Price premiums for eco-labelled seafood: effects of the MSC certification suspension in the Baltic Sea cod fishery

Abstract

1. Introduction

Sustainability labels, or eco-labels, on food products are important in many markets. In the seafood market, certification programmes have been introduced as a market-based tool for promoting sustainable fishing practices. The criteria of the certification schemes vary between eco-labels, but they generally focus on stock sustainability and method of harvest.¹ Eco-labels may therefore provide otherwise unobservable information to the consumers about the environmental sustainability of the fish product. The premise is that consumers with a preference for sustainable seafood will shift their demand towards labelled fish that, in turn, generates price premiums of eco-labelled products. In this way, producers are rewarded for fishing in a sustainable way (Gudmundsson and Wessells, 2000).

The use of eco-labels in fisheries has increased dramatically in the recent decade. The world’s leading standard for sustainable fishing traceability, the Marine Stewardship Council (MSC), founded in 1997, comprises around 300 fisheries in more than 30 countries. Seafood from MSC certified fisheries represents around 10 per cent of the global harvest of wild capture fisheries (Marine Stewardship Council, 2016). The widespread use of certification programmes and eco-labels in fisheries, and their effects on environment and seafood markets, has generated much research interest in recent years. A number of studies have examined whether certification causes environmental improvements (e.g. Jacquet et al., 2010; Froese and Proelß, 2012; Martin et al., 2012; Agnew et al., 2013; Christian et al., 2013; Agnew, Gutierrez and Hoggarth, 2014; Opitz et al., 2016). This research has documented some improvements in e.g. stock status and by-catches, but there is no clear consensus across studies about the environmental benefits of MSC certification.

Another branch of research has examined the existence of price premiums for eco-labelled fish products. Evidence from stated preference surveys and field experiments show that consumers generally express a preference for eco-labelled seafood (Johnston et al., 2001; Brécard et al., 2009; Uchida et al., 2013; Chen, Alfnes and Rickertsen, 2015). In line with these findings, studies using scanner data typically confirm the existence of price premiums in the retail market for eco-labelled fish products (Roheim, Asche and Insignares, 2011; Sogn-Grundvåg, Larsen and Young, 2013; Sogn-Grundvåg, Larsen and Young, 2014; Asche et al., 2015a; Bronnmann and Asche, 2016; Asche and Bronnmann, 2017) as well as for organically farmed seafood (Asche et al. 2015a; Ankamah-Yeboah, Nielsen and Nielsen, 2016; Carlucci et al., 2017). The price premiums in most studies are in the range of 10–15 per cent for MSC-labelled fish, but the premium varies between countries and fish products.

A price premium at the retail level does not, however, necessarily imply a premium at the producer level. Studies on price transmission along the food chain have emphasised that producers may not fully benefit from a price increase at the retail level because of, for example, oligopolistic behaviour and market power in the retail and processing industries, menu costs etc. (see e.g. Meyer and von Cramon-Taubadel, 2004; Vavra and Goodwin, 2005; Bakucs, Falkowski and Fertő, 2014). Whether or not price premiums for eco-labelled fish are transmitted down the value chain to the producer level is an important issue because a price premium signals a return on investment in sustainable fishing practices. On the other hand, absence of a price premium may discourage fishermen to engage in certification as they, in the long run, need proof of market benefits to justify the costs of certification (e.g. Goyert, Sagarin and Annala, 2010).² This has recently been underscored as leading Alaskan salmon processors in January 2012 withdrew from MSC certification as the industry organisation no longer perceived the MSC certification costs as justifiable (Alaska Fisheries Development Foundation, 2012).³ However, there is surprisingly little evidence on the existence, and extent, of price premiums for eco-labelled fish at the producer level. The few existing studies do not provide strong evidence of price premiums in contrast to the results from retail-level studies (Wakamatsu, 2014; Blomquist, Bartolino and Waldo, 2015; Stemle, Hirotsugu and Roheim, 2016).

