J Agric Econ. 2024;00:1–24. | 1wileyonlinelibrary.com/journal/jage Received: 8 October 2023 | Revised: 26 January 2024 | Accepted: 25 March 2024 DOI: 10.1111/1477-9552.12582 O R I G I N A L A R T I C L E Are farm input subsidies a disincentive for integrated pest management adoption? Evidence from Zambia Justice A. Tambo1 | Lenis Saweda O. Liverpool- Tasie2,3 This is an open access article under the terms of the Creative Commons Attribution-NonCommercial-NoDerivs License, which permits use and distribution in any medium, provided the original work is properly cited, the use is non-commercial and no modifications or adaptations are made. © 2024 The Authors. Journal of Agricultural Economics published by John Wiley & Sons Ltd on behalf of Agricultural Economics Society. 1CABI, Delémont, Switzerland 2Department of Agricultural, Food, and Resource Economics, Michigan State University, East Lansing, Michigan, USA 3International Institute of Tropical Agriculture (IITA), Ibadan, Nigeria Correspondence Justice A. Tambo, CABI, Rue des Grillons 1, Delémont, Switzerland. Email: j.tambo@cabi.org Funding information CABI Development Fund (CDF) Abstract Input subsidy programmes (ISPs) remain a popular but contentious policy tool to promote agricultural intensification, food security and poverty reduction across Africa. Although previous studies have explored the impact of ISPs on various smallholder outcomes, no studies have analysed the impact of recent ISPs on pest management. This is particularly important given the increasing pest challenges due to climate change and the recent surge in pesticide use in low- income countries and its associated negative consequences for human and environmental health. Thus, this study assessed the effects of ISPs on smallholder adoption of sustainable pest management practices, using data from 1048 smallholder maize plots across major maize- producing zones of Zambia and a control function regression approach. We find consistent evidence that input subsidy receipt is negatively associated with smallholders' adoption of environmentally friendly and sustainable pest management strategies. Participation in the Zambia ISP (particularly the flexible e- voucher system) encourages synthetic pesticide use, at the expense of sustainable practices. We also find that farmers consider synthetic pesticides and biopesticides as substitutes and are more likely to adopt sustainable pest management when they have tenure security and access to financial resources. Given the human and environmental health consequences associated with synthetic pesticide use, it would be important to leverage input subsidy schemes to promote the adoption of safer and more sustainable alternatives to synthetic pesticides. Beyond input subsidies, policies that improve tenure security and financial access for www.wileyonlinelibrary.com/journal/jage mailto: https://orcid.org/0000-0001-8895-7066 http://creativecommons.org/licenses/by-nc-nd/4.0/ mailto:j.tambo@cabi.org http://crossmark.crossref.org/dialog/?doi=10.1111%2F1477-9552.12582&domain=pdf&date_stamp=2024-04-23 2 | TAMBO and LIVERPOOL- TASIE 1 | INTRODUCTION Crop pests cause significant economic losses and pose a serious threat to food security, par- ticularly in low- income food- deficit countries (Pratt et al., 2017; Savary et al., 2019). In maize, for instance, yield losses caused by pests are estimated to be 22.5% at the global level and 30% in sub- Saharan Africa (SSA) (Savary et al., 2019). Pesticides, which include insecticides, herbi- cides and fungicides, constitute an important input for reducing pest- induced crop losses. Over the past decade, the volume of pesticide use in low- income countries grew by 153% (Shattuck et al., 2023), driven by a combination of factors, including increasing pest pressures, subsidy schemes, lack of advice on alternative methods of pest control, increasing labour costs and falling prices of generic pesticides (Haggblade et al., 2022; Williamson et al., 2008). Although pesticides play a significant role in protecting crop yields, they can cause serious human health and environmental problems (Jepson et  al.,  2014). For example, it is estimated that 44% of farmers worldwide suffer pesticide poisoning every year (Boedeker et al., 2020). In the wake of outbreaks of devastating pests, such as fall armyworm (FAW), Spodoptera frugiperda, many developing- country governments have gravitated to the provision of syn- thetic pesticides to farmers through subsidy schemes (Day et al.,  2017; Tambo et al.,  2020). Given the potential dangers of synthetic pesticides and the limited use of protective equipment by smallholder farmers when handling pesticides, several studies have argued that subsidies for pesticides should be geared towards the promotion of sustainable pest management strategies, such as integrated pest management (IPM) (Bateman et al., 2021; Day et al., 2022; Grovermann et al., 2017; Tambo et al., 2021). IPM involves the use of a combination of pest management techniques, including biological, cultural, mechanical and pest monitoring practices, as well as pesticides, which should be considered as a last resort (Deguine et al., 2021). It has been suggested that governments should never subsidise synthetic pesticides as it may disincentivise farmers from investing in acquiring IPM skills (Pingali & Rosegrant, 2001). This is also under- scored in Article 8.1.3 of the International Code of Conduct on Pesticide Management devel- oped by the United Nation's Food and Agriculture Organisation (FAO) and the World Health Organisation (WHO), which states that: ‘governments should ensure that any pesticide sub- sidies or donations do not lead to excessive or unjustified use which may divert interest from more sustainable alternative measures’ (FAO & WHO, 2014). However, empirical evidence on whether input subsidy programmes (ISPs) encourage or disincentivise the use of sustainable alternatives to synthetic pesticides for pest control is lacking. Farm input subsidies have been a major agricultural policy intervention in many developing countries over the past decades and have received renewed attention in recent years, partly due to global food price crisis, reported early successes of Malawi's ISP, political motives and the emergence of innovative subsidy models (so- called ‘smart subsidies’) (Hemming et al., 2018; Holden,  2019; Jayne et  al.,  2018; Ricker- Gilbert, Jayne, & Shively,  2013). For instance, it is estimated that 10 SSA countries have together spent up to US$1 billion annually on ISPs in recent years (Jayne et al., 2018). The massive government expenditures on ISPs, coupled with smallholders can promote the adoption of sustainable pest management practices. K E Y W O R D S fall armyworm, farm input subsidies, integrated pest management, pesticides, smallholder farmers, Zambia J E L C L A S S I F I C A T I O N Q12, Q16, Q18 14779552, 0, D ow nloaded from https://onlinelibrary.w iley.com /doi/10.1111/1477-9552.12582 by IN A SP - N E PA L , W iley O nline L ibrary on [23/04/2024]. See the T erm s and C onditions (https://onlinelibrary.w iley.com /term s-and-conditions) on W iley O nline L ibrary for rules of use; O A articles are governed by the applicable C reative C om m ons L icense | 3 ARE FARM INPUT SUBSIDIES A DISINCENTIVE FOR INTEGRATED PEST MANAGEMENT ADOPTION? EVIDENCE FROM ZAMBIA debates about their effectiveness have stimulated a growing body of research aimed at un- derstanding the impacts of ISPs on various outcomes, such as modern input use (largely fer- tiliser and improved seeds), crop acreages and yields, crop diversification, deforestation, food prices, farmer income, nutrition and rural poverty (e.g., Abman & Carney, 2020; Chibwana et al., 2012; Khonje et al., 2022; Kuntashula & Mwelwa- Zgambo, 2022; Liverpool- Tasie, 2014; Mason et al., 2013; Ricker- Gilbert, Mason, et al., 2013; Takeshima & Liverpool- Tasie, 2015; Wossen et al., 2017). However, no recent studies (that we are aware of) have analysed the im- pact of ISPs on synthetic pesticide use and pest management more broadly. Thus, this article aims to contribute to filling this gap in the ISP literature by assessing the effect of Zambia's farmer input support programme (FISP) on farmers' choice of pest man- agement strategies. We also examine the differential effects of two subsidy modalities: a tradi- tional model that provides subsidised fertiliser and hybrid maize seed in fixed proportions and a flexible model that allows beneficiary farmers to choose from a variety of subsidised inputs. We use survey data from 837 Zambian smallholder maize producers who have been contend- ing with FAW, a new invasive pest that is causing substantial damage to maize production in many countries across Africa, Asia and Oceania. Zambia has suffered severe FAW infestation (Stokstad, 2017), with a reported average maize yield loss of 40% in the early years of its out- break in the country (Day et al., 2017). It is also estimated that if left uncontrolled, the FAW pest has the potential to destroy nearly US$160 million worth of maize annually in Zambia (Rwomushana et al., 2018). Given that maize is the primary food crop in Zambia, it is crucial for farmers to adopt appropriate management strategies to combat the pest. Our paper makes three main contributions to the literature on farm input subsidies. First, we add to the limited evidence on the effect of ISPs on the adoption of sustainable intensifica- tion practices. The existing related literature has primarily focused on whether ISPs promote the use of soil fertility management practices, with mixed findings (Holden & Lunduka, 2012; Khonje et al., 2022; Kim et al., 2021; Koppmair et al., 2017; Morgan et al., 2019). For instance, Khonje et al. (2022) showed that participation in ISPs is significantly associated with the adop- tion of organic fertiliser and conservation practices in Malawi, while Morgan et al. (2019) found negative effects of ISPs on the adoption of fallowing and intercropping practices in Zambia. By contrast, we focus on ISP effects on sustainable pest management outcomes. This is par- ticularly important because of the recent surge in pesticide use in low- income countries and its associated negative consequences for human and environmental health leading to a call for sustainable approaches to pest management. It is also important given the expected expansion of pest incidence, disease transmission and risk of invasive insect species due to global warm- ing and climate change (Bebber et al., 2013; Deutsch et al., 2018; Diffenbaugh et al., 2008). Second, we contribute to the literature on obstacles to IPM uptake in developing countries. Despite evidence that IPM interventions are associated with positive economic, health and en- vironmental outcomes (Githiomi et al., 2019; Midingoyi et al., 2019; Norton et al., 2019), adop- tion of the technology among developing- country farmers is still very low (Alwang et al., 2019; Parsa et  al.,  2014), with lack of favourable government policies and support argued as one of the major factors hindering widespread adoption (Day et al., 2022; Parsa et al., 2014). We analyse empirically whether government policy on input subsidies encourages or discourages IPM adoption. Although some studies conducted several decades ago showed that pesticide subsidy policies affect pesticide use and IPM adoption in developing countries (Farah, 1994; Repetto, 1985), they did not apply rigorous impact