Murky Water Short YY Full Text: A Virtual Kingdom of Robots

For the past 10 years, we’ve struggled to find the sewers of the world’s capital markets, and the “face-to-face” case has always been one-handed and ubiquitous. After a year-long investigation, a new scam has surfaced, with about 90 per cent of YY’s live-streaming business data falsified. From the beginning, it became clear to us that the YY live broadcast business was a complete fraud. It’s a mirage-like ecosystem. In high-income performance, the actual real performance, only a small part of the disclosure data. Channels that claim to exist independently. In fact, it is mainly controlled by YY in order to carry out continuous false transactions.

A large fan base is almost entirely generated by YY’s internal network virtual robots (contributing 50% of gifts), external virtual robots, and host brush gifts. Therefore, we conclude that 90% of the YY live streaming business is fraudulent, as does YY’s international live streaming business, Bigo.

That’s why we have questions about Baidu’s acquisition of YY Live:

What will Baidu do in the future? Does Baidu really want to grow through an almost all-counterfeiting business?

For the past decade, we’ve thought that “Chinese companies” are losing far more than most investors think. Many people think we’re too cynical, but is YY Live really the only rat in Chinese companies? Or is it a general disregard for U.S. law among executives representing Listed Companies in China?

In April, Baidu’s ethics committee issued a statement on the arrest of the former CFO: “Baidu is determined to crack down on all violations of the law and discipline, and will have zero tolerance for any act that touches the red line of professional ethics.” “

Now, we want to say to Baidu:

“Where’s your due diligence when you buy a completely fake business with nearly 7% of your market value in cash?” How could you not have detected such a large-scale fraud? If Baidu thinks it respects market ethics and the law, let’s see. “

01 YY fakes are available

We shorted the group because we concluded that YY Live was a multi-billion dollar commercial fraud.

We have concluded that YY’s actual business is only a small part of the size of its reporting, and that the user metrics, revenues, and cash balances reported by the company are fraudulent. Our conclusions apply not only to YY Live, but also to YY’s online dating business, Bigo, and a number of other businesses.

Specifically, about 84 per cent of the consolidated revenue reported by YY was fraudulent. Even if we change our underlying assumptions to those that are more favorable to the company, consolidated fraudulent income is about 73%.

We observed three main ways for YY to commit fraud:

1. Robotic users from YY’s own servers – In our data sample, the value of gifts generated by fake users of YY servers accounts for about half of the value of all gifts.

2. The host recycles gifts to the system through another paid user account. As far as we know, those excellent anchors who claim to earn tens of millions of dollars a year are usually paid a fixed salary of no more than 2.5 million yuan (about $350,000) a year.

The union that manages the anchors is part of the plan. The big unions are mainly owned by YY’s former employees, who are clearly defrauding. The financial statements of the five unions published by prC credit bureau show that total revenues in 2018 were only 15 per cent of YY’s claim.

Therefore, we conclude that YY Live, Bigo and YY’s online dating businesses are essentially fraudulent:

1.YY Live’s revenue is fraudulent. YY Live said its live streaming business accounted for about 95.8 percent of total third-quarter revenue.

2. About 80% of YY online dating revenue is fraudulent. Online dating accounts for about 20% of YY Live’s revenue.

3. Approximately 80 per cent of Bigo’s income is fraudulent. Even applying assumptions that are more favorable to the company generates about 60% of false revenue.

4. Bigo’s Singapore parent company changed auditors three times in its first four years of service. Bigo also received three consecutive ongoing business opinions from its auditors in 2016-2018. In August 2019, a few months after YY completed its acquisition, Bigo made a significant recap of its 2017 financial position. These patterns of fact support our conclusion that Bigo is also deceptive in nature.

5. Bigo has rotted since its birth. YY bought Bigo from Chairman Li Xueling, a scam in which Chairman Li took at least $156 million from YY shareholders. But in fact, it was YY who created Bigo, not Li Xueling.

6. Bigo consolidates a Chinese mainland revenue from the company. Based on our preliminary investigation and discussions with former Bigo employees, we believe this is completely false. As of the third quarter of 2020, we estimate that this fraudulent Chinese business still accounts for 13.4% of Bigo’s reported revenue.

We have been studying YY’s business for more than a year and have taken a two-pronged approach. The first step is to collect and analyze macro data on 116 million transactions through automated means. The second aspect is traditional murky water techniques, namely, the review of documents and accounts, clandestine field investigations and human resources.