The purpose of this paper is to add to the literature on MSC price premiums at the producer level. The paper contributes to the literature in two ways. First, the paper takes a new and more precise approach to empirically identify the price premium compared to previous studies. Instead of focusing on the effects of gaining a new MSC certificate, the paper analyses the effects of a suspension. The benefit of this approach is that the effect of a suspension on the ex-vessel price is likely to be immediate. In contrast, the price premium of gaining a new certificate may take many months, perhaps years, to be realised after certification, which makes empirical identification of the premium difficult. The paper’s focus on a certification suspension alleviates this problem, hence enabling a more precise identification method.

Second, the paper pays particular attention to potential heterogeneity of the MSC price premium. If the premium varies between fish products, which is suggested by studies at the retail level (e.g. Asche et al. 2015a; Asche and Bronnmann, 2017), MSC certification is likely to provide an incentive for sustainable fishing in some cases, but not necessarily in all. Thus, the way the premium is transmitted to producers and how it maintains incentives may be more complex than assumed in previous studies. This paper extends the literature by examining whether the MSC price premium differs across different sizes of fish, since different body sizes have different demand and use (e.g. Zimmermann and Heino, 2013).

To quantify the MSC price premium at the producer level, we use a detailed dataset of the Swedish Baltic cod fishery including 9,720 daily ex-vessel price observations from January 2014 to December 2016. The MSC certification was suspended in December 2015 because the scientific assessment was insufficient to provide advice on stock status and reference points. Thus, it was not a decision by the fishing industry to leave the MSC certification, which makes the Baltic cod fishery an appropriate case study for evaluating the price premium. The Baltic cod fishery also provides an ideal situation for examining the price premium because not all landings from the Baltic were qualified for certification. In particular, we make use of the fact that only the eastern Baltic was certified according to the MSC standards. Cod fishing in the western Baltic was not MSC certified and is hence used as a control group. Finally, the price premium is operationalised as the price difference between the two groups in a panel data difference-in-difference (DID) model.

2. Background and previous literature

The MSC is an international organisation aiming to contribute to the health of the world’s oceans by recognising and rewarding sustainable fishing practices. The interest in MSC certification programmes has increased dramatically in the recent decade; in the last 8 years, from 2008 to 2016, the number of certified fisheries increased from 26 to 286 (Marine Stewardship Council, 2016). As the number of certified fisheries has grown, so has the market for MSC-labelled fish products. Today, there are more than 20,000 MSC-labelled products on sale globally with a retail value of US$ 4.6 billion annually (Marine Stewardship Council, 2016).

Whether the producers receive a price premium for providing MSC certified seafood to the market is an important issue. Producers of certified products and those considering assessment for certification must perceive that the benefits of certification are large enough to offset the costs. However, there is scant empirical evidence on MSC price premiums at the producer level. Wakamatsu (2014) analyses the effect of MSC certification on ex-vessel prices for flathead flounder in three Japanese fish markets and finds no evidence of a price premium. On the other hand, Wakamatsu finds that the market for flounder has become more segmented (less influenced by other markets) after the introduction of the MSC certification. Bellchambers, Phillips and Pérez-Ramírez (2016) discuss the lack of a MSC price premium of lobster in the western Australian and the Mexican Baja California fisheries. The authors note that the vast majority of both lobster species sold are not tagged with the MSC logo due to the additional cost associated with its use. Using a DID analysis of MSC-certified and non-certified landings, Blomquist, Bartolino and Waldo (2015) find no evidence of a price premium in the Swedish Baltic cod fishery after its certification in 2011. Stemle, Hirotsugu and Roheim (2016), also using a DID methodology, analyse potential MSC price premiums from three different fisheries: salmon and halibut in Alaska, and flathead flounder in Japan. The authors found mixed results depending on fish species. A positive premium is found for chum and pink salmon and for flathead flounder, while no premium is found for halibut and chinook, coho and sockeye salmon.

Thus, in contrast to the results from retail-level studies, the few existing studies do not provide strong evidence of a MSC price premium at the producer level. However, estimating the price premium of MSC certification is challenging. Previous studies of the price premium, such as Blomquist, Bartolino and Waldo (2015) and Stemle, Hirotsugu and Roheim (2016), specify DID models assuming an immediate price premium after the date of certification. This approach may result in underestimated price premiums if it takes time for the price premium to be realised. Indeed, before the MSC label can be used on fish products, an assessment must take place at each step in the supply chain. Processors must be certified against the MSC’s chain of custody standard before they can label the fish products as MSC-certified.⁴ Similarly, restaurants who want to use the MSC label on their menu need a licence agreement. The point in time when the product obtains the MSC label at the consumer level may therefore not be same as the date of the fishery certification, which may lead to bias when estimating the price premium at the producer level.

