VoxEU Column Frontiers of economic research

The effects of terrorism on well-being and economic activity

Terrorism wreaks a terrible cost on societies. This column quantifies some of the effects by employing daily data on individual, self-reported emotional feelings combined with time allocation data from the American Time Use Survey. The focus is on the days before and after the 2013 Boston Marathon bombing. The data show a significant drop in well-being, driven by the responses of women and Massachusetts residents. Hours worked were not affected.

Terrorist acts are likely to have negative effects on the economy, although these are usually hard to quantify. Abadie and Gardeazabal (2003) show that terrorism had a large and negative effect on local entrepreneurship and GDP in the Basque region of Spain. On the other hand, Becker and Rubinstein (2011) (see also Waxman 2011) conclude that only occasional users reduced their consumption of goods and services (such as bar consumption and bus rides) that were potentially targeted by terrorists during the ‘Al-Aqsa’ Intifada, while the consumption of frequent users was not significantly affected. More recently, Metcalfe et al. (2011) estimated the effect of the 9/11 attacks in New York on the subjective well-being of the British, using an index of psychological well-being (the 12-item General Health Questionnaire) in British Household Panel Survey data. In particular, Krueger (2007), using data from Wisconsin, found that 9/11 temporarily increased sadness while reducing enthusiasm. In this column, we present our findings concerning the effect of an isolated terrorist act, the immediate impact of which on Americans’ well-being and time allocation is likely to resemble that for occasional users in Becker and Rubinstein (2011) for Israel.

The economic effects of the 2013 Boston Marathon Terrorist Bombing

In this column, we present the first evaluation of the effects of the 2013 Boston Marathon bombing on Americans’ subjective well-being and time allocation, using daily information on individual emotional feelings and time allocation from the American Time Use Survey (ATUS) Well-Being Module.1

There were two explosions at the 2013 Boston Marathon that caused the death of three spectators and a policeman, while 264 other spectators were injured. The city of Boston was paralysed until the terrorists were captured by the police. The bombing received wide coverage in the news in the US and globally. While we therefore expect any effect to be stronger in the geographical area around Boston, individuals in other parts of the US are also likely to have been affected.

We rely on a direct daily measure of subjective well-being collected in association with the carrying out of daily activities. ATUS respondents fill in an activity diary covering a 24-hour period, and report their experienced emotional feelings with respect to three randomly drawn activities (out of all of the activities reported over the 24 hours). This measure of subjective well-being is more focused than broader recall questions regarding life satisfaction that likely correspond to much longer time periods (Kahneman et al. 2004, Kahneman and Krueger 2006, Krueger and Mueller 2012). The ATUS survey and the Well-Being Module were in the field continuously both before and after the Boston marathon attack, so that we can evaluate the immediate impact of the bombing. While we do not observe the same individual before and after the attack, the activity diaries collected on the days just before and just after the bombing are likely to be very similar in every respect, except for the occurrence of the terrorist attack.

The average subjective well-being effect is large (see Figure 1). Our estimates indicate a large and significant drop in happiness after the bombing, varying in size from -0.80 (one week after the event) to -0.36 (further away in time) (see Clark and Stancanelli 2016 for details). The effect is largest in the days closer to the bombing, as expected, but remains sizable over a month later. The average person in the sample reported an average happiness score of 4.40 in their three randomly drawn activities in the ten days preceding the bombing. The bombing therefore reduced Americans’ feelings of happiness by about 9-20%. This is largely driven by the responses of those living in Massachusetts and other states that are geographically close to Boston, for whom happiness fell by about 1.5 points, which is a very sizable drop on a scale of 1 to 6. By way of contrast, residents living in other states were not affected, except for women.

The negative effect of the bombing on happiness appears to be larger than the estimated effect of a percentage point drop in GDP, although such comparisons are hard to make (Clark 2011, Stevenson and Wolfers 2008).2 It is also larger than the drop in individual well-being caused by unemployment (Alan Krueger and Andreas Mueller 2012).

Looking at the impact of the bombing on Americans’ allocation, we consider hours worked, household work, active leisure (such as playing sports, cycling and walking), and time spent watching television or listening to the radio (for details of the results, see Clark and Stancanelli 2016). We find a significant rise in hours worked, which actually reflects the weekly allocation of working time, as hours worked increased also on the Monday a week earlier and a year earlier, on Monday 16 April, the day of the 2012 Boston Marathon. Moreover, this positive effect of the day of the bombing on hours worked is not statistically significant for states close to Massachusetts, as the Marathon Monday is a holiday in most of those states. There is some limited evidence that active leisure hours dropped everywhere by about 20 minutes and by an hour in Massachusetts and other states close to Boston. This is in line with the prediction that outdoor activities would fall after terrorist attacks (Becker and Rubinstein 2011). Neither household work nor time spent watching television and listening to the radio was affected.

