7.4.2.2 Services
Services include a wide variety of human needs, activities and systems, related both to meeting consumer needs and to employment in the service activities themselves. This section includes brief discussions of possible climate-change effects on trade, retail and commercial services, tourism and risk financing/insurance as illustrations of the implications of climate change – not implying that these sectors are the only ones that could be affected, negatively or positively.
7.4.2.2.1 Trade
Possible impacts of climate change on inter-regional trade are still rather speculative. Climate change could affect trade by reshaping regional comparative advantage related to (a) general climate-related influences (Figure 7.2), such as on agricultural production, (b) exposure to extreme events combined with a lack of capacity to cope with them, and/or (c) effects of climate-change mitigation policies that might create markets for emission-reduction alternatives. In an era of increased globalisation, small changes in price structures (including transportation costs) could have amplified effects on regional economies and employment. Beyond actual climate-change impacts, a perception of future impacts or regulatory initiatives could also affect investment and trade.
Climate change may also disrupt transport activities that are important to national supplies (and travellers) as well as international trade. For instance, extreme events may temporarily close ports or transport routes and damage infrastructure critical to trade. Increases in the frequency or magnitude of extreme weather events could amplify the costs to transport companies and state authorities from closed roads, train delays and cancellations, and other interruptions of activities (O’Brien et al., 2004). It appears that there could be linkages between climate-change scenarios and international trade scenarios, such as a number of regional and sub-regional free trade agreements, although research on this topic is lacking.
7.4.2.2.2 Retail and commercial services
Retail and other commercial services have often been neglected in climate-change impact studies. Climate change has the potential to affect every link in the supply chain, including the efficiency of the distribution network, the health and comfort of the workforce (Chapter 8), and patterns of consumption. Many of the services can be more difficult to move than industrial facilities, because their locations are focused on where the people are. In addition, climate-change policies could raise industrial and transportation costs, alter world trade patterns, and necessitate changes in infrastructure and design technology. As one example, distribution networks for commercial activities would be affected in a variety of ways by changing winter road conditions (e.g., ACIA, 2004) and negatively affected by an increase in hazardous weather events. Strong winds can unbalance high-sided vehicles on roads and bridges, and may delay the passage of goods by sea. Transportation routes in permafrost zones may be negatively affected by higher temperatures which would shorten the winter-road season (Instanes et al., 2005). Coastal infrastructure and distribution facilities are vulnerable to inundation and flood damage. In contrast, transportation of bulk freight by inland waterways, such as the Rhine, can be disrupted during droughts (Parry, 2000). Further, climate variation creates short-term shifts in patterns of consumption within specific retail markets, such as the clothing and footwear market (Agnew and Palutikof, 1999). However, most impacts entail transfers within the economy (Subak et al., 2000) and are transitory.
Perishable commodities are one of the most climate-sensitive retail markets (Lin and Chen, 2003). It is possible that climate change will alter the sourcing and processing of agricultural produce; and climate-change policies (e.g., a carbon tax or an emissions offset payment) may further alter the geographical distribution of raw materials and product markets.
7.4.2.2.3 Tourism
A substantial research literature has assessed the consequences of climate change for international tourist flows (e.g., Agnew and Viner, 2001; Hamilton et al., 2005), for the tourist industries of nations (Becken, 2005; Ceron and Dubois, 2005), destinations (Belle and Bramwell, 2005), attractions, such as national parks (Jones and Scott, 2007; Chapter 14, Section 14.4.7), and tourism activities (Perry, 2004; Jones et al., 2006) or sectors of tourism such as ski-tourism (e.g., Elsasser and Burki, 2002; Fukushima et al., 2003; Hamilton et al., 2003).
