Response Agro-Industrialization Otherwise

Contents lists available at ScienceDirect

Global Food Security

journal homepage: www.elsevier.com/locate/gfs

Not just for the wealthy: Rethinking farmed fish consumption in the Global South

Ben Beltona,⁎, Simon R. Bushb, David C. Littlec

a Department of Agricultural, Food and Resource Economics, Michigan State University, 446 W Circle Dr, East Lansing, MI 48824, United States b Environmental Policy Group, Wageningen University, P.O. Box 8130, 6700 EW, The Netherlands c Institute of Aquaculture, University of Stirling, Stirling FK9 4LA, UK

A R T I C L E I N F O

Keywords: Aquaculture Domestic consumption Diet transformation Food security Fish

A B S T R A C T

Aquaculture’s contributions to food security in the Global South are widely misunderstood. Dominant narratives suggest that aquaculture contributes mainly to international trade benefiting richer Northern consumers, or provides for wealthy urban consumers in Southern markets. On the supply side, the literature promotes an idealized vision of ‘small-scale’, low input, semi-subsistence farming as the primary means by which aquaculture can contribute to food security, or emphasizes the role of ‘industrial’ export oriented aquaculture in undermining local food security. In fact, farmed fish is produced predominantly by a ‘missing middle’ segment of commercial and increasingly intensive farms, and overwhelmingly remains in Southern domestic markets for consumption by poor and middle income consumers in both urban and rural areas, making an important but underappreciated contribution to global food security.

1. Introduction

Fish1 is a rich source of vitamins, minerals, fatty acids and high quality protein, playing an essential role in the diets of billions of consumers, many of them poor, malnourished and living in low and middle income countries (Thilsted et al., 2016; HLPE, 2014; Kawarazuka and Béné, 2011).

Fish utilized for human consumption is obtained from a continuum of sources running from capture fisheries (the harvest of naturally re- producing fish populations), to aquaculture (breeding and farming under controlled conditions). Global capture fisheries output peaked in the mid-1990s, and has plateaued or declined since (cf. FAO, 2016a; Pauly and Zeller, 2016). In contrast, aquaculture has boomed, growing at an average rate of 8.2% per annum over the past three decades. As a result, farming now provides more than half of the fish destined for direct human consumption (FAO, 2016b).

The growth trajectories of capture fisheries and aquaculture are often juxtaposed to make the case that sustained and rapid aquaculture development is vital to the future food security of fish dependent po- pulations in Southern nations (e.g. Barange et al., 2014).

A counter-narrative (which we term ‘economic geography’), holds that aquaculture largely fails to meet the needs of poor and

undernourished Southern consumers. The narrative asserts that most farmed fish produced in Southern countries is destined for export to Northern markets (McIntyre et al., 2016; Ponte et al., 2014), and that farmed fish remaining in domestic markets is consumed primarily by wealthy urbanites (Beveridge et al., 2013; Bush, 2004; Ahmed and Lorica, 2002; Lewis, 1997). A related argument is that aquaculture production is concentrated in Asia and does little to address the needs of malnourished populations in Africa (Hall et al., 2013; Golden et al., 2016).

A second pair of narratives sets up contrasting visions around aquaculture's supply side. The first emphasizes the predominance and desirability of low intensity ‘small-scale’ fish farming that contributes directly to household food security and producer incomes (e.g. Bondad- Reantaso and Subasinghe, 2013). The second frames environmental degradation and social dislocation associated with the rise of ‘in- dustrial’ export-oriented aquaculture as compromising the food security of communities in Southern fish producing nations (e.g. Nayak and Berkes, 2011; van Mulekom et al., 2006).

We argue that despite their influence in shaping science, policy and popular perceptions, none of these narratives adequately account for the current diversity of aquaculture in the Global South, nor its ag- gregate ‘macro’ effects on food security. The remainder of this paper

https://doi.org/10.1016/j.gfs.2017.10.005 Received 9 March 2017; Received in revised form 2 October 2017; Accepted 27 October 2017

⁎ Corresponding author. E-mail address: [email protected] (B. Belton).

1 The terms ‘fish’, ‘aquatic animals’ and ‘seafood’ are used interchangeably as a shorthand for edible aquatic animals. Aquatic plants, algae, non-edible aquatic animals (e.g. corals, sponges), and aquatic mammals, are excluded all calculations in the paper.

Global Food Security 16 (2018) 85–92

2211-9124/ Published by Elsevier B.V.

T

makes this case. First we demonstrate that, contrary to the focus on international

trade, farmed fish is overwhelmingly consumed domestically in Southern aquaculture-producing nations, and is increasingly widely available and readily accessible to low-income urban and rural con- sumers in these markets. Second, we address supply side arguments by challenging the dominant narratives linking aquaculture and food se- curity and the prescriptions for promoting aquaculture that arise from them. We conclude by highlighting the need for future research and policy to pay closer attention to existing patterns of aquaculture de- velopment and their contributions to Southern food security.

2. International trade vs. domestic consumption

Seafood is among the most highly internationally traded food commodities (e.g. Asche et al., 2015; Tveterås et al., 2012). Fish and shellfish exports from developing countries exceed the value of coffee, rubber, cocoa, tea, tobacco, meat, and rice combined (Smith et al., 2010) and trade in fish products accounts for 10% of all agricultural exports (Gephart et al., 2016). In 2012, 37% of global fish production was exported (Kobayashi et al., 2015), with an estimated value of $129 billion (HLPE, 2014).

The scale of the international seafood trade and its apparent ten- dency to move large quantities of fish away from poor food insecure Southern nations to wealthy food surplus countries renders it con- troversial (HLPE, 2014). For example, Smith et al. (2010) contrast the status of large net exporters of seafood (e.g. China, Indonesia, Vietnam, Thailand, India, and Myanmar) possessing moderate to high levels of undernourishment, with the largest net importing markets (e.g. the United States and European Union), which are wealthy and well- nourished.

Asche et al. (2015) and Béné et al. (2015a) provide thorough sy- nopses of the debate over whether international trade in seafood has positive or negative effects on fish consumption and poverty. Our intent in the present paper is not to contribute to the literature on seafood trade. Rather, we argue that an emphasis on international trade has obscured the contributions made by farmed fish to domestic food se- curity in the main Southern aquaculture producing countries.

To demonstrate this point, we estimate the volume of fish origi- nating from aquaculture and capture fisheries that are traded inter- nationally, or remain in country for domestic consumption, for the ten largest aquaculture producing developing countries in the world – Bangladesh, Brazil, China, Egypt, India, Indonesia, the Philippines,

Myanmar, Thailand and Vietnam (FAO, 2016a). Together these coun- tries accounted for 87% of global production of farmed aquatic animals and 43% of global capture fisheries landings in 2013. They were also home to 51% of the global population, and 52% of all malnourished individuals (Table 1).

