French heatwave disaster of August 2003 (thing)

Blame Management
In the ensuing public out roar, a number of reasons were given as reasons to why so many preventable deaths occured. This reflects not just on blame management carried out by involved parties with political agendas and no interest in declaring responsibility, but also the number of factors at play. These factors can be separated between pre-existing conditions (the lack of air-conditioning), and those that would only have come into play as the crisis evolved (the activation of Plan Blanc ). Identifying potential and actual problems in any system (risk assessment), and determining how they should be dealt with (risk management) is a scientific discipline in itself, but in practice it may be corrupted by both political and subjective influences, as well as simply having incomplete information.

However political realities demanded that the fault be quickly placed on one person. Dr. Abenhaim resigned on the 18th in forced circumstances, making the point that the government had ignored his repeated requests for additional resources. Mattei, his boss, said that Abenhaim failed to provide sufficient warning in enough time for him to react, although Abenhaim, along with several media commentators and members of the political opposition, considered the resignation was arranged to deflect attention away from the government.

The department had ample evidence of an impending disaster: the dominant, stationary anti-cyclone weather pattern would have allowed forecasters to predict with solid confidence that the heatwave would continue for several days, and France had suffered killer heatwaves in recent years (Murray 2003). In fact a public health advisory was issued on the 8th, four days before deaths peaked. Abenhaim did admit that some mistakes were made in the handling of the crisis, but he draw more attention to ongoing issues, such as the lack of air-conditioning in nursing homes. A Ministry of Heath report on the heatwave tabled in September cited compartmentalisation throughout the health system, hindering 'anticipation, organisation and coordination' at the beginning of the crisis (Crabbe 2003:773). This still does not excuse Mattei’s decision on the 11th not to activate Plan Blanc , or even adopt more gravitas in the interview from his villa, when at that time his ministry had already noted overfilling wards. However it illuminates the problem of the ‘wait-and-see’ method of active surveillance, when an authority effectively decides to intervene only after a situation deteriorates to the requisite level causing enough bodies to meet a benchmark.

Still, there was considered commentary about the various root causes of the disaster. Dr. Pelloux stated that many lives could have been saved if hospitals were adequately staffed and resourced (Service France 2003), adding the following day that France’s public health policy needed an ‘autopsy’. The Greens used the heatwave as an argument for combatting global warming. Within Raffarin’s own government the 35-hour work week and the number of public holidays afforded to French workers was criticised. And among some political and intellectual circles there was a call for the public to reexamine how French society treated their elderly and other vulnerable, invisible people living in the margins. The French government later in a measure designed to be a symbolic and practical gesture of atonement (bourne by the public and criticised by some unions), abolished one of France’s eleven public holidays and committed the savings and taxes generated to a special fund intended for the welfare of the elderly and disabled.

Analysis

The Risk Chain
At this point, it is prudent to adopt a risk analysis methodology that models cause- and-effect, to appreciate the full situation. Fone and Young (2000:81-3) have developed the ‘risk Chain’, that breaks down a situation where a loss (or gain) results, into six sequentially interlocking links:

Environment -> Hazards -> exposure -> peril -> Outcome -> Consequence

The Environment. The circumstances in which a loss occurs; e.g.: the weather, labour relations etc.
The Hazard/risk Factor. The conditions within an environment that influence the probability of a loss occuring, and/or its magnitude if one ensures; e.g.: heat.
Exposure. The extent to which somebody or something experiences the hazard/risk factor, being a function of frequency (likelihood of occurrence) and magnitude (severity of a loss); e.g.: how often people labour in the mid-day heat ?
Peril. The process where the exposed hazard contributes to a loss; e.g.: hyperthermia.
Outcome. A quantifiable result that ensues from peril; e.g.: death.
Consequence. What long-term effects result from a loss, not necessarily what a risk regime is designed to avoid; e.g.: political backlash.

The point of the Risk Chain is to show the various points at which a government can prevent and respond to a disaster. Solutions directed at "early" links are preventative and pro-active, while "later" links are reactive. It is not easy for governments to respond early, either by infringing liberties or spending money, when the looming signs of a crisis are not evident (how hard has it been to ban chlorofluorocarbons ?).

