Correspondence
Letter couriered from Shanghai and confirmed delivered.
The letter described abuses on a jigger (including the death of one crew member and disembarkation of a second for medical care after mistreatment on board, and multiple indicators of forced labor among the ship’s crew) which has transshipped with a Chishan reefer. It also said that Chishan vessels were observed fishing in North Korean waters in violation of UN sanctions.
The Outlaw Ocean Project emailed the Chishan Group had received persons from the Xinjiang region under the Chinese government's labor transfer program and that the company's international customers include businesses in the U.S., where the importation of goods produced from state-imposed forced labor is prohibited. The email asked: Can you confirm or deny if any Uyghurs from Xinjiang were employed to work at Chishan Group factories at any time over the past five years? Does the Chishan Group have any comment or statement to make in response to this email?
Guogang Li, a lawyer for Haibo Seafood Co. and Shandong Haidu Ocean Product Co., sent a letter to The Outlaw Ocean Project:
Dear Ian Urbina:
Please allow me to introduce myself, Guogang Li, a lawyer from Tahota (Beijing) Law Firm (please see Attachment 1 for my biography), entrusted by Haibo Seafood Co., Ltd. (hereinafter “Haibo”, please see Attachment 2 for the Power of Attorney) and Shandong Haidu Ocean Product Co., Ltd. (hereinafter “Haidu”, please see Attachment 3 for the Power of Attorney). I would like to send the following comments below to you regarding the report in the “Outlaw Ocean Project” related to the Uyghur laborers in the seafood processing factories for your consideration.
I. Among the laborers in Haibo and Haidu, there are no ethnic minorities from Xinjiang. Since we were entrusted by Haibo and Haidu in October 2023, we have conducted a cross-check on the employee rosters and payrolls of Haibo and Haidu, as well as physically interviewed the employees at the company, and found that Haibo and Haidu do not employ any workers from Xinjiang and there has never been forced labor in Haibo and Haidu. It would be appreciated if you and your team could visit Haibo and Haidu and conduct the audit to identify whether ethnic minority laborers from Xinjiang exist. Any unverified information or data should not be used to support conclusions or as an application for requesting import restrictions from the federal government of the United States, since this could harm downstream and upstream industries as well as consumers in the United States.
You mentioned the Uyghur Forced Labor Prevention Act (UFLPA) many times, and I used to be the director in China’s Ministry of Commerce (MOFCOM), dealt with counterparts from the United States Department of Commerce and U.S. Customs and Border Protection, and also worked as a trade and customs lawyer at WHITE & CASE LLP in the United States and assisted a number of Chinese companies to respond to CBP’s WRO and the UFPLA. My understanding is that the purpose of both the WRO and the UFLPA is to eliminate any forced labor that may exist in the manufacturing or supply chain of a product through customs measures, and not to penalize a company or even an industry. It would be contrary to the starting point of WRO and the UFLPA if CBP were to impose sanctions on Chinese companies based on the principle of punishing, rather than eliminating forced labor. Otherwise, it implies that the applications to the sanctions have been influenced by certain competing interests and is intended to be used to harm competitors.
II. The SMETA audit report and the SYSCO Quality Assurance audit report respectively found no forced labor in Haibo and Haidu.
In August 2023, an independent third-party auditor conducted a SMETA 2-pillar audit of Haibo on Labor Standards, and Health and Safety at the request of our foreign client. In September 2023, an independent third-party auditor conducted a SYSCO Quality Assurance audit of Haidu. According to the audit, Haibo and Haidu do not engage in forced labor in any aspect of wages, working hours, or recruitment, in accordance with the internationally recognized audit standards.
I noted that you mentioned in your email that social audits have not been able to identify the use of Xinjiang workers in many seafood factories in Shandong and so the impartiality of the social audits is questioned.
My reminder is that SMETA and SYSCO Quality Assurance adhere to audit standards. The audits include individual working hours, child labor, workers’ wages and income, and other internationally recognized forced labor standards. Additionally, a worker analysis is conducted, including the identity of the worker, whether they are local employees or migrant employees, as well as whether they are of Chinese nationality, etc. A SMETA audit does not identify whether the employees come from a particular region or country, such as Xinjiang of China or North Korea. The objectivity and reliability of the SMETA and SYSCO Quality Assurance audits cannot be questioned on the basis of not auditing the identities of Xinjiang and North Korean employees, as that is not the audit standard or content. Moreover, from what I understand, no auditing firm audits employees’ ethnicity or whether they are from Xinjiang or North Korea; instead, they only verify whether the company complies with internationally recognized forced labor standards. I would like to reiterate and emphasize that no forced labor was found in Haibo or Haidu, according to SMETA and SYSCO Quality Assurance audit reports.
