Skip to main content
Log in

Did natural disasters affect population density growth in US counties?

  • Original Paper
  • Published:
The Annals of Regional Science Aims and scope Submit manuscript

Abstract

This paper examines the long-run effects of natural disasters on population density growth across US counties during the period of 1960–2000. Detailed data for measuring the number and intensity of three types of major natural disasters (earthquake, tornado, and hurricane) are collected and incorporated into the empirical models. We do not find any significant adverse long-run growth effects of natural disasters. Weak evidence of minor tornadoes being positively correlated to growth is provided. Results also indicate that disasters have negligible indirect effects on county population density growth through impacting the county characteristics.

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.

Institutional subscriptions

Fig. 1
Fig. 2
Fig. 3

Similar content being viewed by others

Notes

  1. As our conceptual model in the appendix will show, it is based on the assumption that consumers are homogeneous and migration across regions is free and costless.

  2. Table 9 in the appendix reports the test statistics. For regressions on growth rates of population density, employment density, and per capita income, all statistics are greater than 372. Table 10 presents OLS results for population density growth for two 20-year periods.

  3. Later in this paper, we will report result from using a different spatial weight matrix.

  4. Changes in county boundaries are documented by Bureau of Census. For more information, see http://www.census.gov/geo/www/tiger/ctychng.html. Most of such changes are that a county is split into two or more counties. We recombine such “split” counties so that counties in the data sample have constant boundaries in our sample.

  5. The website is http://www.ngdc.noaa.gov/nndc/struts/form?t=101650&s=1&d=1. We also checked data available on the website http://earthquake.usgs.gov/research/data/centennial.php. Earthquakes from either source are included in our dataset.

  6. It is available online at http://www4.ncdc.noaa.gov/cgi-win/wwcgi.dll?wwEvent~Storms.

  7. NOAA Coastal Services Center identifies coastal counties which are affected by hurricanes from 1900 through 2000; details are available online at http://maps.csc.noaa.gov/hurricanes/pop.jsp (accessed February 24, 2007) and http://www.aoml.noaa.gov/hrd/hurdat/Data_Storm.html. In a reply to the author’s query, Edward Rappaport at NOAA noted in an email that “we're unaware of any study assessing the meteorological impact (e.g., category) for inland counties.” Missing information for affected inland counties in the dataset may bias our results as often the next county inland is hit just as badly as the coast.

  8. We also tried the measures which weight the number of disasters in the years equally. Qualitatively, the same results are obtained.

  9. We use log of (1 + the disaster measures) in the model and add “1” because the disaster measures equal zero for many counties.

  10. The focus of this paper is on the effects of natural disasters. It is noted that the regression results confirm the well-known facts that counties with nicer weather and western counties attract more people in the past decades. Estimated coefficients for those variables are also statistically significant in income and employment regressions. Results are available from the author upon request.

  11. Data on income inequality in 1959 for several counties are missing, so we have a slightly smaller data sample in this round of regression exercises. We note that missing values for socioeconomic characteristics in year 1960 in several counties can bias the estimates if they are not random. However, we do not have information to evaluate it.

  12. Rappaport and Sachs (2003) use coastal proximity measures (i.e., harbor proximity and the ratio of a county’s shoreline to its total area) to distinguish the two effects. They find that the coastal concentration of economic activity “derives primarily from a productivity effect but also, increasingly, from a quality of life effect.” Shapiro (2006) uses data on wages, rents, and house values to calibrate a model and find that “roughly 60% of employment growth effect of college graduates is due to enhanced productivity growth, the rest being caused by growth in the quality of life.”.

  13. This result of presidential declaration having adverse impacts is contrary to our expectation. Possible reasons include that only those most costly events are declared. Another possibility is that the estimate could be biased. We should also note that the inclusion of presidential declarations in the regression with natural disaster can be potentially problematic because, among other reasons, it is not a perfect measure of relief efforts and there are various determinants of disaster payments (Garrett and Sobel 2003). The detailed results are available upon request from the author.

  14. As the number of counties hit by earthquake or hurricane is very small, we exclude only those affected counties with the highest or lowest measures for the two disasters.

  15. I thank a reviewer for pointing it out.

  16. In the empirical analysis, we focus on population density growth, but not population growth. The main reason is that land areas for some counties in our sample have changed slightly even when we make the county boundaries consistent over the years. Note that change in population density is equal to change in population when the land area is constant between 2 years. In that case, our model is the same as the one developed by Glaeser et al. (1995) and many others in the regional literature.

