Which reference groups matter and how? : a relative income information experiment with administrative data

Received wisdom holds that income rank matters for life satisfaction, but causal evidence on the nature and impact of income comparisons is limited. We randomize individuals from a representative sample of mid-career Finns to receive personal rank information from one of several reference groups. We find strong evidence of the effect of rank information on income satisfaction, but weaker effects on life satisfaction, and some evidence of real effects in experimental and administrative data. Effects are strong in narrow reference groups and weak and insignificant in the national one. Finally, we discuss the implications for income transparency policies.