WPS8256 Policy Research Working Paper 8256 Insuring Well-being? Buyer’s Remorse and Peace of Mind Effects from Insurance Kibrom Tafere Christopher B. Barrett Erin Lentz Birhanu T. Ayana Development Research Group Agriculture and Rural Development Team November 2017 Policy Research Working Paper 8256 Abstract This paper estimates the causal effects of index insurance resolution of uncertainty. Insurance coverage currently in coverage on subjective well-being among livestock herders force generates subjective well-being gains that are signifi- in southern Ethiopia. The randomization of incentives to cantly higher than the buyer’s remorse effect of an insurance purchase index-based livestock insurance and three rounds that lapsed without paying out. Given the temporal cor- of panel data are exploited to separately identify ex ante relation in insurance purchase propensity, failure to control welfare gains from insurance that reduces risk exposure for potential buyer’s remorse effects can bias downward and ex post buyer’s remorse effects that may arise after the estimates of welfare gains from current insurance coverage. This paper is a product of the Agriculture and Rural Development Team, Development Research Group. It is part of a larger effort by the World Bank to provide open access to its research and make a contribution to development policy discussions around the world. Policy Research Working Papers are also posted on the Web at http://econ.worldbank.org. The authors may be contacted at ktafere@worldbank.org. The Policy Research Working Paper Series disseminates the findings of work in progress to encourage the exchange of ideas about development issues. An objective of the series is to get the findings out quickly, even if the presentations are less than fully polished. The papers carry the names of the authors and should be cited accordingly. The findings, interpretations, and conclusions expressed in this paper are entirely those of the authors. They do not necessarily represent the views of the International Bank for Reconstruction and Development/World Bank and its affiliated organizations, or those of the Executive Directors of the World Bank or the governments they represent. Produced by the Research Support Team Insuring Well-being? Buyer’s Remorse and Peace of Mind Effects from Insurance Kibrom Tafere* Christopher B. Barrett Erin Lentz Birhanu T. Ayana Keywords: Index insurance, subjective well-being, vignettes, pastoralists, Ethiopia JEL codes: D60, I31, I38, G22 * Corresponding author: Kibrom Tafere. Email: ktafere@worldbank.org Kibrom Tafere: Development Research Group, World Bank; Christopher B. Barrett: Charles H. Dyson School of Applied Economics and Management, and Department of Economics, Cornell University; Erin Lentz: Lyndon B. Johnson School of Public Affairs, University of Texas at Austin; Birhanu T. Ayana: International Livestock Research Institute, Addis Ababa, Ethiopia. This work was made possible, in part, by support provided by the US Agency for International Development (USAID) Agreement No. LAG-A-00-96-90016-00 through Broadening Access and Strengthening Input Market Systems Collaborative Research Support Program (BASIS AMA CRSP), the Department of Foreign Affairs and Trade through the Australia Development Research Awards Scheme under an award titled “The human and environmental impacts of migratory pastoralism in arid and semi-arid East Africa”, and CGIAR Research Programs on Climate Change, Agriculture and Food Security (CCAFS) and Dryland Systems. All views, interpretations, recommendations, and conclusions expressed in this paper are those of the authors and not necessarily those of the supporting or cooperating institutions. We thank Liz Bageant, Munenobu Ikegami, Andrew Mude, Megan Sheahan, Kazushi Takahashi, IBLI enumerators, and seminar participants at the University of Connecticut, Cornell University, University of Texas at Austin, the 2014 AAEA annual meetings, and NEUDC 2014 for helpful comments on earlier versions. 1 Introduction Uninsured risk exposure in low-income rural communities is widely believed to cause se- rious welfare losses and to distort behaviors, potentially even resulting in poverty traps (Rosenzweig & Binswanger, 1993; Morduch, 1994; Carter & Barrett, 2006; Dercon & Chris- tiaensen, 2011; Barrett & Carter, 2013; Santos & Barrett, 2016). However, standard insur- ance products are routinely unavailable due to moral hazard and adverse selection problems and high transaction costs in infrastructure-poor areas (Besley, 1995). In response to the lack of affordable standard insurance products, there has been a significant push to expand index insurance offerings in the developing world over the past decade.1 Index insurance attempts to mitigate adverse selection, moral hazard and high transaction cost concerns by writing contracts not on policyholders’ realized losses but, instead, on a low- cost, observable indicator – the ‘index’ – believed to be strongly correlated with actual losses. There is, however, little empirical evidence demonstrating that index insurance generates welfare gains for poor, rural households.2 Indeed, the low uptake of index insurance products in a range of countries suggests that perhaps many prospective buyers believe index insurance does not deliver welfare gains (Gin´ e, Townsend, & Vickery, 2008; Binswanger-Mkhize, 2012; Cole et al., 2013).3 Index insurance uptake may even cause welfare losses for buyers for at least two reasons. First, high commercial loadings by insurers can drive premium rates above actuarially fair levels. Second, when the index does not closely track policyholders’ actual losses, the imperfect correlation creates “basis risk” that can result in uninsured losses despite the purchase of insurance. This can lead to uninsured catastrophic loss despite a premium payment; as a result, index insurance will not stochastically dominate remaining uninsured (Jensen et al., 2016). 1 See Chantarat, Mude, Barrett, and Carter (2013) for an extensive discussion of these issues as they apply to a setting very similar to the one we study, and Miranda and Farrin (2012); Smith (2016) and Jensen and Barrett (2017) for broader reviews. 2 Janzen and Carter (2013); Karlan, Osei, Osei-Akoto, and Udry (2014) and Jensen, Barrett, and Mude (2016, 2017) are notable recent exceptions. 3 Gin´e et al. (2008) report that take-up rate of a rainfall insurance product in Andhra Pradesh, India was very low, at just 4.6 percent. They argue this might reflect the short history of the product. Similarly, Cole et al. (2013) find that the take-up rate of livestock insurance among the untreated general population in Andhra Pradesh and Gujarat, India, is close to zero. Binswanger-Mkhize (2012) argues that there is low demand for index insurance because better-off farmers have already self-insured through diversification of their portfolios and informal social networks, while the poor face liquidity constraints that limit their participation. Karlan et al. (2014), on the contrary, find that at an actuarially fair price, almost half of the farmers in their sample from northern Ghana demand index insurance and purchase coverage for more than 60 percent of their acreage. 2 Estimating the welfare effects of insurance coverage is complicated because insurance pro- duces two potentially opposite effects on the welfare of buyers. Holding insurance before the resolution of uncertainty generates ex ante well-being effects. Insurance may increase ex ante welfare for risk averse agents prior to the realization of stochastic events that may otherwise impose substantial losses. These ex ante well-being effects of insurance may differ from, and be partly offset by, the ex post well-being effects of lapsed insurance that did not pay any indemnity. Ex post effects arise after the resolution of uncertainty. The same insurance that is ex ante welfare improving may prove ex post welfare reducing, in a later period, once the risk has passed and a purchaser realizes with perfect hindsight that she could have foregone the premium payment without consequence. In this case, the buyer has “lost” her premium and would have been unambiguously better off financially had she not bought insurance coverage after all. If insurance purchase is positively correlated over time, this then raises the possibility that buyer’s remorse can confound valuation of insurance coverage, biasing downwards estimates of the value of current insurance coverage following periods without indemnity payments, when insurance purchase lost the insuree money. In this paper we take a novel approach to estimating the welfare impact of insurance on a poor, rural population, exploring whether index insurance coverage improves subjective well-being (SWB) and disentangling the potentially distinct effects of current and lapsed insurance coverage. The analysis of gains from insurance coverage has typically relied on either relatively weak tests of stochastic dominance or strong assumptions about utility functions (Williams, 1988; Feldman & Dowd, 1991; Halek & Eisenhauer, 2001; de Janvry, Dequiedt, & Sadoulet, 2014). Recent innovations in SWB measurement, however, permit relaxation of many of the strong assumptions on which such analyses rely. Further, measures of SWB often yield deeper insights beyond the traditional income and expenditure based well- being measures (Ravallion, 2012; Krueger & Stone, 2014). Indeed, conventional measures of well-being may underestimate the true value of a program. A program can have significant effects on SWB even if it does not generate observable material or physical impacts (Devoto, Duflo, Dupas, Parient´ e, & Pons, 2012; Finkelstein et al., 2012; Ludwig et al., 2013). As a result, SWB measures have become increasingly popular in welfare assessment (Frey & Stutzer, 2010; Clark, 2003; Fafchamps & Shilpi, 2008; Graham, 2012; Ravallion, Himelein, & Beegle, 2013; Kaminski, 2014; Krueger & Stone, 2014). Several features of our data enable us to estimate the ex ante and ex post SWB effects of index insurance. First, the project’s experimental design enables us to use an instrumental 3 variables method to overcome potential selection issues in index-based livestock insurance (IBLI) uptake. We exploit the randomization of incentives to purchase IBLI, newly intro- duced in southern Ethiopia by a commercial underwriter in August 2012. The novelty of the product obviates the potential confounding of past, unobserved experience with IBLI on buy- ers’ reported SWB. Second, three-round panel data enables us to control for time-invariant household unobservable characteristics that might affect both SWB and IBLI uptake. Third, no indemnity payouts occurred during this period.4 Without indemnity payments, we ex- ploit the considerable intertemporal variation in households’ IBLI uptake to isolate the causal effect of IBLI on SWB. We use coverage active during a survey round to capture ex ante welfare effects and coverage that had lapsed by the time of the survey to capture ex post impacts. These data offer an unprecedented opportunity to estimate the SWB effects of insurance that arise purely from ex ante risk reduction and to disentangle them from ex post buyer’s remorse effects. We find that current IBLI coverage improves SWB. Lapsed IBLI contracts that did not pay indemnities have a negative effect on SWB, consistent with the buyer’s remorse hypothesis. Although both effects are statistically significant, the welfare gains of current coverage sig- nificantly exceed the adverse buyer’s remorse effects. Our results are robust to a range of alternative estimators, corrections to address concerns on the measurement of SWB, variable definitions, model specifications and variations in the relevant panel sub-samples analyzed. Further, we show that the estimated SWB gains from insurance are downwardly biased if one omits control for lapsed insurance coverage that generates buyer’s remorse. The implication is that, despite premiums set above actuarially fair rates, IBLI improves buyers’ SWB even over a period when pastoralists in southern Ethiopia lose money on the policy. The ex ante peace of mind effect dominates any ex post buyer’s remorse. In other words, even an insurance policy that does not pay out still improves people’s perceptions of their well-being. The remainder of the paper is organized as follows. The next section presents the study setting and discusses IBLI and its contract design. Section 3 discusses the sampling and experimental design. Section 4 reports summary statistics of the data. Section 5 introduces our estimation strategy. Section 6 details our vignette correction strategy, following best current practice in the SWB literature. Section 7 reports our main results. Section 8 presents 4 The first IBLI indemnity payments – on 509 contracts yielding total payments of ETB 526,000 (approx- imately $26,225) – occurred in October-November 2014, after the period covered by our data. 4 a range of robustness checks. Section 9 concludes. 