In most markets, corporate fraud occurs in isolation. It involves just one company and perhaps its associates.
In May 2017, the Hong Kong activist investor David Webb cast the spotlight on the “Enigma Network,” which contained 50 companies in a complex stock manipulation scheme. It was a breathtaking example of syndicated corporate fraud on a grand scale, and the largest exposed in Hong Kong. The following year, Webb alleged to have found another network, this time involving both private and state-owned entities.
Is the sophistication of corporate fraud in Hong Kong a step above other markets? It's possible, but that's not the subject of this article. The objective of this piece is to examine the 50 companies in the Enigma Network and how each stacks up on our accounting Manipulation Risk Analyzer.
For the uninitiated, Transparently.AI’s MRA is an AI-powered tool that has been trained to detect red flags in a company’s financial statements. The system assigns to a company a 0-100% risk score and an A+-to-F rating which together indicate the extent to which the company in question is manipulating its accounts.
What the tool indicates for the Enigma Network companies is nothing short of horrific. We will come to that. But first, some background.
Collusion and corporate fraud
Corporate fraud can involve dozens of companies acting in collusion as a syndicate, typically to manipulate voting rights, share prices or, of course, financial statements. Collusion offers safety in numbers, making it easier to engage in fraudulent activity and making it more difficult for prosecutors to prove wrongdoing.
When a single company engages in fraud, there are limits as to how much it can manipulate its accounts by exaggerating revenue, or hiding expenses and debt. A firm is always constrained by the size of its balance sheet, by the amount of cash it generates and by its ability to obfuscate transactions.
When multiple companies become involved, manipulation becomes easier because fraudulent entries can be spread across multiple entities in small amounts. If the members of a syndicate and their associated companies are seemingly unrelated to a company engaged in a deception, it will be virtually impossible for an auditor to identify these related party transactions.
If, for example, 40 companies are willing to exaggerate their expenses slightly to exaggerate the sales or lower the expenses of a single target, one can imagine how difficult that might be to trace, let alone prove. If those same companies are further willing to purchase shares of the target company in a pump-and-dump operation, one can readily imagine how easily a syndicate can manipulate share prices.
Difficult to prove
Alternatively, management can use the voting rights of syndicate partner companies to push through actions that will be detrimental to shareholders. It is not unknown, for example, for Hong Kong-listed companies to sell associates or assets to entities owned by the controlling shareholders at prices that might be detrimental to minority shareholders, or even without the knowledge of shareholders.
The sale of Alipay to the controlling owner of Alibaba in 2011 is a possible example of this kind of activity.
Our blog repeatedly demonstrates the difficulty of proving guilt in cases of accounting fraud or corporate fraud more generally. Incriminatory evidence takes years to gather. Much of it is often destroyed and unavailable to prosecutors. Moreover, fraudulent companies weave complex corporate structures to hide fraudulent dealings. As a consequence, even though companies and executives are frequently charged and fined by regulators, admission of guilt is rare and convictions are even rarer.
If conviction is difficult with just one company, how much more difficult to demonstrate guilt if dozens of companies are involved, each with affiliated companies participating. Such a maze would be difficult for a prosecutor to untangle, let alone demonstrate guilt.
Of course if the syndicate companies are domiciled in China, the web becomes doubly difficult to untangle. Hong Kong has no formal treaty allowing the extradition of Chinese nationals. Chinese managers at risk of being charged with fraud in Hong Kong simply have to wait it out in China.
The Enigma Network
Experienced investors in Hong Kong have long been aware of an association among certain groups of companies, especially those engaged in construction or investment. Some of these corporate linkages are benign, some are not. In May 2017, David Webb published an article entitled "Enigma Network: 50 stocks not to own," on his website.
The Enigma Network was a group of 50 connected Hong Kong listed companies with significant overlap in ownership, much of it undisclosed. In 2018, Webb alleged to have found another network, this time involving both private and state-owned entities.
In relation to the Enigma Network, Webb identified numerous companies with dangerously concentrated ownership. The Securities and Futures Commission had previously issued concentration warnings on eight of these.
The report also identified many cross holdings below the normal 5% disclosure threshold among the enigma companies which, “when aggregated, provide significant voting power when the companies seek approval from "independent" shareholders to do something that might not make sense to others.”
Webb’s primary concern was corporate governance and the protection of minority interests. Specifically, he focussed on agency conflicts between controlling and minority shareholders. These conflicts arise because controlling shareholders can expropriate minority shareholders by using firm cash and resources to pursue private interests.
Those familiar with Hong Kong will know that this is a prevalent issue.In the past, some controlling owners have made more money trading their company’s shares than from the company’s operations.
Stock crash
Once the news of the network broke, investors anticipated problems and began to dump enigma companies. Some companies suffered catastrophic share-price losses of more than 90%. A further quarter fell by almost 50%. The fallout led to one of the worst small-cap crashes in the history of the Hong Kong market. Sentiment towards small caps in Hong Kong remains sour to this day.
