Early impact of COVID-19 pandemic on the molluscan shellfish supply chain in China.

A farm-level survey and a market survey were conducted in China to study the impact of the COVID-19 on the shellfish supply chain in the early stage of the pandemic in 2020. 114 farms of 36 cites participated in an online survey from mid-July to late August, and 23 markets in five cities were investigated from June to November. The series results are expected to provide an insight for the fisheries and aquaculture sector to mitigate the impact of the pandemic on participants in the shellfish supply chain and to improve of the ability of the government to respond to such events in the future. Shellfish farmers and sellers have experienced a sharp drop in profits, largely due to shrinking demand. Relying on local employees and suppliers, shellfish farming activities have rapidly resumed since the gradual reopening from February. Nevertheless, demand has been slow to recover due to public concern about contamination of seafood with the virus. Additionally, we analyzed the ability and attitude of farms of different size to cope with the pandemic and the plight of shellfish sellers, and discussed government supports for different size farms and improvements in seafood distribution channels.

Introduction

The COVID-19 pandemic has triggered a huge and unpredictable impact throughout the molluscan shellfish (hereafter the shellfish) supply chain in China, the largest producer and consumer of shellfish in the world (FAO, 2020a, 2020b; Peng et al., 2019; Peng et al., 2021), due to the various effects of widespread virus containment measures on demand, logistics, prices, business operations and planning (FAO, 2020c; Xu et al., 2021). The closure of foodservices in February 2020 and restrictions on restaurants seating capacity after reopening have led to a significant reduction in demand for fresh and alive shellfish (FAO, 2020c). Although the prevention efforts have weakened with the decrease in the number of confirmed cases (IMF, 2020), the demand was sluggish to recover. The first COVID-19 cases were officially linked to a seafood market in Wuhan (Djekic et al., 2021; Jalava, 2020), and re-emergent outbreaks in Beijing, Dalian and Qingdao were also linked to a seafood market, a seafood processing plant and imported frozen seafood, respectively (Han et al., 2020). Although COVID-19 does not infect aquatic species (Godoy et al., 2021), the virus has been detected on packaging materials and storage environments (Han et al., 2020), and isolated from imported frozen seafood package surface. These events that were broadcasted across the country, has raised public concerns about the safety of seafood, including shellfish, which may have exacerbated the plight of supply chain participants during the pandemic (Perillo et al., 2021; Xinhua, 2020a).

China contributed 75.9% to global shellfish production and 75.7% to consumption in 2018, with aquaculture accounting for 95.8% of its production (FAO, 2020a, 2020b). And China is the major contributor to production growth, as its shellfish production increased from 86,900 tons in 1950 to 15,081,753 tons in 2018, contributing 80.0% of the global production growth (FAO, 2020a). Unlike other counties (FAO, 2020c; Pedroza-Gutiérrez et al., 2021; Teng, 2020; White et al., 2020), China is nearly self-sufficient in shellfish. In 2018, China's shellfish import and export only accounted for 0.8% of the domestic consumption and 1.5% for the domestic production, respectively (FAO, 2020a), and most shellfish was delivered to individual consumers though wholesale markets and wet markets. Even though, a considerable part of export was contributed by processing trade (GACC, 2020), that is, by importing and processing raw shellfish for re-export (Feenstra & Hanson, 2005). The huge domestic production and consumption implies that it’s unlikely for China to fill in the gap between supply and demand by import from the international market, not to mention its market independence. Meanwhile, in view of the global pandemic and shrinking commodity trade (WTO, 2020; Xu et al., 2021), it is obviously not a feasible option to alleviate any shocks created by declining domestic demand through increasing export. China has to figure out and implement its mitigation strategies to cope with sharply reduced demand, price shock and an uncertain outlook facing its shellfish industry (FAO, 2020c).

The effects of COVID-19 pandemic on shellfish industry may depend on the supply chain activities, species, and the size of farms, so do participants’ ability to respond to lockdown and severe demand shocks (Menhat et al., 2021; Milanes et al., 2021). If farms rely on local algae/feed and seeds, the short-term, for example, a month, lockdown may have marginal impact on shellfish supply capacity, while shrinking demand will reduce sales and revenue of retailers and wholesalers, and the effect can be transmitted upstream along the supply chain (Lee et al., 1997). For farms involved in downstream stages (e.g., processing), they may suffer from the demand shock earlier and more severely than those just engaged in upstream activities (e.g., breeding and cultivation of bait organisms). In view of aquaculture production cycle, it may last for a few months or multiple years in duration depending on species farmed (FAO, 2020c) and farmers can change production schedule, such as to delay harvesting if the harvest time is flexible. Nevertheless, the shellfish that remains in the growing phase will eventually be harvested, thus going into the market in the late 2020 or 2021. In addition, large farms may prepare themselves well in terms of cash flow and financial opportunity to address market uncertainties. Small and medium farms, however, may not be so lucky, often feeling helpless when facing difficult market situations (Bennett et al., 2020; FAO, 2020c).

In this study, an online farm-level survey and a field market survey were conducted for the purpose of providing suggestions to mitigate the impact of the pandemic on participants in the shellfish supply chain and to improve the ability of the government to respond to such events in the future. First, from mid-July to late August 2020, 114 shellfish farms in 36 cities of mainland China were invited to finish an online questionnaire. We discussed the differentiated situation faced by shellfish farms in the early stage of the pandemic, as well as response strategies and desired government supports. Then, 23 markets in five cities were visited from June to November 2020. We investigated the business situation of shellfish sellers and discussed market renovation.

Material and methods

Farm-level survey

Survey process

An online survey of shellfish farms located in China's coastal regions and the Yangtze River Basin was conducted from mid-July to late August 2020. We are quite familiar with 16 operators of shellfish farms in Shandong province, of which 5 were selected to be interviewed by phone in early July, prior to the formal investigation, for the purpose of refining the questionnaire we designed. The questions and options of the questionnaire were then adjusted according to insights derived from the pre-survey.

Since there were still sporadic new confirmed cases of COVID-19 in China, questionnaires were filled through the online platform Wenjuanxing (https://www.wjx.cn/app/survey.aspx), and the link to the questionnaire was sent to the respondents via WeChat with their consent. Farms were contacted in advance to analyze availability for participating in the survey. To ensure a representative sample, shellfish farms in each coastal province of mainland China were invited to participate in this survey, as marine aquaculture contributed 98.66% of China's shellfish production in 2018 (FAO, 2020a). Besides geographical distribution, the only criterion was that the farms operate in at least one stage of the shellfish production: algae/feed production, seed production, nursery, intermediate cultivation, grow-out and processing. We did not limit shellfish species and the proportion of farms that culture each shellfish species, as farms often raise multiple species.

