Bio-based and Applied Economics 7(3): 265-277, 2018 ISSN 2280-6180 (print) © Firenze University Press ISSN 2280-6172 (online) www.fupress.com/bae DOI: 10.13128/bae-7679 Short Communication A preliminary test on risk and ambiguity attitudes, and time preferences in decisions under uncertainty: towards a better explanation of participation in crop insurance schemes Attilio ColettA1, elisA GiAmpietri2, FAbio GAetAno sAnterAmo3,*, simone severini1, sAmuele trestini2 1 Università degli Studi della Tuscia, Department of Agriculture and Forest Sciences (DAFNE), Viterbo, Italy 2 Università degli Studi di Padova, Department of Land, Environment, Agriculture and Forestry (TESAF), Pado- va, Italy 3 Università degli Studi di Foggia, Department of Agricultural, Food and Environmental Sciences (SAFE), Fog- gia, Italy Date of submission: 2018 26th, July; accepted 2019 22nd, July Abstract. The exposure of farmers to different (and increasing) risks has been recog- nized by the EU policy, which supports several risk management tools through the Common Agricultural Policy (CAP). Despite the vulnerability of the agricultural sec- tor, and the attention paid at the EU level, the uptake of such tools is generally low across EU countries. The Italian case is emblematic: the uptake of subsidized crop insurance contracts is low, limited to few products, and concentrated in few areas. Coherently, the interest of policy makers toward explaining these characteristics and in gaining insights on the interventions that may help promoting participation is intense. This contribution investigates behavioral aspects linked to choices under risk and ambiguity, and account for time preferences in order to mimic the scenario faced by the potential adopters of the subsidized crop insurance contracts in Italy. Data are collected through questionnaires submitted to students from agricultural colleges in three administrative regions located in northern, central and southern Italy. Results show that attitude toward risk, ambiguity, and impatience are correlated with the intrinsic characteristics of respondents. In addition, some of those attitudes may help explaining decisions under uncertainty. Despite the empirical analysis is preliminary and focused on students, it allowed to validate a promising methodological approach capable of explaining farmer’s willingness to adopt (or renew) insurance contracts. By accounting for (currently under-investigated) behavioral aspects, it is likely to prove useful to re-design or implementing, more effectively, the current policies. Keywords. Insurance, subjective probabilities, risk preferences, choice experiment. JEL codes. D81, D83, Q18. *Corresponding author: fabio.santeramo@unifg.it 266 Attilio Coletta et alii 1. Introduction Risk affects all economic activities, and the agricultural sector shows specific factors that make yields, input and output prices highly variable. The increased volatility of these variables was shown in recent years, and it is possibly due to frequent adverse phenomena and extreme climatic events. At European level all countries are affected, and Italy seems one of the most spoiled country. The Italian agricultural sector is largely exposed to risky events, as shown by Trestini et al. in 2017. Among EU members, from 1998 to 2006 Italy registered the highest number of farms experiencing a decline in farm income exceeding -30% (on average) (European Commission, 2009); moreover, 35% of Italian farmers expe- rienced income decrease events from 2007 to 2013 (European Commission, 2017). According to the economic theory, price volatility should incentivize farmers to adopt risk management tools (RMT): put differently, the increasing uncertainty should increase the latent demand for RMT. The increasing uncertainty and the availability of new instru- ments introduced by the 2008 CAP Health Check should have favoured the diffusion of these policy instruments (e.g., mutual funds and subsidized insurance contracts). How- ever, the implementation of risk management tools is limited, and the adoption of these instruments is currently rather scarce. Such a contingent scenario is worrisome, provided that a correct use of risk management policies would allow EU countries to increase the resilience of their agricultural sector to external shocks. The EU Regulation 1305/2013 promotes three types of measures, respectively under art. 37, 38 and 39: crop insurance, mutual funds, and the income stabilization tool. The Italian Ministry has budgeted a large amount of financial resources to promote these measures