At lakeside markets along Lake Victoria, Africa’s largest lake by area, choosing fish to eat is part of daily life. A customer might pick a tilapia based on its freshness, size, price, and farmed or non-farmed origin, much like shoppers do in markets and supermarkets around the world. Whether grilled tilapia or salmon fillet, fish are a key source of healthy fats, such as omega-3s, essential for heart health. But these nutrients are invisible. We can’t see what’s inside the fish displayed at the market.
Risks are similarly invisible; as waters grow increasingly polluted, fish become more likely to contain harmful substances such as heavy metals or pesticides. Fish are thus a “bundled” product: the same tilapia that provides essential nutrients might also carry a health risk. This creates a dilemma. Consumers face a difficult trade-off.
Vendors increasingly use labels and nutritional guidelines to help consumers “see inside the fish”, to consider both nutrient benefits and contamination risks. But even if we know the exact levels of every nutrient and toxin, how do we weigh those trade-offs? Which matters more? This complexity is difficult to communicate effectively.
One problem is that, depending on individual background and cumulative intake, the prospective gains from nutrient intake or risks from ingesting contaminants may vary.
Another challenge is that how we resolve this trade-off can change throughout our lives. At one point our choices will be self-motivated, at another, we will have our children in mind, or pregnancy or a health condition will come into consideration. The balance between benefit and risk shifts in relation to these and other life situations.
Past experience shows how difficult this balance can be. A warning issued in the United States about mercury in fish targeted at pregnant women also led many other, not-at-risk consumers to reduce their own fish intake – losing important nutritional benefits.
To better understand how the information communicated in labels affects consumer choices about these trade-offs, our international research team ran an experiment with fish consumers in communities around Lake Victoria in Kenya.
Kenya’s fish markets: a microcosm of a global challenge
Fish is both an important source of income and a regular part of diets in western Kenya: 93% of our participants had eaten fish in the past week, making it an ideal environment for studying real-world decisions.
Add to that the fact that the lake is under pressure. As in many aquatic environments around the world, runoff from farms and wastewater fuel harmful algal blooms, dense green layers that spread across the water. In the wake of these blooms, bacteria produce toxins, known as microcystins.
Microcystins are harmful to humans when absorbed in too-high amounts through drinking water or fish tissue. Maximum intake guideline values set by the World Health Organization (WHO) exist for one cyanobacterial toxin, microcystin-LR, with stricter thresholds for, e.g. children.
In our experiment, participants were asked to choose between different hypothetical fish purchase options. Each option came with simple labels showing nutrient levels (healthy fats) and toxin levels: low, medium or high.
This allowed us to analyse how people weigh up these trade-offs when both benefits and risks are known and clearly communicated.
What we found
At the individual level, our results are reassuring. Most consumers show careful consideration for both nutrient and toxin labels. But they react more strongly to warnings about risk than to information about benefits. In other words, potential harm weighed more heavily than potential gain. This matters. If risk warnings dominate too strongly, people may avoid fish altogether. This highlights the importance of complementing risk guidance with information on benefits in cases of bundled products, like fish, where a nuanced, balanced response is preferable over a large avoidance reaction that may crowd out essential nutrient intake.
Our study also shows that participants’ reactions are more than just the sum of the “nutrient and the contaminant equation”, and indicates that consumers do not treat nutrients and contaminants as two separate factors that can simply offset each other. If a fish is high in nutrients, this does not simply cancel out higher contamination levels.
Instead, the two are evaluated together: benefits matter less when risks are higher. This seems rational, given that a balanced diet would be characterised by a good status in both dimensions. And it suggests that the two labels could support each other rather than compete for limited attention.
A hidden inequality
While many participants made nuanced choices, not everyone responded in the same way.
Some consumers paid close attention to both nutrients and contaminants. Others responded much less to either type of information. These less responsive consumers were more likely to have lower incomes, less education, or were less awareness of environmental risks. Those groups are often already more vulnerable and less healthy.
This raises a broader concern at the societal level.
If more responsive consumers begin avoiding certain fish, prices for those fish might fall, reflecting lower demand. This could make unhealthier fish more affordable – and therefore more attractive – for less information-responsive consumers. In that case, health risks could become concentrated among the already disadvantaged. In other words, better information might not automatically lead to better outcomes for everyone. To what degree this occurs in practice remains a matter for further research.
What can be done?
Helping consumers navigate these trade-offs starts with whether, and how, credible information is presented.
Researchers are working to uncover a systematic relationship between observable fish traits – such as species or size – and nutrient or contaminant levels. While fish-specific labelling will most likely remain out of reach, such systematic links could be used to support simple guidance for consumers’ choices.
Overall, many consumers display nuanced, highly rationally justifiable choices when the benefits and risks appear side-by-side.
Our results support trusting consumers to make nuanced decisions in line with their own specific needs and concerns. But the findings also underscore that nutritional information campaigns need to concentrate on communicating both sides of the coin – and to be vigilant concerning distributional hazards.
But the burden shouldn’t fall only on consumers and on those who oversee labelling.
The difficult trade-offs people face at the market are rooted in environmental pollution problems. Pollution reduces the quality of fish as food, and burdens consumers with increasingly difficult choices. Reducing environmental pollution is therefore essential – not only for ecosystems, but for food security.
This article was co-written with Christopher B. Barrett (Charles H. Dyson School of Applied Economics and Management, Cornell University), Kathryn J. Fiorella (Department of Public & Ecosystem Health, Cornell University),
Christopher M. Aura (Kenya Marine and Fisheries Research Institute, Kisumu), Hezron Awandu (Kenya Marine and Fisheries Research Institute, Kisumu), Fonda J. Awuor (Kenya Marine and Fisheries Research Institute, Kisumu), Patrick Otuo (Kenya Marine and Fisheries Research Institute, Kisumu).
The Axa science philanthropy is now part of the Axa Foundation for Human Progress, which brings together the commitments of Axa Group and Mutuelles d’Assurances in the fields of Science, Nature, Solidarity, and Culture. Before 2025, the global science philanthropy was held by the Axa Research Fund, which has supported over 750 projects around the world since its inception back in 2007. To learn more, visit Axa Foundation for Human Progress.
A weekly e-mail in English featuring expertise from scholars and researchers. It provides an introduction to the diversity of research coming out of the continent and considers some of the key issues facing European countries. Get the newsletter!
