15 lessons for humanitarians responding to the Somalia crisis

The humanitarian crisis in Somalia continues to deteriorate as five consecutive failed rainy seasons  have produced the worst drought in 40 years (UNICEF, 2022). With a likely sixth season of below-average rainfall from March to June 2023, over 8 million people are projected to be food insecure with 727,000 people facing catastrophic hunger levels by June 2023 (OCHA, 2022a; OCHA, 2023). As a result, about 1.8 million children under the age of five could be acutely malnourished and over 500,000 severely malnourished between August and July 2023 (OCHA, 2022b).

Drought is also causing severe water shortages leading to outbreaks of diseases such as cholera and acute watery diarrhoea. Since January 2022, Somalia has recorded 13,383 suspected cases of cholera with 73 associated deaths (UNICEF, 2022).

Moreover, conflict and insecurity continue to intensify vulnerabilities and exacerbate displacement. As people search for food, water, pasture and other services, drought displacements have increased three times since January 2022 with many at risk of eviction and secondary displacement (OCHA, 2022b). The prolonged and intense drought conditions have resulted in excess cumulative fatalities and malnutrition, and excess mortality will continue to build unless humanitarian assistance is scaled up and sustained in critical sectors (OCHA, 2023).

This synthesis of lessons learned builds on evaluations and learnings from previous drought responses in Somalia to provide essential lessons to inform the ongoing planning, programming and activities for the current Somalia drought response. Interviews with frontline staff discussing the key learning challenges faced in the response were used to identify the themes addressed here. This synthesis aims to provide evidence-based lessons for humanitarian efforts to minimise drought impacts and lay a base for resilient recovery.

Applying these lessons to ensure that humanitarians learn effectively from previous responses requires that the right financing is in place. Many of the lessons speak to strengthening the quality of humanitarian response, which can help to end severe need more quickly and support sustainable pathways to recovery.

At the time of publication, the UN-coordinated appeal for Somalia is only 23.5% funded. As seen in wider ALNAP research, cuts in funding force agencies to make crucial trade-offs in quality, while also placing strain on social cohesion between those who receive support and those who are cut from distribution lists. In short, funding shortages have significant programmatic impacts that make it difficult to apply lessons learned and can affect the overall quality and effectiveness of the response.

The global community and humanitarian system still have the opportunity to avert a wider catastrophe in Somalia but only if adequate funding is provided to heed previous lessons and avoid repeating previous mistakes.

Deliver an integrated emergency response

Studies have shown the ‘food bias’ in previous drought responses – which neglected essential services such as water, sanitation, protection and health sectors – had devastating consequences (Devereux et al., 2017: 18). During droughts, disease, not malnutrition, is the primary cause of death. These deaths are largely the result of water-borne and communicable diseases to which drought-affected populations are exposed (Devereux et al., 2017: 15).

Lesson 1: Response planning should continue to include other sectors alongside food security and nutrition

WASH

• Responses should prioritise water, sanitation and hygiene services. Supply clean drinking water (through deepening existing wells, digging more boreholes or using water tankers), provide safe sanitation facilities (such as toilets/latrines and waste disposal sites) and promote public health (CARE et al., 2016: 36; Devereux et al., 2017: 16). As a part of hygiene interventions consider delivering menstrual hygiene management (MHM) information, MHM-sensitive latrines and hand-washing facilities, and expanding the coverage and frequency of dignity kit distribution (Ndhlovu et al., 2018: 69).
• Water and hygiene education needs to be accompanied by the provision of basic sanitation facilities to harness the full benefits of public health promotion. These may include a pit latrine with a slab, a ventilated improved pit latrine, a flush toilet, a pour-flush toilet or a composting toilet (IFRC, 2021).
• Boreholes are more effective for water service provision than shallow wells. Boreholes can effectively reduce incidences of waterborne diseases, while shallow wells and unprotected water sources can increase them (IFRC, 2021).

