Description

Pentamaster International (PIL), a 63.9% owned listed subsidiary of Pentamaster Corporation Berhad (PCB), provides automated testing and solutions to manufacturers in the semiconductor, telecommunications, automotive and consumer electronics sector. PCB is a stock market darling back home in Malaysia and trades on its domestic stock exchange at a price to earnings multiple of around 39x given its high growth profile. However, on the Hong Kong exchange, PIL trades at around 10x earnings, or 9x earnings on a cash-adjusted basis, at a significant discount to its own parent (despite contributing slightly more than 100% of the parent's earnings), as well as compared to global peers.  I believe that the investor today can purchase a fractional ownership in a fast-growing business at a heavily discounted price, with high returns on capital, significant tailwinds, diversified revenue streams, and run by conservative management who has built up a significant net cash balance across the years. This situation reminds me oddly of the Korean preferred shares discount situation, where a significant discount exists with no good reason except perhaps for lack of familiarity within the investor base in Hong Kong.

General idea of the thesis

PCB is a listed company in Malaysia, which owns 63.9% of PIL (subject of the writeup) that is listed in Hong Kong. The corporate structure above indicates 3 other 100% owned subsidiaries, namely Origo, Smart Solutions, and InnoTeq. To better understand the contribution, one would have to look at the individual PCB and PIL financials.

PCB financials

PIL financials

In essence, PIL contributes to almost 100% of PCB’s topline, with the other 3 subsidiaries contributing less than 0.1% of revenue of PCB whilst being slightly loss-making at the same time, hence PIL actually contributes to more than 100% of the consolidated earnings of PCB. This is not a holding company discount situation in case this gets confusing. PCB is actually the holding coy, whereas PIL is the main operating subsidiary, yet Mr Market Malaysia ascribes a valuation of almost 4x higher compared to Mr Market Hong Kong. Of course, the equation is not as straightforward given there could be an argument for PCB being overvalued rather than PIL being undervalued. One should note though that the reason why PCB has been trading at relatively high valuations is because it is fast growing, with topline that has increased at a CAGR of about 28% since the semiconductor industry’s recovery in 2012 to 2021. PIL’s valuation currently stands at a steep discount to its peers, which shall be addressed later stage in this article

In order to understand how PCB / PIL got to where it is currently, I’ve written a little industry background for readers to better understand the context.

Brief history of Malaysia’s semiconductor industry

Malaysia first entered the electronics assembly industry in the 1970s with the 2nd Malaysian Plan in 1971-75. The plan intended to push for the creation of several Free Trade Zones (FTZs) to generate light export manufacturing jobs. It was with this back drop that several western semiconductor firms decided to setup shop in Msia This was as opposed to lower cost locations like Thailand and Indonesia  due to a few reasons: Malaysians perceived to have good work ethics, as well as proficiency with the English language. Aside from that, carrots were dangled in the form of tax free years, tariff exemptions, less bureaucracy etc, as well as waived National Economic Plan requirements (domestic shareholdership requirements).

The first large investments into Malaysia and Penang, which is termed the ‘Silicon Valley of the East’, came from 8 different companies which were initially termed the ‘8 Samurais’, namely Intel, Robert Bosch, Clarion, AMD, HP, Litronix, Hitachi and National Semicon. In 1972, Intel opened its 1st ever offshore assembly plant in Penang, which was better known for its agriculture industry back then. These ‘8 Samurais’ were looking for engineers back then, so locals started to seriously look into engineering as a field of study.

The founder of Pentamaster, Chuah Choon Bin, was one of the Malaysians who decided to further studies in the engineering field back then.

Back-end processes were historically seen as a ‘less exciting’ segment of the industry, although things are changing now, but this was one of the main reasons why these processes were outsourced offshore in the past. Jobs were created and exports soared with semiconductor business accounting for 20% of GDP at 1 stage. For some perspective, in 1977, US semicon imports from Malaysia was $658mn, which was actually larger than Sgp ($591mn) and S. Korea ($510mn), and 3x larger than Taiwan’s $210mn.

In the mid 1980s, an economic downturn and oversupply in integrated circuits resulted in huge layoffs. In Penang, employment went from around 19k in 1983 to 13.1k in 1986, with protests amidst these layoffs. However, local companies started becoming more efficient.

