Economic Benefits and Costs of Biological Control: Coffee Berry Borer, Erythrina Gall Wasp, and Fireweed in Hawai‘i

Executive Summary

This report evaluates the economic costs and benefits associated with biological control efforts targeting three invasive species in Hawai‘i: coffee berry borer (CBB, Hypothenemus hampei), erythrina gall wasp (EGW, Quadrastichus erythrinae), and fireweed (Senecio madagascariensis). Utilizing historical data, expert interviews, and cost-benefit analyses, the study quantifies potential economic impacts and returns on investment (ROI) for ongoing and proposed biocontrol programs.

Key findings indicate substantial economic benefits from biocontrol interventions:

1. Coffee Berry Borer (CBB):
   • The release of the biocontrol agent, Phymastichus coffea, is projected to significantly reduce existing management costs such as spraying and strip-picking.
   • Under baseline assumptions (full replacement of current management practices), the net present value (NPV) of benefits over 50 years is estimated at $141.6 million with a benefit-cost ratio (BCR) of 42.0.
   • Even for conservative scenarios (25-50% effectiveness), NPV remains strongly positive ($32.8–$69.1 million), with benefit-cost ratios ranging from 10.5 to 21.0.
   • Breakeven is expected within two years of biocontrol implementation.
2. Erythrina Gall Wasp (EGW):
   • Existing biocontrol measures (Eurytoma erythrinae) have substantially reduced EGW impacts, protecting native wiliwili (Erythrina sandwicensis) forests from severe ecological and cultural damage.
   • An additional biocontrol agent (Aprostocetus nitens) pending release is anticipated to further improve control.
   • The economic analysis projects an NPV of approximately $26 million over a 50-year period, with a high benefit-cost ratio of 24.0 and a breakeven point of 2 years.
   • Significant benefits include maintaining cultural and ecological functions of dryland forests, successful establishment of forest preserves, and avoided costs of potential listing as a federally endangered species.
3. Fireweed:
   • The moth Secusio extensa was introduced to reduce fireweed populations affecting pastureland and cattle ranching profitability.
   • Depending on the scenario considered (either management cost savings or avoided cattle industry losses), the biocontrol program’s NPV ranges from $2.6–$21.2 million, with BCR estimates between 2.0 and 9.2.
   • Breakeven points vary from approximately 2 to 13 years post-release, with regional effectiveness influencing outcomes significantly.

Overall, the analysis illustrates the potential of biocontrol to be a highly cost-effective invasive species management strategy in Hawai‘i. However, results demonstrate variability across target species in terms of economic magnitude, certainty, and breakeven timeline. These findings reinforce the necessity of careful planning and sustained monitoring for maximizing the economic and ecological effectiveness of biocontrol interventions.

Introduction

The utilization of biocontrol, employing natural adversaries to combat invasive species, dates back over a century in Hawai‘i. After enacting its first laws for the implementation of biocontrols in 1890, Hawai‘i introduced a total of 679 species between 1890 and 1965 for this purpose. A study by Funasaki et al. (1988) revealed that 86.4% of these introduced species had no adverse effects on non-target organisms, a remarkable statistic considering the absence of common practices such as host specificity testing and comprehensive risk assessment procedures during that period.

The 1980s marked a critical juncture for Hawai‘i concerning the unintended consequences of biological control introductions. Works by Howarth (1983) and Gagne and Howarth (1985) highlighted population declines and even extinctions among Lepidoptera species in Hawai‘i, attributing them to parasitism by deliberately introduced biological control agents. Subsequent to these publications and others on the non-target effects of biological control introductions, many nations implemented stricter laws and regulations governing the practice of biological control (Messing and Brodeur, 2018), heralding a significant transformation in the landscape of biocontrol both in Hawai‘i and globally.

Though preventing the arrival of invasive species is generally preferable to reacting after the fact, the former proves challenging due to numerous potential pathways for introduction. Moreover, detection often occurs only once an invasion is well underway, rendering eradication an unrealistic management strategy. Depending on the stage of invasion and the characteristics of the invasive species, conventional control methods (e.g., chemical, mechanical, cultural), may prove ineffective or prohibitively expensive. Inaction, however, carries its own costs as the invasion continues to spread, wreaking havoc on ecosystems and undermining their ecological, cultural, and economic value. Biological control agents, with their capacity for self-dispersion and self-sustainment across diverse ecosystems, may offer a more cost-effective long-term solution, provided adequate precautions are taken to mitigate impacts on non-target species.

Given the potential risks associated with non-target effects, it comes as no surprise that biocontrol regulations have tightened globally, particularly in regions like Hawai‘i with high rates of endemism and numerous threatened and endangered species. While necessary, these regulations increase the time and cost involved in obtaining permits for importing and releasing biocontrol agents. Compounded by a lack of adequate containment facilities for non-target screening and research on prospective biocontrol agents, these constraints underscore the importance of proposed research aimed at examining historical and ongoing biocontrol efforts in Hawai‘i. By focusing on quantifying the economic costs and benefits of such interventions, where possible, this research endeavors to inform future management decisions affecting various sectors. We hypothesize that the return on investment to classical biocontrol in Hawai‘i is generally high, similar to what has been observed in other locations.

This study identifies three invasive species—fireweed (S. madagascariensis), erythrina gall wasp (Q. erythrinae), and coffee berry borer (H. hampei)—for which biocontrol agents have been released or are in the permitting process. Unchecked, these invasive species inflict damage across multiple sectors, including ecologically and culturally significant natural resources (erythrina gall wasp), farming (coffee berry borer), and ranching (fireweed). Consequently, enhancing our comprehension of the benefits and costs of biocontrol interventions through this research will facilitate informed invasive species management decisions in the future.

To this end, the research team conducted interviews with relevant agencies and affected individuals statewide to gather crucial economic data, encompassing costs associated with prospecting for and collecting biocontrol agents, testing facilities, research personnel, release, monitoring, as well as avoided future management costs and commercial damages. Additionally, data from scientific literature and local experts on the efficacy and establishment rates of biocontrol agents, infested acreage, and rate of spread of invasive species were employed to develop counterfactual scenarios and to estimate the economic benefits of biocontrol for each case study.

Background Information

Coffee Berry Borer (Hypothenemus hampei)

The coffee berry borer (H. hampei), commonly referred to as CBB, is recognized as the most economically harmful insect pest of coffee worldwide. It causes damages exceeding $500 million annually and affects the livelihoods of 25 million rural households involved in coffee production globally (Vega et al., 2003). CBB reduces not only the yield of infested coffee farms but also the quality of the coffee produced (Le Pelley, 1968). While the reduction in output has a substantial impact on the bottom line of coffee farmers, the decline in quality—particularly changes to the flavor profiles of recoverable coffee beans—should not be overlooked. This is especially critical in regions like Hawai‘i, which command premium prices on the global market due to their unique origins and historically high-quality coffee (Kinro, 2003).

Native to Central Africa, CBB was first detected in the Kona coffee-growing region of Hawai‘i in 2010 (Burbano et al., 2011). It rapidly spread to the hundreds of coffee farms on Hawai‘i Island, and not long thereafter, it was found on the neighboring islands of O‘ahu (HDOA, 2014), Maui (HDOA, 2016), Lāna‘i (HDOA, 2020a), and Kaua‘i (HDOA, 2020b). Although data on statewide CBB infestation levels are not regularly collected, Lee et al. (2023) extrapolated that approximately 6,500 of the 6,900 (94%) coffee-bearing acres in Hawai‘i were infested by 2021, with potential CBB reservoirs also growing in wild coffee populations on the Big Island, Lāna‘i, and other areas across the state. CBB spreads quickly not only between farms but also within farms, with some Kona farmers reporting infestation levels of up to 80-95% of their harvested crop shortly after the initial outbreak in Kona (Woodill et al., 2017; Johnson et al., 2020; Lee et al., 2023). A recent study supports this observation, finding that poorly managed CBB populations on Hawai‘i farms multiply rapidly and can infest more than 90% of the coffee crop by the end of a season (Johnson & Manoukis, 2020). This suggests that the economic damages caused by CBB are potentially very large in the absence of effective management.

According to the USDA Census of Agriculture, Hawai‘i had 1,111 coffee farms in 2022, down from 1,477 farms in 2017 (USDA NASS, 2024a). In the 2022-23 season, 23.6 million lbs of coffee cherry were produced on 7,500 acres (USDA NASS, 2024b), a small fraction of the 25 million acres planted worldwide. However, the $54.1 million value generated by those 7,500 acres makes coffee one of the most economically significant agricultural crops in the state, second only to seed crops in value (Aristizábal et al., 2023b). Thus, the already realized and future potential economic impacts of CBB are significant, as are the expected benefits of a safe and effective biocontrol agent.

