Hide table of contents

Epistemic Status - tentative.  We are confident in our statement of the problem, but accurately evaluating prospective solutions is challenging.

Importance, tractability, neglectedness: Around 768 million people worldwide are undernourished, and this is both directly bad, and has many negative flow-on effects to health, employment, and political unrest.  Humanity knows how to produce ample food to feed everyone, so ending hunger is scientifically and technologically very tractable, with the main obstacles being political, especially wars.  Globally $47 billion was spent on agricultural research in 2016 (of which a small proportion would be on pests and diseases work), so agriculture is significantly less neglected than most EA projects, however there remain under-resourced gaps in current work.  Specifically, much of this funding is corporate and so not especially value-driven, and most money is spent on existing issues rather than addressing future threats and tail-risks.


This work is focused on human-centric near-termist suffering and death prevention.  Crop pests and diseases have always taken a significant toll on agricultural production, and despite rapid advances in scientific understanding and technological prowess over the previous decades, resultant yield losses are generally still in excess of 20%.  Predictable, consistent losses would be manageable, but there is significant spatial and temporal heterogeneity in pest and disease impacts.  A confluence of climatic, economic, and biological factors can cause total loss of a crop in an area.  It is these more extreme cases that are of particular concern to food security.  As well as causing significant direct harm to many crops from extreme weather and drought, climate change will also alter the distribution of many pest species.  This will be dangerous for food security, as farmers will often be unprepared and ill-equipped to contain novel pests in a region.  Invasive alien species, including those damaging to crops, are also becoming more common due to ever-increasing anthropogenic changes to ecosystems, including through trade introducing new species inadvertently.  Several interventions to combat pests and diseases are available.  Genetic modification technology has great promise, as researchers can create new crop varieties with multiple pathogen resistance mechanisms, conferring longer-lasting resistance compared to traditional plant-breeding approaches.  Early warning systems are also key to inform farmers of what pests and diseases are most likely to pose an imminent threat, and allow for timely action, yet this is only worthwhile when coupled with effective management strategies.  Integrated pest management takes a holistic approach, simultaneously employing many chemical, biological and physical safeguards to protect the crop.  Post-harvest losses can also often be easily prevented with improved storage technology, and employing diverse crop rotations or intercropping, especially including legumes, can improve resilience.

Food Systems Handbook Background

The Food Systems Handbook is an independent non-profit project focused on alleviating food insecurity. We aim to identify promising and efficient interventions and share our findings with our growing network. In practice we conduct literature reviews, conduct interviews and organise roundtable discussions with experts and other food system actors on specific food insecurity drivers and interventions. The emerging insights result in reports which we disseminate with our network of academics, international aid agencies, non-profit organisations, humanitarian aid actors, industry, governments, multilaterals and food system thinkers from around the world. 

Volunteer with Us!

Researchers are deep diving into current and future issues that could lead to acute food insecurity and famine. They read literature about the global food system and the mechanisms that can lead to famine. Further research will go into possible interventions. Researchers contribute to ranking and comparing interventions to help identify the most promising solutions that we want to highlight and advocate for. If you are interested in joining us, read our Volunteer Researcher role description for more details, and feel free to comment on this post or send us a message with any questions.



The Food Systems Handbook (FSH) team is researching drivers of food insecurity and potential solutions. As part of that work we have been reviewing the impacts from crop diseases and pests on food security and the impacts on the livelihoods of farmers. To further our understanding an online roundtable with experts was held on the 5th of August, 2021. Over a dozen disease experts and food system actors came together to discuss existing problems and potential changes and new challenges. The discussion was moderated by Duncan Williamson and the highlights and additional insights have been combined with our research in this report.

This report covers the negative impacts from crop diseases and pests and describes today’s situation briefly. Following is a summary of potential changes emerging from climate change. A look at potential solutions and preventive approaches follows.

Disclaimer: Due to data gaps and different standards in reporting there are uncertainties in the statistics and values cited here. Additionally, as a general recommendation, regions, governments and institutions need to build sufficient functional capacity as a basic step before implementing the recommendations discussed below.

Crop Diseases & Pests Today – How bad is the Problem?

Crop diseases modify or interrupt a crop’s vital functions. Infectious agents, such as bacteria, fungi, viruses or parasitic flowering plants, can reproduce on or within the host, so can be transmitted between plants. The term pest refers to organisms, often insects, which can cause damage to a crop.