In 1985, the Plaza Accord resulted in a depreciation of the USD against JPY, Deutsche mark, and also caused appreciation of NTD and SGD against USD. This affected the competitiveness of exports from these countries. A similar backdrop about a decade later would actually result in the rise of Pentamaster and play an important in its rise. On the back of this, the government of Malaysia decided to devalue the MYR and extended investment tax credits to increase competitiveness.

Meanwhile, Malaysia launched its 1st Industrial Master Plan covering 1986-1995, with an attempt to try to move up the industry value chain. In the same year, the Malaysian Institute of Microelectronics Systems (MIMOS) was established to try develop a local semiconductor industry rather than huge dependence on western counterparts. 

In 1995, MIMOS launched SilTerra to try to build a TSMC equivalent of its own, although hiccups occurred and it was only launched in 2000 with the completion of its 1st fabrication plant. In contrast, in history that we’re more familiar with, Morris Chang launched TSMC in mid 1980s and in just about 3 years in 1987, TSMC was able to launch 3 micrometre process that was 1-2 generations behind then leading technology.

As history would have it, the Asian financial crisis occurred around 1997, which caused an economic crisis in the region, and Southeast Asian currencies to depreciate substantially against the USD. It was during this period that Malaysian capital equipment companies actually started to take hold as an opportunity arose where semiconductor companies, including the likes of Intel, started to source locally given USD imports were now markedly more expensive. A local supply chain started to take its form in Penang, as foreign companies in Malaysia started to try out local Malaysian firms as a much more cost-effective alternative. Local assembly and testing companies started to grow, into what we know now as the Big 4 OSAT (Outsourced Assembly and Testing) players, namely Inari Amertron, MPI, Unisem and Globetronics. 

Fast forward to present-day, in 2020, Malaysia’s electrical & electronics (E&E) exports were RM 386bn, which continues to be a sizable portion of exports (39% of Malaysia’s total exports), even more than oil and gas and palm oil, which are 2 exports that Malaysia is very well known for. Semiconductor devices and electronic ICs had an export value of RM 239bn, about 62% of E&E exports. InvestPenang’s CEO Datuk Loo Lee Lian has been promoting investments in other areas, citing the need to diversify from E&E so as to even out business cycles given the cyclicality of the semiconductor industry in the past. Medical technology might be an offshoot of the E&E and automation industry, and that is part of the reason why Pentamaster has been venturing into the medical field. Penang has 6 of the top 30 global medical device companies by sales. Healthcare companies in clusters and industrial parks in central Malaysia are rubber material-based, but medical device companies in Penang like Abbott, Boston Scientific, Dexcom and KLS Martin are electronic parts-based. By venturing into the medical field, this is part of an overall attempt to help medical device companies localise their supply chains. A localised ecosystem is very important, which is now established in Penang. For e.g. Lam Research setup a new plant in Batu Kawan, and the localised ecosystem has played a huge part in this decision.

Brief history of Pentamaster

I’ve compiled below a chronological timeline of the development of PCB/PIL across the years. Although this is not crucial to the thesis of the investment, I believe understanding the history of the company and how it came about till today will help the investor better understand several important developments and the DNA of the company, including: the reasons behind current growth trajectory, whether historical margin lows could potentially be hit again, and why the company sits on the amount of net cash it has currently. There were several turning points in the company’s history where PCB pivoted and adapted to changes in the macro-environment. 

- 1991: Penta Electronics was setup

o Chuah Choon Bin had been an engineer at Intel and National Semicon previously for 4 years

o Saw an opportunity, was involved in automation projects with these 2 MNCs back then

o Saw Intel buying a lot of equipment from all over the world, Intel was looking for a local supplier of equipment back then, hence took the plunge

o Penta Electronics back then, stood for Penang Technology Automation (PenTA)

- 1997: turning point due to Asian Financial Crisis

o Initially no track record

o AFC resulted in currency depreciation, importing from US and abroad became super expensive

o Pentamaster started to grow from then as MNCs started to look locally

- 2003: listing on the 2nd board of Bursa Malaysia. 

o Was transferred to the Main Board the following year.

