Experimentality of Markets: Natural Disaster Risk Markets, Reinsurance, and the Sociology of Catastrophe Modelling

该项目围绕社会技术认知实践和风险设备如何创造市场以及这些实践和设备的社会技术风险治理如何进一步塑造市场的问题展开。市场关注的是自然灾害风险的金融市场，在这个市场上交易各种结构性金融产品，即所谓的“保险关联证券”（ILSs），在这个市场上，自然灾害有效地变成了金融经济产品。这个市场的核心是“灾难建模”，即灾难性风险评估的认知实践，其中无数潜在的自然灾难性事件被人为建模并转化为常见的金融风险措施。这种做法是20世纪中期开始的地球科学（地震学、地震工程、气象学等）和保险业的风险评估方法的融合。将其应用于巨灾保险定价中对保险公司大规模自然灾害的财务不确定性进行管理。然而，它主要适用于再保险公司，即保险公司的保险公司，在发生自然灾害造成巨大经济损失时，再保险公司保证保险公司的偿付能力，从而保证保险公司的理赔能力。管理这些风险的核心问题是一个认知问题：地震或飓风等自然灾害将在何时、何地以及有多严重是不可知的。由于自然灾害是高度个体化和罕见的，因此人们正在模拟“想象的”灾难性事件的人工历史，并将其用于预测可能的经济损失。这些建模实践的输出在常见的金融投资组合风险量化中呈现。通过将自然灾害与常见的金融市场行为“联系起来”，金融工程师们从20世纪90年代初开始将其建立在高风险的保险相关证券（如“巨灾债券”）的基础上。它们允许资本市场投资者，而不是再保险公司，为潜在的赔付提供资金，同时从高利率中获利。自那以后，自然灾害的风险越来越多地转移到一般的金融市场体系上——如今，大约20%的巨灾再保险被放在这里，而不是再保险公司传统风险管理实践的“屏蔽”空间。这种通常被称为“保险和资本市场的融合”的转变将如何改变全球灾难性风险管理，以及它对全球系统性金融稳定可能产生的后果，本身就高度不确定。这个基于人种学和访谈的项目试图回答这样一个问题，即ils市场究竟是如何通过这种灾难建模的认知实践实现的。更重要的是，这个项目将灾难建模作为一种“实验”实践，它不一定产生可行的真理，但创造了一个认知空间，一种鲍德里亚的“超现实”形式，ils可以合法地存在。这种形式的实验是基于灾难建模的关系性、分散性和分布式实践，以及使其合法化的社会技术斗争。因此，分析的主要单位将是灾难建模的“实践社区”，包括再保险公司、灾难建模公司、监管机构、保险公司金融公司和评级机构。核心问题将是，这种固有的、持续的实验性是否会破坏市场的认知状态，而市场最终取决于这种认知状态。或者它是否打开了一个“有序”的空间，永久地调整关于自然灾害的金融“本体论”的谈判，作为一种新的风险管理形式，以及在极不确定的市场中对认知风险模型进行社会技术治理。

This project revolves around the question of how socio-technical epistemic practices and devices of risk create markets and how markets are then further shaped by socio-technical risk governance of these practices and devices. The market focused on is the financial market for natural disaster risk in which a variety of structured financial products are traded, so-called 'insurance-linked securities' or ILSs - a market where natural catastrophes effectively are rendered financial economic products. At the centre of this market is 'catastrophe modelling', the epistemic practice of catastrophic risk assessment in which myriads of potential natural catastrophic events are artificially modelled and translated into common financial risk measures. This practice emerged as a confluence of risk assessment methods from earth science (seismology, earthquake engineering, meteorology, etc.) and the insurance industry starting in the mid-20th century. It is applied to manage insurance companies' financial uncertainties of large-scale natural catastrophes in disaster insurance policy pricing. Chiefly, however, it is applied by reinsurance firms, the insurers of insurers, who guarantee the solvency of, and thus the ability of claims payments by, insurance companies in case of huge financial losses due to occurring natural catastrophes.The central problem of managing these risks is an epistemic one: the unknowability of when, where, and how severely a natural catastrophe, such as an earthquake or a hurricane, will unfold. As natural catastrophes are highly individual and infrequent, an artificial history of 'imagined' catastrophic events is being simulated and used for projecting possible financial losses. The outputs of these modelling practices are rendered in common financial portfolio risk quantification. By making natural catastrophes 'connectible' to common financial market practices, financial engineers based on it highly risky insurance-linked securities like 'catastrophe bonds' starting in the early 1990's. They allow capital markets investors, instead of reinsurance companies, to provide the capital for potential pay-outs while profiting on high interest rates. The risk of natural catastrophes has since been increasingly loaded onto the general financial markets system - today about 20% of catastrophe reinsurance is placed here, instead of the 'shielded' space of reinsurance companies' traditional risk management practices. How this shift, generally referred to as the 'convergence of insurance and capital markets', will alter global catastrophic risk management and what consequences it might have for systemic global financial stability is in itself highly uncertain.This ethnography- and interview-based project tries to answer the question of how exactly the ILSs market was enabled by this epistemic practice of catastrophe modelling. More importantly though, this project frames catastrophe modelling as an 'experimental' practice, which does not necessarily produce viable truths, but creates an epistemic space, a form of Baudrillardian 'hyper-reality', in which ILSs legitimately can exist. This form of experimentality is based on the relational, decentralised, and distributed practice of catastrophe modelling and the socio-technical struggles in rendering it legitimate. Thus, the major unit of analysis will be that of 'communities of practice' of catastrophe modelling throughout reinsurers, catastrophe modelling firms, regulators, ILSs financial firms, and rating agencies. The core question will be, whether this inherent and constant experimentality destabilises the epistemic state of the market, on which it ultimately hinges. Or whether it opens up an 'ordered' space of permanently adjusting negotiations about the financial 'ontology' of natural catastrophes as a new form of risk management and socio-technical governance of epistemic risk models in an extremely uncertain market.