VoxEU Column Frontiers of economic research Macroeconomic policy

Busts hurt more than booms help: New lessons for growth policy from global wellbeing surveys

How do macroeconomic changes affect people’s wellbeing?  This column presents evidence that the life satisfaction of individuals is between two and eight times more sensitive to negative economic growth than it is to positive economic growth. Engineering economic ‘booms’ that risk even short ‘busts’ is unlikely to improve social wellbeing in the long run.

How do macroeconomic fluctuations affect individual welfare? Nobel laureate Robert Lucas famously suggested that the cost of business cycles in terms of consumption is insignificant (Lucas 1987, 2003). Such findings paved the way for a straightforward growth policy that tends to be evaluated on the basis of how much the economy has grown rather than how the economy has grown. Yet, the psychological impact of the growth trajectory – whether volatile or smooth – is understudied even though it is felt strongly by ordinary citizens and results in markedly different reports of life satisfaction.

The analyses reported in our new paper (De Neve et al. 2014) reveal this important asymmetry in the way that individuals experience positive and negative macroeconomic fluctuations.

We find evidence that the life satisfaction of individuals is between two and eight times more sensitive to negative growth as compared to positive economic growth. People do not psychologically benefit from expansions nearly as much as they suffer from recessions.

These results suggest that policymakers seeking to raise wellbeing should focus more on preventing busts than inculcating booms. Our results also offer an explanation for why increases in GDP do not always pay off in increases in happiness – the modest happiness gains accrued over years of growth can be wiped out by just a single year of contraction.

Measuring sensitivity to positive and negative growth

We employ wellbeing reports as a proxy for ‘experienced utility’ to show that a greater welfare weight is placed on national income losses than equivalent gains. To obtain a global sense for how individuals react to positive and negative growth we use wellbeing measures across three large data sets. We use Gallup World Poll1 data drawn from 151 countries, BRFSS2 data taken from a representative sample of 2.5 million US respondents, and Eurobarometer3 data that cover multiple business cycles over four decades.

Generally, we find that economic growth is significantly associated with wellbeing. However, when split across positive and negative economic growth, we find that this result is mostly driven by the negative growth years.

Recession years are significantly associated with losses in wellbeing. We do not, however, obtain an immediate relationship between positive growth years and wellbeing.

In the Eurobarometer model shown in the Figure 1 below, a 10% economic expansion is associated with a 0.07 standard deviation increase in life satisfaction, while a 10% contraction is associated with a 0.41 standard deviation decrease in life satisfaction – about six times the magnitude. On the four-point life satisfaction scale used by Eurobarometer (very satisfied, fairly satisfied, not very satisfied, and not at all satisfied), these standard deviations correspond to a 0.09 point increase and a 0.55 point decrease, respectively. In short, individuals experience losses much more acutely than equivalent gains in a macroeconomic setting.

This basic finding is further put to the test in De Neve et al. (2014), where individual controls are introduced (including unemployment data), lagged variables, and individual expectations. Generally, the asymmetric experience of positive and negative growth is robust to virtually all specifications and holds up neatly in country-year and quarterly US state-fixed effect models across the data sets.

Figure 1. Economic growth and change in life satisfaction

Why do individuals experience more strongly macroeconomic losses than equivalent gains?

There are several possible avenues to address this important follow-up question. A deep-rooted mechanism could be that individuals simply react more strongly to negative developments. Humans’ disproportionate sensitivity to negative stimuli and the general finding that ‘bad is stronger than good’ (Baumeister et al. 2001, Kahneman and Tversky 1979) may have an explanation rooted in evolutionary biology, since in terms of survival the avoidance of threats is more important than a missed opportunity (McDermott et al. 2008).

Periods of economic contraction not only involve a loss of national income but also an increase in economic uncertainty (Bloom 2009, 2014), which is arguably intensified by the disproportionate coverage of negative news about macroeconomic trends compared to respective positive trends (Soroka 2006). Volatile business cycles and the resulting uncertainty are also attention-seeking stimuli (Wiggins et al. 1992). Luechinger et al. (2010) highlight the role of economic insecurity in increasing angst and stress by showing that the subjective wellbeing of employed individuals working in the public sector—who in general enjoy more job protection—is less acutely affected by economic shocks than comparable workers in the private sector.