but, despite a great attention and a large turmoil, the experiences on mutual funds and Income Stabilization Tool are scant (Severini et al., 2018; Trestini et al., 2018), and subsidized single crop insurances are still the most adopted RMT. However, the subsidized insurance programs are not always sto- ries of success. In Italy, participation in crop insurance programs is low, heterogeneous, and (recently) declining (Santeramo, 2019), making it a pressing issue for policymakers. This decline is also associated to recent policy changes. The last CAP reform has moved the support to RMT to the Rural Development Policy, changing the administrative rules of the system. In Italy this transition has resulted in a lack of familiarity with the rules, in delays in payments for subsidies and indemnifications and, at the end, in a reduced uptake of crop insurance schemes. The current literature falls short in explaining the peculiarities of crop insurance adoption in Italy, and more precisely, it has not explored the potential role of ambiguity aversion and time preferences on participation in crop insurance programs. Understanding the behavioral aspects of potential adopters of RMT is crucial to both design and implement effective policy interventions and avoid low and sparse uptake. The Italian case is an emblematic one and it allows to focus on long-standing issues that need to be solved at national and EU level. The Italian (subsidized) crop insurance system is characterized by high adoption rate in the north, and low participation rate in central and south regions. Apart from the main drivers of farmer behavior under uncertainty and of adoption of risk management tools, several attitudinal aspects are likely to matter. Departures from rationality and non-coherent choices with respect to risk perception help explaining farm- 267Towards a better explanation of participation in crop insurance schemes ers’ choices. A recent study (Sutter et al., 2013) suggests that attitudes toward ambiguity, due to incomplete information, as well as differences in risk perception, and in time pref- erences are likely to play a pivotal role for decisions under uncertainty. This paper is a preliminary attempt to assess the validity of an empirical methodol- ogy to evaluate if and how behavioral factors (risk and ambiguity attitudes and time pref- erences) may affect the decision-making process under uncertainty. Our setup has been inspired by the framework faced by potential adopters of crop insurance. The analy- sis, conducted on a sample of students of agricultural disciplines allows to conclude on whether the methodological approach is worth replication to a set of Italian farmers, rep- resentative of the latent demand for crop insurance contracts. The analysis is divided in two steps. First, we investigate how socio-economic charac- teristics tend to influence risk aversion, ambiguity aversion and time preferences. Second, we explore how socio-economic characteristics as well as risk aversion, ambiguity aversion and time preferences may help explaining choices under uncertainty (smoking, practicing sport and playing lottery). 2. On Italian insurance market and factors affecting farmers’ adoption 2.1 The Italian market for subsidized crop insurance contracts Risks linked to natural disasters have been recognized since long-time in agriculture as unexpected sources of losses for farmers, especially for those highly vulnerable that are not adopters of risk management strategies. The shift from ex post compensations to ex- ante measures, and to subsidized crop insurance contracts, has been a concrete effort to promote the diffusion of risk management strategies. According to ISMEA (2018), the Italian market (2004-2010) is characterized by a limited adoption of insurance contracts. Subsidized insurance market reached a maxi- mum of 265,000 contracts in 2008, followed by declines in the number of contract sub- scriptions. Differently, total compensation rose constantly, signalling the low (economic) sustainability of the system, exacerbated by an adversely selective participation process: as contacts’ prices rise, farmers with lower probability of facing adversities quit the mar- ket, contributing to the increase of the total amount of compensations paid by insur- ers (and by public funds). Since 2010 the public contribution to contracts decreased to 65% (according to EU Reg. 73/2009) and has been devoted (since 2014) to contracts that cover at least three climatic adversities. These changes do not seem to push the mar- ket too far. Last (public) data