The science of sustainable seafood, explained

Mislabeled seafood in the U.S.

The 2018 Oceana Canada study was only the most recent of a series of similar studies published by Oceana (see my earlier piece here). In 2016, Oceana released a report that summarized and mapped seafood fraud from, “more than 200 published studies covering 55 different countries, on every continent except Antarctica, in order to reveal the global scope of seafood fraud.” Oceana found relevant studies by searching Google Scholar and Google News with relevant search terms. Further, legal cases involving seafood fraud in the United States were collected in NOAA Law Enforcement or Department of Justice press releases and archives. The final collection was displayed on this map.

In this post, I will attempt to contextualize Oceana’s findings in the US specifically. I aim to challenge the report title that this map study “reveal[ed] the global scale of seafood swapping”, and instead suggest that it provided an unrepresentative view of mislabeling in the context of actual seafood consumption trends in the U.S

Considering a True Rate of Seafood Mislabeling in the U.S.

The vast majority of the studies collected by Oceana (>75%) were from Europe or the US and, “the bulk of the studies [were] conducted after 2005.” Globally, Oceana reported the weighted average mislabeling rate was 19%, but in the U.S., it was 28%. The most commonly mislabeled species in the US studies referenced were snapper, grouper, and salmon. Across all studies referenced (US and abroad) Oceana reported mislabeling in “all 200+ studies reviewed except one.”

Pins were placed on the world map to indicate the location of each mislabeling study collected for this report. The pins were color coded to indicate the extent of mislabeling, and to indicate if the study was an Oceana study or a study from another source like news media or peer reviewed literature. Dark red indicated studies showing mislabeling rates from 75%-100%; lighter red indicated mislabeling rates from 50%-75%; dark pink indicated rates from 25%-50%; light pink indicated rates from 0%-25%; a white pin with black pinstripes indicated a study featuring “other examples of fraud”; and a blue Oceana logo pin indicated an Oceana study.

Screenshot from Oceana

Measuring the scope of the studies referenced by using the interactive map proved difficult. The amount of studies for each mislabeling rate category, corresponding to the color and design of each pin, was not listed. Thus, the only way to quantify each category would be to laboriously count each pin. Even if each pin was counted one by one, multiple pins were often used for a single study, if that study occurred in multiple geographic locations. There were 210 studies cited, but I found over 150 pins in the US alone. For example, 14 pins were used to map the metropolitan areas from a single Oceana study from 2014. Yet for some reason pins were not included for rural areas in this same study, so the pin placements on the map seemed arbitrary. Showing multiple pins for a single study gave the map a much more congested look, suggesting more mislabeling instances than were actually found. Depicting mislabeling results on a map was an interesting idea, but the execution was sloppy and makes me wonder what the motivation for this map is, if not to provide useful data.

The analysis of this map study mirrored the three-point analysis of health, economics and conservation put forward in the 2018 Canadian study. The health section focused heavily on escolar again; the economic section displayed the best analysis and the most compelling data, primarily on pangasius substitutions this time; and the conservation section again contradicted the good arguments in the economic section and changed the topic to illegal, unreported, and unregulated fishing (another major campaign of Oceana’s).

The most important point to remember when considering the results and analysis from in this map study is that these numbers were derived from news stories, reports, peer reviewed literature and legal cases seeking mislabeling. Many of these studies were conducted by Oceana itself—we explained in detail in our previous post that Oceana sought out fraudulent seafood species in its research. Almost all the other cited studies were conducted in order to investigate reports of fraud—not to measure the rate of fraud across all sectors or species. That is a critical distinction Oceana muddled in its press releases and conclusions. A 28% rate of mislabeled seafood in the US is derived from findings inspired only when mislabeling was significant enough to warrant an investigation. This begs the question—what is the true rate of seafood mislabeling in the US, across all species consumed?

To begin answering this question and put Oceana’s results in proper perspective, I sorted through the 86 studies found on Oceana’s interactive map within the geographic boundaries of the US. I did my best to avoid counting twice for any repeated pins, but again, since multiple pins were used for a single study if that study took place in multiple metropolitan areas, it was challenging.

For each of the 86 studies I determined what the species of focus was (were). I determined this by examining the abstract (if applicable) and the paragraph describing the story provided by Oceana when a pin was clicked on the interactive map.

Screenshot from Oceana

By far the most common species to be included in a mislabeling study in the U.S was red snapper (33% of all U.S studies). In total I identified 22 species of focus, but only 8 appeared in at least 5% of the studies. Notable species not often featured were tuna (any tuna species, species of focus in only 1 study), yellowtail (1 study), and halibut (3 studies). The most commonly consumed seafood in the U.S, shrimp, was only a species of focus in 5 studies.