III. The UFLPA’s Rebuttable Presumption does not mean that the use of workers from Xinjiang constitutes forced labor, as determined by the 11 indicators of forced labor of the International Labor Organization (ILO).
According to your email, the UFLPA presumes that all imports of products produced in whole or in part from Xinjiang are the result of forced labor by the government, and the presumption also applies to the government’s labor transfer program. In my opinion, your understanding of the UFLPA is incomplete. According to the UFLPA, it is specifically referred to as “Rebuttable Presumption”, which means that despite the presumption that products from Xinjiang involve forced labor, the presumption can be rebutted. No matter whether the product is manufactured in Xinjiang or produced by laborers from Xinjiang, it is not in violation of the UFLPA if it is demonstrated that none of the 11 indicators of forced labor exist.
The rebuttable presumption of UFLPA does not apply to Haibo and Haidu since neither Haibo nor Haidu has laborers from Xinjiang, and the SMETA and SYSCO Quality Assurance audit reports found Haibo and Haidu did not have any of the 11 indicators of forced labor of the ILO.
I understand your concern regarding forced labor in marine fishing and seafood as director of the Outlaw Ocean Project, and we also pay more attention to this. We are looking forward to developing a good relationship with you and hope to improve the understanding and clarify the misunderstanding between us. Moreover, we would like to invite you and your team to conduct an on-site audit of Haibo and Haidu to verify if forced labor exists.
Additionally, Haibo and Haidu reserve the right to take further legal action against any behavior that negatively impacts its reputation.
Thank you. Sincerely,
Guogang Li
Counsel Tahota (Beijing) Law Firm
The Outlaw Ocean Project responded and asked further questions:
Dear Mr. Guogang Li,
Thank you for your letter from November 6, 2023. This letter comes as the latest in a series of exchanges in which representatives at Rongcheng Haibo Seafood Co., Ltd. and/or Shandong Haidu Ocean Product Co. Ltd have claimed that no Xinjiang workers have been at the plants. You can review that back-and-forth for Haibo and Haidu on our website. In prior replies on this matter, we itemized the diversity of evidence indicating the existence of Xinjiang workers at these plants. We resurface that material again here:
Your November 6, 2023 letter says that the Chishan Group newsletter citing Xinjiang workers at their plant was incorrect due to a reporting error on the Chishan Group’s part. Again, the explanation provided is simply not credible. We must make clear that our evidence does not rest upon a single document, but on dozens of sources. We found articles in the state news media from 2021, in both Uyghur and Chinese, that describe government labor transfers to Rongcheng Haibo. Three newsletters published by the Chishan Group also describe the use of Xinjiang labor at Haibo, and also detail government labor transfers to Rongcheng Haibo. An additional four newsletters describe Xinjiang workers at other Chishan Group enterprises, including Shandong Haidu, another processing plant that exports to the United States. Detailed in these articles are high-level meetings in 2020 and 2021 between corporate executives, including Rongcheng Haibo senior management, and Chinese government representatives.
In February 2021, the Chishan Group published a newsletter in which an article celebrated the arrival of Xinjiang workers at Rongcheng Haibo. The article included an image of the workers arriving at the site.
In May 2021, a Chishan Group newsletter said the company had held meetings with government officials in which the transfer of additional laborers from Xinjiang was discussed. Present at the meeting was Haibo's general manager, as well as representatives of a Xinjiang Zhongtai Group subsidiary, a company on the Uyghur Forced Labor Prevention Act Entity List for "working with the government of Xinjiang to recruit, transport, transfer, harbor or receive forced labor or Uyghurs, Kazakhs, Kyrgyz, or members of other persecuted groups out of Xinjiang." The article said the Chishan Group was "looking forward to the arrival of laborers from Xinjiang" and said that the meeting concluded with an agreement between the Bayingolin Mongol Autonomous Prefecture Human Resources and Social Security Bureau and the Xinjiang Zhongtai Group subsidiary to facilitate the transfer of "qualified Xinjiang migrant workers to Chishan enterprises as soon as possible."