referred to 2015 (ISMEA, 2018) depicts a similar picture: from 2010 to 2015 contracts have decreased by 20% (from 210,000 to 168,000), while the insured area remained unaltered (+5%); the insured value raised by 20% as well (from 4.8 to 5.6 billion euro), and it has generated a 4% increase in the premium paid by farm- ers and through public funds (from 279 to 381 mil euro). The geographical distribution of contracts tends to be concentrated in northern regions, which account for more than 80% of the insured value (ISMEA, 2018). In addition, only few products account for most of the total insured value: indeed, apple, corn, rice, grapes, and tomatoes account for 2/3 of the covered value. 268 Attilio Coletta et alii 2.2 On the drivers of crop insurance uptake The identification of the drivers of crop insurance uptake is still open and vivid (Enjolras et al., 2011; Santeramo et al., 2016). More important, there has been a limited effort in investigating how farmers’ behavioral aspects may help explaining the adoption and/or renewal of crop insurance contracts, exception made for Menapace et al. (2015). Key drivers of uptake are the age and the income level: Ogurtsov et al. (2009) found a positive correlation for age and adoption of crop insurance contracts, while Wąs and Kob- us (2018), Liesivaara and Myyrä (2017) and van Winsen et al. (2016) suggested that the opposite is true; as for the income level, Menapace et al. (2015) found a positive correla- tion with uptake, while Wąs and Kobus (2018) and Farrin et al. (2016) concluded on the opposite direction for correlation. Ambiguous results have also been found for risk aversion, which has been found posi- tively correlated with age, according to Nielsen et al. (2013) and van Winsen et al. (2016), and negatively correlated according to Franken et al. (2017) and Goldstein et al. (2008). Heterogeneous results are also reported for the farm size, positively correlated with risk awareness in Franken et al. (2017), and negatively correlated with risk awareness accord- ing to van Winsen et al. (2016). Furthermore, the low participation level may be due to a low level of familiarity with the instrument (Santeramo, 2018 and 2019; Santeramo et al., 2016). Subscription of new contracts tend to be influenced by size, degree of crop diversification and irrigated area (Enjolras and Sentis, 2011; Finger and Lehmann, 2012); moreover, Santeramo et al. (2016) argued that farmers tend to consider crop diversification (and irrigation) and insurance contracts as alternate management strategies with a high degree of substitutability. The policy framework is also playing a role: for instance, greening requirements push toward crop diversification to help preserving the environment; measures of income support (e.g. direct payments or agri-environmental measures) are aimed at reducing famers’ income instability and may prove substitutes for other risk management tools (Severini et al., 2017). A contingent scenario, faced by Italian farmers, is that the bureaucratic aspects related to subscription and reimbursement procedures, and the delays in refunds (ISMEA, 2018), may have discouraged participation and renewal of crop insurance contracts. From 2010 to 2014 the share of new adopters (14%) of (subsidized) crop insurance contracts has exceeded the number of farmers who gave up (11%). Differently, and possibly due to the delays in payments and to the (perceived) ambiguity of the newly adopted rules, in 2015 the quitters overcame new adopters, and the net balance between new entrants and leavers was largely negative (-11%). 3. Methodology and data collection The above presented scenario has emphasized the importance of focusing on three specific aspects: risk aversion, ambiguity aversion, and time preferences. This paper inves- tigates how attitudes toward uncertainty (risk and ambiguity) as well as time preferences influence risky decisions. The dataset includes data on 50 students from three different universities (Faculty of Agricultural Sciences) in Italy: namely, the University of Padova 269Towards a better explanation of participation in crop insurance schemes (Padova) in the North, Tuscia University (Viterbo) in Central Italy and University of Fog- gia (Foggia) in the South. The research is part of a wider ongoing study aiming at investi- gating Italian farmers’ decision making under uncertainty: particularly, the broader aim is to study the factors influencing the insurance schemes’ uptake. The experimental method- ology is inspired by the canonical Holt and Laury (2002) choice