Skip to main content
Log in

The impact of earthquakes on economic activity: evidence from Italy

  • Published:
Empirical Economics Aims and scope Submit manuscript


Although earthquakes are large idiosyncratic shocks for affected regions, little is known of their impact on economic activity. Seismic events are rare, the data are crude (the Richter scale measures the magnitude, but says nothing of the associated damages), and counterfactuals are often entirely absent. Using a geophysical methodology devised to gauge seismic damages (the so-called Mercalli scale), we study the evolution of output and employment following seismic events in 95 Italian provinces from 1986 to 2011 for a total of 22 earthquakes. Our identification strategy relies on ideal counterfactuals: ex ante identical neighboring provinces that only differ ex post in terms of damages. We show that following an earthquake, the observed contraction of output and employment is generally small or even negligible. In some cases, the net effect on output and employment can be positive because the stimulus from the reconstruction activities more than compensate for the destruction of physical capital. Finally, we show that the effects on economic activity are nonpersistent, do not spill over from the epicentral region to the neighbors, and tend to be reabsorbed within 2 years from the event.

This is a preview of subscription content, log in via an institution to check access.

Access this article

Price excludes VAT (USA)
Tax calculation will be finalised during checkout.

Instant access to the full article PDF.

Fig. 1
Fig. 2
Fig. 3
Fig. 4
Fig. 5
Fig. 6
Fig. 7
Fig. 8
Fig. 9
Fig. 10
Fig. 11
Fig. 12

Similar content being viewed by others


  1. See Cavallo et al. (2010) for an excellent review on this point.

  2. Italy is one of the most seismic countries in the world being located in between the Eurasian and the African plate. Statistically, the country experiences a significant earthquake every 4 and a half years. Thanks to a long history of records the National Institute of Geophysics and Volcanology (INGV) provides the information on all recorded episodes.

  3. The correlation between the moment-magnitude and the severity (and extension) of the damages is zero across provinces affected by the same event because there is only one magnitude for each earthquake measured at the epicenter, while the damages vary greatly across provinces.

  4. For a paper on the frontier on how to identify the effects on business dynamics generated by floods.

  5. In this paper, differences in social capital across provinces are captured by the provincial fixed effect given their persistency over time. Although we control for this factor, the analysis of its direct impact goes beyond the scope of this paper and we reserve to investigate this aspect in more detail in future research.

  6. For an excellent survey of the literature see Cavallo and Noy (2009) and Hochrainer (2009).

  7. For the definition of ‘Building Type I’ see “Appendix”.

  8. For the definition of ‘Window Type I’ see “Appendix”.

  9. The \(R^{2}\) of the regression is 0.81.

  10. The Department of Civil Protection is a structure of the Prime Minister’s Office which coordinates and directs the national service of civil protection. When a national emergency is declared, it coordinates the relief on the entire national territory following natural disasters or catastrophes. In this case, the council of ministers declares the ‘state of emergency’ by issuing a law by decree and identifies the actions to be undertaken.

  11. The implicit assumption is that—conditional on Mercalli ranks—the damages are uniformly distributed across types of buildings (especially ‘productive’ vs. ‘nonproductive’). For privacy issues the details about the damages reported by each affected building are not publicly available. However, partial information is available for the 2009 ‘Aquilano’ event. For this earthquake, the distribution of damages severity across types of buildings is indeed uniform. Furthermore, disruption to economic activity might arise even if productive buildings are not directly affected. (Roads might be damaged, Internet connection might be interrupted, etc.)

  12. The distance is calculated as an unweighted average of the distance of each municipality in the province from the epicenter.

  13. Although we have been able to construct the longest time series of provincial GDP growth available at the moment for Italy, the panel structure still contains a large N and a small T. Therefore, typical asymptotic properties of fixed effect panel data model estimators (such as within-the-group) applies in this case.

  14. For the period 1986–1995 we use the estimates released by the statistical office of the ‘Taglicarne Institute’ as in Acconcia et al. (2011).

  15. On average a province is composed by 73 municipalities.

  16. Allowing for more lags in the model does not change our results.

  17. The same evidence applies to robustness checks.

  18. Stock–Yogo weak ID test critical values for single endogenous regressor are between 22 and 5 according to the maximal IV size.

  19. In our models the joint null hypothesis is that the instruments are valid, i.e., uncorrelated with the error term.

  20. Under the null, the Sargan–Hansen test statistic is distributed as Chi-squared in the number of (\(L-K\)) overidentifying restrictions, where \(L=\) instruments and \(K=\) endogenous regressors. A rejection casts doubt on the validity of the instruments.

  21. The five municipalities below level V were Bastia Umbra, Fratta Todina, Monte Castello di Vibio, Paciano, and Scheggia e Pascelupo.