References

  • Albala-Bertrand JM (1993) Political economy of large natural disasters. Oxford University Press, New York

    Google Scholar 

  • Beeson PE, DeJong DN, Troesken W (2001) Population growth in the U.S. Counties, 1840–1990. Reg Sci Urban Econ 31:669–699

    Article  Google Scholar 

  • Bernknopf RL, Brookshire DS, Thayer MA (1990) Earthquake and volcano hazard notices: an economic evaluation of changes in risk perception. J Environ Econ Manag 18:35–49

    Article  Google Scholar 

  • Borjas GJ (2010) Labor economics, 5th edn. McGraw-Hill, New York

    Google Scholar 

  • Bosker M, Brakman S, Garretsen H, Schramm M (2007) Looking for multiple equilibria when geography matters: German city growth and the WWII shock. J Urban Econ 61:152–169

    Article  Google Scholar 

  • Brakman S, Garretsen H, Schramm M (2004) The strategic bombing of German cities during WWII and its impact on city growth. J Econ Geogr 4:201–218

    Article  Google Scholar 

  • Camerer C, Kunreuther H (1989) Decision processes for low probability events: policy implications. J Policy Anal Manag 8:565–592

    Article  Google Scholar 

  • Cavallo E, Galiani S, Noy I, Pantano J (2013) Catastrophic natural disasters and economic growth. Rev Econ Stat 95:1549–1561

    Article  Google Scholar 

  • Ciccone A, Hall RE (1996) Productivity and density of economic activity. Am Econ Rev 86:54–70

    Google Scholar 

  • Cochrane HC (1975) Natural hazards and their distributive effects. Institute of Behavioral Science, Boulder

    Google Scholar 

  • Concannon P, Brooks H, Doswell III CA (2000) Climatological risk of strong and violent tornadoes in the United States. Second In: Conference on environmental applications. American Meteorological Society

  • Cullen JB, Levitt SD (1999) Crime, urban flight, and the consequences for cities. Rev Econ Stat 81:159–169

    Article  Google Scholar 

  • Davis DR, Weinstein DE (2002) Bones, bombs and break points: the geography of economic activity. Am Econ Rev 92:1269–1289

    Article  Google Scholar 

  • Doswell CA, Edwards R, Thompson RL, Hart JA, Crosbie KC (2006) A simple and flexible method for ranking severe weather events. Weather Forecast 21:939–951

    Article  Google Scholar 

  • Emanuel K (2005) Increasing destructiveness of tropical cyclones over the past 30 years. Nature 2005:686–688

    Article  Google Scholar 

  • Findley S (1994) Does drought increase migration? A study of migration from rural Mali during the 1983–1985 drought. Int Migrat Rev 28:539–553

    Google Scholar 

  • Fitzpatrick PJ (1999) Natural disasters: hurricanes. ABC-CLIO Inc, Santa Barbara

    Google Scholar 

  • Florax RJGM, Folmer H, Rey SJ (2003) Specification searches in spatial econometrics: the relevance of Henry’s methodology. Reg Sci Urban Econ 33:557–579

    Article  Google Scholar 

  • Garrett TA, Sobel RS (2003) The political economy of FEMA disaster payments. Econ Inq 14:496–509

    Article  Google Scholar 

  • Geipel R (1982) Disaster and reconstruction. Allen and Unwin, London

    Google Scholar 

  • Glaeser EL, Shapiro JM (2002) Cities and warfare: the impact of terrorism on urban form. J Urban Econ 51:205–224

    Article  Google Scholar 

  • Glaeser EL, Scheinkman JA, Shleifer A (1995) Economic growth in a cross-section of cities. J Monet Econ 36:117–143

    Article  Google Scholar 

  • Greenwood M (1975) Research on internal migration in the United States: a survey. J Econ Lit 13:397–433

    Google Scholar 

  • Haines MR and the Inter-university Consortium for Political and Social Research. Historical, demographic, economic, and social data: The United States, 1790–2000 [Computer file]. ICPSR02896-v2. Hamilton: Colgate University/Ann Arbor: MI: Inter-university Consortium for Political and Social Research [producers], 2004. Ann Arbor, MI: Inter-university Consortium for Political and Social Research [distributor] (2005)

