Food, electricity and the household cost of climate action

A refined global wheat model examines drought, heatwaves and their combination over 1981–2015. The authors report average modelled yield losses of 9.7, 5.8 and 13.3 per cent respectively. Irrigation reduces the damage in their analysis, but its protective effect weakens as extremes intensify. These are historical modelling results, not a forecast for the coming Kenyan harvest.

The domestic connection is substantial. The Kenya National Bureau of Statistics reports that imports supplied 89.8 per cent of Kenya’s wheat supply in 2025. That is a supply share, not the share of every household’s food bill. It establishes why assessment of overseas crop risk belongs alongside domestic production policy. It does not establish a price increase this week.

Irrigation sits within two distinct food-security choices. At farm level, its return depends on the additional yield a water supply can support relative to other crop and management options during difficult seasons. At national level, domestic production, supplier diversification, storage and purchasing arrangements address different risks to availability and cost. Import diversification can reduce dependence on one origin, but it cannot remove a shock affecting several producers.

Water and power change the farm calculation together. A pump can make water available while adding electricity costs; a loan can finance equipment while requiring repayment through a poor season. The return turns partly on which constraint the investment actually relaxes. Kenya’s Water Resources Authority regulates abstractions and allocation, including protection of basic human needs and the environment. Developers, farmers and lenders control different parts of the investment and operating decision.

A global assessment identifies cropland reclaimed from wetlands during 2000–2022 and compares production and flood exposure. Nearly half had lower yields than existing croplands. Its authors estimate that this land contributed 0.1 per cent of total crop production, while its average flood-exposure probability was 23 times that of all croplands. The figures concern this identified class of converted land, not agriculture as a whole.

The finding changes the comparison behind agricultural expansion. Converting a wetland may add hectares without adding much dependable food. It can also place crops and associated investment in a highly exposed location. The global result does not establish that every converted parcel performs poorly, or settle the case for an individual Kenyan proposal.

The wheat and wetland studies address different responses to pressure on food supply. The first tests crop performance under extremes. The second questions where production is expanded. Together they strengthen the case for gains on existing farms as an alternative to opening new land. The balance depends on yield potential, water, access, tenure and livelihoods. Land classified as uncultivated may already support people whose uses and interests are absent from a production map.

Kenya already has a relevant public decision route. The National Environment Management Authority explains that activities likely to harm wetland management require environmental assessment under the applicable regulations, and that sustainable wetland use belongs in land-use planning. The research can inform the evidence requested through that process. It does not replace statutory assessment or establish an automatic prohibition.

For investors and lenders, flood exposure and operating costs can erode the return from additional output. Gains on existing farmland may carry a different balance of production, finance and livelihood effects.

An analysis published as a Comment uses United States power-system, temperature and emissions data to examine electricity during extreme weather. The authors find increased fossil generation during temperature extremes, including systems with low annual fossil shares. Their argument is that the annual generation mix can conceal dependence on thermal power during demanding periods. The article is an analytical Comment, not a trial of competing procurement options.

Cooling protects people from heat while increasing electricity demand. The supply response affects emissions, air quality and cost. Efficient buildings and equipment, flexible loads, storage and networks can contribute different services.

South Africa supplied a concrete policy decision on the same date. Electricity Minister Kgosientsho Ramokgopa proposed prioritising 4,600 megawatts of battery storage and 5,000 megawatts of gas-to-power under the first section 34 determination associated with the Integrated Resource Plan 2025. These are proposed capacities, not completed plants. The two technologies cannot be assessed as though they provide an identical service.

Storage shifts electricity between hours and has a finite discharge duration. Gas supplies additional power and requires fuel. Standby capacity can provide valuable insurance against disruption, but its payment and fuel obligations affect future bills. The relative value changes with the operating conditions.

For Kenya, the institutional connection is specific. The Energy and Petroleum Regulatory Authority sets and reviews tariffs and approves relevant power-purchase and network-service arrangements. Those decisions connect a reliability service with developer obligations and consumer costs. A national annual renewable share alone cannot settle the commercial or operational value of a particular contract.

A European Union (EU) modelling study compares climate-policy packages combining prices, standards, subsidies and transfers. It distinguishes distribution between income groups from differences among households with similar incomes. The authors find that carbon pricing with recycled revenue can offset regressive average effects, while standards reduce variation within income groups. In their model, combining instruments can balance these objectives more effectively.

The distinction helps explain what a household average leaves out. Two families with similar incomes may face different transport needs, housing conditions or opportunities to change equipment. A payment based only on income may therefore leave very different costs after support. These examples illustrate the policy-design problem; the study does not measure their prevalence in Kenya.

The public choice extends beyond the amount of compensation. A price signal may encourage lower energy use without making an affordable alternative available immediately. Standards can improve products or buildings but require compliance spending. Subsidies can help finance a change, while reaching households differently according to eligibility, ownership or the capacity to apply. These interactions shape the burden remaining after support.

For public finance, a subsidy and a transfer have different purposes. One can help change an asset or activity; the other can support a household’s current budget. Financing, timing and administration affect what each delivers. Regulators and sector authorities also control standards that a finance ministry cannot deliver through payments alone. Businesses make the equipment, lending and service decisions that determine whether households can use the support.

The EU results provide a method for local appraisal, rather than a package to copy. Modelled welfare effects do not prove how voters will respond.

A study combines evidence from flood events, insurance records, 57 Tribal Hazard Mitigation Plans and an interview with a tribal representative. It reports that 27 of 574 federally recognised tribes participate in the US National Flood Insurance Program. That is participation by tribal governments, not the percentage of tribal households insured. The programme is administered by the Federal Emergency Management Agency.

The paper identifies problems involving mapping, administrative requirements, affordability and capacity. Its observational evidence does not isolate the causal contribution of each barrier. Nor does one interview represent the experience of every Tribal nation. Its contribution is to make the route into recovery resources part of the resilience question.

A funding instrument can be formally available while the institutions or people most exposed struggle to use it. Offering a cheaper policy may not resolve an eligibility or mapping problem. Improving an application process may not make premiums affordable. The appropriate response depends on which constraint is decisive, and on the institution with power to change it.

In this US setting, federal programme rules and tribal decisions sit in a particular relationship shaped by sovereignty and land arrangements. It would be misleading to translate that relationship directly into Kenyan national or county administration. The narrower relevance lies in the distinction between low demand and barriers involving eligibility, access costs, information or payment conditions.

For an African programme, the cost of participation forms part of the cost of protection. Administrators control eligibility and assistance; financial providers control specified product and servicing decisions. Local evidence would establish where the decisive barrier lies in a named scheme.