A closer look at the various alleged high-income anchors shows the prevalentness of fraud in YY:

1. The most successful pop music of the day represents the Modern Brothers in a popular shopping district for free concerts, but the host’s “fans” pay to be present. Today, the Modern Brothers rarely appear on YY Live, but they continue to receive gifts through the platform when they stop broadcasting. Many of these gifts come from YY’s servers. Analysis from 96 randomly selected data showed that 97.9% of the sample gift income may be false.

2. During the COVID-19 city of Wuhan City, our random sample of YY paid users showed that about 87.5% of the gift revenue may have been fraudulent. Most fraudulent paid users are connected to the YY server’s mobile device id (IMEIs), while a significant proportion of Wuhan paid users display IP addresses that inexplicably jump from one city to another during the blockade.

3. Star anchor Lao Li is a scam in itself. “Old Lee” was the most-gifted anchor in YY Live’s 2019 annual competition. However, our data show that at the December event, Lao Li’s mobile device was shared with his biggest contributor, giving away $7 million in gifts, or about 40 percent of the total. On top of that, the devices associated with him sent more gifts than Lao Li received, exposing a network of related accounts that recycled gifts and inflated YY’s total revenue.

Top Bigo anchor RCT’Khan is said to receive $50,000 a month in gifts, but all he does is sit at his desk and do paperwork. Even as his cameras were aimed at the walls and ceilings, the traffic in his live room was full. Until recently, he acknowledged that the contributors to the gift were himself. (His other major contributors are almost certainly part of a gift recycling network.) )

Through extensive interviews with senior members of the YY ecosystem, we concluded that YY Live and Bigo’s gift income were in fact fake.

Murky Water Short YY Full Text: A Virtual Kingdom of Robots

02 YY Live 90% of its revenue is false

We analyze and sample YY user data and gift revenue in three ways to estimate the proportion of paid users on YY Live and their associated revenue. We conclude that using these three methods, about 90% of YY Live’s gift revenue is fake.

The first is to analyze the nearly 100,000 paying users tracked by XHL. We found that almost half of all YY gifts can be classified by value as fake users from YY-related. We take a set of fake user core groups containing YY IP addresses in the data set and track IMEI shares that radiate out of that set.

The second method of false user identification is to randomly select 96 Wuhan paid users during the COVID-19 lock-up period in China. We found that 87.5% of these Wuhan paid users were obviously false users.

In the third method, we took 96 paid users of modern live streaming, and as mentioned above, about 97.9 percent of the modern brothers’ sample were fake users.

We further verified 90 per cent of the previous false income estimates and compared the income reported by the head union to the China Administration for Industry and Commerce (SAIC) with YY’s claims. We found an 85.9% difference between the YY income of the top five unions and their income in China’s credit report.

Our data analysis and collection methods have two components. With google Chrome web development tools, we are able to track 88 data points per transaction, including paid user name, paid user YY ID, transaction time, gift name, gift unit price, gift ID, number of gifts, recipient name and YY ID, paid user IP address, and device IMEI.

Device IMIES tells us that about half of YY live gift revenue comes from YY’s own servers. The device IMEIs also showed us the universality of gift round trips on YY Live. We use a third-party data analytics service owned by YY Insider, called XHL, to track the number of gifts given and received per account. (The founder and owner of XHL is the founder of YY and remains a shareholder in two of its key VIE.) )

For Bigo, we developed custom code to take advantage of mobile apps and collect Data on Bigo’s top 50 anchors and their paid users. Starting in October 2019, we will use the Google Chrome web developer tool to collect data from YY Live. From November 27, 2019 to February 4, 2020, we expanded our data monitoring to include more than 115.6 million gift transactions sent by 1.2 million paying users to 207,000 anchors.

Over the next few months, our researchers excavated this vast database to check the connections between anchors and paying users, by inferring, and examining details of unconventional transactions with sufficiently fine data (i.e., robot searches during the 2020 COVID-19 lock-up in China).

We also use XHL’s database to check the total gift revenue and gift details of more than 180,000 YY paid users, more than 120,000 anchors, and connect with their respective channel owners.

Our estimates of revenue fraud are also based in part on XHL data. We believe that XHL data match YY’s internal accounting data for two reasons.

