TLDR:

I came up with selecting criminally involved participants of cognitive behavioral therapy programs that would reduce crime rate only if the beneficiaries received a relatively small transfer ($200). For institutional changes free education, slaughter training which includes animals as family members welfare reminders, and parental responsibility norms that may imply some family planning.

I am curious if there are better ways of maximizing unconditional cash transfer cost-effectiveness and realistic institutional change improvement ideas that would support the additionality of targeted transfers and also if there are any organizations already working on any of this.

Reasoning:

Reading the GDLive Africa until a perception of saturation, it seems that recipients are spending about

• 30% on housing (including own and larger houses and furniture), example: Decla
• 25% on livestock (including goats and cows to generate income from milk and offspring sales), example: Erick
• 20% on school fees (up to university level, including returning to school, for children and recipients), example: Chepngetich
• 10% on food (including wheat and maize flour, oil, sugar, and vegetables), example: Selina
• 6% on debt repayment (excluding school fees), example: Zawadi
• 3% on agricultural inputs, example: Mkwavi
• 3% on income generating assets (excluding livestock and agricultural inputs), example: Desma
• 3% on curative healthcare, example: Joan

Some research listed on the GD website reports increases in the value of (column 2, p. 2021):

• (67% of total listed items value increase) Non-land assets ($302)
• (19%) Livestock ($83)
• (12%) Durable goods ($53)
• (2%) Savings ($10)

and (p. 14, subset of 401 recipients)

• 31% on livestock
• 22% on housing
• 14% on education
• 10% on food
• 8% on agricultural spending
• 6% on other consumption
• 4% on assets
• 4% on business
• 1% on healthcare.

→ This shows that people mostly invest in livestock and housing (together ~55%), education fees (~15%), and food (~10%).

Recipients report life changed in terms of (p. 15):

• (24%) Living a happier life
  • I wonder to what extent this is an ‘objective’ assessment by the recipients/about the meaning of this expression (I visited a GD office in Uganda and apparently anecdotally, a specifically observed couple got ‘happier’ meaning that they gained weight - of course which is ok but perhaps purchasing more maize is not the way toward virtuous cycle out of poverty)
• (15%) Purchased livestock
• (5%) More food available
• (5%) Set up a business
• (Under 5%): Improvements in health, farming business, relationships, confidence, access to education, ability to travel, assets, ability to communicate

Here, improved housing is not explicitly included. 

→ From this data it can be inferred that people purchase livestock (perhaps to gain income) but besides that nothing much changes.

Monthly spending on the following items increases by (column 2, p. 2016):

• Non-durable expenditure ($37)
• Food ($19)
  • Cereals ($2)
  • Meat & fish ($5)
• Medical expenditure ($3)
• Social expenditure (such as “weddings, funerals, recreation”) ($2)
• Education expenditure ($1)

The low monthly increase in spending on housing, education, and livestock that contradicts other evidence may be caused by these items already purchased.

→ It seems that people do not continue to invest into income-generating assets, business, upskilling, preventive healthcare, but after they secure a place to stay, livestock to keep generating income, and education for some period of time, they spend mostly on consumables.

Based on a conversation with GiveDirectly during their Office Hours, and receiving further materials, it seems that for conditional transfers for students, transfers of more than $1/month make no subjective wellbeing difference perhaps due to the perception of a burden (p. 19).

Women who were receiving $24 bi-monthly transfers for 36–48 months (just enough to buy an additional meal every day to all family members (Section 2)) felt actually a bit better about their family and more hopeful about the future (0.19–0.25 standard deviations - which actually does not seem so much when looking at the first graph of effect size 0.2) (Does money buy happiness? Evidence from an unconditional cash transfer in Zambia). This can be because they could feel continuously supported in the hardship of finding work or income to take care of family.

This study shows that transfers of $200 after cognitive behavioral therapy (CBT) significantly reduced crime rate for criminally engaged people at least one year after the program (while CBT or transfers only had time-limited effects) (Reducing Crime and Violence: Experimental Evidence from Cognitive Behavioral Therapy in Liberia). Since entire communities suffer from crime, CBT with relatively small transfers to criminally involved individuals may be cost-effective in terms of supporting an environment where further growth can take place.

A stress management program that involved breathing exercises and the “get going, keep doing” (p. 12) motto found no impact on intended beneficiaries, regardless of whether they received a $1076 PPP unconditional transfer but the transfer increased psychological wellbeing among program households.

An additional information is that GiveDirectly so far served about 0.06% of the globally poor. (A Fermi estimate involving that 1/10 people lives in extreme poverty a GD transfer can be about $1,000, and GD so far delivered over $500 million). The 99.94% of people may be surviving due to local means or struggle with improved prospects.

→→ So, one way to select beneficiaries can be finding a locally sound cognitive behavioral therapy program for criminally involved individuals that does not include a cash transfer and including a relatively low, about $200, cash transfer. This still enables recipients to do whatever they want with the funds so should be consistent with the GD fundamental values.

An institutional improvement that may be worthwhile is free schooling. Since many beneficiaries are struggling to find finances to eat food, then even though they may value education highly, since it improves their families’ prospects, they may refrain from spending on education.

Of course, affordable/free natural material for housing can be great too, but this can be circumvented by family planning (in conjunction with the responsibility to provide for the family on the parents, not only in terms of food, but also in terms of spacious room in the house to allow for study).

Livestock and other animal farming should be supported only if the non-human animals are practically family members in any families (then, trading and ownership can be understood by the animals as relocation/changing jobs which are providing animal products as opposed to abuse or a stressful environment). In addition, slaughter methods should involve consciousness loss and somewhat benefit the animal to not suffer from old age. For example, some traditional methods should be acknowledged but modern alternatives more considerate to the animals offered. This information can be shared in the form of slaughter training.