2 Study Setting and Research Design The study area is Borana zone of Oromia region in southern Ethiopia. It is a vast pastoralist land mass consisting mainly of arid and semi-arid agro-ecological zones with a bimodal rain- fall pattern and four distinct seasons: long rainy (March-May), long dry (June-September), short rainy (October-November), and short dry (December-February) seasons. Mobile pas- toralism is the primary source of income and sustenance, with limited cereals cultivation for own consumption. Cyclical movement of livestock in search of forage and water characterizes the livestock production system in the zone (Coppock, 1994; Berhanu, 2011). There are widespread concerns that more frequent droughts, perhaps associated with cli- mate change, are making pastoralism more tenuous (Barrett and Santos 2014). Catastrophic droughts in the 1980s and 1990s resulted in herd losses of over 35% (Desta & Coppock, 2002; Lybbert, Barrett, Desta, & Coppock, 2004). These catastrophic droughts, which are covari- ate within a community, also put pressure on informal social insurance mechanisms, such as iqub (rotating savings and credit associations (ROSCAs)) membership. Informal commu- nity networks facing high and widespread herd losses can no longer sufficiently mitigate the effects of shocks and are in decline (Lybbert et al., 2004; Santos & Barrett, 2011). Formal insurance might effectively transfer drought risk out of the pastoral system to underwriters, thereby cushioning pastoralists against catastrophic herd loss shocks. However, conventional indemnity insurance can be prohibitively costly to establish and sustain in this environment. Droughts that trigger payouts could bankrupt under-diversified insurers. Moral hazard and adverse selection problems and associated high monitoring costs, as well as high transaction costs in infrastructure-poor areas compound the challenges of delivering standard insurance products (Besley, 1995). IBLI was developed for precisely such an environment. Originally designed for and suc- cessfully piloted in the neighboring region of northern Kenya beginning in January 2010, IBLI makes indemnity payouts based on an observable, exogenous index of rangeland con- ditions, as reflected in Normalized Difference Vegetation Index (NDVI) measures generated by remote sensors on satellite platforms. An IBLI policy provides indemnity payouts when pasture vegetation falls below a contractually stipulated threshold level that reflects the on- 5 set of drought conditions that typically lead to excess livestock mortality (Chantarat et al., 2013). IBLI was piloted in 2012 in eight woredas 5 of Borana zone located directly across the border from the Kenyan region where IBLI first piloted. The index for IBLI Borana is calculated at the woreda level as a cumulative deviation of periodic NDVI readings for each IBLI sales period.6 Accordingly, the IBLI premium rate differs across woredas and by livestock species but is the same for all buyers insuring the same livestock species within a woreda, irrespective of individual loss experience. The woreda specific premium rates are applied to the value of herd that an IBLI buyer chooses to insure to establish the total amount that must paid for IBLI coverage. IBLI contracts are sold in two sales periods prior to the start of the short and long rainy seasons. The first IBLI contracts were sold in August-September 2012 (sales period 1). Contract sales were repeated in January-February 2013 (sales period 2), August-September 2013 (sales period 3) and January-February 2014 (sales period 4). The duration of contract coverage is 12 months. A contract sold in January 2014 covers March 2014-February 2015, while one sold in August 2013 covers October 2013-September 2014. Households can augment their coverage by acquiring new contracts in subsequent sales periods. Index readings for each sales period are announced and indemnity payments made to policyholders, if the contractually stipulated strike rate is triggered, at the end the season (See Figure 1). Figure 1 here As with all index insurance products, the substantial basis risk associated with IBLI could leave livestock loss uninsured due to imperfect correlation between the drought predicted by the index and losses experienced at the household level (Jensen et al., 2016). Animal losses due to covariate shocks that are not covered by IBLI, such as animal disease unrelated to rangeland conditions, as well as idiosyncratic shocks such as wildlife predation or injury, are common. Nonetheless, recent impact evaluations of the original IBLI pilot in northern Kenya find income and productivity gains, on average, for IBLI policyholders (Jensen et al., 2016, 5 Woreda is a third-level administrative division in Ethiopia, below region and zone. The eight woredas of Borana zone covered in our sample are Arero, Dhas, Dillo, Dire, Miyo, Moyale, Teltele, and Yabello. 6 For a more detailed discussion of the construction of the IBLI Borana index, see ILRI-IBLI (2013). 6 2017). But in that setting, significant indemnity payouts had occurred in the second year in which contracts were sold following the catastrophic 2011 regional drought, so average indemnity payouts substantially exceeded average premium expenses. Those results could, therefore, be purely the result of stochastic ordering of loss events and associated indemnity payments. Those indemnity payouts had sizable behavioral and welfare effects (Janzen & Carter, 2013). Because there were no indemnity payments in southern Ethiopia, our study isolates the welfare effects of insurance that arise purely from reduced ex ante risk exposure, that is, just the peace of mind effects that arise from buyers’ risk aversion, abstracted from the complication of indemnity payments. The Ethiopia IBLI pilot and associated data enable us to get at these important issues in a novel way that sheds considerable light more generally on the value of insurance coverage. 3 Data and Descriptive Statistics 3.1 Data and Study Design A baseline survey (R1) was designed and fielded in February-March 2012 before IBLI was developed or announced. Data on a broad range of household characteristics, livestock and other assets, livelihood activities, consumption, social networks, expectations and subjective well-being were collected. A year later, following sales period 2, a follow-up survey round (R2) of the original sample households was fielded in March-April 2013. Following sales period 4, a third round (R3) of survey data was then conducted in March 2014 from the same respondents as the first two survey rounds. We therefore have pre-experiment baseline data (R1), followed by two survey rounds (R2 and R3) with the same respondents. In R2, IBLI contracts purchased in sales periods 1 and 2 were in force. In R3 contracts from sales period 1 and 2 had lapsed but contracts purchased in sales periods 3 and 4 were in force (Figure 2). Figure 2 here The sampling was clustered at the reera level.7 Reeras were purposively selected based on geographic distribution, variation in market access, and agro-ecological variation across the 7 Reera is the fourth level administrative division in Oromia region below zone, woreda, and kebele. 7 eight woredas of Borana zone in our sample. Inaccessible reeras were excluded for logistical reasons. In each reera, households were grouped into three livestock holding classes (high, medium and low), measured in tropical livestock units (TLU).8 Fifteen percent of households were randomly selected in each reera such that a minimum of 25 households were selected with a balanced representation of the three TLU classes (terciles). In the event 15 percent of households in a reera yields less than 25 households, neighboring reeras were combined to form a bigger study site, resulting in a total of 17 study sites (ILRI, 2014). The baseline sample included 515 households. In R2, 476 of the original (baseline) households were re-interviewed. Households that had dropped out were replaced by households from the same study site and TLU class. If replacements could not be found in the same TLU class, households in the adjacent TLU class were picked. Thus, 32 new replacement households were surveyed from the original population lists for a total of 508 households in R2. In R3, 500 R2 households and 14 replacement households were surveyed. In selecting replacements in R3 priority was given to original households (those sampled in R1 but missed in R2). Of the 14 R3 replacements, 10 were original households and 4 were new households. Seven households had missing SWB measures or key independent variables and were dropped from the sample. The final estimation sample includes 550 unique households and 1,530 observations (515 in R1, 504 in R2 and 511 in R3), of which 465 households were surveyed in all three rounds, 50 households were surveyed in two rounds (8 in R1 and R2, 12 in R1 and R3, and 30 in R2 and R3), and 35 households were surveyed only once. A detailed treatment of potential attrition bias in the data and relevant corrections is presented in the Appendix. To encourage IBLI uptake, various combinations of premium discount coupons and informa- tion interventions through audio tapes of a poem or comic books were randomly implemented in each of IBLI sales period (Table 1). Information was delivered via caricature represen- tation of IBLI in comic books or audio tapes of a poem about IBLI recited in the local language, Oromifa, to sub-samples of respondents in sales period 1 and 2.9 The encourage- ment design in sales periods 3 and 4 did not include information intervention. All four sales 8 TLU is a measure used to aggregate livestock across species in relation to a common average metabolic weight such that 1 TLU = 1 cattle = 0.7 camels = 10 goats or sheep, collectively called ‘shoats’. 9 In the comic book information treatment, a randomly selected sub-sample of respondents was provided with a caricature representation of the IBLI product prepared by the underwriter, Oromia Insurance Com- pany (OIC). The contents of the material were first read to the sample households, then they were encouraged to look/read through it as many times as they wished. In the audio tape information treatment, development agents (DAs) were asked to play a tape that explains IBLI in Oromifa to a randomly selected sub-sample of respondents (for more details on the information interventions see ILRI (2014). 8 periods included randomized distribution of premium discount coupons. Prior to each sales period, all communities received a basic briefing that described the IBLI product. In each study site, 80 percent of respondents were randomly selected to receive discount coupons that would allow them to purchase IBLI at a discounted price for up to 15 TLUs. Discount coupon recipients were evenly distributed across discount levels of 10, 20, 30, 40, 50, 60, 70, and 80 percent. The remaining 20 percent of respondents did not receive discount coupons.10 The two information treatments – comic book and audio tape – were randomized in six sites each (in 12 of the 17 study sites, overall), with no overlap in assignment. Within the sites selected for information treatment, about 50 percent of respondents were randomly selected for treatment. In total, 20 percent of respondents received information treatment. The ran- domized assignment of respondents into information treatments and discount coupons with varying discount levels was implemented independently for each sales period. By creating exogenous variation in IBLI uptake and in the effective premium faced by prospective buyers, IBLI’s randomized encouragement design allows a rigorous analysis of the causal impacts of IBLI on SWB. All sample households in our study sites had opportunities to insure against drought-related livestock loss. Yet, only 22 percent and 21 percent of households surveyed in R2 and R3, respectively, reported buying IBLI coverage. In both R2 and R3, IBLI purchases were lower in the January-February sales period than in the August-September sales period. Of the 504 households surveyed in R2, 130 purchased IBLI in sales period 1 and 94 in sales period 2. Similarly, of the 514 households surveyed in R3, 150 purchased IBLI in sales period 3, but only 62 in sales period 4. This difference might arise due to seasonality in household liquidity.11 Or this may simply reflect the seasonality arising due to the initial launch of IBLI in August-September 2012, combined with the contracts’ 12 month duration. Because IBLI contracts cover a full year but policies are sold in two sales periods each year, households can augment their coverage or allow contracts to lapse. Of the 130 IBLI buyers in sales period 1, 23 buyers augmented coverage further by buying additional policies in sales period 2, 53 allowed their policy to lapse after a year, and 77 extended their coverage in sales 10 As part of a separate project, however, 10 respondents received IBLI coverage for up to 15 TLUs free of charge (100 percent discount) in each sales period. 11 Extended dry conditions often lead to stress sales and collapse of livestock markets, which in turn limits ability to raise the necessary liquidity to insure against shocks (Barrett, Chabari, Bailey, Little, & Coppock, 2003; Lybbert et al., 2004). 9 period 3. The remaining 71 buyers in sales period 2 were first time buyers. Likewise, 73 of the 94 IBLI buyers in sales period 2 allowed their contracts to lapse and 21 renewed their contract in sales period 4. Among the 150 households who bought IBLI policies in sales period 3, 33 households bought additional coverage in sales period 4. The considerable intertemporal variation in households’ IBLI coverage, combined with the experimental design behind the IBLI pilot, enable us to disentangle the causal effects of current and lapsed insurance policies on respondents’ SWB. 4 Summary statistics Table 1 reports baseline treatment-control covariate balance tests on assignment to pre- mium discount coupon in sales periods 1 and 2. There is very little pre-treatment difference in subjective well-being, wealth, expected livestock loss, various household characteristics, and group membership between those who purchased insurance and those who did not, con- firming that the randomization was successful.12 Detailed variable definitions are provided in Appendix Table A2. To complement these results, we also conducted formal joint or- thogonality tests and found that selection into treatment is uncorrelated with observable household characteristics (Appendix Table A3). Joint significance tests from pooled OLS (linear probability model) regression of treatment dummies (discount coupon, audio tape and comic book) for the August-September and January-February sales periods on house- hold income, livestock and non-livestock assets, expectations of future rangeland conditions, and various individual and household characteristics suggest that treatments are randomly assigned. We cannot reject the joint null of zero partial correlation of all covariates in these regressions. Apart from the discount coupon regression in the August-September sales pe- riod, pre-treatment differences in covariates between treatment and control households are statistically insignificant in almost all cases. Table 1 here Table 2 reports summary statistics on key dependent and independent variables by insurance status.13 The top four rows show that households who had IBLI coverage in R2 and/or R3 12 Covariate balance tests on comic book and poet audio tape information treatments and discount coupon receipts in sales periods 3 and 4 also show that treatment assignment was indeed random. Findings are available upon request. 13 Table 2 presents the averages of the variables in R2 and R3, during which IBLI was available for purchase. 