A total of 114 farms in 36 cities of 12 provinces in mainland China were included in the survey (Fig. 1 ). In 2018, the total shellfish aquaculture production of these 12 provinces was 14,563,801 tons (including 14,439,302 tons of marine aquaculture production and 124,499 tons of freshwater aquaculture production) (BOF, NFTEC, & CSOF, 2019), accounting for 99.51% of China's shellfish aquaculture production, 83.17% of global shellfish aquaculture production and 72.39% of global shellfish production (FAO, 2020a). The farm characteristics are presented in Table 1 .

Fig. 1

Farm and market survey locations.

Table 1

Farm characteristics.

Farm characteristics	Overall (n = 114)	Farm Size
		Small (n = 30)	Medium (n = 61)	Large (n = 23)	Correlation Coefficient
Years in business
0–2	7 (6.1%)	1 (3.3%)	5 (8.2%)	1 (4.3%)	0.237*
3–5	11 (9.6%)	6 (20.0%)	5 (8.2%)	0
6–10	28 (24.6%)	10 (33.3%)	14 (23.0%)	4 (17.4%)
>10	68 (59.6%)	13 (43.3%)	37 (60.7%)	18 (78.3%)
Operating revenue in 2019 (Million ¥)
<0.5	16 (14.0%)	16 (53.3%)	0	0	0.756**
0.5–1	16 (14.0%)	9 (30.0%)	7 (11.5%)	0
1–5	50 (43.9%)	5 (16.7%)	39 (63.9%)	6 (26.1%)
5–10	27 (23.7%)	0	15 (24.6%)	12 (52.2%)
10–200	5 (4.4%)	0	0	5 (21.7%)
>200	0	0	0	0
Number of employees in 2019
1–5	21 (18.4%)	19 (63.3%)	2 (3.3%)	0	0.864**
6–10	20 (17.5%)	10 (33.3%)	10 (16.4%)	0
11–50	43 (37.7%)	1 (3.3%)	41 (67.2%)	1 (4.3%)
51–100	16 (14.0%)	0	8 (13.1%)	8 (34.8%)
>100	14 (12.3%)	0	0	14 (60.9%)
Farm species
Oysters	59 (53.4%)	14 (46.7%)	31 (50.8%)	9 (39.1%)	0.093
Scallops	39 (35.3%)	10 (33.3%)	19 (31.1%)	10 (43.5%)	0.063
Clams, cockles, ark shells	33 (32.6%)	2 (6.7%)	23 (37.7%)	8 (34.8%)	0.234*
Mussels	10 (8.1%)	3 (10.0%)	6 (9.8%)	1 (4.3%)	−0.062
Abalone	12 (10.5%)	1 (3.3%)	7 (11.5%)	4 (17.4%)	0.158
Other molluscs	40 (34.8%)	12 (40.0%)	19 (31.1%)	9 (39.1%)	−0.017
Other aq. animals	37 (35.7%)	6 (20.0%)	20 (32.8%)	11 (47.8%)	0.200*

Note: Correlation Coefficient. Spearman's rank correlation coefficients between farm size (small = 1, medium = 2 and large = 3) and the scores of farm characteristics (see Supplementary Data).

aCorrelation is significant at the 0.05 level (2-tailed).

bCorrelation is significant at the 0.01 level (2-tailed).

Questionnaire used for the survey

A questionnaire has been developed in Chinese language to analyze whether the COVID-19 pandemic has affected farms and stages of the shellfish aquaculture production chain. The questionnaire was designed in four parts.

The first section collected business information related to the farms, including location, years in business, farming species and operating revenue and number of employees in the previous year (2019).

More specific characteristics of farms were analyzed in the second section. Stages involved in, the native place of employees, algae/feed sources, seeds sources and customer distribution were asked to analyze the ability of farms to cope with logistics restrictions. Since logistics restrictions are based on levels of administrative divisions, the options of these multiple-choice questions included ‘within the city’, ‘other cities within the province’, ‘other provinces’ and ‘abroad’.

The third section was dedicated to identifying the pandemic impact on farms and their activities. First, we asked the month they resumed to work and the pandemic impact on farms and production stages, including labor recruitment, algae/feed production or purchase, seed production (hatchery and seed collection) or purchase, nursery, intermediate cultivation, grow-out, harvesting, processing, logistics and sales (Yang et al., 2016), and then the changes in production, the number of employees and per capita labor costs this year compared to previous years. As bivalve growers generally do not need to feed their brood (FAO, 2021), the question about ‘algae/feed production or purchase’ would be skipped, if feeding was not required in their shellfish farms. And if nursery, intermediate cultivation, grow-out or processing was not involved in the farm, the corresponding questions would be skipped as well. The respondents had the option to rate the degree of the pandemic impact on farms and stages according to a five-point Likert scale from 1'very low’, 2 ‘low’, 3 ‘moderate’, 4 ‘high’ to 5 ‘very high’. To ensure that a same score chosen by respondents refers to the same level of shock on their farms, situations corresponding to the scores were added to the description of the question (Table 2 ).

Table 2

The situations of farms corresponding to the degrees of the pandemic impact.

Score	Degree of impact	Situation of farms
1	Very low	No apparent impact
2	Low	Problems can be solved by adjusting plans
3	Moderate	Plans are disrupted and profits are significantly reduced
4	High	Facing certain losses
5	Very high	On the verge of bankruptcy

Finally, the fourth section explored the problems encountered by farms and their concerns for the coming year, responses strategies, and desired government supports.

Working definition of large, medium and small farms

In this study, combining the collected data and China's regulations on the standards for classification of small and medium-sized enterprises (SMEs) (NBS, 2017), a score method was created to measure farm size according to operating revenue and number of employees. Farms with revenue below 0.5 million yuan, 0.5–1 million yuan, 1–5 million yuan, 5–10 million yuan and 10–200 million yuan were assigned a farm size score of 1–5 points respectively (no farms with revenue greater than 200 million yuan participated in the survey). And farms with 1–5, 6–10, 11–50, 51–100 and more than 100 employees were added 1–5 points to their size scores, respectively. Those with scores between 2 and 4 were small farms, 5–7 were medium farms, and 8–10 were large farms.

Based on this approach, the 114 farms participating in the survey were classified into 30 small farms, 61 medium farms and 23 large farms (Fig. 2 ). Farm attributes by size are shown in Table 1. To explore the differential impact of the pandemic on farms of different sizes, we calculated the Spearman's rank correlation coefficients (Spearman, 1904) between size and farm characteristics. As correlation coefficient shows the linear relationship between variables (Mukaka, 2012), non-linear relationships, especially the distinctiveness of medium farms were described separately.

Fig. 2

Farm size defined by operating revenue and number of employees. Bubble sizes are proportional to the numbers of farms (the values in the bubbles).

It should be noted that China has implemented different level of tax and fees reductions and financial supports for large enterprises and SMEs (MOF & STA, 2020; MOHRSS, MOF, & STA, 2020; PBC, CBIRC, MOF, NDRC, & MIIT,2020; Zhang, 2020). According to classification standards for SMEs in China (NBS, 2017), Fishery SMEs are those with operating revenue bellow 200 million, which means all farms participating in the survey were SMEs.