In this paper, we overcome this problem by focusing on suspension of a MSC certificate, rather than on gaining a new certificate. The main advantage of this approach compared to e.g. Blomquist, Bartolino and Waldo (2015) is that, in case of a suspension, fish caught after the date of suspension may no longer be sold as MSC certified and cannot carry the MSC label. In other words, there is no time lag between (suspended) certification of the fishery and (suspended) labelling of the fish product. A suspended certification implies that the marketer of the fish product must drop the MSC label, or the supplier must find another market that does not require MSC certification. In both situations, we would expect a sudden drop in ex-vessel prices after suspension if there is a price premium for providing MSC certified fish. The strategy of using a suspended certification to identify the price premium relies on the assumption that the suspension is exogenous with respect to prices. That is, if the fishery withdraws from MSC certification because of lacking benefits, this strategy is certainly inapplicable (because of reversed causality). However, as we discuss further below, in this study the exogeneity assumption can be deemed to hold as the MSC certification was suspended because of insufficient scientific assessment not related to economic incentives.

3. The Swedish Baltic cod fishery and MSC certification

The cod fishery is one of the most important fisheries in Sweden. In 2015, the landings of cod constituted around 10 per cent of the total catch value of fish and seafood in Sweden (Swedish Agency of Marine and Water Management, www.havochvatten.se). The absolute majority of the Swedish cod catches are taken in the Baltic Sea, where the fishery targets two different stocks: the eastern (ICES subdivisions 25–32) and western Baltic (subdivisions 22–24) cod stocks.⁵ In total, 37,342 tons of cod were landed from the eastern Baltic in 2015, of which around 12 per cent were landed by Swedish vessels (ICES, 2017a). The fishery in the western Baltic is smaller, in total 13,418 tons of cod in 2015 (ICES, 2017a). The quotas for the Baltic cod stocks are set by the EU each year but member states have great flexibility to allocate national quotas among their vessels. The Swedish Baltic cod quota is allocated to different fleet segments depending on gear types and regulated by yearly catch rations for each vessel.⁶ In 2015, around 75 per cent of the total Swedish cod catches were made by vessels using demersal trawls. The demersal trawlers fish for cod in both the western and the eastern Baltic and the catch are sold to buyers mainly in Sweden, Denmark and Poland, where it is processed and distributed to a variety of European markets.

The MSC assessment process for the eastern and western Baltic cod fishery started in January 2010. The catches in the western Baltic were at that time well above the target level for sustainability (above the Maximum Sustainable Yield target) and the western Baltic fishery was excluded from the MSC assessment. The eastern Baltic cod fishery received its MSC certification in June 2011. The MSC certificate covered fishing vessels holding licenses to fish for the eastern Baltic cod with the following gear types: trawl, longline and traps.⁷ Three months after certification, in September 2011, the Swedish food processor Findus was first in line to supply frozen cod fillet from the Baltic to the Swedish market, and about 1 month later McDonald’s announced that MSC-certified cod from the Baltic was used in the production of the Filet-O-Fish in Sweden.⁸ Since then, the major food processors and retail chains in Sweden (such as Coop, ICA and Lidl) have adopted policies to increase their supply of MSC labelled seafood. For example, in 2014, Lidl Sweden announced that all of their wild-caught frozen fish were MSC certified.

