The OceanLink Partners Fund’s unaudited gross and net returns for the second quarter of 2019 were 9.15% and 7.35%. This puts our gross and net returns YTD at 61.81% and 49.37%, and returns since inception (2017/06/01) at 111.53% and 83.14%, respectively.
| Returns | OceanLink, Gross | OceanLink, Net | MSCI China All Shares | MSCI World, Net |
|---|---|---|---|---|
| 2017 (7 months) | 30.08% | 23.85% | 22.85% | 11.04% |
| 2018 | 0.50% | -1.00% | -23.27% | -8.71% |
| 2019 YTD | 61.81% | 49.37% | 17.58% | 16.98% |
| Cumulative | 111.53% | 83.14% | 10.83% | 18.57% |
| Annualized | 43.28% | 33.70% | 5.06% | 8.52% |
Given your and our long-term focus and the inherent volatility in our investment strategy, what happens to the quoted value of the portfolio in a month, a quarter, or even a year should be of little significance to you, and to us. As we have asked all of you to commit to a three year investment, our preferred KPI for OceanLink is a three-year CAGR, so we encourage you to sit back and skip the first paragraph of our future investor letters until the Q2 issue of year 2020.
Testing Champions
Continuing on the Service Champion theme from our previous letter, we chose the Testing, Inspection and Certification (TIC or testing in short) industry as our third case study. The TIC industry is a $230 billion market globally and provides an essential service to virtually every major sector in the economy. China’s TIC market, at $40 billion, is the largest at 18% of the world’s total. It’s also the fastest growing at over 10% a year, driven by rising demand for quality, growing complexity in manufacturing processes, and increasing regulatory standards. Third-party testing makes up half of the industry, with the rest being done in-house or at government agencies. Outsourcing penetration is steadily rising as the third-party model offers impartiality and significant cost advantages over the in-house model, allowing third-party service providers to enjoy even faster growth (~15% per year).
In the previous letter, we described Service Champions as those befitting five key characteristics: industry fragmentation, complex operations, returning customers with high information asymmetry, scale benefits and uneconomic small players. The third-party TIC industry in China strikes us as one that scores particularly well on these criteria:
- Industry Structure: China’s testing market is extremely fragmented with nearly 40,000 participants, the vast majority of which only operate a single testing lab. The largest player in the industry only has 2% market share, and the top 10 combined add up to less than 10% of the overall market.
- Nature of Operations: The testing business model is quite complex, involving a great deal of scientific know-how and deals extensively with government regulations and industry-specific requirements. A typical third-party testing lab also requires numerous government certifications, a B2B sales function, a quality control function, and a procurement function. Usually, a large menu of different test services (chemical, optical, etc) must be provided to serve a vertical. Given the complexity, it’s unlikely that most small players can operate at high standards.
- Customer Lifecycle: Testing is a frequent, recurring business with a long customer life cycle, and is subject to a great deal of information asymmetry. Unethical behavior is rampant, as there are many ways for small operators to cut corners and save on costs (ex. using low quality machines or under-trained employees, skipping testing procedures, or outright issuing fake reports).
- Benefits to Scale: There are also significant scale advantages inherent in the business model. For example, an inaccurate test result or a dishonest report can lead to the wrong products being released into the market and significant reputational damage, so most testing clients, especially large enterprises, are extra risk averse when it comes to choosing testing providers – a dynamic that favors the biggest brand names. Also, for consistency and ease, customers always prefer to use fewer vendors; tests are also time-sensitive, as a day’s delay in test results is a day wasted not being able to sell the product. As such, large clients prefer “one-stop shop” providers who have a national network of local labs with sufficient service radius to ensure timely delivery and processing of test samples. In addition, the standard-setting boards of many industries consist entirely of scientists working for the biggest TIC companies; their ability to dictate how standards are set can help their companies develop and bring new services to market faster than their smaller rivals. Furthermore, only the leading players have enough volumes to justify building large scale central labs, which operate at lower costs and are often the only facilities capable of running certain niche tests, giving them a monopoly in many product areas. There are many other minor advantages (procurement savings, R&D, ability to optimize network utilization, etc) in this model that are too obvious to expand upon, but the one consistent theme that we have observed throughout our research is the presence of a positive feedback loop, with scale advantages leading to further scale. It’s no wonder that almost all of the leading TIC players globally were founded in the 1800s.