10 report higher SWB – by any of the four different measures discussed in the next section – compared to their counterparts who have had no IBLI coverage in any of the survey rounds. Rows 5-9 show that IBLI purchase is strongly positively correlated with the discount coupon and information treatments. In each sales period, about 93 percent of IBLI contract holders had received discount coupons.14 Similarly, households who received information treatments (comic book or audio tape) were more likely to buy IBLI. As expected, higher discount rates are strongly correlated with IBLI uptake. These simple descriptive statistics suggest that the random, exogenous assignment of discount coupons and information treatments are suitable predictors of IBLI adoption. Table 2 here Insured and uninsured households are not distinguishable by observable characteristics, apart from number of TLU owned, which is weakly statistically significant. The value of non- livestock assets, annual income, expected livestock loss, gender and age of household head, household size and composition, and membership in iqub groups vary insignificantly be- tween those that purchased insurance and those who did not. These findings on observable characteristics do not rule out potential differences based on unobservable characteristics. However, so long as such unobservable differences are time invariant, we can control for them using a fixed effects estimator. Concerns that time varying characteristics may determine IBLI adoption nonetheless remain. We exploit the random assignment of discount coupon and information treatments, each strongly correlated with IBLI uptake, to address these concerns. 5 Estimation strategy A key challenge in evaluating policy interventions where respondents can voluntarily “opt- in” is that selection into the program may not be random. Rather, participation could be systematically correlated with respondents observable and unobservable characteristics. Peoples’ SWB is likely correlated with their subjective assessment of risk, their planning horizons, and other unobserved factors that influence insurance uptake. The experimental 14 Since survey rounds 2 and 3 were preceded by two sales periods each, a household who purchased IBLI in sales period 2 but had received discount coupon in sales period 1 is reported to have received discount coupon for the survey round, hence the slightly higher figures in Table 2. 11 design features of IBLI’s impact evaluation, including randomized exposure to various in- formation treatments and randomized distribution of premium discount coupons, allow us to address the selection bias associated with insurance uptake choices. We first estimate selection into IBLI using randomized encouragement treatments as instruments. We then estimate the effect of instrumented IBLI on SWB. This approach allows us to derive unbiased and consistent causal estimates of IBLI’s impact on SWB. IBLI uptake by household i in village v , sales period s, and survey round t is estimated using the linear probability model (LPM)15 as: P r(IBLIivt = 1) = ω + γs Divst + φs Aivst + µs Civst + ηs Pivst + ζXivt + κt + τi + εivt (1) The randomly assigned treatments include dummy variables for receiving a randomly as- signed premium discount coupon (D) in the first sales period (August-September 2012), the second sales period (January-February 2013), or both; dummy variables for receiving ran- domly assigned extension treatments in either audio tape (A) or comic book (C ) form in the first, the second or both sales periods, and a woreda specific continuous measure of the ran- domly discounted IBLI premium rate (P ) in the first and second sales periods. These are all randomly assigned to households and should have no direct effect on SWB, only an indirect effect through their impact on inducing IBLI uptake. The lone possible exception is P , since price variation has a (very modest) real income effect conditional on someone purchasing IBLI and thus could plausibly have some direct effect on SWB. A series of covariates, X , that may influence the uptake of IBLI are included as controls, including household herd size and income, expectation of livestock death, gender, age and educational attainment of household head and household composition. Household fixed effects (FE), τ , which control for, among other things, time invariant optimism or pessimism of individual respondents and survey round fixed effects, κ, are also included. We use the randomized coupon distribution and information treatments to instrument for the purchase of IBLI coverage in the first stage estimation. When applied to R2 data, equation (1) predicts current uptake, IBLI iv2 , based on purchases in sales periods 1 and 2. There were no lapsed contracts in R2. When applied to R3 data, it predicts current uptake, IBLI iv3 , based on purchases in sales periods 3 and 4. We use the IBLI iv2 predicted value 15 To avoid the “forbidden regression” problem associated with non-linear models such as logit or probit, we use an LPM to predict an endogenous dichotomous variable in the first stage of an instrumental variables (IV) regression (Angrist & Pischke, 2008; Wooldridge, 2010). 12 to capture lapsed contracts in R3. In the second stage of our estimation, the predicted IBLI coverage is used to estimate the causal effect of IBLI on SWB in the second stage of our estimation. SWB includes ordinal responses to the question “on which step do you place your current economic condition,” ranging from 1 (very bad) to 5 (very good). The construction of our SWB measure and related robustness checks are discussed in more detail below. The second stage ordered logit regression includes predicted IBLI uptake, number of TLUs owned (TLUO), predicted lapsed IBLI uptake the probability of having acquired an IBLI contract that has lapsed (IBLIL), a series of controls X , household fixed effects χ, and survey round fixed effect λ. SW Bivt = α + β IBLI ivt + θT LU Oivt + σ IBLILivt + δXivt + λt + χi + ivt (2) The coefficient estimate on predicted IBLI uptake, β measures the effect of IBLI coverage on the extensive margin – the ordered log-odds estimate of possessing IBLI contract(s) on SWB. We expect that effect to be positive, reflecting the welfare gains from insurance in a risky setting. The coefficient estimate on IBLILivt , σ measures the effect on SWB of an IBLI contract that was in force in R2 but had lapsed in R3. Since contracts in force are controlled for, this coefficient estimate isolates the ex post SWB effect of insurance that did σ < 0). not pay, i.e., buyer’s remorse, and it is expected to be negative (ˆ A finding that βˆ > |σ ˆ | indicates that even if insurance does not pay out, in expectation, the positive peace of mind effect exceeds the negative buyer’s remorse effect, and hence IBLI improves expected welfare. If policy purchases – and therefore current and lapsed policies – are correlated over time, failure to include lapsed contracts in equation (2) would lead to omitted relevant variable bias of the β estimate, presumably downwards due to negative buyer’s remorse effects. To capture the intensive margin of IBLI coverage, i.e., the marginal effect of increasing the volume of IBLI uptake by a unit, we re-estimate equation (1) replacing the IBLI uptake dummy variable with volume of TLUs insured (TLUI). The first stage equation for the negative censored continuous variable TLUI is estimated using Tobit as: ˜ +γ T LU Iivt = ω ˜s Aivst + µ ˜s Divst + φ ˜s Pivst + ζX ˜s Civst + η ˜ ivt + κ ˜i + ε ˜t + τ ˜ivt (3) We construct predicted values for current and lapsed IBLI coverage using the same approach 13 as we did for the discrete uptake variable earlier. The second stage ordered logit regression then includes predicted TLU insured and predicted lapsed TLU insured instead of predicted IBLI uptake to identify the causal effect of buying an additional TLU of IBLI coverage on SWB. SW Bivt = α ˜ LU Oivt + σ ˜T LU I ivt + θT ˜+β ˜ ivt + λ ˜ T LU Livt + δX ˜t + χ ˜i + ˜ivt (4) The second stage regression equations of both IBLI uptake (equation 2) and quantity of TLU insured (equation 4) include generated regressors. Conventional standard errors of the estimated coefficients would be biased downwards. To account for the lower variation in the predicted uptake and volume of TLUs insured, we estimate the standard errors using a bootstrapping method where both the first and second stage are included for every bootstrap sample. Further, to account for spatial correlation of observations estimated standard errors are clustered at the village (reera ) level in all regressions. There are at least two possible mechanisms through which IBLI coverage could influence SWB. The first effect is the gross non-monetary benefits or costs associated with coverage, represented by the coefficient estimate on the instrumented IBLI, β ˆ, net of instrumented lapsed IBLI, σˆ , (βˆ+ˆ σ ), or the coefficient estimate on instrumented TLU insured, β ˆ ˜, multiplied by the number of TLUs insured net of the coefficient on instrumented lapsed TLU insured, ˆ ˜ , multiplied by the number of lapsed TLUs insured, (β σ ˆ ˜ × T LU I t + σ ˆ ˜ × T LU Lt ). Purchasing insurance may reduce stress about possible adverse outcomes, which could lead to higher levels of SWB (β ˆ > 0), while greater coverage may lead to higher SWB (β ˆ ˜ > 0). Conversely, if the basis risk on the product is high such that IBLI uptake is more like a lottery ticket than a conventional indemnity insurance policy, IBLI uptake could increase stress and reduce SWB (β ˆ < 0). For the same reason, greater IBLI coverage may cause higher stress and lower SWB (β ˆ ˜ < 0). The second influence on SWB arises from the net monetary benefit or cost of IBLI coverage on SWB. If net income or wealth influences SWB, as many studies suggest (Frey & Stutzer, 2010; Graham, 2012), then IBLI will also affect SWB through the premium amount paid for IBLI, which reduces net income or wealth, and any indemnity payment received in the event that the IBLI policy pays out, which increases net income or wealth, ceteris paribus. This effect is captured by the coefficient estimate on the number of TLUs owned, θ ˆ ˜, multiplied by the net flow of funds associated with the period-specific net indemnity payments (indemnity receipts 14 minus premium payments) associated with the predicted IBLI uptake volume, converted into TLU units at prevailing livestock prices, NI.16 We therefore estimate the aggregate effect of IBLI on SWB as: ˆ ˜ × T LU I ivt + σ ∆SW B ivt = β/s ˆ ˆ ˜ × N Iivt ˜ /s × T LU Livt + θ/s (5) The point estimate β ˆ ˜ in equation (5) reflects the SWB benefit of a unit of free IBLI with no indemnity payment. Likewise, the coefficient estimate σ ˆ ˜ measures the SWB loss due to a unit of free IBLI that has expired without payout. Note, however, that β ˆ ˜, σ ˆ ˜ and θˆ ˜ measure effects on SWB in log-odds scales while SWB is measured in ordinal Likert scale. It is necessary to harmonize the units in which these coefficients and SWB are measured before one can calculate the overall effect of IBLI on SWB. We use the fact that the logistic and Normal distributions are similar, except at the tails of the distribution, to convert the coefficients from log-odds units to Normal equivalent deviates. The effects measured in log- odds and their corresponding standard errors can be converted to approximate effects in Normal equivalent deviates by dividing by the standard deviation of the logistic distribution √ s = π/ 3 (Hasselblad & Hedges, 1995; Chinn, 2000). Given that during R2 and R3 there were no indemnity payments but respondents paid for IBLI, our estimates provide a lower bound, reflecting the SWB associated with insurance coverage in the absence of any payout, i.e., a period in which insurance represents an unam- biguous financial loss. A finding that ∆SW B ivt > 0|N Iivt < 0 would therefore represent a strong finding with respect to the welfare effects of index insurance in this setting.17 6 SWB and vignette correction Subjective measures of welfare are becoming increasingly popular but pose methodological challenges (Krueger & Schkade, 2008; Ravallion, 2012). Respondents may have different reference points when answering a subjective question, making interpersonal comparisons TLU−Premium per TLU 16 N I = Indemnity per Price per TLU × T LU I is the TLU equivalent wealth gained or lost due to IBLI purchase. 17 Estimates for ∆SW B are obtained by evaluating equation (5) at the average TLUs insured and NI. The price per TLU is obtained by weighting livestock prices from Haro Bake livestock market (the largest livestock market in Borana zone) with the TLU conversion units of each species (Table A1). 15 problematic. To address any latent heterogeneity problems that might hinder interpersonal comparisons of subjective welfare, we adjust the subjective measures of well-being using hypothetical vignettes that provide an explicitly standardized reference point for all respon- dents’ comparisons in order to bring objective and subjective assessments into alignment (van Soest, Delaney, Harmon, Kapteyn, & Smith, 2011; Krueger & Stone, 2014).18 Interpersonal comparisons using SWB data can be challenging due to potential unobserved heterogeneity in respondents’ reference points, which may depend on socio-economic con- ditions, and other observable and unobservable characteristics. Such latent heterogeneity in subjective well-being measures may render interpersonal comparisons meaningless and invalidate inference from subjective welfare regressions (King, Murray, Salomon, & Tandon, 2004; van Soest et al., 2011; Beegle, Himelein, & Ravallion, 2012; Ravallion et al., 2013). King et al. (2004), King and Wand (2007), van Soest et al. (2011) and Beegle et al. (2012) suggest an approach for correcting latent heterogeneity problems that involves measuring the interpersonal incomparability of responses itself. Respondents are asked to assess their own circumstances relative to a set of hypothetical individuals described by short vignettes on the same scale. Responses to the hypothetical vignettes are then used to construct an interpersonally comparable welfare measure as respondents’ reference points have been exogenously standardized. The validity of this approach relies on two key assumptions: response consistency, and vignette equivalence. Response consistency requires that each respondent use response categories for a particular concept in the same way when self- assessing as when assessing hypothetical individuals. Vignette equivalence is the assumption that each respondent perceives the level of the variable represented by a particular vignette on the same unidimensional scale. That is, the variable being measured by vignettes should have a consistent meaning among respondents (King et al., 2004). Following King et al. (2004), the reported SWB measures are corrected using a simple non- parametric approach. For notational ease, we momentarily suppress the village and time dimensions of the data. Suppose SW Bi is the categorical self-assessment for respondent i(i = 1, ..., n), and Vij is the categorical survey response for respondent i on vignette j (j = 1, ..., J ). For respondents with identical vignette ordering (i.e. Vi,j −1 < Vij ) the vignette adjusted measure of subjective well-being is given as:19 18 As discussed further below, we test the robustness of our core results by re-estimating our model for direct (unadjusted) SWB responses and for responses to a similar SWB question that asks people about their well-being relative to other Borana pastoralists. The core findings prove stable. 