Putting the well-being effect of the 2013 Boston bombing in perspective

We also consider another tragic event – the fatal shooting of 20 children and six staff members at Sandy Hook elementary school on 14 December 2012. This event also received a great deal of media coverage in the US and abroad. Using a similar approach as for the Boston bombing, and comparable data drawn from the 2012 ATUS, we find no impact of the Sandy Hook shooting on individual well-being or time allocation (Clark and Stancanelli 2016). This finding is in line with Kip Viscusi (2009), who points to much stronger population reactions to deaths from terrorism than from other causes. Terrorist attacks that can potentially hit anyone may affect societal well-being via increased feelings of fear (Becker and Rubinstein 2011). The Boston bombing is also more likely to have reminded Americans of the dramatic events of 11 September 2001, when almost 3,000 people died in an unprecedented Al-Qaeda terrorist attack, than did Sandy Hook.

The short versus long-run effects of the 2013 Boston bombing

To sum up, we find a significant drop in Americans’ happiness following the Boston terrorist attack, driven by the responses of women and of residents of Massachusetts and surrounding states. Outdoor activities such as active leisure activities (playing sports, for example) fell after the bombing, as expected, while hours worked (and hence economic activity) were not affected. 

To test for the long-run effects of the Boston bombing, we use data for the preceding year 2012 in addition to the 2013 data, and compare the level of happiness of Americans across the 2012 Boston Marathon and the 2013 Boston Marathon, as well as including all the months available until December 2013 in the analysis (over six months after the Boston Marathon attacks. The details of the approach are presented in Clark and Stancanelli (2016). Happiness fell in the immediate aftermath, but in the longer run (a year later) the impact becomes smaller and insignificant. Terrorism in countries where such attacks are rare therefore has a sharp, but not permanent, impact on well-being.  


Abadie, A and J Gardeazabal (2008), “Terrorism and the world economy”, European Economic Review 52(1): 1-27.

Abadie, A and J Gardeazabal (2003), “The Economic Costs of Conflict: A Case Study of the Basque Country”, American Economic Review 93(1): 113-132.

Becker, G S and Y Rubinstein (2011), “Fear and the Response to Terrorism: an Economic Analysis”, CEP Discussion Paper No 1079.

Clark, A E and E Stancanelli (2016), "Individual Well-Being and the Allocation of Time Before and After the Boston Marathon Terrorist Bombing", PSE mimeo.

Clark, A (2011), “Income and Happiness: Getting the Debate Straight”, Applied Research in Quality of Life, 6(3): 253-263.

Frey, B, S Luechinger and A Stutzer (2007), “Calculating Tragedy: Assessing the Costs of Terrorism,” Journal of Economic Surveys 21(1): 1-24.

Kahneman, D, A B Krueger, D A Schkade, N Schwarz and A Stone (2004), “A Survey Method for Characterizing Daily Life Experience: The Day Reconstruction Method”, Science: 1776- 1780, 3 December.

Kahneman, D and A B Krueger (2006), “Developments in the Measurement of Subjective Well-Being”, Journal of Economic Perspectives 20(1): 3-24.

Krueger, A B  (2007), What Makes a Terrorist, Princeton University Press. 

Krueger, A B and A Mueller (2012), “Time Use, Emotional Well-Being and Unemployment: Evidence from Longitudinal Data”, American Economic Review, papers and proceedings 102(3): 594-599. 

Metcalfe, R, N Powdthavee and P Dolan (2011), “Destruction and Distress: using a quasi-experiment to show the effects of the September 11 attacks on mental well-being in the United Kingdom”, Economic Journal 121(550): 81-103.

Stevenson, B and J Wolfers (2008), “Economic Growth and Subjective Well-Being: Reassessing the Easterlin Paradox”, Brookings Papers on Economic Activity: 1-87.

Viscusi, W K (2009), “Valuing Risks of Death from Terrorism and Natural Disasters”, Journal of Risk and Uncertainty 38(3): 191-213.

Waxman, D (2011), “Living with terror, not living in Terror: The Impact of Chronic Terrorism on Israeli Society”, Perspectives on Terrorism 5(5-6): 1-13.


1 This survey data was not available at the time of the 9/11 terrorist attacks.

2 Stevenson and Wolfers (2008) estimate a gradient of about 0.40 between subjective well-being and GDP growth (specified as the logarithm of GDP per capita), using either World Values Survey or Gallup World Poll data and a variety of empirical approaches, as well as in 1972-2006 data from the US General Social Survey.

6,515 Reads