Health

From Bangladesh in 1974 to Somalia today, acute watery diarrhoea and cholera have been significant causes of death in droughts and famines (Devereux et al., 2017: 18).  Since January 2022, there have been at least 3,720 suspected cases of cholera and 6,216 confirmed cases of measles in drought-affected regions across Somalia. Acutely malnourished children are being admitted to stabilisation facilities at a higher rate, according to health partners, who attribute this to the rise in suspected cases of cholera and AWD (OCHA, 2022a). In such circumstances, having plans for responding to AWD can save lives. Studies show that the best course of action is a combination of fast action teams and epidemic disease surveillance (Devereux et al., 2017: 18). Measures include:

• vaccinating children against communicable illnesses, especially measles (Devereux et al., 2017: 16); and
• the use of ready-to-use therapeutic meals (RUTF) and the establishment of community-based management of malnutrition. These are effective practices in the treatment of acute malnutrition. To safeguard therapeutic rations, it is important to have a blanket supplementary ration in place. Having a steady supply of RUTF is crucial for emergency preparation (Devereux et al., 2017: 17).

Protection

Drought also results in significant protection risks such as conflicts over water and pastures, gender-based violence and displacements (UNHCR, 2022a). Protection analysis continues to reveal that internal displacement in Somalia is a primary source of humanitarian and protection needs. Currently, in Somalia, protection remains vital for the most vulnerable groups, particularly women, children, people with disabilities, the elderly and members of minority clans. Humanitarian actors can do more to protect vulnerable populations (UNHCR, 2022b).

• Mainstream protection in all interventions. Follow humanitarian principles in all targeting processes and consider vulnerable groups, including: minority clans, children, adolescent girls, women, elderly people without support, people with disabilities, unaccompanied and separated children, female- or child-headed households, pregnant and lactating women, and chronically ill individuals (UNHCR, 2022a).
• Ensure that distribution and registration centres are safe, inclusive and close enough to people, and that staff are aware of protection risks, including referral mechanisms (UNHCR, 2022a).
• People with disabilities face increased risks in times of armed conflict and drought. All sectors must devote resources to and gather disaggregated data on sex, age and disability (UNHCR, 2022b).
• Support minority-led organisations and organisations for people with disabilities that are already engaging in and contributing to protection-related programming (UNHCR, 2022b).

Back to top

Lesson 2: When using multi-purpose cash, pay attention to factors in the current Somali context to ensure effectiveness and efficiency

Ensure quality relationships with financial service providers, monitor local markets and adjust transfer values accordingly, and coordinate with other agencies providing multi-purpose cash (MPC).

While global inflation has impacted the cost of many items imported to Somalia, supply chains have remained generally resilient, making the context suitable for a cash-based response. But there are a number of factors that humanitarian actors need to consider when designing and implementing their own MPC intervention.

• While overall supply chains remain strong, some areas of Somalia are subject to localised supply chain disruption and therefore require ongoing monitoring (WFP, 2022).
• Multiple actors are engaged in MPC support in Somalia, alongside development-financed social protection systems. To ensure efficiency and effectiveness, and in particular to ensure consistency in transfer values, agencies should coordinate. To do so they can reach out to the Cash Working Group.

Lesson 3: Consider how marginalisation and vulnerability impact targeting

Very different factors, including marginalisation, make households vulnerable to drought. Targeting should take such complexities into account.

• The targeting of humanitarian response should be informed by a careful analysis of the causes of vulnerabilities and assessments of who is most likely to be vulnerable (Steets et al., 2019; Devereux et al., 2017). For example, to address marginalisation in the 2017 Somalia drought response, DFID targeted assistance by blending generic IPC classifications with the knowledge of ‘social exclusion and historical disempowerment’ (DuBois et al., 2018: 27). Prioritising IPC classifications 3 and 4 can lead to insufficient attention to IPC 2, thus missing the opportunity to prevent ‘large numbers of people from declining and eventually requiring more forms of intervention’ (DuBois et al., 2018: 18).
• Enhance targeting practices that reduce and address levels of vulnerabilities (CARE et al., 2016). Community-based targeting is one way of doing this, because it can ensure a high degree of participation and transparency (Devereux et al., 2017). Community-based targeting during the 2019–2020 Zambian drought response ‘considered gender and inclusion dimensions in prioritising poor female-headed households, pregnant women, lactating females, child-headed households, households with orphans and the elderly’ (IFRC, 2021). In addition, targeting practices can also consider household sizes.
• Mapping minority rights organisations in the operations areas enables the identification of aid diversion and exclusion (Maxwell et al., 2016; UNHCR, 2022a). This practice can solve inclusion and exclusion errors, reinforced local patronage structures and gender bias during community-based targeting (IFRC, 2021).
• Consider elements of gender. Crises, such as drought, offer opportunities to make local systems gender-inclusive, for example through engaging women in planning and decision-making around aid interventions (Mercy Corps, 2022). In addition, staff awareness of gender issues and more human and financial resources allocated to gender mainstreaming are needed to address the inequalities women face (Ndhlovu et al., 2018).
• Prioritisation should be based on the severity of needs rather than ‘what type of shock or emergency people experienced, how recently they experienced it, or what their status is’ (Steets et al., 2019: 76). In addition, consider extending assistance to those who have not yet recovered from previous shocks.

