Skip to main content

A trend analysis of normalized insured damage from natural disasters

Abstract

As the world becomes wealthier over time, inflation-adjusted insured damages from natural disasters go up as well. This article analyzes whether there is still a significant upward trend once insured natural disaster loss has been normalized. By scaling up loss from past disasters, normalization adjusts for the fact that a hazard event of equal strength will typically cause more damage nowadays than in past years because of wealth accumulation over time. A trend analysis of normalized insured damage from natural disasters is not only of interest to the insurance industry, but can potentially be useful for attempts at detecting whether there has been an increase in the frequency and/or intensity of natural hazards, whether caused by natural climate variability or anthropogenic climate change. We analyze trends at the global level over the period 1990 to 2008, over the period 1980 to 2008 for West Germany and 1973 to 2008 for the United States. We find no significant trends at the global level, but we detect statistically significant upward trends in normalized insured losses from all non-geophysical disasters as well as from certain specific disaster types in the United States and West Germany.

This is a preview of subscription content, access via your institution.

Fig. 1
Fig. 2
Fig. 3
Fig. 4
Fig. 5
Fig. 6
Fig. 7
Fig. 8
Fig. 9

Notes

  1. 1.

    Hazards are events triggered by natural forces. They will turn into natural disasters if people are exposed to the hazard and are not resilient to fully absorbing the impact without damage to life or property (Schwab et al. 2007).

  2. 2.

    One has to keep in mind that the NatCatSERVICE data base was set up as an insurance industry-related loss data base that is organized according to the most significant hazardous impact involved with a disastrous event. Hence the disaster subtype is nothing else than a significant type of hazard that has caused a significant proportion of the loss. But any subtype given does not exclude another subtype to be additionally involved while the event occurred. For instance, among the convective events associated with a positive loss there have been 185 events reported where tornados have caused significant insured loss. Definitely, this does not exclude tornados occurring also with some of the 213 hailstorm events that have been reported to have caused losses from hail. Nor does it exclude tornados occurring with the 765 reported tempest storm events. Hence, the subtype tornado does not comprise all the tornado events occurred, but those where tornado was the most significant type of hazard produced by the thunderstorm cell. In order to include comprehensively all the tornado losses, one would have to integrate over all the convective hazards (i.e. flash flood, hailstorm, lightning, tempest storm, tornado), but will at the same time integrate all losses from convective events. Another example of disaster subtypes that often are linked to each other is the ensemble of drought, heat wave and subsidence

  3. 3.

    GDP might also be positively affected by large disasters as repair and reconstruction increase GDP.

  4. 4.

    It has also changed over time (see D’Adda and Scorcu 2003). Nevertheless Krugman (1992: 54f.) concludes that “there is a remarkable constancy of the capital-output ratio across countries; there is also a fairly stable capital-output ratio in advanced nations. These constancies have been well known for a long time and were in fact at the heart of the famous Solow conclusion that technological change, not capital accumulation, is the source of most growth.”

  5. 5.

    Furthermore, comparability of insured losses over time and space could be limited by differences and changes in insurance conditions which affect the insured risk and the size of losses, such as maximum coverage and deductibles (Changnon 2009a; Botzen et al. 2010).

  6. 6.

    For the US, due to lack of data no similar analysis could be undertaken on a market-wide basis. Most likely, if data had been available such an analysis would have shown a lower correlation because of market cycles and premia adjustments after large disasters (Munich Re, personal communication).

  7. 7.

    Alternatively, one can keep all values in USD and then apply the US GDP deflator for normalization purposes. The two approaches lead to practically identical results.

  8. 8.

    Since we use GDP at different levels of spatial resolution for calculating insurance penetration on the one hand and for wealth adjustment on the other for West Germany and the US, GDP does not drop out of Eq. (3). As a consequence, Eqs. (2) and (3) rather than Eq. (4) are used for normalizing insured losses in Germany and the US.

  9. 9.