lists and, more specifical- ly, by the approach proposed by Sutter et al. (2013). In order to elicit individual preferenc- es related to risk aversion, ambiguity aversion and time preferences, respondents received a structured questionnaire with three experiments and ten control questions. More specifically, the first and the second experiments (Fig. 1) are made by a list of 11 choices with two options each: at any given choice respondents choose between a sure payoff (option A), and a gamble (option B). The sure payoff is iteratively decreased (from 100 to 1€) so to elicit the indifference point between the lottery and the sure payoff. The lottery has been simulated by extracting a random number from a uniform distribution ranging from 1 to 100 being the number 50 excluded (in order to have symmetrical prob- ability distributions between the two outcomes). In the first experiment, aimed at eliciting risk preferences, respondents may win (for instance) 100€ if the randomly extracted num- ber ranges between 1 and 49, or nothing, if the randomly extracted number is larger than 51. In order to get respondents acquainted with the functioning of the lottery, respond- ents have been exposed to a computer simulation of ten random draws from 1 to 100 (the extraction of the number 50 implies a further extraction), and have been informed on the cases in which they would have won the lottery. The second experiment, aimed at eliciting ambiguity aversion, compares the choices for a sure payoff and a (ambiguous) lottery. The lottery pays out if, by extracting two random draws, the second extraction gives a larger number than the one extracted in the first place. The ambiguity arises by a peculiarity: the result of the first extraction is not revealed, whereas only the second extraction (and the outcome of the lottery) is revealed. For instance, by drawing the number 20 and succes- sively the number 35, the lottery results in a winning outcome. Finally, in the third experiment aiming at measuring time preferences (Fig. 2), respondents received two lists (blocks) of ten choice sets each. Each choice set consisted in two sure payoffs (A and B) that respondents may receive in different periods: option A is a “early payoff ” of 100€, whereas option B is a “late payoff ” which is increased from 100€ to 190€. Depending on respondents’ preference for receiving a sure payoff earlier (i.e., “now”) or later (i.e., “in 12 months”), we elicited respondents’ attitude in delaying the win (or, put differently, their impatience). Prior to the survey, we paid attention to ensuring that participants were able to under- stand the questions, and that the experiments were correctly explained. We design a ran- Figure 1. Example of a choice list for experiment 1 (risk attitude) and 2 (ambiguity attitude) Option A Option B 1 Sure payoff (100€) Lottery 2 Sure payoff (90€) Lottery 3 Sure payoff (80€) Lottery Source: own elaboration. 270 Attilio Coletta et alii dom lottery incentive system (Cubitt et al., 2019), often used in individual choice experi- ments, to motivate respondents to reveal their true preferences: at the end of the experi- ments we ran a real lottery with the ten percent of (randomly selected) respondents: if their questionnaires did not present incoherent answers (as found in all cases), they played the game included in the questionnaire with the possibility of winning part of the money of the bet (more precisely, 10% of the money at stake), in case of favourable outcome. The individual Certainty Equivalent (CE) has been calculated for experiment 1 and 2 (CEr and CEa, respectively), as midpoint between the two consequent payoffs for which the interviewee switched from option A (i.e., sure payoff) to option B (i.e., gam- ble). Accordingly, CE represents the payoff that makes the individual indifferent between receiving the sure amount and gambling. To measure risk attitude (experiment 1), we cal- culated the coefficient of risk aversion (r) as follows (Sutter et al., 2013): r = 1− CEr π (1) with π representing the prize of the gamble (i.e., 100€). This coefficient ranges from 0 to 1, with values of r larger than 0.5 indicating risk aversion, whereas smaller than 0.5 risk lov- ing and equal to 0.5 risk neutrality. Moreover, in the second experiment we measured the coefficient of ambiguity attitude (a) as follows: a = CEr − CEa CEr + CEa (2) The coefficient a ranges from -1 to 1, with negative numbers representing ambigu- ity loving, 0 standing for ambiguity neutrality and positive numbers indicating ambigu- ity aversion. As regards the third