assessment methods, as we do in this article. Lessons drawn from this study can be useful in leveraging recent ISPs to promote IPM. Finally, the large body of literature examining the effects of farm input subsidies in SSA has mostly focused on traditional maize- centric ISPs that provide fertiliser and improved seeds to beneficiary farmers, with evidence of some positive effects (for reviews, see Hemming et al., 2018; Jayne et al., 2018; Ricker- Gilbert, Jayne, & Shively, 2013). Given the increasing recognition that such traditional subsidy schemes are insufficient to achieve sustainable agricultural intensification 14779552, 0, D ow nloaded from https://onlinelibrary.w iley.com /doi/10.1111/1477-9552.12582 by IN A SP - N E PA L , W iley O nline L ibrary on [23/04/2024]. See the T erm s and C onditions (https://onlinelibrary.w iley.com /term s-and-conditions) on W iley O nline L ibrary for rules of use; O A articles are governed by the applicable C reative C om m ons L icense 4 | TAMBO and LIVERPOOL- TASIE in SSA, alternative subsidy schemes, such as ‘smart’ or flexible programmes have emerged (Jayne et al., 2018). However, as noted by Mason et al. (2013) and Ricker- Gilbert, Jayne, and Shively (2013), there is little knowledge on the relative effectiveness of traditional and ‘smart’ subsidy schemes. To our knowledge, only Mason, Kuteya, et al. (2020) has empirically compared the effectiveness of the two ISP implementation modalities. They found that the traditional FISP in Zambia outper- formed the flexible e- voucher programme in terms of timely availability of fertiliser and farmer adoption of hybrid maize seed and fertiliser, but with no differential effects on crop acreage and diversification, at least in the short run. We complement this previous work by assessing how the two FISP modalities in Zambia influence pest management choices. The rest of the paper is organised as follows. The next section provides a brief background on the FISP in Zambia, and Section 3 describes the data used in this study and the estimation methods. The empirical results are presented in Section 4, including the determinants and effects of FISP participation, as well as the differential effects of the traditional and e- voucher subsidy schemes. The last section concludes by summarising key findings and their implications for policy. 2 | FARM ER IN PUT SU PPORT PROGRA M M E ( FISP) IN ZA M BIA The FISP has been implemented in Zambia since the 2009/2010 agricultural year, building on an earlier large- scale ISP in the country called the Fertiliser Support Programme (2002–2009).1 The subsidy programme aims to improve access to modern agricultural inputs, increase agri- cultural productivity and income, improve household food security, and reduce poverty among smallholder farmers (Mason et al., 2013). The FISP is the main agricultural development pol- icy in Zambia, with nearly three- quarters of the country's agricultural sector budget devoted to the programme in 2022 (Mulenga et al., 2021). Around one million farmers benefit from the programme each year. Eligibility criteria for FISP participation include: membership in a reg- istered farmer organisation or cooperative; being a registered smallholder farmer cultivating not more than 5 hectares of land; having the capacity to pay the farmer contribution of 400 ZMW; and not being a beneficiary of any other government support programme (National Assembly of Zambia, 2022). For several years, Zambia has implemented a traditional FISP model (now referred to as direct input supply) where subsidised fertilisers and improved seeds are delivered to benefi- ciary farmers through farmer groups or cooperatives. In recent cropping seasons, the FISP participants received a predetermined input pack comprising 10 kg maize seed and 300 kg of inorganic fertiliser under the traditional model. As part of diversification efforts, the ben- eficiary farmers may also receive seeds of legume crops, such as groundnut, sorghum and soybean. The traditional FISP model is beset by several challenges, including high implemen- tation cost, leakage of fertiliser inputs into commercial channels, crowding out private sector participation, poor targeting of beneficiaries and blanket fertiliser recommendation (Kuteya et al., 2016; Mason et al., 2013). In response to these challenges, an electronic voucher (e- voucher) system of delivering the FISP was piloted in the country in 13 and 39 districts during the 2015/2016 and 2016/2017 ag- ricultural seasons, respectively, while the remaining districts continued with the traditional FISP. Under the e- voucher modality, beneficiary farmers receive vouchers redeemable at pri- vate agro- input shops for the agricultural inputs of their choice. An e- voucher is worth 2100 ZMW, with contributions of 1700 ZMW and 400 ZMW from the government and benefi- ciary farmer, respectively. The inputs redeemed by the e- voucher recipients include improved maize and legume seeds, fertilisers, pesticides, veterinary drugs and farm equipment (Kuteya 1For a historical overview of Zambia's ISPs, see Mason et al. (2013). 14779552, 0, D ow nloaded from https://onlinelibrary.w iley.com /doi/10.1111/1477-9552.12582 by IN A SP - N E PA L , W iley O nline L ibrary on [23/04/2024]. See the T erm s and C onditions (https://onlinelibrary.w iley.com /term s-and-conditions) on W iley O nline L ibrary for rules of use; O A articles are governed by the applicable C reative C om m ons L icense | 5 ARE FARM INPUT SUBSIDIES A DISINCENTIVE FOR INTEGRATED PEST MANAGEMENT ADOPTION? EVIDENCE FROM ZAMBIA et al., 2016). The e- voucher programme was rolled out nationwide during the 2017/2018 agricul- tural season. However, due to several implementation challenges, 40% of the FISP beneficia- ries were reverted to the traditional model during the 2018/2019 agricultural season. The share of e- voucher beneficiaries was further scaled down in the subsequent agricultural seasons, with even the implementation of only the traditional FISP in the past two seasons. However, the government has committed to rolling out the e- voucher and traditional FISPs in 43 dis- tricts and 73 districts, respectively, in the 2023/2024 agricultural season (National Assembly of Zambia, 2023). Although most studies on the targeting and impacts of farm input subsidies in Africa have focused on Malawi's ISP, Zambia's FISP has also received considerable attention in the litera- ture. On the positive side, studies have shown that farmer participation in Zambia's FISP is sig- nificantly associated with higher maize productivity (Mason et al., 2013; Ngoma et al., 2021), greater production and dietary diversity (Kuntashula & Mwelwa- Zgambo, 2022), and increased income and reduced poverty (Mason & Smale, 2013; Mason, Wineman, & Tembo, 2020). On the other hand, other studies have found that the FISP in Zambia crowds out commercial fertiliser and maize seed demand (Mason & Jayne, 2013; Mason & Ricker- Gilbert, 2013; Xu et al., 2009), disincentivises the adoption of sustainable soil management practices (Morgan et al., 2019), and is not cost- effective (Mason et al., 2013). We extend these studies by examining how the FISP influences sustainable pest management decisions. 3 | DATA A N D M ETHODS 3.1 | Data This study uses survey data collected from a representative sample of 837 households and 1048 maize plots across the three major maize- growing agro- ecological zones (AEZs I, IIa and III) of Zambia.2 These AEZs also cover seven (Central, Copperbelt, Eastern, Luapula, Lusaka, Muchinga and Northern) of the 10 provinces of Zambia (Figure 1). The data were collected primarily for the purpose of assessing farmers' management of the invasive FAW pest; hence, this influenced the sampling strategy used. Based on the severity of FAW infestation and the importance of maize production, 12, 24 and 14 agricultural camps were selected from AEZs I, IIa and III respectively.3 Then within each camp, about 10 to 20 households were randomly selected from a list of maize- producing households provided by local extension workers. Overall, our sample consists of 145 (187), 364 (475) and 328 (386) households (maize plots) from AEZs I, IIa and III, respectively. The surveys were conducted between August and September 2019 and covered the 2018/2019 agricultural season of Zambia. The data were collected by trained enumerators through face- to- face interviews with the sampled households using tablet- based questionnaires. The questionnaires included modules on household demographic characteristics, plot- level charac- teristics, maize production decisions, FAW infestation and management practices, and access to institutional support services, including farm input subsidies. 2There are four AEZs in Zambia. AEZ I: hot and drought- prone region (annual rainfall <800 mm); AEZ IIa: soils and rainfall are more favourable for farming (annual rainfall = 800–1000 mm); AEZ IIb: sandy soils with annual rainfall of 800–1000 mm; and AEZ III: high rainfall area (annual rainfall >1000 mm). 3A camp is the lowest level of agricultural administration in Zambia (i.e., province, district, block and camp). 14779552, 0, D ow nloaded from https://onlinelibrary.w iley.com /doi/10.1111/1477-9552.12582 by IN A SP - N E PA L , W iley O nline L ibrary on [23/04/2024]. See the T erm s and C onditions (https://onlinelibrary.w iley.com /term s-and-conditions) on W iley O nline L ibrary for rules of use; O A articles are governed by the applicable C reative C om m ons L icense 6 | TAMBO and LIVERPOOL- TASIE 3.2 | Empirical approach As mentioned earlier, the goal of this study is to estimate the effect of Zambia's FISP on the adoption of sustainable crop protection practices. There are several mechanisms through which FISP can possibly influence farmers' choice of pest management strategies. For instance, when faced with an invasion of a serious pest such as FAW, FISP participants may be inclined to redeem their vouchers for a subsidised package of inputs including synthetic pesticides for im- mediate control of the pest, at the expense of adopting sustainable non- chemical pest control options. It is also possible that the FISP beneficiaries may invest the money saved from not buying fertilisers and improved seeds at market prices on synthetic pesticides or environmen- tally friendly substitutes to pesticides. Subsidised fertilisers can provide the nutrients needed for the growth of healthy plants that can withstand pest attack, potentially reducing the need for pesticides. However, extensive and inappropriate use of fertilisers can result in reduced pest resistance or increased pest damage (Altieri & Nicholls, 2003; Harrison et al., 2019), which can spur the demand for pest- control products. Given that farmers are encouraged to use complementary pest management practices for FAW control in line with the principles of IPM (Day et al., 2017; Harrison et al., 2019), we first examine FISP effects on the extent of adoption of management practices against FAW. This can be expressed as: where Nip represents the number of IPM practices adopted by household i on plot p against FAW. The FAW management practices adopted included: (1) timely planting to limit FAW infestation; (2) monitoring of FAW infestation; (3) regular weeding to remove alternative host plants; (4) fertilisation to support healthy plant growth so that the maize plants can withstand pest infestations; (5) intercropping with non- host legume crops; (6) crop rotation to disrupt the pest cycle; (7) trap cropping to attract FAW away from the maize crop; (8) (1)Nip = �0 + �1FISPi + � 2 Pp + � 3 Hi + �4Ap + �ip F I G U R E 1 Map of Zambia showing the survey locations. 