Economic growth and happiness in the long-run

The central finding of this research also allows us to revisit the longstanding debate on the relationship between economic growth and wellbeing. When considering the available longer term data that cover entire business cycles, it would appear that wellbeing reports have not risen in most of the world’s economically developed nations, despite having their real GDP almost doubling over the past four decades (see Figure 2 below). This result echoes research by Easterlin et al. (2010). The ‘paradox’ resulting from the conflicting findings in the short-term (using annual growth data that incorporates both positive and negative growth years) versus long-term (real GDP data that cover one or more business cycles) can perhaps be better understood in light of the results on macroeconomic ‘loss aversion’ discussed here. Recessions can rapidly undo the wellbeing gains from longer expansionary periods and lead to an insignificant relationship between national income and average wellbeing in the long-run.

Figure 2. Aggregate life satisfaction for a number of countries by survey year

Greece may illustrate this phenomenon. Figure 3 shows the available Eurobarometer data for Greece, demonstrating the asymmetric experience of positive and negative economic growth. Real GDP grew over 50% from 1981 until 2008, while life satisfaction edged up by about 5 to 10% overall (with most gains in wellbeing made over the latter decade of stable growth). The recession that started in 2008 led to a decrease in average wellbeing that erased all prior gains. Average wellbeing in Greece now stands at a level below historical records, despite real GDP remaining at a level well above historical figures. The psychological costs of the recession cut even deeper than the (negative) growth numbers would indicate.

Figure 3. Subjective wellbeing (SWB) and real GDP for Greece, 1980-2012

Implications for growth policy

A better understanding of this asymmetric sensitivity has macroeconomic policy implications and addresses a long-standing debate. On the one hand, a typical reading of the income-happiness paradox suggests that further growth in the developed world is a futile means to the end of improving social wellbeing. On the other, researchers who find evidence of a positive relationship between wellbeing and GDP typically take from this that further economic growth is good for society.

Our findings suggest a more nuanced perspective; policy designed to engineer economic ‘booms’, but that risks even relatively short ‘busts’ is unlikely to improve social wellbeing in the long-run. Steady positive growth that minimises the risk of economic contraction seems the most likely route to improving general wellbeing.


Baumeister, R F, E Bratslavsky, C Finkenauer and K D Vohs (2001), “Bad is Stronger than Good”, Review of General Psychology 5 (4), 323.

Bloom, N (2009), “The Impact of Uncertainty Shocks”, Econometrica 77 (3), 623-685.

Bloom, N (2014), “Fluctuations in Uncertainty”, Journal of Economic Perspectives 28 (2), 153-175.

De Neve, J-E, G Ward, F De Keulenaer, B Van Landeghem, G Kavetsos and M Norton (2014), “Individual Experience of Positive and Negative Growth is Asymmetric: Global Evidence Using Subjective Well-being Data”, LSE Centre for Economic Performance Discussion Paper No. 1304.

Easterlin, R A, L A McVey, M Switek, O Sawangfa and J S Zweig (2010), “The Happiness-Income Paradox Revisited”, Proceedings of the National Academy of Sciences 107(52): 22463-22468.

Kahneman, D and A Tversky (1979), “Prospect Theory: Analysis of decision under risk”, Econometrica 47(2), 263-291.

Lucas, R (1987), Models of Business Cycles, New York: Blackwell.

Lucas, R (2003), “Macroeconomic Priorities”, American Economic Review 93 (1), 1-14.

Luechinger, S, S Meier and A Stutzer (2010), “Why Does Unemployment Hurt the Employed? Evidence from the life satisfaction gap between the public and the private sector”, Journal of Human Resources 45 (4), 998-1045.

McDermott, R, J H Fowler and O Smirnov (2008), “On the Evolutionary Origin of Prospect Theory Preferences”, Journal of Politics 70 (02), 335-350.

Soroka, S N (2006), “Good News and Bad News: Asymmetric responses to economic information”, Journal of Politics 68 (2), 372-385.

Wiggins, S, P Whyte, M Huggins, S Adam, J Theilmann, M Bloch, S B Sheps, M T Schechter, and M R Hayden (1992), “The Psychological Consequences of Predictive Testing for Huntington’s Disease”, New England Journal of Medicine 327 (20), 1401-1405.


1 The Gallup World Poll is a large-scale repeated cross-sectional survey initiated in 2005 covering more than 150 countries.

2 The Behavioral Risk Factor Surveillance System (BRFSS) contains quarterly data obtained through telephone surveys and is carried out by the Centre for Disease Control and Prevention (CDC) in the US. In total, there are 1,197 state-quarter observations from 2005/1 to 2010/4.

3 The Eurobarometer is an opinion survey carried out on behalf of the European Commission that has typically, though not always, taken place twice a year. The data we use come from an unbalanced panel of the so-called EU-15, for which the minimum time-series is 18 years and the maximum 39 years.

3,779 Reads