19 In our data, rescaling of self-assessments relative to vignettes does not generate vector responses, which 16     1 if SW Bi < Vi1      2 if SW Bi = Vi1   3 if Vi1 < SW Bi < Vi2 V SW Bi = (6)    . .      . .   2J + 1 if SW Bi > ViJ The hypothetical vignettes used in this study involve households that fall in one of three well-being rungs: low, middle and high, which were constructed in consultation with local field researchers knowledgeable about the local socio-economic conditions in the study area. The lowest (poor), middle, and highest (rich) rungs were represented by a family that “has no livestock and does not eat meat except on special occasions,” a family that “has a dozen of shoats [goat and sheep], but no camel or cattle and can eat meat only once a month,” and a family that“has a lot of shoats and several camels and cattle and can eat meat whenever they choose,” respectively. The cross tabulation of SWB measures and vignette corrected SWB measures in Appendix Table A4 shows that vignette corrected SWB measures largely mirror SWB, particularly at the lower end of the scale. For example, in panel (a), out of the 120 observations with SWB score of one (very bad), 27 are rescaled to one and 93 to two on the vignette adjusted SWB. Similarly, of the 88 observations with SWB scores of five (very good), none is rescaled one, and only five to two on the vignette adjusted SWB. We observe similar correspondence between SWB relative to Borana pastoralists and its vignette corrected equivalence in panel (b). To test the robustness of our results to potentially unstable responses, we re-estimate the model using alternative SWB measures vignette corrected SWB relative to Borana pas- toralists and SWB relative to Borana pastoralists. The SWB relative to Borana pastoralists variable is similar to the SWB measure, but respondents are asked to gauge their life relative to other Borana pastoralists. The anchoring of subjective well-being questions reduces the likelihood that respondents may have different reference groups in mind when responding (Ravallion, 2012). are associated with inconsistent vignette ordering or correspondence of self-assessment with more than one vignette responses. As a result, the standard class of econometric methods for ordered dependent variables is suitable for our analysis. 17 7 Results We first discuss the estimated vignette-corrected SWB effects of IBLI on the extensive mar- gin, followed by discussion of results on the intensive margin. Table 3 presents the first stage panel fixed effects LPM estimates of equation (1) (columns 1-2) and panel random-effects (RE) Tobit model of equation (3) (columns 3-4). Column 1 shows results from a basic model with just randomized discount coupon and audio tape and comic book information extension treatments in sales periods 1 and 2. In column 2, in addition to the randomized discount coupon and information treatments in column 1, we include a broad range of household char- acteristics, wealth measures, IBLI knowledge, expectations of livestock loss, membership in iqub, and survey round fixed effects. The parameter estimates of both models show that randomized treatments had positive effects on IBLI uptake and, thus, can serve as suitable instruments. Receiving a discount coupon and the amount of the discount were especially strong predictors of IBLI uptake. Receiving a discount coupon in sales period 1 increases the probability of buying IBLI policy by over 20 percent. This effect is even greater for the discount coupon in sales period 2 – it increases the odds of buying IBLI by about 24 percent. Moreover, having received discount coupons in sales period 1 increases the prob- ability of buying coverage for recipients of discount coupons in sales period 2. Besides the price effect of discount coupons, which is captured by the coefficient estimates on discount values, the discount coupon had informational value, offering holders a physical reminder of the insurance product. Conditional on the amount of discount received and other covariates, receiving a discount coupon had an independent positive effect on IBLI uptake. Table 3 here Randomized provision of audio tape and comic book information treatments also had a positive, albeit weaker, effect on IBLI uptake. The audio tape treatment had a positive and statistically significant effect in sales period 2. The comic book treatment, however, had an effect on IBLI uptake only when offered in both sales periods, suggesting the effectiveness of repeated exposure to this informational approach. Both Sargan (χ2 (24) = 75.36, prob > χ2 = 0.000) and Basmann (χ2 (24) = 79.17, prob > χ2 = 0.000) over-identification tests fail to reject the null hypothesis that our instruments are valid. The Wald test for joint significance of all instruments also strongly rejects the null of jointly insignificant instruments (χ2 (9) = 137.8, prob > χ2 = 0.000). Thus, this first stage appears to successfully instrument for endogenous IBLI uptake. 18 IBLI uptake relates to our control variables in the expected ways. Uptake is positively correlated with knowledge about IBLI and wealth (livestock and non-livestock assets), but only number of TLUs owned is statistically significant. Income and iqub membership are negatively but statistically insignificantly correlated with IBLI uptake. The later suggests that iqub may crowd out IBLI. We also find that male headed households are less likely to buy IBLI and that larger households are more likely to buy IBLI.20 We find similar results when estimating a Tobit model for volume of TLUs insured to study IBLI uptake at the intensive margin (columns 3-4, Table 3). Receiving discount coupons and the size of the discount carried by the coupon are strong predictors of the volume of TLUs insured. The audio and comic book information treatments were also found to be positively, but relatively weakly, related to the volume of IBLI coverage. The number of TLUs owned is positively related to volume of coverage. In line with the IBLI uptake results in columns 1 and 2, we find that IBLI knowledge influences the volume of uptake. Respondents with more correct answers to questions about the particulars of the IBLI contract are more likely to buy IBLI, a result consistent with ambiguity aversion (Gharad, 2013). Iqub membership reduces the volume of TLUs insured, as such traditional institutions lower the demand for other forms of insurance. Table 4 reports second stage ordered logit regression results of the effects of IBLI on vignette corrected SWB. Panel (a) shows the effects of IBLI in log-odds units. While these results are concise and more convenient for presentation purposes, their interpretation may not be straight forward. In panel (b), we present the corresponding marginal effects of the main results in panel (a). Columns 1-3 show the extensive margin effects of IBLI uptake on SWB. Since randomized discount coupon and information treatments were used as instruments for the potentially endogenous IBLI uptake in stage one, the coefficient on IBLI measures the causal effect of IBLI on SWB. We find that IBLI has a strong positive effect on SWB, presumably because insurance coverage reduces risk exposure for risk averse buyers. The full model in column 3 shows that IBLI uptake increases the log-odds of reporting higher SWB by 0.86. That is, IBLI buyers are 2.4 (≈ e0.86 ) times more likely to report higher SWB than lower SWB. The probability estimates in panel (b) make this point more clear. IBLI reduces the probability of reporting lower SWB (SW B ≤ 3) by 11 percent and increases the probability of reporting higher SWB (SW B ≥ 5) by 11 percent. Our results are robust to 20 As a robustness check, we also estimate a probit selection model. The results are strongly consistent with the LPM (Table A5). 19 the inclusion of income, wealth, a range of demographic and household characteristics, and household composition variables. Table 4 here At the time of the R3 survey implementation, IBLI policies from sales period 1 and sales period 2 had already lapsed without payout. Thus, the coefficient estimate on IBLIL captures the negative ex post SWB effect of having bought an insurance policy that did not pay out. Indeed, the negative and statistically significant coefficient estimate on IBLIL indicates buyer’s remorse. Having bought an IBLI contract that lapsed without pay out reduces the log-odds of reporting high SWB by 0.42, which indicates that buyers of lapsed IBLI contracts are 1.5 (≈ 1/e−0.44 ) times more likely to report lower SWB than higher SWB. In probability units, having bought a lapsed IBLI contract increases the probability of reporting low SWB (SW B ≤ 3) by 5 percent and decreases the probability of reporting high SWB (SW B ≥ 5) by 5 percent. More importantly, the magnitude of the IBLIL coefficient is statistically significantly smaller than that of IBLI . This suggests that people are comforted by insurance coverage, and the positive ex ante effect trumps the negative ex post regret they feel once they realize that they paid for insurance that, in retrospect, they did not ultimately need. As expected, SWB is positively correlated with various wealth measures. Both livestock and non-livestock assets are positively related to SWB. Male headed households are more likely to report higher SWB than their female headed counterparts. Household size is negatively correlated with SWB. We find similar results for the volume of TLUs insured (columns 4-6). Vignette corrected SWB is increasing in the predicted number of TLUs insured. In the full model in column 6, a unit increase in the volume of TLUs insured increases the log-odds of reporting higher SWB by 0.14, which translates to 1.15 times more likelihood of reporting higher SWB than lower SWB. The corresponding column in panel (b) shows that an additional unit of TLUs insured reduces the probability of reporting low SWB (SW B ≤ 3) by 2 percent and increases the probability of reporting high SWB (SW B ≥ 5) by 2 percent. Yet, as IBLI policies lapse without paying, the more TLUs one had insured, the greater the buyer’s remorse one experiences. A unit increase in lapsed TLUs insured reduces the log-odds of reporting higher SWB by 0.07. An IBLI buyer with a unit more lapsed TLUs insured is 1.08 times more likely 20 to report lower SWB than higher SWB. That is, an additional unit of lapsed TLUs insured increases the probability of reporting low SWB (SW B ≤ 3) by 0.9 percent and reduces the probability of reporting high SWB (SW B ≥ 5) by 0.8 percent. As is the case with IBLI uptake, the positive effect of greater volume of TLUs insured statistically significantly exceeds the negative remorse it causes when the contract fails to pay out. We also find that livestock and non-livestock wealth are positively correlated with SWB, while household size is negatively correlated with SWB. Appendix Table A9 presents the regression results from estimating equations (2) and (4) using only currently active IBLI policies. Omission of IBLIL leads to a considerably smaller, yet still statistically significant, point estimate on IBLI . This finding underscores the prospective omitted relevant variable bias on the ex ante SWB impact estimate that arises due to autocorrelation in insurance demand if one does not separately control for lapsed policies. In other words, econometric estimates of the gains from insurance will likely underreport the welfare effects of insurance coverage if the research design does not permit the researcher to disentangle the ex ante and ex post effects of insurance. The net aggregate SWB effect of IBLI is presented in Table 5. The estimated ∆SW B is positive and statistically significant in the number of TLU insured. The point estimate sug- gests that insuring an extra TLU increases vignette corrected SWB by 0.2 points, although these units have no specific informational content given the ordinal nature of the dependent variable. But this magnitude indicates that, assuming a constant marginal SWB effect of IBLI, insuring about five TLUs bumps a household up by one rung on the SWB Likert scale, from, for example, “very bad” to “bad” or “good” to “very good”, on average. So, even in- surance policies that did not pay out generate SWB gains. Given the actual financial losses experienced by households that purchased insurance policies in these poor communities in southern Ethiopia, this finding is important and reassuring. Table 5 here 8 Robustness checks We complete several robustness checks to test whether our findings are sensitive to various specifications and variable definitions. First, we re-estimate our model for vignette corrected 21 SWB relative to Borana pastoralists, a refinement of our dependent variable (Appendix Table A6). The results are consistent with our main findings, suggesting response stability – that the phrasing of questions had little impact. As in the model for vignette adjusted SWB in Table 4, buying IBLI leads to higher SWB scores. The estimated coefficients on predicted IBLI as well as lapsed IBLI in the two models are comparable. As expected, the coefficients on predicted lapsed IBLI are negative and statistically significant. But, the positive effect of possessing IBLI policies is significantly higher than the negative buyer’s remorse effect. The number of TLUs owned is positively related to SWB. As before, greater household size is associated with lower SWB. Non-livestock assets and gender are, however, statistically insignificant. The results of the regression of SWB relative to Borana pastoralists on the volume of TLUs insured are also consistent with our main results in Table 4. The positive effect of active contracts exceeds the negative buyer’s remorse effect of lapsed coverage. The difference is statistically significant. We then estimate our model using raw SWB, which has not been vignette corrected, for IBLI uptake and volume of TLUs insured (Table A7). The results are consistent with our main findings – SWB increases with IBLI uptake/ volume of TLUs insured, and livestock and non- livestock wealth. Lapsed IBLI contracts cause remorse, hence negatively impact well-being. Male household heads are more likely to report higher SWB than female household heads. However, the coefficients on predicted IBLI and predicted TLUs insured are not statistically different from the absolute value of the corresponding coefficients on lapsed predicted IBLI uptake and predicted TLUs insured. We also estimate our model for the balanced panel subsample to verify that the differential weighting of households in the unbalanced panel sample does not influence our estimates. Results for the balanced panel household sub-sample are presented in Tables A8. Again, we find that all of the estimated coefficients are consistent with our main results in Table 4. Predicted IBLI coverage and TLU insured increase vignette adjusted SWB, while lapsed contracts reduce it. The magnitudes of the positive effects of IBLI remain significantly higher than the negative estimated buyer’s remorse effects, and comparable to what we find in Table 4. As before, SWB rises with wealth and decreases with household size. Male household heads report higher SWB. The multiple robustness checks we conduct strongly suggest that the positive ex ante SWB effects of IBLI coverage, and the negative ex post SWB effects of buyer’s remorse in response to a lapsed policy that did not pay out, are robust to both definitions of subjective well-being 22 measures, various specifications, and variations in the relevant panel sub-sample. The effects of wealth, gender and household size are also consistent throughout. These results give us more confidence in the robustness of our results. 