Since CBB was detected in Hawai‘i over a decade ago, it is not surprising that farmers have adopted various pest management strategies in the years since, even in the absence of biocontrol. In 2011, the GHA strain of the entomopathogenic fungus Beauveria bassiana (sold commercially as BotaniGard® ES and Mycotrol® ESO) was approved for use as an insecticide on Hawaiian coffee (Lee et al., 2023). Since 2014, the USDA and the Hawai‘i Department of Agriculture (HDOA) have spent more than $2 million in total on subsidies and spray equipment to incentivize farmers to monitor and apply B. bassiana if they have CBB (Johnson et al., 2020). More recently, the USDA allocated approximately $5 million to a cooperative area-wide program aimed at better understanding coffee plant phenology, CBB flight activity, berry infestation, mortality from B. bassiana, current farm management practices, and weather patterns. This research is intended to inform and develop effective integrated pest management (IPM) strategies for CBB in Hawai‘i (Johnson et al., 2020).

Farmers who regularly follow IPM strategies, including field sanitation measures such as efficient harvesting practices, strip-picking trees at the end of the season, and proper application of B. bassiana, have observed reductions in CBB infestation levels. In the Kona and Ka‘ū districts of Hawai‘i Island, for example, average infestation levels at commercial farms fell from 20.5% in 2015 to 3.7% in 2022 (Aristizábal et al., 2023a). By comparing modeled statewide coffee yield, price, and revenue with and without CBB management over the period 2011 to 2021, Lee et al. (2023) estimated that these IPM practices (B. bassiana application, monitoring, and cultural controls) generated $251 million in economic benefits.

Given the proven effectiveness of current IPM practices, one may wonder whether a biocontrol agent for CBB is warranted. The answer depends on how the avoided management costs—ideally, the biocontrol would greatly reduce, if not eliminate, the need for spraying and other labor-intensive sanitation practices—compare to the costs of collecting, researching, rearing, releasing, and monitoring the biocontrol agent. The ceiling for these potential benefits is high in Hawai‘i, due to the relatively high cost of labor. Hawaiian-grown coffee is 5 to 7 times more expensive to produce than the global average (Aristizábal et al., 2017), and labor makes up 39% of total costs for farms (Woodill, 2014). Meanwhile, B. bassiana spray costs alone were estimated to be 5 to 12% of profits (Aristizábal et al., 2023a).

P. coffea is currently being considered for introduction into Hawai‘i as a biocontrol agent for CBB. The parasitoid wasp, which kills CBB females before they can penetrate and damage the coffee endosperm, has already been introduced in at least 12 other countries (Johnson et al., 2020). Lab tests have shown high host specificity (Yousuf et al., 2021), which is one of, if not the most important aspects to consider for the introduction of biocontrol agents. The Final Environmental Assessment for the field release of P. coffea in Hawai‘i was published in 2023 (USDA APHIS, 2023), with a decision and finding of no significant impact. Efforts are being made to import and establish P. coffea in Hawai‘i at the time of this writing, but the effects of that introduction will not be available for our analysis. Thus, benefit calculations will be hypothetical, based on the best available information. It should also be noted that, while other biocontrol agents for CBB have been considered by the research community in Hawai‘i, our study focuses on P. coffea in light of its recent permit approval and impending release.

Erythrina Gall Wasp (Q. erythrinae)

The erythrina gall wasp (EGW), Q. erythrinae, is a gall-forming eulophid wasp native to Madagascar (HDOA, 2023). Its life cycle spans approximately 20 days from egg to adult, with each female carrying an average of 322 eggs (Xu et al., 2009). This pest was first detected in Hawai‘i in Mānoa in April 2005, spreading quickly throughout the rest of the state within six months (HDOA, 2023). The invasion has impacted dry forests, which are estimated to currently cover approximately 260,000 acres, less than 1-5% of the original extent prior to human settlement of Hawai‘i (Javar-Salas et al., 2020; Pau et al., 2009; Sakai et al., 2002).

The EGW inflicts damage on the endemic E. sandwicensis (wiliwili), causing severe galling on leaves that can lead to defoliation and eventual death, with some populations initially showing greater than 40% mortality (Yalemar et al., 2016). The impact of EGW extends to other Erythrina species, often utilized for landscaping purposes, including E. crista-galli and E. variegata (HDOA, 2023). While not extensively used commercially, wiliwili trees hold significant cultural value historically, especially in traditional practices like fishing and crafting of outriggers (ama), surfboards, and lei (HDOA, 2023). The EGW’s invasion timeline spans from its initial appearance in 2005 to the present, with the highest proportion of Erythrina losses in the initial years after detection (HDOA, 2023).

In response to EGW’s statewide infestation and threat to wiliwili’s existence, HDOA began exploratory efforts in search of a biocontrol agent while targeted management practices for controlling EGW infestations were attempted. Before the introduction of biocontrol, imidacloprid was the primary method of control, and is still utilized by various entities including Native Hawaiian plant nurseries and local businesses (Hui Kū Maoli Ola, personal communication, 2023). Xu et al. (2008) recommended imidacloprid injections into the tree trunk every 9-12 months for mature trees. However, since 2008, biocontrol measures, specifically the use of E. erythrinae, have been implemented.

The Eurytomidae parasitoid wasp, E. erythrinae, native to Tanzania, Ghana, and South Africa, plays a crucial role in controlling EGW in its native habitat (HDOA, 2023). The larvae of E. erythrinae feed on multiple EGW larvae and therefore tunnel into adjacent galls for further predation (Yalemar et al., 2016). After host-specificity testing and permit approval, over 8,000 E. erythrinae were released in late 2008 across O‘ahu, Maui, Kaua‘i, Moloka‘i, and Hawai‘i Island (HDOA, 2023). Field effectiveness varied based on location and environmental conditions, with parasitism rates ranging from 20% to 100% (U.S. Forest Service, 2014). Strong efficacy of this biological control agent (BCA) was seen, with signs of recovery in wiliwili trees within just a few months post-release (Yalemar et al., 2016). By 2011, 90% of sampled wiliwili trees had full canopies (HDOA, 2023). However, flowering and seed production are still affected by EGW as galls that form on flowers and fruits are smaller and less easily parasitized by E. erythrinae. Therefore a second, complementary biocontrol has been extensively studied and proposed for release.

A. nitens is an ectoparasitoid native to Tanzania and South Africa (HDOA, 2023). Its larva serves as a biological control agent by feeding on a single EGW pupa each (HDOA, 2023). A. nitens is currently being reared here for initial releases, the timing of which depends on the final approval from the Board of Agriculture (BOA) and the issuance of permits by the Animal and Plant Health Inspection Service (APHIS), with an expected timeline of several months (Oishi, 2023); at the time of writing A. nitens has not been approved for release. Once the permit request is approved, the colony of A. nitens will be transferred from HDOA Plant Pest Control Branch’s Containment Facility to its Insect Rearing Facility for mass rearing (Oishi, 2023). Releases across the state at EGW-infested Erythrina would continue until establishment of the BCA (Oishi, 2023).  Note that E. erythrinae established well upon initial release and did not require secondary releases (M. Wright, personal communication). The post-release monitoring plan for A. nitens entails monthly visits to release sites for the initial five years, followed by reduced visits every two to three months over the subsequent five years (L. Kaufman, personal communication).

Although the direct monetary costs of EGW’s devastation of wiliwili and other Erythrina species can be difficult to explicitly quantify, the survival of wiliwili trees is crucial for the perpetuation of both ecologically important dry forests and Hawaiian cultural practices. Parcels, including the Waikōloa and Palamanui Dry Forest Preserves, have been set aside for protection in large part because of the presence of large wiliwili trees. Additionally, the annual Wiliwili Festival attracts locals and visitors, generating revenue for the Waikōloa Dry Forest Initiative. The avoided cost of listing wiliwili as endangered if it were to near extinction is an additional benefit of BCA success.