Seafood markets survey

In China, seafood trade mainly take place in wholesale markets and wet markets, which are the linkage between farmers and consumers in China (Zhong et al., 2020). In wholesale markets, which are mainly located outside major urban areas, wholesalers sell fresh fruits, vegetables, meat and seafood mainly to retailers, caterers, and some to consumers. Some wholesale markets also serve as distribution centers. Wet markets, a traditional form of food retail, also known as public markets in China, dominate the retail of fresh food in Chinese cities (Bai et al., 2008; Chen, 2019b; Maruyama et al., 2016; Si et al., 2019; Zhang & Pan, 2013; Zhong et al., 2020). Unlike developed countries (Burch et al., 2013; Wrigley & Lowe, 2002), despite the rapid expansion of modern supermarket chains in urban China, wet markets have maintained their popularity which primarily rests in their competitive advantage with regards to a particular kind of freshness corresponds with consumers’ consumption habits and food culture (Zhong et al., 2020). However, these markets have been blamed in some reports for the virus transmission during the pandemic. Frozen seafood and humid environment were thought to have created the conditions for the virus to lurk and spread in seafood markets (Wang, 2020).

Before the pandemic, China had been renovating these markets over the past decades to improve market environment. First, trading venues have been transformed from the open-air to the indoor. Second, markets are divided into several sections such as fruits & vegetable, meat & eggs, seafood, deli, and non-agricultural products, where stalls sell the corresponding products. Third, markets are equipped with appropriate infrastructure according to characteristics of products sold in each section. For example, in seafood section, the floor is covered with non-slip tiles, and drainage facilities are superior to other sections. Nevertheless, the outbreaks in Wuhan and Beijing were directly linked to a wet market mainly selling seafood and a seafood section of a wholesale market, respectively, which has raised public concerns about buying seafood in such markets and led to a decline in demand. And the fact shows the necessity of further improvement of seafood distribution channels.

In seafood market survey, we visited 13 wholesale markets and 10 wet markets in five cities including Qingdao, Dalian, Nanjing, Jinan and Beijing (Fig. 1) from June to November 2020 to verify the feedback from farms about the pandemic impact on shellfish demand and to understand the improvements made by the markets to prevent the virus transmission. The months we visited each city, the markets investigated and the number of stalls interviewed in each market are listed in Appendix. All five cities are capital (Beijing), provincial capitals (Nanjing and Jinan) or sub-provincial cities (Qingdao and Dalian) with a population of more than 5 million (NBS, 2019). And there were seafood-related outbreaks in Beijing, Dalian and Qingdao, while there were none in Nanjing and Jinan. At least one wholesale market and one wet market were investigated in each city. The selected wholesale markets were the local seafood distribution centers, and the selected wet markets were the most popular seafood retail markets. We observed the layout and environment of the market, and interviewed shellfish sellers about their business. If there were more than 10 shellfish stalls in the market, we would stop after interviewing the shellfish sellers of ten stalls. And in some of wholesale markets, managers were interviewed about the rent waiver (SASAC, 2020) and the details of market renovation implemented or planned.

Results and discussion

Farm characteristics

The characteristics of the farms that participated in the survey are displayed in Table 1. Farms in business for more than 10 years represented 59.6% of the sample, followed by farms for 6–10 years (24.6%), 3–5 years (9.6%) and 0–2 years (6.1%). Majority of the farms cultured oysters (53.4%), followed by scallops (35.3%), clams cockles, ark shells (32.6%), mussels (8.1%) and abalone (3.3%). And one-third of the farms cultured other shellfish (34.8%) and other aquatic animals (35.7%).

In addition to the indispensable activities of harvesting and logistics, the farms were involved in 2.7 production stages on average (Table 3 ). Most of farms were involved in grow-out (79.8%), followed by seed production (50.9%), intermediate cultivation (48.2%) and nursery (43.9%), while only a few of farms were involved in algae/feed production (29.8%) and processing (20.2%).

Table 3

Production characteristics.

Production characteristics	Overall (n = 114)	Farm Size
		Small (n = 30)	Medium (n = 61)	Large (n = 23)	Correlation Coefficient
Production stage
Algae/feed production	34 (29.8%)	4 (13.3%)	23 (37.7%)	7 (30.4%)	0.149
Seed production	58 (50.9%)	8 (26.7%)	37 (60.7%)	13 (56.5%)	0.227b
Nursery	50 (43.9%)	5 (16.7%)	30 (49.2%)	15 (65.2%)	0.343c
Intermediate cultivation	55 (48.2%)	11 (36.7%)	27 (44.3%)	17 (73.9%)	0.239b
Grow-out	91 (79.8%)	24 (80.0%)	48 (78.7%)	19 (82.6%)	0.018
Processing	23 (20.2%)	4 (13.3%)	11 (18.0%)	8 (34.8%)	0.172
Other	3 (2.6%)	1 (3.3%)	2 (3.3%)	0	−0.064
Num of stagesa	2.7 ± 1.4	1.9 ± 1.0	2.9 ± 1.4	3.4 ± 1.5	0.371c
Native place of employees
The city	89 (78.1%)	20 (66.7%)	49 (80.3%)	20 (87.0%)	0.172
Other cities within the province	37 (32.5%)	10 (33.3%)	21 (34.4%)	6 (26.1%)	−0.046
Other provinces	36 (31.6%)	8 (26.7%)	19 (31.1%)	9 (39.1%)	0.089
Abroad	0	0	0	0	–
Sources of algae/feed
No need to feed	57 (50.0%)	15 (50.0%)	28 (45.9%)	14 (60.9%)	0.062
Algae/feed production	34 (29.8%)	4 (13.3%)	23 (37.7%)	7 (30.4%)	0.149
Purchase inside the city	15 (13.2%)	5 (16.7%)	7 (11.5%)	3 (13.0%)	−0.043
Purchase in other cities within the province	30 (26.3%)	9 (30.0%)	17 (27.9%)	4 (17.4%)	−0.091
Purchase in other provinces	8 (7.0%)	2 (6.7%)	3 (4.9%)	3 (13.0%)	0.073
Purchase abroad	0	0	0	0	–
Source of seeds
Hatchery	65 (57.0%)	12 (40.0%)	39 (63.9%)	14 (60.9%)	0.160
Seed collection	28 (24.6%)	2 (6.7%)	17 (27.9%)	9 (39.1%)	0.263c
Purchase within the city	20 (17.5%)	5 (16.7%)	9 (14.8%)	6 (26.1%)	0.073
Purchase in other cities within the province	50 (43.9%)	10 (33.3%)	30 (49.2%)	10 (43.5%)	0.083
Purchase in other provinces	44 (38.6%)	14 (46.7%)	23 (37.7%)	7 (30.4%)	−0.114
Purchase abroad	0	0	0	0	–
Distribution of customers
The city	54 (47.4%)	11 (36.7%)	34 (55.7%)	9 (39.1%)	0.039
Other cities within the province	86 (75.4%)	21 (70.0%)	44 (72.1%)	21 (91.3%)	0.156
Other provinces	75 (65.8%)	14 (46.7%)	40 (65.6%)	21 (91.3%)	0.315c
Abroad	6 (5.3%)	1 (3.3%)	1 (1.6%)	4 (17.4%)	0.189b
East Asia	6	1	1	4	–
Other Asia, nes	1	0	0	1	–
Europe	1	0	0	1	–
North America	1	0	0	1	–
Latin America	1	0	0	1	–
Oceania	0	0	0	0	–
Africa	0	0	0	0	–

Notes: Correlation coefficient. Spearman's rank correlation coefficients between farm size (small = 1, medium = 2 and large = 3) and production characteristics (no = 0, yes = 1).

