Bibliographic record
Abstract
6.1 Background 6.2 Importance of the insurance market in climate change adaptation 6.3 Scope of risk and other considerations in the New York City area 6.4 Research on climate change, underwriting, and adaptation planning 6.5 Reducing greenhouse gas emissions: new opportunities for the insurance industry 6.6 Conclusions and recommendations This chapter provides an overview of the private insurance industry's key role in developing flexible climate change adaptation pathways by transferring and mitigating risk. It also provides a discussion of climate change–related risk and adaptation from the standpoint of the property and casualty insurance industry, both in general and specific to New York City. This perspective, while not inclusive of all viewpoints, is pertinent to discussions of risk, particularly given the New York City Panel on Climate Change (NPCC)'s risk management approach to Flexible Adaptation Pathways. Insurance industry tools and risk experience, on the one hand, and the work of government adaptation planners and decision makers, on the other, can be mutually supportive. The insurance industry is often referred to as the “canary in the coal mine” in discussions of climate change. Current research indicates that as climate change impacts become evident, both hurricane and nor'easter frequency and intensity will change (see Climate Risk Information (CRI), Appendix A). Any increase in the frequency or intensity of weather-related catastrophic events will increase insurance risk and directly affect property and casualty premiums. Thus, the insurance industry follows closely the scientific research on the likelihood of changes in frequency and intensity of these storms as a result of climate change. Currently, a major driver of natural catastrophe losses globally is weather-related extreme events (Fig. 6.1), with large contributions from North Atlantic tropical cyclones and European winter storms. Together with sea level rise, these storms could result in significantly higher insured and economic losses in the future if, as is expected, climate change results in more intense storms that hit densely populated coastal areas. Insured global catastrophe losses, 1970–2007.Source: Swiss Re Sigma 1/08. Recently, climate-related events, such as extreme flooding in the United Kingdom, heat waves in France, and prolonged drought in the southwestern United States, have also caused concern among insurers. Projected future changes in the underlying climate system will also affect the life and health insurance industries, as changes in precipitation and temperature globally will shift the distribution of infectious diseases. Many insurance companies recognize the scientific consensus that climate change is already occurring and anthropogenic activities, such as combustion of fossil fuels and tropical deforestation, are having a discernible influence on global climate. While it is not possible to attribute individual weather-related catastrophes, such as Hurricane Katrina, to anthropogenically induced climate change, the Intergovernmental Panel on Climate Change (IPCC) and other scientific assessments have identified relationships between climate change and changes in the frequency or severity of extreme events (IPCC, 2007). For New York City, the primary near-term risk from weather-related disasters is coastal flooding from nor'easters, powerful coastal storms occurring during the autumn and winter months. Historical nor'easters have reached intensities comparable to category 1 or 2 hurricanes on the Saffir-Simpson scale. These extratropical cyclones bring hurricane-like conditions: strong winds, storm surge, and beach erosion. Significant damage to New York City and the surrounding metropolitan area has already occurred in the past. The nor'easter of December 1992 demonstrated the susceptibility of the New York City infrastructure to flooding and storm surge. Over $1 billion of damage occurred in New York City, the transit system of lower Manhattan was inundated, and the highest storm surge since modern record keeping was recorded at the Battery. In the future, storm surges from nor'easters and hurricanes will be exacerbated by already ongoing sea level rise. Insurance and reinsurance transfer risk from an individual policyholder or primary insurance provider to a larger risk-sharing community. The insurance company facilitates this transfer of risk among insured individuals in a way that attempts to be equitable and cost-effective for customers while maintaining solvency and shareholder value for the company. Premiums are set to represent an insured's contribution to the overall shared risk; therefore, to be equitable, those with greater risk should pay higher premiums than those who contribute less to the overall risk. The insurance industry studies risk in actuarial terms, meaning that the mathematical and statistical traits of each risk are carefully monitored. The risk absorbed by the insurance industry is reduced by basing the terms and conditions of insurance policies on actuarially derived premiums and deductibles or by limiting coverage based on risk. Furthermore, private insurance companies are not required by law to write in high-risk areas and will cease to offer coverage in an area deemed high risk. On