Consider the specific risks and needs faced by internally displaced populations

By June 2022, more than a million people in Somalia, Kenya and Ethiopia had already been evicted from their homes due to the crippling effects of drought (NRC, 2022). The drought emergency has already impacted 6.1 million people, of which 771,400 have been forced to leave their homes in search of pasture, water and food. Children and women make up more than 80% of the people who have been displaced (OCHA, 2022a). People displaced by drought in 2022 were among the more than 3.5 million people who are internally displaced in Somalia. (NRC, 2022). 

Lesson 4: Consider the health risks within IDP camps

Humanitarian actors may do more harm than good if they do not consider the health risks associated with attracting large numbers of people into IDP camps, where IDPs are at an increased risk of communicable and vector-borne diseases such as malaria and leishmaniasis (Cantor et al., 2021: 4; Devereux et al, 2017: 15). Currently, the risk of outbreaks of diseases such as cholera and measles have increased in Somalia, especially in IDP camps, due to widespread water shortages and poor access to sanitation and hygiene facilities (UNSC, 2022). There are several ways to mitigate health outbreaks in camp settings.

• Establish safe sanitation facilities – such as toilets or upgraded latrines – and waste disposal, particularly in IDP camps but also anywhere drought affected communities are gathered in large numbers (Devereux et al, 2017: 16).
• Vaccinate children against communicable diseases, especially measles, which accounted for a large portion of child deaths during the famines in the Darfur and Somali regions (Devereux et al, 2017: 16).
• IDP health interventions such as water treatment have been proven to reduce diarrhoea incidence by about 90% and prevalence by over 80% (Cantor et al., 2021: 4).
• For malaria prevention within IDP camps, studies have found that the most cost-effective malaria control strategy is symptom-based screening though mass drug administration, mass screening and treatment. Programmes to distribute insecticide-treated nets are also crucial in preventing malaria in IDP settings (Cantor et al., 2021: 4).
• Across all health interventions, humanitarian actors must engage local community health workers to ensure effectiveness and up-take (Cantor et al., 2021: 4).

Lesson 5: Improve social cohesion among IDPs and host communities

As displacement becomes protracted, social relations between displaced people and host communities can become more strained. The influx of IDPs into urban areas and the increased social interactions between displaced people and host communities as a result of living close together puts a strain on public services that are already stretched thin, leading to rivalry for resources and job opportunities, all of which can cause tension and conflict (Holloway and Sturridge, 2022: 7). There are several ways to promote social cohesion.

• Extend targeting practices to include both IDPs and host communities. When humanitarian aid is perceived to be targeted only at IDPs it can have unintended consequences within the host community (Commins et al., 2022). There are several ways to extend targeting practices so that the needs of all communities are met.
  • Dispel myths about displaced people and explain why helping displaced families is morally beneficial as well as socioeconomically profitable (i.e., that displaced households can bring benefits; Commins et al., 2022).
  • When more programming is offered for host towns in response to a population influx, use programme messaging and content to promote good attitudes toward IDPs (Commins et al., 2022).
  • To assist host communities impacted by the surge of displaced people, identify areas where greater resources are being targeted at host communities (Commins et al., 2022).
• Implement interventions to foster social cohesion. Interventions to improve social cohesion can be divided into three categories.
  1. Direct-contact interventions. The goal of direct-contact interventions is to increase social cohesion by facilitating face-to-face interactions between groups of individuals. This could happen through projects that require host and displaced communities to use the same facilities, or through programmes that include both groups as participants at the same time (Holloway and Sturridge, 2022: 18).
  2. No-contact interventions. These are initiatives and programmes that foster better relationships without putting the two groups in direct contact. This can be achieved by raising awareness of other groups or by addressing the factors that have in the past contributed to social tension between groups (Holloway and Sturridge, 2022: 18).
  3. Advocacy initiatives. Advocacy efforts enhance social cohesiveness by influencing public opinion about IDPs through the media or promoting legislative change to promote more friendly conditions (Holloway and Sturridge, 2022: 18).