The indispensable activities of harvesting and logistics and “other” production stages are not counted in “Num of stages”; and the correlation coefficient is between farm size and the number of stages.

Correlation is significant at the 0.05 level (2-tailed).

Correlation is significant at the 0.01 level (2-tailed).

Except sales, all production activities of farms were carried out domestically, and most activities occurred within the province where the farms were located. Some companies in China delayed the resumption of work, as employees who had returned hometowns to celebrate Chinese New Year were unable to return to work due to the lockdown. But majority of shellfish farms recruited workers within the city (78.1%) where the farms located, followed by other cities within the province (32.5%) and other provinces (31.6%). Half of farms did not need to feed their shellfish (50.0%), and the rest mainly produced algae/feed by themselves or purchased within the province. Only 7.0% of farms bought algae/feed in other provinces. Regarding breeding, two approaches were employed for seed production: hatchery (57.0%) and seed collection (collecting seed from natural resources by providing suitable substrates) (Yang et al., 2016) (24.6%). Farms also purchased seeds from other farms located in the same city (17.5%), other cities within the province (43.9%) or other provinces (38.6%). No farms had foreign employees, or purchased algae/feed or seeds from abroad.

The distribution of farm customers (including wholesalers, caterers, consumers, and farmers) was wider than native places of employees and sources of algae/feed and seeds. Most customers were from other cities within the province (75.4%) and other provinces (65.8%), followed by the city (47.4%) and abroad (5.3%). And most of international business occurred in East Asia.

Pandemic effects on shellfish farming

Based on the Likert scale used (from 1 ‘very low’ to 5 ‘very high’), the pandemic had a moderate impact on shellfish farms (3.1) (Table 4 ), with the average level of situation could be described as ‘Plans were disrupted and profits were significantly reduced’ (Table 2). 66.67% of the farms reported a decline in production compared to the same period of previous years, and 17.54% of them reduced production by more than 50% (Table 5 ). 34.21% of the farms did not shutdown during the pandemic, and most of the rest achieved resumption in February–April, with 92.11% of the farms resuming work as of April (Fig. 3 a).

Table 4

The COVID-19 pandemic impact on shellfish farms.

The changes compared to previous years	Overall (n = 114)	Farm Size
		Small (n = 30)	Medium (n = 61)	Large (n = 23)	CorrelationCoefficient
Production
Decrease >50%	20 (17.5%)	5 (16.7%)	13 (21.3%)	2 (8.7%)	−0.132
Decrease 20%–50%	28 (24.6%)	7 (23.3%)	12 (19.7%)	9 (39.1%)
Decrease <20%	28 (24.6%)	8 (26.7%)	17 (27.9%)	3 (13.0%)
No change	30 (26.3%)	8 (26.7%)	15 (24.6%)	7 (30.4%)
Increase	8 (7.0%)	2 (6.7%)	4 (6.6%)	2 (8.7%)
Number of employees
Decrease >50%	13 (11.4%)	4 (13.3%)	6 (9.8%)	3 (13.0%)	0.055
Decrease 20%–50%	31 (27.2%)	7 (23.3%)	19 (31.2%)	5 (21.7%)
Decrease <20%	22 (19.3%)	2 (6.7%)	11 (18.0%)	9 (39.1%)
No change	44 (38.6%)	14 (46.7%)	24 (39.3%)	6 (26.1%)
Increase	4 (3.5%)	3 (10.0%)	1 (1.6%)	0
Per capita labor costs
Decrease	14 (12.3%)	3 (10.0%)	11 (18.0%)	0	0.013
No change	32 (28.1%)	10 (33.3%)	16 (26.2%)	6 (26.1%)
Increase <10%	29 (25.4%)	5 (16.7%)	16 (26.2%)	8 (34.8%)
Increase 10%–20%	18 (15.8%)	6 (20.0%)	7 (11.5%)	5 (21.7%)
Increase >20%	21 (18.4%)	6 (20.0%)	11 (18.0%)	4 (17.4%)

Notes: Correlation coefficient. Spearman's rank correlation coefficients for farm size (small = 1, medium = 2 and large = 3) and the score of changes in production, number of employees and per capita labor costs (see Supplementary Data).

Table 5

The changes compared to previous years.

Impact	Num	Overall	Farm Size
			Small	Medium	Large	Correlation Coefficient
Month of work resumption a	114	2.6 ± 1.5	2.6 ± 1.4	2.6 ± 1.7	2.5 ± 1.2	−0.045
Impact on farm	114	3.1 ± 1.0	3.2 ± 1.2	3.1 ± 0.9	3.0 ± 0.9	−0.096
Impact on production stage
Labor recruitment	114	2.6 ± 1.4	2.4 ± 1.5	2.7 ± 1.4	2.8 ± 1.4	0.110
Algae/feed production or purchase	57 b	2.0 ± 1.2	2.0 ± 1.4	2.2 ± 1.1	1.7 ± 1.1	−0.027
Seed production or purchase	114	2.4 ± 1.4	2.5 ± 1.4	2.4 ± 1.3	2.4 ± 1.4	−0.036
Nursery	50 c	2.5 ± 1.4	2.8 ± 1.1	2.4 ± 1.4	2.5 ± 1.5	−0.022
Intermediate cultivation	55 c	2.5 ± 1.5	2.5 ± 1.6	2.6 ± 1.6	2.5 ± 1.5	0.015
Grow-out	91 c	2.6 ± 1.5	2.5 ± 1.6	2.6 ± 1.5	2.6 ± 1.6	0.025
Harvesting	114	2.6 ± 1.5	2.6 ± 1.5	2.6 ± 1.4	2.9 ± 1.8	0.042
Processing	23 c	3.3 ± 1.5	3.3 ± 1.5	2.8 ± 1.7	4.0 ± 1.1	0.244
Logistics	114	2.8 ± 1.5	2.4 ± 1.3	2.9 ± 1.5	3.1 ± 1.7	0.142
Sales	114	3.6 ± 1.4	3.5 ± 1.5	3.5 ± 1.4	3.7 ± 1.5	0.039

Notes: Correlation coefficient. Spearman's rank correlation coefficients between farm size (small = 1, medium = 2 and large = 3) and scores (see Supplementary Data).

No Shutdown = 1; No Resumption = 8.

