Investing in disaster risk reduction (DRR) measures at farm level is a crucial way to proactively reduce risk exposure at local level and enhance, from the bottom up, the resilience of farming families to natural hazards. Also, the fact that many DRR good practices add value to production even in non-disaster contexts provides an additional incentive to incorporate greater consideration of risks from natural hazards into existing agronomic and natural resources management practices on farms.

A key data gap, up to now, has been a lack of evidence regarding the amount of disaster-induced losses in agriculture that could be avoided by investing in preventive DRR good practices on farms.

The present study makes a significant contribution to answering this question, by analysing the benefits that improved farm-level DRR good practices offer farmers versus technologies and approaches they used in their fields previously. This analysis is based on comprehensive data collected from ongoing and completed FAO field projects in multiple countries, in various world regions. The study developed and applied a systematic methodology to quantify, on a case-by-case basis, how much damage and loss can be reduced through the implementation of DRR good practices at farm level. Various types of hazards were considered (mainly floods, dry spells/drought, and storms). The performance of DRR good practices under both hazard-induced stress and in non-hazard conditions was examined, when both scenarios could be recorded during the study period.

Highlights:

• The farm-level use of practices and technologies intended to reduce disaster risks provides farmers with economic and social benefits that are significantly higher than the benefits they gained from previously used practices.
• Disaster risk reduction (DRR) good practices in agriculture are highly context- and location-specific. Not all have the potential for wider upscaling; rather, targeted upscaling – driven by evidence – should be pursued.
• To truly qualify as a good practice, DRR measures must offer added value in both hazard and non-hazard situations – that is, they must increase agricultural productivity even in the absence of hazards.
• On average, the DRR good practices analysed in this study generated benefits 2.2 times higher than practices previously used by farmers under hazard conditions. The average observed benefit–cost ratio (BCR) was 3.7 in hazard cases – under non-hazard conditions this rose to 4.5. Benefits included both increases in agricultural production as well as avoided hazard-associated damage and loss.
• Prevention and DRR measures in agriculture are especially useful in avoiding or reducing damage and loss from high- to medium-frequency events – which occur with low or medium intensity. Greater emphasis in agriculture sector strategies is needed on farm-level DRR as an effective and relatively low-cost way to prevent and mitigate the types of disasters that most frequently affect vulnerable smallholders. 
• Agricultural development policy, planning and extension work should treat DRR as a priority. Farm-level DRR should not only be mainstreamed in a deliberate manner, but widely promoted and implemented at much larger scales.
• The upscaling of good practices can considerably increase farm productivity and enhance the resilience of smallholder farmers to natural hazards, bring broader benefits at regional and national levels, and contribute to the achievement of the global development agenda articulated by the Sustainable Development Goals, the Sendai Framework for Disaster Risk Reduction 2015–2030, and the Paris Climate Agreement.
• There are two suitable but different paths for upscaling farm-level DRR good practices in agriculture. The first is at a smaller- and incremental scale, through farmer-to-farmer replication, which requires lower investment and institutional support. The second path is through larger-scale efforts in which government or private sector support are needed to promote uptake of good practices at scale. Crucially, both pathways depend on good infrastructure as well as an enabling environment. This means that new initiatives and investments aimed at meeting those critical needs for upscaling are necessary.