- Profitability of Small Players: Finally, most small operators would be uneconomic without cutting corners. We’ve been consistently told that a proper testing lab with all its fixed costs needs at least RMB 10 million just to break even. With an average RMB 7 million in sales, many small operators would not be able to survive if they did not resort to cost cutting measures that would impact the quality of their test results.
These dynamics make the TIC industry an ideal breeding ground for Service Champions in China, just like it has been in the West for decades. The TIC industry is actually relatively under-followed in the US due to a lack of large domestically listed companies. Most of the world’s testing champions are instead European companies with global operations: SGS, Intertek, Bureau Veritas – while obscure names in the US, they have produced some of the best “compounders” globally in the past two decades. In fact, Eurofins, the second largest French player, has been the single highest returning stock in Europe in the last twenty years, having compounded more than 200 times its IPO price.
What makes a “testing champion” such a powerful business model?
For one, third-party testing is a GDP+ growth industry, and has consistently been for over a hundred years. With rising living standards and productivity, consumers demand higher quality, safety and more complex products, companies demand higher volumes of precise data as input in business decisions, and governments create more regulations and standards to ensure better protection of social interests. In a perfect world, almost everything will be tested and measured, and until we reach that day, testing intensity (testing as % of GDP) still has room to go up.
Additionally, similar to software, the mission-critical nature of third-party tests and their relatively low-ticket costs mean that service providers are often endowed with above-average pricing power. This is especially the case in many verticals where the choice of testing vendors is dictated by the clients’ downstream partners (retailer, OEM, or government), who usually limit their options to a small “approved list” of established brands. Customers rarely switch vendors, with churn rates usually in the single digits.
The low price-to-value ratio also means third-party testing demand has low sensitivity to the overall economy. Most major testing companies are further diversified across multiple verticals (SGS, the biggest of them all, has a presence in every single vertical) making their revenue trends even less volatile throughout a business cycle. As a result, these companies almost never experience revenue declines (all of the major testing players enjoyed positive organic growth and expanding profit margins in 2008 and 2009).
Crucially, the TIC industry faces no serious technological disruption risk. In essence, third-party TIC companies are like accounting firms – they serve a gatekeeper role in protecting consumers from sub-standard products, and they are in the business of selling trust. As long as businesses, consumers and government agencies don’t fully trust each other, TIC companies have an important role to play, no matter how testing technologies evolve over time.
Finally, because testing companies are excellent businesses, it’s an even better business to roll them up! The scale advantages that naturally accrue to the leading global players make them ideal consolidators of subscale and regional testing institutions which still account for most of the industry. It’s no surprise then that the biggest testing companies in the world have all been roll-ups, with some making over a dozen acquisitions a year. These acquisitions tend to be very low risk and show improvements quickly, making them highly value-accretive efforts.
Centre Testing International (CTI)
CTI is the largest domestic player in China’s third party TIC industry, with 130 labs and 90,000 customers. Founded in 2003 by co-founders Wan Feng and Guo Bing, CTI was one of the earliest domestic private companies to enter into the TIC market. A pioneer in the testing of hazardous substances in electronic products, CTI used its early mover advantage to develop a reputation that later allowed it to branch into the environmental, food safety, life sciences, consumer and industrial verticals. Today, CTI is three times the size of the next biggest domestic competitor, and widely regarded to be the most reputable Chinese player in the TIC space.
The man responsible for much of CTI’s early achievements, Guo, left the company in 2014 following a disagreement in strategy with Wan, who was the larger shareholder. Wan took over CEO duties and set out to aggressively expand CTI’s presence, spending significant sums to build out new labs and extend into new verticals. Unfortunately, Wan was not a good CEO. To support the expansion, Wan made revenue growth the primary KPI in evaluating manager performance, and pretty much ignored everything else. While this allowed the company to scale quickly, it also resulted in inefficient allocation of resources and a bloated cost structure. There was so little emphasis on profitability that the company never measured even basic efficiency metrics such as capacity utilization and revenue per employee. The culture under Wan became chaotic and political, as managers fought over bonuses and territories, and were promoted on the basis of favoritism rather than ability. The result was that the wrong people rose to the top and talented managers were marginalized. Performance followed, with operating margins falling every year, from 20%+ in 2013 to 9% in 2017.