Farms where feeding was not required are excluded.

Only farms involved in the stage are included.

Fig. 3

The COVID-19 impact on shellfish farms. (a) Percentage of farms resuming work. (b) The average scores of the impact on farms and stages. (c–f) The native place of employees (c), sources of algae/feed (d) and seed (e), and distribution of customers (f) for farms unaffected (score = 1), less affected (score = 2 or 3), and more affected (score = 4 or 5) in recruitment, production or purchase of algae/feed and seed, and sales. Values in parentheses are the numbers of farms. Local = within the city the farm was located, Provincial = in other cities within the province, Domestic = in other provinces.

By production segment, the impact on the production activities increased from upstream to downstream. As shown in Fig. 3b and Table 4, sales, the stage at the end of shellfish production, was most affected (3.6), followed by processing (3.3) and logistics (2.8), the other two post-harvest stages. In rearing segment, harvesting (2.6), grow-out (2.6) and intermediate cultivation (2.5) were less affected than the stages in the post-harvest segment. In pre-production segment, nursery (2.5), the production or purchase of seed (2.4) and the production or purchase of algae/feed (2.0) were affected to a lesser extent than rearing and post-harvest segments, except for labor recruitment (2.6).

The pandemic had a greater impact on cross-regional activities. Employees of farms in which recruitment was not affected (score = 1) were mainly local inhabitants, while a higher proportion of farms that were more affected (score = 4 or 5), hired workers from other cities within the province and other provinces (Fig. 3c). This phenomenon also occurred in the production or purchase of seed and algae/feed and sales (Fig. 3d–f), i.e., the more affected, the more reliance on cross-regional activities.

Problems, concerns, strategies, and desired government supports

Table 6 Displays farms’ problems (financial pressures and other problems), concerns for the coming year, strategies to cope with the pandemic, and desired government supports.

Table 6

Problems, concerns, strategies and desired government supports.

Content	Overall (n = 114)	Farm Size
		Small (n = 30)	Medium (n = 61)	Large (n = 23)	Correlation Coefficient
Financial pressures
Labor costs	22 (19.3%)	6 (20.0%)	12 (19.7%)	4 (17.4%)	−0.021
Rent, materials & infrastructure costs	45 (39.5%)	14 (46.7%)	26 (42.6%)	5 (21.7%)	−0.162
Decrease in operating revenue	68 (59.6%)	17 (56.7%)	39 (63.9%)	12 (52.2%)	−0.019
Loan repayment	21 (18.4%)	6 (20.0%)	10 (16.4%)	5 (21.7%)	0.008
Hard to get loans	20 (17.5%)	3 (10.0%)	15 (24.6%)	2 (8.7%)	0.014
Other financial pressure	4 (3.5%)	3 (10.0%)	1 (1.6%)	0	−0.196a
No financial pressure	21 (18.4%)	3 (10.0%)	10 (16.4%)	8 (34.8%)	0.207a
Problems
Government calling for shutdown	31 (27.2%)	8 (26.7%)	16 (26.2%)	7 (30.4%)	0.025
Shortage of epidemic prevention supplies	6 (5.3%)	1 (3.3%)	5 (8.2%)	0	−0.032
Lack of technical support	18 (15.8%)	2 (6.7%)	8 (13.1%)	8 (34.8%)	0.249b
Seafood price fluctuations	88 (77.2%)	23 (76.7%)	49 (80.3%)	16 (69.6%)	−0.047
Facilities maintenance, repair and renewal	12 (10.5%)	1 (3.3%)	6 (9.8%)	5 (21.7%)	0.198a
Fisheries diseases	39 (34.2%)	10 (33.3%)	23 (37.7%)	6 (26.1%)	−0.041
Weather disasters	44 (38.6%)	10 (33.3%)	26 (42.6%)	8 (34.8%)	0.021
Other problems	11 (9.6%)	5 (16.7%)	4 (6.6%)	2 (8.7%)	−0.105
No problems	5 (4.4%)	0	3 (4.9%)	2 (8.7%)	0.146
Concerns for the coming year
Rise in labor costs or recruiting difficulty	51 (44.7%)	12 (40.0%)	29 (47.5%)	10 (43.5%)	0.031
Rise in rent or materials cost	29 (25.4%)	8 (26.7%)	17 (27.9%)	4 (17.4%)	−0.064
Rise in logistics cost or transport restrictions	50 (43.9%)	7 (23.3%)	29 (47.5%)	14 (60.9%)	0.184
Facilities maintenance, repair and renewal	18 (15.8%)	1 (3.3%)	12 (19.7%)	5 (21.7%)	0.264b
Drop in shellfish demand due to public concerns for seafood safety	71 (62.3%)	19 (63.33%)	37 (60.7%)	15 (65.2%)	0.009
The effect of shrinking demand transmitted to upstream along the supply chain	37 (32.5%)	5 (16.7%)	19 (31.1%)	13 (56.5%)	0.282b
Price drop or fluctuation	80 (70.2%)	20 (66.7%)	43 (70.5%)	17 (73.9%)	0.054
Rise in stock level	29 (25.4%)	9 (30.00%)	13 (21.3%)	7 (30.4%)	−0.009
Regulation tightening	12 (10.5%)	1 (3.33%)	9 (14.7%)	2 (8.7%)	0.077
Other concerns	4 (3.5%)	1 (3.33%)	2 (3.3%)	1 (4.35%)	0.017
No concerns	1 (0.9%)	(0.00%)	1 (1.6%)	0	0.011
Response strategies
Changing the scale of production	65 (57.0%)	14 (46.7%)	37 (60.7%)	14 (60.9%)	0.106
Changing the number of employees	38 (33.3%)	4 (13.3%)	24 (39.3%)	10 (43.5%)	0.232a
Changing aquaculture practice	54 (47.4%)	12 (40.0%)	31 (50.8%)	11 (47.8%)	0.062
Exploring new sources of algae/feed and seed	11 (9.6%)	3 (10.0%)	5 (8.2%)	3 (13.0%)	0.028
Expanding to upstream and downstream	16 (14.0%)	1 (3.3%)	10 (16.4%)	5 (21.7%)	0.187a
Price promotions	28 (24.6%)	6 (20.0%)	15 (24.6%)	7 (30.4%)	0.081
Developing new products	28 (24.6%)	8 (26.7%)	14 (23.0%)	6 (26.1%)	−0.010
Broadening distribution channels	48 (42.1%)	8 (26.7%)	26 (42.6%)	14 (60.9%)	0.234a
Quitting shellfish aquaculture	11 (9.6%)	7 (23.3%)	4 (6.6%)	0	−0.279b
Other measures or plans	3 (2.6%)	1 (3.3%)	1 (1.6%)	1 (4.3%)	0.013
No measures or plans	8 (7.0%)	1 (3.3%)	5 (8.2%)	2 (8.7%)	0.077
Desirable government supports
Providing epidemic prevention supplies	10 (8.8%)	2 (6.7%)	7 (11.5%)	1 (4.3%)	−0.015
Recruitment support	25 (21.9%)	4 (13.3%)	17 (27.9%)	4 (17.4%)	0.052
Agricultural insurance subsidies	28 (24.6%)	2 (6.7%)	17 (27.9%)	9 (39.1%)	0.263b
Social insurance premiums cut or deferral	23 (20.2%)	4 (13.3%)	10 (16.4%)	9 (39.1%)	0.203a
Technical support	42 (36.8%)	11 (36.7%)	19 (31.1%)	12 (52.2%)	0.092
Logistics support	41 (36.0%)	6 (20.0%)	22 (36.1%)	13 (56.5%)	0.255b
Helping explore sales channels, dock production and marketing	72 (63.2%)	21 (70.0%)	39 (63.9%)	12 (52.2%)	−0.122
Credit support	51 (44.7%)	14 (46.7%)	28 (45.9%)	9 (39.1%)	−0.048
Tax cut or deferral	40 (35.1%)	7 (23.3%)	26 (42.6%)	7 (30.4 %)	0.071
Other supports	4 (3.5%)	2 (6.7%)	1 (1.6%)	1 (4.3%)	−0.056
No desirable supports	2 (1.8%)	0	2 (3.3%)	0	0.016

