Average trust did not change with social assistance in Kenya
Among informal-economy households, there was no average change. In counties with restrictions, some estimates pointed to greater trust.
The key idea
Average trust did not change. Some estimates were more favorable in counties with restrictions.
The study, in plain language
Financial assistance covers immediate expenses, but the question remains whether it also affects how much people trust and cooperate. The team used repeated national surveys of households linked to Kenya's informal economy, from 2018 and December 2020, before and after the first COVID-19 wave, and compared households with and without social assistance.
On average, the sample showed no change in measures of cohesion. When counties with restrictions were examined separately, some estimates pointed to greater trust in government and greater willingness to cooperate voluntarily. This favorable pattern applied to that subgroup, not all households; other changes during the period may have contributed, so it cannot be attributed to the assistance.
What they found
- Across all households, there was no visible change in trust or cooperation.
- In counties with restrictions, some estimates pointed to greater trust in government and greater willingness to cooperate.
- The assistance covered about a quarter of household spending, and outcomes varied with local conditions.
Something to think about
When assistance shows no average change, what differences in context might that average hide? Some counties looked different here, but the study cannot say why.
How they studied it
The researchers compared households with and without social assistance before and after the first COVID-19 wave, using repeated national surveys of households linked to Kenya's informal economy: 1,188 households in 2018 and 2,608 in December 2020. They did not follow the same households or randomly assign assistance. The complete-case analysis included 3,416 household heads. They analyzed counties with restrictions separately. These were repeated surveys, not a panel following the same households, and the parallel-trends assumption needed for causal attribution could not be verified.
What the study cannot show
- Results for particular groups may reflect changes that occurred at the same time as the assistance.
- The finding applies to this program and context, not all social assistance.
The original study
Social Protection and Social Cohesion in Times of the COVID-19 Pandemic: Evidence from Kenya
Source under CC BY 4.0. English adaptation and explanation by PaperFlow, with AI assistance. The text has been reorganized for a general audience; it is not the complete paper and does not imply endorsement by its authors.
One study alone does not represent a field's consensus. Educational content does not replace professional advice. Report a possible error.
