Upgrading rice value chains: Experimental evidence from 11 African markets

Global Food Security ∎ (∎∎∎∎) ∎∎∎–∎∎∎

Contents lists available at ScienceDirect

Global Food Security journal homepage: www.elsevier.com/locate/gfs

Upgrading rice value chains: Experimental evidence from 11 African markets Matty Demont a,n, Maïmouna Ndour b a b

International Rice Research Institute (IRRI), Los Baños, Laguna, Philippines Africa Rice Center (AfricaRice), Saint-Louis, Senegal

art ic l e i nf o

a b s t r a c t

Article history: Received 24 July 2014 Received in revised form 19 September 2014 Accepted 3 October 2014

Previous research has advanced the hypothesis that improving food security in Africa will require upgrading rice value chains in order to increase the quality-based competitiveness of domestic rice relative to imported rice in urban markets. We review the experimental evidence in support of this hypothesis. During 2008–2012, a research program of framed field experiments based on experimental auctions was implemented to study consumers’ revealed preferences for rice quality attributes in 11 African markets. The experimental results suggest that domestic rice can compete with imported rice in urban markets if its intrinsic and extrinsic quality attributes are better tailored to urban consumer preferences. This is important for policy makers who are currently implementing ambitious national rice development strategies throughout Africa. & 2014 Elsevier B.V. All rights reserved.

Keywords: Value-chain upgrading Agricultural policy Quality Labeling Experimental auctions Willingness to pay

1. Introduction In the wake of the 2008 food crisis, the Coalition for African Rice Development (CARD) was launched with the aim of doubling rice production in sub-Saharan Africa within 10 years, i.e., from 14 million tonnes in 2008 to 28 Mt in 2018. The governments of African CARD members have subsequently developed ambitious national rice development strategies (NRDS) (CARD, 2014) in order to reduce reliance on rice imports which have been increasing in sub-Saharan Africa since the sixties at an annual rate of 2.2% to reach 53% of total rice consumption in 2011 (Demont, 2013; FAO, 2014a). The CARD initiative follows the perspective of the food sovereignty movement which argues that long-term food security cannot rely on dependency on food imports, but must be built on investment in agriculture (Laroche Dupraz and Postolle, 2013) and food value chains (FAO, 2012; UNCTAD, 2009; World Bank, 2008). A food value chain is defined here as “the full range of farms and firms and their successive coordinated value-adding activities that produce particular raw agricultural materials and transform them into particular food products that are sold to final consumers and disposed of after use” (FAO, 2014b, p. 6). In order to identify priority areas and investment gaps for the development of the rice sector in Africa, Demont (2013) reviewed the NRDS that have been developed under the CARD initiative and

n

Corresponding author. Tel.: þ 63 25805600; fax: þ 63 25805699. E-mail address: [email protected] (M. Demont).

advanced the hypothesis that in order to improve food security in Africa, more attention will need to be given to rice value-chain upgrading in order to increase quality-based competitiveness of domestic rice. Value chain upgrading is defined as a set of strategic and innovative actions and investments aiming at improving the performance (competitiveness) of a value chain. FAO (2014b) further distinguishes between what is being upgraded or what the upgrade aims to achieve. The first category of upgrading includes technology (e.g. varietal improvement or modernization of processing infrastructure), organization (e.g. collective purchase of inputs by a farmer group or collective decision-making in irrigation), network (e.g. horizontal and vertical coordination and integration), and institution (e.g. improved trade or seed laws). The second category includes process (e.g. introducing food-safety protocols or good agricultural practices—GAP), product/market (e.g. from traditional to modern retail), and function (e.g. farmers integrating transport to the market or processors integrating custom milling services into their activities). Demont’s (2013) hypothesis is that a focus on productivity alone may be insufficient to render domestically produced rice competitive against imported rice, particularly in import-biased food markets where imports have established the quality standards against which domestic rice now has to compete. However, in some cases upgrading quality may be insufficient for achieving competitiveness of domestic rice. Due to persistently inferior postharvest quality, domestic rice suffers from a bad reputation, particularly in major African cities which could otherwise be huge markets for rice growers and which are now served by Asian rice

http://dx.doi.org/10.1016/j.gfs.2014.10.001 2211-9124/& 2014 Elsevier B.V. All rights reserved.

Please cite this article as: Demont, M., Ndour, M., Upgrading rice value chains: Experimental evidence from 11 African markets. Global Food Security (2014), http://dx.doi.org/10.1016/j.gfs.2014.10.001i

