ABR Holdings (ABR) was founded in 1979 to operate Swensen’s restaurants in Singapore. It now runs Swensen’s brands of restaurants, Tip Top Curry Puffs snacks and Season’s brands of bakery and café. The Swensen’s restaurants (25 outlets) and Tip Top Curry Puffs snacks (8 outlets) are operated in Singapore, while Season’s bakery (22 outlets) and café (6 outlets) are operated in Malaysia. ABR also manages four Swensen’s restaurants under its franchise model in Malaysia and is venturing to open corporate-owned Swensen’s restaurants there this year.
Swensen’s restaurants contributed 80% of FY15 revenue. Unfortunately, we don’t have the detail breakdown of the sales and profitability of each business. Here’s the geographical breakdown of the revenue.
|Rest of Asia||1,215||408||59|
Source: ABR’s annual reports
Divestment of The Cocoa Trees business and Restructuring
In early 2012, ABR sold its Cocoa Trees retail business for 100m cash and paid special dividend of 60.3m or 0.30 per share. The share price was 0.60 before the announcement of the sale. After the announcement, the share price rose sharply to 0.90 and fell back to 0.60 after paying the dividend of 0.30 per share. Existing shareholders reaped 50% return from the special dividend.
The Cocoa Trees contributed more than half of the Group revenue and was even more profitable than Swensen’s business. After selling The Cocoa Trees, ABR restructured its businesses by closing several non-performing businesses, including the loss-making business in China. As a result, ABR’s financial performance has a drastic change after FY12.
Tough Retail Market
During 2013 – 2015, ABR’s revenue was relatively flat, while the operating margin declined slightly. This is during the period in which Singapore retail market is facing a very challenging condition with rising rental cost and shortage of manpower supply as the Government tighten up the immigration policy. The margin in FY14 and FY15 would have gotten worse if without the government grant. ABR received government grant of 579k in FY14 and 997k in FY15.
Malaysia, on the other hand, experienced currency depreciation, rising rental and labor costs and implemented GST of 6% in April 2015. SGD-MYR exchange rate was at 2.5 at the start of 2013 and rose to 3.0 at the end of 2015. I believe part of the depreciation was due to Najib’s corruption scandal.
So, both countries are experiencing tough economic condition. They may take several more years to recover, and we can’t predict it. ABR’s performances in the past 3 years are considered reasonably good given such difficult retail market.
ABR is not staying put despite the difficult economy. Its Tip Top Curry Puff business opened three new outlets in 2015, increasing the revenue by 64% in that year. It opened one new outlet in January 2016 and plans to open another outlet at Sengkang Mall in Q4 of 2016. However, Tip Top Curry Puff should be just a small contributor to the revenue and profitability. So, I won’t be excited at this.
Its Swensen’s Malaysia opened one corporate-owned Swensen’s restaurant in KL in May 2015, just one month after the implementation of GST. It should be suffering losses in 2015 given the bad description of the business in the FY15 annual report. Swensen’s Malaysia plans to open three corporate-owned restaurants in Malaysia in Q3 of 2016. Two of the restaurants will be in KL, and one in JB.
ABR is already running four Swensen’s restaurants in Malaysia under its franchise model. Given this franchising experience, they should have some access to the business performance of the franchisees, and this information will be valuable to decide whether or not to open corporate-owned outlets in Malaysia to earn higher return. This new venture should contribute positively to FY16 revenue and profits, however, I don’t expect the contribution to be meaningful.
As for Swensen’s restaurants in Singapore and Season’s business in Malaysia, ABR’s focus is to consolidate the business and streamline the operation. Last year, it closed two Swensen’s restaurants and opened a new one at Sun Plaza. For Seasons, it renovated several outlets. These two businesses probably contributed more than 90% of the revenue and profits, but they are not expected to grow this year. Therefore, we can expect the overall business prospect to be quite muted.