The trade component of the FAO Fishstat J database (FAO, 2016b), on which we base our analysis, does not specify whether internationally traded seafood products originate from capture fisheries or aquaculture. Following the methodology set out by Bush et al. (2013), we estimated the share of internationally traded aquatic animal products derived from each source, working on the assumption that the shares of farmed and wild fish species groups in exports from each country are propor- tional to the shares of farmed and wild fish of these species groups in national production.2

National fish production is reported by FAO in live weight equiva- lents (the weight of freshly harvested fish prior to any processing). The quantity of fish products traded internationally is reported in nominal terms - i.e. as the volume of fish traded post-processing (if any). To estimate the live weight equivalent (LWE) of each internationally traded product listed in Fishstat J we assigned conversion factors for similar categories of product, obtained from published sources (FAO, 2015; Bush et al., 2013; European Commission, 2011; Tacon et al., 2006). For each country, reported aquaculture production was divided by the apparent LWE of aquaculture exports to estimate the share of farmed fish exported and the share remaining as domestic food supply. The same procedure was followed for capture fisheries production and exports.

Fig. 1, reveals the extent to which excessive focus on international trade in seafood has inflated perceptions of its significance. The vast majority of fish farmed and landed in the ten most important Southern aquaculture producing nations is not exported. Eighty-nine percent of the farmed fish produced in these countries is consumed in their do- mestic markets. The share of capture fisheries landings exported is al- most double that of farmed fish, and exceeds that of aquaculture in seven of the ten countries, but is still modest at 22%.

In eight of the ten countries, apparent domestic consumption of

Table 1 Population, undernourishment, and aquaculture and fisheries production for selected countries.

Country Population (millions)a

Prevalence of Undernourishment (% of population)a,b

Undernourished population (millions)c

Aquaculture production (t)d

Capture fisheries production (t)d

Aquaculture as a share of fish production (%)

Bangladesh 161 16 26 1,859,808 1,550,446 55 Brazil 208 5 10 472,829 765,287 38 China 1371 9 123 42,694,335 16,274,939 72 Egypt 92 5 5 1,097,544 356,858 75 India 1311 15 197 4,549,607 4,645,182 49 Indonesia 258 8 21 3,819,517 6,103,001 38 Myanmar 54 14 8 926,175 3,786,840 20 Philippines 101 14 14 815,008 2,335,004 26 Thailand 68 7 5 1,052,701 1,843,747 36 Vietnam 92 11 10 3,203,326 2,803,800 53 Subtotal 3714 418 60,490,850 40,465,104 60 World 7347 11 808 69,296,511 93,763,656 42 Subtotal as share of

world (%) 51 n/a 52 87 43 n/a

Notes: a World Bank (2016). b Undernourishment refers to the percentage of the population whose food intake is insufficient to meet dietary energy requirements continuously. c Calculated from data in columns 2 and 3. d Production data for 2013 (FAO, 2016b).

2 At the country level, production data were categorized by the “ISSCAAP species groups” reported by FAO. Exports were categorized by “ISSCAAP commodity divisions”. Species divisions and commodity divisions were then combined under five aggregate “ISSCAAP commodity groups” (crustaceans, freshwater and diadromous fishes, marine fishes, miscellaneous aquatic animals, and molluscs, including cephalopods) to enable comparison across countries and product categories. The complete dataset used, including all calculations, is available for download (see Belton et al., 2017b).

B. Belton et al. Global Food Security 16 (2018) 85–92

86

farmed fish exceeds 90% of total national aquaculture production. Only in Thailand and Vietnam do aquaculture exports exceed domestic consumption. Both these countries are also major exporters of capture fisheries products, and have fish supplies per capita well in excess of the global average of 20.1 kg, at 24.8 kg/capita/year and 32.7 kg/capita/ year, respectively (FAO, 2016a, 2016c). Their seafood exports are surplus to domestic consumption needs, and do not divert food away from consumers at home.

To address the possibility that extrapolating the proportion of aquaculture and capture production to exports could bias results, we performed an alternative calculation using the most conservative as- sumptions possible with respect to aquaculture's contribution to do- mestic fish supplies. For this estimate, for each country, we attributed 100% of exports to aquaculture in species groups where production of farmed fish exceeded exports. For species groups where export volumes exceeded farmed fish production, we assumed that 100% of farmed fish was exported, with capture fisheries making up the gap between farmed fish production and total exports. Our original and alternate estimates are presented together in Table 2. The alternate estimate has little impact on the overall results: even under the most stringent assump- tions, domestic consumption of farmed fish equals or exceeds 90% of production in seven countries and stands at 84% in one more, with only 15% of farmed fish exported overall.

These results are supported by data presented in FAO (2016a),

indicating that freshwater fish (by far the most important category of fish produced in the ten selected countries) account for just 4.8% of international trade in fish by volume. Shrimp (the second most im- portant species group produced in the ten countries) makes up 6% of world seafoood trade. In contrast, marine fish, which originate over- whelmingly from capture fisheries, account for 68.7% of global trade.

3. New geographies of consumption

The ‘economic geography’ critique of aquaculture's contribution to food security is epitomized by Golden et al. (2016), who state that fish farmed in the Global South is, “mostly exported to the wealthy coun- tries of Europe and North America, or consumed by the growing middle-classes in the megacities of these economies” (p. 318). Beveridge et al. (2013, p. 1075) also hypothesize that “aquaculture producers in developing countries tend to target the production of larger-sized fish, aimed at middle-class urban regional and international markets, presumably in the expectation that the higher absolute and relative prices such fish command increase profits”.

Analysis presented in the previous section undermines the claim that aquaculture is strongly export oriented. The following subsection provides evidence that farmed fish produced for Southern domestic markets is now widely accessible to low-income rural and urban con- sumers. The changing characteristics of urban and rural demand, and

Fig. 1. Aquaculture and capture fisheries exports and apparent domestic consumption in selected countries (2011). Source: Authors’ calculations from FAO (2016b). (This analysis is based on data for 2011, the most recent year for which both production and trade data was available in the Fishstat J database).

Table 2 Estimates of aquaculture's contribution to domestic consumption and exports in selected countries (2011). Source: Authors’ calculations from FAO (2016b).