the basis of actuarial principles, a property at higher risk from weather-related events will pay more for insurance than a property located in an area or built with recognizably reduced exposure to weather-related risk. For example, properties located in designated flood plains or hurricane zones generally pay higher premiums than those elsewhere. Thus, increased weather-related risks will lead to higher rates, higher deductibles, and limited coverage, providing incentives for adaptation or risk reducing measures. The insurance company might stipulate measures that could be undertaken to reduce risks and associated premium costs, such as building retrofits, stricter new building codes, and location of structures. Furthermore, the risk landscape will change, as coastlines shift due to sea level rise and the severity and frequency of individual hurricanes and extratropical cyclones fluctuates. As discussed in Chapter 3 and the CRI (Appendix A), climate hazards likely to increase include severe floods, intense precipitation events, and extreme heat. These changing climate hazards and their associated impacts will directly influence insurance risk and premiums. As an integral part of underwriting, insurance companies factor in historical weather and climate patterns in insurance industry catastrophe models. These are computer models, which combine statistics, mathematics, physical science, and economics to assess the financial risk posed by wind, earthquake, flood, and other perils. Past patterns are the basis for assumptions made about the current probabilities of catastrophic events within these models, which use computer-assisted calculations to estimate the losses likely to be sustained by a portfolio of properties due to a catastrophic event. Climate change poses a unique risk-related challenge. Because of the inertia of the climate system and long lifetimes of key greenhouse gases (GHGs), there are multidecadal lags in the climate system. (For a fuller discussion, see Chapter 3 of this volume.) These lags make it challenging to project the timing and magnitude of climate change impacts far into the future. Furthermore, future GHG emissions paths are difficult to predict, yet these emissions scenarios also have a major influence on the magnitude and timing of future climate change impacts. Thus, there is a timing mismatch between insurance policies that are generally written for 1 year (3 years at most) and climate change impacts that are projected with large uncertainty for decades in the future. At this time, insurance companies do not have enough information to form a basis for raising premiums today to reserve for possible future increases in climate change–related risk. Insurance premiums will rise on a sustained basis only after impacts of increased GHGs are clearly in place. In the meantime, valuable time that could be spent implementing adaptation measures may be lost. Therefore, to address this significant lag, local and/or federal governments have a role in promoting adaptation measures. They can do so through stricter building codes, land-use planning regulations, and strengthening of ecosystems in anticipation of future climate change impacts (see Chapter 5 and Climate Protection Levels (CPL) Appendix C). Insurance executives often intone that climate change poses both risks and opportunities for the insurance industry—but the point bears repeating. Balancing the climate hazards discussed in this report are new business opportunities. The need to mitigate GHG emissions and create a “low carbon” economy leads to opportunities to provide new or expanded insurance and risk management products to help deploy technological solutions, such as renewable energy, low carbon fuels, carbon capture and storage (i.e., trapping and sequestering the GHGs from burning fossil fuels), energy efficiency, and sustainable land-use and forestry practices. These measures are relevant to a discussion of adaptation because they will reduce GHG emissions and thus contribute to lessening the potential scope of adaptation actions by helping to avoid the worst impacts of climate change in the long term. In addition, some of these mitigation measures, especially in the land-use and forestry sector, may also themselves contribute to climate change adaptation as well. There are several ways in which the private insurance market can contribute to climate change adaptation. It can: Help maintain long-term insurability and provide incentives for adaptation through risk-based premium pricing; Use insurance risk-evaluation tools (e.g., catastrophe models) to help policy makers and adaptation planners better understand and assess the financial implications of climate change; Encourage or spearhead research aimed at focusing output from the global climate models to be more useful to insurance underwriters and adaptation planners; Support government adaptation efforts; and Provide educational information on climate change–related risks and increase awareness among customers. To survive, insurance providers need to charge accurate, risk-based premiums so that property owners who build or own property in high-hazard areas will bear the costs of their actual risks. Whether private or public, insurance pools that underestimate risks or use subsidies to mask the true cost