9 Conclusions Interest in the study of subjective well-being (SWB) has increased in recent years, as has research on index insurance in rural areas of the developing world. To date, much of the SWB research in low-income countries has focused on the relationship between SWB and income or assets. There is limited understanding of how institutional factors, access to services, or policy-related issues influence SWB, if at all (Fafchamps & Shilpi, 2008). Furthermore, few studies link policy-related variables, such as uptake of index based livestock insurance (IBLI), with changes in SWB (Kaminski, 2014). This study addresses that important gap in the literature while simultaneously making an important contribution to disentangling the ex ante and ex post welfare effects of insurance by isolating the buyer’s remorse effect that arises from lapsed insurance policies. Empirically, we demonstrate that index insurance such as IBLI, which reduces the drought-related risk faced by pastoralists in southern Ethiopia has the potential to impact not only material well-being – e.g., by replacing lost assets and reducing adverse coping behaviors – but also to improve non-material well-being, by providing valuable peace of mind for risk averse buyers even if they can reasonably anticipate experiencing buyer’s remorse if a policy lapses without payout. We use three rounds of annual household panel data collected between 2012 and 2014, bracketing the introduction of IBLI in southern Ethiopia, and randomized encouragements to buy the product to identify the causal effect of IBLI on SWB. We separate out the ex ante SWB effects of current coverage from the ex post buyer’s remorse effect, exploiting the fact that some households had purchased IBLI in the second survey round and those policies had lapsed by the third survey round. We also show that if buyer’s remorse effects exist and there is any persistence in insurance purchases, such that current and lapsed coverage are positively correlated, then ignoring lapsed policies results in downwardly biased estimates of the well-being effects of insurance. We find that current IBLI coverage has a strongly positive and statistically significant effect on SWB. We also find statistically significant evidence of a buyer’s remorse effect. The 23 negative buyer’s remorse effect is considerably smaller in magnitude than the positive effect of IBLI coverage, however, suggesting that the comfort people derive from insurance coverage more than compensates for any regret they suffer once they realize they did not need coverage. Therefore, in our survey sample, insurance purchase is ex ante optimal, on average. This could reflect the nature of the sample we study. Pastoralists in southern Ethiopia’s Borana Zone face decline in informal social insurance institutions at a time when pastoral livelihoods are becoming more risky. As a result, Borana pastoralists may experience greater well-being as a result of having access to index insurance, even if it did not pay out in the short-term. These results suggest that for people with precarious livelihoods, even an imperfect, commercially priced insurance policy that does not pay out can leave them feeling better off. Our findings also show that estimations of the welfare effects of insurance ought not ignore potential ex post impacts. Prior purchases of insurance may induce buyer’s remorse once a buyer realizes that, in retrospect, costly insurance proved unnecessary. Survey-based SWB measures can capture all of these prospective effects without resorting to strong assumptions about the arguments and functional form of utility functions. SWB measures seem especially appropriate to establishing the impacts of commercial insur- ance. Commercial insurance policies, including IBLI, intrinsically involve a tradeoff between material and non-material well-being if policies are priced above actuarially fair premium rates so as to cover the costs of and ensure a profit margin for the underwriter. Theory suggests that actuarially fair insurance is welfare enhancing, regardless of whether it pays out, because most people are risk averse and insurance mitigates risk. But when insurance is not actuarially fair, and perhaps especially if it offers incomplete coverage, as is inevitably the case with index insurance products subject to basis risk, the ex ante expected monetary loss (because premiums exceed expected indemnity payments over time) and the ex post buyer’s remorse that might result if no insurable loss occurs, might negate the oft-assumed benefits of insurance. 24 References Angrist, J. D., & Pischke, J.-S. (2008). Mostly harmless econometrics: An empiricist’s companion. Princeton university press. Barrett, C. B., & Carter, M. R. (2013). The economics of poverty traps and persistent poverty: Empirical and policy implications. Journal of Development Studies , 49 (7), 976-990. Barrett, C. B., Chabari, F., Bailey, D., Little, P. D., & Coppock, D. L. (2003). Livestock pricing in the northern Kenyan rangelands. Journal of African Economies , 12 (2), 127-155. Beegle, K., Himelein, K., & Ravallion, M. (2012). Frame-of-reference bias in subjective welfare. Journal of Economic Behavior & Organization , 81 (2), 556-570. Berhanu, W. (2011). Recurrent shocks, poverty traps and the degradation of pastoral- ists’ social capital in southern Ethiopia. African Journal of Agricultural and Resrouce Economics , 6 (1), 1-15. Besley, T. (1995). Chapter 36 savings, credit and insurance. In T. N. Srinivasan & J. Behrman (Eds.), Handbook of development economics (Vol. 3, p. 2123-2207). Ams- terdam: Elsevier. Binswanger-Mkhize, H. P. (2012). Is there too much hype about index-based agricultural insurance? Journal of Development Studies , 48 (2), 187-200. Carter, M. R., & Barrett, C. B. (2006). The economics of poverty traps and persistent poverty: An asset-based approach. Journal of Development Studies , 42 (2), 178-199. Chantarat, S., Mude, A. G., Barrett, C. B., & Carter, M. R. (2013). Designing index-based livestock insurance for managing asset risk in northern Kenya. Journal of Risk and Insurance , 80 (1), 205-237. Chinn, S. (2000, November). A simple method for converting an odds ratio to effect size for use in meta-analysis. Statistics in medicine , 19 (22), 31273131. Clark, A. (2003). Unemployment as a social norm: Psychological evidence from panel data. Journal of Labor Economics , 21 (2), 323-351. Cole, S., Gin´e, X., Tobacman, J., Topalova, P., Townsend, R., & Vickery, J. (2013, January). Barriers to household risk management: Evidence from india. American Economic Journal: Applied Economics , 5 (1), 104-135. Coppock, D. L. (1994). The Borana plateau of southern ethiopia: synthesis of pastoral research, development and change, 1980-91 (No. 5). ILRI. 25 de Janvry, A., Dequiedt, V., & Sadoulet, E. (2014). The demand for insurance against common shocks. Journal of Development Economics , 106 , 227-238. Dercon, S., & Christiaensen, L. (2011). Consumption risk, technology adoption and poverty traps: Evidence from Ethiopia. Journal of Development Economics , 96 (2), 159-173. Desta, S., & Coppock, D. L. (2002). Cattle population dynamics in the southern ethiopian rangelands, 1980-97. Journal of Range Management , 55 (5), 439–451. Devoto, F., Duflo, E., Dupas, P., Parient´ e, W., & Pons, V. (2012, May). Happiness on tap: Piped water adoption in urban morocco. American Economic Journal: Economic Policy , 4 (4), 68-99. Fafchamps, M., & Shilpi, F. (2008). Subjective welfare, isolation, and relative consumption. Journal of Development Economics , 86 (1), 43-60. Feldman, R., & Dowd, B. (1991). A new estimate of the welfare loss of excess health insurance. American Economic Review , 81 (1), 297-301. Finkelstein, A., Taubman, S., Wright, B., Bernstein, M., Gruber, J., Newhouse, J. P., . . . (2012). The oregon health insurance experiment: Evidence from the first year*. Quar- terly Journal of Economics , 127 (3), 1057-1106. Fitzgerald, J., Gottschalk, P., & Moffitt, R. (1998). An analysis of sample attrition in panel data: The michigan panel study of income dynamics. The Journal of Human Resources , 33 (2), 251-299. Frey, B. S., & Stutzer, A. (2010). Happiness and economics: How the economy and institu- tions affect human well-being. Princeton University Press. Gharad, B. (2013). Ambiguity aversion decreases demand for partial insurance: Evidence from African farmers. Gin´e, X., Townsend, R., & Vickery, J. (2008). Patterns of rainfall insurance participation in rural india. World Bank Economic Review , 22 (3), 539-566. Graham, C. (2012). Happiness around the world: The paradox of happy peasants and miserable millionaires. Oxford University Press. Halek, M., & Eisenhauer, J. G. (2001). Demography of risk aversion. Journal of Risk and Insurance , 68 (1), 1-24. Hasselblad, V., & Hedges, L. V. (1995). Meta-analysis of screening and diagnostic tests. Psychological bulletin , 117 (1), 167-178. ILRI. (2014). Ibli borena household survey codebook april 2014 draft [Computer software manual]. ILRI-IBLI. (2013). Index-based livestock insurance (ibli) trainers manual [Computer software 26 manual]. Janzen, S. A., & Carter, M. R. (2013, December). After the drought: The impact of microinsurance on consumption smoothing and asset protection (Working Paper No. 19702). National Bureau of Economic Research. Jensen, N. D., & Barrett, C. B. (2017). Agricultural index insurance for development. Applied Economic Perspectives and Policy , 39 (2), 199-219. Jensen, N. D., Barrett, C. B., & Mude, A. G. (2016). Index insurance quality and basis risk: Evidence from northern kenya. American Journal of Agricultural Economics , 98 (5), 1450-1469. Jensen, N. D., Barrett, C. B., & Mude, A. G. (2017). Cash transfers and index insurance: A comparative impact analysis from northern kenya. Journal of Development Economics , 129 , 14-28. Kaminski, J. (2014). Subjective wealth and satisfaction with policy reform: Evidence from the cotton reform experience in Burkina Faso. Journal of African Economies , 23 (4), 528-581. Karlan, D., Osei, R., Osei-Akoto, I., & Udry, C. (2014). Agricultural decisions after relaxing credit and risk constraints. Quarterly Journal of Economics , 129 (2), 597-652. King, G., Murray, C. J. L., Salomon, J. A., & Tandon, A. (2004). Enhancing the validity and cross-cultural comparability of measurement in survey research. American political science review , 98 (1), 191-207. King, G., & Wand, J. (2007). Comparing incomparable survey responses: Evaluating and selecting anchoring vignettes. Political Analysis , 15 (1), 46-66. Krueger, A. B., & Schkade, D. A. (2008). The reliability of subjective well-being measures. Journal of Public Economics , 92 (89), 1833-1845. Krueger, A. B., & Stone, A. A. (2014). Progress in measuring subjective well-being. Science , 346 (6205), 42-43. Ludwig, J., Duncan, G. J., Gennetian, L. A., Katz, L. F., Kessler, R. C., Kling, J. R., & Sanbonmatsu, L. (2013, May). Long-term neighborhood effects on low-income families: Evidence from moving to opportunity. American Economic Review , 103 (3), 226-231. Lybbert, T. J., Barrett, C. B., Desta, S., & Coppock, D. L. (2004). Stochastic wealth dy- namics and risk management among a poor population. Economic Journal , 114 (498), 750-777. Miranda, M. J., & Farrin, K. (2012). Index insurance for developing countries. Applied Economic Perspectives and Policy , 34 (3), 391-427. 27 Morduch, J. (1994). Poverty and vulnerability. American Economic Review , 84 (2), 221-225. Ravallion, M. (2012). Poor, or just feeling poor? on using subjective data in measuring poverty (Working Paper No. 5968). The World Bank. Ravallion, M., Himelein, K., & Beegle, K. (2013). Can subjective questions on economic welfare be trusted? evidence for three developing countries [Working Paper]. (6726). Rosenzweig, M. R., & Binswanger, H. P. (1993). Wealth, weather risk and the composition and profitability of agricultural investments. Economic Journal , 103 (416), 56-78. Santos, P., & Barrett, C. B. (2011). Persistent poverty and informal credit. Journal of Development Economics , 96 (2), 337-347. Santos, P., & Barrett, C. B. (2016, September). Heterogeneous wealth dynamics: On the roles of risk and ability (Working Paper No. 22626). National Bureau of Economic Research. Smith, V. H. (2016). Producer insurance and risk management options for smallholder farmers. World Bank Research Observer , 31 (2), 271-289. van Soest, A., Delaney, L., Harmon, C., Kapteyn, A., & Smith, J. P. (2011). Validating the use of anchoring vignettes for the correction of response scale differences in subjective questions. Journal of the Royal Statistical Society: Series A (Statistics in Society) , 174 (3), 575-595. Williams, J. R. (1988). A stochastic dominance analysis of tillage and crop insurance practices in a semiarid region. American Journal of Agricultural Economics , 70 (1), 112-120. Wooldridge, J. M. (2010). Econometric analysis of cross section and panel data. Cambridge, MA: MIT Press. 