 

Mr. Chung’s Instagram feed and Facebook feed feature pictures of Emart’s No Brand and Peacock R&D labs, new product creations, Mr. Chung’s sourcing trips to the U.S., Europe, and China, and Mr. Chung’s visits to overseas retail concepts such as Target’s and Whole Foods’ new stores, as well as his Facebook commentary on these topics for those who read Korean.

 

Compared to other retail CEOs in Korea, I feel that Mr. Chung is a clear cut above as an operator and visionary. His research-driven belief about the long-term optimal online structure and desire to move quickly away from the uneconomic Instacart online model give me confidence that Emart Inc. is not a melting ice cube.

 

Conclusion

 

When viewed on an SOTP basis, Emart Inc is extremely cheap given the underlying business potential of online and Traders, as well as the opportunity to monetize relatively low-profit assets such as Samsung Life, the hotels, and Shinsegae Duty Free.

 

I believe Emart Inc is on a path to aggressively increase its total Korean grocery market share from ~14% today to ~20% over the next 4 years or so. Most of this will come from online, which will have a profitable margin structure as it scales. Increases in square footage of Traders and No Brand concepts as well as positive same-store sales in Traders will also contribute to this, and offset flat to potentially negative real comps in Emart brick+mortar hypermarkets.

 

When viewed on a non-SOTP basis, I still feel that today’s 163,500 KRW share price reflects an attractive entry point of ~9.5x Q1 run-rate consolidated net income and slightly under 9x seasonally-adj run-rate parent net income. In my model, the multiple to consolidated net income for Dec ’19 rises to 19-20x, backed by solid earnings growth of 20-30%/year in the 2018-2019 timeframe.

 

I feel this implied Dec ’19 consolidated is reasonable given that high-multiple businesses such as Online, Traders, and the malls will be much higher slices of Emart Inc’s value at that time. In addition, the KOSPI today trades at ~14x stagnant earnings. The earnings growth is certainly worth 5-6 extra multiple points – the hard part is getting the growth itself.

 

 

The rich multiples are noted in both the high multiples of cited retail comps BGF Retail (24x NOPAT based on aggressive Q2E run-rate) and GS Retail (also 24x NOPAT on aggressive Q2E run-rate), as well as other large-cap Korean growers like LG H&H (26x run-rate NOPAT, growing at 17% top-line and 30% in OP).  

 

 

Note

 

For the multiples in the top section:

 

• FCF this year and next year is negative due to the Traders buildouts, online DC buildouts, and mall buildout. Maint FCF is moderately above IFRS earnings.

• EV/EBIT: using consolidated EV - higher than parent EV due to more debt - and consolidated EBIT. I view parent-only debt as more relevant though, given this is where the relevant covenants sit + a material % of the subsidiary debt is on the hotels, which can hold a high amount of debt despite minimal EBITDA (and I already factored the hotels' debt into the disposal proceeds calc).

• P/E - using consolidated earnings (lower than parent this year) because this is what the Korean Street focuses on. 

Catalyst

          Continued improvement in disclosure.

o   Emart recently began disclosing in-depth CapEx breakdowns, forward CapEx estimates, and detailed site-by-site data in the quarterly reports.

o   I am hoping for mall-REIT-type disclosures around the 1st mall after it opens, so that this high-multiple cashflow can be valued correctly.

-          Continued acceleration in releasing private-label SKUs over the next year, which will strengthen all the core businesses by saving consumers money at good quality points, and thus increasing consumer preference toward Emart’s retail channels.

o   Of note, No Brand cola and 3 other sodas were just introduced in June at a ~49% discount vs. branded colas. 5 weeks later, an Emart executive was quoted in the local press saying that sales and customer feedback are quite good so far.

o   No Brand sales tripled from 21b KRW in 2H ’15 to 64b KRW in 1H ’16 on the strength of more SKUs and higher “brand” recognition. Emart expects to continue rapid growth in private label, which is both GPM%-accretive and aids in customer loyalty and preference.

-          Improvement in capital allocation and capital structure.

o   In the model above, I assume Emart is extremely conservative with debt and keeps net debt levels flat despite materially increasing overall OP. If Emart is incrementally aggressive in terms of using very-cheap leverage, appreciation potential through 2019 is well over 3x due to the capability to buy back material amounts of stock and/or accelerate the buildout of Emart Online distribution centers and Traders locations. (Or even pay large dividends if so desired.)

-          Online revenue growth acceleration and transformation of the online segment from OP-negative to significantly OP-positive.

o   Acceleration in growth plus continued reiteration of the recent new “5t KRW sales in 2020” online-segment guidance would force the entire sellside (and also myself) to massively revamp valuation models.

-          Continued OP growth of Traders - from new-store openings, comps, and increase in private-label mix.

-          Net-income-negative segments are disposed of or stop losing money.

-          Successful initial openings of new concepts will prove to the investment community that Emart has multiple avenues to deploy cash at acceptable to good ROICs.

o   Starfield Hanam mall opening in two months.

o   No Brand supermarkets opening in a few months.

o   Emart store-within-stores (Electromart, The Life, new food courts) started opening over the last year in pilot quantities and a greater number will open over the next year.