This figure shows the eight most common “species of focus” featured in a seafood mislabeling study on Oceana’s interactive map in the U.S. Red snapper was by far the most common. The percentage of all studies from the map in the U.S to feature each “species” is listed in red font. Almost 1 in 3 stories cited on this map in the U.S regarding mislabeled seafood featured red snapper as a primary focus of the study.

The studies that focused on red snapper featured high rates of mislabeling, similar to those reported in the 2018 Oceana Canada study. The rate of mislabeling for red snapper on average in the studies cited was 63%. Note that this average was calculated after excluding legal cases that did not evaluate a rate of mislabeling. See one such example of a legal case here.

However, to estimate the rate of mislabeled seafood in the US, shrimp and salmon are important indicators since they are the two most consumed seafoods in the country. Of the five citations on the map with shrimp as a species of focus, only one was a study measuring mislabeling rates (the other four were legal cases). This one study was a national study by Oceana that determined 33% of shrimp were mislabeled in the cities tested. Salmon had an average mislabeling rate of 29% across five studies.

The National Fisheries Institute calculated the following list of top ten most consumed seafoods in the US for 2016:

Courtesy NFI

When looking at this top-ten list and comparing it to the most often mislabeled species from Oceana studies, you can see the literature review on the interactive map is not representative of actual consumption trends. Red Snapper and Grouper—by far the two most common species of focus in mislabeling literature—are not to be found on the top-ten list. About half of Oceana’s US mislabeling studies contained one of those two species.

But, if we do take Oceana’s findings and cross reference them with this top-ten list, we can get a very general maximum estimate of the four species that were represented in the literature review on the map – shrimp, salmon, cod and crab. Before doing so let me reiterate some important qualifiers for this estimate: the reports pulled for this map study are inherently aimed at finding seafood mislabeling as I described earlier; weighted averages were impossible because some of the studies referenced did not provide the necessary data to make those determinations; some of the map citations provided inactive links; and most importantly, the Ns for these four species are too small to make a representative determination. Lastly, since only four of the NFI’s top-ten most consumed species were species of focus in this map study in the US, I will consider the other six species (canned tuna, tilapia, Alaska pollock, pangasius, catfish and clams) as 0% mislabeled. Counting these six species as 0% is a reasonable way to measure them since they are typically lower cost species that do not have an economic incentive for purveyors to commit fraud.

With all those considerations in mind, here is how the mislabeling rate would look for the top-ten most consumed seafood species in the U.S:

Of the 15lbs consumed on average per capita, about 2.5lbs are mislabeled according to this estimation. That is only a 15.5% mislabeling rate, considerably lower than the 28% rate Oceana reported for the US after this map study. Even though we took a low estimation with the six species counted as 0% mislabeling, I consider the mislabeling rates for the four other species in this top ten list to be far higher than is truly representative. With this in mind, the 15.5% rate should be considered an extreme high estimate for mislabeling across all seafood consumed per capita in the US.

Even if we take this 15.5% mislabeling rate with a massive grain of salt for all its potential flaws, we are still overlooking the fact that certain types of salmon and shrimp were targeted in the studies referenced on Oceana’s map. The 2014 Oceana shrimp study targeted only shrimp labeled as “Gulf shrimp”, omitting the massive amounts of imported, farm-raised shrimp that Americans typically eat when they choose shrimp. The same can be said for the salmon studies where products labeled as “wild salmon” were the focus, omitting testing on products labeled as farmed Atlantic salmon. Once again the aim in existing seafood mislabeling studies is to find fraud, not represent the problem in the context of the entire seafood system.

My analysis leaves me wondering: what was the goal for Oceana with this map study and the seafood fraud campaign in general? There is certainly value in exposing mislabeled seafood, but focusing on species like snapper and grouper—two species not in the top ten most consumed seafoods in the U.S—is not the correct way to understand the true scope of seafood mislabeling, and to frame the study as “new Oceana report and interactive map reveal global scale of seafood swapping” is misleading. The extremely general estimations I made on the rate of mislabeling across the most consumed seafood species in the US show how little this map study can effectively inform this topic. Policy makers will struggle to use Oceana’s findings to inform any practical policy changes, and the truly fraudulent actors can dismiss the findings as unrepresentative.  Hopefully future mislabeling studies—from Oceana or elsewhere—will measure mislabeling in the context of actual consumption patterns.

Jack Cheney

Jack Cheney

Jack has sourced, sold, cooked, and sustainably certified seafood over the past 10 years. In addition to his contributions to Sustainable Fisheries UW, he is working to increase traceability into supply chains and educate consumers, chefs and retailers on the value of environmentally sustainable seafood. He earned a Master's in Marine Affairs from the University of Washington in 2015.

Share this story:

Share on facebook
Share on twitter
Share on pinterest
Share on reddit
Share on email
Share on linkedin

Subscribe to our newsletter:

Read more:

Leave a Reply