14779552, 0, D ow nloaded from https://onlinelibrary.w iley.com /doi/10.1111/1477-9552.12582 by IN A SP - N E PA L , W iley O nline L ibrary on [23/04/2024]. See the T erm s and C onditions (https://onlinelibrary.w iley.com /term s-and-conditions) on W iley O nline L ibrary for rules of use; O A articles are governed by the applicable C reative C om m ons L icense | 7 ARE FARM INPUT SUBSIDIES A DISINCENTIVE FOR INTEGRATED PEST MANAGEMENT ADOPTION? EVIDENCE FROM ZAMBIA handpicking/crushing of the pest; (9) uprooting and destruction of infested plants; (10) ap- plication of ash or sand on the FAW larvae; (11) use of biopesticides; and (12) application of synthetic pesticides. Thus, Nip ranges from 0 to 12. FISP is a dummy variable that is equal to 1 if the household participated in FISP during the 2018/2019 agricultural season, and zero otherwise. Thus, the main coefficient of interest is �1, which measures the average effects of FISP on the adoption of IPM practices. P is a vector of plot- level covariates, such as plot size, plot tenure security, distance of plot from homestead, and perception of plot steepness and quality. H represents household- level characteristics, such as the number of children and adults in the household; age, gender, education level of household head; household asset wealth; as well as access to institutional support services, such as credit, farmer group, extension, and input suppliers. A captures differences in agro- climatic conditions. � is a vector of coefficients associated with the explanatory variables, and � is the random error term. A description of the explanatory variables can be found in in Table 1. We estimate Equation (1) with a negative binomial regression model, given the count nature of the dependent variable. We are particularly interested in investigating how the FISP influences the adoption of pest management categories, including monitoring, cultural control, mechanical or physi- cal control, and pesticides. These pest management strategies are among the major compo- nents of IPM (Dhawan & Peshin,  2009).4 Monitoring involves regular scouting of crops after germination to check for signs and symptoms of pest to aid in better pest management decision- making. Cultural control involves the use of any of the following practices: timely planting, regular weeding, fertilisation, legume intercropping, crop rotation or trap crop- ping. Mechanical control includes handpicking of larvae and egg masses, destruction of infested plants or the application of ash or sand. In terms of pesticides, we distinguish be- tween synthetic pesticides and biopesticides. Biopesticides are lower- risk crop protection products (compared to synthetic pesticides), and they include natural plant derivatives or extracts (such as neem and pyrethrum) or microbial pest control agents (such as bacteria and virus) (Bateman et al., 2021). We use a multivariate probit (MVP) model to estimate the effects of FISP on the adop- tion of the five different FAW management strategies. Unlike univariate models such as probit and logit, the MVP model accounts for potential interdependence between the five binary outcomes by estimating them jointly (Greene, 2012). Thus, the MVP model estimates the influence of the FISP and other covariates on the adoption of the five FAW manage- ment options simultaneously, while accounting for the possibility that adoption of any par- ticular strategy could be correlated with adoption of the other four strategies. This can be specified as: where y∗ ipk indicates farm household i's latent propensity to use FAW management strategy k on plot p, and yipk represents the actual adoption of FAW management strategy k by household i on plot k. �im is a vector of multivariate normally distributed error terms. The rest of the symbols are as earlier defined. The MVP model also generates an error term correlation matrix, which gives an indication of whether any two FAW management strategies used by the farmers are comple- ments (positive correlation) or substitutes (negative correlation) to each other. In estimating Equations (1) and (2), we recognise that FISP participation was not based on random assignment. Hence, it is possible that some unobservable factors such as motivation 4Other IPM components include host- plant resistance and biological control using parasitoids or predators. However, these options were not available for adoption by Zambian farmers as at the time of the study. (2) y∗ ipk = 𝛼0 + 𝛼1FISPik + � 2 Ppk + � 3 Hik + 𝛼4Apk + 𝜀ipk; k = 1, 2, … , 5 yipk = 1 if y∗ ipk > 0 and 0 otherwise 14779552, 0, D ow nloaded from https://onlinelibrary.w iley.com /doi/10.1111/1477-9552.12582 by IN A SP - N E PA L , W iley O nline L ibrary on [23/04/2024]. See the T erm s and C onditions (https://onlinelibrary.w iley.com /term s-and-conditions) on W iley O nline L ibrary for rules of use; O A articles are governed by the applicable C reative C om m ons L icense 8 | TAMBO and LIVERPOOL- TASIE TA B L E 1 Descriptive statistics for study variables. Variable Description Mean SD Years in the village Number of years household has lived in the village 27.72 15.35 Age Age of household head (years) 50.60 13.01 Gender Gender of household head (1 = male) 0.70 0.46 Education Years of formal education of household head 7.83 3.39 Children Number of children in the household 3.29 2.14 Adults Number of adults in the household 4.00 2.12 Maize area Size of maize plot (hectares) 1.39 1.77 No. of maize plots Number of maize plots cultivated by household 1.50 0.80 Plot distance Distance from household to maize plot (km) 1.77 3.87 Sloped plot Plot is sloped (1/0) 0.47 0.50 Fertile plot Plot is perceived to be of good quality (1/0) 0.33 0.47 Tenure security Household has secure rights over the plot (1/0) 0.91 0.28 Off- farm activity Household member has off- farm job (1/0) 0.48 0.50 Credit access Household has access to credit (1/0) 0.12 0.32 FISP participation Household participates in FISP (1/0) 0.80 0.40 Asset index Household asset index using principal component analysis 0.00 1.60 TLU Household livestock holding in Tropical Livestock Unit (TLU) 3.27 5.40 Distance to agro- shop Distance from household to the nearest agro- input shop (km) 15.24 13.54 Extension access Household has access to extension services (1/0) 0.85 0.35 Farmer group A household member belongs to a farmer group (1/0)a 0.88 0.32 AEZ I Household is located in agro- ecological zone I 0.17 0.38 AEZ IIa Household is located in agro- ecological zone IIa 0.44 0.50 AEZ III Household is located in agro- ecological zone III 0.39 0.37 Observations Number of observations 1048 aFarmer groups include farmer associations and cooperatives. 14779552, 0, D ow nloaded from https://onlinelibrary.w iley.com /doi/10.1111/1477-9552.12582 by IN A SP - N E PA L , W iley O nline L ibrary on [23/04/2024]. See the T erm s and C onditions (https://onlinelibrary.w iley.com /term s-and-conditions) on W iley O nline L ibrary for rules of use; O A articles are governed by the applicable C reative C om m ons L icense | 9 ARE FARM INPUT SUBSIDIES A DISINCENTIVE FOR INTEGRATED PEST MANAGEMENT ADOPTION? EVIDENCE FROM ZAMBIA and entrepreneurial ability influence both the decision to participate in FISP and our outcome variables. Thus, using negative binomial regression and MVP models to respectively estimate Equations (1) and (2) may yield biased estimates. We applied a control function approach to test and address this potential selection bias problem (Terza et al., 2008; Wooldridge, 2015). The control function approach follows two steps. First, a probit model for FISP participation is estimated to obtain the generalised residual. In the second step, Equations (1) and (2) are estimated using negative binomial regression and MVP models, respectively, while including the generalised residual from the first- stage regression as an additional covariate. A statistical significance of the generalised residual variable implies that the FISP participation decision is endogenous. Thus, the FISP effects are estimated using negative binomial regression and MVP models combined with the control function approach. In estimating Equation 2, we used bootstrap- ping to adjust the standard errors for the two- step estimation procedure. Moreover, we used the number of years that the household head has lived in the village as an exclusion restriction in the control function approach. A similar instrumental variable has been used in several studies examining the impact of farmer participation in input subsidy programmes in Africa (Bezu et al., 2014; Chibwana et al., 2012; Ngoma et al., 2021; Ricker- Gilbert et al., 2011; Wossen et al., 2017). The argument is that the number of years of living in a village improves social cap- ital, which can enhance access to FISP. The first- stage regression results in Table 3 show that the number of years a household has lived in a study village variable significantly affects FISP participation, but it is not statistically significant when included as an additional covariate in the outcome regressions (see Table S1). This provides some support to the validity of our exclu- sion restriction variable (Di Falco et al., 2011). Nonetheless, given the challenge of establishing causality with cross- sectional data, we interpret our regression results as associations rather than causal effects. 4 | RESU LTS A N D DISCUSSION 4.1 | Descriptive results Table 1 reports summary statistics of the control variables in our regression models. A typical household in our sample consists of four adults and three children and is headed by a middle- aged male farmer who has lived in the study village for nearly three decades. The sample com- prises smallholder maize- growing households who cultivate roughly two maize plots, with an average size of 1.40 hectares per plot. A majority of the households have secure rights over their maize plots, and only a third of the plots are deemed to have good soil fertility. Approximately half of the households engage in off- farm income earning activities, while only 12% of them had access to credit during the agricultural year 2018/2019. Eighty percent of the households participate in the FISP, which is our main variable of interest. Additionally, over 80% of the households have access to agricultural extension services and are members of farmer groups or associations. Table 2 presents the descriptive statistics for the outcome variables, disaggregated by FISP participation and whether the household is located in an e- voucher or a traditional FISP dis- trict. On average, households implemented three different fall armyworm (FAW) management practices on a maize plot. This is noteworthy, as smallholder farmers are encouraged to com- bine different practices to tackle FAW, in line with the tenets of IPM (Day et al., 2017; Harrison et al., 2019). This is also consistent with previous evidence showing that Zambian maize farm- ers have adopted a mix of pest management practices in the wake of the FAW pest outbreak in the country (Kansiime et al., 2019; Tambo et al., 2020). The most common IPM techniques include cultural and mechanical control practices, which were implemented on at least half of 14779552, 0, D ow nloaded from https://onlinelibrary.w iley.com /doi/10.1111/1477-9552.12582 by IN A SP - N E PA L , W iley O nline L ibrary on [23/04/2024]. See the T erm s and C onditions (https://onlinelibrary.w iley.com /term s-and-conditions) on W iley O nline L ibrary for rules of use; O A articles are governed by the applicable C reative C om m ons L icense 10 | TAMBO and LIVERPOOL- TASIE the maize plots in our sample. Synthetic pesticides and biopesticides, which are relatively more costly control options, were each applied on about a quarter of the plots. The level of adoption of the specific FAW management practices is presented in Table S2. TA B L E 2 Descriptive statistics for outcome variables. Outcome Full sample (n = 1048) FISPa (n = 841) Non- FISP (n = 207) E- voucherb (n = 746) Traditional (n = 302) No. of IPM practices used (0–12) 3.17 3.21 3.04 3.16 3.21 Use of pest monitoring (%) 42.46 42.57 42.03 42.76 41.72 Use of cultural control (%) 51.34 52.32 47.34 50.80 52.65 Use of mechanical control (%) 52.19 51.37 55.56 47.05*** 64.90 Use of synthetic pesticides (%) 33.59 35.08** 27.54 35.52** 28.81 Use of biopesticides (%) 31.20 32.10 27.54 31.50 30.46 aFISP participants are compared with non- FISP participants. bHouseholds located in e- voucher FISP districts are compared with those located in traditional FISP districts. ***p < 0.01; **p > 0.05; p < 0.1. TA B L E 3 Determinants of FISP participation. Coefficients Standard error Marginal effect Years in the village 0.009** 0.004 0.002 Age 0.008* 0.004 0.001 Gender −0.036 0.127 −0.007 Education 0.001 0.018 0.001 Children 0.006 0.025 0.001 Adults −0.025 0.028 −0.005 Maize area 0.128** 0.052 0.023 Number of maize plots 0.132*** 0.037 0.024 Off- farm income 0.004 0.109 0.001 Credit access 0.284 0.189 0.052 Asset index 0.045 0.043 0.008 TLU −0.007 0.012 −0.001 Distance to agro- shop 0.004 0.004 0.001 Extension access 0.113 0.148 0.021 Farmer group 2.185*** 0.160 0.401 AEZ IIa −0.243 0.163 −0.044 AEZ III −0.072 0.171 −0.012 Constant −2.064*** 0.396 Observations 1048 LR chi2 334.85*** Note: Probit regression results. Dependent variable = FISP participation (1/0). ***p < 0.01; **p > 0.05;*p < 0.1. 14779552, 0, D ow nloaded from https://onlinelibrary.w iley.com /doi/10.1111/1477-9552.12582 by IN A SP - N E PA L , W iley O nline L ibrary on [23/04/2024]. See the T erm s and C onditions (https://onlinelibrary.w iley.com /term s-and-conditions) on W iley O nline L ibrary for rules of use; O A articles are governed by the applicable C reative C om m ons L icense | 11 ARE FARM INPUT SUBSIDIES A DISINCENTIVE FOR INTEGRATED PEST MANAGEMENT ADOPTION? EVIDENCE FROM ZAMBIA The results in Table 2 also show that pesticides were sprayed on a greater number of plots of FISP participants than of non- participants, with statistically significant differences in terms of the use of synthetic pesticides. Similarly, a significant number of maize plots of households located in e- voucher FISP districts received synthetic pesticides, as compared to plots of house- holds residing in traditional FISP districts. On the other hand, mechanical control measures, such as handpicking of FAW larvae, were performed more in traditional FISP districts than in e- voucher districts. Overall, these unconditional summary statistics may be suggestive that the FISP beneficiaries, particularly participants of the flexible e- voucher scheme, are more likely to use synthetic pesticides for maize pest control. The econometric evidence for these potential effects of FISP is presented in Section 4.3. 4.2 | Determinants of FISP participation Table 3 shows the first- stage probit regression results on the determinants of household par- ticipation in the FISP. We find that the number of years a household has lived in a study village is significantly and positively correlated with participation in the FISP, suggesting that our instrument is relevant. This finding is in line with previous studies on input subsidy participa- tion in Malawi (Bezu et al., 2014; Chibwana et al., 2012; Ricker- Gilbert et al., 2011), Nigeria (Wossen et al., 2017) and Zambia (Ngoma et al., 2021). The results also show that the likelihood of participation in FISP increases with the age of household head, which may also be reflective of the effect of social capital. None of the financial capital and wealth- related factors (off- farm job, credit, asset index and livestock holding) is significantly associated with FISP participa- tion, indicating that resource- rich and resource- poor households are equally likely to partici- pate in the programme. Similarly, the input subsidy programme is inclusive of both male and female- headed households, which is consistent with evidence from Zambia and several other African countries (Jayne et al., 2018). The results in Table 3 also indicate that the number of maize plots cultivated and maize plot size exert significant effects on FISP participation. This suggests that participation in the FISP is higher among households for whom maize production is of particular economic importance. This is expected, as the Zambia FISP is largely a maize- centric intervention, especially the traditional FISP where beneficiary farmers receive subsidised maize inputs. In their review of empirical literature on input subsidy programmes in Africa, Jayne et al. (2018) also found that households with more land are more likely to benefit from input subsidies. Finally, the results show that farmer group membership has the strongest effect on FISP participation. This is not surprising because farmer group or cooperative membership is a key eligibility criterion for farmer participation in the FISP (Mason et al., 2013). 4.3 | FISP effects on adoption of IPM practices Table 4 displays the results of the negative binomial regression model performed with the con- trol function approach, which show the effects of FISP participation on the number of IPM practices used. We find that after controlling for potential confounding factors, the FISP par- ticipants adopt roughly four less IPM practices, compared to their non- participant counter- parts. This implies that the FISP intervention disincentivises the use of multiple IPM practices, which is troubling given the calls for the promotion of IPM as a sustainable approach to the management of FAW and other crop pests in smallholder farming systems (Day et al., 2017; Hruska, 2019). This result resonates with the finding of Morgan et al. (2019) that the FISP in Zambia discourages the use of sustainable intensification practices among smallholder farm- ers in the country. 14779552, 0, D ow nloaded from https://onlinelibrary.w iley.com /doi/10.1111/1477-9552.12582 by IN A SP - N E PA L , W iley O nline L ibrary on [23/04/2024]. See the T erm s and C onditions (https://onlinelibrary.w iley.com /term s-and-conditions) on W iley O nline L ibrary for rules of use; O A articles are governed by the applicable C reative C om m ons L icense 12 | TAMBO and LIVERPOOL- TASIE The results for the control variables in Table 4 are informative. We find that a greater num- ber of IPM practices are likely to be adopted on larger and fertile maize plots, albeit the effects are significant only at the 10% level. Results also show that having a secure right over a plot is significantly associated with the adoption of 0.60 additional IPM practices. This is possibly because the tenure secure households may be more likely to care about the future benefits from their plots by investing in sustainable agricultural practices (such as IPM) that can generate yield benefits even in the long term. The positive role of tenure security in agricultural invest- ment has been documented in several studies (e.g., Gebremedhin & Swinton, 2003; Tambo & Mockshell, 2018). We also find that access to credit and off- farm activities, which can help households to relax their financial constraints, are strongly significantly correlated with IPM adoption. In particu- lar, households with access to off- farm activities and credit respectively adopt 0.46 and 0.68 ad- ditional IPM practices on their maize plots. An interesting finding is that households living far from agro- input dealers are more likely to adopt a higher number of IPM practices. In other words, being in closer proximity to agro- input dealers where farmers obtain pesticides reduces the likelihood of IPM adoption. This is likely because profit- driven agro- input dealers tend TA B L E 4 FISP effects on the number of IPM practices used. Coefficients Robust SE Marginal effect FISP participation −1.243** 0.483 −3.954 Age 0.001 0.002 0.003 Gender 0.004 0.057 0.012 Education 0.01 0.008 0.032 Children 0.019* 0.011 0.061 Adults 0.007 0.012 0.023 Maize area 0.040* 0.021 0.127 Distance to maize plot −0.008 0.008 −0.025 Sloped plot −0.035 0.047 −0.111 Fertile plot 0.085* 0.048 0.272 Tenure security 0.188** 0.076 0.597 Off- farm job 0.144*** 0.045 0.460 Credit access 0.215*** 0.074 0.684 Asset index −0.008 0.017 −0.025 TLU −0.001 0.005 −0.002 Distance to agro- shop 0.004** 0.002 0.013 Extension access 0.547*** 0.092 1.738 Farmer group 0.801** 0.351 2.546 AEZ IIa −0.442*** 0.068 −1.547 AEZ III −0.327*** 0.067 −1.207 Generalised residual 0.685*** 0.261 2.178 Constant 0.665*** 0.198 Observations 1048 LR test of alpha = 0 172.71*** Note: Results from a negative binomial regression model performed with the control function approach. Dependent variable is the number of IPM practices adopted, ranging from 0 to 12. ***p < 0.01; **p > 0.05; *p < 0.1. 14779552, 0, D ow nloaded from https://onlinelibrary.w iley.com /doi/10.1111/1477-9552.12582 by IN A SP - N E PA L , W iley O nline L ibrary on [23/04/2024]. See the T erm s and C onditions (https://onlinelibrary.w iley.com /term s-and-conditions) on W iley O nline L ibrary for rules of use; O A articles are governed by the applicable C reative C om m ons L icense | 13 ARE FARM INPUT SUBSIDIES A DISINCENTIVE FOR INTEGRATED PEST MANAGEMENT ADOPTION? EVIDENCE FROM ZAMBIA to provide crop health advice and sell pesticide products to farmers, which may decrease the incentive to apply alternatives to chemical pest control. This also lends support to arguments that pesticide industry interference and heavy promotion of pesticides by salespeople under- mine efforts to promote IPM in developing countries (Parsa et al., 2014). Conversely, access to agricultural extension services (a key source of agricultural information), is significantly associated with the adoption of roughly two additional IPM practices. Farmer group, another source of information, enhances the adoption of nearly three additional IPM practices. These results give credence to the assertion that lack of information is a major obstacle to IPM adop- tion among smallholders in low- income countries (Alwang et al., 2019; Parsa et al., 2014). The results also show that agro- climatic conditions are significantly related to the uptake of IPM practices. Maize plots in AEZ IIa and AEZ III are less likely to have multiple IPM practices, compared to those in AEZ I. A plausible explanation is that AEZ I is a low- rainfall and drought- prone region; hence, households located in this zone may be more likely to adopt certain IPM practices to protect their yield from weather and pest shocks. Given that FAW infestation rates tend to be higher in low- rainfall environments (Early et  al.,  2018; Varella et al., 2015), farmers located AEZ I are more likely to experience the devastating effects of the pest, which may increase their likelihood of implementing multiple management practices. Lastly, the statistical significance of the generalised residual variable confirms the endogeneity of the FISP variable (Wooldridge, 2014). The results for the effects of FISP on the uptake of various IPM strategies, which were es- timated using a multivariate probit model combined with the control function approach, are presented in Table 5. We find that households that participate in FISP are about 106 percent- age points less likely to monitor their fields for signs of pest infestation. This is a disturbing finding because regular field inspection or scouting for pest incidence and severity of infes- tation is a highly important IPM practice that help farmers to decide whether management is needed, and to act at the right time to limit the spread of pests. Similarly, the results show that FISP beneficiary households are 125 percentage points less likely to adopt cultural controls, which are good agronomic practices that double as sustainable pest prevention methods, such as timeliness in planting, disrupting pest cycle through crop rotation, intercropping of maize with non- host leguminous crops, and removal of alternate pest host plants through regular weeding. This result is also in line with evidence that Zambia's FISP crowd out the use of sus- tainable soil management practices, including crop rotation and maize- legume intercropping (Morgan et  al.,  2019), but contrasts with the finding of Koppmair et  al.  (2017) that farmer adoption of natural resource management practices is generally unaffected by Malawi's FISP. The results also show that FISP participation is not significantly related to the adoption of mechanical control methods and biopesticides. On the other hand, participation in FISP has a significantly positive effect on the use of synthetic pesticides for FAW control. In particular, the FISP participants have a 67 percentage points higher probability of controlling FAW using synthetic pesticides. Unfortunately, synthetic pesticides are associated with adverse effects on animal, human and environmental health; hence, the need to reduce farmer reliance on syn- thetic pesticides by promoting the use of a combination of non- chemical pest management practices, in line with IPM. Overall, the results in Table 5 demonstrate that the FISP incentiv- ises the use of synthetic pesticides, at the expense of environmentally friendly and sustainable pest management strategies. Further analysis shows that during the 2018/2019 agricultural sea- son, the average number of pesticide applications against FAW by FISP participants was 2.6 times more compared to non- participants, possibly suggesting that FISP beneficiaries are more likely to use pesticides intensively (see Table S3).5 5Due to data limitations, we are unable to estimate FISP effects on the quantity of pesticide used per unit area, which is a more appropriate measure of pesticide use intensity. 14779552, 0, D ow nloaded from https://onlinelibrary.w iley.com /doi/10.1111/1477-9552.12582 by IN A SP - N E PA L , W iley O nline L ibrary on [23/04/2024]. See the T erm s and C onditions (https://onlinelibrary.w iley.com /term s-and-conditions) on W iley O nline L ibrary for rules of use; O A articles are governed by the applicable C reative C om m ons L icense 14 | TAMBO and LIVERPOOL- TASIE T A B L E 5 E ff ec t of F IS P o n ad op ti on o f d if fe re nt I P M s tr at eg ie s. M on it or in g C ul tu ra l c on tr ol M ec ha ni ca l c on tr ol S yn th et ic p es ti ci de s B io pe st ic id es M ar gi na l ef fe ct R ob us t S E M ar gi na l ef fe ct R ob us t S E M ar gi na l ef fe ct R ob us t S E M ar gi na l ef fe ct R ob us t S E M ar gi na l ef fe ct R ob us t S E F IS P p ar ti ci pa ti on − 1. 05 6* ** 0. 30 8 − 1. 24 7* ** 0. 29 8 − 0. 09 5 0. 28 9 0. 67 2* * 0. 26 4 0. 38 3 0. 25 1 A ge 0. 00 1 0. 00 1 0. 00 1 0. 00 1 0. 00 0 0. 00 1 − 0. 00 2 0. 00 1 − 0. 00 1 0. 00 1 G en de r 0. 01 0 0. 03 6 − 0. 02 3 0. 03 6 − 0. 00 5 0. 03 6 0. 04 0 0. 03 4 − 0. 05 0 0. 03 4 E du ca ti on 0. 00 5 0. 00 5 0. 00 2 0. 00 5 0. 00 6 0. 00 5 0. 00 5 0. 00 5 0. 00 7 0. 00 5 C h il d re n 0. 01 3* 0. 00 7 0. 01 5* * 0. 00 7 − 0. 00 2 0. 00 7 0. 00 1 0. 00 7 − 0. 00 5 0. 00 7 A du lt s − 0. 00 7 0. 00 7 0. 00 3 0. 00 8 0. 01 4* 0. 00 8 − 0. 00 3 0. 00 7 0. 00 5 0. 00 7 M ai ze a re a 0. 01 8 0. 01 3 0. 03 3* * 0. 01 3 0. 00 9 0. 00 6 0. 00 8 0. 00 8 − 0. 00 5 0. 00 7 D is ta nc e to m ai ze pl ot − 0. 00 6 0. 00 4 − 0. 00 6 0. 00 4 − 0. 00 3 0. 00 5 − 0. 00 6* 0. 00 4 0. 00 2 0. 00 3 Sl op ed p lo t − 0. 03 8 0. 02 9 0. 00 0 0. 03 0 − 0. 00 1 0. 03 0 0. 03 9 0. 02 8 − 0. 01 6 0. 02 8 F er ti le p lo t 0. 03 5 0. 03 1 0. 05 6* 0. 03 1 0. 02 7 0. 03 2 0. 02 6 0. 03 0 0. 09 1* ** 0. 02 9 T en u re s ec u ri ty 0. 09 9* 0. 05 2 0. 09 1* 0. 05 2 0. 04 6 0. 05 5 0. 02 1 0. 05 1 − 0. 13 6* ** 0. 05 1 O ff - f ar m a ct iv it y 0. 07 6* * 0. 03 0 0. 05 2* 0. 03 0 0. 08 7* ** 0. 03 0 0. 04 3 0. 02 9 0. 06 9* * 0. 02 8 C re d it a cc es s 0. 07 7* 0. 04 7 0. 14 1* ** 0. 04 9 − 0. 01 0 0. 04 8 − 0. 02 5 0. 04 4 0. 02 2 0. 04 4 A ss et in de x − 0. 00 9 0. 01 2 − 0. 00 4 0. 01 2 − 0. 01 5 0. 01 2 0. 00 5 0. 01 1 0. 00 6 0. 01 1 T L U 0. 00 0 0. 00 3 0. 00 2 0. 00 3 − 0. 00 6* 0. 00 3 0. 00 2 0. 00 3 0. 00 4 0. 00 3 D is ta nc e to ag ro - s ho p 0. 00 1 0. 00 1 0. 00 3* * 0. 00 1 0. 00 1 0. 00 1 − 0. 00 0 0. 00 1 0. 00 0 0. 00 1 A cc es s to e xt en si on 0. 20 4* ** 0. 04 4 0. 24 5* ** 0. 04 3 0. 10 0* * 0. 04 4 0. 30 1* ** 0. 04 8 0. 03 7 0. 04 1 F ar m er g ro up 0. 75 5* ** 0. 22 8 0. 86 4* ** 0. 21 8 − 0. 01 0 0. 21 3 − 0. 45 4* * 0. 19 4 − 0. 28 8 0. 18 5 A E Z I Ia − 0. 31 9* ** 0. 04 3 − 0. 28 2* ** 0. 04 2 − 0. 09 8* * 0. 04 6 − 0. 06 2 0. 04 4 0. 05 4 0. 04 3 A E Z I II − 0. 25 8* ** 0. 04 5 − 0. 17 5* ** 0. 04 3 0. 12 4* ** 0. 04 7 − 0. 08 8* * 0. 04 4 − 0. 09 2* * 0. 04 2 G en er al is ed re si du al 0. 55 7* ** 0. 16 7 0. 69 8* ** 0. 16 4 0. 03 0 0. 15 6 − 0. 35 9* * 0. 14 0 − 0. 17 4 0. 13 2 O b se rv at io ns 10 48 10 48 10 48 10 48 10 48 N ot e: A ll t he f iv e de p en de nt v ar ia bl es a re b in ar y ou tc om es , a nd t he p ar am et er s w er e jo in tl y es ti m at ed u si ng a m u lt iv ar ia te p ro bi t m o de l p er fo rm ed w it h th e co nt ro l f u nc ti on a pp ro ac h. A bb re vi at io n: S E , s ta nd ar d er ro r. ** *p < 0 .0 1; * *p > 0 .0 5; * p < 0 .1 . 14779552, 0, D ow nloaded from https://onlinelibrary.w iley.com /doi/10.1111/1477-9552.12582 by IN A SP - N E PA L , W iley O nline L ibrary on [23/04/2024]. See the T erm s and C onditions (https://onlinelibrary.w iley.com /term s-and-conditions) on W iley O nline L ibrary for rules of use; O A articles are governed by the applicable C reative C om m ons L icense | 15 ARE FARM INPUT SUBSIDIES A DISINCENTIVE FOR INTEGRATED PEST MANAGEMENT ADOPTION? EVIDENCE FROM ZAMBIA Table  6 reports the types of pesticides sprayed against FAW by the pesticide users in our sample. The most commonly used pesticide is azadirachtin, which is a biopesticide. We find that most of the synthetic pesticides used are moderately hazardous, as classi- fied by the World Health Organisation, based on acute risk to human health (WHO, 2020). Alarmingly, about 5% of the pesticide users applied highly hazardous pesticides such as methamidophos, dichlorvos and monocrotophos, which can have acute and severe or irre- versible effects on human and environmental health. Unfortunately, some of the pesticide users do not wear standard personal protective equipment (PPE) when spraying these haz- ardous pesticides (Table 7). For instance, around half of the pesticide users in our sample did not wear masks, gloves or coverall when handling or spraying pesticides, and only 24% of them wore goggles. Table  7 shows that about 12% of the farmers who sprayed pesti- cides did not wear any protective apparel when working with pesticides. The high use of hazardous pesticides without the necessary protective gear underscores the importance of leveraging the FISP to promote the use of sustainable, non- chemical alternatives or a shift from synthetic pesticides to biopesticides, which are considered less toxic to health and the environment (Bateman et al., 2021). The error term correlation matrix of the multivariate probit model, which gives an indi- cation of complementarities and trade- offs between the five IPM options, confirms that the TA B L E 6 Pesticides used for fall armyworm control. Active ingredient WHO toxicity classa Full sample (n = 505) FISP (n = 417) Non- FISP (n = 88) Azadirachtin N 42.77 43.17 40.91 Cypermethrin II 18.22 18.47 17.05 Lambda- cyhalothrin II 10.50 11.51 5.68 Abamectin II 6.14 5.28 10.23 Monocrotophos Ib 4.36 4.55 4.31 Profenofos + Cypermethrin II 3.37 3.60 2.27 Dimethoate II 2.18 2.40 1.14 Chlorpyrifos + Cypermethrin II 2.18 2.40 1.14 Lambda- cyhalothrin + Thiamethoxam III 1.98 1.92 2.27 Chlorothalonil U 1.58 1.92 0.00 Emamectin Benzoate N 1.58 1.92 0.00 GS- omega/kappa- Hxtx- Hv1a N 1.39 1.44 1.14 Profenofos II 1.19 0.96 2.27 Alpha- cypermethrin II 0.79 0.96 0.00 Flubendiamide N 0.99 1.19 0.00 Atrazine III 0.79 0.72 1.14 Chlorantraniliprole + Lambda- cyhalothrin II 0.79 0.96 0.00 Dichlorvos Ib 0.79 0.96 0.00 Emamectin Benzoate N 0.59 0.72 0.00 Malathion III 0.59 0.48 1.14 Methamidophos Ib 0.40 0.48 0.00 Note: Values are percentages of pesticide users. aWHO recommended classification of pesticides (WHO, 2020): Ia = extremely hazardous; Ib = highly hazardous; II = moderately hazardous; III = lightly hazardous; U = unlikely to present acute hazard; N = not classified. 14779552, 0, D ow nloaded from https://onlinelibrary.w iley.com /doi/10.1111/1477-9552.12582 by IN A SP - N E PA L , W iley O nline L ibrary on [23/04/2024]. See the T erm s and C onditions (https://onlinelibrary.w iley.com /term s-and-conditions) on W iley O nline L ibrary for rules of use; O A articles are governed by the applicable C reative C om m ons L icense 16 | TAMBO and LIVERPOOL- TASIE promotion of biopesticides may reduce the demand for synthetic pesticides (Table 8). We see a negative and significant correlation between the use of synthetic pesticides and biopesticides, suggesting that farmers perceive substitutability between these two groups of pesticides. Thus, farmers who use biopesticides are less likely to opt for synthetic pesticides for FAW control. Tambo et al.  (2020) reported a similar finding for Ghana. The results in Table 8 also show complementarities between the monitoring of pest infestation and the use of cultural and me- chanical control practices, as well as synthetic pesticides for FAW control. The integration of different pest management tactics is consistent with the expectation of the IPM paradigm. However, it is hoped that synthetic pesticides will be used as a last resort, and if pesticide used is required, farmers will select the products with the lowest risk to human and environmental health. 