Preventative Actions at the Environment Level
Just as prevention is better than cure, controlling the risk at the earliest juncture is more likely to be cheaper and safer than reacting later once a crisis brews. In the context of the risk chain an ‘environment’ might be defined as an entire identifiable and delineatable domain of interrelated entities and processes underpinned by physical and/or social laws, which by virtue of its own existence creates hazards. Regardless if it is abstract (like the global bond market or consumer confidence), or material (a stretch of road, the weather), an environment by its nature would likely to affect a large number of different people and organisations. Metaphorically, it is the source where ‘root causes’ originate, or the ‘swamp’ that needs to be ‘drained’.

However, in practice for many types of threats governments do not often adopt preventative measures at early stages along the chain, that is, by changing the environment to eliminate a risk.

Unclear responsibility. In some ways an environment can also be considered a ‘public good’ (or in the case of hazardous environments , a ‘public responsibility’ ?). This unfortunately creates the effect where ownership of a problem becomes diffused, and there is no clear line of responsibility directed to addressing a risk, save from whatever decision comes from a consensus of concerned stakeholders.

Conflicting interests. Different people with different values, perceptions and objectives often operate in the same environment. Changing the environment in order to address the risk concerns of one group of stakeholders might conflict with the interests of another group (e.g.: in natural resource management, clearing forests to create firebreaks might suit local residents but antagonise environmentalists). Governments in this position dislike being forced to apply the leadership needed to forge a compromise that ensures the long-term sustainability of an environment for the common good, as the decisions chosen may cost it political support.

Value judgements. Social environments may consist of the constructs or influence of other forms of social organisation, like the family or clergy, that the government may prefer to keep separate from its own secular, rule-based governing processes (e.g.: evangelising patriotism, family values or multiculturalism amongst society)

Cause-and-effect. The number of degrees of separation between the environment and the peril sometimes obscures their relationship, especially if the effect is not physically apparent (e.g.: asbestos), or is subject to a lengthy time delay (e.g.: obesity), or if through ignorance or political design people blame a correlated but non-causal attribute instead (e.g.: violent crime and ethnicity). At the very least it makes the link contentious and weakens the argument for collective action.

The Political ‘Environment’
We can also more abstractly define ‘environment’ as the social milieux where the problem of aged care resides. In this case the ‘environment’ consists of a series of overlapping rights and responsibilities exercised by all parties concerned, each with their own set of interests, and underpinned by social conventions and law. The offspring of the aged generally want the ability to balance their home and employment life with responsibilities to their parents. Private providers of aged care services may want to operate as profitability as a minimal level of quality will avail, while managers of public providers may not even have profit as an incentive to provide a safe, dignified existence for its aged residents. The state wants to maintain the public’s trust that it gives security for all its citizens (either directly, or by creating the necessary conditions in society), and at the same time provide it and other areas of government concern with an adequate level of funding. Depending on the depth of social capital in a community, neighbours may have some attachment for each other. And the aged themselves desire to be safe, comfortable, autonomous and needs-fulfilled, at least cost to themselves personally. France, like many countries, has a mixed aged care system where family-based care is underpinned by institutions funded through three notoriously complicated caisses, or insurance funds, which also provides for pensions, home help, some hospitalisation and a myriad of other benefits (Giarachi 1992:195). While a mixed health system shares the burden on both families and the state, two problems emerge:

Perceived Roles and Moral Hazard. Citizens can misconstrue the state’s role. Rather than being seen as the 'savior at the last resort' because of its perceived role in guaranteeing the security of the population, citizens might instead be tempted to form the view that the state will take full responsibility. As a result of this premise a ‘moral hazard’ emerges when citizens modify their behaviour and withdraw their participation, such as when families frequently left their aged parents at day-care centres. From the other end, the government may see aged care as a private matter for which it is obliged to provide some assistance, but it may overestimate the capacity and will of families in being the primary carers. This is not to say that governments will deliberately seek to delude themselves of the competency of these primary carers, to pretend a problem doesn’t exist; however at an operational level it might be perfectly rational for managers to allocate their limited resources towards problem cases that have no chance of helping themselves (e.g.: the isolated elderly), while presuming those with access to families will muddle through.

