ILO, Crozet CASE STUDIES MAKING PUBLIC PRIVATE PARTNERSHIPS WORK IN INSURANCE

Transcription

1 ILO, Crozet CASE STUDIES MAKING PUBLIC PRIVATE PARTNERSHIPS WORK IN INSURANCE March 2015

2 TABLE OF CONTENTS CASE STUDY RSBY IN INDIA... 2 CASE STUDY AGRICULTURAL CATASTROPHIC INSURANCE (SAC) IN PERU CASE STUDY CADENA IN MEXICO CASE STUDY BANCA DE LAS OPORTUNIDADES IN COLOMBIA

3 2

4 CASE STUDY RSBY IN INDIA In 2008, the Ministry of Labour and Employment (MoLE) launched Rashtriya Swasthya Bima Yojana (RSBY), an in-patient health microinsurance scheme. The program is sponsored by the central and state governments and mainly targets the below poverty line (BPL) population, even though occupational groups such as domestic servants and construction workers were added in mid As of 2014, the scheme operates in 479 of the 672 districts in India across 29 states and union territories. As India increases the number of poor individuals covered by the scheme, challenges remain in improving quality of services, avoiding fraud, and guaranteeing the program s sustainability. BACKGROUND India has high per capita out-of-pocket health expenditures and very low public health spending for lowincome citizens. The lack of financing options leads to poor health outcomes and poverty traps. As of 2010, approximately 240 million people, 19 percent of the population, were covered by a publicly funded health-financing scheme with around 25 percent covered by some form of public or private health insurance. The Indian Prime Minister unveiled the RSBY scheme as a major initiative to provide access to health care and financial protection for the BPL population. In the past, the government provided health insurance cover to selected beneficiaries at either the state level or central level, but most of these schemes didn t achieve their objectives due to either design or implementation issues. Based on this history, the government decided to design a health insurance scheme that avoided those earlier mistakes. STRUCTURE: A MULTIPLE-STAKEHOLDER SCHEME PLANNING The design and implementation of the scheme required input from a broad set of stakeholders. The MoLE oversaw the scheme s development and formed a task force in 2007, including representatives from the insurance industry, relevant ministries, and the technology industry, as well as international organizations such as the World Bank and GIZ. This task force evaluated past experience with different health microinsurance (HMI) schemes in India and elsewhere. A number of RSBY s features drew on past experience from public and private insurance schemes. One major question for the RSBY task force was whether the government had the capacity to implement and administer the program in its entirety, or whether it should consider outsourcing certain functions, given the substantial growth of India s private insurance market over the past decade. Since neither private insurers, public insurers, nor the MoLE had the capacity to administer a scheme with 355 million eligible members, the task force decided to design a public-private partnership between the government (central and state levels), insurers, and hospitals (public and private). ADMINISTRATION The central coordinating and policymaking agency for RSBY is the MoLE, particularly the Office of the Director General Labour Welfare. MoLE plays a major role in decisions on scheme structure and implementation and also drafts standard documents, defines operational processes, and monitors implementation. At the state level, the scheme is implemented through the state nodal agency (SNA). The SNA was created to be the main supervisory and implementing agency for the scheme at the state level and is involved in contracting insurance companies through a competitive bidding process in accordance with 3

5 the guidelines issued by MoLE. The insurance companies are responsible for empaneling hospitals and ensuring that their institutional infrastructure meets specific criteria such as requirements for biometric scanners, smart card readers, and software. Insurance companies are also responsible for ensuring the enrolment of beneficiaries. They are also required to establish an RSBY helpline and call centre in each district, as well as providing preenrolment information, education, and communication activities. After enrolment, information primarily becomes the responsibility of the state. Like most new government-sponsored health insurance schemes, RSBY has yet to establish the institutional arrangements to perform functions such as oversight, supervision, transparency, regulation, consumer information and protection, and data management and analysis. RSBY relies heavily on insurance companies as intermediaries to perform these functions. OPERATIONS The SNA selects insurers through a competitive bidding process according to a set of eligibility criteria. The state government transfers its share of the premium to the SNA and informs the central government. Then, the central government electronically transfers the premium to the SNA, which is responsible for paying the insurer after the total number of enrolments is confirmed. The task force recognized that success would rely on effective enrolment processes, as well as incentives for implementers to enrol people. They decided to outsource enrolment to private insurers, who receive premium revenue based on enrolment, as well as to reduce manual, paper-based procedures through the use of smartcards to store client information, thus enabling more efficient management of a high number of policies. Even though central and state government jointly finance the premiums for enrolees, many of the functions of administering RSBY fall to the insurers, which are contracted for one year to incentivize healthy competition. They are responsible for empanelling hospitals, enrolling beneficiaries, and processing claims. Annually, an electronic list of eligible BPL households is provided to insurers by each participating state. The insurance company, with the help of district officials, prepares an enrolment schedule for each village with four months to enrol qualifying families in each district. To communicate and market the RSBY scheme and enrolment campaigns, insurance companies are required to provide grassroots outreach prior to enrolment. In addition, the BPL list is posted in each village at the enrolment station and other prominent places before the enrolment campaigns take place. The date and location of enrolment campaigns are also publicized beforehand. Mobile enrolment stations are established at local centres such as public schools at each village at least once a year. Insurers must equip enrolment stations with the technology needed to collect the fingerprints and photographs of household members to be covered, as well as a printer for the smartcards. After paying the registration fee, all enrolees receive the card along with an information packet describing benefits, hospitals in the network, and other relevant information. The process normally takes less than 10 minutes. At the end of enrolment, the insurer sends a list of enrolled households to the state nodal agency. The list of enrolled households is maintained centrally and the insurer is paid once data provided from the insurer and government records is matched. This enrolment process imposes substantial logistic and administrative requirements, but given the massive scale of the operation, the costs are relatively low, at about US$3 per family or less. These costs 4

6 are borne by the insurer within its quoted premium (which averages about US$12 per family). Insurers must also set-up a district office in each of the covered districts to coordinate activities, manage a district kiosk (for modifications to the smart cards), and operate a 24-hour call centre with toll free help line to handle queries related to benefits and operations of the scheme, including information on providers, individual account balances, and interaction with other stakeholders. According to the standardized process laid down by RSBY, once the beneficiary reaches the empaneled hospital, a designated RSBY helpdesk manned by hospital staff is the first point of contact. The helpdesk registers and validates the beneficiary s details, using the information encoded on the smart card and the patient s thumbprint. If hospitalization is required, the staff checks the treatment required and the availability of funds (within the Rs. 30,000 yearly cap for the family). If the required service is not listed in the predefined packages, a prior authorization is required, and the hospital helpdesk coordinates with the insurer. The hospital then proceeds to block the required amount on the smart card. On completion of the treatment, the amount is confirmed and deducted from the insured amount stored on the chip. The hospital also sends an electronic report to the insurer or TPA for reimbursement of the claim. The government maintains responsibility for regulation, information systems, and ultimate oversight. One of the key strengths of RSBY is its real-time monitoring and evaluation system enabled by the program s technology. The scheme s management information system was built centrally for all state led RSBY schemes to enable the collection of standardized information on all daily transactions at hospitals. Data from all districts flow to the central government at periodic intervals. A separate set of pre formatted tables are generated for the insurer, the state, and the central government. The collection of this information allows the insurer to track claims, transfer funds to the hospitals, and investigate suspicious claims through on site audits. The government is able to monitor utilization of the program by members and to some extent, begin to measure the impact of the program. The central government posts much of this data on the RSBY website. Enrolment, hospitalization, and claims data are analysed at the national level regularly and shared with state nodal agencies. In addition, the government of India is planning to produce periodic reports on selected indicators to be shared with all the stakeholders in the near future. This table summarizes the roles and responsibilities of different stakeholders in RSBY program: Role of central government Role of state government/state Nodal Agency Role of insurers/ TPA Oversight of the scheme X X Financial management and X X planning Definition of benefits package X X Hardware specifications (e.g. X systems, smartcard, etc.) Selection of insurers X Empanelment of hospitals, X based on RSBY guidelines Premium payment X X Outreach and marketing to X X beneficiaries Enrolment and collection of fee X X Processing claims X Monitoring state-level utilization X X X Monitoring national RSBY X 5