M. Demont, M. Ndour / Global Food Security ∎ (∎∎∎∎) ∎∎∎–∎∎∎

2

exporters. If the quality factor alone was at work, investment in upgrading of intrinsic quality would be sufficient. Should preferences for imported rice persist after quality upgrading, however, additional investment would be required in extrinsic quality cues such as packaging and branding in order to dedifferentiate domestic rice from imported rice. Once rice value chains are upgraded, investment in domestic rice productivity is expected to become more effective in improving food security in Africa. In this article, we review the experimental evidence in support of this hypothesis. In 2008, the Africa Rice Center (AfricaRice) initiated a program of experimental economics research in order to generate empirical evidence for supporting rice value-chain upgrading in Africa (Tollens et al., 2013). A protocol for ‘framed field experiments’ (see Harrison and List, 2004 for a classification of field experiments) based on experimental auctions was designed and implemented in 11 urban markets throughout Africa during 2008–2012. Experimental auctions are useful tools for eliciting consumers’ willingness to pay (WTP) for products or product attributes that are not yet available on the market. Because real products and real money are exchanged in experimental auctions, participants have increased incentives to reveal their true value for the products or attributes under research (Lusk and Shogren, 2007). The experiments aimed at eliciting consumers’ revealed preferences for upgraded rice quality attributes and assessing under which conditions they are willing to substitute imported for local rice (Akoa Etoa et al., in press; Costello et al., 2013; Demont et al., 2013a, 2013b, 2013c, 2012; Ouedraogo et al., 2013; Zossou et al., 2013). Shoppers were randomly approached in urban markets and invited to take part in an experiment during which they bid on alternative rice types under controlled conditions (e.g., intrinsic visual and sensory quality of the product and extrinsic quality cues like labeling and information). The advantage of field experiments over classical surveys is that they allow empirical testing of whether and under what conditions rice value chains can be upgraded, providing crucial information on how resource-poor farmers can tap into huge urban consumption zones. The experimental results suggest that domestic rice can compete with imported rice in urban markets if its intrinsic and extrinsic quality attributes are better tailored to urban consumer preferences. We conclude that quality upgrading, labeling, and information are crucial sources of competitiveness of domestic rice value chains in Africa.

2. Design of the framed field experiments 2.1. Choice of the markets Demont (2013) distinguishes between three categories of CARD countries for which different upgrading strategies may apply: (i) coastal countries characterized by dominant consumer preferences for imported rice (Benin, Cameroon, Côte d’Ivoire, Ghana, Liberia, Nigeria, Senegal, and Togo); (ii) coastal countries characterized by dominant consumer preferences for local rice (Guinea, Kenya, Madagascar, Mozambique, Sierra Leone, The Gambia, and Tanzania); and (iii) landlocked countries (Burkina Faso, Central African Republic—CAR, Democratic Republic of the Congo—DR Congo, Ethiopia, Mali, Rwanda, Uganda, and Zambia). The three groups are shown in different colors in Fig. 1. Senegal is a special case: the region north of the Gambia River to the Senegal River valley (supplying the major cities Saint-Louis and Dakar) can be categorized under Group 1, while the southern Casamance region (supplying Kolda) can be categorized under Group 2, together with The Gambia, where rice is an indigenous crop grown for centuries.

Demont’s (2013) main argument is that urban consumers in coastal countries are typically more susceptible to developing food preferences biased toward imported rice, particularly if the primary consumption zone is close to a port. In some coastal countries, urban markets are already marked by dominant importbiased rice preferences to the point of paying price premiums for imported relative to local rice (Group 1), while in other coastal countries, consumers have preserved their preferences for local rice reflected through the price premiums they are paying for local rice, despite their exposure to imported rice (Group 2). The bias toward locally produced rice endows Group 2 countries with a ‘comparative advantage in demand’ in the development of their rice sectors relative to Group 1 countries. The term was coined by Lusk et al. (2006) to argue that differences in the demand side can play an equally important role in comparative advantage. Improving food security through domestic rice production in Group 1 countries requires rice value chains to be upgraded following a judicious investment portfolio, whereby quality upgrading needs to precede upgrading of productivity (Demont, 2013; Demont and Rizzotto, 2012). The reason is that investments in productivity only become profitable if there is a market for absorbing the surplus created (Barrett, 2008). If production and processing infrastructure are unable to assure minimal quality standards, increased production volumes will rapidly saturate local markets and erode prices, jeopardizing farmers’ incentives to invest in productivity (Demont, 2013; Demont and Rizzotto, 2012). The need for sequencing becomes less critical when one moves to countries that are somewhat shielded from import bias through (i) cultural barriers, such as coastal countries endowed with a comparative advantage in demand (Group 2), and (ii) physical barriers such as being landlocked (Group 3) (Demont, 2013). In order to generate empirical evidence for the hypothesis advanced by Demont (2013), we conducted 14 framed field experiments in 11 major African markets across six CARD member countries (Benin, Burkina Faso, Cameroon, Senegal, The Gambia, and Uganda) and one non-member country (Mauritania) and across the three categories (Groups 1–3). The locations of the experiments are mapped in Fig. 1. Most of the framed field experiments were carried out on an urban market in or near the (densely populated) commercial capitals (Dantokpa market in Cotonou, Yaoundé, Nouakchott, Dakar, Serre Kunda, Ouagadougou, and Kampala), most of them near the port (Cotonou, Nouakchott, Dakar, and Serre Kunda). For comparison, experiments were also carried out in important provincial towns in rice production zones (Glazoué, Malanville, Saint-Louis, and Kolda). The majority of the experiments were conducted in Group 1 countries (Benin, Cameroon, Mauritania, and Senegal) (Akoa Etoa et al., in press; Costello et al., 2013; Demont et al., 2013a, 2013b, 2013c, 2012; Zossou et al., 2013). However, despite their comparative advantage in demand, some of the Group 2 countries are also exposed to massive rice imports and so local rice needs to compete against the latter quality-wise. Therefore, experiments were carried out in Senegal’s southern Casamance region (Kolda) and The Gambia to assess consumer response to newly introduced improved rice varieties, such as NERICA (Demont et al., 2013a). Like the cultural barriers (i.e., favorable preferences for local rice) in Group 2 countries, the landlocked position of Group 3 countries does not entirely shield them from rice imports. Burkina Faso, for instance, imports 60% of its rice consumption needs via the ports of Côte d’Ivoire, Ghana, and Togo (USDA, 2012) and we deemed it useful to assess the competitiveness of local rice (e.g., from the Bagré region) on the major urban consumption market of the country, i.e., the capital Ouagadougou (Ouedraogo et al., 2013). Finally, although varietal quality improvement can be a major driver for rice productivity (e.g. Kijima et al., 2011), its success also crucially hinges on urban markets, where improved rice