Notes: Correlation coefficient. Spearman's rank correlation coefficients for farm size (small = 1, medium = 2 and large = 3) and farm characteristics (no = 0, yes = 1).

Correlation is significant at the 0.05 level (2-tailed).

Correlation is significant at the 0.01 level (2-tailed).

In terms of financial pressures, majority of farms experienced a decrease in operating revenue (59.6%). Some farms were plagued by operating costs, e.g., rent, materials & infrastructure costs (39.5%), labor costs (19.3%) and loan repayment (18.4%), and it was hard for a few farms to get loans (18.4%). And there were still 18.4% of farms without financial pressures.

Most of farms were suffering from seafood price fluctuations (77.2%), followed by weather disasters (38.6%), fish diseases (34.2%), government calling for shutdown (27.2%), lack of technical support (15.8%), and facilities maintenance, repair and renewal (10.5%).

For the coming year, farmers were most concerned about price (price drop or fluctuation, 70.2%) and demand (drop in shellfish demand due to public concerns for seafood safety, 62.3%, and the effect of shrinking demand transmitted to upstream along the supply chain, 32.5%). Nearly half of farmers worried about the rises in labor costs or recruiting difficulty (44.7%) and logistics cost or transport restrictions (43.9%), followed by rent or materials cost (25.4%), stock level (25.4%), difficulties in facilities maintenance, repair and renewal (15.8%) and regulation tightening (10.5%).

Farmers had generated and implemented COVID-19 mitigation strategies amidst vanishing demand, falling prices and an uncertain outlook (FAO, 2020c). The most popular measure was delaying harvesting or reduce harvest volume (57.0%), followed by changing aquaculture practices (47.4%), exploring sales channels (42.1%), changing the number of employees (33.3%), price promotions (24.6%), developing new products (24.6%). It should be noted that a small portion of farms (9.6%), mainly small farms, were quite negative and planned to quit shellfish aquaculture.

Faced with declining demand, most of farms expected the government to help explore sales channels, dock production and marketing (63.2%), and introduce or extending support measures in credit (44.7%), technology (36.8%), logistics (36.0%), taxation (35.1%), agricultural insurance (24.6%), recruitment (21.9%) and social insurance premiums (20.2%).

Differential impact on farms of different sizes

This part focused on the attributes that were significantly correlated (P > 0.05) with farm size. The sample consisted of 30 small farms (26.3%), 61 medium farms (53.5%) and 23 large farms (20.2%). The farm characteristics, effects of the pandemic, problems, concerns, strategies and desired government supports for each farm size and the Spearman's rank correlation coefficients between these attributes and farm sizes are displayed in Table 1 and Table 3, Table 4, Table 5, Table 6 .

Larger farms had higher revenues (ρ = 0.756) and more employees (ρ = 0.864) (Table 1), which is consistent with the definition of farm size in this paper. And they had been in operation for longer periods of time (ρ = 0.237), with 78.3% of large farms that had operated for more than 10 years, a higher proportion than 60.7% of medium farms and 43.3% of small farms. The number of production stages involved in the farm was also related to the farm size (ρ = 0.371) (Table 3). Large farms were involved in 3.4 production stages on average, higher than 2.9 for medium and 1.9 for small. And the proportions of large farms involved in each production stage were higher than that of small farms. Customers of large farms spread over a wide geographical area (Table 3). There were correlations between farm size and whether the farm had customers in other provinces (ρ = 0.315) and abroad (ρ = 0.189), which might explain the interest in logistic supports for larger farms (ρ = 0.255).

Problems and concerns varied by farm size. First, financial situation of large farms was not so critical as that of small farms. Whether the farm was free of financial pressure was correlated to its size (ρ = 0.207). There was no financial pressure for 34.8% of large farms, compared to 16.39% and 10.00% of medium and small farms respectively, probably because larger firms usually had more adequate funding to deal with short-term crises and had easier access to bank financing (Andrieu, Staglianò, & van der Zwan, 2018; Demirgüç-Kunt et al., 2020). Second, large farms were more concerned about technology and facilities. There were correlations between the size of farms and whether they were lack of technical support, had problems in facilities maintenance, repair and renewal, and concerned about facilities maintenance, repair and renewal for the coming year. Farms with above problems and concern accounted for 34.8%, 21.7% and 21.7% of large farms, while 13.1%, 9.8% and 19.7% of medium, and 6.7%, 3.3% and 3.3% of small, which suggested a higher level of mechanization and more reliance on technology on large farms. Third, larger farms were more concerned about the transmission of the effect of shrinking demand to upstream along the supply chain (ρ = 0.282). A higher percentage of large farms (56.5%) have this concern compared to medium (31.1%) and small farms (16.7%), probably because large farms involved in more production stages had realized the impact of the pandemic on the industry from a broader perspective. In terms of response strategies and desirable government supports, there were positive correlations between farm size and whether the farm changed the number of employees (ρ = 0.232), expand to upstream and downstream (ρ = 0.187), broaden distribution channels (ρ = 0.234) and desire agricultural insurance subsidies (ρ = 0.263), social insurance premiums cut or deferral (ρ = 0.203) and logistics support (ρ = 0.255), while there was a negative correlation between farm size and quitting shellfish aquaculture (ρ = −0.279). In addition to the linear relationships measured by the correlation coefficients, compared to large and small farms, a higher percentage of medium farms were more concerned about getting loan (24.6%), regulation tightening (14.7%), recruitment support (27.9%), and tax cut or deferral (42.6%).