experiment, we calculated the Future Equivalent (FE) of the fixed payoff as the midpoint between the two consequent later payoffs where the interviewee decided to switch from option A to B. The larger the FE, the larger the aver- sion for delayed payments (i.e., impatience). Finally, in order to control for the main driv- ers of decisions under uncertainty, we collected information on age (age), gender (gen- der), number of university credits achieved (ECTS credits), average grade (max 30) (aver- age grade), and on whether the respondent does not have a technical high school degree (degree), on smoking habits (being a smoker), on habits to practice physical activity (sport practicing), and on habits to play lottery or sport betting at least once a month (playing Figure 2. Example of a choice list for experiment 3 (time preference). Option A Option B 1 Receive 100€ today Receive 100€ in 12 months 2 Receive 100€ today Receive 110€ in 12 months 3 Receive 100€ today Receive 120€ in 12 months Source: own elaboration. 271Towards a better explanation of participation in crop insurance schemes lottery). Finally, we recorded whether the respondent is owner (or son of the owner) of a farm (family farm) and, whether the respondent has ever worked on a farm even for a short period of time (farmworker). The empirical strategy is admittedly simple, yet rigorous and comparable with the approach suggested in Sutter et al. (2013). First, we use a linear regression to conclude on the effects of some socio-demographic variables on: i) the coefficient of risk aversion (r), ii) the coefficient of ambiguity aversion (a), iii) time preferences (i.e., future equivalent at 12 months). Second, we use a linear regression to investigate how risk aversion, ambigu- ity aversion and time preferences (FE_12m) influence behaviors characterized by decisions under uncertainty: i) being a smoker; ii) sport practicing; iii) playing lottery. 4. Hypothesis testing and results As shown in Table 1, the sample consists of 78 observations, mostly male students (78%). Most participants have not a technical high school background (51%), are not smokers (64%), practice sports activities (60%), and do not play lotteries (80%). The aver- age number of credits acquired by sampled students is 132, while the average grade is 26. In terms of coefficients of risk aversion and risk ambiguity, we have quite heterogene- ous results: the coefficient of risk aversion ranges from 0.05 to 0.95 and the coefficient of ambiguity aversion ranges from -0.50 to 0.83. Similarly, we have time preferences com- puted at 12 months ranging from 105 to 185. The sample is mainly composed of risk averse (51%) and ambiguity averse students (51%), whereas the future equivalent shows a greater impatience for risk neutral and ambiguity averse subjects (Table 2). We regress attitudes toward risk and ambiguity on control factors (Table 3). The con- sidered observable characteristics do not allow to explain these attitudes. Regarding risk aversion, only the variable “degree” is positively correlated with risk aversion, regardless of students’ career characteristics (number of credits acquired) and average grade, and of respondent’s social characteristics (gender, age, farm owner and farming experience). There are no significant coefficients in the case of ambiguity aversion. Results seems to be in line with studies (e.g. Sutter et al., 2013) that refer risk attitude and ambiguity not influenced by ordinarily observable characteristics. As shown in Table 4, we also found a positive significant correlation between the degree of impatience and gender, degree and past experience in farm work, showing that males with non-technical degree are less impatient, while subjects who already had a work experience related to agricultural sector are more impatient. Conversely, we did not find any relevant effect for risk and ambiguity aversion. In general, we found that attitudes toward uncertainty (risk aversion, ambiguity aversion, and impatience) are correlated with intrinsic characteristics of the students, hereafter referred as control factors. Following Sutter et al. (2013) we use the control factors (age, gender, degree, ECTS credits, average grade, family farm, and farmworker) and the attitudes toward risk, ambi- guity and time, to explain decisions under uncertainty. We regress “being a smoker”, “sport practicing” and “playing lottery” on control factors and variables on attitudes. We found that average grade and risk aversion are statistically significant having a negative effect on being a smoker, whereas