Item Bangladesh Brazil China Egypt India Indonesia Myanmar Philippines Thailand Vietnam Total

Total aquaculture production (million t) 1.52 0.44 38.62 0.99 3.67 2.72 0.82 0.77 1.20 2.85 53.59 Exports (first estimate)

Total aquaculture exports (estimated LWE) (t) 60,475 2993 2,918,521 5985 193,093 231,391 36,613 19,187 935,535 1,681,781 6,085,575 Share of exports in aquaculture production

(%) 4 1 8 1 5 9 4 3 78 59 11

Share of aquaculture consumed domestically (%)

96 99 92 99 95 91 96 97 22 41 89

Exports (second estimate) Total aquaculture exports (estimated LWE) (t) 130,071 5782 4,005,385 10,820 361,248 429,382 66,510 40,070 837,868 1,895,821 7,782,956 Share of exports in aquaculture production

(%) 9 1 10 1 10 16 8 5 70 67 15

Share of aquaculture consumed domestically (%)

91 99 90 99 90 84 92 95 30 33 85

B. Belton et al. Global Food Security 16 (2018) 85–92

87

the effects of aquaculture on fish price stability and supply are also examined.

3.1. Access and availability

Contrary to the prevailing view, aquaculture produces a wide range of species of low or moderate market value. Reports from the early stages of aquaculture development in Bangladesh indicated a bias to- ward producing high value Indian major carps that were not accessible to the rural poor (Lewis, 1997). However, although Lewis’ observation accurately reflected the situation in the mid-1990s, by 2011 three low value farmed species (pangasius, silver carp and tilapia) were each eaten in greater quantities in Bangladesh than rohu (the most popular Indian major carp) or any single species of fish originating from capture fisheries (Hernandez et al., 2017).

Low value farmed fish species now dominate or make large con- tributions to domestic fish supply in most of the 10 countries assessed above (e.g. tilapia and walking catfish in Thailand and Indonesia, silver carp in China, tilapia in Egypt, pangasius in India). At the same time, production of more expensive carnivorous species in these countries has often stagnated at low levels (e.g. barramundi and grouper in Thailand, Indonesia and Vietnam). Given that the vast majority of consumers in these countries belong to low or middle income brackets, such a pattern is to be expected. Aquaculture could not have sustained its extreme growth rate over the last three decades if it catered only to demand from a small wealthy segment of consumers.

Aquaculture's growth has driven down the real price of most farmed fish species produced in large volumes, making them increasingly ac- cessible to lower income consumers. The tendency for supply to in- crease and production to remain profitable even as prices fall, is the result of productivity growth arising from improvements in efficiency (Asche et al., 2009). This pattern has been a defining feature of the global aquaculture boom over the past 30 years.

Examples of declines in the real price of farmed fish are numerous. In Egypt, from 2000 to 2010 inflation adjusted prices fell by 46% for catfish, 38% for tilapia, and 31% for mullet (Macfadyen et al., 2011). International prices for tilapia and pangasius fillets fell by around 40% from 1995 to 2007 and 2002 to 2007, respectively (Asche et al., 2009). Nominal prices received by carp farmers in India's main producing state, Andhra Pradesh, remained roughly constant from the mid-1980s to 2014, representing a huge reduction in real price (Belton et al., 2017a).

Depletion of wild fish stocks has been another driver of aquacul- ture's growth. Pervasive habitat degradation and overexploitation of fish stocks have caused declines in the supply per capita of fish from inland and marine capture fisheries in major aquaculture producing countries. Non-farmed fish have become increasingly expensive, both in real terms and relative to farmed fish, often reversing their relative accessibility to consumers. For example, in Myanmar the price of sna- kehead and hilsa (two of the most important species harvested from inland and marine capture fisheries, respectively) increased at an average rate of 2.9% and 5.5% each year from 2008 to 2014. In con- trast, the real price of rohu, the main species farmed, fell at an average rate of 0.5% per annum (Belton et al., 2015).

In Bangladesh, the average real price of fish from inland capture fisheries increased 4% between 2000 and 2010, while that of marine capture fish jumped 42%. The average price of farmed fish fell over this period, to become 10% cheaper than the inland capture fish that once dominated supply (Toufique et al., 2017). A similar pattern is illu- strated in Fig. 2. From January 2012 to June 2015 the nominal price of Bangladesh's four main farmed fish species remained fairly constant. In contrast, the price of two formerly cheap and abundant small fish species harvested from inland capture fisheries (puti and mola) rose, to exceed of all but one of the main farmed species by the end of the period. These trends are linked to a 35% reduction in the consumption of inland capture fish by extreme-poor households and 37% reduction in consumption by non-poor households in Bangladesh from 2000 to 2010. In contrast the consumption of farmed fish increased 152% and 88% for the same groups over this period (Toufique and Belton, 2014).

3.2. Urbanization and shifting patterns of demand

Contrary to assumptions implicit in the economic geography nar- rative, food security is no longer solely, or even predominantly, a rural concern. The share of urban dwellers already exceeds those in rural areas, and is expected to rise to 66% by 2050, fueling unprecedented urban demand for food (Chicago Council, 2016). Many urban in- habitants in Southern countries, even among the middle classes, occupy precarious positions. For example, sixty percent of Africa's 350 million strong middle class is considered vulnerable to slipping back into poverty, and thus food insecurity (Chicago Council, 2016). Results from the Gallup World Poll's ‘Food Insecurity Experience Scale’ survey reflect this urban precarity. Respondents in urban areas of Asia feel more food insecure than those in rural areas - a difference of about three

Fig. 2. Average monthly nominal retail prices of four farmed and two non-farmed fish species in Bangladesh, January 2012-May 2015. Note: Non-farm species identified by dashed line. Source: Unpublished data collected by WorldFish on a weekly basis from 25 rural and urban retail markets in six districts in South, Southwest, North and Northwest Bangladesh.

B. Belton et al. Global Food Security 16 (2018) 85–92

88

percentage points for Asia as a whole (21% in rural areas versus 24% in urban), and eight percentage points in South Asia (FAO, 2016b).

Demand for non-staple foods in Southern cities is growing in line with ‘Bennett's Law’ - that the proportion of calories derived from starchy staples falls with rising income as consumers diversify the food consumption bundle to include higher-priced calories (Timmer et al., 1983). As a result, the share of food expenditure allocated to fruits, vegetables, meat, dairy and fish rises disproportionately with income (Reardon et al., 2014). By virtue of higher average urban incomes and the increasingly large share of urban inhabitants in the population, demand for food is highly concentrated in urban areas, which already account for 65% of the value of the entire Asian food economy (Reardon and Timmer, 2014). This means that demand for all fish - both farmed and wild - is greater in urban areas than rural.