What triggered our interest in a seemingly mismanaged company with a rather mediocre founder? While CTI isn’t a particularly efficient organization under Mr Wan’s leadership, the company did one thing well – it had always insisted on maintaining integrity in its test results, and always used the best testing equipment and technologies to ensure quality, so its reputation among customers was never damaged by its internal mismanagement. CTI’s rapid growth also allowed it to build out the most extensive lab network in its largest verticals, which reinforced the company’s industry standing. In some sense, the business still had a healthy core intact, if only it could find the right person to revive it.
The revival came last year. As CTI’s performance deteriorated, Wan realized that his shortcomings would hinder the company’s development and CTI needed a more capable leader, which led to Wan firing himself from the CEO position. His replacement was Shentu Xianzhong (yes, that’s a four syllable Chinese name), the former Head of SGS China.
The Turnaround Story
To provide some context on the importance of this hire, SGS was once a similarly mismanaged family-run business with a bloated cost structure and a host of managerial issues struggling with profitability. In 2001, SGS had become largest TIC company in the world following an aggressive roll-up campaign, yet the company was shrinking organically and earning a meager 5.9% operating margin. To improve its performance, SGS’s controlling shareholder, the Agnelli family, brought in a young CEO named Sergio Marchionne who had a reputation as a turnaround specialist. Against deep skepticism from the investment community, Sergio immediately outlined a long-term plan for SGS to improve operating margin to 10% and increase top-line growth to double digits. In support of the plan, he made several important changes: first, he flattened the organization by removing five layers of management. Then he re-oriented the KPIs to profitability metrics such as margins and ROIC, in the process forcing the organization to become cost-conscious and return-focused. Finally, to create a merit-based culture, Sergio replaced the old guard with young, entrepreneurial managers who delivered results.
These changes marked the beginning of a textbook case study of a perfect turnaround effort – in just one year, Sergio had blown past his 10% margin goal; by the time he transitioned to Chairman in 2004, SGS was making a 14% operating margin and growing at 15% organically, just 24 months into Sergio’s tenure. Shentu was one of the young people that Sergio promoted. Having joined SGS China in 1990 as its third employee – a quality control inspector, Shentu had steadily risen up the ranks through hard work and leadership. His former colleagues told us that Sergio was impressed by Shentu’s abilities and promoted him to run the biggest product line in China in 2002. Five years later, Shentu became the first head of SGS China, and spent the next eight years building it into the No.1 testing institution in the country. Today SGS China accounts for 15% of SGS’s global revenue and nearly 30% of its profits. In 2015, Sergio promoted Shentu to EVP in charge of running Consumer and Retail, SGS’s most profitable division, where he sat on the company management committee and gained significant experience in overseas M&A.
It’s clear from our reference calls that Shentu’s management philosophy was deeply influenced by Sergio. Like Sergio, Shentu has been described to us as being strategic, pragmatic and results-oriented. Able to remember even the names of frontline employees, he inspires loyalty and unity, and is a firm believer in the power of a good culture. We interviewed over 70 industry experts and can hardly recall any negative comments on Shentu; even people he had fired were singing his praise.
Yearning to return to China, Shentu joined CTI as CEO in June of 2018. At a rare interview, he revealed the primary motivation behind the move: “After so many years working in the TIC industry, I observed one phenomenon: every country with a large testing market has cultivated a world-class testing institution… As China has now surpassed 10% of the global testing market, it should be able to develop not one, but several testing companies with global scale and reputation. As one of the earliest participants in the Chinese testing market, with the type of experience that I have accumulated, it’s fair to say that I now have the confidence to take on the responsibility of building a globally competitive testing institution in China. This is a dream that I have always held. With CTI as a platform, I believe the time to do this has arrived.”
Implementing the Vision
Several of our sources similarly informed us that Shentu had aspired for many years to build the SGS of China. If he succeeds, the upside is significant – SGS is almost twenty times the size of CTI; given the potential offered by the Chinese market, it’s not inconceivable for the leading Chinese player to eventually reach a similar size as its Swiss counterpart. To reach this goal, Shentu must first make deep, impactful changes to revitalize CTI.
Many of the needed adjustments mirror the changes that Sergio had made at SGS 17 years ago. First, he would re-organize the leadership team and rebuild the corporate culture, by cutting management layers and promoting capable managers into more important roles. The restructuring has been swift – within six months, Shentu had replaced most of CTI’s senior managers. We spoke to one of those executives, who confirmed with us that the managers that were removed had indeed been the worst people in the company, before revealing that he invested in CTI shares immediately after Shentu fired him… (He was not alone; in fact, quite a few managers we interviewed who worked at competitors became CTI shareholders after hearing news of Shentu’s appointment). To hold people accountable, Shentu also implemented formal performance rankings where the bottom employees are fired every year.