4.4 | Differential effects by FISP modality Finally, we examine the heterogeneous effects of the FISP by differentiating households ac- cording to the type of FISP operating in their districts. As mentioned earlier, there are two FISP modalities in Zambia: a f lexible e- voucher programme and a traditional FISP or di- rect input supply model. Unlike the traditional FISP where the government distributes a fixed package of maize seed and fertiliser to beneficiary farmers, the participants of the e- voucher programme are allowed to choose inputs of their choice from registered agro- dealers. During the 2018/2019 season, 45 of the 116 Zambian districts were under the traditional FISP, and the rest were under the e- voucher programme. Table 9 summarises the estimation results of the FISP effects on the adoption of IPM strategies, based on whether a household is located in an e- voucher or a traditional FISP district.6 The full estimation results of the multivariate probit model with control function approach are presented in Tables S4 and S5. The results show that FISP participation has a negative and significant effect on the adop- tion of the IPM options of regular monitoring and cultural control in both the e- voucher and traditional FISP districts. In other words, the FISP beneficiaries are less likely to regularly monitor their fields for pest infestation or adopt cultural control techniques, irrespective of the type of Zambia's FISP they are exposed to. However, the marginal effects suggest that house- holds located in traditional FISP districts have a greater probability of not adopting these two IPM practices than those located in the e- voucher FISP districts. Specifically, FISP par- ticipants in e- voucher districts are about 51 and 73 percentage points less likely to implement 6Given that a district in Zambia can only participate in either the traditional or e- voucher FISP modality, but not both in an agricultural season, the district location of a FISP beneficiary household reflects participation in the FISP modality that is being implemented in that district. TA B L E 7 Use of PPE items. PPE Full sample (n = 505) FISP (n = 417) Non- FISP (n = 88) Goggles 23.76 23.50 25.00 Mask 49.11 49.40 47.73 Gloves 50.69 49.40 56.82 Coverall 51.09 51.56 48.86 Rubber boots 75.84 76.98 70.45 None 12.08 10.79 18.18 Note: Values are percentages of pesticide users. 14779552, 0, D ow nloaded from https://onlinelibrary.w iley.com /doi/10.1111/1477-9552.12582 by IN A SP - N E PA L , W iley O nline L ibrary on [23/04/2024]. See the T erm s and C onditions (https://onlinelibrary.w iley.com /term s-and-conditions) on W iley O nline L ibrary for rules of use; O A articles are governed by the applicable C reative C om m ons L icense | 17 ARE FARM INPUT SUBSIDIES A DISINCENTIVE FOR INTEGRATED PEST MANAGEMENT ADOPTION? EVIDENCE FROM ZAMBIA T A B L E 8 C om pl em en ta ri ty a nd s ub st it ut ab il it y of t he I P M o pt io ns . M on it or in g C ul tu ra l c on tr ol M ec ha ni ca l c on tr ol S yn th et ic p es ti ci de B io p es ti ci de M on it or in g 1 C u lt u ra l c on tr ol 0. 85 0 (0 .0 21 )* ** 1 M ec ha n ic al c on tr ol 0. 45 8 (0 .0 43 )* ** 0. 44 2 (0 .0 44 )* ** 1 Sy nt he ti c p es ti ci de 0. 10 4 (0 .0 52 )* * 0. 07 0 (0 .0 51 ) 0. 18 0 (0 .0 51 )* ** 1 B io p es ti ci de 0. 01 0 (0 .0 53 ) 0. 02 8 (0 .0 52 ) − 0. 00 3 (0 .0 52 ) − 0. 79 0 (0 .0 32 )* ** 1 N ot e: R es u lt s in t h is t ab le a re t he c or re la ti on m at ri x ge ne ra te d fr om t he m u lt iv ar ia te p ro bi t m o de l e st im at es in T ab le  5 . ** *p < 0 .0 1; * *p > 0 .0 5; * p < 0 .1 . 14779552, 0, D ow nloaded from https://onlinelibrary.w iley.com /doi/10.1111/1477-9552.12582 by IN A SP - N E PA L , W iley O nline L ibrary on [23/04/2024]. See the T erm s and C onditions (https://onlinelibrary.w iley.com /term s-and-conditions) on W iley O nline L ibrary for rules of use; O A articles are governed by the applicable C reative C om m ons L icense 18 | TAMBO and LIVERPOOL- TASIE cultural controls and field monitoring, respectively, while their counterparts in the traditional FISP districts are nearly 150 percentage points less likely to adopt these two IPM options. It is possible that participation in the FISP (particularly the traditional model) leads to greater use of inorganic fertilisers (Jayne et al., 2018; Mason, Kuteya, et al., 2020), which can influence pest populations (Altieri & Nicholls, 2003; Harrison et al., 2019). Given that monitoring and cultural control are pest prevention practices, farmers may have less incentive to implement them on plots with high pest infestation and damage levels. More use of inorganic fertilisers is also labour intensive, and thus households might be reducing the time spent on monitoring and other cultural control practices to give more time for fertiliser application and associated practices. Table  9 also indicates that FISP participation is significantly associated with the use of synthetic pesticides only among households who can benefit from the e- voucher model. In particular, beneficiary households in the e- voucher FISP districts have a 60 percentage points higher likelihood of spraying synthetic pesticides to control FAW. Furthermore, the estima- tion results in Table S3 show that the average number of pesticide sprays per season by FISP participants in the e- voucher districts is 4.5 times more significantly higher than their non- participant counterparts, whereas there is no statistically significant difference in pesticide spray frequency between FISP participants and non- participants in the traditional districts. These results are intuitive because the e- voucher FISP beneficiaries have the possibility of redeeming their vouchers for a wide range of subsidised farm inputs, including pesticides, whereas the traditional FISP participants have to purchase the pesticides, which can be pro- hibitively expensive for many Zambian smallholders who are used to receiving other inputs at subsidised rates. Overall, these results may be suggestive that the e- voucher programme, which allows farmers access to subsidised pesticides, crowds out the use of environmentally friendly pest prevention practices. 5 | CONCLUSION Input subsidy schemes are a popular but contentious policy tool used by several African gov- ernments in efforts to promote agricultural intensification and achieve food security and pov- erty reduction goals. This study has assessed the effects of input subsidies on smallholder adoption of sustainable pest management practices, which is important because increasing pest pressure has fostered widespread and indiscriminate use of synthetic pesticides that pose TA B L E 9 Differential effects by two FISP types. IPM strategy (1) E- voucher FISP districts (2) Traditional FISP districts Marginal effect Robust SE Marginal effect Robust SE Monitoring −0.731** 0.293 −1.472*** 0.419 Cultural control −0.506* 0.293 −1.454*** 0.439 Mechanical control −0.018 0.296 −0.523 0.412 Synthetic pesticides 0.596** 0.301 0.325 0.350 Biopesticides 0.490 0.316 −0.459 0.324 Observations 746 302 Note: The results in columns 1 and 2 are based on data from maize plots in districts under the e- voucher and traditional FISP programme, respectively. Each column presents results from a multivariate probit model performed with the control function approach. Each row represents a binary outcome variable. The treatment variable is FISP participation (1/0). The full regression results can be found in Tables S4 and S5. ***p < 0.01; **p > 0.05; *p < 0.1. 14779552, 0, D ow nloaded from https://onlinelibrary.w iley.com /doi/10.1111/1477-9552.12582 by IN A SP - N E PA L , W iley O nline L ibrary on [23/04/2024]. See the T erm s and C onditions (https://onlinelibrary.w iley.com /term s-and-conditions) on W iley O nline L ibrary for rules of use; O A articles are governed by the applicable C reative C om m ons L icense | 19 ARE FARM INPUT SUBSIDIES A DISINCENTIVE FOR INTEGRATED PEST MANAGEMENT ADOPTION? EVIDENCE FROM ZAMBIA high risks to human and environmental health. Specifically, we examined whether Zambia's farmer input support programme (FISP) influences the use of sustainable integrated pest man- agement (IPM) practices against fall armyworm (FAW)—an invasive pest that is causing mas- sive economic damages to maize production in many African and Asian countries in recent years. We used data from 1048 smallholder maize plots across the major maize- producing zones of Zambia. Regression results based on control function approaches showed that FISP participation significantly reduces the likelihood of a household adopting IPM practices. More specifically, households that participate in FISP are about 106 percentage points less likely to regularly monitor their farms for early detection pest infestation and are 125 percentage points less likely to adopt preventative cultural measures, such as intercropping and rotation with non- host plants, field sanitation and trap cropping. Moreover, while FISP participation is significantly associated with a 67 percentage points higher probability of a household controlling FAW using synthetic pesticides, it is not significantly related to the adoption of biopesticides, which are considered a safer and low- risk alternative to synthetic pesticides. Worryingly, some of the pesticide- using farmers do not wear protective clothing against pesticide exposure, even though most of the synthetic pesticides used are moderately and highly hazardous. We also found ev- idence that farmers consider synthetic pesticides and biopesticides as substitutes, suggesting that the promotion of biopesticides is likely to discourage the use of synthetic pesticides. A disaggregated analysis suggested that the significant effects of FISP on increased use of synthetic pesticides is likely driven by the flexible e- voucher programme where beneficiary farmers can redeem vouchers for a wide range of subsidised farm inputs, including synthetic pesticides, from registered agro- dealers. Being located in a traditional FISP area, where the government distributes a fixed package of maize seed and fertiliser at subsidised rates to ben- eficiary farmers, has no significant effect on pesticide use. We also found heterogeneous ef- fects of the subsidy programmes on the adoption of non- chemical IPM practices. Results also showed that households with access to secure land tenure, credit and off- farm income earning activities, as well as those in low rainfall environments, where FAW infestation tend to be greater, are more likely to invest in IPM practices. Overall, the findings imply that the receipt of input subsidies through the Zambia FISP (flexible e- voucher system, in particular) encourages the use of synthetic pesticides, at the ex- pense of environmentally friendly and sustainable pest management strategies. The findings underscore arguments that while ISPs can raise crop yields in the short run, they may inadver- tently discourage the adoption of sustainable intensification practices and consequently un- dermine long- term yield and income gains (Jayne et al., 2018; Morgan et al., 2019). Moreover, the increased synthetic pesticide use induced by subsidies can have immediate and long- term effects on human health (such as respiratory, reproductive and neurological disorders), non- target species and biodiversity (Rani et al., 2021). Given the human and environmental health consequences associated with synthetic pes- ticides, it would be important to leverage the FISP to promote safer and more sustainable alternatives to synthetic pesticides. For example, the FISP beneficiaries who choose to re- deem their vouchers for pesticides could be required to opt for biopesticides rather than synthetic pesticides. Alternatively, higher subsidy amounts could be offered to those using biopesticides to encourage adoption. A good example that might be worth emulating is China's ‘Green Pest Control’ policy that provides subsidies on lower- risk crop protection products, including biopesticides (Wei et al., 2019). However, such a strategy can only be effective