The surprise news to me came 3 days after I bought the shares. ABR bought a property in Australia for A$20m and plans to redevelop it into residential houses for sale. Out of the A$20m paid, 1m was for commission. So high commission?? The Company already mentioned in 2013 that it planned to invest in property market but hadn’t found a good opportunity. I didn’t expect them to really do it. With 82m cash, ABR does have the resources to invest in another business. This property investment will take at least few years to yield any result. In the meantime, cash on hand will deplete substantially. Instead of investing outside their core business, I’d prefer the management to use the excess cash to buy back shares or just pay the dividends, thus reducing their overall capital.
On ABR’s balance sheet sit investment properties worth 3.5m, comprising of
- Freehold shop unit at City Plaza (size 25 sqm)
- Freehold shop unit at Far East Plaza (size 39 sqm)
- Leasehold HDB shop unit (85 years remaining lease) at Changi Village Road (size 358 sqm)
- Leasehold apartment unit in Ascott Towers Indonesia (size 159 sqm)
All these properties are rented out and generated rental income of 282k in FY15 while incurring the costs of 276k. Hence, the operating profit was a negligible 6k. That’s very poor return on the property rental. The costs of these properties were 5.6m with accumulated depreciation of nearly 2.1m.
What we should take note is that the fair value of these properties is 10.4m as stated in FY15 annual report. That’s nearly 3x the book value. Similar to cash, the income generated by this asset is next to nothing, but the value of this asset is high. These properties are not used in the core business of operating restaurants. If ABR is to sell them, they should generate one-off income of more than 9m.
Part of the PPE consists of:
- Freehold land with net book value of 366k (cost 420k)
- Building and structural improvements with net book value 6.9m (cost 12.4k)
- Leasehold property with net book value of 1.1m (cost 2.3m)
- Leasehold improvements with net book value of 3.7m (cost 13.6m)
This book value of 12m on PPE (total cost 28.7m) should refer to the following properties:
- Leasehold shop unit at Thomson Plaza (size of 349 sqm)
- A 4-storey factory building with a basement carpark in Tampines (size 9,780 sqm)
- A double storey factory building in JB (size 3,420 sqm)
- A 3-storey terrace shop in JB (size 178 sqm)
- Perhaps, other properties not listed in the annual reports
Some of these properties were acquired several decades ago when the property prices were generally way below today’s level. The property price level between now and 20 years ago is more than double. Therefore, similar to the investment properties discussed earlier, the fair value of these properties should exceed the book value by a substantial margin.
For example, the 4-storey factory building in Tampines has floor area of 9,780 sqm. This building has 30 years of lease from 1993, with a further term of 30 years. It’s safe to assume that the building has lease of 37 years left. Assuming the price per sqft is just $120, the whole building itself is worth more than 12m. This represents huge hidden assets in the balance sheet. My ballpark figure is more than 10m of hidden assets in these properties. In other words, they are understated by at least 10m.
These properties are currently utilized to run ABR’s business; hence the hidden assets may never be realized, and the market may never pay attention to them. After all, some of these assets have been sitting on the balance sheet for decades and don’t get re-priced at fair value. So, I don’t expect them to get re-priced for the next decade either. On the other hand, they may continue to get depreciated every year. In my valuation, I have excluded this hidden assets.
However, the total hidden assets from all the properties could be worth more than 15m. As a shareholder of the company, you have to be aware that these tangible hidden assets are there. The net cash of 82m plus the fair value of all these properties could be worth more than 120m. Against the current market value of 140m+, it implies that the Swensen’s restaurants, Tip Top Curry Puff and Season’s bakery and café business are valued at just 20m+ while generating more than 7m of profits. This tangible asset gives you assurance that the Company will not go bankrupt, but it will not prevent the share price to fall below the value of tangible assets as short term price movement is driven by emotion.
Last week, ABR announced that it plans to develop residential property in Australia. Its decision to develop property reminded me of Power Root, which manufactures Ah Huat instant coffee mix. Power Root’s property investment developed 64 shop lots in Malaysia and generated high return on investment. Power Root’s FY15 group revenue was nearly RM360m and operating profit was 48.5m. The property business generated revenue of 56.7m and operating profit of 19.6m. Its FY15 earnings were boosted significantly by this one-off profit from the property investment.
