1

Dr Mark Fabian, University of Warwick

EA Global London 2023:

Taking happiness seriously: A rejoinder

Introduction

• Massive props to Michael for platforming this event
• HLI’s work is a step in the right direction for EA – good > more lives
• BUT…‘Happiness’ is a bad definition of wellbeing
• Cognitive Behavioural Therapy (CBT) is a narrow conceptualisation and treatment for ‘mental health’ (individualistic & biomedical)
• A fixation on cost-benefit analysis (CBA) and randomised control trials (RCTs) is actively harmful to good wellbeing analysis

2

What is welfare/wellbeing?

• Fabian, M. (2022). A Theory of Subjective Wellbeing. OUP.
• Popular version on the way in 2024/5 – Wholeness: An Honest Guide to Wellbeing
• Wellbeing: what makes life go well
• The prudential good: what is intrinsically good for someone
• NB: DALY/QALY etc. based on welfare as preference satisfaction

3

Wellbeing theories

• Michael’s list only has canonical theories from analytic philosophy (and no eudaimonia!)
• Other important perspectives:
• Functional accounts from psychology: what sort of organism has the human evolved to be
• Geo/Soci/Epi: Wellbeing as an emergent property of a complex social-environmental structure or system
• NB: both of these perspectives are critical for understanding ‘mental health’
• See also the Capabilities Approach

​

4

Intrinsic vs instrumental

• “Oh, this confuses wellbeing as an outcome with the causes of wellbeing”
• ^False: All wellbeing ‘outcomes’ are intimately bound up with causes, e.g.

Rational, authentic desire satisfaction is guided by affective signals like happiness, not directed at them

• Wellbeing is a process as much as an outcome
• Bishop: Wellbeing is a complex causal network

​

5

This is all getting complicated…

• Can’t we just use ‘life satisfaction’ as a proxy for all this complexity?
• ‘All other things held constant’ for structural/sociological factors?
• This would enable ‘rigorous’ analysis ☺
• Some sympathy, e.g. getting some numbers around value of loneliness, etc.
• But broadly NO.

6

The analysis is not rigorous

• See Bruce Tsai on EA forums: https://forum.effectivealtruism.org/posts/ffmbLCzJctLac3rDu/strongminds-should-not-be-a-top-rated-charity-yet
• Social desirability bias: people thought they were getting money
• CBT treats symptoms not causes e.g. poverty, DV, trauma
• Maybe band aid still valuable? Yes
• PHQ-9 vs reality of ‘mental health’
• External validity is non-existent because there is no engagement with context
• Effective social policy in the UK actively undermined by CBA/unidimensional evals

7

Subjective measures

• Work OK in aggregate
• Validity of LS measures is established quantitatively – ongoing qualitative research strikes me as catastrophic to validity
• LS scales work OK until someone gets more money if they are unhappy – then you’ll start to see people lying

​

8

I RCTs & CBA!

My first book 🡺 🡺 🡺

Advocated for:

• Randomised-control trials
• Cost-benefit analysis
• ‘Evidence-based policy’
• Technocracy (implicitly)

9

There are alternative approaches

10

Complex systems perspective

• I am not suggesting woolly, N=5 qualitative methods or opinion
• Agent-based models
• Causal process tracing
• Mixed methods
• EA needs to step out of the economics/analytic bubble
• Getting back to philosophy of science (and falsification) instead of equating science with experiments
• Being comfortable with ambiguity

11

So just back to deworming?

• These problems are germane to EA, not just HLI
• The ‘effective’ in EA refers to analytical effectiveness more than impact effectiveness.
• Development > Deworming (Pritchett 2019)
• But development is a function of institutions, industrialisation, commerce…
• None of these amenable to RCTs and CBA
• I’d rather be right than certain

12

What’s the upshot?

• EA should think harder about ‘wellbeing’
• Invest in good lives not just more lives
• Use multidimensional concepts/measures
• EA should not use RCTs and CBA like a hammer in search of nails
• ‘Effective’ treatments for complex problems are likely to be messy – tolerate it

13

Thanks

14