The above results reflect the differentiated attitudes of operators from different size of farms in coping with the impact of the pandemic. Despite an uncertain outlook, large farms were motivated to explore sales channels (42.1%) and expand to upstream and downstream (21.7%). Regarding the desired government supports, large farms with wider customers distribution and more employees than medium and small, were more attracted to logistics support (56.5%) and social insurance premiums cut or deferral (39.1%), as well as agricultural insurance subsidies (39.1%). However, small farms handled the pandemic in a negative manner. On the one hand, small farm had less willingness or capacity to expand their operations. Desperate farmers in a quarter of small farms even intended to quit shellfish aquaculture (23.3%), compared to 6.6% of medium and 0 of large. On the other hand, unlike large farms, small farms desired government supports in sales channels (70.0%) and credit (46.7%), which implied that they were suffering more severe stagnant sales and financial constraints than large farms. In addition, compared to large and small farms, a higher percentage of medium farms were more concerned about getting loan (24.6%), regulation tightening (14.7%), recruitment support (27.9%), and tax cut or deferral (42.6%). The divergence in situations and strategies of different size farms in coping with the pandemic, particularly large and small farms, may lead to an increase in the concentration of shellfish farming in the post-pandemic era in China. Although it will hurt small farms, the increased market share of large farms that have utilized more technology and equipment in shellfish aquaculture will improve the efficiency of the industry.

Market survey

Shrinking demand

In the markets we visited, shellfish retailers and wholesalers experienced a significant decline in sales due to market closures and a decrease in customers.

At the retail level, according to the statement of shellfish sellers, public markets in the cities we visited were closed during the lockdown from late January to early February 2020, and consumers turned to supermarkets and online channels to purchase food, including seafood. Although these public markets had not been shut down since then due to the pandemic, whenever there were new COVID-19 cases that were linked to seafood, few customers went to the markets for seafood in the ensuing periods.

After the report of COVID-19 infections linked to seafood stalls in Beijing Xinfadi Agricultural Wholesale Market, the largest wholesale food market in China's capital, on Friday, June 12, over the following weekend “there were almost no customers in the seafood section” in the public markets in all five cities and “the number of people coming to the retail market (non-seafood sections) also dropped by more than half,” retailers said. “While retail sales had recovered to near pre-outbreak levels over the previous weekend.” And after the re-emergent outbreaks in Beijing, we observed that, in seafood sections of public markets in Qingdao, there were almost devoid of customers until Father's Day on June 21 and the Dragon Boat Festival on June 25. And retailers said that the phenomenon had returned to the markets in the cities we surveyed after the seafood-related outbreaks in Dalian in July and Qingdao in September, but “lasted for a shorter period of time”.

At the wholesale level, markets in Beijing and Dalian were severely impacted by the local outbreak, with most wholesalers reporting a drop in turnover of more than half compared to the same period last year. Following the outbreak in June, the Beijing Xinfadi Agricultural Wholesale Market and Jingshen Seafood Markets were closed until September. During that time, the rest of the wholesale markets in Beijing also suspended frozen seafood trading. And wholesalers dealing in live and fresh seafood relocated to other markets, such as Shenghuahonglin Wholesale Market, or continued their business in temporary open-air trading sites. And none of the wholesalers we interviewed who had moved to other markets had intention of moving back to the two markets. In Dalian, after an outbreak linked to a seafood processing plant of Dalian Kaiyang Seafood Co., Ltd. in late July, the nearby Liaoyu International Aquatic Products Market was closed for a month. Although not closed, the rest of seafood markets in Dalian experienced a plunge in sales. At the New Changxing Agricultural & Sideline Products Wholesale Market, hours of operation were reduced from 18 to 7 h “Sales had dropped 70 percent in the two months following the outbreak,” shellfish wholesalers at the market said. “Those who used to import seafood have switched to selling domestic seafood.” At the Tiansheng Seafood Comprehensive Wholesale Market, we observed that only a quarter of seafood stalls were in business. A shellfish wholesaler said that before the pandemic they mainly sold shellfish to consumers and barbecue stalls, but at the moment the former hardly come to the market and the latter had halved their purchases.

In Qingdao, where seafood wholesale markets remained open despite the seafood-related outbreak, shellfish sales dropped by half, as retailers and caterers reduced their purchase. As for the seafood markets in Nanjing and Jinan, without seafood-related outbreaks, most wholesaler reported a 30–50 percent decline in sales, mainly due to a drop in catering purchases.

Despite the decline in sales, few wholesalers reduced prices for promotions. The reason explained by wholesalers in these cities was that farmers refused to reduce prices, which was consistent with the results of our farm-level survey, where only 24.6% of farms were willing to conduct prices promotions.

In addition, despite the government promoting rent exemption and reduction (Chinadaily, 2020; SASAC, 2020), majority of wholesalers did not get rent relief. We asked the management of the Jinan Seafood Market about this issue, and they replied:

All stalls in the market have been sold. Wholesalers in the market are the owners or renters of the stalls. We have appealed to lessors to reduce the rent for their renters. However, it is up to the lessors and the renters to negotiate whether and how much to reduce, and we have no authority to require them to adopt a particular option.

Wholesalers at the Tiansheng Seafood Comprehensive Wholesale Market in Dalian experienced a similar situation:

We rent our stall from the market. The market had promised to waive part of the rent, but there was a delay in action. We were told that the waived rent would be paid by the market rather than a government subsidy.

It follows that the availability of rent relief to wholesalers depends largely on the willingness of lessors of stalls. And the lessors had no incentive to waive the rent for the renters in the absence of subsidies.

In addition to not getting rent relief, wholesalers had to bear the cost of regular PCR tests on their employees and products to ensure their seafood were free of the virus. Due to the increase in operating costs and decrease in sales, some wholesalers suspended their business.

Distribution channel improvements

The results of the market survey showed a decline in demand and the plight of sellers during the pandemic. And the farm-level survey indicated that, the shrinking demand had a more severe impact on farms compared to the disruption of production caused by the lockdown, and that the farms, especially small ones, desired the government to help explore distribution channels. Therefore, dispelling consumers’ fears to foster demand recovery has been the key to mitigating the impact of the pandemic on shellfish supply chain participants.

To allay consumer concerns about the safety of buying and consuming seafood, a series of measures have been implemented to reduce the risk of virus transmission in the markets we investigated. As frozen seafood and humid environment are thought to have created the conditions for the virus to lurk and spread in the seafood market (Wang, 2020), these measures have focused on the separation of live & fresh from chilled & frozen products in the stage of distribution and the improvement of the market environment.

First, after the outbreak in Beijing Xinfadi Wholesale Market, regular PCR tests were conducted in all the wholesale markets we investigated to ensure the seafood on sale was free of the virus. Second, renovations of the markets, especially the seafood sections, have been implemented or planned.

