A number of studies have attempted to measure the impacts of the economic and social aspects of globalization upon obesity. In general, the findings indicate that political and social globalization affect obesity rates by influencing lifestyles through information flows and societal links, while the effects of economic globalization, which includes trade and FDI, were found to have negligible effects on obesity rates (see also Box 2.4).176, 177
A review of 28 studies suggests that increased trade liberalization alone was insufficient to drive overweight and obesity, and that FDI levels were more strongly associated with rising obesity in low- and middle-income countries.178 Assessing the relationship between trade openness and obesity across 175 countries during the 1975–2016 period, one study suggested that the more open a country is to merchandise trade the higher the prevalence of obesity.179
Using a sample of 70 developing countries between 1990 and 2013 and considering only trade in food and agriculture, one study identified a negative relationship between agricultural trade openness and the prevalence of obesity and overweight among adults. The findings suggest that a 1 percent increase in agricultural trade openness would reduce the prevalence of obesity and overweight among adults by approximately 0.5 percent.180 This effect of agricultural trade is found to be due to a decline in the share of fat in favour of carbohydrate.ab Nevertheless, the position of a country on the development path can also affect the relationship between trade and obesity. Estimates presented in this report suggest that for countries at the later stages of the nutrition transition, higher agricultural trade openness can be an important driver of obesity as it can increase the availability of foods high in sugars and/or fats (see Box 2.6, Part 2).
Focusing on the composition of agricultural trade, more particularly on specific foods such as those high in sugars or ultra-processed foods, provides stronger evidence of the relationship between trade and obesity. For example, analysis based on a sample of 116 developing countries between 2000 and 2016 points to obesity being driven by imports of ultra-processed foods with high sugar content rather than agricultural trade in general.181 Another study, using data from 172 countries for the 1995–2010 period finds that, on average, a 50 percent increase in the imports of sugar and energy-dense (ultra-) processed foods would result in a 0.0007 increase in the average BMI.182 At a country level, rapid import growth in South Africa contributed to an increase in the availability of processed foods. Between 1992 and 2010 imports of soft drinks and of (ultra-) processed snacks increased by 92 and 83 percent, respectively. Similar increases took place in countries that participate in the Southern African Development Community.183
The nutrition literature suggests a positive relationship between the high consumption of ultra-processed foods and obesity. Ultra-processed foods can contain large amounts of free sugars and saturated fats, which contribute to a high energy intake.184 Studies indicate that individuals consuming higher shares of ultra-processed foods are more likely to be obese than individuals with low levels of consumption.185, 186 This pattern also holds for children, indicating that a higher consumption of ultra-processed foods is associated with higher increases in adiposity in both children and youth.187, 188 It is possible that the weight is affected by an overall higher intake, which may be facilitated by the combined characteristics of ultra-processed foods such as high palatability, energy density, marketing and convenience.189 Nevertheless, there is no consensus among nutrition experts on the exact metabolic responses to ultra-processed food consumption. Further studies are needed to establish a causal relationship between ultra-processed food consumption and obesity.190
Food trade and obesity in the Pacific islands
The disproportionately high prevalence of obesity in the Pacific islands could be due to a number of factors. Geographical remoteness and susceptibility to food shortages due to weather extremes could have enhanced the islanders’ genetic predisposition to gain weight.191 Nevertheless, anthropological research suggests that, prior to colonization, local technologies for food preservation such as drying and fermenting, and food storage were widespread enough to provide a buffer for production shocks. Instead, globalization, trade and rapid social change, especially in small, closely-knit societies, could provide an alternative explanation for the rise of obesity in the region. Indeed, researchers show that, as in other countries, obesity emerged in the Pacific, for example, in Nauru and the Cook Islands, during the second half of the twentieth century.192
Agriculture in the Pacific islands is under a number of constraints. Agricultural land endowments are limited, which translates into low levels of food production and a low diversity of foods produced that do not meet the needs of a growing population. In many cases, the per capita production of fruits, vegetables and staple foods decreased or remained more or less stable between 1965 and 2015.193 High trade costs due to geographic remoteness add to low competitiveness, and exposure to natural disasters increases risks and hinders investments. For example, one study suggests that in Samoa during the early 1990s, the local food staples sectors of coconuts, fruit and taro were hit by a rapid succession of natural disasters and the spread of the taro leaf blight, resulting in a lasting decline in production capacity.194
As agriculture in the Pacific islands has limited production capacity, it is food trade that has a significant impact on the availability and diversity of food. For example, food imports in Samoa made up around 50 percent of total calories supplied, and only 31 out of the 122 unprocessed and minimally processed food items that were available for consumption in 2020 were produced domestically. In the same year, in Nauru, there were 95 unprocessed and minimally processed food items available while only 19 of those were domestically produced. For Nauru, in 2020 the share of food imports in total calories available for consumption amounted to 73 percent.