First, the founder of XHL, Jin Cao, is the co-founder of YY, was a key executive of YY, is still a shareholder in YY’s two VIE (Beijing Tuda and Guangzhou Huayu) and is still the legal representative of VIE Guangzhou Dorwang. Second, Tiger Tooth, a former subsidiary of YY, has used XHL data as evidence in chinese court proceedings to demonstrate the authority, independence and reliability of its data monitoring platform.

The macro data sources we provide for Bigo are room monitoring and data collection technologies for live traffic, as well as third-party application analytics databases.

In December 2019, we reviewed data on 964,000 YY Live paid users tracked by XHL. Our analysis found that YY controls 24.9% of paying users. Through further analysis, these users account for 48% of our gift revenue sample.

These YY-controlled paid user groups can be identified by cross-referencing our data analysis and paid user information in the XHL database. We define paid users under YY control as: (a) paid users who display YY’s local server or internal network IP when sending gifts, and (b) mobile device IDs (IMEI) that are directly or indirectly linked to paid users under the YY control channel owner.

We received paid user information from XHL, including 964,000 paid users, for a total of $687 million. The number of paid users who met our definition was 241,000, or 24.9%. The reward amounted to $330 million, accounting for 48% of the total reward.

There were 1,382 paid users who showed two YY-controlled behaviors. Of these, 300 paid users had gift information data on XHL, indicating that they had sent $17.5 million in gifts, equivalent to 2.5 percent of all gifts in the XHL data set. This category, although very small in absolute terms, allows us to find two other, larger categories that share IMEI with these users and are closely associated with YY, resulting in the exposing of fake users on the internal network and fake users of YY servers.

The internal network false user is the paid user who displays the IP address of the YY internal network, to On the other hand, the fake users of the YY server are paid users with the local host server IP (

The IP addresses displayed by these fake users can only come from within YY. However, these internal IP addresses are usually blocked by normal IP. Fake users rarely disclose these ips: We suspect that YY often uses VPNs or other software to disguise transactions from within YY.

ImeIs are shared by 204,000 paying users with the above-mentioned false users of the YY intranet and the false users of the YY server. Among them, the XHL gift data of the paid user with the domain name 010.606 recorded 91.1 million yuan of gifts, equivalent to 13.3% of the gifts in the XHL data set. We consider this part of the paid user to be a fake user. To some extent, these fake users are also anchors, and their channel owners are under YY control.

In addition, after data weighting, the ID of 303,000 unique paid users is shared with the host IMEI, which belongs to the owner of the YY-controlled channel. This includes 131,000 paying users with XHL gift data. The 131,000 (13.6%) users gave $222 million in gifts, equivalent to 32.2% of all gifts in the XHL data set. We think these are also fake users.

Combining the gift contributions of these fake users, we found that 48.0% of the above live gift revenue was fraudulent based on IMEI sharing alone.

We determined below that half of the gift revenue from live traffic is almost entirely fraudulent, as well as through additional research techniques. We contacted two paid user samples, which showed that the vast majority of paid users were bots and therefore fake users, making fraudulent gift transactions and showing signs of YY control.

In the table below, we show paid users and their total gifts by category. Starting with fake users of YY servers and fake users of YY’s internal network, we gradually expanded, sending 330 million yuan (48.0 percent of the total) in gifts to 241,000 users (24.9 percent of the total).

While we’ve confirmed that a quarter of paying users are YY-related, that doesn’t mean the remaining three-quarters of paying users and half of gift payments are real. Conversely, as we showed below, most of the remaining paid users have different kinds of links to YY that make them fake users.

During the National Day holiday in October 2020, YY launched a new promotion called “Platform Angel”. The new anchor will receive gifts from paid users controlled by YY, or “angels,” as a reward for regular broadcasts.

Murky Water Short YY Full Text: A Virtual Kingdom of Robots

Platform Angel’s gift-giving time is so precise that it reaches the performer’s screen every day in the same second. Platform angels often think of themselves as coming from the platform, and their precise gift-giving time makes them easy to identify.

Murky Water Short YY Full Text: A Virtual Kingdom of Robots

We collected information on 82 “platform angels” and checked “platform angels” paid users against our database. We found that only 26, or about one-third, of the YY-controlled user ID records identified through IMEI sharing.

The figures show that behind every YY-controlled false user we identify through IMEI sharing, it is likely that two YY-controlled fake users have not been identified through the IMEI method. This coincides with the inference that about 90% of our YY live gift revenue is false.