Next, Shentu set out to create a focus on performance, where every relevant metric would be measured and tracked. The number of KPIs at the lab level soon grew from just one to over twenty. The budgeting and spending processes were changed such that every new capital spend must now be justified by a detailed ROI analysis. Shentu also laid out a medium-term target of 15% net margin (this is an after-tax metric) and publicly stated that any testing company that can’t make 15% in China is not a good company. To get there, Shentu shifted the company’s primary KPI from revenue growth to profit growth. As a result, low return projects have been curtailed and organic growth has been re-prioritized.
These measures have quickly made an impact. Just one year into Shentu’s tenure, CTI has already improved its operating margin by over 5% and grew its revenue by 26% without opening a single new lab.
Future Growth and Valuation
Longer term, as part of CTI’s going global strategy, Shentu plans to acquire several overseas testing labs which are certified by large multinational customers in order to refer their Chinese demand to CTI’s domestic lab network. This will put CTI on a level playing field with the likes of SGS in the lucrative Consumer testing market. To scale CTI’s nascent industrial testing business, Shentu poached away Zeng Xiaohu, SGS China’s former head of Industrial Services. We have high hopes for this hire as reference calls on Zeng came back extremely positive – one of the most capable managers at SGS, Zeng was responsible for building out the company’s entire industrial business in China. In the life sciences vertical, Shentu plans to prioritize CTI’s CRO (outsourced clinical trials) business which is showing early promise, with triple digit growth in the past year. This is a potentially 40% margin business at scale with oligopolistic characteristics. If successful, these initiatives will help CTI achieve bigger revenue mix in higher margin, less competitive verticals.
To align incentives, Shentu purchased a significant amount of CTI shares in the open market, and was awarded 3 million stock options in the first stage of a multi-year stock incentive program. Another 23 million options, accounting for nearly 1.5% of shares outstanding, were given to the other top managers, including several who had followed Shentu from SGS. Crucially, for these options to vest, CTI has to nearly triple its earnings by 2020 (from 2017 levels), a target that we believe the company is already on track to exceeding this year.
When we first discovered the company, CTI was trading at 30x EBIT. How much upside could there be in such an expensive stock? Our research led us to conclude that CTI was deeply under-earning and on normalized margins the stock was actually reasonably attractive. This is mainly because a sizable portion of CTI’s labs are quite new. The lengthy certification process (usually a whole year) and high fixed cost base mean that it generally takes three years for a new lab to hit break-even, and five years to reach a mature level of profitability. With 40% of CTI’s labs under three years old and more than half under five years old, the average utilization rate across the portfolio was only 50%, compared to 80% for its slower growing competitors. Similarly, CTI’s average revenue per employee was only RMB 339k, compared to RMB 400-500k at other leading testing companies. Both of these metrics should converge with the competition as CTI’s new labs mature over time. Depending on the vertical, we estimate that mature labs can earn anywhere between an 18 to 35% EBIT margin. As CTI has completed building out its lab network and will mainly grow organically going forward, these margin levels will start to be reflected on the income statement.
In addition, even CTI’s mature labs are operated sub-optimally. We have heard many anecdotes of gross inefficiency from our channel checks, and expect them to be corrected in the next 2-3 years as Shentu optimizes operations. Overall, we believe a low 20s EBIT margin is a realistic long-term goal, more than double what CTI earned last year. Meanwhile, we expect CTI to grow at 25% organically, a reasonable progression rate in the context of an industry that is growing at 15%. With $500 million of revenue, CTI has only a 1% market share in the Chinese testing industry and 2% market share among third-party providers, and the TAM is only getting larger over time as every five years, the government opens up new verticals for market competition.
If CTI plays a similar consolidator role in China as SGS, Intertek and Eurofins have played in Europe, the company should also have a long runway of accretive M&A opportunities ahead in the coming decades. Many private and SOE labs that specialize in attractive product lines would make good targets, as are overseas labs that can create revenue synergies. With more than ten years of M&A experience under his belt, Shentu should be well prepared to capitalize on these opportunities.