However, the gap in consumption of animal proteins between urban and rural consumers is now smaller than is generally understood. Cross- country analysis of Bangladesh, Indonesia, Nepal and Vietnam shows that urban dwellers in all four countries consume approximately 50% more animal source foods (meat and fish combined) than rural dwellers (Reardon et al., 2014). In China, urban consumers in the poorest wealth quartile (i.e. the poorest 25% of the urban population) eat 40% more fish than the poorest quartile of rural consumers, while the difference between urban and rural consumers across the other quartiles ranges from 21% to 25% (calculated from Chui et al., 2013). In Bangladesh, the difference between urban and rural fish consumption is smaller still, with urban consumers eating 12% more farmed fish and 17% more fish from inland capture fisheries than those in rural areas (Toufique et al., 2017).

Price elasticities for fish of all types tend to be high – estimated at 0.8 on average, with a range of 0.31–1.04 by Naylor (2016) – and are larger for poor consumers than for the better-off. For example, the price elasticity of farmed fish in Bangladesh is reported to range from 0.78 to 1.29 for non-poor and extreme-poor consumers, respectively (Toufique et al., 2017). The poor are therefore more responsive to increases (or decreases) in fish prices than the wealthy, as also demonstrated by Dey et al. (2008) in a cross-country analysis of fish consumption in Ban- gladesh, China, India, Indonesia, Malaysia, the Philippines, Sri Lanka, Thailand and Vietnam. Rural consumers (who are poorer on average than urban) therefore have most to lose from increases in the price of non-farmed fish that accompany the contraction of wild supply, and most to gain, proportionately, from declines in the price of farmed fish that accompany the expansion of aquaculture.

The income elasticity of demand for fish is also relatively high; re- ported as −0.64 by Naylor (2016), and ranging from −0.44 to −0.51 for non-poor and extreme-poor consumers for farmed fish in Bangla- desh (Toufique et al., 2017). This tendency contributes to increasing demand for fish in economies where incomes are rising.

3.3. Price stability

The growth of aquaculture has stabilized fish prices. Food price stability is particularly important for the food security of poor con- sumers, for whom food expenditures constitute a large share of total income (Troell et al., 2014). Farmed fish prices are about half as volatile as those of wild fish on average, reflecting the seasonally variable and often unpredictable nature of fisheries, and the high degree of control over the timing of the production process that is possible in aquaculture (Asche et al., 2015). This pattern is illustrated by Fig. 2, which shows much greater seasonal fluctuations in the price of two the non-farmed species than the four cultured ones.

As aggregate production from capture fisheries appears to have reached the maximum level that can be sustained, increasing demand for wild fish can only increase prices further. However, because the supply of fish from aquaculture has kept pace with demand, substitu- tion has dampened price pressure on wild fish as demand has spilled over to farmed species, partially reigning in divergence in the prices of

farmed and wild fish over time (Tveterås et al., 2012). This finding has profound implications for food and nutrition se-

curity. Rashid et al. (2016) calculate that if aquaculture had stopped growing in 1980, global fish supply per capita in 2013 would have been about half of the actual supply in that year, and 17% less than it was in 1980. As these authors note, “the consequences of such a scenario are easy to imagine: higher prices, lower consumption, and far greater pressure on marine and inland capture fisheries.” The adverse con- sequences would have been particularly severe in Asia (Rashid et al., 2016, p. 1). In other words, the poor in our ten countries would eat far less fish of any kind - wild or farmed - were it not for the growth of aquaculture.

4. Transforming farmed fish supply

Production of farmed fish is as subject to misinterpretation as its consumption. Two narratives dominate development focussed aqua- culture science and policy. The first extolls the benefits of low intensity ‘small-scale’ aquaculture for promoting food security. The second cri- tiques ‘industrial’ forms of production, particularly shrimp farming, for undermining it. Reducing aquaculture to the idealized binary categories of ‘small-scale’ and ‘industrial’ has obscured the contributions of a very large intermediate segment of producers, existing along a gradient be- tween these two poles. This ‘missing middle’ has emerged in response to the changing patterns of demand and diet transformation outlined above, as part of a broader ‘quiet revolution’ in agri-food supply chains that is taking place in most of Asia and parts of Africa (Hernandez et al., 2017). Failure to acknowledge the role played by this set of producers has resulted in ineffective prescriptions for food security focused aquaculture development.

4.1. Polarized production narratives

Promotors of small-scale aquaculture emphasize the cumulative impact and positive nature of the contributions such farms make to food security. For instance, the High Level Panel of Experts on Food Security and Nutrition (HLPE, 2014) asserts that 70–80% of aquaculture pro- duction originates from small-scale farming. Such pronouncements are commonplace (e.g. Tacon et al., 2010; De Silva and Davy, 2010) and form a central tenet of the conventional wisdom on aquaculture. But they are also unsubstantiated - there is no global database of fish farm sizes and characteristics to provide an empirical basis for this claim.

Small-scale aquaculture is generally represented in the literature as low intensity (utilizing limited external feed inputs), and/or integrated with other agricultural systems. As such, it is seen as ecologically effi- cient in terms of a reliance on low trophic level fish species, limited dependence on nutrient inputs, and integration with components of terrestrial farming systems (e.g. through use of crop processing wastes or manures available on farm as feeds or fertilizers, or the stocking of fish in rice fields) (Edwards et al., 2002). The contribution of small- scale aquaculture to food security is usually framed in terms of meeting the subsistence fish consumption needs of rural households, and the generation of supplemental incomes through sales of small marketable surpluses that may be spent on purchases of other food items (e.g. Ahmed and Lorica, 2002).

The ‘small-scale’ narrative has provided the basis for promotion of aquaculture by overseas development projects throughout the tropics for more than 30 years. The assumption that small-scale aquaculture is underdeveloped globally and should be supported as a key strategy for improving food and nutrition security in developing economies is per- sistent. Golden et al. (2016, p. 139) argue for instance that, “when explicitly planned to improve local well-being, aquaculture can be a crucial contribution to local diets and economies … less-intensive and more-diverse forms of aquaculture may have the most potential to meet the nutrition and food-security needs of the poor”.

Literature at the opposite end of the narrative spectrum focusses on

B. Belton et al. Global Food Security 16 (2018) 85–92

89

the role of intensive, industrial aquaculture in undermining local food security. Work on ‘industrial aquaculture’ emphasizes in particular the role that export oriented shrimp aquaculture has played in “threatening both domestic food security and the economic opportunities of local communities” (van Mulekom et al., 2006, p. 546). Studies of shrimp farming provide evidence of a variety of social and environmental im- pacts on local food security, including reduced agricultural production arising of the salinization of rice-dominated terrestrial farming systems (e.g. Paprocki and Cons, 2014); displacement of small-scale capture fisheries from coastal lagoons (Nayak and Berkes, 2011); and, de- struction of mangroves (a critical nursery habitat for wild fish) to make way for shrimp ponds (e.g. Hamilton and Lovette, 2015). These pro- cesses have been a recurrent feature of shrimp farm development in many locations, particularly during ‘boom’ periods.