    This will inevitably create some (small) bias of unknown direction. To test the robustness of our results, we assumed as a shortcut that the share of Western premia was equal to the share of total disaster damage in the entire post-1990 period. Thus estimating, admittedly rather crudely, Western premia and employing these in the normalization leads to qualitatively similar results. In fact, the marginally insignificant upward trend in normalized damage from all storms becomes significant at the 5% level with this alternative premia measure.

  10. 10.

    Personal income is defined as the income received by all persons from all sources before the deduction of personal taxes (BEA 2010) and reported in current USD and converted into constant values with the US GDP deflator. Results are almost identical if we use GDP data at the state level from the same source instead.

  11. 11.

    To cover as many country-years as possible, we extrapolated data on insurance penetration for some missing years such that the analysis is based on a balanced panel of countries. The results are, however, fully robust if only countries with full time series in the original insurance penetration data are included.

  12. 12.

    While landslides are generally geo-physical events, they are regularly triggered by sustained wet conditions in a mountainous region. We dropped the landslides, which were classified as a geo-physical event in the database, but kept those that were recorded as hydrological events. However, none of the former and only five events of the latter resulted in a known insured loss. Similarly, a subsidence might be driven by droughts as a consequence of which moist and welled clay soils lose water and compact. The inclusion of 19 subsidence events with a positive known insured loss in our global sample does not alter the results. For the US and Germany, there are no such events with a positive insured loss.

  13. 13.

    We show no graphs for developing countries separately as insurance penetration is very low and insurance coverage is typically restricted to major cities in middle- and upper middle-income developing countries.

  14. 14.

    Precipitation-related events encompass both floods and wet mass movements.

  15. 15.

    See, however, Schiesser (2003) who reports evidence on increased frequency of strong hailstorm events in Switzerland after 1980 and, similarly, Kunz, Sander and Kottmeier (2009) for the South-West of Germany. Also, Botzen, Bouwer and van den Bergh (2010) find a strong correlation between minimum temperatures (see, similarly, Dessens 1995) as well as precipitation and total agricultural hailstorm damage in the Netherlands. Since there has been higher precipitation and higher minimum temperatures in Northern latitudes, an increase in the frequency and/or intensity of extreme hailstorm events is likely.

References

  1. Barredo JI (2009) Normalised flood losses in Europe: 1970–2006. Nat Hazards Earth Syst Sci 9:97–104

    Article  Google Scholar 

  2. BEA (2010) Regional economic accounts, available at: http://www.bea.gov/regional/index.htm

  3. Botzen WJW, Bouwer LM, van den Bergh JJCJM (2010) Climate change and hailstorm damage: empirical evidence and implications for agriculture and insurance. Resource Energy Econ 32:341–362

    Article  Google Scholar 

  4. Bouwer LM (2011) Have past disaster losses increased due to anthropogenic climate change? Bull Am Meteorol Soc (forthcoming). doi:10.1175/2010BAMS3092.1

  5. Brooks HE, Doswell CA (2001) Normalized damage from major tornados in the United States: 1890–1999. Weather Forecast 16:168–176

    Article  Google Scholar 

  6. Cambridge Econometrics (2010) European regional data. Cambridge, UK

  7. Changnon SA (2001) Damaging thunderstorm activity in the United States. Bull Am Meteorol Soc 82(4):597–608

    Article  Google Scholar 

  8. Changnon SA (2003) Shifting economic impacts from weather extremes in the United States: a result of societal changes, not global warming. Nat Hazards 29:273–290

    Article  Google Scholar 

  9. Changnon SA (2007) Catastrophic winter storms: an escalating problem. Clim Chang 84:131–139

    Article  Google Scholar 

  10. Changnon SA (2009a) Temporal and spatial distributions of wind storm damages in the United States. Clim Chang 94:473–482