In terms of wholesale market renovation, Beijing has been a step ahead of other cities. In the past, wholesale markets served wholesalers, retailers, caterers and consumers, with both wholesale and retail functions. Now, wholesale markets in Beijing are no longer available to individual consumers (Global Times, 2020; Xinhua, 2020b). And a site has been separated from the wholesale market to serve the surrounding residents as a public market, such Yuegezhuang Convenient Service Center separated from Yuegezhuang Wholesale Market and Xinfadi Convenient Food Market from Xinfadi Agricultural Wholesale Market. In wholesale markets, seafood section is divided into live & fresh part and chilled & frozen part, and the storage for the two types of products have also been separated. And some markets have also launched online platforms to trade seafood, including frozen seafood that have been suspended from on-site trading, such as Jingshen E-Home, a WeChat mini program launched by Jingshen Seafood Market (Global Times, 2020). As for wholesale markets in other cities, according to our interviews with market managers, there have been plans to renovate the markets, especially seafood sections.

However, at the retail level, it's difficult to partition seafood section based on product types in public markets, as a stall may sell both fresh & frozen, domestic & imported seafood. As for market environment, after renovations that have been implemented before the pandemic, the markets are divided into several sections, including seafood section, in which the floor is covered with non-slip tiles and the drainage facilities are superior to the other sections. But we observed that workers at seafood stalls would dump wastewater including fish water and ice-melting water in front of the stalls, which made the ground slippery and the air humid. As a result, drainage facilities are convenient for the stalls but do little to improve the market environment. And the layout of sections in some markets is not appropriate. A customer needs to cross several other sections to buy seafood, and sewage is also brought to the sections with poor sewage facilities. To solve these problems, it is necessary to regulate the behavior of the staff in the market and to improve the market layouts in the subsequent renovation.

Additionally, in public markets, if products are found to be defective or tested positive for the virus, it's hardly possible to accurately trace the unpackaged and unlabeled seafood products that are sold to numerous customers without any record of business transactions. In public markets, customers pay as they buy at each stall, instead of settling all at the exit as in supermarkets, so shopping receipts are not applicable to these markets. Nevertheless, electronic payments have created the conditions for recording these fragmented transactions, and new retail channels, such as community group buying, also offer a path to solving the problems at the retail level. Community group buying, a new type of location-based social e-commerce model, is changing how people engage in food shopping (Chen, 2019a; Weinswig, 2020) and have appeared superior in satisfying the consumer needs and easing the panic buying perception during the pandemic (Hao et al., 2020). The model usually centers on a group of residents who live in the same community. Consumers place their order through online platforms, mainly launched or supported by China's internet giants such as Alibaba and Tencent. Products typically from selected suppliers are sent to regional central distribution centers after an order is placed and then delivered to collection points in communities the next day (Weinswig, 2020), which partially replace the role of public markets in retail. And the model can realize the digitalization of retail as commodity information and the process of delivering products from suppliers to consumers are recorded by the platforms, which provides a new solution to achieve the traceability in fresh grocery retail. However, the involvement of e-commerce giants has raised public concerns about consumer privacy, monopoly, and livelihood of vendors in public markets. China government has enhanced regulations of these community group buying platforms (Li, 2020; SAMR, 2020).

Conclusions

A farm-level survey and a market survey were conducted in China to study the impact of the COVID-19 on the shellfish supply chain in the early stage of the pandemic, especially the lockdown and the re-emergent outbreaks linked to seafood in 2020. Relying on local employees and suppliers, shellfish farming activities have rapidly resumed since the gradual reopening from February, and some farms even remained in operations during the lockdown. Nevertheless, shellfish farmers have experienced a significant drop in profits, largely due to shrinking demand. Capacity and attitude of farms to cope with the pandemic varied by farm size. Despite an uncertain outlook, large farms were motivated to explore sales channels and expand their operations to upstream and downstream, while small farms did not have such willingness or ability, with a quarter of small farms even planning to quit shellfish aquaculture. The government is supposed to support farms, especially small farms, in marketing. Large farms also seek supports for technology and logistics. And for medium farm, tax relief and financing facilitation are more attractive. As for the distribution, shellfish sellers were struggling with fewer customers, market closures, and costs for pandemic prevention. Most of them have not received subsidies, including rent waivers. Additionally, to reduce the risk of pandemic transmission in seafood markets, wholesale markets in Beijing have conducted regular PCR tests, separated retail from wholesale markets, and isolated live & fresh and chilled & frozen seafood. And we recommend further improvements of markets in other cities, especially wet markets.

The farm-level survey ended in August 2020 and the market survey in November 2020, while the ongoing pandemic still have an uncertain impact on China's shellfish supply chain. As the recurrences of the pandemic and the late harvest continue to hit the market, the government should avoid prematurely withdrawing support policies for farmers. In the post-pandemic era, market renovations should be implemented in more cities to avoid the spread of disease in wholesale and public markets. And government, e-commerce giants and traditional sellers are supposed to work together to explore the capabilities of electronic payments and e-commerce platforms in enabling traceability of seafood. As for farmers, it's an important proposition to improve flexibility in their supply chains and distribution channels to prepare for another potential crisis. Moreover, the pandemic may also lead to a temporary or permanent transition in shellfish consumption patterns, which was not covered in this study. A series of systematical studies need to be carried out in more countries with different patterns of production, distribution and consumption to fully understand the effects of the pandemic.

Declaration of competing interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

City		Market	Type	Num of interviewed stalls
Qingdao		Chengyang Vegetable & Aquatic Products Wholesale Market	Wholesale	10 + 1
Status	Sub-provincial	Fushun Road Vegetable & Food Wholesale Market	Wholesale	10 + 1
Population	10.07 M	Benxi Road Public Market	Retail	10
Visiting month	Jun–Nov	Nanshan Public Market	Retail	10
		Tuandao Public Market	Retail	5
Dalian		Tiansheng Seafood Comprehensive Wholesale Market	Wholesale	10 + 1
Status	Sub-provincial	Liaoyu International Aquatic Products Market	Wholesale	10
Population	7.45	New Changxing Agricultural & Sideline Products Wholesale Market	Wholesale	4
Visiting month	Sep	Huabei Road Market	Retail	5
Nanjing		Nanjing Agricultural & Sideline Products Logistics Center	Wholesale	6 + 1
Status	Provincial capital	Zhuquan Public Market	Retail	5
Population	7.45 M	Jinxianghe Public Market	Retail	3
Visiting month	Nov	Jianyelu Public Market	Retail	3
Beijing		Yuegezhuang Wholesale Market	Wholesale	10 + 1
Status	Capital	Jingshen Seafood Market	Wholesale	10 + 1
Population	21.89 M	Xinfadi Agricultural Wholesale Market	Wholesale	10
Visiting month	Nov	Shenghuahonglin Wholesale Market	Wholesale	10 + 1
		Yuegezhuang Convenient Service Center	Retail	3
		Xinfadi Convenient Food Market	Retail	3
Jinan		Jinan Seafood Market	Wholesale	10 + 1
Status	Provincial capital	KuangShan Agricultural Products Comprehensive Trading Market	Wholesale	10
Population	9.20 M	Weierkang Meat & Aquatic Products Wholesale Market	Wholesale	8
Visiting month	Nov	Qilishan Agricultural Products Market	Retail	4

Note: “+1” in Num of interviewed stalls indicates that the manager of the market was interviewed.