However, both small-scale and industrial narratives have a tendency to confuse the specific characteristics of their subject matter with the generalized attributes of the entire global aquaculture sector. The small-scale aquaculture narrative vastly overstates the importance of the idealized small, low input, integrated farm. Similarly, the industrial aquaculture narrative often extrapolates from negative food security impacts arising from local struggles over resources that characterize production of specific commodities in specific places and specific times, to divine the aggregate effects of global farmed fish production.

4.2. The missing middle

Contrary to the polarized narratives presented above, empirical evidence points to most global aquaculture output originating from a ‘missing middle’ segment of producers, characterized by five features. They are: (1) highly commercially oriented; (2) span a broad spectrum of scales of production; (3) utilize a diverse range of production tech- nologies (but are increasingly intensifying through use of pelleted feeds); (4) produce multiple, predominantly low and medium value, species and; (5) have emerged in an unplanned manner in response to opportunities created by changing patterns of demand.

For example, Chinese farms producing carps and tilapia are almost entirely commercial, with more than 95% using manufactured feeds, and produce fish in polycultures that contain an average mix of four to six fish species (Chiu et al., 2013). In Andhra Pradesh, India, the lo- cation of the largest concentration of freshwater fish farms outside China, aquaculture is dominated by semi-intensive carp and pellet-fed pangasius polycultures, oriented entirely to the market and spanning a spectrum of farm sizes from small to very large (Belton et al., 2017a). In Bangladesh, a recent survey of 2678 farms identified 14 distinct pro- duction technologies, all of them polycultures, producing a combined total of 54 different species of fish and crustaceans. Production is dominated by carps, tilapia and pangasius. While overwhelmingly below 2 ha in size, farms are strongly commercially oriented, gen- erating large marketed surpluses (Jahan et al., 2015). In Egypt's Nile Delta, the mean size of tilapia farms is 6 ha, 94% of farms use pelleted feeds, and 100% of farms sell their product to domestic wholesale markets (Eltholth et al., 2015). Aquaculture in other major Southern producer countries possesses similar characteristics. For example, a recent report synthesizing case studies from Bangladesh, Chile, Ecuador, Egypt, Indonesia, Mexico, Thailand, Viet Nam and Zambia, concluded that ‘small-scale’ aquaculture contributed less than 30% of farmed fish production in these countries (Phillips et al., 2016).

The missing middle segment of fish producers represented in all of the cases above differs markedly from the idealized small farms portrayed in developmentalist narratives. Although large numbers of households in South and Southeast Asia have benefited from enhanced nutrition and supplementary incomes generated by small farms of the traditional type, these can no longer be considered the dominant mode of production. For instance, even in Bangladesh, a country with more than 4 million small ‘homestead ponds’, two thirds of farmed fish output in 2010 originated from fully commercial farms (Belton and Azad, 2012). Moreover, even

among the homestead pond farm segment the trend is one of intensifica- tion and commercialization, with 38% of farms in Bangladesh using pel- leted feeds in 2014 (Hernandez et al, 2017).

Attempts to establish small-scale aquaculture in the traditional mold have largely failed outside of Asia. For example, concerted efforts by external development agencies in Malawi over several decades have been highly constrained by poor market development and, ultimately, the limited purchasing power of a still largely rural population (Brummett, 2000; Brummett et al., 2008). Even where well-intentioned projects have demonstrated potential for subsistence orientated small- scale production (e.g. Dey et al., 2010), outcomes have rarely been sufficiently attractive to stimulate widespread uptake. Similar failures to establish viable, self-sustaining small-scale aquaculture have been repeated throughout a succession of African and Caribbean states (Leschen and Little, 2014). The potential of integrated rice-fish farming, promoted widely as a contributor to local food security, has also been overstated. Less than 1% of the world's rice fields are deliberately stocked with fish, principally because returns to labor are often un- attractive to farm households (Edwards, 2015).

As Brummett et al. (2008) outline for Africa, the types of aqua- culture that have become most successfully established and produce the greatest volumes of fish for domestic consumers have rarely received direct support from government or donors because they are not per- ceived to represent the poor, nor offer the possibility of generating export earnings. Despite this, ‘immanent development’ (Belton and Little, 2011) of fish farms – the unplanned supply response of large numbers of producers and supporting value chain actors to emergent opportunities presented by rising domestic incomes and diet transfor- mation – is the global norm. Normative prescriptions for small-scale aquaculture ‘planning’ that continue to pervade the literature (e.g. Golden et al., 2016) are thus outdated and unlikely to be effective where implemented.

Aquaculture was once widely considered to have failed in Sub- Saharan Africa (SSA), but is now expanding rapidly in Nigeria, Uganda, Ghana, Kenya, Zambia, and South Africa. Between 2004 and 2014 there was a seven-fold increase in farmed fish production in SSA, con- centrated mainly in these countries, with an average annual growth rate of 21% (Satia, 2017). The growth of aquaculture in this region parallels the recent emergence of a substantial middle class and their demand for more diverse diets (Tschirley et al., 2015).

Fish production serving this market was pioneered mainly by large vertically integrated farms, but the value chain infrastructure estab- lished to facilitate these operations (e.g. hatcheries, feed mills, cold storage) has allowed significant numbers of medium scale producers to emerge in some locations (see Kassam and Dorward, 2017). Poorly developed supply chains and logistics, and consequent difficulties in accessing inputs and end markets, remain a challenge in other areas of SSA (Naylor, 2016). However, where these hurdles have been over- come, the high value urban market segments targeted by large farms are starting to become saturated, leading to some diversification into production and marketing of smaller, cheaper fish that are accessible to consumers in lower income brackets (see Kassam and Dorward, 2017 and Kominsky et al., In press for examples from Ghana and Zambia, respectively). There is also a small but growing role for aquaculture exports from Southern nations to SSA, which is now ranked as China's second most valuable tilapia market after the U.S. (Mao, 2016).

As noted above, one of the overriding trends in aquaculture for the past two decades has been a shift toward more intensive forms of production that utilize formulated pelleted feed diets to achieve higher levels of productivity per unit area of land. Although many extensive and semi-intensive production technologies are likley to remain viable over the medium to long term, there is no reason to expect that the dominant trend will reverse in favor of widespread expansion of the less intensive production systems advocated in the interventionist literature (e.g. Golden et al., 2016; Thilsted et al., 2016).