In December 2015, the MSC certificate for the eastern Baltic cod fishery was suspended. Independent assessment of the fisheries showed that the eastern Baltic cod stock no longer met the requirements of the MSC Fisheries Standard. More specifically, uncertainties with the estimation of the age of individual fish and vast changes in the growth of part of the stock prevented the International Council for the Exploration of the Sea (ICES) to provide an analytical assessment of the stock status in relation to reference points, which is necessary for a biologically sustainable management of the fishery.⁹ The implication was that cod caught in the eastern Baltic (ICES subdivisions 25–32) from 17 December 2015 could no longer be bought or sold with the MSC label. The fisheries producer organisations and processors active in the fishery feared that the suspension of the MSC certificate would limit market access and cause economic losses for the fishermen. This was perceived to be especially problematic for the smaller size classes of cod typically used as raw material in value-added products like McDonald’s Filet-O-Fish, which requires MSC certified fish. The fisheries organisations therefore formed a corrective action plan with the objective to remove the certificate suspension, but without success.¹⁰ Thus, the eastern Baltic cod fishery was MSC-certified in a period of four and a half years before suspension. During this period, the fishery enjoyed enhanced reputation among consumers and was given the green light in the WWF fishery guide for the first time in 8 years. While there is no information on the quantities of landings from the certified fishery that end up being MSC-labelled products, it is clear that MSC certification has been important in gaining market access, at least in Sweden.

Blomquist, Bartolino and Waldo (2015) evaluated the presence of a price premium at the ex-vessel level in the eastern Baltic cod fishery. Using ex-vessel price data covering 6 months before and 18 months after the introduction of MSC, the study found no evidence of a price premium after certification. However, although there was no immediate price effect, a price premium may arise over time as processors, retailers and restaurants obtain their MSC chain-of-custody certification and subsequently increase the supply of eco-labelled fish. For example, in the period 2014–2015, 3–4 years after certification of the eastern Baltic cod, the major European brands of frozen fish products, such as Findus and Birds Eye, declared that all of their wild-caught cod products were sourced from MSC certified fisheries (https://www.msc.org/newsroom/, last accessed 10 August 2017).

4. Data and empirical approach

Information about landings of cod comes from two databases provided by the Swedish Agency for Marine and Water Management (SwAM). The first database consists of sales notes and includes information about prices, buyers and sellers, landing ports and size and quality of the cod. This information is matched to a second database consisting of log-book data with information about the geographical position and fishing gear used for each fishing activity. The information from the latter source is used to discriminate between catches of cod in the MSC-certified eastern Baltic and catches in the non-certified western Baltic. Information about which vessels are MSC certified was obtained from Fiskbranschen (www.certifieratfiske.se). From these, data it is possible to create a binary variable, MSC fishery, which takes the value of one if the landing comes from the MSC-certified fishery in the eastern Baltic. In other words, the variable is set to one if the vessel holds a MSC certification and the cod was caught by MSC-certified gears in the eastern Baltic before the 17 December 2015. Information about which buyers are MSC certified (chain-of-custody certification) was obtained from MSC (www.msc.org) and matched to the sale notes. A second variable, MSC landing, could then be generated which takes the value of 1 if the buyer is chain-of-custody certified and the variable MSC fishery is equal to 1. By constructing the MSC variables in this way, landings of cod from the western Baltic, as well as landings from vessels and buyers that are not MSC certified, are used as a control group in the analysis.

It should also be noted that the variables MSC fishery and MSC landing capture different aspects of a potential price premium. On the one hand, vessels may gain a price premium for supplying fish from a MSC-certified fishery, irrespective of whether or not the buyer holds a chain-of-custody certificate. For example, processors and retailers may have adopted sustainability policies to source seafood from sustainable fisheries. This effect will be captured by the variable MSC fishery. On the other hand, if there is a MSC price premium associated with the eco-label at the consumer level, buyers of fish may be willing to pay fishermen higher prices for MSC certified landings that are also chain-of-custody certified, which will be captured by the variable MSC landing. Because not all cod from the MSC fishery flow through processors/buyers with the chain-of-custody certification, the distinction between the two variables is important.

In addition to the MSC variables, information was gathered on catch method (MSC certified gear types include trawl, longline, trap), as well as quality rating and size class of the fish. Size classes and quality ratings are defined by the European Commission (European Commission, 1996). The size classes include very small (0.3–1 kg), small (1–2 kg), medium (2–4 kg), large (4–7 kg) and very large (over 7 kg).¹¹ In addition, because of the EU landing obligation in force in the Baltic Sea from January 2015, there is one size class for cod under 35 cm, which is the minimum conservation reference size (MCRS). Cod below MCRS must be landed and cannot be marketed for direct human consumption purposes, but may be used for e.g. fish meal, fish oil, animal feed etc. The quality classes include high quality (class E), medium quality (class A) and low quality (class B).¹²