7 information Customer service X The program is designed to bring synergies among partners. The insurers, a majority of which are commercial, for-profit companies, benefit from access to new market segments, with premiums guaranteed by the government. The government is able to leverage the capacity of the insurers and expand enrolment into RSBY faster than it would otherwise. Enrolees benefit from access to subsidized benefits, made easier through new technologies. The mutual benefits of linking HMI to UHC initiatives Advantage to the government Advantage to private insurers HMI informed the design of RSBY and enabled Linking with the public initiative has enabled client centred design. private insurers to develop a new market segment and secure their revenue thanks to access public subsidies. HMI provided data enabling initial pricing of the scheme. Outsourcing key insurance functions to capable insurers enabled fast development of the scheme, proactive adjustment and promotion of innovation. Outsourcing enrolment to insurers has accelerated enrolment. Transferring the financial risk for claims to insurers secured the government s financial obligation to cover BPL population. Competitive bidding to provide insurance services can enable contain cost and improve quality. PRODUCT DESCRIPTION Engagement of state government has enabled better acceptance of the scheme by the population, and supported implementation and scaling up. Including local authorities at the enrolment stage has improved efficiency and reduced fraud, as insurers have access to the list of eligible members and a local government representative validates the identity of enrolees. Partnering with the government on the RSBY initiative has enabled private sector actors to develop IT solutions to target the low-income segment. This can also benefit other business segments. The RSBY package offers benefits up to a maximum level and focuses primarily on inpatient coverage. It provides cover for hospitalization expenses up to Rs. 30,000 for a family of five. Transportation charges are also covered up to a maximum of Rs. 1,000 with Rs. 100 per visit. Expenditures incurred by the beneficiary on tests or medicines prior to hospitalization are also covered for reimbursement. The hospital will also provide medicines and other assistance for up to five days after a hospital stay. Outpatient care, however, is not covered in RSBY, so beneficiaries must pay for any medicines and tests not related to hospitalization. Most forms of hospitalization are covered with fixed close-ended prices as specified in a list of over 700 defined service packages. These package rates are in the form of a national guidance list that state governments can adopt across 18 categories including: dental, ear, nose, and throat, obstetrics and gynaecology, endoscopic, hysteroscopy, neurosurgery, ophthalmology, orthopaedic, paediatric, endocrinology, urology, oncology, and general surgery. The RSBY does not cover congenital external diseases, drug and alcohol induced illness, sterilization and family planning related procedures, or vaccination. Conditions resulting from war and attempted suicide are also not covered. Treatments using alternative systems of medicine are excluded. Other than these, however, all forms of hospitalization are covered. Pre-existing conditions are covered as well as maternity. 6

8 The latter includes automatic coverage for a newborn until the next renewal cycle (even if five members are already covered for that period). The premium for RSBY is different dependent on the district. The central and state governments jointly finance the full premium at 75 percent and 25 percent respectively. In the case of Jammu, Kashmir and northeast states, the central government pays 90 percent, while the state government pays the remaining 10 percent. According to RSBY guidelines, the cap is estimated at up to Rs. 750 per family. So far, however, the health insurance market has been competitive and, in order to gain volume and profits, bidding by the insurers has resulted in lower quotes for premiums, such that the cap has so far not been breached. Recently, new target groups of beneficiaries were introduced by the RSBY, including beneficiaries of the National Rural Employment Guarantee Scheme (NREGA), building construction workers, contractual postmen, railway porters and hawkers, and domestic workers. MoLE has also announced that RSBY will be open to any group of individuals willing to pay the full premium out of pocket. However, MoLE will vet the composition of the group before such groups can be covered, to avoid adverse selection. These new groups will be covered under clearly identifiable, unique IDs (for monitoring their utilization patterns) and will have a different financing structure from the one the BPL group currently follows. This will, however, also make the scheme more complex. The RSBY smartcard enables cashless transactions at empanelled hospitals, as well as paperless billing and payment between the insurer and health-care providers. It also supports portability of benefits across the country. This technology is one of the major success factors of the RSBY program, since it enables the management of a scheme with more than 100 million members. Besides, it helps providing in-patient insurance for migrating individuals as well as for illiterate ones. The RSBY continues to improve and it is using HMI to test new operational models and mechanisms for expansion. Research shows that 67 percent of out-of-pocket expenditure by clients is for outpatient services (especially drugs), meaning that the RSBY benefits package still has a limited impact on overall household health spending. To address this challenge, RSBY is experimenting with more comprehensive benefits, including outpatient services. In partnership with the MoLE, ICICI Foundation, ICICI Lombard, and the ILO s Microinsurance Innovation Facility, two pilot programs were started. Enrolment and eligibility follow the existing RSBY process, but clients have access to up to ten outpatient visits in empanelled outpatient clinics (public and private) per year, per family. Premiums for the outpatient benefits in these two districts during the pilot are being subsidized by the ICICI Foundation. Consultations (and follow up within 7 days), as well as most drugs are covered. Additional pilots are beginning to test other outpatient benefit packages, such as including diagnostics, as well as other provider payment mechanisms. Based on early but encouraging lessons from the pilots, outpatient benefits will be extended to other districts in the future, with funding from the government. This transition will occur as new districts begin to implement RSBY, or as partnerships with insurers to offer RSBY in existing districts are renewed. Additional pilots are planned to test further variations of outpatient benefits, including coverage of diagnostic services. These will also test different provider payment mechanisms. RESULTS In India, the partnerships under RSBY have helped to improve health care access for the BPL population, manage a high volume of policies, and enable a faster rollout. As of January 2013, 54 percent of eligible families are enrolled, although this data does not reflect re-enrolment rates, so the number of actively enrolled families may be lower. Also, an estimated three members per family have enrolled in the scheme. These results are encouraging, but they also highlight the fact that despite an almost fully subsidized 7

9 program, barriers remain to enrolling the informally employed population in the scheme. The variation of penetration rates across districts (from 10 percent in Delhi to 84 percent in Kerala) also raises the issue of equity in access to the scheme. To sustain the positive trend, the scheme will have to achieve a breakthrough in uncovered states and districts. RSBY also needs to expand outreach deeper into partially covered districts and achieve enrolment from all missing and remote villages. States have ownership of the scheme and have created a community to share learning. A total of 18 insurers (4 public, 14 private) are partnering in RSBY. Average enrolment has increased along with the average expense ratio, which increased from 77 percent in the first year of the scheme, to 143 percent for the 47 districts with two full years of experience in mid Clients are satisfied, as they have been given the possibility to access private health facilities where they may perceive higher quality of care. The hospitalization ratio the percentage of enrolled persons who have been hospitalized is 2.55% under the RSBY as compared to the national average of 1.70% for the poorest 40% in India. However, the interstate variation was significant, ranging from a low of 0.08 percent in Chandigarh (and a similar 0.09 percent in Assam) to a high of 5.21 percent in Kerala (and an equally high 4.33 percent in Gujarat). The reasons for these variations require further study, but given the early stage of implementation, contributing factors would be the correspondingly low beneficiary awareness and the limited accessibility of the hospital network. States such as Kerala and Gujarat, with high awareness levels, extensive hospital networks, and more implementation experience in RSBY, stand out with high utilization compared with states such as Assam and Goa where the scheme is relatively new and the hospital networks are restricted. The claim ratios demonstrate similar patterns. Kerala has a high claim ratio of 100 percent, but Nagaland, with a much lower utilization, displays a claim ratio of 149 percent. This suggests that the insurer was way off the mark in predicting potential claims in the state. Other states with evidence of an adverse claim ratio and where insurers lost (or are losing) money are Delhi and Gujarat. Nonetheless, the average claim ratio is a reasonable 67 percent for this entire subset of districts. Considering that the insurers spend an estimated 20 to 30 percent of their premium on enrolment and other administrative costs, the industry probably managed a small profit. In India, about 74% of the private spending on health is for outpatient care, which RSBY doesn t cover, and only 26% is spent on hospitalization. Therefore, the major portion of the out-of pocket expenditure is still falling to the patient despite their enrolment in the program. It would be prudent to say that the RSBY is a well-designed scheme for major ailments but that it does not manage outpatient care, which, in fact, constitutes a large portion of each family s out-of-pocket spending on health care. KEY LESSONS ON BUILDING EFFICIENT PPPS Governments should consider whether outsourcing key insurance functions is a viable way of expanding health coverage. This requires first finding out if there are reliable organizations in the country with the capacity to provide expertise that government lacks. To build an effective PPP, governments should make sure all parties have an interest in the scheme and are engaged in a participatory manner from project inception. Key stakeholders were consulted during early design discussions and invited to contribute and innovate, resulting in strong support from all parties, and increased interest from the private sector. In order for a partnership model to be successful, it is important that all processes are clearly defined at the beginning, with all roles and responsibilities carefully mapped. The task force that designed RSBY spent time developing a very detailed implementation plan. There was enough flexibility in the scheme to enable innovation and rapid improvements: for example, maternity 8