Moving beyond interventionist approaches opens up food security

B. Belton et al. Global Food Security 16 (2018) 85–92

90

questions that are more relevant in the context of the intensification occurring with the rise of aquaculture's ‘missing middle’. Perhaps most important is the long running controversy over whether utilization of marine ‘forage fish’ in pelleted feeds diverts food that might otherwise be used for direct human consumption away from poorer consumers, thereby compromising their nutrition security (e.g. Tacon and Metian, 2013). The extent to which use of fish for animal feeds competes di- rectly with the use of fish for human consumption remains unclear, however (Béné et al., 2015b).

Moreover, advances in the formulation of pelleted feeds mean that feed conversion ratios (an indicator of overall feed use efficiency), and fish-in fish-out ratios (a measure of the efficiency with which fish meal and fish oil are utilized) are improving (e.g. Sarker et al., 2013). Aquaculture is increasingly less dependent on marine ingredients in feeds, with rapid substitution for vegetable derived feed ingredients occurring (Roberts et al., 2015). Furthermore, an estimated 35% of all marine ingredients used in feeds are now sourced from fish processing byproducts, and aquaculture is itself becoming a major source of fish- meal and oil (Little et al., 2016). Partly as a result of these changes, the share of global fish production used as fishmeal declined from an average of 23% in the 1990s to 10% in 2012; a reduction of 10 million tonnes (HLPE, 2014).

Finally, considering the sustainability of the global food system as whole, evidence suggests that the environmental performance of in- tensive aquaculture, as measured by life cycle assessment, can be equal to or better than other animal source foods (Béné et al., 2015b). The question then, is how to build further efficiencies into aquaculture production by working with the missing middle.

5. Conclusion

The rapid rise of aquaculture has focussed attention on the sector's potential to contribute to food security, as well as its perceived failures to do so. As demonstrated here, most science and policy in this sphere lags far behind current empirical realities and fails to recognize the scale and nature of contributions that aquaculture already makes to global food security. In fact, aquaculture development in the main fish farming countries of the Global South has already averted severe de- clines in food and nutrition security.

Part of the disconnect between narrative and reality stems from an excessive focus in the literature on the international seafood trade, and on farmed aquatic commodities produced for export. Export focused research has directed attention toward a relatively small and unusually problematic set of technologies and commodities, and away from a domestic demand led ‘quiet revolution’ in farmed fish production, in which huge improvements in the availability, accessibility, and stability of fish supply have been achieved.

Instead of emphasizing aquacultural reform based on outdated narratives about the planned development of extensive small-scale aquaculture, we argue for recognition that diverse and increasingly intensive forms of commercial fish farming already make important contributions to the diets of low and middle income urban and rural consumers in Southern countries home to more than half the world's population. These latter forms of farming have the greatest potential to augment nutrient supplies from capture fisheries, and will continue to do so, whether or not much-needed improvements in capture fisheries governance and management are achieved. The transformation of fish supply, already far advanced in Asia, is now also beginning to emerge in nascent form in several of the more populous and rapidly growing countries in Africa.

Conflicts of interest

None.

References

Ahmed, M., Lorica, M.H., 2002. Improving developing country food security through aquaculture development—lessons from Asia. Food Policy 27 (2), 125–141.

Asche, F., Roll, K.H., Trollvik, T., 2009. New aquaculture species – the whitefish market. Aquac. Econ. Manag. 13, 76–93.

Asche, F., Bellemare, M.F., Roheim, C., Smith, M.D., Tveteras, T., 2015. Fair enough? Food security and the international trade of seafood. World Dev. 67, 151–160.

Barange, M., Merino, G., Blanchard, J.L., Scholtens, J., Harle, J., Allison, E.H., Allen, J.I., Holt, J., Jennings, S., 2014. Impacts of climate change on marine ecosystem pro- duction in societies dependent on fisheries. Nat. Clim. Change 4 (3), 211–216.

Belton, B., Azad, A., 2012. The characteristics and status of pond aquaculture in Bangladesh. Aquaculture 358–359, 196–204.

Belton, B., Little, D.C., 2011. Immanent and interventionist inland asian aquaculture development and its outcomes. Dev. Policy Rev. 29 (4), 459–484.

Belton, B., Hein, A., Htoo, K., Kham, L.S., Nischan, U., Kham, T., Boughton, D., 2015. Aquaculture in Transition: Value Chain Transformation, Fish and Food Security in Myanmar. Michigan State University (International Development Working Paper 139).

Belton, B., Padiyar, A.P., Ravibabu, G., Gopal Rao, K., 2017. Boom and Bust in Andhra Pradesh: development and transformation in India's domestic aquaculture value chain. Aquaculture 470, 196–206.

Belton, B., Bush, S., Little, D., 2017b. Fish production, export and domestic consumption in Bangladesh, Brazil, China, Egypt, India, Indonesia, Myanmar, Philippines, Thailand, Vietnam. Mendeley Data, v1. ⟨http://dx.doi.org/10.17632/ 9mz86zvrcc.1⟩.

Béné, C., Arthur, R., Norbury, H., Allison, E.H., Beveridge, M., Bush, S., Campling, L., Leschen, W., Little, D., Squires, D., Thilsted, S.H., Troell, M., Williams, M., 2015a. Contribution of fisheries and aquaculture to food security and poverty reduction: assessing the current evidence. World Dev. 79, 177–196.

Béné, C., Barange, M., Subasinghe, R., Pinstrup-Andersen, P., Merino, G., Hemre, G.I., Williams, M., 2015b. Feeding 9 billion by 2050 – putting fish back on the menu. Food Secur. http://dx.doi.org/10.1007/s12571-015-0427-z.

Beveridge, M.C.M., Thilsted, S.H., Phillips, M.J., Metian, M., Troell, M., Hall, S.J., 2013. Meeting the food and nutrition needs of the poor: the role of fish and the opportu- nities and challenges emerging from the rise of aquaculture. J. Fish Biol. 83 (4), 1067–1084.

Bondad-Reantaso, M.G., Subasinghe, R.P., 2013. Enhancing the contribution of small- scale aquaculture to food security, poverty alleviation and socio-economic develop- ment. In: FAO Fisheries and Aquaculture Proceedings No. 31. Rome.

Brummett, R.E., 2000. Factors influencing fish prices in Southern Malawi. Aquaculture 186 (3), 243–251.

Brummett, R.E., Lazard, J., Moehl, J., 2008. African aquaculture: realizing the potential. Food Policy 33 (5), 371–385.