Food processing capacity is also low. Fiji is the only island state with a small food processing sector that serves the domestic market and exports processed foods to other neighbouring island states, while a large part of processed and ultra-processed foods available in local markets are almost entirely imported.195 For example, out of 31 ultra-processed foods supplied in Samoa in 2020 only one was produced domestically.
While trade has contributed to increasing the availability and diversity of foods in the Pacific islands, it has also been cited as an important factor in the rapid increase in obesity and its disproportional high levels.196, 197 Integration into global food markets has helped accelerate the nutrition transition with significant changes in local diets. Traditional local diets composed of mainly fruits, food staples, locally produced animal foods and fresh fish, gave way to increasing amounts of imported animal products of high fat content as well as processed and ultra-processed foods.198
Imports of mutton flaps and turkey tails – relatively inexpensive meats with very high fat content – into the islands and their role in the growing obesity rates have been discussed extensively in the literature.199, 200 Analysts suggest that the low levels of income in many Pacific Small Island Developing States also play an important role in shaping consumption patterns.201, 202 Fiji’s ban on the sale of mutton flaps in 2000 served to raise awareness of the risks of frequent consumption of food items high in fats, but it did not lead to conclusive improvements in the overall local diet (also see Part 5).203
The high share of imported processed foods, processed culinary ingredients and ultra-processed foods in consumption expenditure is also receiving attention in the context of the high prevalence of obesity. For example, between 2014 and 2018 per capita sales of processed foods, soft drinks and vegetable oils increased in most Pacific islands, with Papua New Guinea showing the highest level of increase at 56 percent.204
Indeed, in Oceania the share of ultra-processed foods in total calories imported is very high relative to all other regions, making up 23 percent of all imports in terms of calories and close to 50 percent in terms of import value in 2021. Within Oceania, Australia features the highest share of ultra-processed food imports, followed by Nauru and Tonga (Figure 4.3).
FIGURE 4.3Ultra-processed food imports as share in all food imports (based on energy content), selected countries in Oceania, 2021
Imports of ultra-processed foods by Oceanian countries in 2021 were made up of high shares of pastry, followed by food preparations. Margarine and shortening, sausages and similar products of meat, other meat preparations, and sugar confectionery also contributed to high levels of ultra-processed food imports. Compared with the most traded ultra-processed items at the global level, the high shares of imports of sausages and other meat preparations are striking in some of the Small Island Developing States of Oceania such as Kiribati, Nauru, New Caledonia, Samoa, Solomon Islands and Tonga.