Next, we surveyed YY users in Wuhan to determine if they might be robots. Due to objective factors, Wuhan became a closed city, during which it was almost impossible for non-robot users to move locations. We searched the records for paid users of Wuhan IP addresses, which actively gave gifts at the beginning of the city closure period. A random sample of these active Wuhan paid users showed that about 87.5 percent of them were fake, depending on their transaction details and gift-giving behavior patterns.

If a paid user is considered a false user, the paid user in Wuhan must meet the following fraud characteristics:

Share IMEI with devices connected to false users of YY internal networks

Make IP jump from one geographic location to another

Use IP addresses in a large block of IP to send gifts to controlled channel hosts (i.e., from robot farms)

Send most gifts to the host on the YY Control channel

or other characteristics that exhibit fraud or control behavior.

Murky Water Short YY Full Text: A Virtual Kingdom of Robots

03 The top five unions have an income gap of 85.9 per cent

YY claims that its top unions will earn $1.1 billion in 2018. We have access to local credit reports from these top unions, which show that their total revenue is only $156 million, which means that 85.9 per cent of their revenues are overstated.

According to YY, Yujia, Huashe, Wudi, Chinablue, and IR are the top five unions by revenue in 2018. YY claims to have revenues of more than $1.1 billion in 2018. Below is a screenshot of a video posted on YY’s website that says “2018! YY’s five largest unions earn more than $1.1 billion. “

Murky Water Short YY Full Text: A Virtual Kingdom of Robots

However, we have obtained china credit reports from these unions, showing that their total revenue in 2018 was only 156 million yuan. There is an 85.9 per cent gap compared to YY’s statement. We believe that, based on the review of legal disputes in China, the anchor was paid by the union and our interviews with major union owners, we believe that the credit report revenue figure is the total amount paid to the anchor. Below, we compare YY’s SAIC revenue with YY’s claimed income:

Murky Water Short YY Full Text: A Virtual Kingdom of Robots

In the fourth quarter of 2019, these five unions accounted for about 28.0% of YY’s live streaming revenue.

According to XHL, the five companies are still head unions, and their revenues as a percentage of XHL’s YY live streaming revenue for the fourth quarter of 2019 are as follows:

Murky Water Short YY Full Text: A Virtual Kingdom of Robots

We believe that the 85.9 per cent gap between the $156 million in revenue reported in the credit system report and the $1.1 billion claimed by YY implies large-scale fraud. This approach confirms our conclusion that only 10% of YY’s income is real.

Murky Water Short YY Full Text: A Virtual Kingdom of Robots

04 Rebates and fraud are prevalent

We also found that some of the top YY anchors received most of their gift revenue from their own and YY-controlled robots. One of the top anchors still receives a steady stream of gifts outside of her showtime, much like the robot activity we saw on the Modern Brothers channel. These findings are the basis for our false judgment that about 90% of YY live streaming revenue is made.

Liu Yuning, the lead singer of the “Modern Brothers” band, was live-streaming in Dandong City when many passers-by were stationed to listen.

Despite the hype, we learned from the manager of a YY channel in Dandong that YY paid someone to watch the free show in person.

Even more outrageous things happen in the range of gifts received by the modern brothers. We found that YY has created robots in 58 countries to reward the modern brothers’ live broadcasts.

The robots are likely to make up the majority of the 323,000 paid gift-giving users who visited the band’s channel in January 2020: because we found that almost all of the “so-called paid users” we sampled were actually fake users.

We found that 97.9 ± and 10% of modern brothers’ paid users may be fake. To determine the robot’s activity, we randomly surveyed a random sample of 96 modern brothers fans who gave gifts on the modern brothers’ January 30, 2020 show.

We chose this time as a sample because we believe that the people who give gifts during the show are real. The Modern Brothers were chosen because they are unique in the YY ecosystem, and the owners of the Modern Brothers YY Channel are themselves. As a result, there is virtually no fraudulent power at stake for third parties.

In this sample, we identified false users based on the pattern of regular and repeated gift-giving from them during the days when the “Modern Brothers” were suspended. Some paying users who gave gifts to the modern brothers also showed IP addresses connected to YY servers, further indicating that they were YY-controlled robots.

In a month of particular success for the Modern Brothers, the Modern Brothers ranked 97th on YY and earned $2.9 million in gift revenue. However, YY seems to have hired fake users to get the ranking.

Data from XHL and the YY app show that robots have been running their fixed patterns from July 2020 to September 2020. For example, on July 22 and 23, 2020, the Modern Brothers did not perform, but the channel received $2,631 from 152,000 paying users and $4,007 from 224,000 paying users.