  • Higgins MJ, Levy D, Young AT (2006) Growth and convergence across the United States: evidence from county-level data. Rev Econ Stat 88:671–681

    Article  Google Scholar 

  • Jones EL (1987) The European miracle: environments, economies, and geopolitics in the history of Europe and Asia. Cambridge University Press, Cambridge

    Google Scholar 

  • Kahn ME (2000) Smog reduction’s impact on California county growth. J Reg Sci 40:565–582

    Article  Google Scholar 

  • Kahn ME (2005) The death toll from natural disasters: the role of income, geography, and institutions. Rev Econ Stat 87:271–284

    Article  Google Scholar 

  • Kelejian HH, Prucha IR (2010) Specification and estimation of spatial autoregressive models with autoregressive and heteroskedastic disturbances. J Econom 157:53–67

    Article  Google Scholar 

  • Knapp T, Graves P (1989) On the role of amenities in models of migration and regional development. J Reg Sci 29:71–87

    Article  Google Scholar 

  • Kunreuther H, Pauly M (2006) Rules rather than discretion: lessons form Hurricane Katrina. Working Paper 12503. National Bureau of Economic Research, Cambridge

  • Leiter AM, Oberhofer H, Raschky PA (2009) Creative disasters? Flooding effects on capital, labour and productivity within European firms. Environ Resour Econ 43:333–350

    Article  Google Scholar 

  • LeSage JP, Pace RK (2009) Introduction to spatial econometrics. CRC Press, Taylor & Francis Group, Boca Raton

    Book  Google Scholar 

  • Loayza N (2004) Comment on: ‘The open society assesses its enemies: shocks, disasters and terrorist attacks’. J Monet Econ 51:1071–1075

    Article  Google Scholar 

  • Magat W, Viscusi WK, Huber J (1987) Risk-dollar tradeoffs, risk perceptions, and consumer behavior. In: Viscusi W, Magat W (eds) Learning about risk. Harvard University Press, Cambridge, MA

  • Marchiori L, Maystadt J-F, Schumacher I (2012) The impact of weather anomalies on migration in Sub-Saharan Africa. J Environ Econ Manag 63:355–374

    Article  Google Scholar 

  • McGranahan D (2007) Natural amenities drive rural population change. Agricultural Economic Report No. 781, Economic Research Service, U.S. Department of Agriculture. http://www.ers.usda.gov/Data/NaturalAmenities. Accessed 8 Jan 2007

  • Mellinger A, Sachs JD, Gallup JL (2000) Climate, coastal proximity, and development. In: Clark GL, Feldman MP, Gertler MS (eds) The oxford handbook of economic geography. Oxford University Press, New York, pp 169–194

    Google Scholar 

  • Miguel E, Roland G (2011) The long-run impact of bombing Vietnam. J Dev Econ 96:1–15

    Article  Google Scholar 

  • NOAA (National Oceanic and Atmospheric Administration) (2006) NOAA’s List of Coastal Counties. http://www.census.gov/geo/landview/lv6help/coastal_cty.pdf. Accessed 18 Dec 2006

  • Noy I (2009) The macroeconomic consequences of disasters. J Dev Econ 88:221–231

    Article  Google Scholar 

  • Rappaport J, Sachs JD (2003) The U.S. as a coastal nation. J Econ Growth 8:5–46

    Article  Google Scholar 

  • Rogers JL, John CAST, Coleman R (2005) Do fertility go up after the Oklahoma City bombing? An analysis of births in metropolitan counties in Oklahoma, 1990–1999. Demography 42:675–692

    Article  Google Scholar 

  • Rossi PH, Wright JD, Wright SR, Weber-Burdin E (1978) Are there long-term effects of American natural disasters? Estimations of effects of floods, hurricanes, and tornados occurring 1960 to 1970 on US counties and census tracts in 1970. Mass Emerg 3:117–132

    Google Scholar 

  • Saldaña-Zorrilla S, Sandberg K (2009) Impact of climate-related disasters on human migration in Mexico: a spatial model. Clim Change 96:97–118

    Article  Google Scholar 

  • Shapiro JM (2006) Smart cities: quality of life, productivity, and the growth effects of human capital. Rev Econ Stat 88:324–335

    Article  Google Scholar 

  • Skidmore M, Toya H (2002) Do natural disasters promote long-run growth? Econ Inq 40:664–687