  22. The list of municipalities in the province of Rome involved in the 1997 event is as follows (Mercalli ranks and population in brackets): Ciciliano (V—1,105), Mentana (V—34,326), Montelibretti (V—4,881), Nemi (VI—1,702), Ponzano Romano (V—1,013), Riano (V—6,148), Riofreddo (VI—770), and Roccagiovine (V—293).


Download references

Author information

Authors and Affiliations


Corresponding author

Correspondence to Riccardo Trezzi.

Additional information

We are grateful to Giancarlo Corsetti, Pontus Rendahl, Massimiliano Stucchi, and to the participants of the presentations at the University of Cambridge, at University of Exeter, and at University of Ferrara, for helpful comments and suggestions. We are also particularly grateful to the Italian National Institute of Geophysics and Volcanology (INGV) for providing the requested data. All errors and omissions remain ours. First draft: September 2014. This version: May 2017. Disclaimer: The views expressed in this paper are those of the authors and do not necessarily reflect those of the Board of Governors of the Federal Reserve System and SOSE SpA.


Appendix A: The Mercalli scale—definitions

  • I Instrumental People: Not felt except by a very few people under exceptionally favourable circumstances.

  • II Weak People: Felt by persons at rest, on upper floors or favourably placed.

  • III Slight People: Felt indoors, hanging objects may swing, vibration similar to passing of light trucks, duration may be estimated, may not be recognised as an earthquake.

  • IV Moderate People: Generally noticed indoors but not outside. Light sleepers may be awakened. Vibration may be likened to the passing of heavy traffic, or to the jolt of a heavy object falling or striking the building. Fittings: Doors and windows rattle. Glassware and crockery rattle. Liquids in open vessels may be slightly disturbed. Standing motorcars may rock. Structures: Walls and frames of buildings, and partitions and suspended ceilings in commercial buildings, may be heard to creak.

  • V Rather strong People: Generally felt outside, and by almost everyone indoors. Most sleepers awakened. A few people alarmed. Fittings: Small unstable objects are displaced or upset. Some glassware and crockery may be broken. Hanging pictures knock against the wall. Open doors may swing. Cupboard doors secured by magnetic catches may open. Pendulum clocks stop, start, or change rate. Structures: Some windows Type I cracked. A few earthenware toilet fixtures cracked.

  • VI Strong People: Felt by all. People and animals alarmed. Many run outside. Difficulty experienced in walking steadily. Fittings: Objects fall from shelves. Pictures fall from walls. Some furniture moved on smooth floors, some unsecured free-standing fireplaces moved. Glassware and crockery broken. Very unstable furniture overturned. Small church and school bells ring. Appliances move on bench or table tops. Filing cabinets or “easy glide” drawers may open (or shut). Structures: Slight damage to Buildings Type I. Some stucco or cement plaster falls. Windows Type I broken. Damage to a few weak domestic chimneys, some may fall. Environment: Trees and bushes shake, or are heard to rustle. Loose material may be dislodged from sloping ground, e.g. existing slides, talus slopes, shingle slides.

  • VII Very Strong People: General alarm. Difficulty experienced in standing. Noticed by motorcar drivers who may stop. Fittings: Large bells ring. Furniture moves on smooth floors, may move on carpeted floors. Substantial damage to fragile contents of buildings. Structures: Unreinforced stone and brick walls cracked. Buildings Type I cracked with some minor masonry falls. A few instances of damage to Buildings Type II. Unbraced parapets, unbraced brick gables, and architectural ornaments fall. Roofing tiles, especially ridge tiles may be dislodged. Many unreinforced domestic chimneys damaged, often falling from roof-line. Water tanks Type I burst. A few instances of damage to brick veneers and plaster or cement-based linings. Unrestrained water cylinders (water tanks Type II) may move and leak. Some windows Type II cracked. Suspended ceilings damaged. Environment: Water made turbid by stirred up mud. Small slides such as falls of sand and gravel banks, and small rock-falls from steep slopes and cuttings. Instances of settlement of unconsolidated or wet, or weak soils. Some fine cracks appear in sloping ground. A few instances of liquefaction (i.e. small water and sand ejections).

  • VIII Destructive People: Alarm may approach panic. Steering of motorcars greatly affected. Structures: Buildings Type I heavily damaged, some collapse. Buildings Type II damaged, some with partial collapse. Buildings Type III damaged in some cases. A few instances of damage to Structures Type IV. Monuments and pre-1976 elevated tanks and factory stacks twisted or brought down. Some pre-1965 infill masonry panels damaged. A few post-1980 brick veneers damaged. Decayed timber piles of houses damaged. Houses not secured to foundations may move. Most unreinforced domestic chimneys damaged, some below roof-line, many brought down. Environment: Cracks appear on steep slopes and in wet ground. Small to moderate slides in roadside cuttings and unsupported excavations. Small water and sand ejections and localised lateral spreading adjacent to streams, canals, lakes, etc.