In total, the market value of the 50 network stocks fell by US$6.1 billion, making it arguably Hong Kong’s worst corporate scandal.
The Webb report led to a joint investigation by the Independent Commission Against Corruption and the SFC. Several stocks were suspended, most notably Town Health International Medical and Lerado Financial. Criminal proceedings continue to this day. Owing to the difficulties noted above, the prosecution has enjoyed mixed success obtaining convictions.
On our count, 11 of the original 50 network stocks failed to produce an annual report for 2016. Some were delisted as a consequence. Others were rolled into other entities and able to obfuscate their accounts. Fourteen companies highlighted by the scandal changed their names to distance themselves from the affair. Seven of the original companies are dead. One remains suspended. Most have morphed to some degree.
Manipulation risk and the Enigma Network
If a company pays no heed to proper corporate governance, there is no reason to expect it will act honestly in other respects. Risk of account manipulation and corporate fraud is always elevated in companies with poor governance, which is why governance is a key risk cluster in Transparently.AI’s Manipulation Risk Analyzer.
Figure 1 illustrates the risk scores and grades for the 50 companies in the Enigma Network. In most cases the risk scores are for 2016, the year before David Webb published his report. Where possible, if there was no report for 2016, we have shown the risk score for 2015 or 2014. If there were no prior reports, we have shown the risk score for the first year where a report was issued. The companies were assessed on the model estimated for Asia-Pacific ex-Japan and so the percentile risk shown is versus companies from this region.
Figure 1: Manipulation risk scores and grades of the Enigma Network companies
Source: Transparently.AI
As can be seen, the risk scores of the network companies were nothing short of horrific.
The average and median risk scores were 84% and 86%, respectively. The median company in the network was at the 96th percentile for risk in Asia Pacific ex-Japan, with a risk grade of “F”.
A risk grade of F on the MRA implies substantial issues with accounting quality and transparency. Many of these companies will have been publicly identified as having serious accounting issues and will have already undergone some form of significant market pressure. accounting quality and transparency.
This abysmal performance illustrates the power of the Transparently.AI system in identifying fraudulent companies.
Only one company in the group had a risk grade of B, but this score was for 2017, came after a name change and, of course, the absence of a financial statement for 2016 suggests that the score for that year was likely not a B.
Five companies had a C- grade. Four of the five had been identified as bubble companies by the SEC. Three failed to produce financial statements for 2016.
Had financial statements been available for all companies for 2016, it seems likely that no more than two or three would have enjoyed a risk grade of D or above.
This abysmal performance illustrates the power of the Transparently.AI system in identifying fraudulent companies.
Space prevents us from delving into the detail of the risk clusters exhibited by these companies but we will note that the system issued red flags for every company in relation to corporate governance. Most showed risk of earnings manipulation. The suggestion is that some of the network companies might have been engaged in more than just the expropriation of minority shareholders.
On the Transparently.AI system, virtually all of the network companies were uninvestable and an investor using the system in 2017 or before would have avoided the names in the Enigma Network.
From Convoy to AGBA
Perhaps the most interesting aspect of the Enigma Network group of companies is that most of these companies had a worse risk score in 2023. This would suggest that there are still problems in the group even though some have changed names and morphed into different business models.
The most interesting example of a morphed Enigma Network company is that of Convoy Global Holdings, which was the financial advisor alleged to have been at the centre of the network.
Convoy delisted after failing to present its 2016 accounts but resurfaced in a reverse merger with Nasdaq-listed AGBA Group Holdings.
In September 2023, AGBA announced that it had appointed Bob Diamond, the former CEO of Barclay’s, as its chairman. In the press release, AGBA described itself as “the leading one-stop financial supermarket in Hong Kong,” and “Asia’s leading investment advisor.” Putting these claims in perspective, in 2023, AGBA lost $49 million on revenue of $54 million. Group assets have fallen from $177 million in 2020 to $67 million in 2023.
Nevertheless, AGBA’s share price surged from 50 cents in March to over $3 in April on news that AGBA was moving into the social media business. AGBA recently made an all-stock take-over of the video platform Triller, which AGBA claims is a rival to TikTok.
Triller has insufficient data but is reportedly being sued by Sony and Universal for non-payment of royalties and has been accused of inflating revenue by former employees: An inauspicious beginning. Still, we watch with great interest. If it plays out, this will be a remarkable deal.
For those with investment experience in Hong Kong, some of the deals undertaken by Enigma Network companies also seemed to offer plenty of blue sky. None came to fruition.
Disclaimer: Views presented in this blog are the author’s own opinion and do not constitute financial research or advice. Both the author and Transparently Pte Ltd do not have trading positions in the companies it expresses a view of. In no event should the author or Transparently Pte Ltd be liable for any direct or indirect trading losses caused by any information contained in these views. All expressions of opinion are subject to change without notice, and we do not undertake to update or supplement this report or any of the information contained herein.