Please cite this article as: Demont, M., Ndour, M., Upgrading rice value chains: Experimental evidence from 11 African markets. Global Food Security (2014), http://dx.doi.org/10.1016/j.gfs.2014.10.001i

M. Demont, M. Ndour / Global Food Security ∎ (∎∎∎∎) ∎∎∎–∎∎∎

3

Fig. 1. Sample of 23 CARD member countries and 11 African markets where experimental auctions have been conducted. Group 1: Coastal countries characterized by dominant consumer preferences for imported rice. Group 2: Coastal countries characterized by dominant consumer preferences for local rice. Group 3: Landlocked countries. (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.) Source: Adapted from Demont (2013) and based on CARD (2014).

varieties now have to compete against existing local rice varieties. We illustrate this with the case of the Kampala market in a Group 3 country (Uganda). 2.2. Choice of the products An important aspect of the research question is the choice of the rice types that will be used in the auctions. With the idea of value-chain upgrading in mind, all experimental auctions were based on the endow-and-upgrade method, which requires defining a ‘standard’ rice of inferior quality and eliciting WTP for upgrading to superior-quality alternatives (Demont et al., 2013a, 2013b, 2013c, 2012). We typically chose a standard rice type that was commonly available on the market and then purposely selected a set of alternative rice types to be auctioned such that they differed only in a bundle of quality attributes (intrinsic

or extrinsic) that are relevant to rice value-chain upgrading and quality-based competitiveness of domestic rice relative to imported rice. Intrinsic quality attributes include color, cleanliness, purity, grain shape and size, grain homogeneity, proportion of head rice, aroma, taste, swelling capacity, etc., while extrinsic quality cues include packaging, labeling, branding, reputation and information (exogenous and endogenous). We report the ‘standard’ rice types used during the experiments in Table 1. The choice of the standard depended, first, on the research question. If the focus was on consumers’ valuation of upgraded processing technologies, the standard would typically represent the rice obtained through traditional (e.g., parboiling in Benin and Cameroon) or standard processing equipment (e.g., milling in Burkina Faso, Cameroon, Mauritania, and Senegal), leading to rice with inferior (impure, broken, and mixed) grain quality. If the research focused on the impact of varietal quality

Please cite this article as: Demont, M., Ndour, M., Upgrading rice value chains: Experimental evidence from 11 African markets. Global Food Security (2014), http://dx.doi.org/10.1016/j.gfs.2014.10.001i

M. Demont, M. Ndour / Global Food Security ∎ (∎∎∎∎) ∎∎∎–∎∎∎

4

improvement on competitiveness of local rice with imported rice, a non-improved imported rice type was chosen as a standard and auctioned for upgrading to a set of improved rice varieties including a set of local varieties and an imported upperbenchmark check. For example, in order to study WTP for rice fragrance in Dakar, non-fragrant imported rice was chosen as a standard and auctioned against a set of fragrant rice types, including local varieties (Sahel 177, Sahel 328, and Sahel 329) and imported ‘Thai Hom Mali’ rice. The choice of the standard rice also depended on the context (Demont, 2013). For Group 1 countries, we would typically choose local rice as a standard and then include imported rice as one of the upgrades against which the locally upgraded rice products had to compete. In Group 2 regions such as Senegal’s Kolda region and The Gambia, imported rice was typically used as a standard. A standard was always chosen that was perceived to be equal or inferior to the upgrades in terms of quality and price by the majority of consumers in the market. In some cases (e.g., Mauritania), we split the experiment and switched between local and imported rice to check whether the choice of the standard affected consumers’ relative valuation of rice quality. 2.3. Auction mechanism and procedure For the majority of the framed field experiments, the Vickrey (1961) second-price auction was chosen, where in a group context each participant submits a sealed bid on a product and the highest bidder buys the product at the second price. The detailed procedure is described in Demont et al. (2013a, 2013b, 2013c, 2012). For research exclusively focused on extrinsic quality cues of rice, we used the Becker–Degroot–Marschak (BDM) auction mechanism, where participants individually bid against an unknown predetermined sealed market price (Becker et al., 1964; Costello et al., 2013; Lusk et al., 2001). We focused on women as they are the major decision-makers in rice purchasing throughout Africa (Demont et al., 2012). Vickrey auctions typically involved 8–10 experimental sessions over the course of 4–5 days, organized in a conference hall close to the market. For each session, we randomly recruited 10–15 women from the market, generating samples of 100–135 participants per experiment. BDM auctions were conducted individually in a shop on the market and took 3 days to obtain a sample of 120 participants. All 14 experiments together generated a joint sample of 1610 female consumers from 11 African cities. We started the experimental session in the popular local language with translations to other local languages, French, or English where necessary. Multiple rice products were auctioned simultaneously over multiple bidding rounds, separated by treatments (see Section 2.4). The rice types were presented on colored plates on each participant’s table. Each plate on the individual tables contained one kilogram of the rice type. Where we wanted to assess the impact of an extrinsic quality cue such as a label, we simulated a market setting by presenting the alternative rice types in 25 kg bags on a table at the front of the conference room and distributed a single rice type over two bags, i.e., one unlabeled plain white bag and another labeled bag. We clearly explained to the participants how the rice plates corresponded to the rice bags on the table. During the experiment, participants could examine the visual (purity and homogeneity) and sensory (taste and aroma) quality attributes of the uncooked rice types. We explained the auction procedure to the participants and then proceeded with a training session with other food products in order to familiarize the participants with the auction procedure. We used a two-stage approach to elicit WTP (Haines et al., 1988). For each alternative rice type, we first elicited willingness to upgrade (WTU) standard rice to each of the upgrades as a