  • IX Violent Structures: Many Buildings Type I destroyed. Buildings Type II heavily damaged, some collapse. Buildings Type III damaged, some with partial collapse. Structures Type IV damaged in some cases, some with flexible frames seriously damaged. Damage or permanent distortion to some Structures Type V. Houses not secured to foundations shifted off. Brick veneers fall and expose frames. Environment: Cracking of ground conspicuous. Landsliding general on steep slopes. Liquefaction effects intensified and more widespread, with large lateral spreading and flow sliding adjacent to streams, canals, lakes, etc.

  • X Intense Structures: Most Buildings Type I destroyed. Many Buildings Type II destroyed. Buildings Type III heavily damaged, some collapse. Structures Type IV damaged, some with partial collapse. Structures Type V moderately damaged, but few partial collapses. A few instances of damage to Structures Type VI. Some well-built timber buildings moderately damaged (excluding damage from falling chimneys). Environment: Landsliding very widespread in susceptible terrain, with very large rock masses displaced on steep slopes. Landslide dams may be formed. Liquefaction effects widespread and severe.

  • XI Extreme Structures: Most Buildings Type II destroyed. Many Buildings Type III destroyed. Structures Type IV heavily damaged, some collapse. Structures Type V damaged, some with partial collapse. Structures Type VI suffer minor damage, a few moderately damaged.

  • XII Catastrophic Structures: Most Buildings Type III destroyed. Structures Type IV heavily damaged, some collapse. Structures Type V damaged, some with partial collapse. Structures Type VI suffer minor damage, a few moderately damaged.

Construction types.Buildings Type I: Buildings with low standard of workmanship, poor mortar, or constructed of weak materials like mud brick or rammed earth. Soft story structures (e.g., shops) made of masonry, weak reinforced concrete, or composite materials (e.g., some walls timber, some brick) not well tied together. Masonry buildings otherwise conforming to buildings Types I to III, but also having heavy unreinforced masonry towers. (Buildings constructed entirely of timber must be of extremely low quality and are Type I.). Buildings Type II: Buildings of ordinary workmanship, with mortar of average quality. No extreme weakness, such as inadequate bonding of the corners, but neither designed nor reinforced to resist lateral forces. Such buildings not having heavy unreinforced masonry towers. Buildings Type III: Reinforced masonry or concrete buildings of good workmanship and with sound mortar, but not formally designed to resist earthquake forces. Structures Type IV: Buildings and bridges designed and built to resist earthquakes to normal use standards, i.e., no special collapse- or damage-limiting measures taken (mid-1930s to c. 1970 for concrete and to c. 1980 for other materials). Structures Type V: Buildings and bridges, designed and built to normal use standards, i.e., no special damage-limiting measures taken, other than code requirements, dating from since c. 1970 for concrete and c. 1980 for other materials. Structures Type VI: Structures, dating from c. 1980, with well-defined foundation behavior, which have been specially designed for minimal damage, e.g., seismically isolated emergency facilities, some structures with dangerous or high contents, or new-generation low-damage structures. Windows.Type I: Large display windows, especially shop windows. Type II: Ordinary sash or casement windows. Water tanks.Type I: External, stand-mounted, corrugated iron tanks. Type II: Domestic hot-water cylinders unrestrained except by supply and delivery pipes.

Appendix B: List and definition of control variables

Population: total number of residents at 31 December of each year. Source: ISTAT. Population65: share of population older than 65 years resident at 31 December of each year. Source: ISTAT.Population85: share of population older than 85 years resident at 31 December of each year. Source: ISTAT. Index of young dependency: ratio between the number of people younger than 14 years and people in working age (14–65 years old) at 31 December of each year. Source: ISTAT. Index of senior dependency: ratio between the number of people older than 65 years and people in working age (14–65 years old) at 31 December of each year. Source: ISTAT.

Appendix C: Summary statistics

See Tables 4, 5.

Table 4 List and geophysical characteristics of earthquakes
Table 5 Summary statistics

Appendix D: Robustness checks—tables

See Tables 6, 7, 8, 9, 10, 11.

Table 6 Events with high magnitude—dependent variable: output
Table 7 Events with high magnitude—dependent variable: employment
Table 8 Weighted variables—dependent variable: output
Table 9 Weighted variables—dependent variable: employment
Table 10 Excluding Mercalli lower than V—dependent variable: output
Table 11 Excluding Mercalli lower than V—dependent variable: employment

Rights and permissions

Reprints and permissions

About this article

Check for updates. Verify currency and authenticity via CrossMark

Cite this article

Porcelli, F., Trezzi, R. The impact of earthquakes on economic activity: evidence from Italy. Empir Econ 56, 1167–1206 (2019).

Download citation

  • Received:

  • Accepted:

  • Published:

  • Issue Date:

  • DOI:


JEL Classification