Murky Water Short YY Full Text: A Virtual Kingdom of Robots

Although the normal live broadcast time for the modern brothers is from 7 to 10 p.m., the purple lines shown in the XHL gift transactions below indicate that fake users are sending small gifts 24 hours a day, indicating the robot’s activity time. The images below on the XHL dashboard show data from July 22, 2020, when the modern brothers were not broadcast live.

Murky Water Short YY Full Text: A Virtual Kingdom of Robots

In addition to the modern brothers, some head anchors have a similar situation.

Lao Li, a top anchor at YY, almost all gifts come from YY official related or himself. We found this by analyzing the data for its channels. Data show that in December 2019, eight device IDs associated with Lao Li’s own device contributed more than 80% of the revenue. This device connection is combined with a false user controlled by YY, and overall, the amount the user network rewards the anchor is much higher than Lao Li’s income.

This is YY’s income inflation plan. A common feature of YY head anchors is that false users who share IMEI links to a large number of YY controls.

Lao Li, a former security guard, bened famous via YY Live. Every day, Lao Li hosts variety shows in YY, joking with other anchors and fans. Lao Li calls his 8.4 million followers Li Jiajun, and each of his videos has hundreds of thousands of views. As the third-ranked anchor on the YY platform, Lao Li receives tens of millions of dollars a year in rewards. In December 2019 alone, 28 million yuan (40 percent of the gifts of the month) came from one of Lao Li’s top Tuhao Pingshenshens.

However, our survey shows that Pingshen actually shares Lao Li’s IMEI. We found that 14 other paid users, including eight head-paying users, were also associated with Lao Li. Over a four-month period, the eight accounts far exceeded the rewards that Lao Li received.

Overall, we believe that at least 85% of the gifts reported by XHL in the December 2019 annual competition came from fake users of YY. Our analysis shows that of the $13.7 million in such gifts that XHL obtained in-depth data, YY-controlled fake users contributed a total of $13.1 million, or 96% of the total. We believe that most of the paying users who paid for Lao Li’s award in December 2019 were fake users who gave gifts several times when Lao Li stopped broadcasting.

Murky Water Short YY Full Text: A Virtual Kingdom of Robots

To us, Lao Li is more of a big lie.

Murky Water Short YY Full Text: A Virtual Kingdom of Robots

05 YY China financial data anomalies

We obtained the credit report of YY’s onshore entity and compared it with YY’s filings with the Securities and Exchange Commission (SEC) and found many differences. These variances amount to hundreds of millions of dollars, including a significant cash shortfall in local corporate finances compared to what YY reported to the SEC. This difference, particularly in cash balances, is a trace of manipulation or fraud.

For example, YY’s 2018 filings with the U.S. Securities and Exchange Commission show long-term deposits of $1 billion, as shown in the table below. In our view, this total is likely to be fictitious or inflated significantly, as the main YY onshore entities do not contain these long-term deposits. This difference can be seen in a rough comparison of the two sets of financial statements.

Murky Water Short YY Full Text: A Virtual Kingdom of Robots

In addition, we believe that YY transferred about 1.326 billion yuan to its wholly foreign-owned enterprises (most of which will be illegally transferred to its wholly foreign-owned enterprises), or that the cash of wholly foreign-owned enterprises does not exist at all.

It is calculated that by the end of 2018, YY’s cash balance of wholly foreign-owned enterprises was 1,298 million yuan, the vast majority of which (about 1.16 billion yuan) was overseas. However, given that YY did not show substantial dividends or loan flows overseas in SEC filings, we believe that YY violated China’s capital controls by transferring the vast majority of RMB1,326 million, or that the cash balance of RMB1.16 billion of offshore wholly foreign-owned enterprises is false.

The table below shows the total book cash balance of YY onshore and offshore wholly foreign-owned enterprises of RMB 1,298 million without tiger teeth.

Murky Water Short YY Full Text: A Virtual Kingdom of Robots

To determine the split of the cash offshore, we reviewed YY’s China Credit Report. The credit report shows that by the end of 2018, the cash of wholly foreign-owned enterprises that do not include Tiger Tooth onshore will not exceed 138 million yuan. The table below shows only the cash position of wholly foreign-owned enterprises that are mainly onshore and do not contain tiger teeth

Murky Water Short YY Full Text: A Virtual Kingdom of Robots

Due to the small onshore cash balance at the end of 2018, we know that almost all of the $1.16 billion of the $1,298 million cash balance will come from overseas.