Lesson 6: Support IDPs against secondary displacement and tenure insecurity

By June 2022, the internally displaced population of Somalia were distributed over more than 3,400 camps and informal settlements. Many of these lacked security of tenure, putting these IDPs at a significant risk of eviction, re-displacements and conflict over resources (NRC, 2022). Most of the major IDP camps in Somalia are privately owned, which increases the possibility of forcible evictions. About 80,000 forcible evictions by landowners were documented by protection cluster partners between January and June 2022, with 84% taking place in the Banaadir region (UNSC, 2022). Evictions have emerged as one of the primary catalysts for secondary displacement in cities (IDMC, 2021). Driven by a lack of adequate housing and informal tenure agreements in increasingly crowded areas, there has been an increased demand for land which has seen a corresponding increase in forcible evictions, including the destruction of community and humanitarian assets to an estimated value of US$2,100,000 (NRC, 2022).

Humanitarians should collaborate with developmental and governmental actors to protect IDPs from forced evictions and secondary displacement.

Best practice

Examples of projects that have supported IDPs with forced evictions and shelter include the Puntland Shelter Project, a collaboration between World Vision and the Ministry of the Interior of Puntland in which the Puntland government secured land from private owners and transferred individuals title deeds to IDPs after housing had been constructed by World Vision. The project's multi-sectoral architecture made sure that a community with access roads, access to healthcare and a water supply was developed around the housing programmes. Ultimately, the IDPs who were relocated were integrated into the town of Garowe (IGAD, 2019: 23–24). In another project, NRC and UN-Habitat collaborated with Benadir Regional Administration to support IDPs to address short-term cash requirements, medium-term housing needs and long-term needs for economic inclusion by focusing on increasing the security of tenure for rental homes and providing a variety of support measures, such as rental subsidies, WASH improvements and employment (IGAD, 2019: 24).

Consider the ways in which humanitarian access complicates drought response

Much like the previous droughts in Somalia, al-Shabaab is still a danger to humanitarian efforts and lack of access remains one of the most difficult challenges in drought-affected conflict zones. The group has vacillated in providing humanitarian help due to changing interests and goals, making it difficult for aid organisations and donors to properly negotiate principled access (Kurtzer et al., 2022: 3).

Lesson 7: Address access challenges to prevent the displacement of already marginalised populations

The lack of access to communities in the 2017 response resulted in an approach based on aid hubs, which frequently forced marginalised groups to relocate in search of assistance, marginalising them further. Given the increasingly urbanising Somali population and the de-pastoralisation of Somali society, this is especially concerning (DuBois et al., 2018: 25–27). With a few notable exceptions, aid organisations do not appear to routinely negotiate access with armed groups; some even deliberately avoid doing so. As a result, access becomes dependent on ‘informal’ networks or subcontracting chains. However, studies show that the humanitarian sector in south-central Somalia has a poor understanding of clan and local dynamics, which further impacts access (DuBois et al., 2018: 17–27). Emergency responders can overcome challenges around access using the following approaches.

• Develop institutional legal and operational skills to help staff members as they negotiate the challenges of operating in contexts where armed groups are active. This on-call expertise might be organised in a variety of ways, including the establishment of a counterterrorism focal point within existing structures, the formation of a specialised expert post, or a prospective collaboration with another organisation (IFRC 2021: 4).
• Build the capacity required to reach Somalia’s most vulnerable by investing in appropriate resources to access hard-to-reach communities. This includes the ability to apply humanitarian principles to negotiate access location by location and the possession of a thorough contextual awareness of the politics and social factors that may result in marginalisation and exclusion (DuBois et al., 2018: 27).

Lesson 8: Invest in remote management

Remote management has been one of the most popular ways to manage access issues and reduce risks to international staff (Donini and Maxwell, 2014). It involves the withdrawal or reduction of international staff from the field, thereby shifting responsibility to local organisations and managing programme activities from a different location (Stoddard and Harmer, 2010). However, the lack of preparation and investment in remote management has either transferred risk to local partners or hindered programme efficiency (Donini and Maxwell, 2014). There are several ways to improve remote monitoring tools and systems.