- 2005: 

o entered into RFID Technical Collaboration Agreement with American RFID Solutions for business and technical know-how

o built 130,000 sqft factory building

o equipment contract manufacturing, for equipment manufacturers that supply manufacturing equipment to semicon and medical industries

- 2006: 

o ventured into development of Test and Measurement Systems (TMS) for electrical and electronics industry, with 3y R&D programme

o JV with listed Swiss coy Komax, called Komax Systems Penta (KSP), for manufacturing of auto assembly system, modules and parts for pharma, electromechanical and computer industries

- 2007:

o Established 1st rep office in Shanghai

- 2008: Global Financial Crisis, another turning point but downturn

o Had to move away from contract manufacturing which was lower value-add

- 2010: started to focus on higher value add products / move up value chain

o Started going into higher end testing solutions / automation

o Had around 2 decades of automation experience already, so went into smart sensors, automotive and factory automation segments, which are ‘higher end’ automation technology

o Established Glove Reprocessing Unit (GRU) which caters for healthcare industry

o Reprocess gloves using a fully automated reprocessing and testing system, up to 7x compared to normal single use conventional gloves available in the market

o Disposed of 2 factory buildings to pay down debt. I believe this played a crucial part in why PIL now sits on net cash, as this was a painful period in PCB’s history.

- 2015: 

o sold Pentamaster Engineering and Pentamaster Solutions as suffered losses for a few years

o bought Origo Ventures to enter Smart Home and Building Solutions

o developed i-ARMS (intelligent automated robotic manufacturing system) to address manufacturing industry needs under Industry 4.0 and granted Pioneer status

- 2018: 

o Listing on HKEx

Wanted to go global, saw a huge market in Greater China

To do business in China, the company has to be physically in China 

Chuah Choon Bin believes that business in China grew after HK listing

o I-Hub: Smart and Cost Effective Warehouse logistics. 

Processing and manufacturing time of a business accounts for <20% while over 80% of time is spent on storage, handling and transportation. 

Apparently logistics and storage alone costs 30% of the whole supply chain’s cost.

I-Hub is designed for space-saving and cost efficiency

- 2020:

o Trade diversion due to trade war

o PIL can sell to both China and US given their location / neutrality

o Pentamaster MediQ was setup to leverage on TP Concept to product single-use medical devices for healthcare industry. Focus on intravenous catheters, and dual-safety pen needles. Committed to invest about RM60mio over next 3 years. 

- 2021: 

o focusing on the need for Offshore Support Site (OSS) given the trend of deglobalization / nationalization recently. Having a local presence gives client confidence. Currently OSS in China, Singapore, Taiwan, US and Japan. Will try to setup in Germany to tap into EV market, and manufacturing site in Fujian, China.

Overall business of Pentamaster

PIL focuses on 2 main business segments, namely the Automated Testing Equipment (ATE) segment and Factory Automation Solutions (FAS). PIL actually provides a wide range of solutions, which are listed below.

Below is a bird’s eye view of the product lines of PIL.

ATEs are used to verify specific Device Under Test (DUT) functions according to design specifications, or identify and diagnose faults. This process needs to be automated because of the sheer volume of devices to be tested, as well as the micro sizes of the test points on each DUT. PIL’s focus is on the non-memory and test handler segments. Test-handlers act as transportation mechanisms to move parts to be tested, and to present them to ATE terminals where the test measurements are performed, and also to perform tested parts sorting (called binning). 

Products under the ATE segment include standardised and customised standalone ATE for

i) semiconductor electronic components testing for smart sensors, power modules and integrated circuits

ii) proprietary burn-in test technologies for wafer-level devices and memory modules,

iii) end products testing for consumer electronics, electro-optical products and LEDs

As a ballpark figure, these equipment typically cost around MYR 1-3mn. Due to the importance of testing as a whole, there is a certain amount of stickiness, especially as the requirements and rigours of testing increases with increasingly complicated technologies and amounts of DUTs to be tested. The ATE segment remains the largest contributor to PIL’s topline as well as operating income.

In the FAS segment, PIL customises and automates manufacturing processes by integrating automated assembly and testing modules, material handling equipment, robotics technology, auto inspection and manufacturing executive systems. i-ARMS is their flagship product, which is a factory automation solution system capable of integrating combinations of FAS modules with other technology components, such as vision devices, sensor devices and RFID. Again, such solutions are sticky given the need for users to be familiarised with systems, as well as bespoke solutions required for application specific automation. 

