41. Collective index insurance

Manufacturing trends, electric tractors, US ad-hoc payments etc.

Hi, if you are new here, I am Rhishi Pethe and you became a member of the “Software is feeding the world” community. You will receive this free weekly newsletter (every Wednesday) at the intersection of technology and agriculture/food systems. I work as a product manager on Project Mineral at Alphabet X, focused on sustainable agriculture. The views expressed in this newsletter are my personal opinions.

This week’s newsletter is the “Canopy” version, a summary of current happenings, research and my thoughts on them.

It covers research on collective index insurance and its application for small-holders, manufacturing trends for 2021, electric tractors, energy sorghum for bioenergy, and research on US farm policy and outlook.

Business model evolution: informal risk sharing in collective index insurance

😅 tl;dr (too long, didn’t read): Insurance is one of the best ways to help small-scale farmers. Insurance models like indemnity insurance and index insurance have issues to work effectively in the smallholder space. With enough adoption, collective index insurance programs can resolve a lot of these issues.

Small-scale farmers in poor countries are similar to average Americans. Small-scale farmers are one drought away from absolute poverty. An average American is one large medical expense away from bankruptcy. Both are one big adverse event away from a bad financial situation.

Insurance is a way to protect farmers. Mechanisms to manage risk in small-scale agriculture are foundational to tackle the scourge of chronic poverty and to improve sustainability. A recent research paper (January 2021 - see references for citation) published in “Nature Sustainability” looks at the impact of collective index insurance. It combines the benefits of indemnity insurance (common in the first world) and informal insurance (common in the developing world).

Informal insurance is common in smallholder space. It is based on tradition, and reciprocity to safeguard against isolated losses in the community, though it does not protect against common shock. The “Agricultural insurance for smallholder farmers: Digital innovations for scale” report from GSMA (May 2020), provides a summary of challenges.

*Basis risk is the possible mismatch between losses and triggering of the index.

So how can/does collective index insurance work? How does it solve the problems of moral hazard, adverse selection and basis risk prevalent in indemnity-based and index-based insurance.

Collective index insurance alleviates basis risk through within-group informal transfers, as shown below. Local monitoring, networks, and social norms reduce basis risk by pooling the excessive payout to protect against uncovered losses. It does suffer from adverse selection, as it requires a minimum fraction of contributors to average out basis risk, and for individuals to take up insurance. 

Image source

As the diagram shows below, it combines good qualities of indemnity, informal, and index based insurance, while leaving behind negatives (except adverse selection). The problem of adverse selection becomes less severe if enough people adopt insurance. 

Image source

The efficacy of peer monitoring (a feature of informal insurance) is a big factor in determining the success of consolidated index insurance adoption, as can be seen from the chart below (m is a measure of efficacy of peer monitoring)

Fig. 4: Peer-monitoring efficacy and success of CII adoption

💡Key takeaway: Collective index insurance and informal transfers can constitute a practical instrument to improve sustainability, both economic and environmental in developing countries.

Share Software is Feeding The World

Manufacturing trends for 2021

😅 tl;dr (too long, didn’t read): Manufacturing is adjusting to the COVID reality. Some of the key trends in manufacturing in 2021 and beyond are around employee safety, connected workforce, IoT adoption, localized production, and predictive maintenance.

The Association of Equipment Manufacturers recently published a report on 5 manufacturing trends to watch for in 2021 (and beyond).

COVID-19 and employee safety

Workplace safety and compliance, will remain a prominent concern for manufacturers well into 2021 and beyond. Meat processing plants with their alleged lack of concern for worker safety took a beating during the middle of the pandemic. COVID cases are still rising in many parts of the world. Manufacturers will have to make significant investments in time and resources to maintain employee safety, access to vaccines, availability of PPE etc. to protect their workers, and ensure continuous operations.

Connected workforce

The adoption of remote tools and a connected workforce has been forced upon us by the pandemic. The process of digitization of processes to improve utilization of physical assets and digital collaboration tools will help manufacturers boost productivity, while enhancing the quality of their products and the quality of life of their employees.

The adoption of digital processes will act as a clear separator between manufacturers who do and those who don’t.

Internet of Things (IoT)

The Internet of Things (IoT) is a read thing (compared to blockchain), as far as manufacturing and the food/agriculture sectors go. With a trend towards a connected workforce, and digitization of processes, IoT will act as a catalyst. Today one in three manufacturing production processes now incorporate smart devices and embedded intelligence. Adoption of IoT within manufacturing requires a strong network infrastructure for communication, and right skill sets in terms of human resources.