associated with risk are not sustainable, at least not from an insurance perspective. Public insurance options, such as those currently practiced in the United States, are not always actuarially sound. These underfunded state solutions promote up-front affordability, but they also encourage risky—and potentially expensive—economic activity. There are several examples in the United States of government-run insurance pools that do not set premiums at an actuarially sound, risk-based level, thereby jeopardizing the long-term sustainability of the insurance coverage. Two of them are the State of Florida's Citizens Property Insurance Corporation and the National Flood Insurance Program (NFIP). Although the New York State shows no signs of initiating a state-run insurance pool, the case of Florida illustrates the challenges of maintaining a private insurance market in a shifting risk landscape. In Florida, following 2005, the year of Hurricane Katrina and several other major hurricanes, catastrophe models and other analytical studies indicated that higher premiums were in order. However, Florida regulators, in the belief that they were promoting affordable insurance and fairness to owners of Florida coastal properties, set a cap on the premium rates that could be charged. Insurance companies, deciding that they could not be profitable at the prescribed rates, began pulling out of the market. The state government then relied on a state-funded plan to provide insurance for properties that private carriers would not cover. The public plan, known as the Citizens Property Insurance Corporation (Citizens), is now the largest homeowners’ insurance company in the state, with 1.3 million policies, many of which are located in areas at high risk of hurricanes. The state of Florida is presently $2.3 billion in debt, so paying out claims is likely to be difficult in the event of a major catastrophe, resulting in claims greater than what Citizens can pay out. The state is now weighing options, including raising the premiums on Citizens’ property-holders by as much as 10% a year, to encourage policyholders to switch to private insurance carriers. It appears that some of the Florida coastal property owners have been temporarily shielded from paying to insure for the true risks from major hurricanes.1 Instead of discouraging development of vulnerable coastal areas, the insurance encourages development where the risk is highest. This situation has a considerable economic cost, not to mention threatening fiscal trouble for the state in the event of a large loss. A second example where the government has affected the pricing of the private insurance market, and thus may be failing to signal to property owners the true cost of ownership, is the NFIP, which is in debt because of flood claims from the recent hurricanes. As of 2009, the NFIP was $19.2 billion in debt to the U.S. Treasury. The NFIP has been in place since 1968 to provide flood insurance in locations where private insurance is not available and to decrease federal disaster relief outlays by offering flood insurance at reduced rates. For the first 37 years, the program paid out approximately $10 billion in claims, in total. However, in 2005, claims were $21 billion primarily due to Hurricanes Dennis, Katrina, Rita, and Wilma. Now, Congress is facing the task of reforming the NFIP to manage the debt and keep the program in tact. The concern is that even a restructured NFIP may continue to allow homeowners to purchase insurance at a price that does not include the full extent of flood risk, thus signaling the ongoing and increasing challenge of coastal land-use planning in high-risk flood zones. Some of the standard insurance industry tools for risk evaluation can help climate change adaptation planners, particularly the industry's loss valuation, or catastrophe models. Currently, underwriters use catastrophe models as one of several tools to guide them in setting insurance premiums for the near term (e.g., the upcoming year); the models can also be used to estimate financial losses from projected future increases in frequency or severity of climate change–related catastrophes. Some modeling groups are already doing this. Catastrophe models combine statistical and deterministic methods to estimate the economic losses for insured properties in a specific location from a variety of natural catastrophes. They use a wide range of information to generate potential losses from natural catastrophes (hurricanes, tornadoes, earthquakes, winter storms, and floods), and the probability of these losses occurring. The insurance industry uses catastrophe models as tools to assess the risks posed by natural catastrophes and to help determine insurance rates and coverage in specific locations for the next year. Catastrophe models consist of four modules: event, hazard, vulnerability, and financial analysis.2 The Event Module describes the occurrence of the extreme events and their locations. It is used to provide an accurate representation of the probability of occurrence of all events likely to cause damage at any given location; The Hazard Module incorporates the intensity of the events at each location; The Vulnerability Module calculates the damage at a given location for each event in the event module; and The Financial