Bush, S.R., 2004. A Political Ecology of Aquatic Living Resources in Lao PDR (Ph.D. thesis). School of Geociences, University of Sydney, Sydney.

Bush, S.R., Belton, B., Hall, D., Vandergeest, P., Murray, F.J., Ponte, S., Oosterveer, P., Islam, M.S., Mol, A.P.J., Hatanaka, M., Kruijssen, F., Ha, T.T.T., Little, D.C., Kusumawati, R., 2013. Certify sustainable aquaculture? Science 341 (6150), 1067–1068.

Chicago Council, 2016. Growing Food for Growing Cities: Transforming Food Systems in an Urbanizing World. The Chicago Council on Global Affairs, Chicago.

Chiu, A., Li, L., Guo, S., Bai, J., Fedor, C., Naylor, R.L., 2013. Feed and fishmeal use in the production of carp and tilapia in China. Aquaculture 414–415, 127–134.

De Silva, S.S., Davy, B.F., 2010. Success Stories in Asian Aquaculture. Springer Netherlands, Dordrecht.

Dey, M.M., Garcia, Y.T., Kumar, P., Piumsombun, S., Haque, M.S., Li, L., Radam, A., Senaratne, A., Khiem, N.T., Koeshendrajana, S., 2008. Demand for fish in Asia: a cross-country analysis. Aust. J. Agric. Resour. Econ. 52, 21–338.

Dey, M.M., Paraguas, F.J., Kambewa, P., Pemsl, D.E., 2010. The impact of integrated aquaculture–agriculture on small-scale farms in Southern Malawi. Agric. Econ. 41 (1), 67–79.

Edwards, P., 2015. Aquaculture environment interactions: past, present and likely future trends. Aquaculture 447, 2–14.

Edwards, P., Little, D.C., Demaine, H., 2002. Rural Aquaculture. CABI Publishing, Wallingford.

Eltholth, M., Fornace, K., Grace, D., Rushton, J., Häsler, B., 2015. Characterisation of production, marketing and consumption patterns of farmed tilapia in the Nile Delta of Egypt. Food Policy 51, 131–143. http://dx.doi.org/10.1016/j.foodpol.2015.01.002.

European Commission. 2011. The EU system for fisheries controls conversion factors. Annex XIII European Union Conversion Factors for Fresh Fish. ⟨http://ec.europa.eu/ fisheries/cfp/control/conversion_factors/table01.pdf⟩. (Accessed 14 July 2016).

FAO, 2015. CWP Handbook of Fishery Statistical Standards. (Accessed 14 July 2016). ⟨www.fao.org/fishery/cwp/search/en⟩.

FAO, 2016a. The State of World Fisheries and Aquaculture 2016. Food and Agriculture Organization of the United Nations, Rome.

FAO. 2016b. FishStat J database. Available for download at: ⟨http://www.fao.org/ fishery/statistics/software/fishstatj/en⟩. (Accessed 18 July 2016).

FAO. 2016c. Food Balance Sheets. Food and Agriculture Organisation of the United Nations, Rome. ⟨http://faostat3.fao.org/download/FB/CL/E⟩. (Accessed 18 July 2016).

FAO, 2016b. Asia and the Pacific Regional Overview of Food Security: Investing in a Zero Hunger Generation. Food and Agriculture Organization of the United Nations Regional Office for Asia and the Pacific, Bangkok.

Gephart, J.A., Rovenskaya, E., Dieckmann, U., Pace, M.L., Brännström, A., 2016.

B. Belton et al. Global Food Security 16 (2018) 85–92

91

Vulnerability to shocks in the global seafood trade network. Environ. Res. Lett. http://dx.doi.org/10.1088/1748-9326/11/3/035008.

Golden, C., Allison, E.H., Cheung, W.W., Dey, M.M., Halpern, B.S., McCauley, D.J., Smith, M., Vaitla, B., Zeller, D., Myers, S.S., 2016. Fall in fish catch threatens human health. Nature 534 (7607), 317–320.

Hall, S.J., Hilborn, R., Andrew, N.L., Allison, E.H., 2013. Innovations in capture fisheries are an imperative for nutrition security in the developing world. Proc. Natl. Acad. Sci. 110 (21), 8393–8398.

Hamilton, S.E., Lovette, J., 2015. Ecuador's mangrove forest carbon stocks: a spatio- temporal analysis of living carbon holdings and their depletion since the advent of commercial aquaculture. PLoS One. http://dx.doi.org/10.1371/journal.pone. 0118880.

Hernandez, R., Belton, B., Reardon, T., Hu, H., Zhang, X., Ahmed, A., 2017. The “Quiet Revolution” in the Fish Value Chain in Bangladesh. Aquaculture. http://dx.doi.org/ 10.1016/j.aquaculture.2017.06.006.

HLPE, 2014. Sustainable Fisheries and Aquaculture for Food Security and Nutrition. A Report by the High Level Panel of Experts on Food Security and Nutrition of the Committee on World Food Security. Food and Agriculture Organization, Rome.

Jahan, K.M., Belton, B., Ali, H., Dhar, G.C., Ara, I. 2015. Aquaculture technologies in Bangladesh: an assessment of technical and economic performance and producer behavior. WorldFish, Penang, Malaysia. Program Report: 2015-52, pp. 123.

Kominsky, A.M., Genschick, S., Kefi, A.S., Kruijssen, F., 2017. Commercialization and upgrading in aquaculture value chains in Sub-Saharan Africa: the case of Zambia. Aquaculture (In press).

Kassam, L., Dorward, A., 2017. A comparative assessment of the poverty impacts of pond and cage aquaculture in Ghana. Aquaculture 470, 110–122.

Kawarazuka, N., Béné, C., 2011. The potential role of small fish species in improving micronutrient deficiencies in developing countries: building evidence. Public Health Nutr. 14 (11), 1927–1938.

Kobayashi, M., Msangi, S., Batka, M., Vannuccini, S., Dey, M.M., Anderson, J.L., 2015. Fish to 2030: the role and opportunity for aquaculture. Aquacult.Econ. Manag. 19 (3), 282–300.

Leschen, W., Little, D.C., 2014. Study on the Potential of Aquaculture in ACP (African, Caribbean, Pacific) countries: An Overview Report of the Three Regions. Institute of Aquaculture, University of Stirling, Scotland, UK, pp. 76.

Lewis, D., 1997. Rethinking aquaculture for resource-poor farmers: perspectives from Bangladesh. Food Policy 22 (6), 533–546.

Little, D.C., Newton, R.W., Beveridge, M.C.M., 2016. Aquaculture: a rapidly growing and significant source of sustainable food? Status, transitions and potential. Proc. Nutr. Soc. 75, 274–286.

