Estimating the economic consequences of climate change is a tricky task for various reasons. The first one is that climate and weather influence, more or less, almost all human activities from leisure to industrial production. Assessing this influence is made difficult by the multiplicity of these human activities. But even when considering only a few activities, the task remains problematic. Indeed, uncertainties are large in all components of the analysis. The future economy that will be impacted by climate change will differ from today's economy, and even small changes in economic development can make a difference in climate change
J.B. Elsner and T.H. Jagger (eds.), Hurricanes and Climate Change, 361
doi: 10.1007/978-0-387-09410-6, © Springer Science + Business Media, LLC 2009
impacts. For example, a reduction in poverty and an improvement in housing quality can reduce vulnerability to climate change. Climate change impacts will therefore vary with levels of economic development as well as with levels and types of climate change.
But future economies are not the only unknown components: future climate is also uncertain. While there is an agreement about the largest patterns of climate change (IPCC, 2007), local changes are still very uncertain. Unfortunately, impacts are mainly at the local scale, and this uncertainty makes it very difficult to investigate climate change impacts.
Also, all economic agents will respond to climate impacts by implementing adaptation strategies, therefore reducing damages. Their ability to do so, however, is difficult to predict, as it depends on the ability to detect a change in climate conditions in due time, to develop technical or institutional responses to this change, and to implement these responses in an efficient way (Hallegatte, 2007a). Past experience shows that detection failure, ill-adaptation, and over-reactions are common in such situations. This makes the efficiency of adaptation dependent on subtle local factors, which are difficult to measure and predict.
Finally, even when impacts can be estimated with some level of confidence, e.g. an impact on agriculture in a region, the indirect effects of this impact on the entire society or economy is very difficult to assess. For instance, if one important sector of the economy is made unprofitable by climate change, the overall impact on the economy depends on complex interaction of factors including the ability of workers to shift to other economic sectors, the ability of investors and entrepreneurs to create rapidly profitable activities in new sectors, and the ability of the government to facilitate the transitions and support the households in difficulty.
In spite of these difficulties, this paper aims at proposing a methodological roadmap to assess sectoral and regional impacts of climate change. To highlight methodological issues, it focuses on one well-studied phenomenon - hurricanes -, on one period of time - the end of this century -, and on one well-studied region, the U.S. Atlantic and Gulf coastline. The paper proposes an assessment of the additional hurricane losses due to climate change over this region and at this time. It summarizes and combines several analyses from different disciplines, namely climatology, engineering and economy, within an interdisciplinary framework that is presented in Fig. 1. The methodology that is described could then be adapted to other regions and other climate change phenomena (e.g. winterstorms, flood, etc.). Most importantly, the present paper emphasizes the main difficulties in the assessment of the economic impacts due to climate change, and highlights issues on which additional research is needed.
The study of hurricane risk in the U.S. is a good candidate to demonstrate such a methodological roadmap because hurricanes represent a real concern in the U.S. and in the rest of the world. Indeed, many recent hurricane seasons have been particularly active, and the 2005 season remains the most active in recorded history, with 28 tropical storms, 15 hurricanes, 7 strong hurricanes (category 3 and higher); Wilma was the most intense hurricane ever observed over the North Atlantic. Some have argued that the level of hurricane activity in the North Atlantic exhibits
inter-decadal oscillations that fully explain the present level of activity (e.g., Land-sea, 2005), while others have suggested that climate change is responsible for the current situation (Emanuel, 2005; Webster et al., 2005). Regardless, the uninterrupted rise in social vulnerability (e.g., Pielke, 2005; Pielke et al., 2007), combined with the possibility that climate change may increase future hurricane intensity and frequency has raised concerns about the management of hurricane risks in the U.S. Investigating hurricane risks in relation to climate change is also important because of the high cost of coastal protection and because of the lifetime of these infrastructures. This long lifetime makes it essential to anticipate future changes in hurricane risks to avoid infrastructure ill-adaptation and sunk costs (see, e.g., Hallegatte, 2006).
In the course of this century, climate change will modify many identified drivers of hurricane activity (see, e.g., Bove et al., 1998; Murnane et al., 2000; Elsner et al., 2000; Elsner et al., 2001; Jagger et al., 2001; Emanuel, 2005): sea surface temperature, vertical wind shear, global circulation, tropopause temperature, El Nino and NAO properties, etc. The following considers only an increase in potential intensity, which summarizes not the full set of drivers but only the thermodynamical drivers of hurricanes. This paper assumes a global 10-percent increase in potential intensity, which corresponds roughly to a 2°C warming of the Atlantic Ocean
(Emanuel, 2005), all other environmental drivers being unchanged. A more detailed analysis accounting for other parameters will be published in a follow-up article.
One major drawback of the present analysis is that uncertainty is discussed but is not quantified, because only one model is used in each component of the analysis. As a consequence, this paper proposes only one estimate of the climate change influence on hurricane losses. To inform public policy and guide adaptation strategies, however, a best guess is insufficient: an uncertainty estimate, or at least an assessment of confidence level, would also be necessary (on this issue, see the IPCC guidelines, Manning et al, 2004, and a discussion in the conclusion of this article). Another drawback is that climate change will occur in many other forms than through shifts in hurricane activity. Hurricanes in the U.S. coastline represent only a part of the climate change impacts in this region and other perils need to be considered in conjunction (e.g., sea level rise). However, the narrow focus here is consistent with the aim of this paper, which is to demonstrate a methodology to estimate climate change damages rather than to provide new estimates of aggregate climate change damages.
The paper is organized following the causal chain from global climate change to economic damages, which is presented in Fig. 1. Section 2 will describe the socioeconomic, emission and climate scenarios that are used in this article. Section 3 shows how to downscale simulations of global climate change, i.e. how to translate climate change information at the pertinent scale of analysis. Section 4 translates the changes in hurricane risk into changes in direct losses caused by hurricanes in the U.S. Atlantic and Gulf coastline. Section 5 translates these changes in direct losses into indirect economic impacts, taking into account macroeconomic feedbacks and limitations in reconstruction capacity, and provides a mean to estimate total macro-economic costs of the change in hurricane activity. Section 6 concludes from a methodological and substantive point of view, and highlights the need for future research.
Was this article helpful?