Analysis Module utilizes insured value and policy terms, such as deductibles, inurings, and limits and applies these to exposures at specific locations to calculate the insured loss for insurance company portfolios. The event and hazard modules incorporate all the potential events and their magnitudes that might occur at each location. The input regarding the distribution of these events is based on historical data, such as the National Hurricane Center Best Track data, a 159-year archive of all tropical cyclone tracks in the North Atlantic. The vulnerability module combines event and hazard module output with data on building quality, age, construction materials, building type, occupancy type, and other factors through the use of vulnerability functions—equations that prescribe the degree of damage associated with the event parameters. This module assesses how badly each location is damaged in an event. Vulnerability curves are developed through a combination of engineering modeling, observed building behavior in actual historical events, and experimental observations, such as wind tunnel experiments. The financial module combines the damage generated by the vulnerability module with the coverage terms of the insured locations to calculate total and insured losses for insurance companies. Calculated losses are mostly property losses but also include business interruption and building contents. The final output of a catastrophe model is the Loss Frequency Curve (LFC), which summarizes the loss potential. The reliability of the LFCs depends on the reliability of the individual modules and their interrelationships. The expected occurrence of hazards included in catastrophe models is based on the premise that an adequate statistical profile of catastrophic events can be derived from historical data, and that the near-future behavior of these events can be estimated from these data. Since statistically significant climate change impacts related to disaster frequencies are, for the most part, not discernible in recent data, trends related to climate change are weak at best in most current catastrophe models. The insurance industry's models can be adapted to assess potential losses from natural catastrophes in the future when climate change impacts become more evident. For example, the frequency and intensity of hurricanes, rain storms, floods, and winter storms in catastrophe models can be increased, under specific assumptions, to simulate climate change forecasts; possible economic losses based on these forecasts can then be calculated. The economic losses can be based on current data or on estimates of the future value and characteristics of properties in specific areas. Using current property value and characteristics data to project the impacts of future climate change losses provides a valuation benchmark based on today's property characteristics and valuations. However, these current financial values and physical vulnerabilities will almost certainly change with time. The benefit of catastrophe models lies in providing a basis and an approach for evaluating what future losses might be, on the basis of certain assumptions about climate change impacts. They can give policy makers a chance to consider “what if” scenarios, which can be useful for adaptation planning purposes. However, given the typically high degree of uncertainty in timing and extent of the future climate change impacts, it is difficult to use the loss valuation results based on future climate change scenarios to make concrete adaptation investments without having better guidance on the probability and timing of these in the climate state can potentially result in historical data and not of current and Therefore, and companies might become on global climate models to and relevant hazard for catastrophe models. climate models, as in Chapter directly incorporate and to provide long-term of the climate such as sea and precipitation The output of the climate models of forecasts of climate such as and sea and These are long-term for the decades and are large areas. While in current climate models are to simulate the that some extreme events, such as the severe storms and that are for the insurance industry catastrophe models. Some of the uncertainty in global climate models is related to of of scientific such as the of of to better understand such is The climate model forecasts also with large associated with underlying assumptions, including future GHG emissions economic and in to global on climate change the current of near-term to of due to and large associated the climate models need to be so as to be more useful for the insurance industry in their future and for adaptation planning in As climate models and their forecasts for with reduced uncertainty may be which will insurance as as adaptation There is presently an intense on climate model which will help to increase the of climate model results for underwriters and adaptation is for a plan of climate change adaptation. There are several ways in which state, and federal governments can a role in the private insurance market and the public State can long-term insurability of property by with the insurance industry to maintain a private insurance market with risk-based premiums. at state, or federal could adaptation measures, such