dichotomous ‘adoption’ variable (willing to upgrade or not). We then elicited willingness to pay (WTP) for upgrading standard rice into each of the upgrades for which a non-zero WTU was elicited. Responses were recorded privately for each participant. During and after the auction rounds, survey questionnaires were administered to the participants to collect socio-demographic data and further information on consumer preferences and awareness of the alternative rice types used in the experiment. Finally, we randomly selected one rice type and one of the bidding rounds as binding, deducted the second price from the participation fees of the winning bidders, and distributed the rice and the adjusted participation fees to the participants. 2.4. Treatments In experiments, a treatment is something that researchers administer to experimental units in order to assess the impact of the treatment on some key variables of interest, i.e., WTU and WTP in our case. Three types of treatment were used before or between the individual auction rounds. First, information treatments were typically administered as ‘between-subjects’ treatments at the onset of an experimental session. In the experimental auctions on improved parboiled rice in Benin, for example, a sample of 270 participants was equally split into a control group and two treatment groups, which were treated with a short video projection or a short radio transcript on an improved technology or rice product (Zossou et al., 2013). Second, sensory tests were inserted between the individual auction rounds as ‘within-subjects’ treatments, during which the participants could experience the sensory quality attributes (aroma, taste, texture, and stickiness) and observe the swelling capacity of the cooked rice types. Finally, ‘collective induction treatments’ were inserted between the individual auction rounds as ‘within-subjects’ treatments. During a collective induction treatment, participants attempt to reach consensus on valuation for the products under research. The treatment acts as an aggregator of endogenous information and preferences and was introduced by Demont et al. (2013c) for eliciting word-of-mouth (WOM) communication among participants and capturing the influence of endogenous information on consumers’ acceptability (WTU) and valuation (WTP) of upgraded rice quality attributes.

3. Results and discussion 3.1. Willingness to upgrade and willingness to pay The experimental auctions elicited urban consumers’ WTU standard rice to rice with superior intrinsic and extrinsic quality attributes and the relative price premiums (WTP) they were willing to pay for these attributes over different auction rounds. In order to review the empirical evidence collected during 14 experiments in 11 African markets and compare the general magnitudes of the WTU and WTP levels, we analyzed the raw WTU and WTP data in the 14 datasets resulting from the experiments. In Table 1 we report the range of WTU rates and WTP price premiums among different quality attributes. As mentioned before, WTU was recorded as a dichotomous ‘adoption’ variable (willing to upgrade or not) and the coefficients in the table represent the change in the ‘market share’, i.e. the percentage of participants in the sample who were willing to upgrade to rice with superior intrinsic and extrinsic quality attributes. WTP price premiums were averaged over auction rounds and expressed as the percentage price increase participants in the sample were willing to pay relative to the price of standard rice. By exposing consumers to a set of rice attributes controlled through the choice

Please cite this article as: Demont, M., Ndour, M., Upgrading rice value chains: Experimental evidence from 11 African markets. Global Food Security (2014), http://dx.doi.org/10.1016/j.gfs.2014.10.001i

M. Demont, M. Ndour / Global Food Security ∎ (∎∎∎∎) ∎∎∎–∎∎∎

of the rice products and the treatments and subsequently recording the change in their purchase behavior through the metrics WTU and WTP, we believe that we have adequately captured the values of those attributes. The data are analyzed more rigorously through econometric methods in the papers referenced below the table. A first observation that can be made from the results (Table 1) is that the more one moves from Group 1 (Benin, Cameroon, Mauritania, and Senegal) to Group 2 (Kolda in Senegal and The Gambia) and Group 3 countries (Burkina Faso and Uganda), the more the attention of the experimental auctions shifts from processing to varietal quality improvement. Obsolete and substandard (relative to imported rice) processing technologies (parboiling, milling, sorting, cleaning, grading, etc.) are often the primary constraint to quality-based competitiveness in importbiased countries. Due to mixing of rice varieties at the milling stage, any incentives for pre- and postharvest quality upgrading and adoption of varieties with improved quality characteristics are largely annihilated (Demont et al., 2013b; EUCORD, 2012). In some Group 1 countries such as Senegal, substantial improvements have been made in value-chain upgrading (e.g., postharvest processing infrastructure and vertical coordination); as a result, increased attention to varietal quality improvement and even packaging and branding is observed. This is reflected in the focus of the experimental auctions during 2008–2012, which progressively shifted from processing (Demont et al., 2013b, 2013c) to varietal quality improvement (Demont et al., 2013a) and branding (Costello et al., 2013). Local fragrant rice from the Senegal River valley has now been introduced in urban markets where consumers are used to