impatience has a slight positive effect on the 272 Attilio Coletta et alii same characteristic (Table 5). Impatience seems to play a slight role on sport practicing too, being instead negatively correlated. Regarding playing lottery, a significant positive correlation emerged for gender (all respondents that practice gambling are males), num- Table 1. Descriptive statistics of the sample (N = 78). Variable Type   % Mean Std Min Max Age Continuous     23.39 2.22 20 29 Gender Dummy 1 = male 78.20     0 = female 21.80         Degree1 Dummy 1 = yes 51.30         0 = no 48.70         ECTS credits2 Continuous     131.51 57.99 23 300 Average grade (max 30) Continuous     25.72 2.04 21 29.7 Family farm Dummy 1 = yes 28.20     0 = no 71.80     Farm worker Dummy 1 = yes 61.50     0 = no 38.50     Being a smoker Dummy 1 = yes 35.90     0 = no 64.10     Sport practicing Dummy 1 = yes 60.30     0 = no 39.70     Playing lottery Dummy 1 = yes 20.50     0 = no 79.50 r Continuous     0.48 0.16 0.05 0.95 a Continuous     0.08 0.22 -0.50 0.83 FE_12m Continuous     146.54 20.83 105 185 1 Subjects without a technical high school background (“Liceo” in Italy). 2 ECTS credits express the volume of learning based on the defined learning outcomes and their asso- ciated workload. 60 ECTS credits are allocated to the learning. Table 2. Risk and ambiguity attitude (%) and future equivalent (N = 78). Category % Average FE_12m1 Risk averse 51.3% 146.50 (20.07) Risk neutral 24.4% 149.21 (24.79) Risk seeker 24.4% 143.95 (18.83) Ambiguity averse 51.3% 148.00 (20.78) Ambiguity neutral 19.2% 147.00 (23.36) Ambiguity seeker 29.5% 143.70 (19.84) 1 Standard deviations are reported in parentheses. 273Towards a better explanation of participation in crop insurance schemes Table 3. OLS - Risk Aversion (r) and Ambiguity Aversion (a). Dep. Var. . Risk Aversion (r) Ambiguity Aversion (a) β S.E. P>|t| β S.E. P>|t| Age 0.003 0.010 0.756 -0.012 0.013 0.355 Gender 0.033 0.046 0.476 0.008 0.063 0.900 Degree 0.068 0.040 0.088* -0.082 0.055 0.137 ECTS credits -0.001 0.001 0.185 0.001 0.001 0.448 Average grade 0.008 0.010 0.456 0.015 0.014 0.307 Family farm -0.032 0.046 0.486 -0.036 0.064 0.568 Farmworker 0.006 0.042 0.888 0.024 0.058 0.678 cons 0.278 0.328 0.401 -0.036 0.457 0.983 Obs 78 78 Prob > F 0.574 0.695 Adj R2 -0.017 -0.031 Note: *** Significant at the 1% level; ** Significant at the 5% level; * Significant at the 10% level. Table 4. OLS - Impatience (FE_12m). Dep. Var. Future equivalent 12 months (FE_12m) β S.E. P>|t| Age 0.039 1.234 0.975 Gender -9.918 5.831 0.094* Degree -8.656 5.146 0.097* ECTS credits -0.037 0.053 0.484 Average grade 0.998 1.330 0.455 Family farm -7.290 5.873 0.219 Farmworker 9.760 5.331 0.072* r 11.212 16.183 0.491 a 9.264 11.626 0.428 cons 127.054 42.084 0.004 Obs 78 Prob > F 0.206 Adj R2 0.045 Note: *** Significant at the 1% level; ** Significant at the 5% level; * Significant at the 10% level. 274 Attilio Coletta et alii ber of credits acquired (with a positive slight coefficient close to zero) and being part of a family involved in farming activities. Average grade shows negative correlation indeed. Respondents showing little risk aversion and high levels of impatience smoke more, whereas less impatient individuals practice sport more. Men are found to play lottery more than women. As shown by “ECTS credits”, students up to date with credits play lottery more, whereas “average grade” shows that best students play lottery and smoke to a less- er extent. Interestingly, the higher the impatience (i.e., subjects who have a higher future equivalent with 12 month-delay condition), the less they practice sport. Lastly, ambiguity aversion coefficients don’t show significant relations with the analysed dependent variables. To summarize, both observable characteristics and behavioral characteristics (risk aversion, ambiguity aversion and time preferences) help explaining choices under uncer- tainty, particularly smoking and playing lottery. It is important to note that, as expected, risk aversion is negatively correlated with smoking while impatience is positively corre- lated with smoking while negatively with practicing sport. 