Fireweed (Senecio madagascariensis)

Fireweed (Senecio madagascariensis), a daisy-like invasive species endemic to Madagascar, was first identified in Hawai‘i circa 1980 (HDOA, 2010). Initial infestations were recorded on the islands of Hawai‘i (Big Island) and Maui, but the species has since exhibited rapid proliferation. By 2010, it had colonized an estimated 400,000 acres across the state (Ramadan et al., 2010). This accelerated spread is largely attributed to the plant’s prolific reproductive capacity, with individual plants capable of producing approximately 30,000 seeds annually. These seeds demonstrate remarkable resilience, remaining viable for up to 50 years. Dispersal occurs through various vectors, including wind, animals, human activity (e.g., vehicles and clothing), facilitating the plant’s wide and efficient colonization. Despite mitigation efforts, fireweed distribution has remained largely unchanged since 2010, favoring elevations between 1,500 and 11,000 feet with annual rainfall around 20 inches and limited canopy cover (Ramadan et al., 2010).

The ecological and economic consequences of fireweed invasion in Hawai‘i have been significant. Livestock, particularly cattle, have suffered fatalities from ingesting the plant, which contains toxic pyrrolizidine alkaloids (Thorne et al., 2005; Brooks, 2013). In addition to direct toxicity, fireweed’s aggressive growth has rendered agricultural lands unsuitable for cultivation, reducing farm productivity and profitability by an estimated 15–50% (Ramaden et al., 2023). The application of herbicides, while sometimes necessary for control, poses the risk of collateral damage to surrounding vegetation (HDOA, 2010). Notably, forage production in affected areas has declined by as much as 40%, significantly impacting the cattle industry through reductions in available grazing land and herd sizes over time (Ramaden et al., 2023; HDOA, 2010). These adverse effects have persisted for over four decades, dating back to the early 1980s (Motooka et al., 1999). Even in non-lethal cases, the ingestion of fireweed by cattle has been linked to reduced quality of milk and meat, thereby diminishing the commercial value of livestock products. The economic burden is further compounded by the increased logistical demands of relocating cattle for grazing and the inability of veterinarians to isolate sick animals in infested areas (C. Martin, personal communication).

Management efforts for fireweed in Hawai‘i primarily encompass chemical, biological, and grazing-based strategies. Herbicidal treatment has been employed, though its efficacy is limited due to the large scale of infestation and the financial burden associated with application across extensive acreage. Additionally, herbicides become less effective once fireweed reaches maturity, often necessitating increased concentrations for control (Department of Agriculture and Fisheries Queensland, 2024). Grazing management using sheep and goats has demonstrated potential, as these animals are not susceptible to the plant’s toxins and can aid in its suppression. However, this method remains underutilized by ranchers (M. Thorne, personal communication). Control efforts are further complicated by the presence of other pests, such as the two-lined spittlebug, which compound the ecological challenges (M. Thorne, personal communication).

A promising biocontrol agent, Secusio extensa, a moth species native to Madagascar and first described by Butler in 1880, has been introduced to mitigate fireweed proliferation. Although Secusio primarily targets cape ivy, it also consumes fireweed foliage. The State of Hawai‘i approved its release in 2010, with permits finalized in 2012. The initial release consisted of 50 adult moths and 500 larvae at Haleakalā Ranch, followed by the dissemination of 2,000 additional individuals to ten other ranches on Maui in 2013 (HDOA, 2013). The moths were propagated in captivity, with the goal of producing approximately 10,000 eggs every 40 days across 10 rearing cages. The Hawai‘i Department of Agriculture provided funding for the initial three years, with the project overseen by Mark Thorne on both Maui and the Big Island (M. Thorne, personal communication).

Research indicates that S. extensa can reduce fireweed populations without causing substantial harm to non-target species (Ramadan et al., 2010). Nevertheless, its efficacy appears to differ between islands, with more favorable outcomes on Maui, potentially due to more cohesive ranching areas. In contrast, on the Big Island, challenges such as greater population dispersion and the moth’s attraction to urban lighting have impeded its success (M. Thorne, personal communication). Given the longevity of fireweed’s seed bank, it may take several decades to fully assess the long-term impact of S. extensa. Nonetheless, anecdotal evidence points to a reduction in fireweed density in some areas (M. Thorne, personal communication). The development and approval process for this biocontrol initiative involved 13 years of research, along with considerable investment in equipment, staffing, materials, and logistics.

Benefit-Cost Framework

The economic assessment reflects biological benefits and costs, or those that scientists consider when exploring possible biocontrol agents; for example, which structures on the target invasive species that the agent attacks (i.e. reproductive, stems, leaves), which affects the efficacy of the agent and thus the benefits of biocontrol, or if the agent can easily become established in the field, which reduces the cost of implementing biocontrol and increases the rate of accumulated potential benefits.

In this assessment, the benefits of biocontrol are defined as cost-savings in comparison with current control methods and avoided economic damages. For example, if current control costs $1 million annually and biocontrol achieves the same efficacy as this control, then the benefit of biocontrol would be $1 million per year. These cost-savings represent a lower-bound estimate of the benefits of biocontrol, because they do not take into account all of the benefits associated with lowering the population of the target invasive species, which may include ecological and cultural values that are often difficult to quantify. This assessment assumes that the three invasive species should be managed and examines the benefits of using biocontrol in comparison with current/past control methods. Several benefit scenarios are evaluated, representing varying levels of biocontrol efficacy depending on the invasive species being considered and the mode of action of the biocontrol agent.

There are two main classes of biocontrol costs: pre-release costs and post-release costs. Pre-release costs of biocontrol include all of the expenses incurred prior to the release of the agent: foreign exploration, quarantine and testing, initial release, as well as facility and lab space/personnel. These costs are paid once for every agent and are independent of the amount of agent released or how much the agent is used in the future. The high up-front pre-release costs reflect the risk associated with introducing new species into native environments; the large investments in host-range testing reduce the risk that an introduced agent will have unintended detrimental effects. Post-release costs of biocontrol include money spent on subsequent inoculations of the agent (if necessary), money spent to monitor the biological control programs, and costs to renew biological control permits. Both of these costs are incurred over time. Because biocontrol programs require such high initial costs and incur potential benefits over long periods of time, it is important to conduct an economic assessment over an appropriate time horizon.

The high fixed costs will typically outweigh the benefits over the first years of a biocontrol program. Therefore, neglecting to evaluate the net benefits over time can inaccurately portray biocontrol as economically unjustifiable. Because these net benefits are considered over time, discounting must be used to find the present value of net benefits. If the present value of net benefits is positive, then biological control should be implemented. The time horizon for the analysis was 50 years from the initiation of research on potential biocontrol agents for the target invasive species in each case.

While not exhaustive, Figure 4 illustrates several hypothetical examples of biocontrol costs and benefits over a 50-year management period. As noted above, the typical case involves high up-front pre-release costs, followed by a flattening of cumulative costs in the post-release period (Figure 4a). Even if benefits exceed costs on an annual basis shortly after release, it may take decades for cumulative benefits to catch up with cumulative costs. However, after the breakeven point–around year 40–cumulative net benefits continue to increase indefinitely.

A more prolonged pre-release period (Figure 4b) typically results in higher cumulative pre-release costs, which may be due to challenges such as difficulty finding an appropriate agent, issues with quarantine and testing, or delays in permitting. However, this alone does not indicate an economically undesirable management option, as costs must be weighed against future benefits. If the target invasive species is currently causing significant damages and/or affected parties are incurring substantial costs to mitigate those damages without the biocontrol, then the post-release benefits (avoided damages) will accumulate rapidly and offset cumulative costs. As shown, the steep cumulative benefit curve in Figure 4b results in a breakeven point shortly after year 30, nearly a decade earlier than the typical case.

In contrast, even when the pre-release period is fairly short and the cumulative cost curve remains relatively flat (Figure 4c), the breakeven point may never be reached or may occur later than in the typical case. This outcome may arise due to difficulties in successfully establishing the biocontrol agent in the field or lower than expected efficacy in controlling the target invasive species.

Figure 4. Hypothetical cumulative benefits vs. cumulative costs of biocontrol over 50 years for (a) quick release and late breakeven point, (b) slow release and early breakeven point, and (c) moderate release and no breakeven point.

Data

Coffee Berry Borer (H. hampei)

Costs
Although pre-release costs for biocontrol agents typically include those related to the search for and collection of potential agents, this particular case is somewhat unique in that Hawai‘i obtained P. coffea specimens from Cenicafé, a national coffee research center in Colombia, for host-specificity testing. Hawai‘i will also obtain wasps from Cenicafé in larger numbers for future field releases (USDA APHIS, 2023). As such, the total costs for P. coffea may be low relative to those of potential biocontrol agents for other invasive species, where search and collection costs are typically incurred by the state. Although costs for search and collection are not available, the analysis includes the estimated $20,000 to $30,000 annual cost of maintaining the P. coffea colony at Cenicafé since 1996 (M. White, personal communication). Taking the midpoint ($25,000) and adjusting the value to 2023 dollars results in an annual cost of almost $48,000. The cost of non-target host screening, which includes facility/lab space and personnel, has been estimated at $750,000 (Hawaii News Now, 2023).