10 coverage was added after one year to improve client value; the smart cards have become even smarter, with greater capacity to capture more information on clients, and processes to empanel hospitals have been improved. ON CREATING INCENTIVES FOR THE DIFFERENT PARTIES Insurers earn on the number of individuals enrolled, so they have incentives to enrol as many families as possible. The entities outsourced by insurers to conduct enrolment, however, face additional logistics costs and therefore have little incentive to advertise and enrol potential beneficiaries in remote villages or in villages with few BPL families. This can prevent enrolment in certain regions and cause the divergence in conversion rates among states. ON THE ROLE OF TECHNOLOGY Technology has unlocked the potential to effectively service millions of insured members. RSBY s success strongly relies on the development of a relevant technology platform that has been adopted by service providers and promotes efficient enrolment and back office processes with paperless claim management. For the first time, IT applications are being used for social sector scheme on such a large scale. This will ensure a smooth data flow regarding service utilization periodically. However, technology alone cannot do much: good processes are also required. ON THE NEED FOR STRONG CONTROL MECHANISMS The increase in claims ratio and incidence rate in some areas is due to increased utilization of health care, but also possibly to increased fraud. This threatens the financial sustainability of the scheme and makes a case for strong monitoring and controls, as well as for more systematic audits of potential malpractice. With limited claims experience and interest to enter a huge underserved market segment, insurers may have bid low, but as more data becomes available, premiums could increase. ON SCHEME GOVERNANCE So far, RSBY has not been mandated by law, and only involves contracts between state governments and insurers on the one hand, and insurers and hospitals on the other. ON IMPROVING RESULTS In-patient care can solve a lot of problems, but out-of-pocket expenses are also related to outpatient care. Though RSBY offers in-patient care, more benefits are still needed for the coverage to be really appropriated to the needs of low-income households. The difference in hospitalization rates among states demonstrates that more work is needed to increase awareness and understanding of the scheme. CONCLUSIONS AND FUTURE CHALLENGES The RSBY scheme shows how HMI can inform the design of a national scheme and support testing of new approaches and products. It also makes a case for building intelligent public-private partnerships, where all actors deliver to the best of their capacity. RSBY has improved access to health-care facilities, especially once private hospitals were empanelled in the scheme. It has also created technology-enabled enrolment, eligibility verification, and claims processes. The time may come when the scheme should be endorsed by an act of Parliament, to ensure citizens rights are recognized and that citizens play a stronger role in the scheme. To be sure, the case of India is unique for a number of reasons: it is a fast growing economy, with increasing fiscal capacity to enable investment in a massive health insurance scheme, both in terms of premium subsidies and infrastructure development; the Indian IT industry is well developed, a factor that enabled the scheme to deploy smart cards, effective policy administration, and claims management systems; the insurance sector is well developed and sufficiently strong to manage large volumes of policies and claims. Even if the Indian example cannot and should not be duplicated everywhere, India s 9

11 experience is an example of strong government involvement, careful partnership, and engaged partners one that other countries may be able to replicate. Still, challenges remain to increase the quality and sustainability of the program. Measures are required, such as expansion of benefits, investigation into frauds in enrolment and utilization especially in high utilization districts and sample surveys to investigate patient satisfaction, patterns in procedures, claims denials, renewal, usage by migrants, gender bias, and hospital capacity and infrastructure. 10

12 Seguro Agrícola Catastrófico (SAC) 11

13 CASE STUDY AGRICULTURAL CATASTROPHIC INSURANCE (SAC) IN PERU Catastrophic insurance in Peru, also known as Seguro Agrícola Catastrófico (SAC), is a semi-parametric and public-private partnership insurance scheme designed to protect small farmers with a higher economic dependence on agriculture and fewer economic assets to confront negative weather events. The SAC considers an event as a catastrophe when the average return of the insured crop (expressed in kilos per planted hectare) is less than 40% of the recorded historical performance. The SAC started operating in 2009, with its first insured crop season between August 2009 and July So far, it has worked across eight regions of the country that are both the poorest as well as where most of the population is dedicated to agricultural activity. Although this insurance mechanism has created an agricultural risk management system, there is still much improvement to do both on technical and operational aspects. GETTING STARTED INITIAL PUBLIC ATTEMPTS In 2003, the Peruvian government created the National Commission for the Development of Agricultural Insurance to assess the potential to implement an agricultural insurance policy. The committee prepared a technical paper that outlined the main design criteria and implementation of agricultural insurance and its regulation. The committee proposed agricultural insurance that would be operated by the private sector with the additional participation of the public sector to develop studies and an information system. The public sector would also finance a potential grant aimed to promote the development of an insurance market for small and medium farmers. In 2004, the Technical Committee for the Development of the Agricultural Insurance coordinated and implemented the strategy proposed by the National Commission to develop the necessary studies and rules to implement agricultural insurance in Peru. In parallel, the Ministry of Agriculture received technical support from the State Agency of Agricultural Insurance of Spain (ENESA) to develop technical studies to help design a weather index insurance scheme. In 2005, the Technical Committee received additional technical assistance from the Commodity Risk Management Group (CRMG) of the World Bank to further the development of the weather index insurance program. These last studies led to the formal design of a weather index insurance for tanguis cotton on the coast of Peru. Unfortunately, both initiatives failed to thrive because of the lack of political support. Subsequently, in 2006, the Agricultural Insurance Development Unit (UDESA) was created within the Ministry of Agriculture with the aim of implementing insurance schemes based on the previously studies and designs. Unfortunately, this unit also did not receive the necessary political support to operate successfully. However, the General Office of Planning of the Ministry of Agriculture received new technical assistance for the development of agricultural insurance in Peru, which ended in an agreement to develop a pilot project in the regions of Junin and Lambayeque. But, once again, this initiative failed. PRIVATE EXPERIMENTS Meanwhile, in 2005, COPEME, a consortium of private organizations for the promotion of small and micro enterprise development with the support of USAID and GIZ, contracted the services of GlobalAgRisk with the goal of developing micro-finance institutions (MFIs) in Peru to efficiently manage the risk associated 12

14 with their agricultural portfolios. Although this initiative did not materialize immediately, it led to the development of ENSO Business Interruption Index Insurance, a weather index insurance based on sea temperature that was designed to protect the MFI s portfolios in the presence of El Niño. The product was marketed by La Positiva Seguros, a private insurance company. This insurance was made possible through the initiative of GlobalAgRisk, this time supported by Bill & Melinda Gates Foundation and the Instituto de Estudios Peruanos, a private institution dedicated to research, teaching, and dissemination of social studies of Peru and other Latin American countries. In March 2012, the first insurance policy was sold to MFI Caja Nuestra Gente in the Ica region in the south coast. Almost at the same time, the BASIS Research Program on Poverty, Inequality, and Development of the University of Wisconsin sponsored a study to investigate the feasibility of developing weather index agricultural insurance to protect small farmers against risks that affect their income and consumption. Then the BASIS / AMA developed the pilot program Insurance by Index Average Yields for cotton producers in Pisco Valley, Ica region. This pilot program sought to protect cotton farmers who obtained credit in the MFI Caja Señor de Luren, offering a discount of 0.25% on the monthly interest. The program also developed a financial education plan for farmers so they could know about the benefits, advantages, and disadvantages of having agricultural insurance. RENEWED PUBLIC EFFORTS The SAC began with the creation of the Guarantee Fund for the Countryside-GFC (Fondo de Garantía para el Campo) in December 2006, with Law No This fund was created with a total budget of USD$ million and was intended to secure the loans granted by financial institutions to organized small and medium farmers who direct their activity towards domestic and/or international markets. In April 2007, with Law No , the purpose of the GFC was expanded to finance mechanisms of agricultural insurance. The insurance system would reduce the exposure of farmers such as peasant communities, native communities, and small and medium farmers to weather risks and the presence of pests that affect production and profitability. This law also changed the name of the fund to the Guarantee Fund for the Countryside and Agricultural Insurance (FOGASA). In December 2007, with the enactment of Law No , the implementation and operation criteria of FOGASA were set. The law determined that the Directing Council of the fund would be formed by the Minister of Agriculture (MINAG) or his representative, who shall be the chairman of the fund; the Minister of Economy and Finance (or designee); and a representative of the Presidency of the Council of Ministers. And the technical secretariat of the fund would be in charge of the Directorate of Agricultural Insurance and Capitalization (DCSA) from the MINAG. This Law established that the resources of FOGASA would be administered by COFIDE, which would act as trustee of the fund. The resources of the fund were authorized by: Law No , the interests the fund generates, the contributions made by regional and local governments, private funded grants and contributions, resources for technical cooperation grants, and any budget approved by the Annual Budget Law. In January 2008, the Ministry of Agriculture hired the Mexican company LatinRisk to develop a new study to design and implement agricultural insurance in Peru. Although this study laid out the foundation for the development of the SAC, it did not take into account the several previous experiences in the subject, and proposed a very general scheme, extremely similar to the one implemented in Mexico (Agrosemex). In August 2008, the government approved the general terms of the agricultural insurance, later released as Agro Protégé, and the specific conditions of a package of insurance schemes proposed by 13

15 LatinRisk, including the SAC. Then in September 2008, the operating rules of the trust fund for the agricultural insurance (now called Agro Protege) were approved, and the SAC began operations in August THE INSURANCE DESIGN AND MECHANISM The SAC is not considered commercial insurance because it does not cover production costs. Nor is a disaster fund that compensates for the whole damage. Its goal is to reduce the vulnerability of agriculture among small and medium farmers to extreme weather events. The risks covered are: drought, excess humidity, frost, low temperatures, flood, avalanche, hail, fire, strong winds, high temperatures, and pests. Although the SAC is geared to farmers, regional governments are actually the direct beneficiaries of the state subsidy since they negotiate, contract, and implement the SAC. The central government finances the premiums through FOGASA and the regional governments hire the SAC after a selection process. The directing council of FOGASA approves the co-financing table that sets the SAC s mechanisms. This council also has a technical secretariat, which provides administrative support to the board and oversees the Regional Governments in the implementation of the insurance. Operative scheme The Agricultural Catastrophic Insurance (SAC) operation is annual and begins with the approval of the cofinancing table, which is made by the Department of Agricultural Insurance and Capitalization (DCSA) and FOGASA, with the following attributes: Specifies the regions that will receive the premium subsidy Defines the maximum premium rates to be paid Sets the maximum grant for each region Defines the maximum acceptable trigger 14

