Taking Happiness Seriously: Can We? Should We? Would It Matter If We Did? A Debate
Dr Michael Plant
London EAG
May 2023
For £1,000 you could:
Double the annual income of 1 household
Provide 250 bednets
Treat 10 women �for depression
We want to do ‘the most good’. But how do we assess what works?
Deworm 1,300 children
Two paths to measuring impact
A new, radical idea?
“the care of human life and happiness [...] is the first & only legitimate object of good government” - Jefferson, 1809
“the greatest happiness of the greatest number is the foundation of morals and legislation” - Bentham, 1776
“Happiness is the meaning and the purpose of life, the whole aim and end of human existence” - Aristotle
What’s new? Data
The case for taking happiness seriously
Therefore, we should take happiness seriously, set priorities using the evidence on subjective wellbeing.
QALYs (Quality-Adjusted Life Years)
A well-established measure of health,
combining quality & quantity of life.
1 QALY = 1 year of healthy life
Based on hypothetical trade-offs, not evidence on people’s experiences
GiveWell’s moral weights
GiveWell (2020) 2020 update on GiveWell's moral weights
2. Happiness can be measured
Let’s replace the Objective Indicators approach with:
Wellbeing-Adjusted �Life Years (WELLBYs)
1 WELLBY equals:
UK Office of National Statistics (2021)
Can we rely on subjective measures?
Wealth
Health
Being in a partnership
What your friends say
How often you smile
Suicide
Kahneman and Krueger (2006)
Layard (2020) Can We Be Happier?
World Happiness Report 2022
Challenges with happiness data
3. Our predictions about happiness are often wrong
We’re often wrong (about happiness)
Ord (2013) The Moral Imperative toward Cost-Effectiveness in Global Health
3. Our predictions about happiness are often wrong
But, does it matter? HLI investigates
| GiveWell’s ‘objective indicators’ approach |
Cash transfers (GiveDirectly) | 1 |
Therapy for depression (StrongMinds) | >1 ? (no analysis) |
Deworming (Deworm the World) | |
Anti-malarial bednets (AMF) | 13 |
GiveWell figures come from their CEA (2022)
Effects over time of cash and therapy
(Author’s own figure)
Summary of cash v therapy results
| WELLBYs /treatment | Cost�(USD) | WELLBYs /$1000 |
Cash transfers (GiveDirectly) | 9 | $1,220 | 7.3 |
Therapy for depression (StrongMinds) | 10 | $170 | 62 |
Ratio (therapy vs cash) | 5% more effective | 14% of�cost | 8x more cost-effective |
Deworming: unclear effects
Figure from Dupret et al. (2022)
GiveWell figures come from their CEA (2022), and SWB re-analysis of StrongMinds and AMF (2023). HLI figures from reports on StrongMinds (2021), GiveDirectly (2021), updated in (2022) and (2023), AMF (2023) and deworming (2023).
| GiveWell (‘objective indicators’ approach, pre-2023) |
Cash transfers (GiveDirectly) | 1 |
Therapy for depression (StrongMinds) | >1 ? (no analysis) |
Deworming (Deworm the World) | |
Anti-malarial bednets (AMF) | 13 |
SWB approach (HLI) |
1 |
8 |
≈0 (no clear effect) |
Accounting for uncertainty
GiveWell figures come from their CEA (2022), and SWB re-analysis of StrongMinds and AMF (2023). HLI figures from reports on StrongMinds (2021), GiveDirectly (2021), updated in (2022) and (2023), AMF (2023) and deworming (2023).
| GiveWell (‘objective indicators’ approach, pre-2023) |
Cash transfers (GiveDirectly) | 1 |
Therapy for depression (StrongMinds) | >1 ? (no analysis) |
Deworming (Deworm the World) | |
Anti-malarial bednets (AMF) | 13 |
SWB approach (HLI) |
1 |
8 |
≈0 (no clear effect) |
SWB approach (GiveWell’s revised numbers) |
1 |
2 |
? (no GW analysis) |
Comparing life-extending (bednets) versus
life-improving (cash, therapy, worms)
Two tricky philosophical issues:
It makes a big difference
Where bednets are more cost-effective than treating depression
Where respondents seem to put the neutral point
Figure from Plant et al. (2022) (annotated)
GiveWell figures come from their CEA (2022), and SWB re-analysis of StrongMinds and AMF (2023). HLI figures from reports on StrongMinds (2021), GiveDirectly (2021), updated in (2022) and (2023), AMF (2023) and deworming (2023).
| GiveWell (‘objective indicators’ approach, pre-2023) |
Cash transfers (GiveDirectly) | 1 |
Therapy for depression (StrongMinds) | >1 ? (no analysis) |
Deworming (Deworm the World) | |
Anti-malarial bednets (AMF) | 13 |
SWB approach (HLI) |
1 |
8 |
≈0 (no clear effect) |
SWB approach (GiveWell’s revised numbers) |
1 |
2 |
? (no GW analysis) |
1 - 12 (depends on assumptions) | 9 (uses GiveWell’s ‘house view’) |
Wrapping up
References