The final component of pre-release costs is research grants, which are instrumental in identifying a suitable biocontrol agent, in this case, P. coffea. Over a dozen grants related to CBB research, totaling more than $4 million, have been recorded in the University of Hawai‘i Office of Research Services system since 2014. However, only $987,000 (M. White, personal communication), or roughly 24% of the total, has been dedicated specifically to P. coffea, with the remainder going toward B. bassiana research, CBB survey and monitoring, and the development of a systems approach to CBB management. After adjusting to 2023 dollars, $1.15 million in pre-release research costs are included in our analysis.

Since P. coffea has not been released yet, estimating post-release costs is challenging. Of the $4 million in University of Hawai‘i research grants related to CBB, roughly $44,000 has been spent on developing post-release management plans. In addition, subsequent releases may be necessary if P. coffea fails to establish after the initial release. Because we do not have a clear understanding of the likelihood of successful establishment, subsequent release costs are not included in our analysis. However, post-release monitoring will be required, regardless of the likelihood of establishment, in order to better understand the effectiveness of P. coffea at controlling CBB. This will inform future management decisions, including planning for subsequent releases if necessary. We assume that $10,000 (in 2023 dollars) will be spent annually on monitoring, beginning in the expected year of release (2025) and continuing through the end of the planning horizon (2046).

Benefits
Ideally, the benefits of a biocontrol agent like P. coffea would account for both the avoided (status quo) management costs associated with CBB, as well as the reduced direct impacts on coffee production and profitability due to improved CBB control. The latter is particularly challenging in this case because it is difficult to isolate the impacts of CBB using historical data, given the many other factors that affect both the quantity and quality of coffee produced in Hawai‘i. For example, Woodill et al. (2019) observed declines in statewide coffee production and yield from 2007 to 2012 but pointed out that while CBB is likely a contributing factor, the declines had already begun prior to CBB’s detection in 2010. Furthermore, changes in regional rain patterns and drought, ongoing volcanic events on Hawai‘i Island, and other pests like coffee root-knot nematode and coffee leaf rust likely contributed to the downward trend. Additionally, price increases have coincided with declines in production over the same period, partly due to the CBB-induced decrease in supply but also likely driven by other factors, such as increased product differentiation and the growth of internet sales. Given these challenges, we focus on avoided management costs in this analysis.

Because our benefit calculation is limited to avoided management costs, it should be viewed as a conservative estimate. On the other end of the spectrum, the oft-cited study by Leung et al. (2014) used the State of Hawai‘i Input-Output Economic Model to estimate the economy-wide impact of CBB, under the assumption that the reduction in coffee production during the 2011/12 and 2012/13 growing seasons, relative to the 2010/11 season, was due entirely to CBB. Scaling their estimated $12.7 million decrease in crop value, $25.7 million in lost sales, $7.6 million reduction in household earnings, and 380 lost jobs to our 50-year management timeline and calculating the present value could be considered an upper bound for the benefits of P. coffea. However, due to the reasons discussed above, observed decreases in coffee production and yield during that time period cannot be entirely attributed to CBB. Thus, the true benefits of the biocontrol, including both the avoided management costs and the reduced impacts on coffee production and profits, likely lie somewhere in between.

We focus on the costs of two current CBB sanitation practices that would be at least partially avoidable upon the successful release and establishment of P. coffea: (1) spraying of B. bassiana throughout the growing season and (2) strip-picking at the end of each season. As previously mentioned, a B. bassiana subsidy has been available for coffee farmers since 2014. One way to estimate the annual expenditures on B. bassiana would be to tally up the subsidy payments. However, this approach is challenging given that the subsidy rules have changed over time. From 2014 to 2016, up to 75% of the costs were reimbursed with a $9,000 cap per farmer per year, whereas most recently (2018 to present), 50% of the costs are reimbursable with a $6,000 cap. To complicate matters, Aristizábal et al. (2023a) notes that in recent years, only a small percentage of the total farms in the state have taken advantage of the program. Therefore, rather than attempting to estimate total expenditures on B. bassiana and its application from the subsidy program, we instead extrapolate these management costs going forward using data on harvested coffee acres, the share of farms spraying in recent years, the average spray cost per acre, and the average number of sprays per season.

USDA survey data (USDA NASS, 2024b) suggest that the number of acres bearing coffee in the state has been fairly stable over the past decade. For our analysis, we assume that 7,320 acres, the average over the past five years, will be harvested annually going forward. The reported share of growers spraying B. bassiana ranges from 80% to 95% (Johnson et al., 2020; Lee et al., 2023). We conservatively assume that 85% of farmers will continue to apply B. bassiana in the absence of an effective CBB biocontrol agent. The reported cost of spraying ranges from $106 to $240 per acre (Aristizábal et al., 2017; Hollingsworth et al., 2020; Woodill et al., 2021; Aristizábal et al., 2023a), or $193 on average in 2023 dollars. Lastly, the reported number of sprays per season has been quite variable, both geographically and over time, ranging from 1 to 12 (Aristizábal et al., 2017; Hollingsworth et al., 2020; Johnson et al., 2020; Woodill et al., 2021; Aristizábal et al., 2023a,b; Johnson and Manoukis, 2024). In recent years, however, the range appears to have converged to around 4-7 sprays per year, likely due at least in part to the latest research-based IPM recommendations for optimal spray timing. Our calculations are based on the assumption of five sprays per season on average. To summarize, the avoided future costs of spraying are estimated assuming 5 annual sprays on 85% of 7,320 harvested coffee acres, at a cost of $193 per acre per year.

The second avoided management cost, strip-picking (collecting and properly disposing of all remaining berries after the regular harvest to reduce CBB infestation in the following season), is estimated using information on the number of harvested coffee acres, the share of farms strip-picking at the end of each season, the amount of labor required per acre for strip-picking, and the hourly wage for farm labor. Woodill et al. (2019) and Johnson et al. (2020) reported that 80% of farms in 2017 and 82% of farms in 2018, respectively, strip-picked at the end of each season. We assume that this practice will continue on 80% of harvested acres in the absence of a biocontrol for CBB. Strip-picking is a fairly labor-intensive task. Woodill et al. (2017) reported a range of 15 to 25 hours of labor required per acre. Thus, our analysis assumes a labor requirement of 20 hours per acre for strip-picking. Finally, reported hourly wages for off-peak farm labor vary from $10.50 to $15.00 (Aristizábal et al., 2016, 2017, 2023b; Woodill et al., 2017, 2021). Given that wages have remained fairly stable over the past decade and that the minimum wage in Hawai‘i has been $14 per hour since January 2024, the analysis assumes a wage of $15 per hour in 2023 dollars. In summary, the avoided future costs of strip-picking are calculated assuming that the sanitation activity occurs on 80% of the 7,320 harvested acres, with labor inputs of 20 hours per acre, at a cost of $15 per hour.

Erythrina Gall Wasp (Quadrastichus erythrinae)

Costs
When EGW was first detected in Hawai‘i, Hawai‘i researchers began exploration for biocontrol agents by travelling to the pest’s native range in Africa. After collecting potential biocontrol agents, different options were extensively tested and characterized in quarantine facilities and research labs on island. A number of research grants have been awarded to UH Mānoa professors for their work on testing the effectiveness and host-specificity of the two EGW BCAs. Initial grants helped establish protocols and methodologies for ongoing evaluation. The consistent financial support from the Department of Land and Natural Resources (DLNR) has been pivotal in ensuring the long-term success and sustainability of biological control efforts against EGW in Hawai‘i.

Exploration costs, including search and collection of potential biocontrol agents, are estimated at less than $120,000, while the overall project cost for EGW biocontrol, encompassing exploration, cataloging, rearing, and evaluation of both BCAs, is around $700,000 (Grossman, 2023). Subsequent release and monitoring costs are covered by HDOA and TSTAR grants (U.S. Forest Service, 2014). We estimate possible post-release monitoring costs for A. nitens in line with those of E. erythrinae.