Macfadyen, G., Nasr Allah, A., Kenawy, D., Ahmed, M., Hebicha, H., Diab, A., Hussein, S., Abouzied, R., El Naggar, G., 2011. Value-Chain Analysis of Egyptian Aquaculture. Project report 2011-54. The WorldFish Center, Penang, Malaysia (84 pp).

Mao, G.F. 2016. Inside China’s Current Tilapia Trade. March 3, 2016. ⟨http://www. seafoodsource.com/all-commentary/inside-china-s-current-tilapia-trade⟩. (Accessed 15 Feburary 2017).

McIntyre, P.B., Liermann, C.A.R., Revenga, C., 2016. Linking freshwater fishery man- agement to global food security and biodiversity conservation. Proc. Natl. Acad. Sci. USA 113 (45), 12880–12885.

Nayak, P.K., Berkes, F., 2011. Whose marginalisation? Politics around environmental injustices in India's Chilika lagoon. Local Environ. 15 (6), 553–567.

Naylor, R.L., 2016. Oil crops, aquaculture, and the rising role of demand: a fresh per- spective on food security. Glob. Food Secur. 11, 17–25. http://dx.doi.org/10.1016/j. gfs.2016.05.001.

Paprocki, K., Cons, J., 2014. Life in a shrimp zone: aqua- and other cultures of Bangladesh's coastal landscape. J. Peasant Stud. http://dx.doi.org/10.1080/ 03066150.2014.937709.

Pauly, D., Zeller, D., 2016. Catch reconstructions reveal that global marine fisheries

catches are higher than reported and declining. Nat. Commun. http://dx.doi.org/10. 1038/ncomms10244.

Phillips, M.J., Subasinghe, R.P., Tran, N., Kassam, L., Chan, C.Y., 2016. Aquaculture Big Numbers. FAO Fisheries and Aquaculture Technical Paper 601. Food and Agriculture Orgaization of the Untied Nations, Rome.

Ponte, S., Kelling, I., Jespersen, K.S., Kruijssen, F., 2014. The blue revolution in Asia: upgrading and governance in aquaculture value chains. World Dev. 64, 52–64.

Rashid, S., Minot, N., Lemma, S., 2016. Does a “Blue Revolution” Help the Poor? Evidence from Bangladesh (IFPRI Discussion Paper 01576). International Food Policy Research Institute, Washington D.C.

Reardon, T., Timmer, C.P., 2014. Five inter-linked transformations in the Asian agrifood economy: food security implications. Glob. Food Secur. 3 (2), 108–117.

Reardon, T., Tschirley, D., Dolislager, M., Snyder, J., Hu, C., White, S., 2014. Urbanization, Diet Change, and Transformation of Food Supply Chains in Asia. Michigan State University, Global Center for Food Systems Innovation, Lansing.

Roberts, C.A., Newton, R.W., Bostock, J.C., et al., 2015. A Risk Benefit Analysis of Mariculture as a Means to Reduce the Impacts of Terrestrial Production of Food and Energy. SARF 106. Scottish Aquaculture Research Forum (SARF)/World Wildlife Fund (WWF). SARF, Pitlochry, UK.

Sarker, P.K., Bureau, D.P., Hua, K., Drew, M.D., Forster, I., Were, K., Vandenberg, G.W., 2013. Sustainability issues related to feeding salmonids: a Canadian perspective. Rev. Aquac. 5 (4), 199–219. http://dx.doi.org/10.1111/raq.12013.

Satia, B.P., 2017. Regional Review on Status and Trends in Aquaculture Development in Sub-Saharan Africa – 2015. FAO Fisheries and Aquaculture Circular No. 1135/4. Food and Agriculture Organization of the United Nations, Rome.

Smith, M.D., Roheim, C.A., Crowder, L.B., et al., 2010. Sustainability and global seafood. Science 327, 784–786.

Tacon, A.G.J., Metian, M., 2013. Fish matters: importance of aquatic foods in human nutrition and global food supply. Rev. Fish. Sci. 21 (1), 22–38.

Tacon, A.G.J., Hasan, M.R., Subasinghe, R.P., 2006. Use of Fishery Resources as Feed Inputs for Aquaculture Development: Trends and Policy Implications. FAO Fisheries Circular. No.1018. Food and Agriculture Organization of the United Nations, Rome.

Tacon, A.G.J., Metian, M., Turchini, G.M., De Silva, S.S., 2010. Responsible aquaculture and trophic level implications to global fish supply. Rev. Fish. Sci. 18 (1), 94–105.

Thilsted, S.H., Thorne-Lyman, A., Webb, P., Bogard, J.R., Subasinghe, R., Phillips, M.J., et al., 2016. Sustaining healthy diets: the role of capture fisheries and aquaculture for improving nutrition in the post-2015 era. Food Policy 61, 126–131.

Timmer, C.P., Falcon, W.P., Pearson, S.R., Agriculture, W.B., Economics, R.D. D., Division, P. 1983. Food Policy Analysis.

Toufique, K.A., Belton, B., 2014. Is aquaculture pro-poor? Empirical evidence of impacts on fish consumption from Bangladesh. World Dev. 64, 609–620.

Toufique, K.A., Farook, S., Belton, B., 2017. Managing fisheries for food security: im- plications from demand analysis. Mar. Resour. Econ. 33 (1). http://dx.doi.org/10. 1086/694792.

Troell, M., Naylor, R.L., Metian, M., Beveridge, M., Tyedmers, P.H., Folke, C., Arrow, K.J., et al., 2014. Does aquaculture add resilience to the global food system? Proc. Natl. Acad. Sci. USA 111 (37), 13257–13263.

Tschirley, D., Reardon, T., Dolislager, M., Snyder, J., 2015. The rise of a middle class in East and Southern Africa: implications for food system transformation. J. Int. Dev. 27, 628–646.

Tveterås, S., Asche, F., Bellemare, M.F., Smith, M.D., Guttormsen, A.G., Lem, A., Lien, K., Vanuccini, S., 2012. Fish is food: the FAO's fish price index. PLoS One 7 (5), E36731.

van Mulekom, L., Axelsson, A., Batungbacal, E.P., Baxter, D., Siregar, R., de la Torre, I., 2006. Trade and export orientation of fisheries in Southeast Asia: under-priced export at the expense of domestic food security and local economies. Ocean Coast. Manag. 49, 546–561.

World Bank, 2016. World Bank. ⟨https://data.worldbank.org/indicator/SN.ITK.DEFC. ZS⟩. (Accessed 18 February 2016).

B. Belton et al. Global Food Security 16 (2018) 85–92

92