16 Establishes and specifies the steps and time required for the operation of insurance from the insurance registration, submission and approval of applications for co-financing, to final payment of premiums to insurers selected for each region. Establishes criteria of competition between insurance companies: o o o Rate premiums Risks to cover Compensation trigger The SAC is currently available in eight regions initially identified using specific poverty criteria. The regional government through its agricultural agencies must enrol and validate the list of small and medium farmers. But the distribution of the fund between these regions do not respond to any technical and validated reason that could be found in official documents or even in interviews with government functionaries, and the same happens for the distribution of resources between crops and areas within each region. The lack of transparent criteria for this selection has to do with an even bigger problem, which is the absence of a clearly defined objective of the SAC, and the subsequent methodology for identifying beneficiaries that meet that goal. The original law that created the fund (Law N.29148, second article) talks about the need to reduce exposure of agricultural producers to weather risk and presence of plagues that reduce their production, and mentioned small and medium producers as well as peasant and native communities as beneficiaries of the insurance. The operative norm for the functioning of the fund defines small and medium producers based on a combination of a maximum amount of yearly sales and hectares. However, that definition is inapplicable in practice because it has been proved that there are serious inconsistencies with this definition. And even if there is a better definition of the beneficiaries of the SAC, it turns out to be irrelevant as the contract of the insurance policy for a specific region is done based on the sowing intentions for specific crops, regardless of the type of producers who cultivate them. Once again, the selection of crops that can be insured was initially chosen by FOGASA without a clear selection mechanism, but later this decision was left to each regional government. There are no homogeneous criteria for the selection of crops to be insured in all regions. The determination of the maximum premium rates to be paid (14.5%) as well as the trigger that activates the insurance mechanism (average return of the insured crop, expressed in kilos per planted hectare, is less than 40% of historical performance) were initially suggested by the Latin Risk 2008 consultancy, without a technical explanation, and have not being modified since then. Unfortunately, Peru does not have a proper risk map that identifies the main risks associated with each crop in each geographical area. Even if this information were available, it still wouldn t be possible to improve premium rates calculations using available historical yield information. Currently, the selection process of the insurance company is not officially regulated, which could result in problems between insurers, brokers, and representatives of the regional governments. In addition to that, the short deadline for submission of proposals (5 days) discourages the possible involvement of new companies, limiting the potential development of a bigger market for agricultural insurance. It is also not clear that the evaluation indicators generate the correct incentives for insurance competition and improvement. The type of risk to cover is almost irrelevant as the determination of payment is based on yield reduction, with the premium rates and trigger set arbitrarily. 15

17 THE OPERATION OF THE SAC When a disaster happens, the community must report through their leaders or the local media to the agricultural agency staff. That agency will make a report of the incident and send it to the regional agricultural office, a department of the regional government, which then directs the report to the insurance company. A company adjuster is then responsible for verifying any damages or losses incurred by assessing the area to determine whether it should be covered or not by insurance, how many hectares must be indemnified, and the value that corresponds to every crop. Sometimes the damages found are contrasted with the report of installed areas and affected areas managed by the agricultural agency. While there are adjustment procedures established in the policies, there is no uniformity in this process, leaving everything to the goodwill of the insurance system and the regional government. The adjustment process does not generate data that enables any future determination that it is developing properly. Lack of budget limits the ability of agricultural agencies to generate improvements in statistical information recovered. Plus, the agricultural agencies receive the news of a disaster and must prepare the report for each insurance company. However, there is no formal budget to enable them to do it. This limits its activities and generates high operating costs for insurers, which in turn affects the ability to obtain better prices for insurance premiums. The process of transmission of information of a catastrophic event from the producer to the insurance company needs to be improved, as it has started to be done in some regions where the insurance companies are installing online reporting systems. There is also a strong inconsistency related to the incompatibility between the methodology used for determining the historical crop performance and the methodology used to estimate the crop performance in the presence of a catastrophic event. On the one hand, to estimate the average yield per hectare, the annual production is divided by the harvested areas of that year using information produced by the Ministry of Agriculture through a method of qualified informants which is basically the opinion of field experts about the sowing area, yields, and prices for each crop in each SEA. With this historic annual average information, the insured yield is estimated by multiplying it by the 40% trigger. The methodologies used are completely different and incompatible, and they can easily underestimate or overestimate the real affected yield, and hence severely risking the sustainability of the SAC. Once the insurance gets activated, the beneficiaries are named using a process that depends on the community leaders to report the farmers who suffered losses to the agricultural agencies. They are responsible for the registration and validation of the list of beneficiaries of the agricultural insurance and payment is made once the insurance company approves this list. Currently, indemnities are paid through a financial institution or directly transferred to the regions. Unfortunately, the process of registration of potential beneficiaries is done after the company adjuster determines the total lost hectares and this leads to delays between filing the claim and paying those affected of up to seven months. Another major problem in the operation of this insurance is the delay in the approval of the co-financing table. This is supposed to be signed before the start of the agricultural season, but what actually happens is that the agreement is signed after the start of the campaign so any spoilage or damaged crops between the start of the season and the signing of the contract are not covered by the insurance policy. This limits the effectiveness of the insurance and extends to underestimate the claims. The late adoption of the co-financing table allows insurance companies to change their decisions, like choosing to 16

18 Crop season participate or not as they see the evolution of the agricultural season. If the insurance company knows the presence of ex-ante claims, it will evaluate the cost-benefit of participating. RESULTS As we can see in the next table, the amount of premiums, total insured amount, and insured surface area has not changed much during the first 4 years of the SAC operation. The regions attended are also the same: Ayacucho, Apurímac, Huancavelica, Cusco, Cajamarca, Huánuco, Pasco and Puno. Premium Operation outcomes of SAC Total insured amount Insured surface area (ha) Amount indemnified 1 Loss ratio 2 Number of beneficiaries 2009/2010 US$ 13,697,115 US$ 76,734, ,069 US$ 3,353, % 31, /2011 US$ 14,275,242 US$ 85,138, ,210 US$ 8,554, % 99, /2012 US$ 14,279,589 US$ 86,400, ,108 US$ 3,491, % 37, /2013 US$ 15,525,396 US$ 93,938, ,149 US$ 4,199, % 55,945 TOTAL US$ 57,777,342 US$ 342,212,287 1,796,536 US$ 19,597, % 224,000 The regions with higher loss ratio in all this period were Puno, Pasco, Ayacucho, and Huancavelica with 75.07%, 81.78%, 81.71%, and 40.98% respectively. The insured sum per hectare was in between 400 and 550 soles (around US$140 and US$200), which appear to fall short to the real losses suffered. During the campaign, for example, the Ministry of Agriculture found that many communities from Huancavelica lost nearly all their crops (corn, potatoes, barley, beans, peas, among others) so it was established that the insurer would indemnify the farmers with US$197 per hectare of crop damage. However, some farmers receive amounts ranging from 31 to 59 dollars for extensions of their plots (an average of 0.25 hectares). Farmers invested at least US$ 784 in cereals to US$ 2,353 in potatoes per hectare. Besides the low amount, the compensation payments must be claimed in the provinces, forcing farmers to travel long distances and spend money on tickets and travel expenses. Consequently, the amount of insurance coverage not only generates a negative and pessimistic sentiment in the farmer, but also prevents their capitalization. In addition to this, the distribution of resources between regions seems insufficient. Without a clear policy of crop priority, the coverage amount per hectare will fall short in order to cover all the insured crops. Two possible scenarios appear: either they stop paying some producers, or the amount of compensation is reduced per producer. A study about perceptions of the SAC in Puno and Ayacucho, found that on average 88.6% of the households who received a payment were poor and 73.7% were extremely poor. This means that despite the focalization problems mentioned before, at least in these regions it seems to be a good targeting by poverty conditions. However, the same study found that only 17% of the beneficiaries interviewed considered the amount of compensation to be fair, mainly because it was an unexpected payment (66.7%) or because it helps with their expenses (24.1%). The 82.9% who consider it unfair mentioned the insufficiency of the payment to cover the needed expenditures to confront a catastrophic event (94%) as their main reason. And it is consistent with the fact that the compensation received is equivalent only to the 37.6% of their agricultural annual expenses. This information also reflects the fact that many producers do not know that they are covered by the SAC or the benefits that 1 The exchange rate soles to dollars considered is from the 31 st of December of each year. 2 The loss ratio is the percentage expression of the existing relationship between the cost of claims and the net premium (excluding taxes) of the policy. 17

19 they can get from it. At present, the most important channel to spread the news about the SAC is the community leaders who receive information directly from their agricultural agency. In 2014, under Ministerial Resolution No. 076, insurance companies that apply to the SAC and offer a communication plan that includes TV spots and radio announcements besides brochures, will be rewarded with more points in the selection process. If beneficiaries do not know they are insured, it will be very difficult to observe changes in behaviour associated with lower risk. Insurance companies make information campaigns for the local and regional public counterpart to know about this product, but the government that subsidizes the premiums should complement this practice. FUTURE CHALLENGES This document mentions several limitations and challenges currently faced by the SAC. But the government is already taking action to improve the design and implementation of the SAC, including: Fund distribution between regions for the next crop season will be done based on an index of disasters from the last four FOGASA campaigns. The SAC with use a new definition of beneficiaries of the program described as "subsistence" producers, according to the priorities established by the agricultural ministry. New directions for the registering beneficiaries in each region, as well as the opening of bank accounts in offices of the National Bank or other financial institutions in order to make direct payments. FOGASA is currently evaluating two international proposals for a study with the following purposes: o o Elaborate a new methodology for the evaluation of catastrophic events, estimation of the trigger, and field adjustment procedures. Risk mapping for better targeting of beneficiaries and crops 18