in Africa if biopesticide products are made readily available—for instance, by addressing registration bottlenecks (Constantine et al., 2020; Day et al., 2022; Srinivasan et  al.,  2019). Another potential strategy is to ensure that the array of items redeemable through the FISP include PPE items and non- chemical pest management inputs, such as biocontrol agents, pheromone traps and pest tolerant varieties. In addition, extension 14779552, 0, D ow nloaded from https://onlinelibrary.w iley.com /doi/10.1111/1477-9552.12582 by IN A SP - N E PA L , W iley O nline L ibrary on [23/04/2024]. See the T erm s and C onditions (https://onlinelibrary.w iley.com /term s-and-conditions) on W iley O nline L ibrary for rules of use; O A articles are governed by the applicable C reative C om m ons L icense 20 | TAMBO and LIVERPOOL- TASIE workers and agro- input dealers from whom smallholder farmers tend to receive pest man- agement information should be trained to offer advice on sustainable pest management and safe pesticide use. Our findings also suggest that beyond input subsidies, policies that improve tenure security and access to financial resources for smallholders (e.g., via improved access to credit and off- farm employment) can contribute to stimulating the adoption of sustainable IPM practices. For future research, it would be interesting to employ panel data approaches to investigate the long- term effects of FISP in terms of adoption of pest management strategies and related out- comes, given that our cross- sectional study was conducted in the early years of the outbreak of FAW during which some farmers may have limited knowledge on the various alternatives to pesticides for its control. It is possible that farmer participation in FISP can trigger be- havioural changes, and the effects may take several years to materialise. Moreover, due to data limitation, we were not able to assess the direct contribution of the traditional and e- voucher FISP programmes to sustainable pest management. This is another important area for further research. ACK NOW LEDGEM EN TS This work was supported by the CABI Development Fund (CDF) and the PlantwisePlus programme. CABI is an international intergovernmental organisation, and we grate- fully acknowledge the core financial support from our member countries (and lead agen- cies) including the United Kingdom (Foreign, Commonwealth & Development Office), China (Chinese Ministry of Agriculture), Australia (Australian Centre for International Agricultural Research), Canada (Agriculture and Agri- Food Canada), Netherlands (Directorate- General for International Cooperation), and Switzerland (Swiss Agency for Development and Cooperation). See www. cabi. org/ about - cabi/ who- we- work- with/ key- donors/ for full details. This study was also made possible with support from the US Department of Agriculture National Institute of Food and Agriculture and Michigan State University AgBioResearch, MICL12077. We also thank the Editor and two anonymous re- viewers for their helpful feedback and comments. CON F LICT OF I N T ER E ST STAT EM EN T The authors declare that they have no competing interests. DATA AVA I LA BI LI T Y STAT EM EN T The data that support the findings of this study are available on request from the correspond- ing author. The data are not publicly available due to privacy or ethical restrictions. I N FOR M ED CONSEN T STAT EM EN T Before starting an interview, a consent statement was read to each respondent. Only those who gave their consent were interviewed. ORCI D Justice A. Tambo  https://orcid.org/0000-0001-8895-7066 R E F ER E NC E S Abman, R., & Carney, C. (2020). Agricultural productivity and deforestation: Evidence from input subsidies and ethnic favoritism in Malawi. Journal of Environmental Economics and Management, 103, 102342. Altieri, M. A., & Nicholls, C. I. (2003). Soil fertility management and insect pests: Harmonizing soil and plant health in agroecosystems. Soil and Tillage Research, 72(2), 203–211. Alwang, J., Norton, G., & Larochelle, C. (2019). Obstacles to widespread diffusion of IPM in developing countries: Lessons from the field. Journal of Integrated Pest Management, 10(1), 10. 14779552, 0, D ow nloaded from https://onlinelibrary.w iley.com /doi/10.1111/1477-9552.12582 by IN A SP - N E PA L , W iley O nline L ibrary on [23/04/2024]. See the T erm s and C onditions (https://onlinelibrary.w iley.com /term s-and-conditions) on W iley O nline L ibrary for rules of use; O A articles are governed by the applicable C reative C om m ons L icense http://www.cabi.org/about-cabi/who-we-work-with/key-donors/ http://www.cabi.org/about-cabi/who-we-work-with/key-donors/ https://orcid.org/0000-0001-8895-7066 https://orcid.org/0000-0001-8895-7066 | 21 ARE FARM INPUT SUBSIDIES A DISINCENTIVE FOR INTEGRATED PEST MANAGEMENT ADOPTION? EVIDENCE FROM ZAMBIA Bateman, M. L., Day, R. K., Rwomushana, I., Subramanian, S., Wilson, K., Babendreier, D., Luke, B., & Edgington, S. (2021). Updated assessment of potential biopesticide options for managing fall armyworm (Spodoptera fru- giperda) in Africa. Journal of Applied Entomology, 145(5), 384–393. Bebber, D. P., Ramotowski, M. A., & Gurr, S. J. (2013). Crop pests and pathogens move polewards in a warming world. Nature Climate Change, 3(11), 985–988. Bezu, S., Kassie, G. T., Shiferaw, B., & Ricker- Gilbert, J. (2014). Impact of improved maize adoption on welfare of farm households in Malawi: A panel data analysis. World Development, 59, 120–131. Boedeker, W., Watts, M., Clausing, P., & Marquez, E. (2020). The global distribution of acute unintentional pesticide poisoning: Estimations based on a systematic review. BMC Public Health, 20(1), 1–19. Chibwana, C., Fisher, M., & Shively, G. (2012). Cropland allocation effects of agricultural input subsidies in Malawi. World Development, 40(1), 124–133. Constantine, K. L., Kansiime, M. K., Mugambi, I., Nunda, W., Chacha, D., Rware, H., Makale, F., Mulema, J., Lamontagne- Godwin, J., Williams, F., Edgington, S., & Day, R. (2020). Why don't smallholder farmers in Kenya use more biopesticides? Pest Management Science, 76(11), 3615–3625. Day, R., Abrahams, P., Bateman, M., Beale, T., Clottey, V., Cock, M., Colmenarez, Y., Corniani, N., Early, R., Godwin, J., Gomez, J., González- Moreno, P., Murphy, S. T., Oppong- Mensah, B., Phiri, N. A., Pratt, C., Silvestri, S., & Witt, A. (2017). Fall armyworm: Impacts and implications for Africa. Outlooks on Pest Management, 28(5), 196–201. Day, R., Haggblade, S., Moephuli, S., Mwang'ombe, A., & Nouala, S. (2022). Institutional and policy bottlenecks to IPM. Current Opinion in Insect Science, 52, 100946. Deguine, J. P., Aubertot, J. N., Flor, R. J., Lescourret, F., Wyckhuys, K. A., & Ratnadass, A. (2021). Integrated pest management: Good intentions, hard realities. A review. Agronomy for Sustainable Development, 41(3), 38. Deutsch, C. A., Tewksbury, J. J., Tigchelaar, M., Battisti, D. S., Merrill, S. C., Huey, R. B., & Naylor, R. L. (2018). Increase in crop losses to insect pests in a warming climate. Science, 361(6405), 916–919. Dhawan, A. K., & Peshin, R. (2009). Integrated pest management: Concept, opportunities and challenges. In R. Peshin & A. K. Dhawan (Eds.), Integrated pest management: Innovation- development process (Vol. 1, pp. 51–81). Springer Science. Di Falco, S., Veronesi, M., & Yesuf, M. (2011). Does adaptation to climate change provide food security? A micro- perspective from Ethiopia. American Journal of Agricultural Economics, 93(3), 829–846. Diffenbaugh, N. S., Krupke, C. H., White, M. A., & Alexander, C. E. (2008). Global warming presents new chal- lenges for maize pest management. Environmental Research Letters, 3(4), 044007. Early, R., González- Moreno, P., Murphy, S. T., & Day, R. (2018). Forecasting the global extent of invasion of the cereal pest Spodoptera frugiperda, the fall armyworm. NeoBiota, 40, 25–50. FAO & WHO. (2014). International code of conduct on pesticide management. Food and Agriculture Organization of the United Nations (FAO) and the World Health Organization (WHO). Farah, J. (1994). Pesticide policies in developing countries: Do they encourage excessive use. World Bank discussion paper No. 238. World Bank. Gebremedhin, B., & Swinton, S. M. (2003). Investment in soil conservation in northern Ethiopia: The role of land tenure security and public programs. Agricultural Economics, 29(1), 69–84. Githiomi, C., Muriithi, B., Irungu, P., Mwungu, C. M., Diiro, G., Affognon, H., Mburu, J., & Ekesi, S. (2019). Economic analysis of spillover effects of an integrated pest management (IPM) strategy for suppression of mango fruit fly in Kenya. Food Policy, 84, 121–132. Greene, W.H. (2012) Econometric Analysis (7th ed.). Prentice Hall. Grovermann, C., Schreinemachers, P., Riwthong, S., & Berger, T. (2017). Smart'policies to reduce pesticide use and avoid income trade- offs: An agent- based model applied to Thai agriculture. Ecological Economics, 132, 91–103. Haggblade, S., Diarra, A., & Traoré, A. (2022). Regulating agricultural intensification: Lessons from West Africa's rapidly growing pesticide markets. Development and Policy Review, 40(1), e12545. Harrison, R. D., Thierfelder, C., Baudron, F., Chinwada, P., Midega, C., Schaffner, U., & van den Berg, J. (2019). Agro- ecological options for fall armyworm (Spodoptera frugiperda JE smith) management: Providing low- cost, smallholder friendly solutions to an invasive pest. Journal of Environmental Management, 243, 318–330. Hemming, D. J., Chirwa, E. W., Dorward, A., Ruffhead, H. J., Hill, R., Osborn, J., Langer, L., Harman, L., Asaoka, H., Coffey, C., & Phillips, D. (2018). Agricultural input subsidies for improving productivity, farm income, consumer welfare and wider growth in low- and lower- middle- income countries: A systematic review. Campbell Systematic Reviews, 14(1), 1–153. Holden, S., & Lunduka, R. (2012). Do fertilizer subsidies crowd out organic manures? The case of Malawi. Agricultural Economics, 43(3), 303–314. Holden, S. T. (2019). Economics of farm input subsidies in Africa. Annual Review of Resource Economics, 11, 501–522. 