See below:

Murky Water Short YY Full Text: A Virtual Kingdom of Robots

From these calculations, we can infer that the cash transferred to YY wholly foreign-owned enterprises during 2018 should essentially go to overseas wholly foreign-owned enterprises. Because onshore wholly foreign-owned enterprises not only have low cash balances, but also do not show the possibility of major cash transfers of large-scale financing activities.

To calculate the amount of YY transferred to an offshore wholly foreign-owned enterprise, we reviewed YY’s cash flow statement.

In the table below, we first summarize the total cash generated by YY VIE in 2018 at $3.38 billion. We then add this figure to YY’s variable interest entity cash at the end of 2017, and if the funds are not transferred, the cash on YY’s balance sheet should be $5.324 billion, which is a serious deviation from YY’s cash balance of $3.997 billion at the end of 2018 without tiger teeth.

The $1.326 billion discrepancy includes transfers from non-tiger-toothed VIE to non-tiger-tooth WFOE, as they disappear from VIE’s cash balance in 2018. YY did not disclose the corresponding material dividends or loans to offshore entities in a 2018 filing with the U.S. Securities and Exchange Commission . However, as we have concluded in the table above, almost all of the transferred cash appears to have gone exclusively to wholly foreign-owned enterprises overseas that do not contain tiger teeth, and our transfer calculation of $1,326 million is as follows:

Murky Water Short YY Full Text: A Virtual Kingdom of Robots

These unsealable offshore cash transfers have raised concerns about the authenticity of YY’s offshore cash balances.

Murky Water Short YY Full Text: A Virtual Kingdom of Robots

06 The acquisition of Bigo was a surprise scam

We believe YY bought Bigo in 2010 for $1.45 billion in cash and stock, a scam that actually bought another company from the chairman. We got Bigo’s local documents in Singapore, which show that it wasn’t Li Xueling who created Bigo first, bigo who was later transferred to Li Xueling.

We have also found that Bigo has retroactively increased Chinese mainland’s revenue, which we believe is largely false, which makes it seem that YY’s acquisition of Bigo was a very wise decision and justified Chairman Li Xueling’s high salary.

We believe Bigo was trying to defraud investors from the start. Since 2015, management has been misleading investors and telling the public that it was Li who founded Bigo. However, this is a lie: Singapore documents show that Bigo Technology Pte.Ltd. was registered on September 11, 2014, with only YY’s BVI entity, Duowan Entertainment Corp.

Murky Water Short YY Full Text: A Virtual Kingdom of Robots

YY’s acquisition of Bigo was the key to a scam that, after several rounds of financing, resulted in the majority of YY’s second offering in 2017. Most of the purchases were paid directly to Li in cash and stock, which included 38.32 million shares of YY B common stock issued, with a total value of about $156 million.

In addition, he is likely to receive a portion of the $343 million in cash and a portion of YY’s 314 million Class A common shares. YY has also used the huge amount of money paid to insiders to generate significant value gains in 2018 and 2019, generating significant book profits.

Even in Bigo’s early years, local audits revealed some information to us. In the first four years, Bigo replaced the Singapore auditor three times. Shortly after the re-acquisition of a 100% stake in Bigo, YY management stated that, despite Bigo’s overall losses, Bigo Live was strong enough to support the initial development of its other businesses.

But Bigo’s Singapore auditors objected, warning bigo’s continued operations in three annual reports from 2016 to 2018. Furthermore, Bigo made a significant recap of its 2017 financial position in August 2019, a few months after the acquisition was completed.

With its financial position at stake, Bigo continues to pack up in preparation for the acquisition. In local Singapore documents disclosed in 2018, Bigo reiterated its financial position by adding a Chinese variable interest entity, Guangzhou Park, which contributed an additional $110 million in revenue.

However, as we have detailed below, we believe that almost all of Bigo Chinese mainland revenues from the market are fabricated. In addition, when we compared the details published by the local audit in Singapore with the financial position of Bigo China Variable Interests entities in the credit reports issued in China, we found many illogical differences.

Bigo’s financial fraud has resulted in a significant fair value revaluation gain for YY, further increasing the value of YY shares. YY’s 2018 round D investment in Bigo resulted in a profit of $989 million for the current period, equivalent to 46.7% of the reported net income for 2018. The acquisition of Bigo in 2019 further enhanced YY’s financial position in the first quarter of 2019, with a profit of $2.67 billion, or 72.2% of net income for 2019.