20 19

21 CASE STUDY CADENA IN MEXICO Within one decade, Mexico transformed the lives of its low-income farmers through a persistent goal of providing agricultural microinsurance and continually tuning its public-private partnership to ensure the program s longevity and success. Agricultural microinsurance is notoriously difficult to offer, with premiums linked to unpredictable weather conditions and costs affecting farmers with little financial cushion when things go wrong. But the Component of Assistance against Natural Disasters (CADENA, for its initials in Spanish) integrates smart subsidies, awareness raising activities, and a flexible operational model that allows a public insurer to compensate for a shortfall in private coverage. These tools have driven rapid expansion of basic catastrophic coverage from just 9% in 2003, when CADENA began this project, to nearly three-quarters of the target population of low-income farmers today. And as Mexico steers toward its goal of universal coverage for all low-income farmers by 2018, it is looking to find creative ways to ensure the program s fiscal sustainability and enhance private sector competition. GETTING STARTED After the passing of Chapter 12 of Mexico s Sustainable Rural Development Law in 2003, CADENA oversees both the private and mutual insurance programs and controls 80% of the approved federal budget for responding to rural contingencies. CADENA is managed by the Secretariat of Agriculture, Livestock, Rural Development, Fisheries, and Food (SAGARPA), and was originally a spin-off of the National Fund for Natural Disasters (FONDEN). Post-catastrophe support for farmers was moved from FONDEN to SAGARPA, after the government recognized it needed to respond differently to farmers facing environmental hazards than general civil protection. CADENA creates a social safety net for vulnerable farmers while it reduces and manages budget outlays through ex-ante insurance rather than direct ex-post assistance. CADENA employs a combination of federal subsidies, standardized tender processes, and training to incentivize states to acquire private catastrophic agricultural risk insurance tailored to the needs of regional farmers. The program also involves a public insurer to cover commercially unattractive areas as well as to drive product innovation. The result has been a massive increase in coverage and decrease in response times for vulnerable populations affected by natural hazards at much lower per-unit costs, as well as the gradual development of a previously untapped insurance market. STAKEHOLDERS SAGARPA, which coordinates the entire CADENA program, must: Define subsidies, target segments, rules, and guidelines Monitor the entire program Manage subsidies Negotiate the terms of centralized coverage options through public insurer Agroasemex Offer annual training seminars to state government officials The states and Federal District determine their participation in CADENA, usually coordinated through 20

22 state agricultural agencies (called state Secretariats of Farming Development (SEDAGRO), in most states), must: Identify areas, risks, and products to insure Purchase coverage through one of the CADENA mechanisms Declare emergencies and hazards as they occur Claim and receive benefits Distribute benefits to covered farmers Identify farmers covered, either through pre-registration or post-hoc verification via inspection. At the local level, states often rely on municipal officials to: Inform constituents about their coverage Alert them if payouts become available after an insured event Organize distribution of payouts Public insurer Agroasemex has a dual role as both a direct insurer for centralized policies and as the primary research and development agency for innovating products and technology. As the only provider under the centralized contracting mechanism, Agroasemex received 57% of gross premiums under CADENA in As a not-for-profit, public entity, it is often willing to invest in new technologies and assume larger risks than the private sector. Agroasemex was the first (and until 2013, only) entity to provide parametric crop insurance. Private insurers play an increasingly important role within CADENA as direct insurers to the states through the federalized mechanism. Three companies are currently active in the program: Protección Agropecuaria Insurance Company (ProAgro) specializes in agricultural insurance, and first participated in It continues to be the most active private insurer with 27% of CADENA premiums in General de Seguros (GS) and Mapfre-Tepeyac joined CADENA in 2005 and 2006, and as of 2013 managed 1% and 7% of CADENA premiums respectively. All three private insurance companies transfer approximately 80% of their risk to private reinsurers. While reinsurers do not contribute directly to CADENA, they are critical to enabling those private insurers to participate. In some cases, private insurers and states were unable to ensure direct coverage due to reinsurers refusal of support based on a high perception of climatological or security risks. The National Confederation of Livestock Organizations (CNOG) plays a unique and important role in providing insurance for livestock through its own insurance fund, which serves livestock associations across the country. CNOG received 8% of gross premiums in Small vulnerable farmers, the target beneficiaries, are currently defined as producers with up to: 20 hectare of annual crops, 10 hectares of fruit crops, or 60 animal units (one animal unit is equivalent to 1 cattle, 1 horse, 5 sheep, 6 goats, 5 swine, 100 birds, or 5 bee hives). While fishermen and fish farmers can also be covered, to date no state has requested coverage for these target segments. SAGARPA has gradually expanded the eligibility criteria over time, most notably the inclusion of producers with access to irrigation or those with commercial insurance who wish to add catastrophic coverage. While some observers worry that this expansion deviates from the initial focus on low-income populations, the changes were necessary to accommodate the different socio-economic profiles between 21

23 northern and southern Mexico. PRODUCTS CADENA subsidizes multiple insurance products, each responding to the different needs and contexts of the states and their target communities. Products fall into two major categories: parametric and traditional. Parametric insurance for both crops and livestock is popular for its efficiency, enabling coverage of dispersed populations in large geographical areas. While eliminating the need for on-site inspections should reduce premium costs, parametric insurance has been relatively expensive under CADENA, perhaps due to Agroasemex s monopoly. As other insurers began to offer parametric products in 2014, premiums should go down. Parametric crop index insurance compares actual rainfall and temperatures to local indices based on historical levels. Excess or insufficient moisture or heat triggers payments to help producers prepare for the next growing cycle. Measurements are provided by weather stations built by the National Water Commission. The low number of weather stations and the rigorous historical databases needed to calculate an appropriate index that meets the standards of reinsurers limit the scope of this insurance. Parametric satellite pasture insurance uses satellite imagery to compare the Normalized Difference Vegetation Index (NDVI) of pastures to levels necessary for each stage of production. Low NDVI levels, meaning insufficient pasture, trigger a payment to cover the cost of purchasing additional food for animals. Traditional insurance for both crops and livestock has grown in importance for areas without weather stations as coverage has deepened. Traditional products determine payouts in various ways. Traditional crop insurance uses a variety of method to protect farmers from losses. The most popular is area yield index insurance, which compares actual yields to an index based on average historical yields for specified crops in a specified geographic area, usually a municipality. Unlike parametric insurance, it requires field inspections at the end of the production cycle (except in cases of total loss events) to verify the yield, which makes it slower and more expensive. But the specific risks causing low yields are more comprehensive than parametric crop insurance, which only focus on moisture and temperature. Traditional livestock insurance verifies weight loss of insured animals according to a visual index, requiring a random sample field inspection. Another disadvantage of this product is that animals have already suffered weight loss at the time of indemnity, unlike the NDVI parametric product. The required inspections generate higher operational costs and slower payouts (especially for remote affected areas), as well as safety risks in dangerous areas. All CADENA products offer support to small farmers, as you can see from these authorized payouts for illustrative products and crops: Rainfed annual crops Irrigated annual crops Fruit trees, coffee, and prickly pear Livestock (for fodder supplement) Livestock (in case of death) MXN 1,300 per hectare MXN 2,200 per hectare MXN 2,200 per hectare MXN 600 per animal unit MXN 1,500 per animal unit While these amounts don t cover all losses, farmers indicate that small amounts can help finance the next cycle after a bad harvest. CADENA insurance products cover a wide range of meteorological risks 22

24 (drought, cold, hail, snow, torrential rain, freezing or low temperatures, severe flooding, tornados, and cyclones) as well as geological events (earthquakes, volcanic eruptions, tidal waves, and landslides), which were added in SUPPORT TYPES CADENA uses two distinct support types and two contracting mechanisms to maximize protection nationwide. The primary support type consists of Catastrophic Insurance. The federal government subsidizes 80-90% of premiums for insurance contracted by or on behalf of state governments, transferring the risk of supporting small farmers to private insurance markets. Insurance for communities classified as highly marginalized receive a 90% federal subsidy, while regular areas receive an 80% federal subsidy. Insurance subsidies represent about 80% of the total CADENA budget. This ex-ante assistance is designed to be cost and time efficient. CADENA also provides ex-post Direct Assistance. The federal government subsidizes 60% of direct payments authorized by state governments to small farmers, for lack of insurance or in excess of coverage obtained. Direct Assistance aids farmers who: Live in states that did not insure Raised uninsurable crops Did not follow typical production cycles SAGARPA has considered insuring its outlays under this support program as well to reduce costs. RESULTS: MASSIVE GROWTH IN OUTREACH THROUGH INSURANCE CADENA has grown exponentially since its inception, primarily due to the expansion of its insurance mechanisms. By prioritizing ex-ante insurance coverage over ex-post Direct Assistance, CADENA represents a massive transfer of risk from the federal and state governments to the private insurance and reinsurance markets. In 2014, CADENA had an approved budget of over USD 229 million for the insurance component, and USD 63 million on the Direct Assistance component, compared to just USD six million for Direct Assistance in Premiums and sums insured have also grown accordingly. This rapid increase in uptake for catastrophic insurance reveals the decade-long evolution of risk management mentalities and practices among state governments a massive paradigm shift in response to clear policy incentives and information. The growth of CADENA has also coincided with a gradual increase in participation by the private sector from just 17% in 2004 to 50% in 2010, after which these levels have remained relatively unchanged. 23