Crops are at risk of being affected by diseases or pests throughout their lifecycle, from seed to plate. For instance, fungal pathogens can attack seeds or sprouts directly, while desert locusts prefer juvenile crops, but swarms can eat large quantities of any green vegetation including more mature crops. If not stored properly, harvested produce may be eaten by various pests, such as rice weevils or even rodents, or become rotten.

Crop pests and diseases are a key cause of food insecurity, resulting in average yield losses of over 20% for many key crops including wheat, rice and maize. With no countermeasures in place there could even be yield losses of 100%, causing regional devastation. However, there is significant spatial and temporal variation. Acute food insecurity occurs especially when losses are unusually extensive, or other factors like conflict, drought or market failure coincide.

To provide some examples: desert locusts harmed the livelihoods of at least 3 million people in East Africa in 2020, and caused extensive damage despite $220 million of spending across the region. Meanwhile, the fungal disease Fusarim wilt is threatening the continued viability of Cavendish Banana production worldwide.

While some pests and diseases have affected particular regions for centuries, of particular concern are invasive alien species that can rapidly transition from absent to causing significant damage in a region. These are species that are present in an area due to human activity - perhaps arriving on a ship - and which have the ability to spread rapidly in the new environment. Such species have the potential to severely impact agriculture, as crops may have no immunity to the disease or coping mechanisms for the pest. Invasive alien species are also difficult to counteract as by definition there will not be existing monitoring and control protocols in that region. As such, global cooperation and knowledge-sharing is particularly key here.

Climate Change – How might it affect crop diseases & pests?

While the direct negative impact of climate change on agriculture is expected to be felt most strongly in the tropics (Gornall et al., 2010), rising global temperatures will likely expand the range of crop pests and pathogens into higher latitudes (Bebber et al., 2013). New research combines the results of several crop and climate models to provide more detail about how rising global temperatures will likely impact the relative burden of crop pests faced in different parts of the world.

Chaloner and colleagues (2021) focused on an RCP 6.0 scenario, a high emissions future where GHG concentrations stabilize by the end of the 21st century (see van Vuuren et al., 2011 for more detail). The team found that as temperatures increase, many crops grown in the tropics will face both a lower absolute pest burden and a lower diversity of pest species, but the inverse will be true for higher latitudes. The Sahel, India, Southeast Asia and Brazil are expected to see the greatest drop in crop pests, while Europe, China and the US are projected to become more suitable to many pests, including novel species, to which adaptation may be particularly difficult. In addition to this geographic variation, the degree to which climate change is projected to impact the burden of diseases and pests differs between crops. For example, tropically grown maize, millet and sugarcane are expected to be negatively affected by warming temperatures, but will probably face a lower burden from pests. Rice, however, is projected to experience an increase in pests in many places, although it is also likely to be positively impacted by warming temperatures to a certain degree, before increased temperatures cause crop failures (see Korres et al., 2017).

In the future, the global spread of pests can be expected to be speeded up by international trade (see Kumar et al., 2021 for details on this process and efforts to combat it).

There are several limitations in the work presented in Chaloner et al., 2021 that point to possible future research directions. Firstly, while the authors included tropical pests in their dataset, they acknowledge that the existing literature is biased toward species that are historically found in the Northern Hemisphere. Secondly, the paper bases its analysis of pest ranges only on temperatures and not moisture owing to uncertainties in both climate projections of future hydrology as well as pest populations’ response to water availability. Thirdly, the modelling used does not account for changes in crop growing seasons or for changes in either total growing area or which crops are grown where. Finally, little is known about the relationship between climate change and pest biology (Ziska and McConnel, 2016), so additional research may advance our understanding in this area as well.

Interventions – What can be done about the problem?

There are several dimensions over which possible interventions can vary. An important one is the time the intervention targets in the progression of a pest or disease. Preventative interventions seek to install safeguards and mechanisms that mean outbreaks do not occur. Given that some significant outbreaks will happen, resilience interventions mitigate the scale and spread of the pest or disease. Finally response interventions accept the reality of a given size of outbreak and seek to limit the damage caused to food production. Each of these are important, and some interventions will not be clearly contained in a single category. Interventions may also be categorised according to the epidemiological features of the pest or disease targeted. Here, the most important interventions tend to target diseases or pests that are endemic to the region and constantly impose high crop loss burdens. Next best are interventions focused on transboundary pests or diseases not yet common in a region, but which have the potential to quickly flare up into a severe outbreak in an unprepared plant population. Lastly, interventions that target already well-managed threats that are under control but continue to exact small costs are less crucial but still generally worthwhile. This priority ranking might differ depending on a region's level of established control methods.