14779552, 0, D ow nloaded from https://onlinelibrary.w iley.com /doi/10.1111/1477-9552.12582 by IN A SP - N E PA L , W iley O nline L ibrary on [23/04/2024]. See the T erm s and C onditions (https://onlinelibrary.w iley.com /term s-and-conditions) on W iley O nline L ibrary for rules of use; O A articles are governed by the applicable C reative C om m ons L icense 22 | TAMBO and LIVERPOOL- TASIE Hruska, A. J. (2019). Fall armyworm (Spodoptera frugiperda) management by smallholders. CAB Reviews, 14(43), 1–11. Jayne, T. S., Mason, N. M., Burke, W. J., & Ariga, J. (2018). Taking stock of Africa's second- generation agricultural input subsidy programs. Food Policy, 75, 1–14. Jepson, P. C., Guzy, M., Blaustein, K., Sow, M., Sarr, M., Mineau, P., & Kegley, S. (2014). Measuring pesticide eco- logical and health risks in west African agriculture to establish an enabling environment for sustainable inten- sification. Philosophical Transactions of the Royal Society, B: Biological Sciences, 369(1639), 20130491. Kansiime, M. K., Mugambi, I., Rwomushana, I., Nunda, W., Lamontagne- Godwin, J., Rware, H., … Day, R. (2019). Farmer perception of fall armyworm (Spodoptera frugiderda JE smith) and farm- level management practices in Zambia. Pest Management Science, 75(10), 2840–2850. Khonje, M. G., Nyondo, C., Chilora, L., Mangisoni, J. H., Ricker- Gilbert, J., & Burke, W. J. (2022). Exploring adoption effects of subsidies and soil fertility management in Malawi. Journal of Agricultural Economics, 73(3), 874–892. Kim, J., Mason, N. M., Mather, D., & Wu, F. (2021). The effects of the national agricultural input voucher scheme (NAIVS) on sustainable intensification of maize production in Tanzania. Journal of Agricultural Economics, 72(3), 857–877. Koppmair, S., Kassie, M., & Qaim, M. (2017). The inf luence of farm input subsidies on the adoption of natu- ral resource management technologies. Australian Journal of Agricultural and Resource Economics, 61(4), 539–556. Kuntashula, E., & Mwelwa- Zgambo, L. (2022). Impact of the farmer input support policy on agricultural production diversity and dietary diversity in Zambia. Food Policy, 113, 102329. Kuteya, A., Lukama, C., Chapoto, A., & Malata, V. (2016). Lessons learnt from the implementation of the e- voucher pilot. IAPRI Policy Brief No. 81. Liverpool- Tasie, L. S. O. (2014). Fertilizer subsidies and private market participation: The case of Kano state, Nigeria. Agricultural Economics, 45(6), 663–678. Mason, N. M., & Jayne, T. S. (2013). Fertiliser subsidies and smallholder commercial fertiliser purchases: Crowding out, leakage and policy implications for Zambia. Journal of Agricultural Economics, 64(3), 558–582. Mason, N. M., Jayne, T. S., & Mofya- Mukuka, R. (2013). Zambia's input subsidy programs. Agricultural Economics, 44(6), 613–628. Mason, N. M., Kuteya, A., Ngoma, H., Tossou, D. A., & Baylis, K. R. (2020). Does shifting to a f lexible E- voucher approach improve input subsidy program outcomes? Quasi- experimental evidence from Zambia's crop forecast surveys. In Feed the future innovation lab for food security policy research paper 165. Michigan State University. Mason, N. M., & Ricker- Gilbert, J. (2013). Disrupting demand for commercial seed: Input subsidies in Malawi and Zambia. World Development, 45, 75–91. Mason, N. M., & Smale, M. (2013). Impacts of subsidized hybrid seed on indicators of economic well- being among smallholder maize growers in Zambia. Agricultural Economics, 44(6), 659–670. Mason, N. M., Wineman, A., & Tembo, S. T. (2020). Reducing poverty by ‘ignoring the experts’? Evidence on input subsidies in Zambia. Food Security, 12(5), 1157–1172. Midingoyi, S. K. G., Kassie, M., Muriithi, B., Diiro, G., & Ekesi, S. (2019). Do farmers and the environment benefit from adopting integrated pest management practices? Evidence from Kenya. Journal of Agricultural Economics, 70(2), 452–470. Morgan, S. N., Mason, N. M., Levine, N. K., & Zulu- Mbata, O. (2019). Dis- incentivizing sustainable intensification? The case of Zambia's maize- fertilizer subsidy program. World Development, 122, 54–69. Mulenga, B. P., Kabisa, M., & Chapoto, A. (2021). Zambia agriculture status report 2021. Indaba Agricultural Policy Research Institute. National Assembly of Zambia. (2022). Ministerial statement on the farmer input support programme status of the 2022/2023 farming season. National Assembly of Zambia. National Assembly of Zambia. (2023). Ministerial statement on the distribution of inputs for the 2023–2024 farming season. National Assembly of Zambia. Ngoma, H., Machina, H., & Kuteya, A. N. (2021). Can agricultural subsidies reduce gendered productivity gaps? Panel data evidence from Zambia. Development and Policy Review, 39(2), 303–323. Norton, G. W., Alwang, J., Kassie, M., & Muniappan, R. (2019). Economic impacts of IPM practices in developing countries. In D. W. Onstad & P. R. Crane (Eds.), The economics of integrated pest management of insects (pp. 140–154). CABI Publishing. Parsa, S., Morse, S., Bonifacio, A., Chancellor, T. C., Condori, B., Crespo- Pérez, V., Hobbs, S. L., Kroschel, J., Ba, M. N., Rebaudo, F., Sherwood, S. G., Vanek, S. J., Faye, E., Herrera, M. A., & Dangles, O. (2014). Obstacles to integrated pest management adoption in developing countries. Proceedings of the National Academy of Sciences of the United States of America, 111(10), 3889–3894. 14779552, 0, D ow nloaded from https://onlinelibrary.w iley.com /doi/10.1111/1477-9552.12582 by IN A SP - N E PA L , W iley O nline L ibrary on [23/04/2024]. See the T erm s and C onditions (https://onlinelibrary.w iley.com /term s-and-conditions) on W iley O nline L ibrary for rules of use; O A articles are governed by the applicable C reative C om m ons L icense | 23 ARE FARM INPUT SUBSIDIES A DISINCENTIVE FOR INTEGRATED PEST MANAGEMENT ADOPTION? EVIDENCE FROM ZAMBIA Pingali, P. L., & Rosegrant, M. W. (2001). Intensive food systems in Asia: Can the degradation problems be reversed? In D. R. Lee & C. B. Barret (Eds.), Tradeoffs or synergies? Agricultural intensification, economic development and the environment (pp. 383–397). CABI Publishing. Pratt, C. F., Constantine, K. L., & Murphy, S. T. (2017). Economic impacts of invasive alien species on African small- holder livelihoods. Global Food Security, 14, 31–37. Rani, L., Thapa, K., Kanojia, N., Sharma, N., Singh, S., Grewal, A. S., Srivastav, A. L., & Kaushal, J. (2021). An extensive review on the consequences of chemical pesticides on human health and environment. Journal of Cleaner Production, 283, 124657. Repetto, R. (1985). Paying the price: Pesticide subsidies in developing countries. Research report No. 2. World Resources Institute. Ricker- Gilbert, J., Jayne, T., & Shively, G. (2013). Addressing the “wicked problem” of input subsidy programs in Africa. Applied Economic Perspectives and Policy, 35(2), 322–340. Ricker- Gilbert, J., Jayne, T. S., & Chirwa, E. (2011). Subsidies and crowding out: A double- hurdle model of fertilizer demand in Malawi. American Journal of Agricultural Economics, 93(1), 26–42. Ricker- Gilbert, J., Mason, N. M., Darko, F. A., & Tembo, S. T. (2013). What are the effects of input subsidy programs on maize prices? Evidence from Malawi and Zambia. Agricultural Economics, 44(6), 671–686. Rwomushana, I., Bateman, M., Beale, T., Beseh, P., Cameron, K., Chiluba, M., Clottey, V., Davis, T., Day, R., Early, R., Godwin, J., Gonzalez- Moreno, P., Kansiime, M., Kenis, M., Makale, F., Mugambi, I., Murphy, S., Nunda, W., Phiri, N., … Tambo, J. (2018). Fall armyworm: Impacts and implications for Africa evidence note update, October 2018. Report to DFID. CABI. Savary, S., Willocquet, L., Pethybridge, S. J., Esker, P., McRoberts, N., & Nelson, A. (2019). The global burden of pathogens and pests on major food crops. Nature Ecology & Evolution, 3(3), 430–439. Shattuck, A., Werner, M., Mempel, F., Dunivin, Z., & Galt, R. (2023). Global pesticide use and trade database (GloPUT): New estimates show pesticide use trends in low- income countries substantially underestimated. Global Environmental Change, 81, 102693. Srinivasan, R., Sevgan, S., Ekesi, S., & Tamò, M. (2019). Biopesticide based sustainable pest management for safer production of vegetable legumes and brassicas in Asia and Africa. Pest Management Science, 75(9), 2446–2454. Stokstad, E. (2017). New crop pest takes Africa at lightning speed. Science, 356(6337), 473–474. Takeshima, H., & Liverpool- Tasie, L. S. O. (2015). Fertilizer subsidies, political influence and local food prices in sub- Saharan Africa: Evidence from Nigeria. Food Policy, 54, 11–24. Tambo, J. A., Kansiime, M. K., Mugambi, I., Rwomushana, I., Kenis, M., Day, R. K., & Lamontagne- Godwin, J. (2020). Understanding smallholders' responses to fall armyworm (Spodoptera frugiperda) invasion: Evidence from five African countries. Science of the Total Environment, 740, 140015. Tambo, J. A., & Mockshell, J. (2018). Differential impacts of conservation agriculture technology options on house- hold income in sub- Saharan Africa. Ecological Economics, 151, 95–105. Tambo, J.A., Romney, D., Mugambi, I., Mbugua, F., Bundi, M., Uzayisenga, B., Matimelo, M. & Ndhlovu, M. (2021) Can plant clinics enhance judicious use of pesticides? Evidence from Rwanda and Zambia. Food Policy, 101(9), 102073. https:// doi. org/ 10. 1016/j. foodp ol. 2021. 102073 Terza, J. V., Basu, A., & Rathouz, P. J. (2008). Two- stage residual inclusion estimation: Addressing endogeneity in health econometric modeling. Journal of Health Economics, 27(3), 531–543. Varella, A. C., Menezes- Netto, A. C., Alonso, J. D. D. S., Caixeta, D. F., Peterson, R. K., & Fernandes, O. A. (2015). Mortality dynamics of Spodoptera frugiperda (Lepidoptera: Noctuidae) immatures in maize. PLoS ONE, 10(6), e0130437. Wei, X., Zhao, L., Qiao, Y., Wang, B., Wan, M., & Toepfer, S. (2019). Implementing Agripolicies on Pesticide Reduction through Subsidies and Plant Clinics in China (pp. 13–25). CABI. WHO. (2020). WHO recommended classification of pesticides by hazard and guidelines to classification (2019th ed.). World Health Organization. Williamson, S., Ball, A., & Pretty, J. (2008). Trends in pesticide use and drivers for safer pest management in four African countries. Crop Protection, 27(10), 1327–1334. Wooldridge, J. M. (2014). Quasi- maximum likelihood estimation and testing for nonlinear models with endogenous explanatory variables. Journal of Econometrics, 182(1), 226–234. Wooldridge, J. M. (2015). Control function methods in applied econometrics. Journal of Human Resources, 50(2), 420–445. Wossen, T., Abdoulaye, T., Alene, A., Feleke, S., Ricker- Gilbert, J., Manyong, V., & Awotide, B. A. (2017). Productivity and welfare effects of Nigeria's e- voucher- based input subsidy program. World Development, 97, 251–265. Xu, Z., Burke, W. J., Jayne, T. S., & Govereh, J. (2009). Do input subsidy programs “crowd in” or “crowd out” commercial market development? Modeling fertilizer demand in a two- channel marketing system. Agricultural Economics, 40(1), 79–94. 14779552, 0, D ow nloaded from https://onlinelibrary.w iley.com /doi/10.1111/1477-9552.12582 by IN A SP - N E PA L , W iley O nline L ibrary on [23/04/2024]. See the T erm s and C onditions (https://onlinelibrary.w iley.com /term s-and-conditions) on W iley O nline L ibrary for rules of use; O A articles are governed by the applicable C reative C om m ons L icense https://doi.org/10.1016/j.foodpol.2021.102073 24 | TAMBO and LIVERPOOL- TASIE SU PPORT I NG I N FOR M AT ION Additional supporting information can be found online in the Supporting Information section at the end of this article. How to cite this article: Tambo, J.A. & Liverpool- Tasie, L.S.O. (2024) Are farm input subsidies a disincentive for integrated pest management adoption? Evidence from Zambia. Journal of Agricultural Economics, 00, 1–24. Available from: https://doi. org/10.1111/1477-9552.12582 14779552, 0, D ow nloaded from https://onlinelibrary.w iley.com /doi/10.1111/1477-9552.12582 by IN A SP - N E PA L , W iley O nline L ibrary on [23/04/2024]. See the T erm s and C onditions (https://onlinelibrary.w iley.com /term s-and-conditions) on W iley O nline L ibrary for rules of use; O A articles are governed by the applicable C reative C om m ons L icense https://doi.org/10.1111/1477-9552.12582 https://doi.org/10.1111/1477-9552.12582 Are farm input subsidies a disincentive for integrated pest management adoption? Evidence from Zambia Abstract 1|INTRODUCTION 2|FARMER INPUT SUPPORT PROGRAMME (FISP) IN ZAMBIA 3|DATA AND METHODS 3.1|Data 3.2|Empirical approach 4|RESULTS AND DISCUSSION 4.1|Descriptive results 4.2|Determinants of FISP participation 4.3|FISP effects on adoption of IPM practices 4.4|Differential effects by FISP modality 5|CONCLUSION ACKNOWLEDGEMENTS CONFLICT OF INTEREST STATEMENT DATA AVAILABILITY STATEMENT INFORMED CONSENT STATEMENT REFERENCES