25 CADENA premiums and sums insured Millions Premiums, state Premiums, federal Sum insured The program s growth and the gradual expansion of eligibility have enabled a massive increase in outreach to the target population. In 2013, CADENA covered 12 million hectares of arable land and 10 million animal units in all states and the Federal district under the insurance component alone, representing approximately 75% and 70% of the estimated potential target populations of growers and ranchers respectively. Prior to CADENA s launch, only 9% of cultivated cropland in Mexico was insured. Millions CADENA coverage of target population Hectares insured Animal units insured CADENA stakeholders unanimously concluded that the rapid expansion of coverage has been the program s greatest achievement; it has also managed to improve the quality of protection as well. For example, the gamut of insurable risks has also grown. CADENA initially insured only non-irrigated corn, 24

26 but today it insures a wide variety of crops including beans, coffee, fruit, and prickly pear, both with and without irrigation. Furthermore, CADENA has reduced post-catastrophe response by as much as six times. Under its predecessor FONDEN, the federal government took an average of 20 weekdays to verify a catastrophic event, and by the time state governments reimbursed local governments, it took farmers up to six months to get their payments. By contrast, parametric catastrophe insurance verifies the claim in a few days, and distributes benefits to farmers in about nine days. Even traditional guaranteed yield products that require field inspections at the end of the productive cycle are able to disburse indemnities in line with producers typical cash flow timelines. KEYS TO SUCCESS: ALIGNING INTERESTS TO BUILD A MARKET FOR PROTECTION Mexico was among the first countries to implement nationwide index-based insurance as part of a larger natural catastrophe response program and social protection agenda. CADENA has become a reference point for governments worldwide that wish to protect low-income producers in the rural sector. Unlike many public-private partnerships, SAGARPA did not directly negotiate alliances between public and private entities. Instead, it has used purposeful public policies to incentivize the emergence of a private insurance market serving state governments and, through them, low-income farmers. This strategy has powered the exponential expansion of coverage to previous excluded populations, while respecting the initiative of the private sector and the autonomy of states. Some of the most essential elements of SAGARPA s strategy include: Smart subsidies Operational clarity and transparency Training and awareness raising Use of a public insurer to lead fast and by example SMART SUBSIDIES SAGARPA has used federal subsidies as a primary policy instrument to encourage states to use insurance to mitigate natural hazards. Insurance has several clear advantages over direct government assistance. It pays more benefits per dollar budgeted and, especially in the case of index insurance, pays benefits much more rapidly. But during the first five years of SAGARPA s catastrophic risk coverage program, the Direct Assistance mechanism dominated the other insurance mechanisms. After nearly a decade of Direct Assistance under the previous FONDEN program, state governments favoured a reactive rather than preventive approach to natural disasters, with no clear incentives to change, which resulted in the federal government subsidizing 70% of damages or costs assessed for both insurance premiums and Direct Assistance outlays. In 2008, SAGARPA increased the federal subsidy for catastrophic insurance premiums to 90% and reduced the subsidy for Direct Assistance funds to 50%. This change marked a turning point for CADENA, catalysing exponential growth in premiums in absolute and relative terms compared to Direct Assistance. You can see in this chart how Catastrophe Insurance overtook Direct Assistance: 25

27 Federal budget utilization for CADENA Millions Direct support budget Insurance budget With greater incentives to act preventively rather than reactively, states demand for insurance grew steadily and was reinforced with each claim paid. While this explosion of demand has increased CADENA s assistance budget substantially, from just USD 6 million in 2003 to USD 292 million approved in 2014 (excluding administrative costs), reliance on insurance over direct assistance secured significant cost savings. For instance, in 2013, total insurance subsidies amounted to USD 22 per hectare of annual crops, compared to USD 220 per hectare spent on Direct Assistance. Similarly, federal and state governments combined spent just USD 10 per animal unit on insurance, compared to USD 60 for direct assistance to livestock losses. More importantly than the actual cost savings, CADENA s incorporation of insurance has protected federal and state budgets from the volatility of requests for public assistance due to unpredictable catastrophic events. Cost savings have benefited state governments as well. In 2011, the state of Chihuahua contracted crop insurance for multiple products under the federalized track and joined the centralized livestock policy to cover cattle. Chihuahua paid just USD 1.3 million of the total USD 5.9 million premium for a sum insured of USD 64.8 million, of which USD 25 million was paid in indemnities for catastrophes experienced. Had the state not contracted insurance and relied solely upon Direct Assistance, it would have had to spend USD 12.5 million for the same level of support, in addition to undergoing a slower payout process. SAGARPA frequently makes this powerful point with its state counterparts, and states savings have been a major motivator of uptake. Since the 2008 change, SAGARPA has adjusted federal subsidies in response to perceived needs. For instance, it has gradually increased the federal premium subsidy to insure non-marginalized areas, in the interest of promoting universal coverage. SAGARPA also increased the federal subsidy for Direct Assistance from 50% to 60% in 2013, in response to the increasing frequency and unpredictability of natural disasters due to climate change. 26

28 While the upward creeping level of subsidies in nominal and percentage terms has raised questions about sustainability, SAGARPA is currently formulating strategies to counteract these concerns. OPERATIONAL CLARITY AND TRANSPARENCY Most stakeholders believe that establishing and enforcing clear rules and processes was critical to CADENA s success. As the central coordinator and the largest contributor of resources, SAGARPA defines the rules and guidelines of operations each year, in consultation with state governments and insurers. The respect that CADENA s annual rules of operation command has allowed smart subsidies to do their job, leaving no room for doubt. But the rules have not been constantly static. Since 2003, CADENA has undergone over 120 changes, most of which focused on: Expanding eligibility criteria Promoting higher private sector participation and greater autonomy on behalf of state governments Giving greater flexibility to Agroasemex to act as direct insurer in markets where the private sector was nervous to enter For most of CADENA s history, federalized contracts followed a rigorous tender process clearly outlined in CADENA s rules of operation, including competing bids by at least two private insurers. SAGARPA s technical committee conducted the evaluation and selection of bidders with technical commentary from Agroasemex. This process allowed for a high degree of transparency and rigor that instilled confidence in private insurers to participate in the program, which was essential to reaching scale. The confidence level has been questioned at times, such as the change in 2012 whereby formal tenders were no longer required for federalized contracts and states were allowed to choose their own insurers without technical oversight from SAGARPA s technical committee. Although this change made the federalized process more flexible and awarded more autonomy to the states, some private insurers believed that transparency and fair competition suffered. Clear rules and transparency have also been crucial to creating buy-in at the grassroots level. For instance, the requirement to clearly identify producer beneficiaries and verify benefit recipients not only lends transparency and legitimacy to the process in general, but it creates buy-in among farmers who can then pressure their local and state representatives to participate. As SAGARPA officials explained, the evolution of CADENA reflects a dynamic learning process in which all stakeholders are fully engaged. At each stage, clarity, transparency, and adherence to the rules was critical, especially to encourage private sector participation. This contrasts with the ad-hoc or outright corrupt operations of previous government-led initiatives. TRAINING AND AWARENESS RAISING Capacity and awareness building among stakeholders has been essential to CADENA s growth. Mexican political and bureaucratic institutions have long suffered from a high degree of turnover, partly due to the Revolution-era prohibition and continuing taboo against re-election. Furthermore, local governments often suffer a lack of internal capacity and awareness, including related to agriculture and insurance. SAGARPA realized it was spending significant resources on educating incoming state and local officials about CADENA and the need for insurance, only to see that buy-in leave with the officials at the end of their term. 27

29 To combat this trend, SAGARPA partnered with the Monterrey Institute of Technology and Higher Education in 2010 to develop basic and intermediate seminars for bureaucrats in state governments, state-level SAGARPA delegations, and other interested agencies. The basic series consists of four consecutive 1.5-day weekend sessions and covers the concepts of climate change, risk management, and insurance, culminating in an overview of CADENA s Catastrophe Insurance component. SAGARPA believes the training program has proven instrumental in the accelerated growth in CADENA insurance coverage. Now, these trained functionaries often take the initiative to educate their superiors on the importance of CADENA and take the lead on contracting coverage. Raising awareness for producer beneficiaries also improves CADENA s results. Previously, few farmers were aware that they had coverage under the scheme, and those who did were unaware that it was primarily through insurance. Now, farmers have become more familiar with CADENA and catastrophic risk insurance, while local governments promote it directly and effectively. SAGARPA also delivers direct training workshops for producers. Direct experience is also a powerful awareness-raising tool. Indemnities are often delivered in mass meetings of farmers, where representatives from affected municipalities, state officials, and SAGARPA managers drive home the importance of the program. Local representatives also typically visit their constituents before each productive cycle, alert farmers to their coverage, and notify them when benefits become available. At one meeting in the state of Jilotepec, one farmer remarked, One day the ward representative came and invited us to a meeting to pick up a benefit from the government. At first we didn t understand, but after several bouts of bad weather, we know what it s about. As part of a new policy in the 2014 rules of operation, producers will be encouraged to augment their coverage by paying 25% of the additional premium. This important step toward sustainability seeks to build upon demand stimulated through over a decade of total subsidies as well as the more recent program to train producers. Finally, SAGARPA is using training- and awareness-building activities to help reduce the loss ratios of catastrophic insurance due to poor farming practices. In Mexico, many communities cultivate crops not ideally suited to the local climate, but new support helps with reconversion to more suitable crops that can be more easily insured. USE OF A PUBLIC INSURER TO LEAD FAST AND BY EXAMPLE SAGARPA s vision has always been to increase private sector participation in CADENA over time. Yet even today, among nearly 300 private insurance companies, only 14 offered some type of agricultural insurance, with only three truly active in the sector. The private sector s participation has increased nominally over CADENA s life, though its percentage participation has hovered around 50% since 2010: 28