Although PIL has its roots in Penang, it has now grown and expanded into several locations globally, given the need for customer touch points locally as solutions are generally bespoke.

Overall local industry background and where PIL stands

There are a few large ATE companies in Malaysia, all of whom focus on slightly different segments and do not specifically compete head on with each other. The termed big 4 ATE companies are ViTrox, Greatech, Pentamaster and Mi Technovation. There are several other local ATE peers as well including QES, Aemulus, MMS Ventures etc.

The ATE segment is likely to grow from smart sensors adoption + 5G, as well as electrification of vehicles. Integration with AI features will likely propel sales of new hardware and devices moving forward. The FAS segment is likely to grow with adoption of automation for industry 4.0 and Internet of Things (IoT)

ATE generally results in reduction of manufacturing time, increasing throughput. However, ATE is expensive. However, the good thing is that the life cycle of ATEs can be about 5-10 years. ATE is also not just about saving time, but also about tweaking parts for lower power consumption and higher accuracy, which brings the product to market faster.

The ATE market is slowly getting saturated with competition, especially for the standard ATE i.e. pick and place machine. However, PIL’s segment is smart sensor testing, which is less crowded due to higher expertise in technical design and development. The likelihood is that in years to come, competition will likely drive down margins here, so diversifying the product portfolio to EV and medical product segments would be prudent, which have higher barriers to entry and stringent requirements. Management has targeted that single-use medical devices will eventually contribute about 1/3 of total revenue.

Here’s a quick overview of the overall semiconductor value chain as well as the Malaysian companies in this ecosystem.

I will not go into specifics here on the description of each value chain segment, as this would make the article unnecessarily long. It should not be particularly tough for the reader to obtain information on the different segments within the value chain as there is a wealth of information out there.

In the past, the Malaysian semiconductor industry was mostly dominated by OSATs like Inari Amertron, MPI, Unisem, Globetronics and KESM. These companies were established to provide outsourced services such as assembly, packaging, testing, to MNCs like Broadcom, Infineon, Intel, Osram, AMD, Agilent etc. These OSATs, alongside other MNCs, created a need for ATE manufacturers like ViTrox, Pentamaster, Greatech. There are now several Malaysian companies with links to global MNCs, including but not limited to:

o Pentamaster (Intel)

o Vitrox (former employee of Agilent)

o Globetronics (Intel)

o Inari (Agilent)

o Eng Tek (Intel)

o LKT (Intel)

o Wong Engineering (Intel)

o Unico (Intel)

In Malaysia, Vitrox appears to be the main listed competitor. Vitrox develops and produces automated vision inspection systems, and is also a very profitable company with decent ROE and strong revenue growth across the years. Elsoft Research researches, designs and develops test and burn-in systems and applications specific systems. Aemulus does design, engineering and development of ATE, and focuses on RF and mixed-signal semiconductor test markets, with a focus on architecture and marketing of testers, rather than manufacturing itself, so the cost structure is different. MMSV Ventures manufactures industrial automation systems, etc. Mi Technovation designs, develops and manufactures wafer level chip scale packaging sorting machines

Overall global landscape and comparisons against peers

On the global stage, the major ATE manufacturers are Teradyne, Advantest, Chroma, Cohu and LTX-Credence (acquired by Cohu in 2018). The industry is relatively consolidated with these players having about 85% of market share. These are the main competitors in each segment. 

o Semiconductor testing

Teradyne

Advantest

Cohu

o System Test

Teradyne

Keysight Tech

Advantest

Test Research

SPEA S.p.A

o Wireless Test

Rohde & Schwarz

Anritsu

Keysight Tech

National Instruments Corp

I’ve compiled some financial figures below for a quick comparison of companies across the board.

Revenues

Gross margins:

Operating margins

Pre-tax margins

There are varying business models for all different companies which result in different margin profiles for each competitor. Nonetheless these figures are presented side by side as a direct comparison for the reader to understand where PIL stands against other competitors.