28 Table 1: Test of treatment-control covariate balance at baseline Sales period 1 assignments Sales period 2 assignments Difference Difference Discount No Discount (Discount– Discount No Discount (Discount– Coupon Coupon No Discount) Coupon Coupon No Discount) (1) (2) (3) (4) (5) (6) Subjective well-being (SWB) 2.869 2.924 -0.055 2.878 2.887 -0.010 (0.058) (0.102) (0.126) (0.057) (0.112) (0.125) SWB relative to Borana pastoralists 2.855 2.866 -0.012 2.886 2.746 0.140 (0.047) (0.088) (0.103) (0.046) (0.096) (0.102) Vignette corrected SWB 3.570 3.741 -0.172 3.556 3.793 -0.238 (0.076) (0.145) (0.167) (0.075) (0.146) (0.165) Vignette corrected SWB relative to 3.628 3.760 -0.132 3.626 3.765 -0.139 Borana pastoralists (0.072) (0.139) (0.159) (0.073) (0.132) (0.158) Number of TLUs owned 14.197 17.048 -2.851 14.058 17.529 -3.471 (1.097) (2.104) (2.424) (0.942) (3.027) (2.405) Non-livestock assets (’000 Birr) 2.672 3.034 -0.363 2.761 2.684 0.078 (0.197) (0.495) (0.464) (0.204) (0.448) (0.461) Annual income (’000 Birr) 20.357 20.106 0.251 19.734 22.512 -2.778 (2.322) (2.087) (4.736) (1.846) (5.888) (4.701) Expected TLU loss (max=52) 15.252 17.019 -1.767* 15.784 14.933 0.852 (0.448) (0.888) (0.996) (0.443) (0.935) (0.991) Gender of household head (Male=1) 0.774 0.818 -0.044 0.773 0.821 -0.049 (0.021) (0.039) (0.046) (0.021) (0.038) (0.045) Age of household head (years) 49.850 49.500 0.350 49.607 50.444 -0.838 (0.902) (1.75) (1.997) (0.914) (1.658) (1.983) Household size (#) 6.324 5.895 0.430 6.189 6.425 -0.237 (0.126) (0.205) (0.271) (0.120) (0.257) (0.269) Non-working age hh members (#) 3.567 3.231 0.337* 3.480 3.576 -0.097 (0.091) (0.163) (0.198) (0.087) (0.197) (0.197) Female hh members (#) 3.122 3.010 0.113 3.064 3.236 -0.173 (0.078) (0.135) (0.169) (0.075) (0.162) (0.168) Iqub (ROSCAs) membership (%) 0.095 0.058 0.037 0.092 0.064 0.028 (0.015) (0.023) (0.031) (0.015) (0.025) (0.032) Observations 411 104 515 409 106 515 Standard errors in parentheses: *** p<0.01, ** p<0.05, * p<0.1. 29 Table 2: Summary statistics - round 2 and 3 values (pooled), by insurance status Difference Insured Uninsured (Insured - Uninsured) (1) (2) (3) Subjective well-being (SWB) 3.192 3.049 0.143** (0.041) (0.036) (0.056) SWB relative to Borana pastoralists 3.250 3.138 0.112** (0.038) (0.034) (0.053) Vignette corrected SWB 4.079 3.714 0.365*** (0.068) (0.058) (0.092) Vignette corrected SWB relative to 4.100 3.792 0.308*** Borana pastoralists (0.065) (0.057) (0.089) Encouragement design Discount coupon 0.932 0.524 0.408*** (0.013) (0.020) (0.027) Audio tape 0.110 0.039 0.071*** (0.016) (0.008) (0.016) Cartoon 0.165 0.085 0.081*** (0.019) (0.011) (0.020) Value of discount coupon (%) SP1 0.353 0.164 0.188*** (0.016) (0.010) (0.018) Value of discount coupon (%) SP2 0.278 0.171 0.107*** (0.016) (0.011) (0.082) Number of TLUs owned 20.592 17.323 3.269* (1.671) (1.050) (1.874) Non-livestock assets (’000 Birr) 4.975 4.630 0.344 (0.480) (0.460) (0.702) Annual income (’000 Birr) 20.932 19.180 1.753 (2.048) (1.168) (2.188) Expected TLU loss (max=52) 13.077 12.989 -0.089 (0.410) (0.362) (0.566) Gender of household head (Male=1) 0.774 0.807 -0.033 (0.021) (0.016) (0.026) Age of household head (years) 50.341 51.884 -1.542 (0.915) (0.726) (1.176) Household size (#) 6.561 6.745 0.183 (0.125) (0.105) (0.167) Non-working age hh members (#) 3.619 3.754 0.134 (0.090) (0.071) (0.115) Female hh members (#) 3.276 3.330 0.055 (0.074) (0.065) (0.101) Iqub (ROSCAs) membership (%) 0.058 0.053 0.005 (0.012) (0.009) (0.015) Observations 381 639 1020 Standard errors in parentheses: *** p<0.01, ** p<0.05, * p<0.1. 30 Table 3: First stage estimates of IBLI uptake and volume of TLUs insured LPM estimates of Tobit estimates of IBLI uptake volume of TLUs insured (1) (2) (3) (4) Discount: SP1 only 0.103** 0.108** 3.124*** 2.731*** (0.044) (0.043) (0.764) (0.773) Discount: SP2 only 0.165*** 0.173*** 2.988*** 2.931*** (0.046) (0.044) (0.765) (0.768) Discount: SP1 & SP2 0.084* 0.085* 2.402*** 2.036** (0.046) (0.043) (0.823) (0.824) Value of discount (%) SP1 0.183*** 0.181*** 3.737*** 4.050*** (0.056) (0.055) (0.892) (0.892) Value of discount (%) SP2 -0.007 -0.011 3.296*** 2.957*** (0.065) (0.061) (0.908) (0.912) Poet tape: SP1 only 0.043 0.063 0.363 0.963 (0.092) (0.087) (1.031) (1.041) Poet tape: SP2 only 0.114 0.131* 2.823*** 2.803*** (0.073) (0.070) (0.977) (0.979) Poet tape: SP1 & SP2 0.129** 0.098 0.836 0.553 (0.063) (0.065) (1.256) (1.268) Comic book: SP1 only 0.078 0.063 0.945 0.820 (0.059) (0.061) (0.891) (0.896) Comic book: SP2 only 0.068 0.079 1.586* 1.231 (0.061) (0.064) (0.923) (0.926) Comic book: SP1 & SP2 0.200*** 0.217*** 2.632*** 2.522*** (0.073) (0.071) (0.787) (0.812) IBLI premium: SP1 - - 0.286 -3.236 (4.189) (18.438) IBLI premium: SP2 0.243 0.017 2.425 4.598 (0.211) (1.014) (2.947) (20.238) IBLI knowledge 0.007 0.508*** (0.006) (0.129) Expected TLUs loss -0.001 -0.004 (0.002) (0.023) Number of TLUs owned 0.002* 0.019** (0.001) (0.008) Asset index 0.131 0.303 (0.128) (0.253) Annual income (’000 Birr) -0.0001 -0.005 (0.0002) (0.005) Household head gender (Male=1) -0.241* 0.592 (0.136) (0.638) Household head age 0.002 -0.041 (0.016) (0.081) Household age squared -0.0001 0.0003 (0.0001) (0.001) Household size 0.075*** -0.009 (0.027) (0.193) Household head schooling -0.003 -0.176 (0.008) (0.124) Iqub membership -0.070 -1.353* (0.049) (0.748) Household composition No Yes No Yes Round dummy No Yes No Yes Constant 0.086 0.482 -8.403*** -8.332*** (0.132) (0.902) (2.280) (3.488) Wald test for joint significance of instruments (χ2 ) 72.4 197.7 58.6 255.9 P-value (0.000) (0.000) (0.000) (0.000) Observations 1,015 1,015 1,015 1,015 Number of households 520 520 520 520 Bootstrap standard errors in parentheses: *** p<0.01, ** p<0.05, * p<0.1. Note: SP1 stands for sales period 1 and SP2 stands for sales period 2. Columns (1) and (2) show LPM estimates of IBLI uptake. The dependent variable IBLI uptake is a dummy variable that takes value 1 if a household buys IBLI and 0 otherwise. Standard errors are clustered at the panel-Reera level. Columns (3) and (4) show Tobit model estimates of volume of TLUs insured. The dependent variable TLUs insured is a non-negative continues variable. IBLI premium for SP1 did not vary between R2 and R3. Thus, it is dropped in the FE LPM regression results in columns (1) and (2). The controls for household composition include number of household members by age group and gender: all/male/female #members≤5, #mem>5&≤15,#mem>15&≤64, and #mem≥65. 31 Table 4: Ordered logit regression: Vignette adjusted SWB estimates using IBLI uptake and volume of TLUs insured IBLI uptake TLUs insured (1) (2) (3) (4) (5) (6) Dependent variable: SWB panel (a) Predicted IBLI/ TLUs insured 0.816*** 0.713*** 0.859*** 0.137*** 0.138*** 0.144*** (0.264) (0.256) (0.290) (0.041) (0.040) (0.041) Predicted lapsed IBLI/ TLUs insured -0.454** -0.439** -0.442** -0.077** -0.071** -0.074** (0.191) (0.197) (0.198) (0.030) (0.031) (0.030) Number of TLUs owned 0.015** 0.012* 0.012* 0.015** 0.012* 0.012* (0.006) (0.007) (0.007) (0.006) (0.006) (0.007) Asset Index 0.283** 0.239** 0.325*** 0.289*** (0.114) (0.120) (0.102) (0.101) Annual income (’000 Birr) 0.003 0.004 0.003 0.003 (0.002) (0.002) (0.002) (0.002) Household head gender (Male=1) 0.733** 0.617* (0.358) (0.334) Household head age -0.042 -0.037 (0.040) (0.038) Household head age squared 0.0004 0.0003 (0.0003) (0.0003) Household size -0.224** -0.193** (0.090) (0.085) Household head schooling 0.055 0.058 (0.051) (0.053) panel (b) Predicted IBLI/ TLUs insured prob(SWB=1) -0.042*** -0.037*** -0.044*** -0.007*** -0.007*** -0.007*** (0.014) (0.014) (0.015) (0.002) (0.002) (0.002) prob(SWB=2) -0.033*** -0.029*** -0.035*** -0.006*** -0.006*** -0.006*** (0.011) (0.011) (0.012) (0.002) (0.002) (0.002) prob(SWB=3) -0.025*** -0.021** -0.026*** -0.004*** -0.004*** -0.004*** (0.009) (0.008) (0.009) (0.001) (0.001) (0.001) prob(SWB=4) 0.005** 0.005* 0.007* 0.001* 0.001* 0.001* (0.003) (0.003) (0.003) (0.001) (0.001) (0.001) prob(SWB=5) 0.052*** 0.045*** 0.055*** 0.009*** 0.009*** 0.009*** (0.018) (0.017) (0.019) (0.003) (0.003) (0.003) prob(SWB=6) 0.027*** 0.023*** 0.028*** 0.005*** 0.004*** 0.005*** (0.009) (0.009) (0.009) (0.001) (0.001) (0.001) prob(SWB=7) 0.015*** 0.013** 0.016*** 0.003*** 0.003*** 0.003*** (0.005) (0.005) (0.006) (0.001) (0.001) (0.001) Predicted lapsed IBLI/ TLUs insured prob(SWB=1) 0.024** 0.023** 0.023** 0.004** 0.004** 0.004** (0.010) (0.011) (0.011) (0.002) (0.002) (0.002) prob(SWB=2) 0.018** 0.018** 0.018** 0.003** 0.003** 0.003** (0.008) (0.008) (0.008) (0.001) (0.001) (0.001) prob(SWB=3) 0.014** 0.013** 0.013** 0.002** 0.002** 0.002** (0.006) (0.006) (0.007) (0.001) (0.001) (0.001) prob(SWB=4) -0.003* -0.003 -0.003* -0.001 -0.001 -0.001 (0.002) (0.002) (0.002) (0.0003) (0.0003) (0.0004) prob(SWB=5) -0.029** -0.028** -0.028** -0.005** -0.005** -0.005** (0.013) (0.013) (0.013) (0.002) (0.002) (0.002) prob(SWB=6) -0.015** -0.014** -0.014** -0.003** -0.002** -0.002** (0.007) (0.007) (0.007) (0.001) (0.001) (0.001) prob(SWB=7) -0.009** -0.008** -0.008** -0.001** -0.001** -0.001** (0.004) (0.004) (0.004) (0.001) (0.001) (0.001) Number of TLUs owned prob(SWB=1) -0.001** -0.001* -0.001* -0.001** -0.001* -0.001* (0.0003) (0.0003) (0.0004) (0.0003) (0.0003) (0.0004) prob(SWB=2) -0.001** -0.0005* -0.0005* -0.001** -0.0005* -0.0005* (0.0003) (0.0003) (0.0003) (0.0003) (0.0003) (0.0003) prob(SWB=3) -0.0004** -0.0004* -0.0004 -0.0005** -0.0004* -0.0004* (0.0002) (0.0002) (0.0002) (0.0002) (0.0002) (0.0002) prob(SWB=4) 0.0001 0.0001 0.0001 0.0001 0.0001 0.0001 (0.0001) (0.00001) (0.0001) (0.0001) (0.0001) (0.0001) prob(SWB=5) 0.001** 0.001* 0.001* 0.001** 0.001* 0.001* (0.0004) (0.0004) (0.0005) (0.0004) (0.0004) (0.0004) prob(SWB=6) 0.0005** 0.0004* 0.0004* 0.0005** 0.0004* 0.0004* (0.0002) (0.0002) (0.0002) (0.0002) (0.0002) (0.0002) prob(SWB=7) 0.0003** 0.0002* 0.0002* 0.0003** 0.0002* 0.0002* (0.0001) (0.0001) (0.0001) (0.0001) (0.0001) (0.0001) Household composition No No Yes No No Yes Round dummy No No Yes No No Yes Observations 1,530 1,530 1,530 1,530 1,530 1,530 Number of households 550 550 550 550 550 550 Cluster bootstrap standard errors in parentheses: *** p<0.01, ** p<0.05, * p<0.1. Note: Panel (a) reports the effects of IBLI uptake and volume of TLUs insured on vignette adjusted SWB in log-odds units. Panel (b) reports the marginal effects for the main results in panel (a) – IBLI/ TLUs insured, lapsed IBLI/ TLUs insured and number of TLUs owned. The marginal effects estimates in panel (b) show the effects of these variables on the probability of reporting one of the seven unique scales of SWB. In column 3 for example, IBLI uptake reduces the probability of reporting SWB=1 by 4.4% and increases the probability of reporting SWB=7 by 1.6%. A unit increase in TLUs owned reduces the probability of reporting SWB=1 by 0.1% and increases the probability of reporting SWB=7 by 0.02%. 32 Table 5: Aggregate effect of IBLI on SWB Variables (1) (2) (3) ∆SW Bivt 0.197*** 0.202*** 0.213*** (0.072) (0.071) (0.073) Observations 1,530 1,530 1,530 Standard errors in parentheses: *** p<0.01, ** p<0.05, * p<0.1. 1 year contract coverage LRLD season coverage SRSD season coverage Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec Jan Feb Sales period for LRLD season Period of NDVI observation for LRLD season Sales period for SRSD season Period of NDVI observation for SRSD season Announce LRLD NDVI value. If below contract strike value, make indemnity payments. Rainy season Dry season Announce SRSD NDVI value. If below contract strike value, make indemnity payments. Figure 1: Temporal structure of IBLI contract Note : LRLD indicates the long rains, long dry season. SRSD indicates the short rains, short dry season. Round 1 (R1) Round 2 (R2) Round 3 (R3) (2012) (2013) (2014) Sales period 1 Sales period 2 Sales period 3 Sales period 4 (SP1) (SP2) (SP3) (SP4) Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec Jan Feb Mar Apr May Jun Jul SP1 coverage SP2 coverage SP3 coverage SP4 coverage Lapsed contract: Lapsed contract: Lapsed contract: None SP1 SP1 & SP2 Survey period Sales period Figure 2: Timeline of IBLI survey and sales periods 33 Appendix A Appendix Tables Table A1: Annual IBLI premium and out of pocket payments Aug-Sept 2012; Jan-Feb 2013; Aug-Sept 2013 Jan-Feb 2014 Premium (Birr) Out of Premium (Birr) Out of pocket pocket Goat/ TLUs pay per Goat/ TLUs pay per Woreda (%) Cattle Camel Sheep TLU insured TLU (Birr)* (%) Cattle Camel Sheep TLU insured TLU (Birr)* Dillo 9.8 488 1,463 68 739 1.3 450 8.6 516 860 69 606 0.4 324 Teltele 8.7 436 1,307 61 660 3.1 385 7.7 462 770 62 543 3.3 236 Yabello 7.5 377 1,131 53 571 3.1 289 6.7 402 670 54 472 2.1 240 Dire 9.5 475 1,424 66 719 1.6 413 8.4 504 840 67 592 1.7 296 Arero 8.6 429 1,287 60 650 2.9 333 7.6 456 760 61 536 4.1 300 Dehas 9.4 468 1,404 66 709 3.0 343 8.3 498 830 66 585 4.0 234 34 Miyo 11.1 553 1,658 77 837 0.9 442 9.8 588 980 78 691 2.5 414 Moyale 11.1 553 1,658 77 837 1.2 566 9.8 588 980 78 691 0.0 - Overall 461 1,382 65 698 2.3 384 489.4 815.7 65 575 2.4 279 Source : Source: ILRI, 2013 and own calculation Note :* Average out of pocket payment per TLU by actual buyers. Table A2: Variable definitions General information Description Round 1 Baseline conducted: March/April, 2012 Round 2 Conducted: March/April, 2013 Round 3 Conducted: March 2014 Sales period 1 August-September 2012; contract active- October 2012-September 2013; Encourage- ment design- discount coupon, poet tape, comic book Sales period 2 January-February 2013; contract active- March 2013-February 2014; Encouragement design- discount coupon, poet tape, comic book Sales period 3 August-September 2013; contract active- October 2013-September 2014; Encourage- ment design- discount coupon only Sales period 4 January-February 2014; contract active- March 2014-February 2015; Encouragement design- discount coupon only Variable Definition SWB An ordinal scale of respondents’ stated perception of their economic condition on a Likert scale ranging from 1=very bad to 5= very good. It is the answer to the question “On which step do you place your present economic conditions?” SWB relative to Bo- Response the question “In general, how do you rate your living conditions compared rana pastoralists to those of other Borana pastoralists?” 