Likely effects of climate change on tourism vary widely according to location, including both direct and indirect effects. Regarding direct effects, climate change in temperate and high latitude countries seems to mean a poleward shift in conditions favourable to many forms of tourism (Chapter 15). This might, for instance, lead to more domestic tourism in north-west Europe (Chapter 12, Section 12.4.9; Agnew and Viner, 2001; Maddison, 2001) and in the middle latitudes of North America (Chapter 14, Section 14.4.7). If winters turn out to be milder but wet and windy, however, the gains to be expected are less obvious (Ceron, 2000). Areas dependent on the availability of snow are among those most vulnerable to global warming (Chapter 11, Sections 11.4.9; Chapter 12, Section 12.4.9; Chapter 14, Section 14.4.7). In summer, destinations already hot could become uncomfortable (Chapter 12, Section 12.5.9). Tropical destinations might not suffer as much from an increase in temperatures, since tourists might expect warm climates as long as indoor comfort is assured – with implications for greenhouse gas emissions (Gössling and Hall, 2005). For low-lying islands, sea-level rise and increasingly frequent and intense weather extremes might become of great importance in the future (Chapter 16, Section 16.4.2). Extreme climate events, such as tropical storms, could have substantial effects on tourist infrastructure and the economies of small-island states (London, 2004).
Indirect effects include changes in the availability of water and costs of space cooling, but at least as significant could be changes in the landscape of areas of tourist interest, which could be positive or negative (Braun et al.,1999; Uyarra et al., 2005; Chapter 14, Section 14.4.7). Warmer climates open up the possibility of extending exotic environments (such as palm trees in western Europe), which could be considered by some tourists as positive but could lead to a spatial extension and amplification of water- and vector-borne diseases. Droughts and the extension of arid environments (and the effects of extreme weather events) might discourage tourists, although it is not entirely clear what they consider to be unacceptable. In tropical environments, destruction due to extreme weather events (buildings, coral reefs, trees and plants) is a concern, but vegetation and landscape tend to recover relatively quickly with the notable exception of eroded beaches and damaged coral reefs. One indirect factor of considerable importance is energy prices, which affect both the cost of providing comfort in tourist areas and the cost of travelling to them (Becken et al., 2001). This effect can be especially significant for smaller, tourist-oriented countries, often in the developing world; for instance, receipts from international tourism account for 39% of GDP in the Bahamas, but only 2.4% for France (World Tourism Organization, 2003).
The environmental context in which tourism will operate in the future involves considerable uncertainties. The range of possible scenarios is great, and there have been some attempts to link the future of tourist activities to SRES scenarios (Chapter 14, Section 14.4.7; Chapter 11, Section 11.4.9) . In these scenarios, tourist reactions to climate change are assumed to be constant, notwithstanding the fact that these responses are currently not satisfactorily understood.
7.4.2.2.4 Insurance
Insurance is a major service sector with the potential to be directly affected by any increase in damages associated with climate change, such as more intense and/or frequent extreme weather events (see Box 7.3). While a number of lines of insurance have some potential to be affected by catastrophe losses, the principal impacts are expected to be on property lines.
Box 7.3. The impact of recent hurricane losses
The US$15.5 billion insurance loss of Hurricane Andrew in 1992 (US$45 billion adjusted to 2005 values and exposures) remains an exemplar of the consequences on the insurance industry of a catastrophe more severe than had been anticipated, leading to the insolvency of 12 insurance companies and significant market disruption. However, after major adjustments, including the widespread use of catastrophe models, the private insurance market re-expanded its role, so that in the four hurricanes of 2004 (with a total market loss of around US$29 billion from the U.S., Caribbean and Gulf Energy sectors) only one small U.S. insurance company failed, and there was little impact on reinsurance rates, largely because state-backed insurance and reinsurance mechanisms in Florida absorbed a significant proportion of the loss. However, a far greater proportion of the US$60 billion of insured losses from the 2005 hurricanes in Mexico, the energy sector in the Gulf of Mexico and the USA fell onto the international reinsurance market, leading to at least two situations where medium-sized reinsurers could not remain independently viable. Following more than 250,000 flood claims in 2005 related to Hurricanes Katrina, Rita and Wilma, the U.S. federal National Flood Insurance Program would have gone bankrupt without being given the ability to borrow an additional US$20.8 billion from the U.S. Treasury.
