When people experience a large negative shock – such as the Great Recession of 2008, the COVID-19 pandemic of 2020, or the Russian invasion of Ukraine – you might expect the experience to be reflected in how they say they feel. This might be picked up in trends in happiness, life satisfaction, and aspects of negative affect like anxiety or stress. But is it?
We address this question with analyses of six micro-data files for individuals across the world. We show that most subjective well-being metrics – sadness, anger, worry, pain, enjoyment - do not shift in predictable ways in response to macro-shocks. Life satisfaction and happiness did decline in the Great Recession and during the COVID-19 pandemic, and in the four countries bordering Ukraine after the Russian invasion in 2022. However, the movements are not large, are relatively short-lived and are not apparent everywhere. This is unsurprising in the light of other studies which have shown well-being recovers rapidly, even after exposure to big negative shocks like natural disasters and terrorist incidents.
When it comes to establishing the effects of shocks on people’s welfare, our study suggests there is value in paying more attention to what people expect to happen in future. People’s expectations of life in general, their financial situation and the economic and employment situation in their country all dropped markedly in the Great Recession and during COVID-19, bouncing back quickly in most cases. However, effects on satisfaction with the economy and democracy persisted for some time.
How does citizens’ well-being look like over the longer run? Here, again, evidence is mixed. There has been secular improvement in life satisfaction in the last decade or so, a trend which coincides with gradual improvement in the United Nations Human Development Index (HDI) which measures institutional features of countries such as their education and welfare systems.
But other trends are not so encouraging. There has been secular growth in negative affect in Europe and, in the United States, in despair as captured by the number of poor mental health days reported in the Behavioral Risk Factor Surveillance System (BRFSS).
What are we to make of these disparate results? The lessons are four-fold. First, well-being is multifaceted so one needs to collect a range of wellbeing metrics to track change over time because trends differ across metrics. Second, one needs high-frequency data to capture rapid movements in well-being after shocks because mean reversion is a feature of much of the data.
Third, as we’ve shown elsewhere, data need to be collected consistently across time within the year to avoid confounding of trends by temporal factors such as seasonality. Finally, traditional wellbeing metrics need supplementing with expectations data because these move strongly with shocks – more so than traditional well-being metrics. Expectations data are also valuable from a policy perspective because, unlike wellbeing measures, drops in expectations are predictive of future negative shocks.
© David G. Blanchflower and Alex Bryson
David G. Blanchflower is Professor of Economics at Dartmouth College
Alex Bryson is Professor of Quantitative Social Science at University College London and IZA Research Fellow
Please note:
We recognize that IZA World of Labor articles may prompt discussion and possibly controversy. Opinion pieces, such as the one above, capture ideas and debates concisely, and anchor them with real-world examples. Opinions stated here do not necessarily reflect those of the IZA.
Related IZA World of Labor content:
Covid-19 and the youth-to-adult unemployment gap by Francesco Pastore
Relationship between recessions and health by Nick Drydakis
Health effects of job insecurity by Francis Green
Photo by Heike Trautmann