General financials across time

For completeness, I’ve included some financials across time wherever available. The industry backdrop I have provided above, as well as company specific developments earlier in the article, should help the reader better understand the context of the financial development of PCB / PIL. 

The lowest gross margins attained were in 2008 during the GFC. In my opinion, these gross margins will likely not be attained in future downturns, primarily because PIL has since moved on from its lower value-added products in the past to a higher value-added product mix. For some context, at 2021’s revenue level, even if gross margins were to hit the trough figure of 7%, PIL will be at a breakeven level.

One of the clearest signs of improving economies of scale has been the general decline in SG&A expenses as a % of revenue across the years. PIL’s scale has allowed for some degree of operating leverage and this is reflected in the improving margins across the years.

The net cash balance that has been built up across the years has been very conservative. At current SG&A expense run-rates, and even taking into account current year capex levels, it would not be inconceivable for the cash balance to last 5 years. As aforementioned, I believe this to be a deliberate move by management, as a result of the GFC experience, and PIL is certainly in a much better position now to weather potential downturns.

ROE has been declining across the past few years as the company’s increasing net cash position has been a drag on returns on capital. As explained earlier in the article, due to their awful experience during the GFC where some asset sales had to occur, PCB had not paid dividends from 2008 to 2017, despite already returning to a net cash position in 2014. The situation changed after PIL’s spin off, with a dividend payout ratio ranging from 12-22% from 2018 to 2021.

Industry exposure is not as concentrated / dependent on semiconductors as one thinks

Given Penang’s semiconductor dependence as well as PIL’s own previous dependence on the semiconductor industry, one would be inclined to think that PIL’s business mix focuses a lot on semiconductors. However, a quick look at the customer segments reveals a pretty diversified revenue mix. The medical devices segment in particular has been a relatively interesting development, which was elaborated upon above where Penang’s push into the medical devices industry has developed opportunities for PIL. The industry mix is not a static situation as things do change from time to time.

PIL’s orderbook has been robust. For FY 2022, their order book reached all-time high of RM 500mn as of May’22, of which 70% was ATE and 30% FAS, and 40% was from the automotive division. 60% are from new customers, with higher ASPs of up to 12%. The order book is expected to be recognized in 6-9 months. The customer concentration is acceptable as well, with the largest customer accounting for 15.0% of revenues and largest 5 customers in aggregate 46.0% in 2021. 

According to management, PIL used to pick up between 1-3 new clients a year. However, because of Covid, they managed to add 10 customers in 2020, of which 7 were from China, which has ambitions to become self-sufficient in chips. PIL setup a WFOE in Apr 2021 to take advantage of this. 80% of PIL’s exports are now to MNCs. Average selling prices are on the rise as well. In 1H2017, PIL delivered 50 test equipment, with a value of >RM 20mio, so the average was around RM 400k. As of around 2022, ASPs are now around the RM 1-3mn range. Due to Covid, clients have been ordering on a lower quantity/piecemeal and Just-In-Time basis, where they used to order in bulk in the past. Unfortunately, this has made it harder to estimate resources required, and also results in a longer production lead time.

Supply chain challenges have prompted local procurement to reduce disruptions

On the procurement front, about 20-30% of supplies for PIL come from China. PIL also buys most components like motors and sensors from Japan and Europe. The company recently changed the design of some equipment to accommodate parts from outside China. PIL used to import precision products from overseas to use in their machines. However, local players including ViTrox, Pentamaster and Walta Engineering combined together in a joint venture to setup the Penang Automation Cluster project to further enhance the local supply chain ecosystem.   

M&A record has been acceptable though new businesses haven’t always worked out

Below are some acquisitions and sales that occurred across PCB/PIL’s history.