In October 2021, a representative of Kashgar’s Human Resources and Social Security Bureau and Shandong’s ‘Xinjiang Aid’ program visited Chishan Group to “investigate the employment environment” of the company, alongside a representative from a state-owned Kashgar company and Xu Wu, director of the Public Employment and Talent Service Center and member of the Human Resources and Social Security Bureau of Rongcheng. Xu Wu oversees the Xinjiang Aid Labor Service Workstation, located within the Shidao Management District, which has provided Xinjiang labor to Chishan Group as well as other Rongcheng seafood enterprises, according to state media.
In September 2022, a Chishan Group newsletter said that Rongcheng Haibo implemented special kitchen arrangements for Xinjiang workers.
In addition to this, four other newsletters reference the transfer and/or accommodation of Xinjiang workers at other Chishan Group companies, including Shandong Haidu, another major exporter to the United States. Again, the practice of cooperation with government authorities to facilitate the transfer of Xinjiang workers is described.
Beyond state media and company materials, we've obtained footage uploaded by Uyghurs to social media that shows them working at Rongcheng Haibo. Video clips uploaded by seven users between March 2022 and June 2023 depict scenes that are verifiably shot from within Rongcheng Haibo, and around the company compound, according to our analysis. The videos include images and clips of the Uyghurs in employee overalls on the processing plant floor.
There is also a clip of a Xinjiang worker at Rongcheng Haibo that was uploaded on the same day in 2022 that SGS undertook a SMETA audit of the site. Our evidence further indicates that social audits have consistently failed to detect Xinjiang labor at numerous Shandong seafood plants over a period of several years.
In this reply we would direct attention to a video which focuses on Haibo and Haidu and their parent company, the Chishan Group, and seeks to explain the methodology and evidence further. Please take a look.
Below are our replies to specific points raised in your most recent letter.
Your letter mentions: “Among the laborers in Haibo and Haidu, there are no ethnic minorities from Xinjiang. Since we were entrusted by Haibo and Haidu in October 2023, we have conducted a cross-check on the employee rosters and payrolls of Haibo and Haidu, as well as physically interviewed the employees at the company, and found that Haibo and Haidu do not employ any workers from Xinjiang and there has never been forced labor in Haibo and Haidu. It would be appreciated if you and your team could visit Haibo and Haidu and conduct the audit to identify whether ethnic minority laborers from Xinjiang exist. Any unverified information or data should not be used to support conclusions or as an application for requesting import restrictions from the federal government of the United States, since this could harm downstream and upstream industries as well as consumers in the United States.”
Reply: Your comment is in the present tense. It says that no Xinjiang workers are at these plants now. The pertinent question, however, is whether any workers from Xinjiang have been employed at either plant in the past 5 years. This question was the focus of our investigation. You also, separately, mention that there has never been forced labor in either plant. It would appear that to come to this conclusion, you interviewed workers to see whether any of them were forced labor. Is this, in your view, an effective method for investigating the possible existence of forced labor? Can you identify who conducted the interviews, the questions they used, and whether managers were present? Is it your view that interviewing present workers would give you an indication of whether past workers were part of a forced-labor program?
As for your generous invitation to visit Haibo and Haidu, thank you. But it is unclear what journalistic purpose that would serve. Though kind, this type of invitation highlights a larger misunderstanding. A pre-scheduled visit (be it by hired auditors, western journalists or a lawyer hired by the plants) is the opposite of a spot check or true inspection. Such a visit would be a public relations event, not an effective means for assessing true conditions. Researchers have pointed out repeatedly that pre-announced visits are among the core flaws in how social auditing is conducted. Obviously, advance notice enables managers to show what they want the world to see.
Your letter mentions: “My understanding is that the purpose of both the WRO and the UFLPA is to eliminate any forced labor that may exist in the manufacturing or supply chain of a product through customs measures, and not to penalize a company or even an industry.”
Reply: No one has claimed, at least to my knowledge, that the goal of WROs or the UFLPA or our investigation is to punish companies. My understanding of these measures is their aim is to stop the import of goods that may be tied to forced labor.
