پیامدهای مالی از حملات طوفان در کارائیب

This paper empirically traces the fiscal impacts of hurricane strikes. To this end, a hurricane damage index is derived from a physical wind field model for a panel of Caribbean countries over 36 years. Results, based on panel VAR and impulse response functions analysis, show that, overall, hurricane strikes exert a short-term impact. Indeed, the study finds that the response of government spending is positive and significant while public investment, debt and tax revenue do not appear to respond (significantly) to hurricane strikes. Moreover, the study finds that Governments respond to hurricane strikes by engaging in short term deficit financing.

Natural disasters are generally associated with considerable economic losses, often causing substantial and sustained disruption to normal economic activity. Particularly vulnerable in this regard are small countries, whose limited budgetary capacity prevents them from establishing sufficient financial reserves to such a relatively large negative shock. Additionally, the high debt level of most small economies limits their ability to access credit in the aftermath of a natural disaster, high transaction costs associated with the relatively small market limits access to private catastrophe insurance, and international assistance is often too little and comes too late. For example, when Hurricane Ivan struck the island of Grenada in 2004 causing over US $800 million in damages, the country was no longer able to finance its public service bill and was forced to introduce a number of revenue enhancing measures and delay efforts of recovery and reconstruction in order to deal with the immediate problem of the fiscal shortfall, thus likely further amplifying the long term effects of the hurricane. However, despite the arguable policy relevance of understanding the immediate fiscal impact of natural disasters, a review of the literature reveals that there are only a handful of studies that have addressed the issue, treating natural disasters as homogenous and using potentially measurement prone assessments of ex-post damages. To be sure, much of the literature on the effect of natural disasters has focused on its impact on gross domestic product (Cavallo et al., 2010, Fomby et al., forthcoming, Hochrainer, 2009 and Loayza et al., 2009Noy, 2009, Noy and Nualsri, 2011, Raddatz, 2007, Skidmore and Toya, 2002 and Strobl, 20111; among others). The evidence provided by these studies is generally mixed; indeed, while some studies find that natural disaster exert an adverse effect on output dynamics (see Hochrainer, 2009, Noy, 2009 and Raddatz, 2007) others suggest a positive growth impact of natural disaster (see Skidmore and Toya, 2002 and Loayza et al., 2009). The fiscal implications of natural disasters, on the other hand, have been under-investigated notwithstanding the fact that natural disasters can have severe implications for public sector finances and provide a major obstacle to recovery (International Monetary Fund, 2009, Inter American Development Bank, 2009 and World Bank, 2001). As a matter of fact, to the best of our knowledge only three studies (Lis and Nickel, 2009, Melecky and Raddatz, 2011 and Noy and Nualsri, 2011) have examined the fiscal effects of natural disasters.2Noy and Nualsri (2011) construct a country level panel data set and find that the fiscal impact of natural disasters depend on the country-specific macroeconomic dynamics occurring in the aftermath of the disaster shocks. For example, while developed countries are characterized by a counter-cyclical fiscal reaction, developing countries respond via procyclical decreasing spending and increasing revenues. Also, Lis and Nickel (2009), similarly in a cross-country panel data context, discover that the negative budgetary impact of extreme weather events can be up to 1.1% of GDP, where the effect is generally larger for developing countries. Melecky and Raddatz (2011), using annual data for a sample high and middle-income countries over 1975–2008 and panel VAR, find that government expenditure increases while revenue experience an insignificant change to climate shocks; moreover, the results show that the budget deficit worsens following the climate shocks. Although much of the studies reviewed, above, use econometrics as their tool for capturing the fiscal effects of natural disasters, it is worth noting that alternative methods, such as the CatSim model (see Hochrainer and Mechler, 2009), have been used to calculate the financial resource gap, and obtain country-level estimates for the most hazard-exposed countries. For example, Mechler et al. (2010) suggest that many highly exposed countries are highly financial vulnerable and experience a resource gap. This present paper is an attempt to add to the limited number of studies that have looked at the public finance implications of natural disasters. In particular, the paper looks at the specific case of hurricane strikes in the Caribbean. Hurricane strikes in the Caribbean are arguably an ideal case study for the question on hand, as hurricanes tend to be frequent, albeit still unpredictable events, and Caribbean countries tend to be small and heavily indebted. However, our contribution to the literature is also with respect to other aspects. Firstly, rather than using measurement prone ex post loss data or simple incidence dummies as proxies for these disaster events – as was done in the literature cited above – we here employ ex ante data on the nature of the striking hurricanes in conjunction with a physical wind field model to develop a proxy of potential damages incurred that will arguably provide a much more accurate measure of large exogenous negative shocks to a small economy. One may want to note in this regard that this approach of using pre-defined information of a natural disaster event to proxy its impact has recently not only gained popularity in academic circles,3 but also appears to have generated interest among policy makers. For instance, the recently established Caribbean Catastrophe Risk Insurance Facility, set up by the World Bank to deal with short term windfalls in public financing, now uses the local maximum wind speed derived from a hurricane wind field model to determine the amount of funds to disperse in the case of a hurricane strike for participating countries. Secondly, we employ our hurricane destruction index within a panel VAR framework that allows us to include other exogenous shocks. The main findings of the paper is that hurricane strikes in the Caribbean countries lead to short run increase in government spending, a worsening of the budget deficit-but an insignificant (in statistical term) change in public spending, tax revenue and debt. These findings are robust to various specification tests. The remainder of the paper is organized as follows. The next section describes the construction of our potential hurricane destruction index. Section 3 describes the data and introduces the econometric methodology. Section 4 presents and discusses the estimation results. Section 5 concludes.

This study analyzed the dynamic effects of severe hurricanes on the main fiscal variables of a panel of Caribbean countries over the period 1970–2006. Indeed, using a physically derived hurricane potential destruction index, based on actual hurricane track data, and applying a panel VAR methodology the paper measures and examine the response of government spending, public investment, debt, tax revenue and budget balance to hurricane Saffir–Simpson Scale 3 shocks. The study finds that the overall impact of the hurricane strikes have fiscal implications, at least in the short run. More specifically, we discover that the response of government spending is positive and significant following the shock, its effect lasting up to two years. Public investment, debt and tax revenue do not appear to respond significantly (in statistical terms) to hurricane shocks. Finally, the evidence shows that the impact of hurricane strikes on the budget balance is negative and significant. Taken overall, this study suggests that Caribbean Governments respond to hurricane strikes by engaging in short-term deficit financing. This result provides some justification for participating in the Caribbean Risk Insurance Facility which was specifically designed to be able to partially finance any public sector shortfalls. This study could be extended in several ways. Subject to data availability it would be fruitful to include other variables such as insurance, remittances, relief aid in the model. The data used in the current study is based on annual data. However, high frequency data (e.g. monthly or quarterly data) will be best suited to investigate the fiscal impact of natural disasters, if these become available over a long enough time period for the region. For instance, it is precisely the very immediate fiscal response that the Caribbean Risk Insurance Facility is intended address. Finally, one should note that our potential damages due to hurricanes are measured in terms of winds only on the grounds that all types of damages are related to the wind speed of a hurricane. However, taking account directly of flooding and storm surge directly, which play a large part in damages, may provide further precision to the analysis. Related to this actual reliable data on direct and indirect losses would allow us to verify our hurricane loss model and provide further confidence in our results.