    Article  Google Scholar 

  • Strobl E (2011) The economic growth impact of hurricanes: evidence from U.S. coastal counties. Rev Econ Stat 93:575–589

    Article  Google Scholar 

  • Tavares J (2004) The open society assesses its enemies: shocks, disasters and terrorist attacks. J Monet Econ 51:1039–1070

    Article  Google Scholar 

  • Thomas DSK, Mitchell JT (2001) Which are the most hazardous states. In: Cutter SL (ed) American hazardscapes: the regionalization of hazards and disasters. Joseph Henry Press, Washington

    Google Scholar 

  • Thompson RL, Vescio MD (1998) The destruction potential index—a method for comparing tornado days. In: Preprints, 19th conference on severe local storms, Minneapolis, MN. American Meteorological Society, pp 280–282

  • Tol R, Leek F (1999) Economic analysis of natural disasters. In: Downing TE, Olsthoorn AJ (eds) Climate, change and risk. Routledge, London

    Google Scholar 

  • Wu J, Gopinath M (2008) What causes spatial variations in economic development in the United States? Am J Agr Econ 90:392–408

    Article  Google Scholar 

  • Xiao Y (2011) Local economic impacts of natural disasters. J Reg Sci 51:804–820

    Article  Google Scholar 

Download references

Acknowledgements

This paper is based on a chapter of my PhD dissertation at Oregon State University. I thank Edward Rappaport for explaining the hurricane data and Jordan Rappaport for sharing his county-level socioeconomic data. Thanks also to Steven Brakman, Munisamy Gopinath, Ingmar R. Prucha, Jordan Rappaport, JunJie Wu, seminar participants at various institutions, three anonymous referees, and the editor for helpful comments and suggestions. Gina Wang provided excellent assistance in assembling the datasets. The research is partially supported by China Institute for WTO Studies, UIBE (Grant #: 13ZXWTO04). The usual disclaimers apply.

Author information

Authors and Affiliations

Authors

Corresponding author

Correspondence to Chunhua Wang.

Appendix

Appendix

1.1 Conceptual framework

Consider an economy composed of a set of counties \( i \in \left\{ {1,2, \ldots , I} \right\} \). Suppose that population in county \( i \) at time \( t \) is represented by \( L_{it} \). Following Ciccone and Hall (1996), we assume that the aggregate production function for county \( i \) is given by

$$ Y_{it} = A_{it} L_{it}^{\gamma } S_{it}^{1 - \gamma } , $$
(A1)

where \( A_{it} \) denotes the level of productivity. \( S_{it} \) denotes total land area. Wage rate for an individual in this county, \( R_{it} \), is the marginal product of labor, thus

$$ R_{it} = \gamma A_{it} L_{it}^{\gamma - 1} S_{it}^{1 - \gamma } . $$
(A2)

A representative individual in county \( i \) at time \( t \) derives utility, \( V_{it} \), from local quality of life and wage rate she received:

$$ V_{it} = \left[ {Q_{it} \left( {\frac{{L_{it} }}{{S_{it} }}} \right)^{ - \alpha } } \right]R_{it} , $$
(A3)

where \( Q_{it} ( {\frac{{L_{it} }}{{S_{it} }}} )^{ - \alpha } \) is an index of local quality of life in which \( Q_{it} \) denotes exogenous environmental amenities such as weather; \( ( {\frac{{L_{it} }}{{S_{it} }}})^{ - \alpha } \) represents congestion effects in this county with \( \alpha > 0 \). This is linked to previous studies, suggesting that amenities play an important role with respect to migration destinations (Knapp and Graves 1989).

Let \( l_{it} \equiv \frac{{L_{it} }}{{S_{it} }} \) denote population density.Footnote 16 Combining (A2) and (A3) gives that

$$ V_{t} = \gamma Q_{it} A_{it} \left( {l_{it} } \right)^{\gamma - 1 - \alpha } . $$
(A4)

Assume that migration across counties is free and costless. In equilibrium, utility levels across counties are equal. Let \( V_{t} \) denote equilibrium level of utility at time period \( t \), then \( V_{t} \equiv V_{it} \) for any county \( i \). Thus, Eq. (A4) implies that,

$$ \log l_{it} = \frac{1}{\gamma - 1 - \alpha }\left[ {\log V_{t} - \log \gamma - \log Q_{it} - \log A_{it} } \right]. $$