Genetic Modification

Introducing transgenes from other species to confer resistance to a pathogen, or otherwise genetically engineering resistance in a crop, is a promising approach to reducing losses from diseases. Plants can also be genetically modified to produce pesticidal molecules to avoid the need for external pesticide application. GM techniques have several key advantages over traditional plant breeding:

  • If no natural population of a crop has the desired trait, then selective breeding cannot produce the trait, but genetic modification can, by introducing transgenes from a species that does have the desired trait.
  • Traditional breeding could fix a new desirable gene in a population over many generations, but genetic modification in theory can be far faster, though regulatory challenges inhibit this rapidity.
  • Relatedly, while breeding approaches often only introduce a single resistance gene, GM techniques make it easier to insert multiple resistance genes to allow for more durable resistance. This could be very valuable because currently, crops with a single resistance gene are often obsolete in 3-5 years as pathogens evolve to overcome the resistance mechanism.


So GM technology is highly promising, and has been successfully used, yet several key challenges remain to be addressed:

  • Crucially, public acceptance of GM crops remains limited, and consequently regulators are reluctant to approve new GM varieties. This reduces the scope of GM crops, and also greatly increases timelines and R&D costs for getting new GM varieties from lab to market. Even when a successful GM crop exists and is in use, as with the ‘rainbow’ papaya variant resistant to the ringspot virus, achieving global adoption remains challenging. In this case, the GM papaya is widely grown in Hawaii but has met significant resistance in many other jurisdictions. Moreover, golden rice has also been very thoroughly tested, yet implementation remains poor.
  • Aside from activist and public concerns over GM technology itself, many stakeholders are more worried about the political economy of the technology. Anti-GM advocates worry that GM crops are another tool to concentrate agricultural power into the hands of a few multinational agrotech corporations, by ensuring farmers are beholden to these companies to purchase seeds and pesticides. These concerns notwithstanding, they are not about the technology itself. Ideally, to complement the profit-driven GM research and development of agribusiness, research universities and government agencies can focus on unprofitable but still important crop improvement, including GM approaches. Another advantage of public GM research is that restrictive intellectual property arrangements could be avoided so that many groups can collaboratively build on each other’s work in an ‘open-source’ way without worrying about patents and access rights. Ideally, this would ensure the benefits of technological progress are reaped more equitably throughout the agricultural economy through improved yields and disease resistance, rather than accruing solely in corporate coffers. That said, industry will always play a large role in agricultural R&D, so food systems practitioners should work with companies and communities to reach mutually beneficial solutions. This is not a zero-sum situation, so all parties should be able to benefit from technological advances.
  • Similar to fears around concentration of agricultural power, others worry that GM crops will focus on a single global variety, at the expense of many localised landraces. This is a legitimate concern, as maintaining cultivar diversity itself mitigates plant pathogen risks. As such, GM advocates should try to separate the technology from necessarily entailing monocultures. As the technology improves and becomes cheaper, it will be more economically viable to produce several GM varieties of a crop suited to different environments and regions.

It was also noted that GM is increasingly a broad class of technologies and definitions vary across countries and over time. Therefore attention must be paid to legal and reputational regional differences, for instance between US and EU regulatory architecture. Specifically, the advent of CRISPR technology which allows very precise modifications to be made to a particular gene, and does not require the introduction of genetic material from other species, could be controlled by different regulations. Currently though, CRISPR remains highly regulated in the EU, greatly limiting its applicability to crop enhancement there.

Panellists at the roundtable generally agreed that how technology is used is at least as important as which technology is used, emphasising the need for nuance and selective rather than wholesale refusals or endorsements of all things GM. Particularly given the key bottlenecks for broader implementation of genetic technologies are regulatory rather than technical issues, policy and governance interventions seem likely to be more impactful than further scientific work. The improved biosafety prospects of CRISPR over previous techniques that were less precise, and transgene-based are promising for advocates of a more favourable regulatory environment.

Early Warning Systems

Forecast for locust swarm spread from March to June 2020. Credit: FAO Locust Watch

Early warning systems (EWS) are in place within many countries for particular pests and diseases, and some are integrated globally. New platforms and data compilations are also coming online. One participant at the roundtable gave the example of the Pest Risk Information Service https://prise.org/.

While it is hard to estimate counterfactuals everywhere, some implementations of EWS lend themselves to academic study. A USDA soybean initiative, for instance, achieved impressive results of an 18% reduction in losses.