• Plan for remote management by investing in remote management programming during the design phase of a regular programme.
• Invest in international staff with backgrounds in remote management or distance management contexts, expertise adjusting to changing access requirements, good communication and distance management abilities, and fluency in the local language (Stoddard and Harmer, 2010).
• By conducting collaborative risk assessments with partner entities, figuring out their respective security management structures, and identifying any additional security requirements, it is possible to avoid risk transfer as a policy priority in remote management. Risk assessments could also be coordinated and carried out at the cluster level to review sector-level coverage and spot any gaps (Stoddard and Harmer, 2010).

Best practice

‘A new generation of remote control and remote monitoring tools and systems arose from the 2011 crisis when presence on the ground was often impossible. The most prominent of these are:

1. talking directly to beneficiaries and other stakeholders by phone and well-established hot lines
2. third-party monitoring, through the contracting of independent monitoring firms staffed by both Kenyan and Somali consultants
3. ensuring that dialogue with partner agencies is in place to triangulate and verify information.

Adopting these approaches required significant investment on the part of aid agencies to ensure that their national staff and Somali partners were able to adapt and use these new mechanisms properly, as they transfer a lot of the responsibility for information reliability and robustness to them. Feedback from project participants was elicited through a “beneficiary hotline”. This cash transfer system also functioned in the al-Shabaab-controlled areas, despite the restrictions they had imposed. Using these different approaches in a complementary manner was the best alternative where direct field monitoring was impossible.’ (Grünewald et al., 2019: 52)

Implement better coordination

Coordination remains one of the most challenging issues when responding to droughts. The Inter-Agency Humanitarian Evaluation of the drought response in Ethiopia found that coordination structures during the 2017 drought response were seen as too onerous, with many duplications between different fora (Steets et al., 2019: 4). For example, coordination across response sectors was addressed in the inter-cluster coordination group, in the Ethiopia Humanitarian Country Team, and in meetings between cluster lead agencies and the Humanitarian Coordinator (Steets et al., 2019: 17). There is also the issue of competition between organisations, which hinders effective inter-cluster coordination. This includes organisations paying higher cash-for-work rates to increase the number of people participating in their projects, and clusters not sharing response information with other clusters (Steets et al., 2019: 18). There are several ways to improve humanitarian coordination.

Lesson 9: Bring coordination closer to the field

Bringing coordination mechanisms closer to the field has been proven to drastically improve coordination. Studies show that the transition from Nairobi-based coordination to more on-the-ground coordination (the Drought Operation Coordination Centre in Mogadishu and coordination hubs in Somali states) was a major turning point in the 2017 response. This change was widely acknowledged to have played a crucial role in organising the response at a more local level, addressing problems such as duplication and improving coverage and access (Grünewald et al., 2019: 54).

Lesson 10: Work respectfully with national and local NGOs

International actors must go beyond simply inviting national NGOs to meetings and ensure that they are involved in high level discussions and decisions. Evaluations show that during the Ethiopia response, through the Consortium of Christian Relief and Development Associations, international NGOs, for instance, promoted improved coordination with national NGOs. The original plan, however, to have a member of the consortium participate in the Ethiopian Humanitarian Country Team did not come to pass. As a result, only two of the national NGOs that were interviewed as part of the evaluation routinely participated in clusters, and many of them did not feel that their organisation was appropriately represented in the coordinating structure (Steets et al., 2019: 19).

Best practice

‘There were significant changes in this respect during the 2016–2017 drought response [in Somalia]. First of all, efforts were made to establish coordination that was closer to the ground. The establishment of the Drought Operation Coordination Centre (DOCC) in Mogadishu brought together actors based not only in Nairobi but also on the ground, and significantly enhanced dialogue between actors and between Mogadishu and Nairobi. Indeed, thanks to excellent video-conference facilities, it improved their ability to jointly address collective problems and to facilitate complex decision-making processes. Similarly, the decentralised DOCCs in Baidoa and Galkayo also provided opportunities for synergy.’ (Grünewald et al., 2019: 51)

Build recovery and resilience into the response

Repeated drought crises in the Horn of Africa have constantly produced the same lessons: that short-term emergency response should be replaced with approaches that build longer-term resilience to drought and reduce the need for humanitarian assistance. While much of this rests on medium- to longer-term engagement from development actors, there are also ways that humanitarian responders can contribute to, rather than undermine, longer-term resilience during a drought response.