Benefits
The benefits of introducing a biocontrol for EGW include avoided future damages and costs. Other benefits include the establishment of Waikōloa Dry Forest Initiative and other dry forest areas and preserves, and the avoided cost of replacing existing Erythrina species. Note that EGW is a broad pest of Erythrina species, and while our report focuses mainly on wiliwili due its additional conservation and cultural importance, other Erythrina spp. trees have been killed and needed to be safely removed, with high replacement costs (Vorsino, 2006; Vargas et al., 2007). Additionally, we consider the avoided costs of listing wiliwili as federally endangered, which would likely have been necessary in the absence of the first BCA.

The Waikōloa Dry Forest Initiative (WDFI), established in 2011, protects 270 acres of lowland dry forest on Hawai‘i Island, and represents one of the best remaining populations of wiliwili in the state. Through community advocacy and a partnership with the Waikōloa Village Outdoor Circle, the Waikōloa Dry Forest Preserve was established to protect the area’s remaining native trees, which also include the endangered uhiuhi (Mezoneuron kavaiense). Restoration efforts, educational programs for keiki, and ongoing conservation initiatives by WDFI have helped ensure the future of Waikōloa’s lowland dry forest (R. Yagi, personal communication). Each year approximately 500 volunteers provide 1,500 – 2,000 hours of service, and approximately 2,000 visitors visit for an average of 2.5 hours (R. Yagi, personal communication). WDFI also hosts an annual wiliwili festival which averages approximately 1,000 visitors in recent years (R. Yagi, personal communication). The preserve’s establishment and the organization’s conservation work serves as a proxy for an extreme lower bound on how wiliwili may be valued economically.

An analysis of Honolulu County’s urban forest (Vargas et al., 2007) found 606 Erythrina spp. trees, with 166 of those being wiliwili and representing 0.38% percent of the total population of urban trees, corresponding to 14% of the native population. The total replacement value of those wiliwili trees was estimated to be $520,000 (2023$). Removal of more than 1,000 Erythrina trees from Honolulu County’s medians and parks cost $690,000 (2023$) (Vorsino, 2006).

The financial aspect of these control methods includes the cost of insecticides like Imicide, which amounts to $116 for 24 capsules, as reported by Hara et al. (2007), and the significant expenditure by the City and County of Honolulu in 2006, allocating $440,000 for the removal of 1,000 dead Erythrina trees affected by EGW, as cited in the Honolulu Star Advertiser (2013) and Vorsino (2006). Hui Kū Maoli Ola purchases their insecticides annually in bulk and sells wiliwili trees in pots of different sizes, with a 2-gallon pot containing a 2-ft tree priced at $40 (Hui Kū Maoli Ola, personal communication, 2023). As the full extent of avoided costs from the biological control of EGW is unclear with respect to how it alters the use of other control methods compared to the counterfactual, we do not include them in our benefit analysis.

Economic valuation, however, provides a limited assessment of wiliwili’s true significance in Hawai‘i. While quantitative metrics such as tree replacement, nonprofit revenue, and other expenditures offer measurable data points, they fail to capture wiliwili’s cultural and ecological importance. Wiliwili occupies a central position in Hawaiian cultural heritage. The tree appears in the Kumulipo, the Hawaiian cosmogonic chant, establishing it as a fundamental element in Native Hawaiian worldview. Traditional accounts link wiliwili to significant mythological narratives, demonstrating its deep integration within Hawaiian cultural identity.

Historical documentation reveals wiliwili’s extensive utilitarian applications in Native Hawaiian material culture. The wood’s buoyancy and lightweight properties made it ideal for constructing ama (outrigger floats), ‘iako (outrigger booms), and papa he‘e nalu (surfboards). Additional uses included water troughs, fishing implements, and ceremonial objects. The seeds, flowers, and bark were utilized for adornment and medicinal purposes, illustrating cultural importance that goes beyond contemporary economic frameworks (HDOA, 2023). The ‘ōlelo no‘eau ‘Pua ka wiliwili, nanahu ka manō’ represents indigenous ecological knowledge correlating wiliwili flowering patterns with increased shark activity (HDOA, 2023).

As a keystone species in dryland forest ecosystems, which contain 25% of all endangered Hawaiian flora, wiliwili provides critical habitat for native flora and fauna (HDOA, 2023). With only 1-5% of Hawaiian dryland forests remaining, the ecological services provided by wiliwili extend far beyond individual tree value (HDOA, 2023). The loss of this species would likely trigger cascading effects throughout these vulnerable ecosystems. The Cultural Impact Assessment from the A. nitens final Environmental Assessment (HDOA, 2023) documents ongoing relationships between contemporary practitioners and wiliwili. The continued harvesting of wiliwili for traditional crafts and the species’ spiritual significance in Hawaiian cosmology represent persistent cultural values.

The decline of wiliwili populations due to the EGW has demonstrably affected cultural practices (HDOA, 2023). Resource scarcity forces adaptation through substitute materials, potentially accelerating cultural knowledge erosion and representing an intangible cultural cost not captured in economic analyses. Wiliwili’s role in defining Hawaiian landscapes contributes to place-based cultural identity. The loss of native plants like it diminishes the ecological characteristics that distinguish Hawaiian environments, affecting aesthetic, spiritual, and historical dimensions of place.

Standard economic assessments focus primarily on direct-use values while overlooking non-use values including existence value, bequest value, and option value. For communities with historical connections to Hawaiian ecosystems, these non-use values constitute the most significant portion of wiliwili’s perceived worth (HDOA, 2023). The proposed biological control intervention represents not merely a cost-effective management strategy but an investment in preserving a culturally significant species that maintains connections between Hawaiian communities and their ancestral knowledge systems (HDOA, 2023). This cultural preservation dimension extends beyond conventional economic valuation methodologies, representing an investment in cultural continuity and indigenous knowledge with intergenerational benefits that resist quantification.

Lastly, one of the largest, unquantifiable benefits from the introduction of a biocontrol agent for EGW was a significant shift in the court of public opinion, and those of conservationists, by its demonstration of how classical biological control can act as a highly effective management tool for preservation of native species and forests (Ramadan et al., 2023).

Fireweed (Senecio madagascariensis)

Costs
Pre-release costs for fireweed included the costs incurred for the collection, rearing, and investigation of the Secusio moth, as well as the lab personnel and equipment to perform host specificity testing. The initiation of the exploratory period for fireweed was in 1999, and continued through state approval in 2010 and the finalization of the release permits in 2012. We use the HDOA Exploratory Entomologist salary plus an additional $50,000 for surveys, travel especially to South Africa and Madagascar, and other expenses for a total of $136,712 spent every year from 1999 until release in 2013 (all costs and benefits reported in 2023 dollars). In 2008 Mark Thorne received a $40,000 pre-release research grant and from 2009-2011 there was an additional $69,000 spent per year from Paul Krushelnycky’s research project evaluating the efficacy and nontarget impacts of the biocontrol agent release. In 2011 Mark Thorne’s USDA research grant provided an additional $258,449 in pre-release testing. Based on available data, total spending for the pre-release category from 1999–2013 is estimated at $2.05 million. For comparison, in Australia where fireweed is also a problem, approximately $2 million is spent annually on control measures (HDOA, 2010).

We have less information about the actual release and post-release costs. We assume $50,000 were spent annually for three years 2013-2016. Initially, HDOA struggled with the scale of rearing Secusio from eggs to caterpillars due to limited facilities and personnel. In response, UH CTAHR Cooperative Extension collaborated with HDOA, setting up rearing programs at two experimental stations on Maui and the Big Island. Following the initial round of funding, the program had to operate on minimal resources for several years after the formal project ended. The release of S. extensa on Maui proved to be more effective due to consolidated ranch areas and less geographic spread. On Hawai‘i Island the widespread nature of ranches and moths’ attraction to urban lights hampered S. extensa establishment. Despite these challenges, there are signs of a naturally sustaining population of S. extensa, indicating some level of success in biological control efforts. Post-release monitoring seems sporadic; we assume $10,000 in efforts beginning in 2013 through the end of the planning horizon (2049) to monitor the population of S. extensa.

Benefits
Our analysis calculates benefits of introducing S. extensa as both the management costs avoided following the 2013 release, as well as avoided potential damages to the cattle industry from the continued widespread presence of fireweed.

Fireweed is currently present on at least 400,000 acres across O‘ahu, Maui, Moloka‘i and Hawai‘i Island (HDOA, 2010). Chemical treatment of fireweed across 5,000 acres in 2011 was approximately $150,000 (Brooks, 2013). While treating the entire extent of fireweed is not feasible, this analysis considers a range of potential treatment areas for future years, 5,000 acres annually, or twice this to represent a potential higher annual treatment scenario. Adjusting to 2023 dollars, chemical treatment of fireweed is $41.15/acre, or $200,739 to $401,478 per year.