As the biggest protagonist of Mr Li’s scam, Bigo has been hiding that a third of its business at the time of the acquisition did not come from overseas, but Chinese mainland. We believe that this Bigo’Chinese mainland income is almost entirely false.

In addition, we believe that these Chinese revenues enrich Bigo’s books through fictitious sales, allowing the business to be acquired at a higher acquisition price, and ultimately Benefit from Li Xueling.

The June 2018 announcement of Bigo Round D financing and the completion of the acquisition of Bigo in March 2019 focused on Bigo Live and Likee, portraying Bigo as a $1.45 billion global live streaming company. The statements highlighted the exclusion of China from Bigo Live’s live-streaming platform, but did not mention that a large portion of Bigo’s revenue comes from China’s audio-streaming app.

However, according to a 6-K report filed with the Securities and Exchange Commission in June 2019, a $166 million in sales came from China, about one-third of Bigo’s total revenue for 2018.

By comparison, there is a nearly similar figure, with $159 million in revenue coming from audio streaming, which means that the source of additional revenue in China is Chinese mainland’s audio live streaming app Hello, while others may come from smaller applications.

As we have described, we suspect that Bigo’s 2017 revenue and 2018 revenue from The Hundred Orchards and related audio streaming business are likely to be false or grossly inflated. According to the statistics we found, about 80% of Bigo’s income is false.

Bigo’s former executives say most of the users are fake, and the robots are directed by Bigo’s headquarters in Singapore or YY’s headquarters in Guangzhou.

Our investigators interviewed a former executive of Bigo India and asked about the way Bigo operated. About 60 per cent of the people in Bigo’s main studio are robots, which are decided by the Chinese teams in Singapore and Guangzhou, the executive said.

He thinks it’s necessary to use robots, giving a reason for the response: “When you first started broadcasting, I couldn’t let you see a blank space in the live room, we had to give you some fans.” “

A former Bigo Middle East executive described the proportion of fake users on Bigo as “for Bigo, only about 30-40% of the audience is real, and about 60-70% of the audience is robots.” “,

Bigo can raise or lower the proportion of robots on the platform, the former executive said: “The proportion of fake users is probably much higher than I said.” For example, in a live music channel, the first 5,000 viewers are usually robots.”

Finally, this was confirmed by a former Bigo executive in Southeast Asia, who told us that 70 per cent were typical of robots on the Bigo channel.

Bigo has a bean recycling mechanism, and Bigo uses unions and host “families” to manage the distribution and distribution of Bigo beans on their networks, creating dynamics similar to YY live streams without the need for implicit control of channels.

Bigo anchors are contracted users, and if they want to be paid, they need to meet a certain target “bean” quota and follow a regular schedule for the game ranking system.

The contracted anchor must keep the daily broadcast between 30 minutes and 2 hours. This ensures that they log in as many days as possible each month, combine their payment rewards with third-party application analytics company metrics, and then rank applications based on engagement metrics.

Bigo has long given the impression of a highly engaged number of users, when in fact most of the users involved are workers working on a set of scheduling parameters.

According to the former Bigo manager we interviewed, the revenue quota system forces the anchor to keep the beans in Bigo. Some anchors can be paid, but others, such as in the United States, Morocco and other countries, bigo’s monthly salaries are paid with beans, which helps Bigo reduce the amount of beans circulating in the ecosystem.

Bigo also offers performers a recycling option that allows them to use beans without leaving the platform.

Performers can help each other achieve their revenue goals by buying discounted beans and gifts for each other in tit-for-tat deals. In this way, when their quotas are low, anchors often seek help from their union or family members to support them and exchange discount gifts.

They usually exchange soy products with other members during the last month of the event. This Bigo activity has created a strong liquidity currency activity.

Sometimes their income far exceeds their total income. However, because Bigo not only changed the identity of the fake user in the head, but also the statistics of the rewarder’s account, it is difficult to make accurate calculations.

Murky Water Short YY Full Text: A Virtual Kingdom of Robots

We believe that the intention behind changing head user data is very obvious, in order to prevent others from discovering the scale and outrageous extent of the Bigo scam.

Bigo also owns a Chinese mainland, but is not mentioned in his earnings conference call and annual filings. We think it avoids mentioning Chinese business because it is essentially an empty shell because its application does not show real activity.