Growth rate in population density between 2 years can be expressed as

$$ \log \left( {\frac{{l_{i,t + 1} }}{{l_{it} }}} \right) = \frac{1}{\gamma - 1 - \alpha }\left[ {\log \left( {\frac{{V_{i,t + 1} }}{{V_{it} }}} \right) - \log \left( {\frac{{Q_{i,t + 1} }}{{Q_{it} }}} \right) - \log \left( {\frac{{A_{i,t + 1} }}{{A_{it} }}} \right)} \right]. $$
(A5)

Natural disasters may contribute to the growth of local productivity, \( A_{it} \), by encouraging the adoption of new production technologies by firms through replacing the damaged machines and equipment in the afflicted regions (Tol and Leek 1999). Affected locations also benefit from higher growth of total assets in firms (Leiter et al. 2009) and a massive inflow of capital for rebuilding, and from technological innovations in the construction sector (Albala-Bertrand 1993). On the other hand, disasters could have adverse productivity effects through destroying physical capital stock, lowering marginal product of labor. Thus, the net effects of natural disasters on local productivity growth are not clear in theory.

In addition to productivity effects, the occurrences of natural disaster events also change the level of quality of life, \( Q_{it} \), by directly destroying amenity structures, damaging transportation system, and causing inconvenience to daily life for people in the affected regions. However, some case studies find that natural disasters improve local quality of life through promoting and reinforcing a sense of regional identity among population in the affected region (Geipel 1982). Thus, the net effects on quality of life are also ambiguous.

To describe the dynamics of productivity and quality of life, \( A_{it} \) and \( Q_{it} \) are assumed to change over time and are functions of natural disaster events and other characteristics of the associated county (Glaeser et al. 1995). Formally,

$$ \log \left( {\frac{{A_{i,t + 1} }}{{A_{it} }}} \right) = D_{it}^{'} \delta_{A} + C_{it}^{'} \pi_{A} + v_{i,t + 1} , $$
(A6)

and

$$ \log \left( {\frac{{Q_{i,t + 1} }}{{Q_{it} }}} \right) = D_{it}^{'} \delta_{Q} + C_{it}^{'} \pi_{Q} + \mu_{i,t + 1} , $$
(A7)

where \( D_{it} \) is the vector of measures of natural disasters and \( C_{it} \) is the vector of some other natural, socioeconomic, and demographic characteristics of the associated county. The error terms in the two processes are represented by \( v_{i,t + 1} \) and \( \mu_{i,t + 1} \), respectively. Neither of the two error terms is correlated with natural disasters or county characteristics. It is well documented in the regional literature that locational fundamentals such as weather and coastal proximity have impacts on regional growth. The growth literature also finds substantial evidence on the importance of human capital, political and social characteristics, industrial composition, and many other factors for population growth. Those related variables are grouped in \( C_{it} \). Combining (A5), (A6), and (A7) yields the expression for population density change between two periods:

$$ \log \left( {\frac{{l_{i,t + 1} }}{{l_{it} }}} \right) = \frac{1}{1 + \alpha - \gamma }D_{it}^{'} \left( {\delta_{A} + \delta_{Q} } \right) + \frac{1}{1 + \alpha - \gamma }C_{it}^{'} \left( {\pi_{A} + \pi_{Q} } \right) + u_{i,t + 1} , $$
(A8)

where \( u_{i,t + 1} = \frac{1}{1 + \alpha - \gamma }[ { - \log ( {\frac{{V_{i,t + 1} }}{{V_{it} }}} ) + v_{i,t + 1} + \mu_{i,t + 1} } ] \).

1.2 Appendix tables

Table 9 Lagrange multiplier test statistics
Table 10 The effects of natural disasters on population density growth: OLS results
Table 11 Spatial Hausman test statistics: OLS versus SEM

Rights and permissions

Reprints and permissions

About this article

Check for updates. Verify currency and authenticity via CrossMark

Cite this article

Wang, C. Did natural disasters affect population density growth in US counties?. Ann Reg Sci 62, 21–46 (2019). https://doi.org/10.1007/s00168-018-0878-1

Download citation

  • Received:

  • Accepted:

  • Published:

  • Issue Date:

  • DOI: https://doi.org/10.1007/s00168-018-0878-1

JEL Classification

Navigation