5

the characteristics of imported fragrant rice (e.g., Thai Hom Mali). This has resulted in huge payoffs for the sector, i.e., 36–44% value addition as revealed by consumers in Dakar (Table 1). Consumers’ WTU to rice with superior intrinsic quality attributes is generally high and indicates a segment of ‘upgraders’ or ‘adopters’ of about two-thirds of the overall population. Although WTU rates were similar, in terms of WTP, varietal quality improvement tended to add slightly more value (about a third) than upgrading of processing (about a quarter) (Table 1). Nevertheless, these ballpark figures indicate that there is a huge market for local quality rice into which rice farmers can tap if varietal and postharvest quality characteristics are tailored to urban consumer preferences. The case of Burkina Faso is worth mentioning: when exposed to local rice from the Bagré region, more than half of the urban consumers in the experiment switched from imported rice to local rice, even without additional investment in postharvest product quality. This comparative advantage in demand translated into substantial price premiums (25%) for ‘localness,’ compared to the smaller premiums (2–11%) for improved processing such as sorting, cleaning, and parboiling (Ouedraogo et al., 2013). Group 1 countries, in contrast, do not benefit from a comparative advantage in demand. Their best short- and medium-term option is to first quality-dedifferentiate local from imported rice by bringing the quality and market presentation of local rice up to the level of imported rice products such that local rice blends into the set of competing rice brands on the market and that foreignness is no longer a search characteristic for superior quality (Demont, 2013). Akoa Etoa et al. (in press) confirm that dedifferentiation is indeed a determining factor in urban consumers’ valuation of

Table 1 Consumers’ willingness to upgrade (WTU) standard rice to rice with superior intrinsic and extrinsic quality attributes and relative price premiums (WTP) they are willing to pay for these attributes in 11 African cities. Country

Benin

City

Glazoué Cotonou Malanville Cameroon Yaoundé Mauritania Nouakchott Nouakchott Senegal Saint-Louis Saint-Louis Dakar Dakar Dakar Kolda The Serre Kunda Gambia Burkina Ouagadougou Faso Uganda Kampala

Year

2009 2011 2011 2012 2009 2009 2008 2012 2009 2011 2012 2012 2010

Methodology Sample size

Willingness to upgrade (WTU)

Willingness to pay (WTP)

Standard rice

Intrinsic

Intrinsic

Origin

Type

Variety (%)

Parboiled Parboiled Parboiled Mixed Mixed 100%B Mixed F100%B Mixed 100%B 58–63 F100%B 100%B 86–88 100%B 54–67

Vickrey Vickrey Vickrey Vickrey Vickrey Vickrey Vickrey BDM Vickrey Vickrey BDM Vickrey Vickrey

100 135 135 120 50 50 99 121 100 120 120 120 100

Local Local Local Local Local Import Local Local Local Import Local Import Import

2012 Vickrey

120

Import 5%B

52

2011 Vickrey

120

Local

67–83

Mixed

Extrinsic Processing (%)

Label (%)

Info (%)

Variety (%)

Extrinsic Processing (%)

85–98 95 87 75–89 32–54 62–81 47–75

6 6 2

14–30 16 18 17–39 34–38 25–26 43–47

27–73

5

32–40

0–2 7–17

Label (%)

a

22–35

b

d d e f g h f d

22–35 32–33 25

b

c

4 2 6 17 4 16

12–19

Info (%)

4–8 6–13

36–44

14

Source

2 2–11

h

i

d

Notes: B¼ broken; F¼ fragrant. WTU was recorded as a dichotomous ‘adoption’ variable (willing to upgrade or not) and the coefficients in the table represent the change in the percentage of the participants in the sample who were willing to upgrade to rice with superior intrinsic and extrinsic quality attributes after being exposed to those attributes. WTP price premiums were averaged over auction rounds and expressed as the percentage price increase participants in the sample were willing to pay relative to the price of standard rice. In Vickrey (1961) auctions, participants submit sealed bids on a product and the highest bidder buys the product at the second price. In BDM (Becker et al., 1964) auctions, participants individually bid against an unknown predetermined sealed market price. a

Demont et al. (2012). Zossou et al. (2013). c Akoa Etoa et al. (in press). d Unpublished dataset. e Demont et al. (2013b, 2013c). f Costello et al. (2013). g Demont et al. (2013b). h Demont et al. (2013a). i Ouedraogo et al. (2013). b

Please cite this article as: Demont, M., Ndour, M., Upgrading rice value chains: Experimental evidence from 11 African markets. Global Food Security (2014), http://dx.doi.org/10.1016/j.gfs.2014.10.001i