1=much worse;...; 5=much better Discount coupon A dummy variable taking value 1 if a household received discount coupon and 0 oth- erwise. Audio tape A dummy variable taking value 1 if a household received additional information treat- ment via audio tape and 0 otherwise. Comic book A dummy variable taking value 1 if a household received additional information via comic book and 0 otherwise. Value of discount The amount of discount received, in percentages, which ranges between 0 and 100%. coupon Number of TLU A standardized measure of livestock holding. It is obtained by multiplying number of owned livestock by the relevant TLU conversion unit for each livestock type. The conversion units used are TLU=1 for cattle, TLU=1.4 for camel, and TLU=0.1 for goats and sheep, collectively called shoats. Non-Livestock assets Value of non-livestock assets in Birr. It includes assets such as bed frame, mattress, chair, table, bicycle, motorcycle, car, cellphone, computer, television, radio, wheelbar- row, grind mill, axe, spade, sickle, hoe, watch, jewelry etc. Expected TLU loss Constructed from a set of questions that ask responds how many of 20 livestock (by type) they expect to die in the coming year. These figures are converted to common TLUs. Thus, results should be read against a total of 52 tropical livestock units. The questions used are “what is the number out of 20 X do you expect to die over the March 2013 to February 2014 period?” X here stands for livestock types. 35 Insurance premium Insurance premium per TLU. Insurance premium vary by livestock type and Woreda. Some household in the sample also received discount. To reflect this variation, premium is calculated as: (1 − %discount) × (cattlepremium × 1+ camelpremium × 1.4+ shoatspremium × 0.1)/3. Cash income Includes cash income (in 1,000 Birr) from sale of livestock and livestock products, crop sales, wages and salaries, business and trading (petty trading, motorcycle services etc), cash for work (bush clearing, pond digging etc), mining etc. Net transfers The value of annual net cash transfers (during the four seasons: long dry, long rainy, short dry and short rainy). It includes both cash and in kind transfers. It is the difference between transfers received and transfers given. Value of food aid The value of annual food aid (in 1,000 Birr) received by households. It is calculated by multiplying the value of monthly food aid by the number of months food aid is received. Non-food assistance The value of annual non-food assistance (in 1,000 Birr). It includes value of annual should feeding, supplementary feeding, income from employment program, and non- food aid. The value of non-food aid consists of non-food aid from government, NGOs, and PSNP program e.g., water, fodder, vaccination, cash transfers via PSNP. Annual Income The sum of annual cash income, value of auto-consumption, net transfers, food aid, and non-food assistance in 1,000 Birr. Price per TLU The average price of a TLU equivalent calculated by weighting prices for shoats, cat- tle, and camel at Haro Bake livestock market in Borana zone by each species’ TLU conversion unit. More specifically, we used Birr 700 for shoats price, Birr 5,000 for cattle price and Birr 15,000 for camel price. The TLU conversion unit for shoats is 0.1, for cattle 1 and camel 1.4. Thus, price per TLU=0.1700 + 15,000 + 1.415,000 = Birr 7,571.4. Asset Index An index constructed from the current value of non-livestock assets using the principal component factor (PCF) method. Household size The number of people who live in the same homestead including people who are away temporarily for less than eight months. Number of non- Includes household members 14 years old and under and 65 years and above. working age house- hold members Iqub membership Iqub is an informal rotating saving and credit organization (ROSCA). The variable takes value 1 if a household member is a member of Iqub, and 0 otherwise. 36 Table A3: Joint orthogonality test for selection into treatment Aug-Sep sales period Jan-Feb sales period Discount Comic Audio Discount Comic Audio coupon book tape coupon book tape (1) (2) (3) (4) (5) (6) Expected TLU loss -0.002 0.001 0.001 0.001 -0.0003 0.004** (0.002) (0.002) (0.002) (0.002) (0.002) (0.002) Number of TLUs owned -0.001** 0.001 -0.0003 -0.001 -0.0004 0.0003 (0.001) (0.001) (0.001) (0.001) (0.001) (0.001) Asset index 0.010 -0.033 -0.038** 0.012 -0.030 -0.029 (0.014) (0.021) (0.018) (0.014) (0.021) (0.018) Annual income (’000 Birr) 0.0002 0.0005 -0.00003 0.0002 0.0004 -0.0004 (0.0004) (0.001) (0.0004) (0.0004) (0.001) (0.0004) Household head gender (Male=1) -0.077** -0.0004 -0.018 -0.043 -0.009 -0.044 (0.034) (0.048) (0.041) (0.034) (0.049) (0.042) Household head age -0.0004 -0.0001 -0.0002 0.00004 0.001 -0.001 (0.001) (0.001) (0.001) (0.001) (0.001) (0.001) Household size 0.018* 0.020 0.011 0.008 0.002 0.019 (0.010) (0.014) (0.012) (0.010) (0.014) (0.012) Household head schooling 0.005 0.006 -0.006 0.002 0.014 -0.010 (0.007) (0.010) (0.009) (0.007) (0.010) (0.009) Number of females in household -0.024** -0.017 -0.010 -0.010 0.003 -0.014 (0.012) (0.017) (0.015) (0.012) (0.018) (0.015) Number of working age household members -0.010 -0.022 0.013 -0.017 0.006 0.004 (0.013) (0.019) (0.016) (0.013) (0.019) (0.017) Iqub membership 0.056 0.111* -0.039 0.015 -0.010 -0.020 (0.049) (0.065) (0.055) (0.049) (0.066) (0.056) Constant 0.909*** 0.130 0.078 0.868*** 0.084 0.070 (0.058) (0.083) (0.071) (0.059) (0.084) (0.072) Observations 968 473 473 968 473 473 Prob > F 0.144 0.463 0.411 0.729 0.840 0.129 R-squared 0.016 0.023 0.024 0.008 0.014 0.035 Standard errors cluster bootstrapped at the Reera level in parentheses: *** p<0.01, ** p<0.05, * p<0.1. Note: Table A3 presents joint tests of orthogonality for the treatment variables. Prior to sales period 1 (Aug-Sep 2012) and sales period 2 (Jan-Feb 2013), discount coupons as well as audio tape and comic book information treat- ments were distributed to randomly selected sub-sample of survey households. Similarly, prior to sales period 3 (Aug-Sep 2013) and sales period 4 (Jan-Feb 2014) discount coupons were distributed to randomly selected house- holds. Columns 1-3 and 4-6 show the linear probability model (LPM) regressions of assignment into discount coupon, comic book and audio tape treatment in the Aug-Sep and Jan-Feb sales periods, respectively, on lagged household characteristics using a pooled sample from rounds 2 and 3. Note that the sample includes baseline households who were re-interviewed in R2 and R2 households re-interviewed in R3. The joint orthogonality test (F-test) is reported in the second bottom row. 37 Table A4: Comparison of uncorrected SWB and vignette corrected SWB a)SWB and vignette corrected SWB Vignette corrected SWB SWB 1 2 3 4 5 6 7 Total Very bad (1) 27 93 0 0 0 0 0 120 Bad (2) 31 30 115 74 15 0 0 265 Neither good nor bad (3) 65 22 147 224 221 5 5 689 Good (4) 29 7 23 85 183 34 9 370 Very good (5) 0 5 0 8 0 58 17 88 Total 152 157 285 391 419 97 31 1532 b) SWB relative to Borana pastoralists and vignette-corrected SWB relative to Borana pastoralists Vignette corrected SWB relative to Borana households SWB relative to Borana households 1 2 3 4 5 6 7 Total Much worse (1) 13 51 0 0 0 0 0 64 Worse (2) 28 32 145 88 21 0 1 315 Same (3) 67 19 154 181 194 5 6 626 Better (4) 31 15 27 92 266 59 13 503 Much better (5) 0 2 0 2 0 13 7 24 Total 139 119 326 363 481 77 27 1532 38 Table A5: Probit model estimates of IBLI uptake Dependent variable: IBLI uptake (1) (2) Discount: SP1 only 1.865*** 1.633*** (0.308) (0.309) Discount: SP2 only 1.861*** 1.754*** (0.367) (0.350) Discount: SP1 & SP2 1.661*** 1.445*** (0.393) (0.368) Value of discount (%) SP1 0.956** 0.955** (0.390) (0.406) Value of discount (%) SP2 0.163 0.030 (0.367) (0.350) Poet tape: SP1 only 0.672 0.827 (0.709) (0.712) Poet tape: SP2 only 1.381*** 1.336*** (0.337) (0.317) Poet tape: SP1 & SP2 0.237 0.115 (0.461) (0.429) Comic book: SP1 only 0.770 0.777* (0.478) (0.434) Comic book: SP2 only 0.344 0.121 (0.426) (0.427) Comic book: SP1 & SP2 0.921* 0.799 (0.508) (0.535) IBLI premium: SP1 2.725 -0.846 (1.738) (8.931) IBLI premium: SP2 1.905 4.095 (1.640) (10.080) IBLI knowledge 0.232*** (0.055) Expected TLUs loss -0.001 (0.012) Number of TLUs owned 0.006 (0.004) Asset index 0.028 (0.097) Annual income (’000 Birr) -0.001 (0.002) Household head gender (Male=1) -0.216 (0.230) Household head age -0.022 (0.026) Household age squared 0.0002 (0.0002) Household size 0.008 (0.057) Household head schooling -0.025 (0.048) Iqub membership -0.362 (0.282) Household composition No Yes Round dummy No Yes Constant -4.647*** -4.191** (1.684) (1.779) Observations 1,015 1,015 Number of households 520 520 Cluster bootstrap standard errors in parentheses: *** p<0.01, ** p<0.05, * p<0.1. 39 Table A6: Ordered logit regression: Estimates for vignette adjusted SWB relative to Borana pastoralists using IBLI uptake and volume of TLUs insured IBLI uptake TLUs insured (1) (2) (3) (4) (5) (6) Dependent variable: SWB relative to Borana pastoralists panel (a) Predicted IBLI/ TLUs insured 0.942*** 0.930*** 1.070*** 0.179*** 0.179*** 0.188*** (0.269) (0.273) (0.290) (0.040) (0.040) (0.041) Predicted lapsed IBLI/ TLUs insured -0.351* -0.336* -0.364* -0.075** -0.072** -0.080** (0.200) (0.199) (0.201) (0.033) (0.032) (0.033) Number of TLUs owned 0.010** 0.009* 0.009* 0.010** 0.009** 0.009** (0.005) (0.005) (0.005) (0.004) (0.004) (0.005) Asset index 0.044 -0.006 0.103 0.060 (0.141) (0.145) (0.123) (0.120) Annual income (’000 Birr) 0.002 0.002 0.001 0.002 (0.001) (0.001) (0.001) (0.001) Household head gender (Male=1) 0.593 0.436 (0.366) (0.337) Household head age -0.044 -0.037 (0.041) (0.038) Household head age squared 0.0004 0.000 (0.0003) (0.000) Household size -0.232** -0.187** (0.093) (0.088) Household head schooling 0.071 0.077* (0.045) (0.046) panel (b) Predicted IBLI/ TLUs insured prob(SWB=1) -0.043*** -0.043*** -0.049*** -0.008*** -0.008*** -0.009*** (0.013) (0.013) (0.014) (0.002) (0.002) (0.002) prob(SWB=2) -0.031*** -0.031*** -0.036*** -0.006*** -0.006*** -0.006*** (0.009) (0.010) (0.010) (0.002) (0.002) (0.002) prob(SWB=3) -0.037*** -0.037*** -0.042*** -0.007*** -0.007*** -0.007*** (0.012) (0.012) (0.012) (0.002) (0.002) (0.002) prob(SWB=4) 0.002 0.002 0.003 0.0004 0.0004 0.001 (0.002) (0.002) (0.003) (0.0004) (0.0005) (0.001) prob(SWB=5) 0.067*** 0.066*** 0.077*** 0.013*** 0.013*** 0.013*** (0.020) (0.021) (0.021) (0.003) (0.003) (0.003) prob(SWB=6) 0.027*** 0.027*** 0.030*** 0.005*** 0.005*** 0.005*** (0.008) (0.009) (0.009) (0.001) (0.001) (0.001) prob(SWB=7) 0.015*** 0.015*** 0.017*** 0.003*** 0.003*** 0.003*** (0.005) (0.005) (0.005) (0.001) (0.001) (0.001) Predicted lapsed IBLI/ TLUs insured prob(SWB=1) 0.016* 0.016* 0.017* 0.003** 0.003** 0.004** (0.009) (0.009) (0.009) (0.001) (0.001) (0.001) prob(SWB=2) 0.012* 0.011* 0.012* 0.002** 0.002** 0.003** (0.006) (0.006) (0.006) (0.001) (0.001) (0.001) prob(SWB=3) 0.014* 0.013* 0.014* 0.003** 0.003** 0.003** (0.008) (0.008) (0.008) (0.001) (0.001) (0.001) prob(SWB=4) -0.001 -0.001 -0.001 -0.0002 -0.0002 -0.0002 (0.001) (0.001) (0.001) (0.0002) (0.0002) (0.0003) prob(SWB=5) -0.025* -0.024* -0.026* -0.005** -0.005** -0.006** (0.014) (0.013) (0.014) (0.002) (0.002) (0.002) prob(SWB=6) -0.010* -0.010* -0.010* -0.002** -0.002** -0.002** (0.006) (0.005) (0.006) (0.001) (0.001) (0.001) prob(SWB=7) -0.006* -0.005* -0.006* -0.001** -0.001** -0.001** (0.003) (0.003) (0.003) (0.001) (0.001) (0.001) Number of TLUs owned prob(SWB=1) -0.0004** -0.0004* -0.0004* -0.0005** -0.0004* -0.0004* (0.0002) (0.0002) (0.0002) (0.0002) (0.0002) (0.0002) prob(SWB=2) -0.0003** -0.0003* -0.0003* -0.0003** -0.0003* -0.0003* (0.0002) (0.0002) (0.0002) (0.0002) (0.0002) (0.0002) prob(SWB=3) -0.0004** -0.0004* -0.0003* -0.0004** -0.0004* -0.0004* (0.0002) (0.0002) (0.0002) (0.0002) (0.0002) (0.0002) prob(SWB=4) 0.00002 0.00002 0.00003 0.00002 0.00002 0.00003 (0.00003) (0.00003) (0.00003) (0.00003) (0.00003) (0.00003) prob(SWB=5) 0.001** 0.001* 0.001* 0.001** 0.001* 0.001* (0.0003) (0.0003) (0.0004) (0.0003) (0.0003) (0.0003) prob(SWB=6) 0.0003** 0.0003* 0.0003* 0.0003** 0.0003* 0.0003* (0.0001) (0.0001) (0.0001) (0.0001) (0.0001) (0.0001) prob(SWB=7) 0.0002* 0.0001* 0.0001* 0.0002** 0.0001* 0.0001* (0.0001) (0.0001) (0.0001) (0.00007) (0.00007) (0.00008) Household composition No No Yes No No Yes Round dummy No No Yes No No Yes Observations 1,530 1,530 1,530 1,530 1,530 1,530 Number of households 550 550 550 550 550 550 Cluster bootstrap standard errors in parentheses: *** p<0.01, ** p<0.05, * p<0.1. Note: Panel (a) reports the effects of IBLI uptake and volume of TLUs insured on vignette adjusted SWB relative to Borana pastoralists in log-odds units. Panel (b) reports the marginal effects for the main results in panel (a) – IBLI/ TLUs insured, lapsed IBLI/ TLUs insured and number of TLUs owned. The marginal effects estimates in panel (b) show the effects of these variables on the probability of reporting one of the seven unique scales of SWB. In column 3 for example, IBLI uptake reduces the probability of reporting SWB=1 by 4.9% and increases the probability of reporting SWB=7 by 1.7%. A unit increase in TLUs owned reduces the probability of reporting SWB=1 by 0.4% and increases the probability of reporting SWB=7 by 0.1% 40 Table A7: Ordered logit regression: Estimates for SWB using IBLI uptake and volume of TLUs insured IBLI uptake TLUs insured (1) (2) (3) (4) (5) (6) Dependent variable: SWB panel (a) Predicted IBLI/ TLUs insured 0.759*** 0.597** 0.646** 0.126*** 0.128*** 0.126*** (0.250) (0.238) (0.282) (0.041) (0.041) (0.034) Predicted lapsed IBLI/ TLUs insured -0.724*** -0.734*** -0.730*** -0.136*** -0.133*** -0.135*** (0.223) (0.230) (0.233) (0.033) (0.034) (0.022) Number of TLUs owned 0.034*** 0.030*** 0.029*** 0.034*** 0.030*** 0.029*** (0.006) (0.006) (0.006) (0.006) (0.006) (0.004) Asset index 0.205** 0.187** 0.231*** 0.213*** (0.086) (0.089) (0.080) (0.055) Annual income (’000 Birr) 0.003 0.002 0.003 0.002 (0.003) (0.003) (0.003) (0.002) Household head gender (Male=1) 0.406** 0.346*** (0.191) (0.131) Household head age 0.023 0.025 (0.021) (0.020) Household head age squared -0.0002 -0.0003 (0.0002) (0.0003) Household size -0.066 -0.051 (0.055) (0.043) Household head schooling -0.019 -0.017 (0.032) (0.024) panel (b) Predicted IBLI/ TLUs insured prob(SWB=1) -0.051*** -0.040** -0.043** -0.008*** -0.009*** -0.008*** (0.017) (0.016) (0.018) (0.003) (0.003) (0.003) prob(SWB=2) -0.077*** -0.060** -0.064** -0.013*** -0.013*** -0.013*** (0.026) (0.025) (0.027) (0.004) (0.004) (0.004) prob(SWB=3) -0.003 -0.002 -0.003 -0.0005 -0.0003 -0.0005 (0.006) (0.005) (0.005) (0.001) (0.001) (0.001) prob(SWB=4) 0.098*** 0.077** 0.083** 0.016*** 0.016*** 0.016*** (0.033) (0.031) (0.034) (0.005) (0.005) (0.005) prob(SWB=5) 0.032*** 0.025** 0.027** 0.005*** 0.005*** 0.005*** (0.011) (0.010) (0.011) (0.002) (0.002) (0.002) Predicted lapsed IBLI/ TLUs insured prob(SWB=1) 0.048*** 0.049*** 0.048*** 0.009*** 0.009*** 0.009*** (0.015) (0.016) (0.016) (0.002) (0.002) (0.002) prob(SWB=2) 0.073*** 0.074*** 0.073*** 0.014*** 0.013*** 0.013*** (0.023) (0.024) (0.024) (0.003) (0.003) (0.003) prob(SWB=3) 0.003 0.002 0.003 0.0005 