Previously the traditional model of family support was quite evident, indeed France’s civil law of 1804 codified the Obligation Alimentaire- the legally enforceable obligation that people were required to ensure the welfare of their needy kin, and could not be disinherited from a parent’s estate. Its influence has considerably weakened since a universal old age pension was established in 1983 (Twigg, J. and Grand, A. 1998:141), which more effectively addressed equity issues for poorer families.

Balancing capacity. A hybrid system is only effective when one side alone has enough capacity to cope with peak demand; if not, then depending on how vital a government values a given service, that side will have to surrender its autonomy and accept the extra workload, or the system is at risk of failure. The main problem in France was that families and aged care workers alike went on holiday at the same time. The responsibilities for the temporary care of the elderly was passed to different providers, who were too overstretched to provide satisfactory care. Ultimately the emergency departments in the public hospitals, who could not refuse patients, bore the brunt of caring for the elderly who at this stage now suffered dehydration as the result of earlier neglect.

Leadership
So, one side in a hybrid system (a) may be tempted to welsh out of its responsibilities, or (b) discover that they do not really have the power to determine how much it is obliged to contribute. While it is tempting for the state to apply laws that thrusts the responsibility of aged care into the private sphere (e.g.: asset testing government pensions), it can be equally tempting for the private sphere (usually the offspring of the elderly) to avail themselves of government resources. Whenever such ambiguity arises, strong, publicly endorsed leadership might at least solve part of the problem. An effective leader can determine which stakeholders should be responsible for managing a risk and at which points, and they can then get their risk management plans endorsed by the stakeholders. At the very least good leaders make stakeholders aware of potential risks and invite them to collaboratively manage it with other stakeholders. Just as importantly, leadership also introduces a human element of intuition, foresight, ethics and accountability into a decision making process that might otherwise be ignored by the two sides of a hybrid system, who do not see themselves as solely responsible for the system’s totality.

Chirac’s comments about the need for French society to be compassionate towards the vulnerable could be interpreted as a message designed to encourage citizens to believe that the welfare of the elderly is their responsibility. Unfortunately his homily was delivered after the heatwave. One may conclude at this point that adopting effective leadership within a chaotic ‘environment’, hindered by unclear responsibilities, value judgements, conflicting interests and invisible causes, is a key preventative action. Failing to acknowledge the existence or importance of a threat is denial.

Preventative Actions at Other Levels
Hazards
Instead of changing the environment, a risk control strategy might directly target the hazard (i.e.: heat). In the case of heatwaves, this obviously means cooling dwellings where vulnerable people live, and in the context of public policy, it means creating the set of circumstances that promotes access to suitably cooled accommodation.

As mentioned, France has a low take-up of air-conditioning. taxes on electricity designed to encourage energy conservation has been blamed by Michaels (2003) as a reason why power hungry air-conditioners are not common in France, a net exporter of electricity. Two facts contradict Michaels, an environmental spokesperson of the Competitive Enterprise Institute : in Europe, France is actually a relatively high user of electricity, with only Slovenia, Switzerland, Belgium and the four coldest Scandinavian countries consuming more power per capita (Nationmaster.com 2003). Furthermore, French electricity is retailed comparatively cheaply at US$0.134/kWH, compared with US$0.146/kwH averaged by other European OECD states (rates are even higher in Spain, Portugal and Italy) (National Energy Information Center 2003). Demand for electricity soared through the heatwave, so clearly it is not a price elastic commodity.

Instead air conditioning is less common in France’s homes and hospitals due to their age. In the case of high-density Paris, it is difficult to install ducted systems in existing apartments, while possibly town planning regulations restrict the use of individual fascia units that disturbs a building’s aesthetics. Falco promised to ensure that 80% of France’s 10,000 nursing homes will contain at least one air-conditioned room by June 2004 (Agence France Press 2004), although one wonders if a national directive will conflict with municipal planning laws.

Chicago, rebuilt after the Great Chicago Fire of 1871, does not share Paris’s architectural legacy (except its own highly functionalist Chicago School traditions) and thus has more modern housing stock equipped with air conditioning. Still, Klinenberg makes the point that electricity prices were a deterrent for poor people to purchase and use air conditioners in the 1995 Chicago heatwave (2002:160). The Center of Disease Control (CDC) observed that 50% of heatwave deaths could have been prevented if the victims had an operating air conditioner. For the impoverished whose energy bills represent a large slice of their incomes, electricity] consumption is highly price-sensitive. The US government runs the Low-Income Home Energy Assistance Program (LIHEAP) which transfers payments to states to subsidise the heating and cooling costs of disadvantaged groups. Like other targetted assistance programs it was subjected to funding cuts since the mid-1980s, and that year a bitter winter depleted Illinois’s LIHEAP budget.