6

M. Demont, M. Ndour / Global Food Security ∎ (∎∎∎∎) ∎∎∎–∎∎∎

quality upgrading of local rice: only those consumers who perceived the locally improved product as being ‘imported’ (two-thirds of the participants) paid significant price premiums for it. This suggests that the major value of the improved processing technology (upgraded parboiler in this case) lies in its ability to successfully dedifferentiate local from imported rice (Akoa Etoa et al., in press). The incremental impact of extrinsic quality attributes on WTU and WTP, such as a local label or information, is generally smaller by an order of magnitude. Labeling and information can serve as extrinsic cues that supplement the use of intrinsic quality cues to reduce quality uncertainty (Akerlof, 1970) and form quality expectations (Verlegh and Steenkamp, 1999). Although the experimental evidence on the impact of labeling on market share (WTU) of local quality rice is mixed and often insignificant (Demont et al., 2013b, 2013c), it has been consistently shown to be a powerful driver for building value (WTP) of a rice product (Costello et al., 2013; Demont et al., 2013a, 2013b, 2013c). Costello et al. (2013) conducted BDM auctions in two major cities in Senegal in order to analyze whether local rice should be marketed either (i) through brands that are perceived as being ‘international’ and hence dedifferentiated from imported rice brands, or (ii) through brands that are clearly differentiated through their ‘localness,’ i.e., by using brand names in local language and symbols referring to Senegal or Africa. They presented the same rice in two differently branded bags and asked participants which rice type they preferred by looking at the bags and checking out the (identical) rice in both bags. They then endowed participants with one kilogram of their non-preferred rice brand and asked them how much they were willing to pay to upgrade their rice to their preferred brand in an auction in which they are bidding against an unknown predetermined sealed market price. They found that urban consumers are willing to pay 16–17% price premiums for their preferred rice brand and can be classified into two market segments with preferences for either brand type. Preferences for ‘local’ brands tended to be more pronounced in the production zones. Some companies attempt to exploit foreignness as a persistent search characteristic for superior quality by marketing local rice through ‘international’ brands without emphasizing the local origin of the product (e.g., ‘Rixel’ by the Senegalese company Vital), while others try to capture both market segments simultaneously through a ‘hybrid’ rice brand. The brand ‘Terral’ introduced in Senegal by the Belgium-based impact investing company Durabilis in 2012, for example, is a combination of a local name and an international (Indian) symbol (Costello et al., 2013; FAO, 2014b). Information has been similarly demonstrated to be a significant contributing factor for raising awareness which results in a higher likelihood of purchase (Demont et al., 2013b) and for building value of a quality rice product (Demont et al., 2012). For example, exposing consumers to a video excerpt or radio transcript on the nutritional value of rice processed through improved parboiling technologies (Fofana et al., 2011; Zossou et al., 2009) significantly increases their WTP for the end-product. Visual information added twice as much value (12%) as auditory information (6%). Important to note is the fact that information had an opposite effect on the competing product: after exposure to the visual and auditory information, consumers significantly shifted away from imported rice and transferred value from the latter to the ‘advertised’ improved local product (Zossou et al., 2013). Endogenous information such as WOM communication can also be a critical factor in the acceptance and value of a new rice product. The collective induction treatment captures the impact of WOM (Demont et al., 2013c), such that: (i) the damage from negative WOM can be anticipated in an early stage of product development and minimized – this may be critical in Group 1 countries where local rice suffers from a bad reputation; and

(ii) the spread of positive WOM (e.g., in Group 2 and 3 countries) can be encouraged (Allsop et al., 2007). WOM has been shown to significantly shift 10–11% of consumers from imported rice to locally improved parboiled rice in Group 1 countries such as Benin, in addition to a 14% gain of market share after exposure to exogenous information transmitted through rural radio. WOM had a dampening effect on value though: while video and radio transferred 6–12% of value from imported rice to locally improved parboiled rice, WOM transferred 2% of the value back to the competing product. Finally, in urban Burkina Faso, WOM significantly boosted the value of raw rice and the acceptance of parboiled rice from the Bagré region (Ouedraogo et al., 2013), which indicates that comparative advantage in demand can be further amplified endogenously through ‘WOM multiplicator’ effects (Hogan et al., 2004). 3.2. Determinants of willingness to upgrade and willingness to pay The determinants of WTU and WTP were analyzed econometrically through a double-hurdle model (Burke, 2009; Cragg, 1971), unless very high WTU levels were observed in which case a Tobit model was used (Demont et al., 2012). Generally, urban consumers in provincial cities closer to the production zones (e.g., Saint-Louis in Senegal and Malanville in Benin) were more likely to upgrade and pay more for rice quality (Demont et al., 2013b; Zossou et al., 2013). WTU and WTP levels tended to be higher in the morning (Demont et al., 2013b, 2013c, 2012; Zossou et al., 2013) and sometimes dropped after tasting. When the decline was uniform over all products, we concluded that the benchmark gained acceptance and value when more information about its sensory attributes was obtained (Demont et al., 2013b, 2013c). Traditionally parboiled rice tended to lose market share after tasting in Cameroon and Burkina Faso (Akoa Etoa et al., in press; Ouedraogo et al., 2013), which is consistent with claims in the literature that traditional parboiling technologies often yield inferior postcooking quality attributes (Behrens et al., 2007). WTU and WTP levels tended to be relatively unaffected by socio-demographic factors, which is consistent with the literature (Lusk and Shogren, 2007), with the exception of a few variables. Younger, wealthier, and more educated women tended to be (i) more likely to upgrade, and (ii) pay higher price premiums for quality rice (Demont et al., 2013b, 2012; Ouedraogo et al., 2013) and sometimes differences between ethnic groups were observed (Demont et al., 2013c, 2012; Zossou et al., 2013). Finally, in Cameroon, we found that urban consumers with higher percapita consumption levels of rice were willing to pay more for quality (Akoa Etoa et al., in press).

4. Conclusion Experimental evidence throughout Africa suggests that urban consumers are willing to upgrade standard rice and pay for rice with superior intrinsic and extrinsic quality attributes resulting from investments in varietal quality improvement, upgrading of processing infrastructure, improved packaging, labeling, and advertising. This provides huge opportunities for value-chain upgrading and connecting poor farmers to urban markets, and is important for policy makers who are currently implementing ambitious NRDS which aim at doubling rice production in subSaharan Africa under CARD. Although varietal quality improvement and upgrading of processing infrastructure lead to similar pay-offs in urban markets, postharvest grain quality needs to be improved before incentives for varietal quality improvement are sufficiently transmitted from consumers to farmers. An enabling environment should be created for private-sector investment in

Please cite this article as: Demont, M., Ndour, M., Upgrading rice value chains: Experimental evidence from 11 African markets. Global Food Security (2014), http://dx.doi.org/10.1016/j.gfs.2014.10.001i