YY reported that Bigo’s asia-China revenue for 2018 was $166 million, of which the audio broadcast portion was $159 million. Bigo’s core products, Bigo and Likee, target video rather than audio and streaming, both of which are focused on markets outside China.

In 2018, Bigo’s financial data in Singapore showed revenue of $538 million. China’s share of revenue during the reporting period was as high as 30.7 per cent of total revenue, a significant amount for a market such as Singapore.

The following table shows the geographic breakdown of Bigo’s revenue, filed on an annual basis:

Murky Water Short YY Full Text: A Virtual Kingdom of Robots

Through our investigators on the mainland and conversations with former Bigo employees, we determined that Chinese mainland people could not download Bigo through the normal App Store. Bigo’s main business in China comes from overseas Chinese signing up and Chinese anchors in foreign countries, so we don’t think the Chinese users of Bigo Live app are anchors and won’t generate substantial revenue for Bigo.

YY’s disclosure convinced us that its Revenue in China came from an app called Hello.

YY’s June 2019 semi-annual report states, “In March 2016, The Park Network launched a live audio APP Hello, an app focused on business in China. “

Our inspection of Bigo China VIE 100 Orchards shows that it Chinese mainland applications and websites operating on the site, most of which have lapsed. The main APP is Hello Voice, a voice social network app with a small market share.

Therefore, we believe that this portion of the revenue Chinese mainland mainly comes from Hello, as the revenue from audio streaming and audio broadcasting is essentially the same.

We found that Bigo’s potential revenue in China has been growing for Bigo, according to the filings. In 2019, YY began disclosing its share of China and non-China live streaming revenue in its quarterly results. YY’s non-Chinese income comes mainly from Bigo and a small amount from Huya and Hago. Assuming that Bigo and its consolidated subsidiaries account for 100% of YY’s non-Chinese revenue, we can calculate Bigo’s lowest onshore China income by subtracting all of YY’s non-Chinese income from Bigo’s total revenue.

From the recently disclosed YY data (which does not now include Huya), we can see that Bigo’s hidden income in China is relatively high, accounting for 13.4% of revenue in the third quarter of 2020, reaching 456 million yuan, an increase of 27.4%.

Murky Water Short YY Full Text: A Virtual Kingdom of Robots

Bigo’s onshore income appears to be limited to VIE’s 100 Orchard-owned apps and investments, which should be largely entirely from Hello and its associated APPS.

But when our investigators spoke to YY anchors and virtual currency traders, they found that they didn’t know much about China’s APP Hello voice.

So far, we have found little evidence that Hello is important to YY in terms of brand name, number of people online, downloads, mau or revenue in China.

As we mentioned below, we wonder if Hello and its associated APPS can really bring in much revenue.

Our investigators conducted a preliminary investigation of Hello and found that there are very few active users. This increases our judgment that the app has little revenue.

In addition, Hello was created by YY, not Bigo, and we discovered it by analyzing the information disclosed by the application.

We monitored several live rooms and found them to be a joke. In our view, not only is the content bad, but it is also mostly lacking in participants and anchors. When we browse the Hello app, we see very few active users, and even in a full-load room, the interaction between the anchor and the user is still minimal, with little or no gift.

Most users seem to be robots, don’t participate in any activities, and neither join nor exit these dead rooms. The following two examples show that the top-level broadcast room on Hello Voice is running at full capacity, but there is no message or language interaction at all.

Murky Water Short YY Full Text: A Virtual Kingdom of Robots

We believe that Hello and its associated applications exist for the convenience of management, who may find it easier and cheaper to generate revenue onshore than offshore.

We further believe that these APPS are designed to help Bigo’s profits increase rapidly, increase its contribution to YY, and generate sustained revenue.

In addition, this increase in income is a way for Li Xueling to raise the valuation of Bigo, by incorporating the potentially false income of the hundred orchards into Bigo, so that more benefits can be obtained from YY.

As of August 2018, hello used YY customer service emails before YY acquired Bigo: This is a red flag for a small company, especially for Bigo in 2018.

Bigo is expected to earn more than Rmb1bn this year. Explain that some of YY’s management has been integrated and are under YY’s common control.

We found more evidence of this actual control in a January 2017 blog post about how to set up an account on Hello Voice and earn a commission from streaming media.

According to documents filed in Singapore, Bai Orchard and Hello Voice should be completely separated from YY from 2016 and under Bigo’s control.

However, Hello Voice users are encouraged to bind their YY account to their Hello voice account and use YY’s payment system.