Roughly two-third of manufacturing executives believe that IoT will increase profitability over the next five years. The profitability will be driven by (see table below) improvements in customer satisfaction, productivity, quality, on-time delivery, reliability, innovation, and a variety of other factors.

Image source

IoT technology will drive value for the industry by allowing organizations to make measured, informed decisions using real-time data in an effort to increase efficiency and positively impact their bottom lines.

Localized production and near-sourcing

AEM considers localized production and near-sourcing a prominent trend for 2021 for manufacturers. Manufacturers can create a network of smaller, flexible factories located near existing and potential customers. One benefit of being closer to customers is to be able to react quickly to changing customer needs, and act as a risk mitigation mechanism against any future disruptions. My view is that we will see this in products with shorter development lifecycles, or where customers place a premium on “local”. We might see this more in food manufacturing, and less in machinery, electronic equipment, and appliance manufacturing.

Predictive maintenance

The combination of IoT, digitization, and ML/AI will help manufacturers, and customers predict equipment failures, and prevent equipment downtime. The benefits of predictive maintenance can be incredible with reduced costs, fewer failures, smaller scheduled downtime, and optimized parts delivery. 

Doing effective predictive maintenance is not easy, and requires the collection and analysis of large amounts of data. It requires investment in special skill sets either inhouse or through an expert. 

For e.g. C3 AI built a business around providing predictive maintenance capabilities using ML/AI and ultimately went public. From the C3 AI website,

💡Key takeaway: COVID is going to accelerate some of the existing trends within manufacturing. None of the trends are surprising, though localized production & near sourcing will depend on the product manufactured, entrenchment of existing supply chains, product development spans etc.

Monarch launches electric tractors

😅 tl;dr (too long, didn’t read): Smaller electric tractors with autonomy will have a big impact on the existing equipment business models, improve productivity, reduce operating and labor costs, and be more sustainable than their diesel counterparts. 

Along with a remote and connected workforce, IoT adoption, etc. autonomy in agriculture equipment will grow in importance. Monarch recently launched autonomous electric tractors which operate in the 40hp to 70hp range.

The tractor can perform pre-programmed tasks without a driver or an operator can use Monarch’s interactive automation features including Gesture and Shadow modes to have the tractor follow a worker on the job.

The video below is their launch video (8:59). You can watch the first 90 seconds to get a good idea of what it does.

The machine collects and analyzes over 240GB of crop data every day it operates in the field. The tractor is backward compatible with existing equipment and is supposed to work with the next generation of smart equipment. The data is collected by sensors, for real time actuation, to understand growth stage, plant/crop health data, and yield estimates. The 40hp continuous power rating is interesting, as the sub-40hp tractor showed the most growth in 2020 according to the US data from AEM.

Electric tractors can offer farming-as-a-service options. This will be true in places like Africa and south Asia, where the level of mechanization is low, and capex investment is challenging. According to Bain’s Farming-as-a-service report for India (2018), only 40% of farmers in India in 2011 (95% in the US) had access to mechanized equipment.

The initial price of the tractor makes it unsuitable for direct purchase for most farmers. The starting price of the tractor is $ US 50,000, whereas a 42 HP tractor from Mahindra costs about $ US 10,000

I would love to understand the operating costs, which are hopefully much lower than a manned-diesel tractor. For a farming-as-a-service model, you need an on the ground dealer and service network to make it work.

💡Key takeaway: Small electric and autonomous tractors have the potential to improve productivity, safety, environmental impact and to offer a different business model.

Share Software is Feeding The World

Monster drink or sustainability monster? 

Image source

😅 tl;dr (too long, didn’t read): Energy sorghum could be an interesting option for bioenergy as it solves some of the problems associated with annual corn and perennial grasses like miscanthus. Additional research is needed to understand other trade-offs.

In the US, corn is an important crop for bio-energy production.

The recent research published in the GCB-Bioenergy journal (see references for citation), “Ecosystem‐scale biogeochemical fluxes from three bioenergy crop candidates: How energy sorghum compares to maize and miscanthus,” compares annually grown corn, perennial grasess like miscanthus (with high bioenergy potential), and energy sorghum.

Even though large perennial grasses like miscanthus have high bioenergy potential, they are not ideal for crop rotation, as they take several years to establish. Corn and other annual crops are easier to manage with traditional farming, but they are tougher on the environment.