As the actuarial analysis of recent loss experience is typically an inadequate guide to catastrophe risk, since the 1990s probabilistic ‘catastrophe’ modelling software has become employed by insurers for pricing and managing portfolios of property catastrophe risk (Grossi and Kunreuther, 2005). At the start of 2006, the five-year forward-looking activity rate employed in the most widely used Hurricane Catastrophe Model was increased relative to mean historical rates with an acknowledgement that some contribution to this increase is likely to reflect climate change (Muir Wood et al., 2006).
Within the risk market, reinsurers tend to be more pessimistic about catastrophe risk-costs than the insurers who are ceding the risk, and this perspective has been highlighted by statements from reinsurers going back more than a decade warning of the potential impacts of climate change (Swiss Re, 2004; Munich Re, 2005). However, in 2006, insurers also began to communicate directly with their policyholders regarding the rising costs of claims attributed to climate change (Allianz and World Wildlife Fund, 2006; Crichton, 2006).
The specific insurance risk coverages currently available within a country will have been shaped by the impact of past catastrophes. Because of the high concentration of losses where, over the past 50-60 years, there have been catastrophic floods, private sector flood insurance is generally restricted (or even unavailable), so that in many developed countries governments have put in place alternative state-backed flood insurance schemes (Swiss Re, 1998).
In both developed and developing countries, property insurance coverage will expand with economic growth. If overall risk increases under climate change, the insurance industry can be expected to grow in the volume of premium collected, claims paid and, potentially, income (where insurers overcome consumer and regulatory pressures to restrict increases in insurance rates, and where catastrophe loss cost increases are appropriately anticipated and modelled). However, market dislocations are also likely, as in 2006 when, unable for regulatory reasons to pass on higher technical hurricane risk costs, U.S. insurers declined to cover homeowners and businesses at the highest-risk coastal locations, thereby undermining the real estate market and forcing government intervention in structuring some alternative insurance provision (Freer, 2006).
After a decade of rising losses (from both natural and man-made catastrophes), insurance is generally becoming more restrictive in what is covered. Insurance rates in many areas rose after 2001 so that, while the 2004 year was the worst (up to that time) for U.S. catastrophe losses, it was also the most profitable year ever for U.S. insurers (Dyson, 2005). However, the years 2001 to 2005 were not so profitable for reinsurers, although increases in prices saw significant new capital entering the market in 2002 and 2005, while 2006 appeared a benign year for losses.
Where increased risk costs lead insurers to reduce the availability of insurance, there will be impacts on local and regional economies, including housing and industrial activity, unless government expands its risk protection roles. In particular in developed countries, governments are also likely to be the principal funders of risk mitigation measures (e.g., flood defences) that can help ensure that properties remain insurable. In the developing world, the role of insurers and governments in offering risk protection is generally limited (Mills, 2004).
The use of insurance is far lower in developing and newly-developed countries (Enz, 2000), as insurance reflects wealth protection that typically lags a generation behind wealth generation. As highlighted by events such as 2005 Hurricane Stan in Mexico and Guatemala, individuals bear the majority of the risk and manage it through the solidarity of family and other networks, if at all. However, once development is underway, insurance typically expands faster than the growth in GDP. With this in mind there has been a focus on promoting ‘micro-insurance’ to reduce people’s financial vulnerability when linked with the broader agenda of risk reduction (ProVention Consortium, 2004; Abels and Bullen, 2005), sometimes with the first instalment of the premium paid by the non-governmental organisation (NGO), e.g., in an insurance scheme against cyclones offered in eastern Andhra Pradesh and Orissa.
For the finance sector, climate change-related risks are increasingly considered for specific ‘susceptible’ sectors such as hydroelectric projects, irrigation and agriculture, and tourism (UNEP, 2002). In high carbon-emitting sectors, such as power generation and petrochemicals, future company valuations could also become affected by threatened litigation around climate-change impacts (Kiernan, 2005). Some specialised investment entities, and in particular hedge funds, take positions around climate related risks, via investments in reinsurance and insurance companies, resource prices such as oil and gas with the potential to be affected by Gulf hurricanes, and through participation in alternative risk transfer products, e.g., insurance-linked securities such as catastrophe bonds and weather derivatives (see Jewson et al., 2005).