Exposure
When a hazard is inevitable (such as heat outside air-conditioned buildings), governments can still try to reduce the likelihood that the community will come in contact with it. Depending on both the severity of the risk and the behaviour of the community, the state might just issue warnings about a hazard (e.g.: like on cigarette packets), or it may seek to totally restrict exposure by imposing physical, legal, economic and administrative barriers (e.g.: erecting fences around power pylons, compelling motorcyclists to wear helmets, taxing cigarettes, or, in the case of heatwaves, controlling class times in schools).

Adams has noticed that people for various reasons treat risk differently and thus have varying attitudes about what role the government should have in reducing their exposure to hazard (1995:141-2).

Individualists. Believe that individuals, rather than the ‘nanny state’, should be free to choose the level of risk they themselves are prepared to be exposed to, and that this level can be determined through market forces.
Hierarchists. Believe that the state (automatically assumed to be knowledgeable and holding the citizen’s best interests at heart) is in the best position to determine the level of risk citizens should be exposed to.
Egalitarians. Believe that both the state and the individual need to come to a consensus over what level of risk is acceptable.
Fatalists. Do not believe risk can be mitigated.

It may be possible to derive a link between each of the four personality profiles above and their attitudes to the optimal level of government intervention (individualists wanting a laissez-faire approach, egalitarians wanting government to strive for equity, hierarchists prefering a clear, rule-governed regime based on meritocracy, and fatalists doubting either governments or individuals can make a difference). Unfortunately, this model does not assume that people or governments have perfect knowledge about a situation, and indeed in France people probably made decisions under-appreciating the lethality of the heatwave. The problem of people trusting organisations that rely on incomplete information is that when these organisations fail, the hierarchists and egalitarians may feel disillusioned. This could be one reason why there was a greater sense of outrage in hierarchist France than individualist, neo-liberal Chicago, where Americans regard misfortune as being the necessary price that an individual might have to pay if they want to participate in a deregulated, open economy, if they are not blaming the victim themselves.

Perils
For most threats risks are mitigated at the peril link, the event where a once potential hazard causes a loss. Unfortunately unlike other threats with enough prominent signs to scare a population into accepting government policy, the dangers of heatwaves do not become apparent until this point.

It makes economic sense when protecting a large population with limited resources and foresight to accept that not every threat can be anticipated and prevented, and instead they should respond once the peril emerges. Reacting at the point of peril provides some protection to a public who do not want their lifestyle limited to activities devoid of risk. This however is predicated upon the nature of the threat - can a situation be effectively dealt with once the event occurs ? (regulators monitor aircraft maintenance more strictly than cars since the consequences of aeroplane’s engine failing are more dire); and, is this option more economical than prevention ? (governments do not carry out mass flu vaccinations because those who succumb can be easily cured).

The most effective method of intervention in France would have been the prompt detection of vulnerable people suffering heat-related conditions and attendance of medical personnel. The state does not actively involve itself with monitoring the health of citizens inside their own private homes. For the aged there is a means-tested ‘Home Help’ (aide ménagère ) service where carers can be dispatched to the homes of the aged for a limited number of hours per week, however the registration procedures have been criticised for being cumbersome and not responding to emergencies (Giarachi 1992:205). Without state support, having friends and family members who could check on their welfare became a matter of survival for the elderly.

If reacting at the peril level is going to be policy, then the government needs to be certain that they can identify a crisis (even with incomplete information and subjective influences), and are prepared to make a pro-active decisions, even if they go against other interests. Unnecessary intervention gives the impression of incompetence or over-reaction, wastes resources and not just disrupts the public, but conditions them against reacting to further alerts. Justifiable intervention (such as activating Plan Blancpromptly) carries political risks, in that it highlights that the government was either unprepared or it understated the crisis in a previous public announcement. The dangerous feature about heatwaves is that its silence makes it too easy for a government not to respond, since it is tempting to leave the public largely ignorant about the magnitude of a disaster. A less conspiratorial explanation could be that governments are more (politically) risk adverse when faced with an uncertain situation. With a logic similar to curious teenagers who throw firecrackers onto bonfires to confirm that an action will lead to a predicted adverse outcome, the French government held off Plan Blanc until it had enough data from the National Institute of Medicine to confirm that, yes, they were indeed in the midst of a crisis, albeit now at a point beyond salvage.