However, EWS are only useful in conjunction with effective mitigation technologies and approaches, as learning of an impending infestation achieves nothing unless farmers are equipped to prevent or manage it. Thus, capacity-building work to improve the ability of poor farmers to respond effectively to disease and pest outbreaks is a necessary complement to expanded EWS. One panellist at the round table, for example, recalled that in their context in sub-Saharan Africa, locust warning systems were in place, and were particularly effective when combined with weather data, to coordinate applications of pesticides and fertilisers with regards to pests and likely rainfall. Without the weather data, pesticides and fertilisers might be applied before rainfall, such that they are washed away, leaving the crops unprotected from the locusts. Therefore, EWS may be less impactful – particularly in low-resource contexts – due to limited capacity to act on that foreknowledge effectively. This means that the effectiveness of upscaling EWS depends on the simultaneous adoption of successful control methodologies. Moreover, any successful EWS will have two key components: a sophisticated modelling and data collection and analysis part that accurately assesses and predicts pest and disease burdens over time; and an extensive grassroots communication network to rapidly and thoroughly disseminate relevant warnings. Installing state-of-the-art monitoring and evaluation technology is useless without the social and community structures in place to inform farmers.

Integrated Pest Management

Many fungicides and insecticides are becoming less effective and simply applying higher doses of chemicals is an unsustainable solution. A range of approaches are available that rely less exclusively on chemical pesticides, and are collectively classed as 'integrated pest management' (IPM). One component of integrated pest management could be biopesticides, where natural predators or parasites are targeted to control a specific pest or disease. As with almost all management interventions though, biopesticides alone are generally not adequately effective in reducing crop losses, so the ‘integrated’ part is key, where several management techniques can complement each other to together provide good protection. Promoting and publicising effective reduced-pesticide management systems should sustainably reduce crop losses.

Integrated pest management approaches resist quantification of expected impacts given the diversity of projects included under the banner, and the rich knowledge-base required to successfully collate complex farm-level considerations into each cropping decision. One meta-analysis found that yields increased by an average of 40.9% over 85 projects, but the reviewers note publication bias among other factors may have inflated the seeming efficacy, and other experts are skeptical gains would be so large.

Post-harvest Storage

About one third of food is wasted or lost globally costing USD 1 trillion, and while consumer wastage is more salient in richer countries, inadequate storage facilities in developing nations cause extensive losses, largely to insects, rodents and fungi. With modern storage techniques, this can relatively easily and cost-effectively be reduced to <2% losses (in the pre-consumer phase). However, historically, about 20 times more research funding has gone towards interventions to increase yield compared to improved storage interventions. Thus, this neglected area could contain some easy wins. Hermetic storage – using airtight, dehumidified and vermin-proof containers for harvested grain – has been identified as one particularly promising intervention for on-farm storage.

Econometric analysis suggests that investing in improved storage and transportation infrastructure to reduce post-harvest losses is important, but probably not the most cost-effective. A 10% reduction in post-harvest losses throughout the developing world by 2030 is modelled to cause a 10.5% reduction in the population at risk of hunger in 2050, representing a cost-benefit ratio of 1:11, compared to 1:32 for generic agricultural research investments, using a 5% discount rate.

Cropping Diversity

When a single genetically homogenous crop provides a large proportion of the nutrition and/or calories of a country or region, that region is especially vulnerable to a single crop disease emerging that is very well targeted to that cultivar. As such, a key preventative intervention is to increase the number of crops grown and the number of sub-species and variants of each crop, to ensure a single devastating new disease only has limited adverse impacts on food security. Existing crops common in a region could be planted together in the same field (intercropping), or different varieties of the same crop could be grown together in a field, both of which tend to reduce the disease burden for all plants. This approach requires extensive localisation rather than universal recommendations, as choosing new crops poorly suited to a local climate and situation would increase diversity but at the expense of yield and profit. 

Use of cultivar mixtures was found to produce a modest 2.2% increase in average yields and an 8% reduction in yield instability. However, gains are larger in more pest and disease prone regions, and will be greater still as climate change progresses. Conservation agriculture practices, of which cultivar diversity is a key part, generally have an average return on investment of 1:2.

Legumes to enhance soil biome

Nitrogen-fixing bacteria in legume nodules are particularly important for enhancing soil accessible nutrient content for other crops as well. Moreover, as factory farming (hopefully) reduces in scale over coming decades – for climate change, animal ethics, zoonotic disease, land and water use reasons, to name a few – legumes could play a key role in providing cheap protein to a reduced-meat world. As such, agricultural research should increase its focus on legumes, including diseases and pests specific to legume crops.