    Article  Google Scholar 

  11. Changnon SA (2009b) Increasing major hail losses in the U.S. Clim Chang 96:161–166

    Article  Google Scholar 

  12. Changnon SA, Changnon JM (1992) Storm catastrophes in the United States. Nat Hazard 6:93–107

    Article  Google Scholar 

  13. Changnon SA, Pielke RA, Changnon D, Sylves RT, Pulwarty R (2000) Human factors explain the increased losses from weather and climate extremes. Bull Am Meteorol Soc 81(3):437–442

    Article  Google Scholar 

  14. Crompton RP, McAneney KJ (2008) Normalised Australian insured losses from meteorological hazards: 1967–2006. Environ Sci Pol pp. 371–378

  15. D’Adda C, Scorcu AE (2003) On the time stability of the output-capital ratio. Econ Model 20:1175–1189

    Article  Google Scholar 

  16. Dessens J (1995) Severe convective weather in the context of a nighttime global warming. Geophys Res Lett 22(10):1241–1244

    Article  Google Scholar 

  17. Downton M, Miller JZB, Pielke RA Jr (2005) Reanalysis of U.S. National Weather Service flood loss database. Nat Hazards Rev 6:13–22

    Article  Google Scholar 

  18. Elsner JP, Kossin JP, Jagger TH (2008) The increasing intensity of the strongest tropical cyclones. Nature 455:92–95

    Article  Google Scholar 

  19. IPCC (2007a) Climate change 2007: the physical science basis. Cambridge University Press, New York

    Google Scholar 

  20. IPCC (2007b) Climate change 2007: impacts, adaptation, and vulnerability. Cambridge University Press, New York

    Google Scholar 

  21. Karl TR, Meehl GA, Miller CD, Hassol SJ, Waple AM, Murray WL (eds) (2008) Weather and climate extremes in a changing climate. Regions of focus: North America, Hawaii, Caribbean, and U.S. Pacific Islands, Report by the US climate change science program and the subcommittee on global change research, synthesis and assessment product 3.3. http://downloads.climatescience.gov/sap/sap3-3/sap3-3-final-all.pdf

  22. Knutson TR, McBride JL, Chan J, Emanuel K, Holland G, Landsea C, Held I, Kossin JP, Srivastava AK, Sugi M (2010) Tropical cyclones and climate change. Nat Geosci 3:157–163

    Article  Google Scholar 

  23. Krugman P (1992) Comment. NBER Macroecon Annu 7:54–56

    Article  Google Scholar 

  24. Kunz M, Sander J, Kottmeier Ch (2009) Recent trends of thunderstorm and hailstorm frequency and their relation to atmospheric characteristics in southwest Germany. Int J Climatol 29:2283–2297

    Article  Google Scholar 

  25. Miller S, Muir-Wood R, Boissonade A (2008) An exploration of trends in normalized weather-related catastrophe losses. In: Diaz HF, Murnane RJ (eds) Climate extremes and society. Cambridge University Press, New York, pp 225–247

    Chapter  Google Scholar 

  26. Min S-K, Zhang X, Zwiers FW, Hegerl GC (2011) Human contribution to more-intense precipitation extremes. Nature 470:378–381

    Article  Google Scholar 

  27. Neumayer E, Barthel F (2011) Normalizing economic loss from natural disasters: a global analysis. Global Environ Change 21(1):13–24. doi:10.1016/j.gloenvcha.2010.10.004

    Article  Google Scholar 

  28. NHGIS (2010) Census of population and housing 1970–2010, available at: http://data.nhgis.org/nhgis/

  29. Nordhaus WD (2010) The economics of hurricanes and implications of global warming. Clim Chang Econ 1:1–20

    Article  Google Scholar 

  30. Pall P, Aina T, Stone DA, Stott PA, Nozawa T, Hilberts AGJ, Lohmann D, Allen MR (2011) Anthropogenic greenhouse gas contribution to flood risk in England and Wales in autumn 2000. Nature 470:382–385

    Article  Google Scholar 

  31. Peterson TC, Zhang X, Brunet-India M, Vázquez-Aguirre JL (2008) Changes in North American extremes derived from daily weather data. J Geophys Res 113:D07113