The questions that researchers and experts pose to me to help me understand the category at issue here are the following: If a worker from North Korea is found in a Chinese seafood plant that exports to the U.S., are the goods from that plant banned from import into the U.S.? Would it be the same if a child laborer were found to be working in a processing plant that exports to the U.S.? What if both types of workers (the North Korean and the child laborer) were interviewed by a lawyer hired by the plant operator or by an official from the Chinese government or by a private social auditing firm? If the child laborer or North Korean worker said in the interview that they were not forced labor — would their testimony be deemed credible in light of the relevant context?
Other points made in the most recent letter broadly claim that the audits done on these plants clear them of the presence of forced labor and raise questions about the definition of forced labor. For these points we would kindly like to hew closely to the wording used in your letter itself.
Your letter says: “A SMETA audit does not identify whether the employees come from a particular region or country, such as Xinjiang of China or North Korea. The objectivity and reliability of the SMETA and SYSCO Quality Assurance audits cannot be questioned on the basis of not auditing the identities of Xinjiang and North Korean employees, as that is not the audit standard or content.”
Reply: We agree on the point that the audits have not identified Xinjiang workers at multiple plants in China. You also say that the audits are not supposed to identify these two categories of workers. This, of course, is precisely the point we sought to make. UFLPA and CATSA presumptions categorize the presence of Xinjiang or North Korean workers as distinctly problematic and categorically worthy of specific attention tied to forced labor, and any audit that did not involve asking about the ethnic identities of workers and where they are from would be irrelevant to rebutting these the presumptions in these laws. Under UFLPA, workers from Xinjiang are categorically seen as part of state-sponsored forced labor because they are not given the true option to decline the work nor to leave the work without penalty. This interpretation is consistent with CBP’s description of labor transfer programs in the UFPLA implementation strategy. That description relates back to the legal definition in the U.S. Tariff Act and the ILO indicators: work that is “performed under the menace of any penalty and for which the worker does not volunteer.”
Your audit review and my reporting both found that the audits did not investigate regional origin or ethnicity of workers. Your position seems to be that the audit is valid because it adhered to auditing standards. But those standards did not include the most important question about the origin of workers.
That said, I might offer a point of nuance. As you know, SMETA self-assessment questionnaires (SAQ), required by the SMETA audit, call for workers to disclose whether they are migrants, as well as their primary languages, and thus if done properly at Chishan would have identified Xinjiang and North Korean workers. That they apparently failed to do this underscores one of the shortcomings of social audits. But the other deeper concern with these types of audits, that experts have pointed out, is whether any of them can be effectively conducted in China where access to plants and independent assessments by NGOs is tightly controlled and organizations (including journalism organizations or ocean conservation NGOs) are expelled from the country if they are critical of the government or companies.
As part of our investigation, we engaged auditors with China-specific expertise from multiple international firms to ask their views on why social audits aren't detecting state-imposed forced labor affecting Xinjiang ethnic minorities. The reasons given included: Chinese audit teams being reluctant to raise the presence of Xinjiang workers as a finding; auditees concealing Xinjiang workers ahead of announced audits, as well as payroll and other documentation; Xinjiang workers interviewed by auditors being reluctant to answer questions candidly; auditors lacking the training and expertise to ask the right questions to investigate state-imposed forced labor.
Social audits have also failed to identify hazards in other countries and industries, such as in a case when hundreds of workers died at Rana Plaza and other sites in Southeast Asia. Our reporting identified audit gaps for seafood that are consistent with reports on other industries. These include audits after an incident (after evidence is cleaned up or removed), advance notice of an audit (to remove evidence), coercion of workers (managers present for interviews, criminal liability for interviews), delegation of illegal work to contractors (outside the scope of an audit), reliance of self-reporting by industry managers, failure to inspect all premises, and failure to audit for known indicators of forced labor.
Thank you again for your kind invitation and thoughtful email. Furthermore, thank you for pointing out that you and your clients are presently going to “pay more attention to” the concerns we raised about forced labor in the seafood supply chain.
As I’m sure you’re aware, Haibo and Haidu are important to the global seafood supply chain, accounting for roughly thirty percent of all squid processed in China, according to the Rongcheng government. These facilities are distinctly important in the U.S. context since they produce at least seventeen percent of all squid imported into the United States, according to trade and NOAA data, which means that the discovery of Xinjiang labor in the plants has direct pertinence to the UFLPA and the Tariff Act.
Take care,
Ian Urbina
The Outlaw Ocean Project