Daily landing observations from 1 January 2015 to 31 December 2016 give a total of 9,720 ex-vessel price observations, of which around 60 per cent derive from the period before the suspension of the certificate (17 December 2015). Mean value and standard deviation of the variables are shown in Table 1, where the descriptive statistics are shown separately for MSC-qualified and non-qualified landings. In order to compare the covariates between the ‘treatment’ and ‘control’ groups, landings that would have been MSC-qualified if not for the suspension of the certificate are included for the whole time period in the column ‘MSC-qualified’. In total, 6,213 out of the 9,720 observations (64 per cent) are landings that are not MSC-qualified. As can be seen from the table, the price is somewhat lower for the MSC-qualified landings, which is opposite to what one may expect if there is a price premium. This may, however, be a result of difference in catch composition between the two groups. For example, the shares of undersized (under MCRS) and low quality (Quality B) cod are lower for non-qualified landings. This is further discussed in the next sections.

5. Model specification

The vessel- and buyer-specific fixed effects, $ηv$ and $γb$⁠, are included to control for permanent (time-invariant) firm-specific effects that may influence the ex-vessel price.¹⁵ The importance of controlling for such effects has been highlighted in previous studies showing that different fishing vessels may receive different prices, depending not only on observed characteristics of the fish (such as size and quality), but also on the specific buyer of the fish (see e.g. Lee, 2014; Asche, Chen and Smith, 2015b; Gobillon, Wolff and Guillotreau, 2017). Thus, the panel data model makes it possible to control for unobserved self-selection problems that may otherwise bias the estimated price premium.

It may also be interesting to analyse the coefficients of the included covariates to see if, for example, larger cod commands a higher price per kilo. Since a constant is included in the equations, the parameters $βj$ should be interpreted as the price deviations from a baseline product with a given set of features. In our model, the baseline is cod of very small size (0.3–1 kg) and medium quality (Quality A), caught by a vessel using trawl and not MSC certified. To estimate the standard errors of the coefficients, we follow the suggestion by Bertrand, Duflo and Mullainathan (2004) and use Arellano's (1987) clustered covariance matrix, which allows for both serial correlation and heteroscedasticity in the errors.

6. Empirical results

The results from the simple DID estimator in equation (1) are given in Table 2. The results show that the price of both MSC qualified and non-qualified landings has increased in the period after the suspension of the MSC certificate. It is also evident that the price of non-certified landings increased more than the price of MSC landings; the average price difference between the two groups before and after the suspension, the DID estimator, is 0.93 SEK and is significantly different from zero at the 5 per cent significance level. Based on the results of Table 2, the loss in MSC price premium is around 9 per cent.

The next step is to examine if this result holds when including the covariates and the fixed effects in the regression model. The results from the regressions of equations (2) and (3) are shown in Models I and II of Table 3, respectively. Looking first at Model I, we see that the coefficient on the variable MSC landing (top panel) is positive and statistically different from zero at the 5 per cent significant level and the DID estimate of the price premium is 0.51 SEK/kg. The premium is somewhat lower than the simple DID estimate, which indicates that the included covariates capture some of the variation in the group-specific time trends. Model II shows the results from equation (3), which includes interaction terms for the size dummies and the variable MSC landing. The interaction terms show if the MSC price premium differs between cod of different sizes, which may be the case if cod of different sizes is used for different products. As can be seen from these results, the coefficient on MSC landing almost doubles while the interaction terms are all negative and statistically different from zero. The likelihood ratio test comparing the restricted (Model I) and unrestricted (Model II) models gives a test statistic of 66.47, which should be compared to the critical value of the $χ2(4)$ distribution of 9.49. Thus, the restrictions imposed by Model I are firmly rejected in favour of Model II.