An important consideration in estimating the benefits of biological control is determining the extent to which the biocontrol agent reduces damages to the economically important resource of interest, in this case, cattle primarily on Maui and Hawai‘i Island. The most recent agricultural baseline study of Hawai‘i reports 765,579 acres of pastureland across the state (Perroy & Collier, 2020), therefore fireweed is likely present on almost half of these lands. The total value of cattle production in Hawai‘i was estimated at $42.5 million (USDA NASS, 2024a). While it is difficult to estimate a precise value for cattle across these pasturelands given the variety of grazing management practices used across ranches and properties, for transparency and in order to show potential avoided damages from the introduction of a successful biocontrol we assume these values are evenly distributed across all pasturelands, and therefore estimate per acre cattle value at $57.53 in 2023 dollars.

Fireweed has a deep taproot and competes for moisture and minerals with grass. When prevalent, a pasture loses between 15-40% of forage production (Thorne et al., 2005; M. Thorne, personal communication). From our conversations with ranchers and extension agents we learned that cattle actually will not willingly eat fireweed; the bigger issue is that there is less forage so the herd size tends to decline over time as a result.

Results

Coffee Berry Borer (H. hampei)

Assuming a discount rate of two percent, the present value (PV) of all biocontrol pre- and post-release costs through the year 2046 is estimated at $3.5 million. In the most optimistic scenario (baseline case), the PV benefit, which includes the full avoided costs of both B. bassiana spraying and strip-picking, totals $145.0 million, resulting in a net present value (NPV) of $141.6 million and a benefit-cost ratio (BCR) of 42.0. More realistically, the biocontrol agent will not completely eliminate the need for other conventional CBB management activities but will reduce their frequency, meaning that the PV benefit will likely be a fraction of what is estimated for the optimistic scenario. Under the assumption that only 50% of the baseline PV benefits are realized, which will occur if farmers spray and strip-pick only half as often as they currently do once P. coffea is successfully released, the NPV and BCR fall to $69.1 million and 21.0, respectively. In the case that only 25% of the baseline benefits are achieved, the NPV is estimated at $32.8 million, and the BCR at 10.5. Results for the three scenarios are summarized in Table 1.

Table 1. Present value cost, present value benefit, net present value, and benefit-cost ratio for the CBB biocontrol agent P. coffea

	Baseline	50% Benefits	25% Benefits
Discount rate	0.02	0.02	0.02
Present value cost	$3,455,519	$3,455,519	$3,455,519
Present value benefit	$145,015,606	$72,507,803	$36,253,902
Net present value	$141,560,087	$69,052,284	$32,798,383
Benefit-cost ratio	42.0	21.0	10.5

In the case of P. coffea, the breakeven point–where cumulative biocontrol benefits catch up with cumulative costs–is projected to occur within two years of release, even in the most conservative scenario (Figure 5). This is due to a combination of factors affecting both benefits and costs. Although the colony rearing and maintenance costs borne by Cenicafé are included in the calculation, the aggregate cost is still underestimated due to a lack of information on other pre-release costs, such as search and exploration, as well as uncertainty regarding the resources required to locally rear and expand the P. coffea colony. The cumulative cost estimate also does not account for potential re-releases if the biocontrol agent fails to establish a self-sustaining colony in the field. At the same time, the benefit (avoided management cost) curves are relatively steep, as coffee farmers are already investing significant resources in controlling CBB through other methods, given the high value of Hawaiian coffee.

Erythrina Gall Wasp (Quadrastichus erythrinae)

With a discount rate of two percent, the net present value of the two BCAs for EGW is $26.0 million, with present value costs and benefits of $1.1 million and $27.2 million, respectively. Most of the costs of the biocontrols have already been incurred but some, including those for continued rearing, release, and post-release monitoring, are projected as A. nitens has not yet been released.

The cost of listing a species as federally endangered was $790,000 in 2016 (FWS, 2016). As the release of E. erythrinae has successfully prevented wiliwili from rapidly nearing extinction, the avoided cost of listing the species is a benefit, which we assume is spread over 13 years, the median timeline to list an endangered flowering plant species (Puckett et al., 2016). A majority of the present value benefit comes from WDFI, which accounts for $3.9 million as of 2024, and we extrapolate its revenue over the coming decades. We assign all of its past and projected revenue as a benefit of the EGW BCA. Avoided costs of insecticide use is not quantified along with replacement costs of Erythina spp. trees in urban Honolulu.

Table 2. Present value cost, present value benefit, net present value, and benefit-cost ratio for the EGW biocontrol agents E. erythrinae and A. nitens.

	Baseline
Discount rate	0.02
Present value cost	$1,100,000
Present value benefit	$27,200,000
Net present value	$26,000,000
Benefit-cost ratio	24.0

Fireweed (Senecio madagascariensis)

Assuming a discount rate of two percent, the present value (PV) of all S. extensa pre- and post-release costs through the year 2049 is estimated at $2.5 million.

We consider two possibilities in terms of potential benefits from the introduction of S. extensa. First, including only the avoided management costs associated with controlling fireweed pre-S. extensa introduction. As discussed in the previous section, we consider a range of treatment acres, either 5,000 acres as treated on Maui in 2011, or double that for a higher range (10,000 acres). Therefore the total benefit from avoided management costs ranges from $5.2 – $10.4 million, resulting in a NPV of $2.6 – $7.8 million and a benefit-cost ratio (BCR) of 2.0 – 4.0.

In the second possibility, rather than assuming the benefit of S. extensa comes from avoiding management of fireweed, we assume that the introduction of the biological control has prevented losses in the cattle industry, according to the calculations described in the Data section above. With a per acre value of $57.35 (in 2023 dollars), a 15-40% reduction in forage acres results in 60,000-160,000 less available acres for grazing, which would result in corresponding losses in cattle value of $344,110–$917,626 annually, even if only 10% of the value was affected by fireweed. If S. extensa is helping offset these losses via protecting those forage acres from invasion, the total benefit of avoided cattle value losses are between $8.9 – $23.8 million, resulting in a NPV of $6.3 – $21.2 million and a benefit-cost ratio (BCR) of between 3.4 – 9.1.

Results for these scenarios (costs are the same across all, but we present four independent benefit possibilities) are summarized in Table 3. Depending on how quickly benefits accumulate, the breakeven year may happen 10-15 years after the biocontrol agent is introduced (Figure 7, Low Avoided Management Costs scenario), or relatively sooner if benefits accumulate faster due to higher levels of damages avoided (Figure 8, High Avoided Cattle Damage scenario).

Table 3. Present value cost, present value benefit, net present value, and benefit-cost ratio for the Fireweed biocontrol agent S. extensa

	Avoided Management Costs (low)	Avoided Management Costs (high)	Avoided Cattle Losses (low)	Avoided Cattle Losses (high)
Discount rate	0.02	0.02	0.02	0.02
Present value cost	$2,507,760	$2,507,760	$2,507,760	$2,507,760
Present value benefit	$5,213,079	$10,426,159	$8,936,348	$23,830,261
Net present value	$2,613,650	$7,826,730	$6,336,918	$21,230,831
Benefit-cost ratio	2.0	4.0	3.4	9.2

Discussion

The benefit-cost analyses for the three biological control programs–targeting CBB, EGW, and fireweed–demonstrate a range of economic outcomes, with notable variation in BCRs, breakeven timelines, and levels of uncertainty. Among these, the programs for CBB and EGW stand out as highly cost-effective, with strong economic justification even under fairly conservative assumptions.

The EGW program exhibits particularly strong performance, with a BCR of approximately 24 and a rapid breakeven point, due to the low cost of implementation and the substantial long-term benefits, including avoided expenses related to endangered species listing and sustained revenue from the Waikōloa Dry Forest Initiative. This case illustrates how early investment in biocontrol can yield both ecological and economic returns, especially when tied to high-conservation-value species.

The CBB biocontrol program, while not yet fully implemented, also shows high economic potential. Depending on the eventual efficacy of the released agent (P. coffea) and how much it reduces the need for labor-intensive control measures like spraying and strip-picking, the BCR ranges from 10 to 42. Importantly, even in the most conservative scenario–assuming only 25% of baseline benefits are realized–the breakeven point is projected to occur within two years of release. However, this rapid payoff is contingent upon successful establishment of the biocontrol agent and its timely release. Delays in permitting, rearing challenges, or failures in population establishment could extend the time to breakeven and reduce net returns, underscoring the need for continued support and monitoring.