    Article  Google Scholar 

  32. Pielke RA Jr., Landsea CW (1998) Normalized hurricane damages in the United States: 1925–1995. Weather Forecast Sept. 1998, pp. 621–631

  33. Pielke RA Jr, Landsea CW, Musulin RT, Downton M (1999) Evaluation of catastrophe models using a normalized historical record. J Insur Regul 18(2):177–194

    Google Scholar 

  34. Pielke RA Jr, Rubiera J, Landsea C, Fernández ML, Klein R (2003) Hurricane vulnerability in Latin America and the Caribbean: normalized damages and loss potentials. Nat Hazards Rev 4(3):101–114

    Article  Google Scholar 

  35. Pielke RA Jr, Gratz J, Landsea CW, Collins D, Saunders MA, Musulin R (2008) Normalized hurricane damages in the United States: 1900–2005. Nat Hazards Rev 9(1):29–42

    Article  Google Scholar 

  36. Raghavan S, Rajseh S (2003) Trends in tropical cyclone impact: a study in Andhra Pradesh, India. Am Meteorol Soc 84:635–644

    Article  Google Scholar 

  37. Schiesser, H-H (2003) Hagel. In: Extremereignisse und Klimaänderung. Bern: Organe consultatif sur les changements climatiques (OcCC). pp. 65–68. Bern. http://www.proclim.ch/4dcgi/occc/fr/Report?859

  38. Schmidt S, Kemfert C, Höppe P (2009) Tropical cyclone losses in the USA and the impact of climate change—A trend analysis based on data from a new approach to adjusting storm losses. Environ Impact Assess Rev 29:359–369

    Article  Google Scholar 

  39. Schwab AK, Eschelbach K, Brower DJ (2007) Hazard mitigation and preparedness. Wiley, Hoboken

    Google Scholar 

  40. Trapp RJ, Diffenbaugh NS, Brooks HE, Baldwin ME, Robinson ED, Pal JS (2007) Changes in severe thunderstorm environment frequency during the 21st century caused by anthropogenically enhanced global radiative forcing. Proc Natl Acad Sci U S A 104:19719–19723

    Article  Google Scholar 

  41. Trapp RJ, Diffenbaugh NS, Gluhovsky A (2009) Transient response of severe thunderstorm forcing to elevated greenhouse gas concentrations. Geophys Res Lett 36:L01703

    Article  Google Scholar 

  42. UNCTAD (2005) Trade and development aspects of insurance services and regulatory frameworks. Geneva: UNCTAD; available at http://www.unctad.org/en/docs/ditctncd200515_en.pdf

  43. US Census Bureau (2010a) Population estimates: housing units, available at: http://www.census.gov/popest/housing/

  44. US Census Bureau (2010b) Historical census of housing tables: home values, available at: http://www.census.gov/hhes/www/housing/census/historic/values.html

  45. Vranes K, Pielke R Jr (2009) Normalized earthquake damage and fatalities in the United States: 1900–2005. Nat Hazards Rev 10(3):84–101

    Article  Google Scholar 

Download references

Author information

Affiliations

Authors

Corresponding author

Correspondence to Fabian Barthel.

Additional information

Equal authorship. The authors acknowledge support from the Munich Re Programme “Evaluating the Economics of Climate Risks & Opportunities in the Insurance Sector” at LSE. All views expressed are our own and do not represent the views of Munich Re. We thank Eberhard Faust, Peter Höppe, Jan Eichner, Nicola Ranger, Lenny Smith and Bob Ward as well as three referees for many helpful comments. All errors are ours.

Rights and permissions

Reprints and Permissions

About this article

Cite this article

Barthel, F., Neumayer, E. A trend analysis of normalized insured damage from natural disasters. Climatic Change 113, 215–237 (2012). https://doi.org/10.1007/s10584-011-0331-2

Download citation

Keywords

  • Gross Domestic Product
  • Tropical Cyclone
  • Disaster Loss
  • Significant Upward Trend
  • Insured Loss