as stricter construction or for and include climate change risk in land-use planning through and other or and other adaptation in high-risk areas, even the full price that will from private insurance are evident. of this can contribute to Flexible Adaptation Pathways. Since these measures can reduce climate related they can help maintain insurability and contribute to lower private insurance premiums. In both the near and long government financial might the form of financial to owners that or other adaptation measures. The public can also the adaptation by helping to determine the level and of for current and climate change risk, by global and data such as National and as as and local and by to in global climate models. This research can lead to better of the key scientific that affect forecasts of climate change and associated impacts. The private could work with governments at all and provide information regarding data and that would benefit all In addition, the government at all could public to increase awareness of and provide from increased risk of natural weather-related catastrophes. While and private property are not within the scope of the Climate Change Adaptation the scope of climate change impacts on all properties in the New York City metropolitan area is since the insured who own and manage these contribute to the New York City and directly on the losses include not only property but also losses from business interruption following a catastrophic event. For example, almost of the insured losses from Hurricane Katrina from business interruption following the while of the losses were related to property damage Thus, the full scope of risk in the New York City metropolitan area is key to providing information for the adaptation While the primary near-term risk for New York City is storm surge from intense nor'easters, which will be exacerbated by sea level rise, the is also to hurricanes. The Insurance Information estimates that a storm as strong as the the 3 hurricane that made in in today and approximately to the in New York City and New insured property losses could be as high as This would be to the insured losses caused by Hurricane The future loss potential could continue to because of development in the to the total economy could resulting from impacts, such as business interruption the New York metropolitan area has been a hit from a hurricane comparable to the since However, to the of the National Hurricane in a is not a of a major hurricane will the New York but The of a hurricane is a damage in the from Hurricane and Hurricane were less significant to their hurricanes during low resulting in storm The of industry insurance record keeping to a more of the loss occurring from a storm the and intensity of the it were to occur However, it is possible to estimate what the industry losses from events occurring in and would be today after for changes in and insurance Property a of the Insurance in City, a of United States losses by state and The is in U.S. of the year of the of and and which into the the estimated cost of of historical hurricanes and nor'easters are calculated. Some extreme events are in While the do not to the insured losses generated by Hurricane Katrina, all losses and most of them are in of This is only of insured economic losses are at least the insured Therefore, the occurrence today of at least of the events and and the December could result in economic losses in of $1 billion in the New York metropolitan expected climate change impacts and the loss potential in New York City is even higher than historical data might A recent for and report a of how much is at the New York is among the in terms of to coastal and second only to in terms of to coastal estimate of sea level rise during the that approximately in about would be affected for a increase in sea level in greater New York Furthermore, some of major and would also be under this a more emissions the in New York by sea level rise Furthermore, the value to sea level rise increases from billion to Risk associated with coastal flooding because of sea level rise is not the only on in the the of hurricanes and nor'easters are as well. A wind damage is on the basis of and and New York is second only to in wind damage potential In the total value of insured coastal properties in New York all of which are located in New York City and was more than $2.3 This of the value of all insured properties in New York This is an estimate of the cost to and their including and business interruption coverage, for all
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame machine prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.
Distilled classifier scores by category (both heads)
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.001 | 0.002 |
| Meta-epidemiology (narrow) | 0.001 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.000 |
| Bibliometrics | 0.002 | 0.003 |
| Science and technology studies | 0.003 | 0.001 |
| Scholarly communication | 0.007 | 0.005 |
| Open science | 0.001 | 0.002 |
| Research integrity | 0.003 | 0.003 |
| Insufficient payload (model declined to judge) | 0.187 | 0.046 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one source (direct Gemma or distilled Codex), not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".