5. Concluding remarks Risk management policies for the primary sector are under the spotlight in the EU: large subsidies have been granted for crop insurance programs and mutual funds. The EU Regulation 1305/2013 establishes rules and funds that may be adopted by Member States to promote participation in crop insurance programs (art. 37), to start and manage mutual funds (art. 38) and to enhance the start of the Income Stabilization Tool (art. 39). Despite Table 5. OLS Estimates on being a smoker, sport practicing, and playing lottery. Dep. Var. Being a smoker Sport practicing Playing lottery β S.E. P>|t| β S.E. P>|t| β S.E. P>|t| Age 0.029 0.028 0.310 -0.022 0.030 0.465 -0.010 0.022 0.643 Gender -0.091 0.136 0.506 0.169 0.146 0.248 0.265 0.108 0.016** Degree 0.139 0.120 0.250 -0.071 0.128 0.584 0.116 0.095 0.225 ECTS credits -0.001 0.001 0.986 0.001 0.001 0.739 0.002 0.001 0.019** Average grade -0.056 0.031 0.074* 0.038 0.033 0.250 -0.063 0.024 0.011** Family farm -0.036 0.136 0.792 0.078 0.145 0.594 0.215 0.107 0.050* Farmworker 0.075 0.125 0.548 -0.033 0.134 0.805 0.021 0.099 0.830 r -0.914 0.371 0.016** 0.232 0.397 0.561 -0.247 0.294 0.404 a -0.394 0.267 0.145 0.242 0.286 0.399 -0.170 0.211 0.423 FE_12m 0.005 0.003 0.093* -0.005 0.003 0.079* -0.001 0.002 0.523 cons 0.850 1.025 0.410 0.636 1.095 0.563 1.767 0.811 0.033 Obs. 78 78 78 Prob > F 0.134 0.590 0.008 Adj R2 0.069 -0.021 0.179 Note: *** Significant at the 1% level; ** Significant at the 5% level; * Significant at the 10% level. 275Towards a better explanation of participation in crop insurance schemes the clear interest of the policymakers, the academic debate seems behind. The economic literature provides several hints to explain farmers’ uptake in crop insurance programs, but several determinants (other than farm size, farmers’ education, relationships with other risk management strategies, and insurance premia) are still under-investigated. In particular, while the literature on insurance programs (i.e. health, car and life insurance) has emphasized the role of information, and of individual attitudes toward uncertainty, ambiguity and impatience, there is little evidence on the role of ambiguity and impatience on farmers’ decision to adopt crop insurance contracts. Based on these premises, we tested the validity of a methodology in exploring how risk and ambiguity aversion, and impatience may influence the decision-making process for risky activities. Our test, conducted on a sample of students, has been calibrated on behavioral aspects that are likely to matter for potential adopters of (subsidized) crop insurance contracts. We asked students involved in university programs related to agricul- tural sciences to declare if they experienced working in a farm. Similarly, we investigated decisions under uncertainty proxying risky decisions such as those related to the adoption of crop insurance programs. We found that the attitudes toward uncertainty (risk aversion, ambiguity aversion, and time preferences) are weakly correlated with some intrinsic characteristics of the students. These attitudes cannot be satisfactorily explained by few observable characteristics. In contrast, we found evidence that attitudes toward risk and impatience may help explain- ing agents’ decisions under uncertainty. This suggests including agents’ attitudes in future research to prevent biased inference due to missing explanatory factors which would lead to ineffective policy recommendations. Despite the analysis is still preliminary and applied to students, the approach we have taken seems promising in explaining potential residual factors that may affect farmer’s willingness to adopt (or renew) insurance contracts. Hence, future research on this latter issue should take into consideration not only farmers’ risk aversion but ambiguity aversion and time preferences as well. These factors may be used to explain the limited (and het- erogeneous) uptake of insurances. Furthermore, the empirical findings may help to bet- ter design and manage future policy measures: understanding the role of time preferences may be useful to address how delayed payments of reimbursements and indemnities may discourage participation. 6. References Cubitt, R., Van De Kuilen, G. and Mukerji, S. (2019). Discriminating Between Models of Ambiguity Attitude: A Qualitative Test. Journal of the European Economic Association. doi: 10.1093/jeea/jvz005/5424161 Enjolras, G. and Sentis, P. (2011). Crop Insurance Policies and Purchases in France. Agri- cultural Economics 42(4): 475-486. European Commission (2009). Income Variability and Potential Cost of Income Insurance for EU. AGRI L.1/L.3/ D (2009). Brussels. European Commission (2017). Risk Management Schemes in EU Agriculture Dealing with Risk and Volatility. Eu Agricultural Markets Briefs No 12, September 2017. Brussels. 