Several studies have examined the impact of trade agreements and, especially, the impact of the accession of the Pacific Small Island Developing States to the WTO on the composition of food imports to assess the trade effects on obesity.ac For example, in Fiji, WTO accession in 1996 resulted in increasing the availability of fresh fruits, vegetables and whole-grain refined cereals, but also that of fats and oils, meat, processed dairy products, energy-dense beverages and processed and packaged foods. These impacts were the result of changes in import tariffs, but also of changes in non-tariff measures, as for example adjustments in the appropriate level of protection of sanitary and phytosanitary measures, which contributed to increasing import volumes.ad, 205
Similar impacts were reported for Vanuatu, which joined WTO in 2012.206 Nevertheless, an analysis across 16 countries, including the United States, Australia, New Zealand and 13 Pacific Small Island Developing States in 2019, many of which are not signatories to WTO, points out that the trends in imports and consumption of processed and ultra-processed foods were dependent on income, suggesting that income growth is the underlying driver of processed food consumption in the region, while trade could be seen as an accelerator of the nutrition transition.207
Regional trade agreements and the composition of food trade
The debate on whether increased trade initiated by trade liberalization promotes the availability of ultra-processed foods and contributes to a high prevalence of obesity has expanded beyond the Pacific. A review of 17 studies on the impact of regional trade agreements (RTAs) on health outcomes, suggests that their implementation was associated with an increased consumption of processed food and sugar-sweetened beverages and correlated with a higher cardiovascular disease incidence and higher BMI, underlining, however, that these linkages were methodologically limited.208
By reducing barriers to trade and investment between signatories, RTAs in the Americas were found to increase the availability of calories, which could contribute to rising obesity.209 By lowering or removing tariff and non-tariff barriers on energy-dense foods, including ultra-processed foods, RTAs in Northern America and Latin America and the Caribbean could influence the composition of traded foods and thus affect the food environment of the signatories.210, 211 For example, a study on food trade impacts on obesity in the context of the United States–Mexico–Canada Agreement found that food imports from the United States to Mexico competed with domestic foods and exerted downward pressure on domestic prices. However, this price-reducing effect was limited to energy-dense foods with low nutritional value only, while the effect of competitive pressure on other “healthier” foods was insignificant.212
Analyses focusing on the impacts of RTAs on the composition of food imports and on obesity are, in general, haunted by several issues. First, as these trade agreements include provisions that lift barriers on both trade and investment flows between the signatories, studies tend to conflate the effect of trade and that of foreign direct investment on the availability and composition of foods in the domestic markets. Second, most studies do not take into consideration the impact of income on the demand for different food imports, thus disregarding income effects and the significant influence of the nutrition transition on dietary patterns. Third, the way foods are classified into “healthy” or “unhealthy”, “obesity-prone” or not varies across studies, making comparisons difficult.
A study carried out for this report investigates the impact of RTAs on the composition of food imports, using the NOVA food classification (see Part 2, Box 2.2).213 Analysing bilateral trade flows of approximately 400 food items across all countries from 1991 to 2017, the study suggests that RTAs have a significant impact on food imports that varies across foods at different processing levels. This variation stems from two sources: (i) the use of non-tariff measures such as sanitary and phytosanitary (SPS) measures and technical barriers to trade (TBT) and their harmonization or mutual recognition across RTA signatories; and (ii) different income effects on food imports of different processing levels.
Regional trade agreements: Impact of sanitary and phytosanitary measures and technical barriers to trade provisions on the import demand for foods classified according to processing level
SPS measures include mandatory standards that focus on additives, contaminants, residues of pesticides or veterinary drugs in foods and beverages, as well as certification and labelling requirements directly related to food safety (for example, food allergens). Technical barriers to trade reflect technical regulations, conformity assessment procedures and standards such as nutrition labelling addressing risks not expressly referring to food safety (for example, information on nutritional content), packaging, grading and quality requirements (see Part 5).214, 215 In fact, as both TBT and SPS measures are widespread in food and agriculture, their effects on food trade can be much stronger than those of tariffs.216 However, the effects of these non-tariff measures on trade can be mixed; food standards can be trade-enhancing, as well as trade-impeding, depending on the measures, food products and countries involved.217 For example, both TBT and SPS measures may restrict trade as they increase trade costs related to compliance. At the same time, they can also expand trade, as they strengthen the demand for a product through better information on its safety and nutritional characteristics. Harmonization of standards and regulations generally promotes trade as higher trade costs associated with diverging requirements cease to exist.218, 219
Modern RTAs include provisions for deeper cooperation in regulation and standards to promote trade among their signatories and foresee a harmonization of TBT and SPS measures or provide for the mutual recognition of domestic measures. For example, the Deep and Comprehensive Free Trade Areas of the European Union with Georgia, the Republic of Moldova and Ukraine suggest that SPS measures by the three countries converge towards the European Union legislation.220
The analysis indicates that both SPS and TBT provisions have positive effects on food imports. RTAs including a high number of SPS provisions tend to increase imports of processed culinary ingredients and ultra-processed foods, while their impact on processed foods is relatively small and that on unprocessed and minimally processed foods insignificant. RTAs with a high number of TBT provisions have a stronger positive impact on food imports, as compared to SPS provisions, but their impact on food imports does not vary across processing levels (Figure 4.4). While mutual recognition of SPS measures has little impact on food imports, harmonization increases food trade, especially for ultra-processed foods and processed culinary ingredients. For TBT measures, both harmonization and mutual recognition increase food trade across processing levels (Figure 4.5).