M. Demont, M. Ndour / Global Food Security ∎ (∎∎∎∎) ∎∎∎–∎∎∎

upgrading traditional and substandard processing infrastructure (improved parboiling technologies, dryers, mills, sorters, cleaners, graders, etc.), while more public resources should be devoted to breeding programs that focus on varietal quality improvement. Since the introduction of fragrant rice varieties in the Senegal River valley, for example, Senegalese rice farmers’ access to urban markets has increased dramatically, where local rice was usually sold at a discount owing to inferior postharvest grain quality and the absence of fragrance. Extrinsic quality attributes like labeling and information also play a powerful role in improving the reputation of domestic rice in urban markets, building value, and reducing quality uncertainty. However, once the product is purchased it is important that it corresponds to urban consumers’ expectations in order to structurally transform medium- and long-term consumption patterns. Double-digit annual growth rates in rice consumption are currently being recorded in several African countries (Rutsaert et al., 2013), and the positive relationship between per-capita rice consumption and quality premiums (Akoa Etoa et al., in press) suggests that the priority in these countries will be to upgrade rice quality such that domestic rice value chains can tap into these rapidly growing market opportunities. Acknowledgements We thank Justin Fagnombo Djagba and Sander Zwart (Africa Rice) for generating the map in Fig. 1. Financing from the European Union, Government of Japan, Canadian International Development Agency (CIDA), Syngenta Foundation for Sustainable Agriculture, United States Agency for International Development (USAID), and Global Rice Science Partnership (GRiSP) is gratefully acknowledged. These donors had no involvement in study design; in the collection, analysis, and interpretation of data; in the writing of the report; or in the decision to submit the article for publication. We would like to thank the editor and three anonymous reviewers for their constructive comments that greatly improved this article. The authors attest originality of the results and assume full responsibility for any remaining shortcomings. References Akerlof, G.A., 1970. The market for “lemons”: quality uncertainty and the market mechanism. Q. J. Econ. 84, 488–500 〈http://www.jstor.org/stable/1879431〉. Akoa Etoa, J.M., Ndindeng, S.A., Owusu, E.S., Woin, N., Bindzi, B., Demont, M., in press. Consumer valuation of an improved rice parboiling technology: experimental evidence from Cameroon. Q. J. Int. Agric. Allsop, D.T., Bassett, B.R., Hoskins, J.A., 2007. Word-of-mouth research: principles and applications. J. Advert. Res. 47, 398–411 〈http://versatilesolutionsbb.com/ web_documents/word_of_mouth_research_jar_dec-07.pdf〉. Barrett, C.B., 2008. Smallholder market participation: concepts and evidence from eastern and southern Africa. Food Policy 33, 299–317. http://dx.doi.org/10.1016/ j.foodpol.2007.10.005. Becker, G.M., Degroot, M.H., Marschak, J., 1964. Measuring utility by a singleresponse sequential method. Behav. Sci. 9, 226–232. http://dx.doi.org/10.1002/ bs.3830090304. Behrens, J.H., Heinemann, R.J., Lanfer-Marquez, U.M., 2007. Parboiled rice: a study about attitude, consumer liking and consumption in São Paulo. Brazil. J. Sci. Food Agric. 87, 992–999. Burke, W.J., 2009. Fitting and interpreting Cragg’s tobit alternative using Stata. Stata J. 9, 584–592. CARD, 2014. (〈http://www.riceforafrica.org〉) (accessed October 6, 2014). Costello, C., Demont, M., Ndour, M., 2013. Marketing local rice to African consumers. Rural 21: Int. J. Rural Dev. 47, 32–34 〈http://www.rural21.com/english/ a-closer-look-at/detail/article/marketing-local-rice-to-african-consu mers-0000657/〉. Cragg, J.G., 1971. Some statistical models for limited dependent variables with application to demand for durable goods. Econometrica 39, 829–844. Demont, M., 2013. Reversing urban bias in African rice markets: a review of 19 National Rice Development Strategies. Global Food Secur. 2, 172–181. http://dx. doi.org/10.1016/j.gfs.2013.07.001.