In contrast, the biocontrol program targeting fireweed presents a more modest economic case. Depending on whether benefits are framed in terms of avoided management costs or protection of cattle grazing capacity, the BCR ranges from 2 to 9, with a breakeven timeline that may extend over 10 years post-release. While these figures still support a positive return on investment, they also reflect a higher degree of uncertainty. Some ranchers have expressed skepticism about the effectiveness of S. extensa, and in areas where the agent’s impacts have been limited, the economic payoffs may be slower and spatially variable. This highlights the importance of region-specific impact assessments and complementary management strategies.

Overall, these cases illustrate the varying dynamics of biocontrol investments in Hawai‘i. While all three programs are projected to deliver positive net economic benefits, the magnitude, speed, and certainty of those net benefits vary significantly (Table 4). This underscores the value of targeted analysis to inform decision-making and guide resource allocation across invasive species management efforts.

Table 4. Summary of present value costs, present value benefits, net present values, benefit-cost ratios, and breakeven points for the three case studies

	Coffee Berry Borer	Erythrina Gall Wasp	Fireweed
Present value cost (millions)	$3.5	$1.1	$2.5
Present value benefit (millions)	$36.3 – 145.0	$27.2	$5.2 – 23.8
Net present value (millions)	$32.8 – 141.6	$26.0	$2.6 – 21.2
Benefit-cost ratio	10.5 – 42.0	24.0	2.0 – 9.2
Breakeven point (years post-release)	1 – 2	2	2 – 13

Conclusion

The economic analyses of three biological control programs in Hawai‘i – targeting the coffee berry borer (CBB), the erythrina gall wasp (EGW), and fireweed – demonstrate consistent positive returns on investment, albeit with varying degrees of economic efficiency. All three programs generate substantial net benefits when evaluated over a 50-year time horizon, supporting the economic case for biological control as a cost-effective management strategy for certain invasive species.

The CBB and EGW programs show particularly strong economic performance, with benefit-cost ratios of 10.5 – 42.0 and 24.0 respectively, and break-even points within two years of the biocontrol agent’s release. These results highlight how targeted biological control can rapidly deliver economic returns, especially for high-value agricultural commodities like coffee and ecologically and culturally significant native species such as wiliwili. The fireweed biocontrol program demonstrates more modest but still positive economic outcomes, with benefit-cost ratios between 2.0 and 9.2 and a more variable breakeven timeline extending up to 13 years post-release.

These findings confirm our hypothesis that the return on investment for classical biological control in Hawai‘i is generally high, akin to observations in other locations. However, the variation in economic performance across programs underscores the importance of species-specific assessments and recognition of implementation challenges. The lack of comprehensive economic records for biological control efforts in Hawai‘i has historically complicated efforts to garner support for these programs. This research addresses that gap by providing empirical evidence of the substantial economic benefits that can be achieved through investment in biological control infrastructure and capacity.

As Hawai‘i faces growing invasive species threats, this economic assessment offers valuable guidance for resource allocation decisions. Future biological control initiatives should be evaluated not only for their potential economic returns but also for their broader ecological, cultural, and social benefits, particularly when targeting invasive species that threaten Hawai‘i’s unique native ecosystems and cultural resources.

Acknowledgements

The UHERO study team is grateful to the Hawai‘i Invasive Species Council for funding this project, and to Leyla Kaufman, Mohsen Ramadan, Mark Wright, and Abel Konan for helping craft the original study approach and choose the invasive species and biological controls of interest. We also want to thank Christy Martin, Mark Thorne, Britt Craven, Elliott Parsons, Robert Yagi, Jen Lawson, Luis F. Aristizábal, Melanie Bondera, Melissa Johnson, and Richard Stevens for hosting field visits and conducting interviews with our study team. The UHERO team also thanks Armando Martinez and Isabela Daszewska for research assistance.

References

Aristizábal, L. F., Johnson, M., Shriner, S., Hollingsworth, R., Manoukis, N. C., Myers, R., Bayman, P., & Arthurs, S. P. 2017. Integrated Pest Management of Coffee Berry Borer in Hawai‘i and Puerto Rico: Current Status and Prospects. Insects, 8(4), 123. https://doi.org/10.3390/insects8040123

Aristizábal, L. F., Johnson, M. A., Mariño, Y. A., Bayman, P., & Wright, M. G. 2023a. Establishing an Integrated Pest Management Program for Coffee Berry Borer (Hypothenemus hampei) in Hawai‘i and Puerto Rico Coffee Agroecosystems: Achievements and Challenges. Insects, 14(7), 603. https://doi.org/10.3390/insects14070603

Aristizábal, L. F., Johnson, M. A., Shriner, S., & Wall, M. 2023b. Frequent and efficient harvesting as an economically viable strategy to regulate coffee berry borer on commercial farms in Hawai‘i. Journal of Economic Entomology, 116(2), 513-519. https://doi.org/10.1093/jee/toad041

Burbano, E., Wright, M., Bright, D. E., & Vega, F. E. 2011. New record for the coffee berry borer, Hypothenemus hampei, in Hawai‘i. Journal of Insect Science, 11(1), 117. https://doi.org/10.1673/031.011.11701

Brooks, L. 2013. Fireweed (Senecio madagasceriensis) Biological Control Release Program in Maui County. University of Hawai‘i Master Gardener Program, https://www.ctahr.hawaii.edu/uhmg/news/V3-Maui-fireweed-biocontrol.pdf

Department of Agriculture and Fisheries, The State of Queensland. 2024. Fireweed (Senecio madagascariensis) factsheet. https://www.publications.qld.gov.au/ckan-publications-attachments-prod/resources/c8619c36-254f-4165-a19e-40fcfdde0933/fireweed.pdf?ETag=6558ecb51b64b1568a50c210f3c14116

Funasaki, G.Y., Lai, P., Nakahara, L.M., Beardsley, J.W., & Ota, A. 1988. A review of biological control introductions in Hawaii: 1890 to 1988. Proc. Hawai‘i. Entomol. Soc. 28, 105–160.

Gagne, W.C., & Howarth, F.G. 1985. Conservation status of endemic Hawaiian Lepidoptera. In: Proceedings 3rd Congress on European Lepidoptera, Cambridge, 982. Societas Euro-paea Lepidopterologica, Karlsruhe, Germany, pp. 74–84.

Grossman, D. 2023. State one step closer to using new wasp to save wiliwili trees. The Maui News.

Hawai‘i Department of Agriculture. 2010. Final Environmental Assessment: Field Release of Secusio extensa (Butler) (Lepidoptera: Arctiidae), for Biological Control of Fireweed, Senecio madagascariensis Poiret (Asterales: Asteraceae), in Hawai‘i.

Hawai‘i Department of Agriculture. 2013. Biocontrol moths released on Maui to fight invasive fireweed. Biocontrol Moths Released on Maui to Fight Invasive Fireweed. https://hdoa.hawaii.gov/blog/news-releases/2013-news-releases/biocontrol-moths-released-on-maui-to-fight-invasive-fireweed/

Hawai‘i Department of Agriculture. 2014. Coffee pest detected on Oahu coffee farm; http://hdoa.hawaii.gov/blog/main/nrcbboahu/.

Hawai‘i Department of Agriculture. 2016. Coffee berry borer confirmed on Maui; http://hdoa.hawaii.gov/blog/main/nr17-1-cbbonmaui/.

Hawai‘i Department of Agriculture. 2020a. Coffee berry borer confirmed on Lanai; https://hdoa.hawaii.gov/blog/main/nr20-14cbblanai/.

Hawai‘i Department of Agriculture. 2020b. Coffee berry borer confirmed on Kauai; https://hdoa.hawaii.gov/blog/main/nr20-13cbb-kauai/.

Hawai‘i Department of Agriculture. 2023. Final Environmental Assessment and Finding of No Significant Impact (FONSI) for the Proposed Statewide Release of Aprostocetus Nitens Prinsloo & Kelly (Hymenoptera: Eulophidae: Tetrastichinae) For Biological Control of the Erythrina Gall Wasp, Quadrastichus Erythinae Kim (Hymenoptera: Eulophidae).

Hawai‘i News Now. 2023. Tiny wasps released to save Hawaii coffee farms from invasive beetle. Hawaii News Now. https://www.hawaiinewsnow.com/2023/05/25/tiny-wasps-released-save-hawaii-coffee-farms-invasive-beetle/

Hollingsworth, R. G., Aristizábal, L. F., Shriner, S., Mascarin, G. M., Moral, R. A., & Arthurs, S. P. 2020. Incorporating Beauveria bassiana Into an Integrated Pest Management Plan for Coffee Berry Borer in Hawaii. Frontiers in Sustainable Food Systems, 4, 22. https://doi.org/10.3389/fsufs.2020.00022

Howarth, F.G. 1983. Classical biological control: panacea or Pandora’s Box? Proc Hawai‘i Ent Soc 24, 239–244.