276 Attilio Coletta et alii Farrin, K., Miranda, M.J. and O’Donoghue, E. (2016). How Do Time and Money Affect Agricultural Insurance Uptake? A New Approach to Farm Risk Management Analy- sis. U.S. Department of Agriculture, Economic Research Service No. 212. Finger, R. and Lehmann, N. (2012). The Influence of Direct Payments on Farmers’ Hail Insurance Decisions. Agricultural Economics 43(3): 343-354. Franken, J.R.V., Pennings, J.M.E. and Garcia, P. (2017). Risk Attitudes and the Structure of Decision-Making: Evidence from the Illinois Hog Industry. Agricultural Economics 48(1): 41-50. Goldstein, D.G., Johnson, E.J. and Sharpe, W.F. (2008). Choosing Outcomes Versus Choosing Products: Consumer-Focused Retirement Investment Advice. Journal of Consumer Research 35(3): 440-456. Holt, C.A. and Laury, S.K. (2002). Risk Aversion and Incentive Effects. American Econom- ic Review 92(5): 1644-1655. ISMEA (2018). Rapporto sulla Gestione del Rischio in Italia: Stato dell’ARTE e Scenari Evolutivi per la Stabilizzazione dei Redditi in Agricoltura. Roma: Romana Editrice. Liesivaara, P. and Myyrä, S. (2017). The Demand for Public-Private Crop Insurance and Government Disaster Relief. Journal of Policy Modeling 39(1): 19-34. Menapace, L., Colson, G. and Raffaelli, R. (2015). A Comparison of Hypothetical Risk Attitude Elicitation Instruments for Explaining Farmer Crop Insurance Purchases. European Review of Agricultural Economics 43(1): 113-135. Nielsen, T., Keil, A. and Zeller, M. (2013). Assessing Farmers’ Risk Preferences and their Determinants in a Marginal Upland Area of Vietnam: A Comparison of Multiple Elicitation Techniques. Agricultural Economics 44(3): 255-273. Ogurtsov, V.A., van Asseldonk, M.A. and Huirne, R.B. (2009). Purchase of Catastrophe Insurance by Dutch Dairy and Arable Farmers. Review of Agricultural Economics 31(1): 143-162. Santeramo, F.G. (2018). Imperfect Information and Participation in Insurance Markets: Evidence from Italy. Agricultural Finance Review 78(2): 183-194. Santeramo, F.G. (2019). I Learn, you Learn, we Gain: Experience in Crop Insurance Mar- kets. Applied Economic Perspectives and Policy 41(2): 284-304. Santeramo, F.G., Goodwin, B.K., Adinolfi, F. and Capitanio, F. (2016). Farmer Participa- tion, Entry and Exit Decisions in the Italian Crop Insurance Program. Journal of Agricultural Economics 67(3): 639-657 Severini, S., Biagini, L. and Finger, R. (2018). Modeling Agricultural Risk Management Policies – The Implementation of the Income Stabilization Tool in Italy. Journal of Policy Modeling 41(1): 140–155 Severini, S., Tantari, A. and Di Tommaso, G. (2017). Effect of Agricultural Policy on Income and Revenue Risks in Italian Farms. Implications for the Upload of Risk Management Policies. Agricultural Finance Review 77(2): 295-311. Sutter, M., Kocher, M.G., Glätzle-Rüetzler, D. and Trautmann, S.T. (2013). Impatience and Uncertainty: Experimental Decisions Predict Adolescents’ Field Behavior. American Economic Review 103(1): 510-31. Trestini, S., Giampietri, E. and Boatto, V. (2017). Toward the Implementation of the Income Stabilization Tool: An Analysis of Factors Affecting the Probability of Farm Income Losses in Italy. New Medit 16(4): 24-30. 277Towards a better explanation of participation in crop insurance schemes Trestini, S., Szathvary, S., Pomarici, E. and Boatto, V. (2018). Assessing the Risk Profile of Dairy Farms: Application of the Income Stabilisation Tool in Italy. Agricultural Finance Review 78(2): 195-208. van Winsen, F., de Mey, Y., Lauwers, L., Van Passel, S., Vancauteren, M. and Wauters, E. (2016). Determinants of Risk Behavior: Effects of Perceived Risks and Risk Atti- tude on Farmer’s Adoption of Risk Management Strategies. Journal of Risk Research 19(1): 56-78. Wąs, A. and Kobus, P. (2018). Factors Differentiating the Level of Crop Insurance at Polish Farms. Agricultural Finance Review 78(2): 209-222. Positive Mathematical Programming and Risk Analysis Quirino Paris The hedonic contents of italian super premium extra-virgin olive oils Luca Cacchiarelli1,*, Anna Carbone2, Tiziana Laureti1, Alessandro Sorrentino1 Corporate R&D and the performance of food-processing firms: Evidence from Europe, Japan and North America Heinrich Hockmann1, Pedro Andres Garzon Delvaux2,*, Peter Voigt3, Pavel Ciaian2, Sergio Gomez y Paloma2 Can menu labeling affect away-from-home-dietary choices? Elena Castellari1,*, Stéphan Marette2, Daniele Moro3, Paolo Sckokai1 A preliminary test on risk and ambiguity attitudes, and time preferences in decisions under uncertainty: towards a better explanation of participation in crop insurance schemes Attilio Coletta1, Elisa Giampietri2, Fabio Gaetano Santeramo3,*, Simone Severini1, Samuele Trestini2