7

Demont, M., Rizzotto, A.C., 2012. Policy sequencing and the development of rice value chains in Senegal. Dev. Policy Rev. 30, 451–472. http://dx.doi.org/10.1111/ j.1467-7679.2012.00584.x. Demont, M., Ndour, M., Zossou, E., 2013a. Can local African rice be competitive? An analysis of quality-based competitiveness through experimental auctions. Cah. Agric. 22, 345–352. http://dx.doi.org/10.1684/agr.2013.0664. Demont, M., Rutsaert, P., Ndour, M., Verbeke, W., 2013b. Reversing urban bias in African rice markets: evidence from Senegal. World Dev. 45, 63–74. http://dx. doi.org/10.1016/j.worlddev.2012.11.011. Demont, M., Rutsaert, P., Ndour, M., Verbeke, W., Seck, P.A., Tollens, E., 2013c. Experimental auctions, collective induction and choice shift: willingness-topay for rice quality in Senegal. Eur. Rev. Agric. Econ. 40, 261–286. http://dx.doi. org/10.1093/erae/jbs021. Demont, M., Zossou, E., Rutsaert, P., Ndour, M., Van Mele, P., Verbeke, W., 2012. Consumer valuation of improved rice parboiling technologies in Benin. Food Qual. Preference 23, 63–70. http://dx.doi.org/10.1016/j.foodqual.2011.07.005. EUCORD, 2012. Rice Sector Development in East Africa. Common Fund for Commodities (CFC), Amsterdam. FAO, 2012. The State of Food and Agriculture: Investing in Agriculture for a Better Future. Food and Agriculture Organization of the United Nations (FAO), Rome, Italy. FAO, 2014a. FAOSTAT Agriculture Data. Available from: 〈http://faostat3.fao.org〉. (accessed 21.8.14). FAO, 2014b. Developing Sustainable Food Value Chains: Guiding Principles. Food and Agriculture Organization of the United Nations (FAO), Rome. Fofana, M., Wanvoeke, J., Manful, J.T., Futakuchi, K., Van Mele, P., Zossou, E., Bléoussi, T.M.R., 2011. Effect of improved parboiling methods on the physical and cooked grain characteristics of rice varieties in Benin. Int. Food Res. J. 18, 697–703. Haines, P.S., Guilkey, D.K., Popkin, B.M., 1988. Modeling food consumption decisions as a two-step process. Am. J. Agric. Econ. 70, 543–552 〈http://www.jstor.org/ stable/1241492〉. Harrison, G.W., List, J.A., 2004. Field experiments. J. Econ. Lit. 42, 1009–1055. http: //dx.doi.org/10.1257/0022051043004577. Hogan, H.E., Lemon, K.N., Libai, B., 2004. Quantifying the ripple: word-of-mouth and advertising effectiveness. J. Advert. Res. 44, 271–280. http://dx.doi.org/ 10.1017/S0021849904040243. Kijima, Y., Otsuka, K., Sserunkuuma, D., 2011. An inquiry into constraints on a green revolution in sub-Saharan Africa: the case of NERICA rice in Uganda. World Dev. 39, 77–86. http://dx.doi.org/10.1016/j.worlddev.2010.06.010. Laroche Dupraz, C., Postolle, A., 2013. Food sovereignty and agricultural trade policy commitments: how much leeway do West African nations have? Food Policy 38, 115–125. http://dx.doi.org/10.1016/j.foodpol.2012.11.005. Lusk, J.L., Bruce Traill, W., House, L.O., Valli, C., Jaeger, S.R., Moore, M., Morrow, B., 2006. Comparative advantage in demand: experimental evidence of preferences for genetically modified food in the United States and European Union. J. Agric. Econ. 57, 1–21. http://dx.doi.org/10.1111/j.1477-9552.2006.00029.x. Lusk, J.L., Fox, J.A., Schroeder, T.C., Mintert, J., Koohmaraie, M., 2001. In-store valuation of steak tenderness. Am. J. Agric. Econ. 83, 539–550 〈http://www. jstor.org/stable/1245085〉. Lusk, J.L., Shogren, J.F., 2007. Experimental Auctions: Methods and Applications in Economic and Marketing Research. Cambridge University Press, Cambridge. Ouedraogo, M., Demont, M., Ndour, M., Kaboré, B.K., Dakkouo, D., 2013. Evaluation du consentement à payer pour le riz local de qualité au Burkina Faso: une analyse par la méthode des enchères expérimentales. In: Proceedings of the Third Africa Rice Congress, Yaoundé, Cameroon, 21–24 October 2013. Rutsaert, P., Demont, M., Verbeke, W., 2013. Consumer preferences for rice in Africa. In: Wopereis, M.C.S., Johnson, D., Ahmadi, N., Tollens, E., Jalloh, A. (Eds.), Realizing Africa’s Rice Promise. CAB International, Wallingford, UK, pp. 293–301. Tollens, E., Demont, M., Sié, M., Diagne, A., Saito, K., Wopereis, M.C.S., 2013. From WARDA to AfricaRice: an overview of rice research for development activities conducted in partnership in Africa. In: Wopereis, M.C.S., Johnson, D.E., Ahmadi, N., Tollens, E., Jalloh, A. (Eds.), Realizing Africa’s Rice Promise. CAB International, Wallingford, UK, pp. 1–23. UNCTAD, 2009. World Investment Report. United Nations, New York and Geneva. USDA, 2012. 2012 West Africa: Rice annual (GAIN Report). USDA Foreign Agricultural Service. Verlegh, P.W.J., Steenkamp, J.-B.E.M., 1999. A review and meta-analysis of countryof-origin research. J. Econ. Psychol. 20, 521–546. http://dx.doi.org/10.1016/ S0167-4870(99)00023-9. Vickrey, W., 1961. Counterspeculation, auctions, and competitive sealed tenders. J. Finance 16, 8–37. http://dx.doi.org/10.1111/j.1540-6261.1961.tb02789.x. World Bank, 2008. World Development Report 2008. The World Bank, Washington, DC. (http://siteresources.worldbank.org/INTWDRS/Resources/477365-1327599046334/ WDR_00_book.pdf). Zossou, E., Demont, M., Van Mele, P., Owusu, E.S., Lebailly, P., 2013. Assessing the impact of video and rural radio on consumer valuation of local improved parboiled rice. In: Proceedings of the European Association of Agricultural Economists 138th Seminar on “Pro-poor Innovations in Food Supply Chains,” Ghent, Belgium, 9–11 September 2013. Zossou, E., Van Mele, P., Vodouhe, S.D., Wanvoeke, J., 2009. The power of video to trigger innovation: rice processing in central Benin. Int. J. Agric. Sustainability 7, 119–129. http://dx.doi.org/10.3763/ijas.2009.0438.

Please cite this article as: Demont, M., Ndour, M., Upgrading rice value chains: Experimental evidence from 11 African markets. Global Food Security (2014), http://dx.doi.org/10.1016/j.gfs.2014.10.001i