Javar-Salas, C. M., Pe‘a, R., Amidon, F., Reeves, M. K., & Miller, S. E. 2020. Hawaiian Islands Dry Forest. In Encyclopedia of the World’s Biomes (pp. 295–327). Elsevier. https://doi.org/10.1016/B978-0-12-409548-9.11890-1

Johnson, M. A., & Manoukis, N. C. 2020. Abundance of Coffee berry borer in feral, abandoned and managed coffee on Hawai‘i Island. Journal of Applied Entomology 144(10), 920-928. https://doi.org/10.1111/jen.12804

Johnson, M. A., Ruiz-Diaz, C. P., Manoukis, N. C., & Verle Rodrigues, J. C. 2020. Coffee Berry Borer (Hypothenemus hampei), a Global Pest of Coffee: Perspectives from Historical and Recent Invasions, and Future Priorities. Insects, 11(12), 882. https://doi.org/10.3390/insects11120882

Kinro, G.Y. 2003. A Cup of Aloha: The Kona Coffee Epic. University of Hawaii Press. https://doi.org/10.1515/9780824840839

Lee, D., Johnson, M. A., Aristizábal, L. F., Shriner, S., Chan, C., Miyasaka, S., & Wall, M. 2023. Economic Benefits from Managing Coffee Berry Borer (Hypothenemus hampei) in Hawai‘i. Insects, 14(4), 350. https://doi.org/10.3390/insects14040350

Le Pelley, & R. H. 1968. Pests of coffee. Longmans.

Leung, P. S., Kawabata, A. M., & Nakamoto, S. T. 2014. Estimated Economywide Impact of CBB for the Crop Years 2011/12 and 2012/13. Brief Report at Request of Hawaii Congressional Delegation. https://www.hawaiicoffeeed.com/uploads/2/6/7/7/26772370/estimated_economic_impact_of_cbb_in_hawaii.pdf

Messing, R., & Brodeur, J. 2018. Current challenges to the implementation of classical biological control. BioControl 63, 1–9.

Motooka, P., Hagai, G., DuPonte, M., Kawabata, A., & Fukumoto, G., 1999. Control of Fireweed (Senecio madagascariensis). CTAHR Publication on Weed Control WC-2. Univ. of Hawaii Press, Honolulu.

Oishi, D. E. 2023. Aprostocetus nitens / Field Release. Department of Agriculture Advisory Committee on Plants and Animals. https://hdoa.hawaii.gov/wp-content/uploads/2018/05/Aprostocetus-nitens-Submittal-PA-Final.pdf

Pau, S., Gillespie, T. W., & Price, J. P. 2009. Natural history, biogeography, and endangerment of Hawaiian dry forest trees. Biodiversity and Conservation, 18(12), 3167–3182. https://doi.org/10.1007/s10531-009-9635-1

Perroy, R. & Collier, E. 2020. 2020 Update to the Hawai‘i Statewide Agricultural Land Use Baseline. https://hdoa.hawaii.gov/wp-content/uploads/2022/01/2020_Update_Ag_Baseline_all_Hawaiian_Islands_v5.pdf

Puckett, E. E., Kesler, D. C., & Greenwald, D. N. 2016. Taxa, petitioning agency, and lawsuits affect time spent awaiting listing under the US Endangered Species Act. Biological Conservation, 201, 220–229. https://doi.org/10.1016/j.biocon.2016.07.005

Ramadan, M. M., Murai, K.T. & Johnson T. 2010. Host range of Secusio extensa (Lepidoptera: Arctiidae), and potential for biological control of Senecio madagascariensis (Asteraceae). Journal of Applied Entomology, doi: 10.1111/j.1439-0418.2010.01536.x https://www.fs.usda.gov/psw/publications/johnson_mt/psw_2010_johnson_mt(ramadan)001.pdf

Ramadan, M. M., Kaufman, L. V., & Wright, M. G. 2023. Insect and weed biological control in Hawaii: Recent case studies and trends. Biological Control, 179, 105170. https://doi.org/10.1016/j.biocontrol.2023.105170

Sakai, A. K., Wagner, W. L., & Mehrhoff, L. A. 2002. Patterns of Endangerment in the Hawaiian Flora. Systematic Biology, 51(2), 276–302. https://doi.org/10.1080/10635150252899770

Thorne, M.S., Powley, J.S., & Fukumoto, G.K. 2005. Fireweed Control: An Adaptive Management Approach. https://www.ctahr.hawaii.edu/oc/freepubs/pdf/prm-1.pdf

USDA Animal and Plant Health Inspection Service. 2023. Field Release of Phymastichus coffea (Hymenoptera: Eulophidae) for the Biological Control of Coffee Berry Borer, Hypothenemus hampei (Coleoptera: Scolytinae), in Hawai‘i. Final Environmental Assessment.

USDA National Agricultural Statistics Service. 2024a. 2022 Census of Agriculture: Hawaii State and County Data. Geographic Area Series 1(11). https://www.nass.usda.gov/Publications/AgCensus/2022/Full_Report/Volume_1,_Chapter_1_State_Level/Hawaii/hiv1.pdf

USDA National Agricultural Statistics Service. 2024b. Hawai‘i Coffee, 2023-24 Season. https://www.nass.usda.gov/Statistics_by_State/Hawaii/Publications/Fruits_and_Nuts/cafean24.pdf

U.S. Forest Service. 2014. Forest Health 2013 Highlights, Hawaii. https://www.fs.usda.gov/foresthealth/docs/fhh/HI_FHH_2013.pdf

Vargas, K. E., McPherson, E. G., Simpson, J. R., Peper, P. J., Gardner, S. L., & Xiao, Q. 2007. City of Honolulu, Hawai‘i Municipal Forest Resource Analysis. Center for Urban Forest Research USDA Forest Service, Pacific Southwest Research Station.

Vega, F. E., Rosenquist, E., & Collins, W. 2003. Global project needed to tackle coffee crisis. Nature, 435, 343. https://doi.org/10.1038/425343a

Vorsino, M. 2006. City will topple 1,000 trees. Honolulu Advertiser.

Woodill, A. J., Nakamoto, S. T., Kawabata, A. M., & Leung, P. 2017. To Spray or Not to Spray: A Decision Analysis of Coffee Berry Borer in Hawaii. Insects, 8(4), 116. https://doi.org/10.3390/insects8040116

Woodill, A. J., Nakamoto, S. T., Kawabata, A. M., Arita, S., & Leung, P. S. 2019. The Impact of CBB on the Economics of Coffee Production in Hawai’i: 2007-2012 USDA Census Analysis. University of Hawaii at Mānoa College of Tropical Agriculture and Human Resources, Insect Pests IP-46. https://www.ctahr.hawaii.edu/oc/freepubs/pdf/IP-46.pdf

Woodill, A. J., Nakamoto, S. T., Kawabata, A. M., & Leung, P. S. 2021. Optimal spraying strategy to combat the coffee berry borer: A dynamic approach. Journal of Agriculture and Food Research, 4, 100125. https://doi.org/10.1016/j.jafr.2021.100125

Xu, T., Jacobsen, C. M., Hara, A. H., Li, J., & Li, Q. X. 2009. Efficacy of systemic insecticides on the gall wasp Quadrastichus erythrinae in wiliwili trees (Erythrina spp.). Pest Management Science, 65(2), 163–169. https://doi.org/10.1002/ps.1663

Yalemar, J., Bautista, R., & Nagamine, W. 2016. Host Specificity and Biological Studies for Aprostocetus nitens (Hymenoptera: Eulophidae), Another Potential Biological Control Agent of the Erythrina Gall Wasp, Quadrastichus erythrinae Kim (Hymenoptera: Eulophidae). Unpublished report.

Yousuf, F., Follett, P. A., Gillett, C. P. D. T., Honsberger, D., Chamorro, L., Johnson, M. T., Giraldo-Jaramillo, M., Benavides-Machado, P., & Wright, M. G. 2021. Limited host range in the idiobiont parasitoid Phymastichus coffea, a prospective biological control agent of the coffee pest Hypothenemus hampei in Hawaii. Journal of Pest Science 94(4), 1183–1195. https://doi.org/10.1007/s10340-021-01353-8