- Acquired TP Concept for about 4x earnings in Sep 2019, 20mio consideration for 15.9mio NAV

- Aggregate Profit Guarantee of RM 12mio for FY2020 and FY 20201

o However, profit guarantee unlikely to reach due to Covid

o Therefore extended to aggregate for 2019-2022          

- Acquired Origo Ventures for RM 5.8m in Oct 2015

o Below NAV, so recognized a bargain purchase arising from acquisition of about RM 2.6m              

- Sold Pentamaster Engineering and Pentamaster Solutions to GEMS Tech Fund for RM 5mio

- Bought 30% of Everready Precision Industrial Corp’s (EPIC) stake for US$6.8mn in 2022

o Mainly in Taiwan, engages in electro-optical operations

o Specializes in designing and manufacturing end-to-end optical integration solutions and components, such as advanced optical precision moulds and lenses  

o Currently loss-making

In terms of new business ventures, PCB’s contract manufacturing focus didn’t exactly work out, but the shift in focus to higher-end products subsequently was a smart move that brought about the growth that PIL has seen today. The Smart Control business as well as the Glove business did not exactly work out as well, so there have been some hits and misses in PCB’s history. Nonetheless, I actually see this as positive, as it shows that management has been forward-thinking, and very proactively trying to diversify revenue streams and strengthen business further.

To further aid their expansion into the medical device sector, PIL has earmarked about MYR 60mio to be spent on medical devices from 2021-2024. The company purchased leasehold land in Batu Kawan in Nov 2021 for MYR 28mn for their Campus 3 expansion, and will add 600k sqft of production space. The main intention is to cater for the growing FAS and medical devices segments, to attain the management target of MYR 1bn topline by FY 2025, which translates to an 18% CAGR growth target from 2021 to 2025. 

Management has been conducting buybacks through PIL and insider purchases as well

There have been a number of share buybacks and insider purchases in the past years. As would be expected, PCB has been conducting share buybacks through PIL instead of buying back PCB shares on the market given the huge gulf in valuation. I’ve highlighted a number of such transactions in the past year or so, although there have been more before August 2021.

PCB has been conducting bonus issues in the past, and management has tried to do the same at PIL as well alongside a reduction in board lot size, perhaps in an attempt to gain investor interest in PIL stock. This has unfortunately not quite turned out quite as management would like, although I do not think that such corporate actions add any value to PIL’s business.

Valuation of PIL is not only at a discount to PCB, but also to its peers

Aside from the obvious valuation gulf between PCB and PIL, PIL also trades at a significant discount to its Malaysian peers as well as global peers. For simplicity, I have used earnings multiples as a comparison, as these numbers should suffice in showing the sharp valuation discount PIL has. 

Amongst the big 4 ATE manufacturers in Malaysia, ViTrox, its closest peer, trades at 37x earnings, Greatech at 41x earnings and Mi Technovation at 18x earnings. For the global players, Terradyne trades at 20x earnings, Advantest 16x earnings, Chroma 19x earnings and Cohu 15x earnings. Perhaps more revealingly, even Innotech Corp, which is a Japanese ATE trading company with a significantly inferior growth and margin profile, trades at about 10x earnings or a slight premium to PIL. At about 9x earnings net cash of ALL liabilities, at the very least I believe the investor today pays a very reasonable price for a company growing fast with tailwinds and earning decent returns on capital.

Risks of PIL’s business

One of the key risks of the business that management has repeatedly outlined is the lack of appropriate human capital. There has been a shortage of talented engineers in Malaysia, which is a worrying issue given that out of 719 staff, >600 are engineers. I believe this to be a non-politically correct reason to explain in public, but in reality, due to the lower wages in Malaysia, many of the talented engineers head to the US, Australia or Singapore for higher-paying jobs. A risk mitigator has been PIL’s setting up of offices worldwide to tap on existing talent locally.

Another risk on the back of investors’ minds would be a cyclical downturn especially in the semiconductor industry. PIL is not a non-cyclical business, even as there have been attempts to diversify revenue streams. However, the recent foray into the medical devices should help PIL during potential tough downturns, and the Fort Knox balance sheet that management has built would certainly help PIL weather the next downturn better than they did during the GFC.

How else can the investor potentially take advantage of the situation?

One of the options that the investor can undertake is to go short on PCB and long on PIL. This could theoretically work while the valuation converges. However, I would not recommend the investor undertake such an action as there are 3 other subsidiaries (Origo, Smart Solutions, InnoTeq) that could certainly have a potential to grow in time. Also, an additional FX risk exists for the investor given PCB is denominated in MYR and PIL is denominated in HKD.

Catalyst

An increase in profile within the HK investor base as PIL's profile grows. Continued share buybacks and paying out of dividends from its net cash pile could help narrow the discount as well.