The results from Model II show that there is a price premium of about 1 SEK/kg for cod of very small size but effectively no, or even negative, price effects for the other size classes. To test this interpretation more formally, we perform two types of Wald tests. First, we test the restriction that the coefficients on MSC landing, MSC fishery and all the interaction coefficients are zero. This yields a χ²(6) test statistic of 25.82, and the null hypothesis of no price effects of MSC is rejected at the 5 per cent significance level. Second, we perform four separate Wald tests of the restriction that the sum of coefficients on MSC landing and each interaction term is equal to zero. These tests assess the null hypotheses that there are no price premiums for the size categories other than the baseline category (very small cod). It turns out that the null hypothesis of no MSC price premium cannot be rejected at the 5 per cent significance level for any of the size categories. The interpretation of this result is that there has been a MSC price premium of 1 SEK/kg, or 11.3 per cent, for landings of cod of very small size (0.3–1 kg) during the period before the suspension of the certificate, but no positive or negative premium for the other size classes.

Looking at the other variables in Model II, we see the variable MSC fishery is not statistically different from zero indicating that there is no price premium for providing fish from a MSC certified fishery, unless the buyer is also chain-of-custody certified. This result is expected given that these landings may not be marketed with the MSC label. Size and quality are important determinants of ex-vessel prices; compared to the price of very small cod (0.3–1 kg) of medium quality (quality A), the price is significantly higher for cod in the larger size classes and the price increases with size category. The regression predicted ex-vessel price for the baseline category is 8.73 SEK/kg, which implies that the price per kilo for large cod (>4 kg) is twice as high as the price for the very small cod (0.3–1 kg). It is also evident that the price is significantly lower for cod below the MCRS, which is expected given that cod in this category is not used for direct human consumption.

The grey line in Figure 1 shows the daily average of $p̃tlsq$ when calculated using all size categories except the very small cod size class. The black line in the figure shows the 15 days moving average and the dashed vertical line indicates the date of suspension. Figure 2 shows the corresponding graphs using only the very small cod size class. Looking first at Figure 1, we see that the price difference fluctuates around zero and there is no clear trend before the suspension of the certificate, which indicates that the control group is appropriate. The fact that $p̃tlsq$ varies around zero during the whole time period is in line with the regression results in Model II of Table 3, which showed no MSC price premium for these size categories.

Fig. 1.

Price differences (SEK/kilo) for all size categories except very small cod (0.3–1 kg).

Open in new tabDownload slide

Fig. 2.

Price differences (SEK/kilo) for very small cod (0.3–1 kg).

Open in new tabDownload slide

Figure 2 shows a different pattern. In the period before MSC suspension, there is a positive price difference for cod of very small size and the price difference is rarely below zero. As can be seen from the figure, the price difference is rather constant at around 2 SEK/kilo and there is no evidence of a trend in the $p̃tlsq$ series before the suspension of the certificate. This provides some confidence that the control group is appropriate and the regression design above is adequate to identify the MSC price premium.¹⁷ As expected from the regression results, after the date of suspension, the price difference drops significantly. It is interesting to note that there is no evidence that the loss of price premium was only temporary, as the price difference varies around zero throughout the entire post-MSC period.

The average price difference in Figure 2 decreases from 2.28 SEK/kg before the suspension to −0.09 SEK/kg afterwards. This result indicates an even larger price premium than the regression results in Table 3. However, it should be noted that the analysis above, in contrast to the regression approach, cannot account for vessel and buyer heterogeneity. If MSC-certified buyers pay a higher price for cod irrespective of MSC certification, there would be positive price differences between MSC and non-MSC landings not associated with the certification as such (see Blomquist, Bartolino and Waldo, 2015). However, neither buyer-specific nor vessel-specific effects could explain why the price difference drops immediately after the date of suspension. Figures 1 and 2 therefore provide convincing evidence that the suspension of the MSC certification indeed caused the ex-vessel price to drop for cod of size 0.3–1 kg.

7. Summary and discussion

This paper analyses the dockside level price effects of losing a MSC certificate. The aim is to investigate whether suspension of a certificate causes a reduction in ex-vessel prices for certified landings relative to non-certified landings. Detailed data from the Swedish Baltic cod fishery, which had its MSC certification suspended in December 2015, are used to estimate the price effects using a difference-in-difference approach. The results show a statistically significant positive price premium for small-size cod prior to the suspension of the certificate, indicating that the market position for the certified landings deteriorated relative to non-certified landings. More specifically, we find that MSC certification suspension caused the ex-vessel price for very small cod (0.3–1 kg) to drop by 1 SEK/kg (around 11 per cent) while the results indicate no price premiums for the other size categories. Although cod in size category 0.3–1 kg yields a low price per kilo compared to the larger size classes, these landings are economically important as they constitute around 70 per cent of total landed weight in the period 2015–2016.