If ABR is also able to generate good return on its property investment, when the project is completed several years later, the profits will provide a strong one-off boost to the earnings. This would have utilized some of the excess cash of 82m very productively. I’m not assigning this property development any value at the moment.
Since divestment of The Cocoa Trees, ABR has been able to maintain a relatively stable operating margins in the past 3 years.
|Gross Margins %||47.4%||46.3%||45.9%|
|EBIT Margins %||8.5%||8.7%||8.4%|
|Net Profit Margins %||7.9%||7.3%||7.5%|
Note: my calculation of EBIT has excluded interest income.
Let’s look at the breakdown of the costs.
|Costs Breakdown as % of revenue||FY13||FY14||FY15|
|Depreciation & Amortization||3.1%||3.1%||3.2%|
|PPE + Inventory write off||0.5%||0.3%||0.9%|
Source: ABR’s annual reports
Labor cost is typically the largest cost for restaurants at about one third of revenue, and there is an increasing trend. The second largest cost is food costs at 25-26% of revenue, which is stable over the past 3 years. The third largest cost is rental at 16-17% of revenue, which dipped in latest year. These three cost items took away 75% of the revenue in total.
The company is not growing its revenue in the past 3 years, and the capex approximately equals the depreciation & amortization.
Return on Capital
We should note that ABR has net cash of 82.2m, which represents nearly $0.41 per share. This is more than half the market cap. Hence, the reported ROA (6.6%) and ROE (7.6%) are very low. After excluding this excess cash, the real picture of the business performance appears. The adjusted ROA is 23% and ROE 44.6%. Such high return on capital indicates that the business is very profitable and should grow by opening more outlets. Swensen’s is likely the cash cow to generate most of the profits.
The book value on fixed asset is 20m and the net working capital (inventories + trade & other receivables – trade & other payables) is -2.4m. Hence, the invested capital is only 17.5m. With operating profit of 8.4m and tax rate of 17%, the ROIC is ~40%.
|Returns on Investments||FY13||FY14||FY15|
|ROA excl. excess cash||22.3%||21.8%||23.0%|
|ROE excl. excess cash||45.1%||43.3%||44.6%|
Source: ABR’s annual reports
The ROIC of 40% is very high, especially so for a restaurant business. However, given the current operating environment, there is little growth opportunity in Singapore. There is a decreasing trend in term of ROIC, and I think this may continue further. But, even if it drops by half, it’s still high. This tells us that existing operation is generating very good return, and it must have possessed some competitive advantage to do so.
I visited the Swensen’s restaurant at Nex mall with my family yesterday. The food is decent, but not fantastic. Half of the menu is ice cream, which is what Swensen’s is known for. My quick survey to walk around the restaurants saw that about a quarter of customers ordered ice creams as desserts, which gave extra sales to the Company on top of the F&B. Ice creams are certainly prepared beforehand and can be served quickly, unlike food, which needs to be cooked. I suppose ice creams might be the key factor of ABR’s higher ROIC.
ABR’s group revenue in Singapore was 87m. If I am to make quick estimation and assume 7m came from Tip Top Curry Puff and 80m came from Swensen’s with 25 outlets, each Swensen’s outlet generated 3.2m sales per year. This is much higher than Japan Foods’s average of 1.3m sales per outlet per year (see this post). Note that Japan Foods has many brands, and this 1.3m is blended average across all outlets. Some better brands might have sales 1.5 – 2.0m per year.
One factor for such higher sales per outlet is Swensen’s higher price per meal. An average set meal at Ajisen Ramen costs about $16, which include side dish and drinks. A main course (not set meal) at Swensen’s costs about the same price. Add on the ice cream and drinks, the typical spending per person is $20-25, which is 25-50% higher than Ajisen Ramen’s.