Energy sorghum behaves more like miscanthus in the way it efficiently captures light and uses water to produce abundant biomass. It has higher nitrogen emissions like corn, but researchers believe careful fertilizer management could reduce those levels.

The researchers conducted ecosystem-scale comparisons of carbon, nitrogen, water, and energy fluxes of Sorghum bicolor with corn and Miscanthus. The summary was as follows:

  • Miscanthus and other large perennials offer the best option for biomass production and carbon sequestration, as they feature extensive underground systems for storing carbon and nitrogen and require less fertilizer than annual crops.

  • Corn is highly productive but requires a great deal of water and nitrogen, and it loses carbon stored in its ecosystem through harvesting and tilling. It is the most efficient to convert light into biomass, despite a lower leaf area index and very high evapotranspiration, though corn facilitates substantial C and N return to soil as residue.

  • Energy sorghum has the highest water use efficiency. 

“Energy sorghum falls somewhere in between. As an annual, it can be easily rotated with other crops like soybeans and corn. It's photoperiod-sensitive, so it produces generous yields of biomass late into the season when grown in regions with long days. And because it is drought-tolerant, energy sorghum can be grown in low-rainfall regions, alleviating the pressure a growing biofuel industry could place on existing land used for food production.”

Additional research and analysis is necessary to understand the interaction between crop type, climate and management to forecast the long‐term sustainability of these key bioenergy crops.

💡Key takeaway: Researchers are looking at different options to find more suitable crops for bio-energy needs. Energy sorghum seems to have the potential to address some of the needs around input use, sustainability, and productivity though more research is required.

Farm policy outlook after 3 years of ad-hoc payments

😅 tl;dr (too long, didn’t read): In the US, ad-hoc payments through the two different government programs have shot through the roof over the last 3 years. There are a couple of key learnings, which should be applied to any future programs of this nature (if they are implemented).

In 2020, farmers in the US received huge amounts of assistance from the government in terms of different programs. Joe Janzen and Nick Paulson from the University of Illinois, analyze data around the MFP (Market Facilitation Program - targeted specific production areas for a handful of crops) and CFAP (Coronavirus Food Assistance Program - more widely distributed across the country), and provided some outlook after three years of ad-hoc payments to farmers since 2018.

The distribution of these ad hoc payments changed over time, becoming more diffuse across commodities and regions. Producers of row crops (mainly corn, soybeans, wheat, and cotton) received a majority of the MFP payments.

For commodities like corn and soy, USDA’s most recent forecast puts 2020 net farm income at $120 billion, second only to the record farm income of $124 billion in 2013 and well above the 10-year average of $93 billion. This has led some to question the necessity of these ad hoc programs, including the type of producers that this aid has gone to.

Some of the policy lessons from the last three years are as follows

  1. Understanding of timelines, even though it is difficult to predict losses in advance is critical.

  2. Payment rates and quantities have to be calibrated to actual economic damage.

  3. Do not try to apply one methodology to measure damages across many commodities.

  4. Program design should not affect farmer acreage allocation decisions.

💡Key takeaway: The last three years have seen historic amounts of ad-hoc payments from the US government to farmers. There are some key policy learnings for the US The biggest lesson is to make sure that payment rates and quantities are calibrated to actual economic damage, as it is not only a political issue, but also a fairness and appropriate resource allocation issue.

So, what do you think?

💗 If you like “Software is feeding the world”, please hit the “Like” button at the bottom or share with a friend.

🙏 If you don’t mind answering 3 questions anonymously (2 are optional), I would love to get your feedback.

References:

Ecosystem‐scale biogeochemical fluxes from three bioenergy crop candidates: How energy sorghum compares to maize and miscanthus Caitlin E. MooreAdam C. von HadenMark B. BurnhamIlsa B. KantolaChristy D. GibsonBethany J. BlakelyEvan C. DracupMichael D. MastersWendy H. YangEvan H. DeLuciaCarl J. Bernacchi

First published: 22 December 2020 https://doi.org/10.1111/gcbb.12788

Janzen, J. and N. Paulson. "IFES 2020: Post-Election Farm Policy Outlook After Three Years of Ad Hoc Payments." Department of Agricultural and Consumer Economics, University of Illinois at Urbana-Champaign, January 7, 2021.

Santos, F.P., Pacheco, J.M., Santos, F.C. et al. Dynamics of informal risk sharing in collective index insurance. Nat Sustain (2021). https://doi.org/10.1038/s41893-020-00667-2