FIGURE 4.4Regional trade agreements: Impact of non-tariff measures on bilateral food trade flows across processing levels
SOURCE: Adapted from Rotunno, L. (forthcoming). Demand for processed foods and deep trade agreements – Background paper for The State of Agricultural Commodity Markets 2024. Rome, FAO.
FIGURE 4.5Income responsiveness of bilateral food trade flows across processing levels
SOURCE: Adapted from Rotunno, L. (forthcoming). Demand for processed foods and deep trade agreements – Background paper for the State of Agricultural Commodity Markets 2024. Rome, FAO.
If a country joined an RTA with the highest number of essential provisions in SPS, this would triple bilateral imports of processed culinary ingredients and would increase imports of ultra-processed foods by 70 percent. Joining a trade agreement with a clause on the harmonization of SPS measures would increase trade in ultra-processed foods by 30 percent. These differing trade impacts arise because foods that are ready to be consumed or used in the food processing industry such as fruits, oils and fats, and ultra-processed products are generally subject to a larger number of regulatory measures than unprocessed or minimally processed foods, of which the largest part is made up of staple foods that need to be cooked.221
TBT effects can be even larger. Joining an RTA with the highest coverage of TBT provisions would increase imports of processed culinary ingredients by 146 percent, imports of unprocessed foods by 140 percent and imports of ultra-processed foods by 90 percent. Such effects reveal the broad reach of TBT provisions, which include nutrition information labelling, across foods (see Part 5).
Recent trade agreements go beyond import tariff reductions and market access and aim at deeper trade integration, focusing on harmonizing non-tariff measures and domestic regulations. Indeed, such deeper trade agreements reduce trade costs related to compliance with multiple and different standards and expand trade between the signatories, especially for products that are subject to a high number of standards and measures.222, 223
Regional trade agreements: Income effects on import demand for foods classified according to processing level
Income effects on the import demand for foods of different processing levels can conflate with the impact RTAs have on the composition of food imports. Income growth is an important driver of the nutrition transition and of the demand for animal source foods, fats and oils, and processed and ultra-processed foods such as meat preparations and soft drinks. These high-value foods tend to be more responsive to income changes as compared to food staples (see Part 1).
Such income effects add to the impact of non-tariff barriers on import demand for foods in the context of RTAs. Indeed, the analysis suggests that the responsiveness of processed and ultra-processed foods to income changes is much higher than that for unprocessed and minimally processed foods. For example, on average, across all countries, food products and RTAs, a 1 percent increase in income can result in a 1.2 percent and a 1.1 percent increase in the demand for processed and ultra-processed food imports, respectively. A 1 percent increase in income would result in a 0.7 percent and 0.8 percent increase in imports of unprocessed and minimally processed products and of processed culinary ingredients, respectively.
The differences in income effects across food imports of different processing levels are important and relevant in the context of nutrition. For a given change in income, the responsiveness of processed and ultra-processed food imports is estimated to be proportionately higher, while that of unprocessed and minimally processed foods is shown to be proportionately lower. On the one hand, as incomes grow, the demand for processed and ultra-processed food imports will be stronger than the increase in income as consumers switch away from staple foods, which make up most of the unprocessed and minimally processed foods. On the other hand, imports of unprocessed and minimally processed foods, as well as processed culinary ingredients that are used as inputs in the food industry, are less responsive to income changes. This behaviour of food import demand, estimated at the aggregated level using bilateral trade flows in the global food market, is entirely consistent with the concept of nutrition transition.
Although the demand for ultra-processed food imports responds strongly to income, a separate RTA effect can be identified through the impact of depth and treatment of non-tariff measures. SPS measures appear to facilitate imports of ultra-processed foods relative to other foods. TBT measures, including nutrition labelling, may affect import demand, leading to a relatively lower expansion of trade in ultra-processed foods compared with the other processing levels. This can have implications for trade policymakers who negotiate RTAs that are increasingly found in the spotlight of the public discourse surrounding nutrition (see Part 5).