Swensen’s average restaurant size may also be bigger than Ajisen Ramen’s on average. Swensen’s restaurant at Nex mall has about 160 seats. If the average table turnover is 2.5x a day, that’s serving 400 customers per day. Multiply by average spending of $20 per person, that’s 8k of sales per day and 240k per month and 2.9m per year. Not far from the calculated average of 3.2m sales per year.
Swensen’s also sells ice cream cakes with animation themes. If each outlet sells 10 cakes per day, 25 outlets sell 250 cakes per day. Multiply that by $40 per cake, that’s 3.6m sales per year.
Swensen’s is a halal certified restaurant. This is an advantage when 13% of Singapore citizens are Malays, and many tourists come from Indonesia and Malaysia.
Throughout this exercise, there are too many assumptions, but it gives us a rough idea on the sales. The assumption of seat turnover of 2.5x is low compared to the reported figure by Ajisen China (see this post). Ajisen China’s restaurants have table turnover of 5x in Hong Kong and 3.4x in China. So, my assumptions in above calculation might be off by a wide margin. A better way to do this is to ask the management if they can provide with more details.
I built a simple 5-year DCF model for ABR, assuming no growth for 5 years, operating profit of 8%, discount rate of 9.5% and terminal growth rate of 2.5%. The FCFF turns out to be around 6m per year. I consider this conservative as the actual FCFF during FY13-FY15 was 7.0-7.5m. My estimated EV is 77 – 86m. With net cash of 82.2m, the estimated equity value is 159 – 168m or 0.79 – 0.84 per share. With a margin of safety of 11-16%, I bought the shares at 0.705 last week.
ABR paid dividends of 5m or 0.025 per share in each of the past 2 years. At the current share price of 0.715, this gives a dividend yield of 3.5%. As the FCFF in each of the past 3 years exceeded 7m, it’s reasonable to expect that ABR could maintain paying the same dividend amount provided that the business does not deteriorate further. If FCFF falls short, it still has 82m of cash on hand. ABR may cut dividend to utilize the capital for further expansion as the ROIC is high.
- If the business can continue to generate the present profit level, the earnings yield is about 4.5%. Dividend yield of 3.5% should be maintainable. In 3 years, the total returns could reach 25 – 30%, provided my current estimate of undervaluation gets corrected. That shall produce an average return of 8 – 9% per year.
- If my estimate of undervaluation is wrong, and current price is the actual intrinsic value, then the potential return is 3.5 – 4.5% per year from the dividend/earnings yield.
- If the business deteriorates as competition and costs keep rising, then we could see loss of 0 to -15%.
- If the business condition improves in 3 years, and the property development in Australia generates good return, we could see rising profit with additional one-off boost. This might give total return of 40+% or 12% per year.
I’m investing in this Company with the expectation of around 8% return per year. This is not high, but given the current investment environment with low return, low inflation and low interest rate, especially in Singapore, I’ll be pleased with 8% return per year. Also, I did not factor in any potential growth for ABR in my valuation. So, if ABR does outperform and starts growing its revenue, I’ll be pleasantly surprised with all the positive development.
My valuation has excluded the potential hidden assets of more than 15m coming from all the properties that ABR owns. I don’t think the fair values of these assets will be built into the share price in foreseeable future.
As mentioned earlier, net cash represents $0.41 per share. After deducting this net cash, my estimated equity value is 0.38 – 0.43. Based on last year’s earnings of 0.038 per share, the P/E is 10 – 11.3x for my estimated value. This P/E range is considered quite low for a company with such high ROIC, so I consider my estimate conservative. This estimation also excludes any upside potential from the properties, which I estimated to have been understated by at least 15m.
I think the Company’s share price is undervalued because the near term outlook of retail business is dim, the profit is declining, it holds excessive cash, which causes the ROE to be below 10%, and general stock market valuation level in Singapore has been low compared to historical level. I hope the undervaluation can be corrected in next 2-3 years.