0.0004 0.0005 (0.006) (0.006) (0.006) (0.001) (0.001) (0.001) prob(SWB=4) -0.094*** -0.094*** -0.093*** -0.017*** -0.017*** -0.017*** (0.030) (0.030) (0.030) (0.004) (0.004) (0.004) prob(SWB=5) -0.031*** -0.031*** -0.031*** -0.006*** -0.006*** -0.006*** (0.010) (0.010) (0.010) (0.002) (0.002) (0.002) Number of TLUs owned prob(SWB=1) -0.002*** -0.002*** -0.002*** -0.002*** -0.002*** -0.002*** (0.0004) (0.0004) (0.0004) (0.0004) (0.0004) (0.0004) prob(SWB=2) -0.003*** -0.003*** -0.003*** -0.003*** -0.003*** -0.003*** (0.001) (0.001) (0.001) (0.001) (0.001) (0.001) prob(SWB=3) -0.0001 -0.0001 -0.0001 -0.0001 -0.0001 -0.0001 (0.0003) (0.0003) (0.0003) (0.0003) (0.0003) (0.0003) prob(SWB=4) 0.004*** 0.004*** 0.004*** 0.004*** 0.004*** 0.004*** (0.001) (0.001) (0.001) (0.001) (0.001) (0.001) prob(SWB=5) 0.001*** 0.001*** 0.001*** 0.001*** 0.001*** 0.001*** (0.0002) (0.0003) (0.0003) (0.0002) (0.0003) (0.0003) Household composition No No Yes No No Yes Round dummy No No Yes No No Yes Observations 1,530 1,530 1,530 1,530 1,530 1,530 Number of households 550 550 550 550 550 550 Cluster bootstrap standard errors in parentheses: *** p<0.01, ** p<0.05, * p<0.1. Note: Panel (a) reports the effects of IBLI uptake and volume of TLUs insured on raw (unadjusted) SWB in log-odds units. Panel (b) reports the marginal effects for the main results in panel (a) – IBLI/ TLUs insured, lapsed IBLI/ TLUs insured and number of TLUs owned. The marginal effects estimates in panel (b) show the effects of these variables on the probability of reporting one of the five unique scales of SWB. In column 3 for example, IBLI uptake reduces the probability of reporting SWB=1 by 4.3% and increases the probability of reporting SWB=5 by 2.7%. In column 6, unit increase in the number of TLUs insured reduces the probability of reporting SWB=1 by 0.8% and increases the probability of reporting SWB=5 by 0.5% 41 Table A8: Ordered logit regression: Vignette adjusted SWB estimates using IBLI uptake and TLUs insured – panel households only IBLI uptake TLUs insured (1) (2) (3) (4) (5) (6) Dependent variable: SWB panel (a) Predicted IBLI/ TLUs insured 0.921*** 0.799*** 0.944*** 0.143*** 0.145*** 0.154*** (0.274) (0.269) (0.300) (0.041) (0.040) (0.042) Predicted lapsed IBLI/ TLUs insured -0.486** -0.449** -0.452** -0.074** -0.066** -0.068** (0.203) (0.190) (0.190) (0.031) (0.031) (0.031) Number of TLUs owned 0.013 0.010 0.009 0.014* 0.009 0.009 (0.008) (0.007) (0.008) (0.007) (0.007) (0.008) Asset index 0.237* 0.186 0.315*** 0.280*** (0.129) (0.135) (0.110) (0.106) Annual income (’000 Birr) 0.004* 0.004* 0.004* 0.004* (0.003) (0.003) (0.003) (0.003) Household head gender (Male=1) 0.801** 0.663** (0.372) (0.335) Household head age -0.049 -0.047 (0.042) (0.038) Household head age squared 0.0005 0.0004 (0.0004) (0.0003) Household size -0.208** -0.173** (0.093) (0.086) Household head schooling 0.034 0.034 (0.059) (0.062) panel (b) Predicted IBLI/ TLUs insured prob(SWB=1) -0.045*** -0.039*** -0.046*** -0.007*** -0.007*** -0.007*** (0.014) (0.014) (0.015) (0.002) (0.002) (0.002) prob(SWB=2) -0.038*** -0.032*** -0.039*** -0.006*** -0.006*** -0.006*** (0.011) (0.011) (0.012) (0.002) (0.002) (0.002) prob(SWB=3) -0.030*** -0.026*** -0.031*** -0.005*** -0.005*** -0.005*** (0.009) (0.009) (0.010) (0.002) (0.002) (0.002) prob(SWB=4) 0.006** 0.005* 0.007** 0.001* 0.001* 0.001* (0.003) (0.003) (0.003) (0.001) (0.001) (0.001) prob(SWB=5) 0.060*** 0.051*** 0.062*** 0.009*** 0.009*** 0.010*** (0.018) (0.018) (0.019) (0.003) (0.003) (0.003) prob(SWB=6) 0.029*** 0.025*** 0.030*** 0.005*** 0.005*** 0.005*** (0.009) (0.008) (0.009) (0.001) (0.001) (0.001) prob(SWB=7) 0.018*** 0.015*** 0.018*** 0.003*** 0.003*** 0.003*** (0.005) (0.005) (0.006) (0.001) (0.001) (0.001) Predicted lapsed IBLI/ TLUs insured prob(SWB=1) 0.024** 0.022** 0.022** 0.004** 0.003** 0.003** (0.009) (0.009) (0.009) (0.002) (0.002) (0.002) prob(SWB=2) 0.020*** 0.018** 0.018** 0.003** 0.003** 0.003** (0.008) (0.008) (0.008) (0.001) (0.001) (0.001) prob(SWB=3) 0.016** 0.014** 0.015** 0.002** 0.002* 0.002** (0.006) (0.006) (0.006) (0.001) (0.001) (0.001) prob(SWB=4) -0.003* -0.003* -0.003* -0.0005 -0.0004 -0.0005 (0.002) (0.002) (0.002) (0.0003) (0.0003) (0.0003) prob(SWB=5) -0.032*** -0.029** -0.029** -0.005** -0.004** -0.004** (0.012) (0.012) (0.012) (0.002) (0.002) (0.002) prob(SWB=6) -0.016*** -0.014** -0.014** -0.002** -0.002** -0.002** (0.006) (0.006) (0.006) (0.001) (0.001) (0.001) prob(SWB=7) -0.009*** -0.009** -0.008** -0.001** -0.001** -0.001** (0.004) (0.004) (0.004) (0.001) (0.001) (0.001) Number of TLUs owned prob(SWB=1) -0.001 -0.0005 -0.0005 -0.001* -0.0004 -0.0005 (0.0004) (0.0004) (0.0004) (0.0004) (0.0004) (0.0004) prob(SWB=2) -0.001 -0.0004 -0.0004 -0.001* -0.0004 -0.0004 (0.0003) (0.0003) (0.0003) (0.0003) (0.0003) (0.0003) prob(SWB=3) -0.0004 -0.0003 -0.0003 -0.0004 -0.0003 -0.0003 (0.0003) (0.0003) (0.0003) (0.0003) (0.0003) (0.0003) prob(SWB=4) 0.00008 0.0001 0.0001 0.0001 0.0001 0.0001 (0.00007) (0.0001) (0.0001) (0.0001) (0.0001) (0.0001) prob(SWB=5) 0.001 0.001 0.001 0.001* 0.001 0.001 (0.001) (0.001) (0.001) (0.001) (0.001) (0.001) prob(SWB=6) 0.0004 0.0003 0.0003 0.0004* 0.0003 0.0003 (0.0003) (0.0002) (0.0003) (0.0003) (0.0002) (0.0003) prob(SWB=7) 0.0002 0.0002 0.0002 0.0003* 0.0002 0.0002 (0.0002) (0.0002) (0.0002) (0.0002) (0.0002) (0.0002) Household composition No No Yes No No Yes Round dummy No No Yes No No Yes Observations 1,395 1,395 1,395 1,395 1,395 1,395 Number of households 465 465 465 465 465 465 Cluster bootstrap standard errors in parentheses: *** p<0.01, ** p<0.05, * p<0.1. Note: Panel (a) reports the effects of IBLI uptake and volume of TLUs insured on vignette adjusted SWB in log-odds units. Panel (b) reports the marginal effects for the main results in panel (a) – IBLI/ TLUs insured, lapsed IBLI/ TLUs insured and number of TLUs owned. The marginal effects estimates in panel (b) show the effects of these variables on the probability of reporting one of the seven unique scales of SWB. In column 3 for example, IBLI uptake reduces the probability of reporting SWB=1 by 4.6% and increases the probability of reporting SWB=7 by 1.8%. In column 6, a unit increase in the number of TLUs insured reduces the probability of reporting SWB=1 by 0.7% and increases the probability of reporting SWB=7 by 0.3% 42 Table A9: Ordered logit regression: Vignette adjusted SWB estimates using IBLI uptake and volume of TLUs insured with omitted lapsed IBLI IBLI uptake TLUs insured (1) (2) (3) (4) (5) (6) Dependent variable: SWB panel (a) Predicted IBLI/ TLUs insured 0.641*** 0.552** 0.704** 0.105*** 0.108*** 0.115*** (0.245) (0.249) (0.282) (0.037) (0.037) (0.038) Number of TLUs owned 0.014** 0.012* 0.012* 0.015** 0.011* 0.012* (0.006) (0.007) (0.007) (0.006) (0.006) (0.007) Asset index 0.286** 0.243** 0.329*** 0.296*** (0.116) (0.123) (0.103) (0.102) Annual income (’000 Birr) 0.004 0.004 0.004 0.004 (0.002) (0.002) (0.002) (0.002) Household head gender (Male=1) 0.747** 0.635* (0.360) (0.333) Household head age -0.046 -0.040 (0.040) (0.038) Household head age squared 0.000 0.000 (0.000) (0.000) Household size -0.229** -0.197** (0.090) (0.085) Household head schooling 0.049 0.051 (0.051) (0.052) panel (b) Predicted IBLI/ TLUs insured prob(SWB=1) -0.033*** -0.029** -0.036** -0.005** -0.006*** -0.006*** (0.013) (0.013) (0.014) (0.002) (0.002) (0.002) prob(SWB=2) -0.026** -0.022** -0.029** -0.004** -0.004** -0.005*** (0.010) (0.010) (0.011) (0.002) (0.002) (0.002) prob(SWB=3) -0.019** -0.016** -0.021** -0.003** -0.003*** -0.003*** (0.008) (0.008) (0.009) (0.001) (0.001) (0.001) prob(SWB=4) 0.004* 0.004 0.005* 0.001 0.001 0.001 (0.002) (0.002) (0.003) (0.0005) (0.0005) (0.001) prob(SWB=5) 0.041** 0.035** 0.045** 0.007** 0.007*** 0.007*** (0.016) (0.016) (0.018) (0.003) (0.003) (0.003) prob(SWB=6) 0.021*** 0.018** 0.023*** 0.003*** 0.004*** 0.004*** (0.008) (0.008) (0.009) (0.001) (0.001) (0.001) prob(SWB=7) 0.012** 0.010** 0.013** 0.002** 0.002** 0.002** (0.005) (0.005) (0.005) (0.001) (0.001) (0.001) Number of TLUs owned prob(SWB=1) -0.001** -0.001* -0.001* -0.001** -0.001* -0.001* (0.0003) (0.0003) (0.0004) (0.0003) (0.0003) (0.0004) prob(SWB=2) -0.001** -0.0005* -0.000* -0.001** -0.0005* -0.0005* (0.0003) (0.0003) (0.0003) (0.0003) (0.0003) (0.0003) prob(SWB=3) -0.0004** -0.0003 -0.0004 -0.0004** -0.0003 -0.0004* (0.0002) (0.0002) (0.0002) (0.0002) (0.0002) (0.0002) prob(SWB=4) 0.0001 0.0001 0.0001 0.0001 0.0001 0.0001 (0.0001) (0.0001) (0.0001) (0.0001) (0.0001) (0.0001) prob(SWB=5) 0.001** 0.001* 0.001* 0.001** 0.001* 0.001* (0.0004) (0.0004) (0.0005) (0.0004) (0.0004) (0.0004) prob(SWB=6) 0.0005** 0.0004* 0.0004* 0.0005** 0.0004* 0.0004* (0.0002) (0.0002) (0.0002) (0.0002) (0.0002) (0.0002) prob(SWB=7) 0.0003** 0.0002* 0.0002 0.0003** 0.0002* 0.0002* (0.0001) (0.0001) (0.0001) (0.0001) (0.0001) (0.0001) Household composition No No Yes No No Yes Round dummy No No Yes No No Yes Observations 1,530 1,530 1,530 1,530 1,530 1,530 Number of households 550 550 550 550 550 550 Cluster bootstrap standard errors in parentheses: *** p<0.01, ** p<0.05, * p<0.1. Note: Panel (a) reports the effects of IBLI uptake and volume of TLUs insured on vignette adjusted SWB when lagged IBLI/ TLUs insured is omitted, in log-odds units. Panel (b) reports the marginal effects for the main results in panel (a) – IBLI/ TLUs insured and number of TLUs owned. The marginal effects estimates in panel (b) show the effects of these variables on the probability of reporting one of the seven unique scales of SWB. In column 3, IBLI uptake reduces the probability of reporting SWB=1 by 3.6% and increases the probability of reporting SWB=7 by 1.3%. In column 6, a unit increase in the number of TLUs insured reduces the probability of reporting SWB=1 by 0.6% and increases the probability of reporting SWB=7 by 0.2% 43 B Attrition Correction There was attrition of some sample households in the follow up rounds of the survey used in this paper. If the sample households who dropped out differ systematically from those who remained in the sample, inference becomes difficult due to attrition bias. In this section, we test whether households who dropped out of the sample introduce attrition bias into our estimates. We find that they do not. Between the baseline and second round survey 40 (about 8% of the sample) households dropped out, and in round three an additional 10 households (2% of sample) dropped out. Yet in round three, 10 of the 40 households who dropped out in round two returned and were re-interviewed. Following Fitzgerald, Gottschalk, and Moffitt (1998), we first check if attrition is random by estimating attrition probit equations for our outcome variables: IBLI uptake and SWB. Then, if attrition is found to be non-random, we make attrition bias correction to our estimates in Tables 4 and 5. We estimate the equations: pr(Aivt = 1) = τ0 + τ1 IBLIivt−1 + τ2 Xivt + τ3 Zivt + ψi + eivt (A1) and pr(Aivt = 1) = τ0 + τ1 SW Bivt−1 + τ2 Xivt + τ3 Zivt + ψi + eivt (A2) where, A is an attrition dummy variable that takes value one if a households attrites in any survey rounds or zero otherwise; X is a vector of household demographic characteristics, household composition, household income and wealth variables, Z is a vector of auxiliary variables that may affect attrition including discount and information treatments, group membership dummies, and exposure to various shocks. The right hand side variables also include lagged IBLI uptake and SWB. Appendix Table A10 presents probit estimates of the probability of attrition with lagged IBLI and SWB equations. Column 1 shows that all of the coefficients are individually insignificant, suggesting that attrition is random. Wald joint test of the group (auxiliary) variables (Chi- squared statistic of 26.08 with 23 degrees of freedom and p-value of 0.297) indicates that these variables are not jointly statistically significantly different from zero. Similarly, column two shows that all of the explanatory variables are statistically insignificant, except for the 44 discount coupon in sales period one, which is significant only at the 10% level. These results also suggest attrition is random. The resulting Chi-squared statistic of a joint Wald test of the group variables and discount coupon in sales period one of 26.37 with 24 degrees of freedom and p-value of 0.335 indicates attrition is random. This leads us to conclude that our estimates of IBLI participation and the effect of IBLI on SWB are likely free of attrition bias, and that no attrition correction is required. 45 Table A10: Attrition probit estimates Attrition on Attrition on IBLI status SWB Dependent variable: Attrition dummy (1) (2) IBLIt−1 -0.219 (0.289) SW Bt−1 0.086 (0.126) Discount: SP1 only -0.868 -0.898* (0.456) (0.545) Discount: SP2 only -0.673 -0.670 (0.695) (0.730) Value of discount (%) SP1 -0.650 -0.692 (0.786) (0.959) Value of discount (%) SP2 -0.128 -0.220 (0.992) (1.416) Comic book: SP1 only 0.430 0.482 (0.409) (0.416) Household head gender (Male=1) -0.317 -0.352 (0.282) (0.325) Household head age -0.011 -0.014 (0.041) (0.048) Household age squared 0.00002 0.00005 (0.0004) (0.0005) Household size -0.306 -0.321 (0.380) (0.472) Household head highest grade -0.053 -0.055 (0.115) (0.098) Number of female household members -0.041 -0.050 (0.104) (0.122) Number of household members under 5 0.266 0.292 (0.389) (0.472) Number of household members between 5 and 15 0.282 0.297 (0.377) (0.475) Number of household members between 15 and 64 0.278 0.297 (0.362) (0.442) Number of TLUs owned 0.002 0.001 (0.007) (0.007) Asset index -0.027 -0.030 (0.217) (0.250) Annual income (’000 Birr) 0.001 0.0003 46 (0.007) (0.009) Net transfers (’000 Birr) -0.041 -0.036 (0.038) (0.036) If household head is village water point group -0.157 -0.189 (0.464) (0.604) If household head is village pasture group 0.006 0.016 (0.401) (0.465) If household head is a member of Iqub 0.726 0.701 (0.534) (0.624) Animal sickness or death 0.016 0.014 (0.259) (0.274) Animal loss or theft 0.083 0.077 (0.277) (0.326) Insecurity/Violence/Fights 0.218 0.223 (0.256) (0.314) Human sickness -0.068 -0.070 (0.261) (0.304) Low prices for animals one wishes to sell 0.134 0.119 (0.214) (0.243) Crop disease -0.137 -0.148 (0.206) (0.258) Lack of food -0.079 -0.063 (0.417) (0.465) High food prices 0.050 0.076 (0.392) (0.524) Land scarcity/disputes 0.063 0.089 (0.283) (0.317) Lack of employment opportunities -0.442 -0.474 (0.320) (0.407) Flood damage 0.013 -0.007 (0.285) (0.329) Constant -0.113 -0.299 (1.051) (1.306) Observations 1,012 1,012 Number of groups (households) 538 538 Cluster bootstrap standard errors in parentheses: *** p<0.01, ** p<0.05, * p<0.1. 47