30 Area covered by public vs. private insurers 100% 90% 80% 70% 60% 50% 40% 30% 20% 10% 0% Private Agroasemex Restrictive requirements on government tendering of private entities, private insurers risk aversion, and lack of technical capacity have all played a part in limiting private sector advancements within the publicprivate partnership. Given these limitations, public insurer Agroasemex has played an important role in rapidly expanding outreach and developing new technology and products. In CADENA s early years, it was doubtful whether SAGARPA could even find enough companies willing to submit bids, which led to it contracting its centralized coverage exclusively through Agroasemex. Aside from its role as a direct insurer under CADENA, Agroasemex is primarily considered a development agency for innovating products and technology as well as a national reinsurer of last resort. Just before the inception of CADENA, Agroasemex developed innovative parametric crop insurance with support from the World Bank. Agroasemex s identity as a public body also facilitated the coordination with national agencies such as the National Water Commission and usage of their infrastructure. By working directly with Agroasemex, SAGARPA was able to cover larger areas more efficiently than would have otherwise been possible with traditional insurance products. On the other hand, private insurers would have been either unwilling or unable to make the necessary investments to develop parametric products at that early stage. Some observers believe Agroasemex s direct insurer role under CADENA has undermined its developmental role. Although the parametric technologies and products it developed exist for the benefit of the market, Agroasemex has declined to share data on them with private insurers. Without direct support from donors or the government, private insurers have resorted to alternative means to develop parametric products copying policies on the market or hiring consultants involved in creating the original Agroasemex product. Given the popularity of parametric insurance, these insurers expect that introducing them will result in a significant increase in market share through the federalized track. This would allow CADENA to break out of the half-and-half public-to-private distribution prevailing since

31 CONCLUSIONS AND FUTURE CHALLENGES Unlike more deliberate public-private partnerships, CADENA demonstrates that by aligning interests and building capacities through policy, private and public stakeholders can achieve synergies catalysing market expansion and greater social protection. Smart subsidies and training offered to state governments stimulate demand, while clear and transparent rules encourage private insurers to assume new risks while giving them a common platform for reaching otherwise unreachable populations. While these factors contributed to success in the first decade, many challenges remain. CADENA s budget continues to expand, causing serious concerns about sustainability. Federal subsidies were slightly raised again in 2012, and farmer eligibility was expanded in SAGARPA will need to coordinate with its key stakeholders to define a reasonable limit to support, and strategize to rationalize its budget within those levels. Changes under consideration to build up long-term sustainability include insuring the federal costs of the Direct Assistance component, strengthening the promotion of private sector competition, and encouraging producers to contribute a part of the premium. Furthermore, although three private insurers are now active in low-income markets where there were previously none, competition is still stifled and some are concerned about the role of Agroasemex as a direct insurer rather than a development agency. To alleviate these woes and encourage even greater private sector participation, SAGARPA should continue to refine its operating rules to create a transparent and rigorous process. It should also work with Agroasemex to help it transition from a role as direct insurer to the fulfilment of its developmental task to innovative products and technology for the benefit of the whole market. These challenges also present an opportunity for SAGARPA to capitalize on the newly cultivated risk management and insurance culture of states and farmers, to deepen their appreciation for transparent and well-planned governance and a healthy relationship with the private sector. For this, the flexibility and creativity of SAGARPA evidenced in the first decade will certainly be of vital importance in the next decade. 30

32 31

33 CASE STUDY BANCA DE LAS OPORTUNIDADES IN COLOMBIA Colombia began a private-public partnership (PPP) through its Banca de las Oportunidades program to stimulate insurance companies to offer insurance linked to conditional cash transfers. Microinsurance enables poor households to protect their fragile livelihoods, assets, and lives against insurable risks and unexpected catastrophes. Without access to risk mitigation tools, insurance, or adequate social security services, poor people are extremely vulnerable and ill-equipped to meet shocks that life inevitably brings. Insurance helps mitigate the effects of adverse shocks that can leave families poorer and unable to get back to their previous economic situation. Poor people, especially those who derive their income from informal activities, tend to be underserved by mainstream commercial and social insurance schemes. GETTING STARTED Although more than 95% of Colombia s population has access to health insurance through the social security system, and about 50% of urban adults have some sort of funeral insurance, the number of people who access or use other types of insurances is very limited. The lowest income population lacks access to insurance because of: unawareness, distrust, absence of adequate products, limited outreach and capillarity of the insurance companies, and limited channels for distribution, claims and payments. The Colombian government established the Banca de las Oportunidades program in 2006 to promote access to financial services for the low-income population of the country. Banca de las Oportunidades: Promotes regulatory reforms to open financial services access to underrepresented sectors Develops projects to unite financial services agencies and potential clients Incentivizes financial services providers to improve coverage, quality, and methodologies for lowincome populations The other agency that works with Banca de las Oportunidades is Red Unidos, Colombia s national strategy for the eradication of extreme poverty through the social and economic inclusion of the poorest of the poor. Its network offers basic social services through 26 public agencies to ensure that eligible low-income families have access to the programs. A key component of this strategy provides 1.5 million families and communities with specialized assistance through 10,000 social workers (or cogestores ) who are trained to motivate families to achieve the 45 basic criteria defined by Red Unidos as necessary to overcome the extreme-poverty trap. These criteria are classified into nine categories: identification, income and employment, health, education, nutrition, housing, family dynamics, savings and credit access, and access to justice. The shared interest of these two programs, as well as Colombia s mandate to prioritize projects for Red Unidos families, led to the design of a joint pilot project to offer microinsurance for families in extreme poverty. To facilitate the distribution and premium repayment process, the agencies leveraged the existing conditional cash transfer (CCT) program deployed by Familias en Acción, which provides cash transfers to poor households if their children attend school and follow preventative health care measures. The program was deployed to the 470,000 families that were already engaged with both Red Unidos and Familias en Acción. In 2010, Banca de las Oportunidades, with Red Unidos support, started a pilot project to stimulate insurance companies to develop and offer life and funeral microinsurance to families in extreme poverty. 32

34 The pilot also sought to motivate insurance companies to reach these families, allowing them to learn and better understand the insurance market for low-income population and identify sustainable models to serve them. It was expected that information learned from the pilot would help other insurance companies develop this market, which could eventually ensure the product s sustainability. STRUCTURE INITIAL STAGES Microinsurance promotion proved to be more challenging than promoting other financial services to low income clients, due to difficulties in product design as well as the insurance companies operational structure and geographical coverage that is primarily focused on medium- and high-income clients, typically in urban environments. Banca de las Oportunidades initial approach was to educate insurance companies in low-income markets by conducting a market study in 2008 on CCT s beneficiaries to show the potential demand, as well as perceptions and opinions on microinsurance. This study showed: what people thought of insurance companies; how access to microinsurance products met the target population s needs in terms of reliability, opportunity, and transaction costs; as well as their willingness to pay a reasonable premium. In spite of these results, the first challenge was to get the insurance companies to participate in the tender process to implement coverage without exclusions for pre-existing conditions, and specific conditions for claims and payments. The initial tender process asked insurance companies to offer life, disability, and funeral microinsurance to the 470,000 beneficiaries who already belonged to both Red Unidos and Familias en Acción. The product guidelines required: Life insurance coverage of USD $1,493 plus USD $498 for other family expenses Funeral insurance for USD $498 Disability insurance for USD $1,493 plus USD $498 for other family expenses Simple, easy to understand policy guidelines No exclusions or preconditions Coverage for suicides and pre-existing conditions Simple and expedited procedure for claims and re-payments As an incentive for insurance companies to explore this new market, Colombia offered to pay the premiums for the first year, as well as for an additional six months for any families that renewed their policies. The tender would be won by the company that obtained the highest score offering the smallest premium for the best product, support, and collection mechanism. In addition, Red Unidos offered its 10,000 social workers to help locate and identify beneficiaries for the insurance company, as well as give families information about the insurance and support the renewals process. In return, the insurance companies had to insure the families, establish a simple and efficient mechanism for claims and payments, and commit to renew at least 75% of the policies during the second year and 50% of the renewed policies during the third year. Failing to accomplish these goals would result in penalties. Despite Banca de las Oportunidades and Red Unidos efforts, three successive tenders had to be conducted since insurance companies failed to participate in the first two. The first tender was declared 33