In the meantime, the competitive environment in China is favorable. Most of CTI’s peers are either mom & pops or government labs. Its closest domestic competitor is an SOE and only one third in size. Chinese subsidiaries of foreign testing companies such as SGS, while comparable in size to CTI, are banned from competing in many verticals due to data security concerns, and suffer from risk-averse decision making at their global headquarters (this was another important reason behind Shentu’s departure from SGS). CTI should be well positioned to take market share from these players.
Risks and Considerations
So what can go wrong? One potential issue is that the 20%+ margins of mature labs in China to which we are underwriting appear higher than elsewhere, as all of the top global testing companies earn EBIT margins of around 15-16%. Will testing margins in China converge with the global level in the long term?
Our research suggests that the high lab margins in China are likely structural. One reason that explains the gap is that the European companies derive significant portions of revenue from low margin verticals such as Oil & Gas and Industrial (8-10% EBIT margin), whereas the mix of testing volumes in China tend to skew toward higher margin verticals such as Consumer and Food Safety (SGS’s global Consumer business earns an EBIT margin of 26% after overhead). Another important factor is the sheer volume of tests in China, which combined with the high density of customers in large cities mean that far more business can be covered in the service radius of an average lab. This translates into bigger labs, greater capacity utilization, and better absorption of fixed costs. For instance, we estimate that SGS generates around $6.7 million of revenue per location in China, almost 3x its per lab revenue of $2.3 million outside China. Finally, labor costs are far lower in China than developed countries. Again, in the example of SGS, the Chinese business generates an average revenue per employee similar to its global average, yet Chinese lab workers are paid as much as 50% less than their developed market counterparts. These savings fall to the bottom line, resulting in higher margins.
A more pressing concern has to do with pricing, as we discovered through channel checks that price inflation has been negative in recent years in CTI’s largest vertical, Environmental. Should pricing continue to weaken, will margins adjust with it?
Our research reveals two primary factors that have affected pricing in the recent years. The first is the entrance of thousands of small independent companies into environmental testing after the government opened up the industry to private competition in 2014. Most of these small operators obtained contracts through local government relationships in relatively under-developed regions. They keep costs low by using low end (less accurate) equipment and many issue fake results without conducting the actual test. The regulators have become aware of this issue and have put regulations in place to force testing vendors into compliance. For instance, government auditors are currently conducting a nationwide audit in which every lab is required to submit all reports issued in the past 3 years for validation. All labs that fail the audit will be ordered to cease operations and lab owners who issue fake reports in the future are liable for life. We expect compliance programs such as these will result in the exit of thousands of subscale, low quality providers, while reducing competitive intensity and restoring pricing discipline.
The second factor is related to the cost structure of the testing industry itself. We estimate that as much as 70% of the cost of operating a lab is fixed, and the marginal cost associated with an incremental test is very low. Even labor cost scales very well – we’ve heard that a lab needs 40 employees to support its first RMB 10 million of revenue, but only 30 to support the next RMB 20 million. Due to operating leverage, as testing volumes exploded in the past three years, the average cost of a test has fallen dramatically, which has led to price falling with it. Most of the industry sources we have spoken with have told us that operating margins of mature labs in the industry have been stable, if not increased in the past three years, even in the face of price deflation. Interestingly, since even “mature” labs can still indefinitely add new capacity (renting new space in separate buildings would still qualify under the existing government certification), there’s theoretically no limit as to how low costs can fall closer to the marginal cost level. As the industry rationalizes and pricing trends turn more favorable (already happened in 2018 with flat pricing), we feel comfortable that mature lab margins should at least remain stable going forward.
The biggest risk in our CTI investment is key man risk. A bet on CTI is a bet on Mr. Shentu’s ability to materially improve profitability and grow the company into a leading global player; much of that upside would be taken away if he left. We believe the most likely scenario that might lead to Shentu’s departure would be a disagreement with the Wan Feng who remains CTI’s chairman and controlling shareholder. Fortunately, with some digging, we were encouraged to hear that Wan, no longer involved in the day-to-day management of the company, has been extremely supportive of Shentu’s new strategy. It was also revealed to us that Shentu and Wan have been close friends for over a decade and Wan had courted Shentu for years prior to his joining. We believe Wan promised to give Shentu at least 8 years of runway to build the company and so far Wan appears very satisfied with Shentu’s leadership. Who wouldn’t be happy to be invested along the Sergio Marchionne of China, anyway? We are excited to see what Shentu can achieve in the next ten years and plan to be long-term shareholders of this growing franchise.