Next Steps

“A problem well stated is a problem half-solved.” - C. Kettering, engineer & inventor

After our initial research and this roundtable adding to our understanding of the diverse problems related to crop diseases and pests, we are now progressing towards interventions and recommendations.

For the solutions where sufficient case studies exist, next actions will involve identifying the key stakeholders that can act on these recommendations. We will disseminate our findings with them and aim to map the roadblocks to application, such as regulatory, financial or practical hurdles. Where applicable we will reach out and facilitate cross-sector communication to advance implementation.

Some of the identified solutions above are highly promising but further studies might not replicate or show smaller effect sizes at scale. Here we aim to disseminate the report with actors who can commence additional studies.

Additional outreach will be targeted at regions that will be prone to new crop diseases and pests in the future due to climate change. Here we have open questions such as: What kind of preparation is needed today? What is already underway?

Lastly, we will continue to iterate our interventions framework and update our recommendations accordingly. We will seek and integrate feedback to improve our writings/output, so they will be helpful for risk managers, policy makers, researchers and other food system actors.

Appendix: Methodology

After an initial survey of the prominent literature in the field, especially some review and theory papers, we collated recommendations from these articles into five main intervention classes. Early warning systems and improved seeds are primarily preventative, post-harvest storage and enhance cultivar diversity interventions focus on resilience, and integrated pest management is response-based. Note we did not complete a formal scoping or systematic review, and are open to the possibility of having missed key ideas, hence the importance of the roundtable to garner other suggestions and approaches.

For each of these five interventions, we scanned the literature to understand broadly the role and applicability of each, and target our search using 20 specific questions. These questions included ones about the current state of the intervention, the different actors involved, possible bottlenecks, the role of FSH in the intervention, the timescale and cost of the intervention, and a scan of case studies and best practices for successful implementation. Through this process, we also noted two possible interventions that we excluded as less promising for us to work on: crop research (not our comparative advantage), and promoting pesticides (not neglected, and of dubious value).

The primary researcher then prioritised the interventions in rough order of importance based on scale, tractability and neglectedness, and any other considerations arising from the research, with a particular focus on what we as FSH could contribute to. This was reviewed and updated by colleagues to get a more robust prioritisation. This work is available here.

Finally, we did a more quantitative, though still highly approximate, analysis of each of the five interventions to determine the possible reduction in global hunger upon successful worldwide implementation, summarised here. This complemented and reinforced the earlier prioritisation, and was a useful reminder of the data and modelling limitations that make such prioritising difficult.

Authors: Oscar Delaney¹², Sahil Shah²³4*

¹ The University of Queensland, Australia

² Food Systems Handbook

³ Alliance to Feed the Earth in Disasters

4 Atlantic Council 


*Corresponding Email: hello@foodsystemshandbook.org 

We, the Food Systems Handbook team, are grateful for the many experts that contributed to the discussion at the roundtable event. Additionally, appreciations go out to the people who helped with feedback, proofreading, additional studies and commenting on early drafts. To name a few: Adin Richards, Aron Mill, Catherine Harries, Duncan Williamson, Rahel Schneider, and Ravi Khetarpal.






More posts like this

Sorted by Click to highlight new comments since:

Very interesting! I am curious as to whether there are any estimates of how much these losses could be reduced quickly, e.g. within a year for a catastrophe.

I’m glad you included cost effectiveness estimates.


representing a return on investment of 1:11, compared to 1:32 for generic agricultural research investments.


These appear to be cost to benefit ratios, rather than ROI, which is percent return per year. With cost to benefit ratios, listing the discount rate would be helpful.

Good point, I have fixed it to now refer to cost-benefit ratios.  They used a 5% discount rate, though they found similar results under 3% and 10%.

I did not come across any research on the rapid reduction of food losses.  Market mechanisms could play a significant role here I imagine, as if the price of food quadrupled after a catastrophe impacting food-production, all actors would be far more motivated to reduce wastage even when it requires extra labour or money.  If a food crisis is looming, governments would also increase their focus on maximising production and minimising wastage, which could also bring significant resources to bear on the problem.  So I think post-harvest losses would be markedly reduced rapidly.  But sadly no quantification or proper research on this that I am aware of.

More from OscarD
Curated and popular this week
Relevant opportunities