35 void because no insurance company submitted a proposal. According to the insurance companies association (Fasecolda), the main pain points were: Insurance companies feared the policy renewal obligation for an unknown market and the legal consequences of failing to achieve it. The difficulties and additional costs required to determine disability insurance applicability. The estimated budget was not large enough to cover 470,000 potential clients. After some discussions with the insurance companies, a second tender process was carried out, with minor changes from the initial design. The most significant changes were that instead of being obligated to renew the policies, the insurance company would make its best effort to renew at least 60% of the policies during the second year, and 60% of those during the third year. Only one insurance company submitted a proposal but it did not meet the requirements and surpassed the budget by four times. Consequently, after two deserted tenders, Banca de las Oportunidades decided to revise the project design and hired a consultant to talk to the different stakeholders (especially insurance companies) to understand the constraints faced by the insurance companies and adapt the proposed product design to the supply s reality. SECOND PHASE Taking into account the lessons learned from the initial experience and the insurance companies concerns, the second phase included significant changes to the tender process, with the main characteristics of the new pilot being: Coverage for 50,000 in 21 cities Only offer life and funeral insurance (no disability insurance nor additional money for family expenses) Contract for 1 year, 100% of the premium covered by Banca de las Oportunidades, with an estimated budget of USD $348,000 (considering a premium of USD $5 plus training expenses) No obligation to renew Contract for 1 year The changes to the tender process proved to be effective. For the third tender process, five insurance companies submitted proposals (Liberty, Mapfre, Solidaria, Positiva, and Mundial) and the contract was awarded to Positiva, a government-owned insurance company. The contract between Banca de las Oportunidades and Positiva was signed on February 2012, and awarded up to USD $348,432 to offer a life insurance product to 50,000 beneficiaries from a list of 54,000 (as it was estimated that about 7% of the beneficiaries would not be found). 34

36 The annual premium to be subsidized was USD $5 for a total of USD $250,000, and a total of USD $98,432 in additional funds to cover training expenses. Positiva had the most outreach with offices in 28 cities, and offered to provide beneficiaries with access to a call centre for help and inquiries, and claims could be paid through a bank account or by bank transfers. The operative process of the contract had the following milestones and main activities: Red Unidos gives to Positiva complete information of the beneficiaries to insure and the corresponding "Cogestores" Positiva trains Red Unidos field staff to distribute the policies Positiva preissues policies according to provided listings Red Unidos distributes policies to their families, have them sign it and assign the corresponding beneficiary and provides information on the claims procedure Policies come into effect on September 1, 2012 for 12 months. The main results from this process were: Of the 50,755 potential beneficiaries 39,201 (77%) actually received the policy because of different reasons: Number % Not found 4,598 40% Already had insurance from 3,723 32% other program Unwilling to accept the policy 949 8% Other 2,284 20% TOTAL not delivered 11,554 RENEWALS Though there was no contractual obligation to renew the policies, it was in the stakeholder s and pilot program s interest to track the policy renewals and the operative process in this market. With sustainability as a key driver in all Banca de las Oportunidades interventions, the main challenge was to get the policies renewed to prove the clients willingness to pay the premium. The payment mechanism was thought to be guaranteed because the clients held savings accounts, but this assumption proved to be incorrect. And since the renewals were not initially stipulated in the contract, this process ended up being even more difficult than the initial stages. To renew the policy, clients had to pay 100% of the policy cost, as there were no longer any government subsidies. The premium s cost continued at USD $5 per year but, as an incentive to clients, Positiva added additional coverage for incapacity if the insured lost a limb, which became a source of confusion. Unlike the initial phase, Red Unidos field officers were not able to distribute the policies, as it was considered a commercial activity in favour of the insurance company, since the beneficiaries now had to pay for the service and it was not initially established in the tender process. Because all beneficiaries had at least one savings account (from Familias en Acción), it was thought that the premium could be easily 35

37 debited from the account, after receiving authorization from the client (signature), but in practice the payment turned out to be the most difficult part and the debit from the savings accounts posed a new challenge. For the renewals process, additional activities that were not included in the contract had to be carried out, which increased the cost to Positiva, including: Promotional campaign Motivation campaign through radio and printed material, to promote policy renewals, two months before the intial policy expired (BdO and Positiva, with Red Unidos support) Training Positiva trains Red Unidos field officers about renewals process Information to beneficiaries Red Unidos gives general information about the renewal process and invited the beneficiaries who were interested in renewing the policy to meetings at Positiva s offices. Debit authorization and policy renewals Positiva organized events/meetings to give general information about the renewals. Those interested in renewing the policies signed authorization to debit premium from the savings accounts, determined their beneficiaries, and received the policy Debit Positiva sends the debit authorizations to the paying bank (where the insured have their savings account) Bank debits the premium value from the accounts (when there is enough balance) The renewal process took longer than expected due to difficulties in reaching an agreement among stakeholders, the lack of consistency among guidelines within the government, planning the renewal process on the road, and defining the administrative and operative details. As a result, even though the initial policies expired in September 2013, the renewal process didn t started until late October and the insurance company had to assume the risks during those two months to avoid leaving the beneficiaries uncovered (there were 4 claims during that period). The insurance company also assumed additional costs for the renewals that were not contemplated in the contract, but decided to do so because of its commitment to the program s success seeing it as strategic for the company as a social responsibility issue. Of the 50,755 initially insured, 11,938 (30%) were interested in renewing the policy and signed the authorization for the insurance company to debit the premium from their savings accounts. Though it was thought that having the beneficiaries hold a savings account would solve the premium collection problems, in the end only 4,327 policies were actually renewed. This part of the process, which was thought to be the easier one, posed unforeseen difficulties that reduced the actual number of renewals like: Some beneficiaries wanted to pay in cash (and actually took cash to the meetings) but the only premium payment mechanism enabled was debit from the savings account. The insurance company only reached an agreement with one of the CCT paying banks, so those who received the CCT payment from the other bank were unable to renew the policy 36

38 When the bank tried to debit the premiums from the accounts of those who had authorized to do so, many did not have enough balance because they tend to withdraw everything as soon as they receive the payment (and the debits were not made immediately after the subsidies were disbursed) Some of the savings accounts were no longer active CONCLUSIONS AND FUTURE CHALLENGES Microinsurance design and implementation may be more challenging than other financial products Uncertainty and lack of information on potential market characteristics and risks limit the penetration for insurance companies, as well as their operational structure and lack of outreach. Reaching an agreement between the insurance companies and the government institutions was harder than expected. It was a long process of negotiations where both sides had to yield and learn. PPPs are key in promoting microinsurance The benefits of the partnership are evident for all sides. Insurance companies (Positiva and Fasecolda on behalf of others) recognize the advantages of leveraging existing government platforms to reach the families in extreme poverty and actually considered it indispensable to work with the government to reach the base of the pyramid. They acknowledge that the government programs that work with this population know them much better and are instrumental in spreading information, gaining confidence of the beneficiaries, and helping in the distribution process. For example, Red Unidos was key in the initial phase of the pilot and its absence posed challenges difficult to overcome in the renewal process. And, since the market is still underdeveloped and the insurance culture is missing, the incentives from the government continue to be required (they are unlikely to go to this market on their own). For the government, it is beneficial to offer a wide portfolio of adequate insurance products to the lowest income population to reduce their vulnerability, help them overcome poverty, and prevent them from falling deeper. From the beneficiaries perspective, they valued the insurance as shown by their willingness to renew the policy, amidst all challenges. As stated by the beneficiaries: We had funeral insurance and when my wife passed away, we only had to get the death certificate and they took care of everything. That (insurance) money has been very useful. First of all, I paid for some religious ceremonies in her memory, then I paid some debts, and invested the rest. If it weren t for that money, the situation would be very different. We would be struggling. - Poliduro Hernández, 68, Soacha. The cogestora told me that since we belonged to Red Unidos, we had received life insurance for my mom, even though she was already sick.(when she passed away ), the insurance was very helpful. I can t imagine what it would have been like if we didn t have the insurance. It helped pay for my daughter s education (which I owed) and to invest in my business. My sister bought a computer for her children. The insurance improved a bit of our life because each of us had a bit left and we are working it, which will help us progress - Sandra Yanuri Botina, 32, with three children 37

39 Pilot designs need to focus on the long run and product sustainability Though from Banca de las Oportunidades point of view sustainability is key, this particular design ended up sacrificing the long run perspective as a result of a long and cumbersome process of discussions with insurance companies. However, at the end of the day, from the perspective of the insurance companies, the pilot was short sighted, with government support for only one year, subsidizing 100% of the premiums (clients value the product less and has a weaker impact on the insurance culture) and not contemplating the renewal process. Complexities in the renewal process should not be underestimated It was assumed that having a savings account as a mechanism to collect premium payments was going to facilitate renewals, but in the end, this process ended up reducing renewals considerably, because: The premium payment procedure for renewals was underestimated: though all beneficiaries had savings accounts, there were still many challenges to make the payments that were overlooked. The lack of planning and foreseeing the renewals process operation ended up in improvisation that posed more challenges 38

40 IMPACT INSURANCE FACILITY Housed at the International Labour Organization, the Impact Insurance Facility enables the insurance industry, governments, and their partners to realise the potential of insurance for social and economic development. The Facility was launched in 2008 with generous support from the Bill & Melinda Gates Foundation, and has received subsequent funding from several donors, including the Z Zurich Re Foundation, the World Bank Group, USAID and AusAID. Facility impactinsurance@ilo.org