The results of this paper have implications for both research and practice. For example, it is interesting to compare our results with those of Blomquist, Bartolino and Waldo (2015), who found no price premium of gaining the MSC certification in the Baltic Sea cod fishery in 2011. While Blomquist, Bartolino and Waldo (2015) show that there was no immediate price effect of the MSC certification, market benefits may build up over time as the supply chain obtain their MSC chain-of-custody certification. It is therefore not surprising that a suspended certificate has a more direct price effect than introducing a new certificate. Another issue highlighted in the current paper is that there is no price premium for providing fish from a certified fishery unless the buyer is also MSC certified. Given the fact that not all fish products from MSC certified fisheries reach the consumer bearing the MSC label, our results highlight the importance to identify certified buyers, and not only landings from certified fisheries, when estimating the price premium at the producer level.

Another contribution of the paper is the analysis of price heterogeneity by examining the MSC price premium for different sizes of fish. Our results show that whether or not a price premium exists depends on the size of cod, and that the positive premium for the small-size Baltic cod may not necessarily show up in the estimation results for the aggregate. A plausible explanation for the differences in price premiums is that different sizes of cod are used for different fish products, which are distributed to different markets with different valuations of the MSC label. This would be in line with previous research that has found eco-label price premiums at the consumer level to be heterogeneous with respect to fish product, retailer and geographical market (e.g. Johnston et al., 2001; Asche et al., 2015a; Bronnmann and Asche, 2016; Asche and Bronnmann, 2017). Our finding of a heterogeneous price premium also at the producer level is an important result, as it suggests a more complex picture of how certification programmes may provide incentives for sustainable fisheries practices. The MSC certification applies to all fish from a fishery, not specifically to a subset of seafood products, which has called into question whether the price premium can be transmitted down to individual fishing vessels (e.g. Asche et al., 2015a). The results of this paper indicate that it can, as vessels seem to receive different price premiums for different sizes of fish.

Previous research on price premiums at the retail level has emphasised that ex ante advice of whether an eco-label would generate market benefits for a particular fishery requires specific information about the supply chain and the markets it serves (Asche and Bronnmann, 2017). The results from the present study underscore this point, as well as indicate that the catch composition of the fishery may be an important factor when assessing potential price premiums at the producer level. This may be particularly important when considering the MSC certification of mixed fisheries where the catch composition, with respect to both size and species, is likely to influence the price premium.

The results of this paper are particularly interesting given the recent changes in fisheries management of Baltic cod. For example, a landing obligation was formulated in the European Union in the new Common Fisheries Policy (CFP) in 2013. This was implemented in 2015 for the Baltic cod fishery and implies that all catches are landed (no discards of cod are allowed). About the same time, the minimum length of cod was reduced from 38 cm to 35 cm. Together, these management reforms have implied larger landings of small-sized cod. Losing the MSC price premium for the small size cod is therefore particularly problematic. Notably, the suspension of the MSC certification is also closely related to management since the cause of the suspension was not over-fishing but rather that the stock assessment was insufficient to provide advice on the stock status.

Overall, the role of eco-labels as an incentive to fish sustainability is not yet well understood. Researchers are only beginning to assess whether market benefits such as price premiums are transmitted down the value chain to the fishers. The results of this paper provide some insights in this issue by illustrating the need to account for heterogeneity of fish products when analysing dockside prices of certified fish. The paper also points out the methodological advantage of examining the effects of losing a MSC certificate, instead of focusing on gaining a new certificate. This approach may also be a useful empirical strategy in other situations where it may take time for market benefits of eco-labels or other certification schemes to materialise. More research is, however, needed to disentangle the complexity of price transmission and economic incentives in seafood markets. Indeed the profitability of eco-label adoption for the fishers, who ultimately affect sustainability, is in the long-run crucial for the proliferation of eco-labels such as the MSC.

Footnotes

Review coordinated by Ada Wossink

References