Lesson 11: Adapt livelihoods support to the changing climate

• Increase climate-smart agricultural assistance and investments to support farming communities’ resilience. An example is the 2019–2020 Zambia drought response, where farmers received seeds and farm inputs for the following growing season. These were complemented by drilling boreholes for safe drinking water and irrigation to improve the livelihoods of the affected communities (IFRC, 2021).
• Support drought-affected pastoralists to develop alternative livelihoods. For those who remain pastoralists, strengthen support systems such as herd diversification, livestock insurance programmes, improved market access and value addition to livestock products (Steets et al., 2019).

Lesson 12: Invest in sustainable water supply and management systems

• Increase investment and interventions around sustainable water supply systems. These measures can be delivered while providing emergency water supplies through temporary measures such as water tankers, which are essential for saving lives. Sustainable water supply measures include drilling and rehabilitating boreholes, constructing rainwater harvesting systems, installing new taps and expanding the geographical coverage of interventions (Ndhlovu et al., 2018).
• Invest in renewable energy solutions such as solar water pumps to ensure the sustainability of water supplies (Ndhlovu et al., 2018). Other measures, such as training water-point managers and operators and ensuring consistent maintenance, are also essential for sustainability (IFRC, 2021).

Best practice

NRC-led Building Resilient Communities in Somalia (BRCiS) installed 127 piped water schemes (solar-powered boreholes) in communities, 80% of which were operational by March 2022, while the remaining 13% (which were mainly surface-water catchments) ran dry (NRC, 2022). Consequently, communities were better able to manage drought.

Lesson 13: Work through and with communities’ own networks

The ‘formal’ aid system needs to recognise, learn and work with ‘informal’ systems. Affected communities are often the first responders during humanitarian crises through informal social protection (support shared within people’s own social networks) for coping, response and recovery (Maxwell and Majid, 2014; Mercy Corps, 2022). Informal social support includes cash and loans, food, water, psychosocial support, information and advice (Mercy Corps, 2022). Humanitarian actors can support informal systems in several ways.

• Responders need to strengthen these social systems as part of emergency response. Social networks can be engaged in collecting and disseminating up-to-date information about the evolving humanitarian crisis (Mercy Corps, 2022). Communities are likely to trust and adhere to information after hearing it multiple times and through various channels.
• Invest in local private or community-led initiatives. For example, in supporting informal initiatives, responders can follow survivor and community-led response practices, which can take the forms of micro-grants, demand-led skill training and mobilisation (Mercy Corps, 2022).
• Improve the capacities of local communities, governments, businesses and organisations, for example, through strengthening local early warning systems and building community contingency funds (Maxwell et al., 2016).

Lesson 14: Work through and with formal social protection systems

Social protection systems and programmes have effectively reduced food insecurity, hunger and poverty and enabled vulnerable groups and individuals to cope with crises (Mercy Corps, 2022). However, during an emergency response, there are high risks of bypassing the formal social protection systems that people rely on when coping with and recovering from shocks, creating a new dimension of vulnerability (Mercy Corps, 2022). Moreover, there is also the issue of aid agencies focusing on expanding social protection systems rather than sustaining support to existing caseloads whose vulnerability, poverty and food security has intensified (Slater et al., 2021: 9).

• Support and invest in the existing social protection programmes. For example, the the Shock Responsive Safety Net for Human Capital Project (SNHCP) – Baxnaano (uplifting Somalia), is still in its early stages. Supporting it can cushion communities against food insecurity and enhance their resilience (Halakhe, 2022).
• Responders should focus on maintaining social protection support for vulnerable households in addition to extending support.

Lesson 15: Support local markets

Local market systems are crucial for supplying essential items and buying farmers’ produce during drought emergencies. However, the direct provision of commodities to crisis-affected people can undermine the functioning of local markets and thus affect the communities’ abilities to cope without aid and hamper resilience and recovery (Mercy Corps, 2022).

• Support local traders when conditions allow, thereby enabling them to restock their shops with essential items and foodstuffs to meet demand (IFRC, 2021). Support can include grants and market linkages.