Outcomes
The losses that come from a peril are the outcome; in the case of heatwaves the outcomes include ailments deleterious to health, and intervention here involves medical treatment.
Unfortunately the government failed to address control risk at this link: due to lax labour laws there were not enough medical workers available, and thus nearly half of all fatalities (42%) occurred in hospitals.

Consequences
The aftermath of many public policy failures often includes a backlash against the government and the institutions perceived responsible. The advent of electronic media quickens the delivery of information and formation of public opinion in democracies, but does not serve as a suitable forum for contemplating underlying issues more broadly. An event can be contextualised by the media to narrate a ‘sexed-up’, sanitised or otherwise biased version of events, and it is this narration that defines how people will interpret the event. Governments are increasingly familiar with how the media operate, and through public relations practitioners can seek to have policy failures interpreted by media outlets in a manner that is politically advantageous. Having failed to control risk at earlier links, the government could only hope to minimise the spread of bad publicity.

The mayor during the 1995 heatwave in Chicago, Richard Daley, knew that a predecessor was deposed over his failure to handle a blizzard. He calculated that it would not be in his interests to give the heatwave more prominence beyond it simply being hot weather, and in the year of Newt Gingrich’s ascent he too wanted to appear fiscally conservative. So mayor Daley publicly censured the city’s Chief Medical Examiner after he had announced a link between the looming death toll and the heat (Klinenberg 2002:173), and in an act one senior paramedic described as murder by public policy, he refused to deploy extra emergency resources. After it became apparent mayor Daley had miscalculated badly, city Hall’s response was to convene a hand-picked mayor’s Commission on Extreme Weather Conditions. This assuaged the rising media interest immediately after the heatwave, and it presented a set of politically safe findings : that the heatwave was a freak cataclysm, that at risk people should rely on community rather than government support structures, and that there was hardly any difference in mortality between whites and African-Americans.

Likewise in France during the disaster the government downplayed the crisis, hoping to frame the situation as purely meteorological. After the heatwave, the French took a different tactic. While Mattei and Abenhaim argued over the amount of warning given, Chirac chose to broaden the public debate over what the proper role of government should be. His comments implying the public had a responsibility to its underprivileged can be seen as promoting a communalist approach to welfare, underpinned by shared values and charity. This contrasts with the bifurcated, mutually antagonistic aspects of France’s industrial relation system, which underpins the welfare state and led, which many in Chirac’s camp believe, to the ill-conceived Loi Aubry. Chirac’s aforementioned eulogy to the Republic after the heatwave acknowledged the French public’s high expectations that their society should protect its vulnerable, yet articulates a Reaganesque Government is the problem message that there are limits to capacity of the public sector to deliver on ‘private sphere’ commitments, explicitly directing the focus on family responsibility.

‘Below the Radar’ - Policy, Risk and Visibility
A heatwave is an interesting kind of disaster to study, since its utter ordinariness, isolated victims from a generally ignorant public and lack of any accompanying physical damage or spectacular imagery, allows the actions of a government uninfluenced by public opinion to be observed. Firstly, one has to make a few assumptions - that the deaths were unintended, that the government seeks to retain the confidence of its citizens, that the government knew that it was not in its interests to draw attention to the disaster when possible as it exposed its own shortcomings, that the rising death toll was known by the government, and that the government (contrary to appearances) thought it had the capacity to respond. The argument ultimately comes down to whether Abenhaim briefed Mattei sufficiently or Mattei ignored his warnings, but on the balance of probabilities we can assume the latter, because (a) the hot weather and its effects on the fires, crop failures etc was already a subject of many pieces of Cabinet agenda in early August, (b) France had suffered killer heatwaves in the recent past, (c) the department issued a health warning on the 8th August, which quite possibly required the minister’s approval, and (d) Mattei is a physician by training and probably would have noticed the dangers himself.

A model is needed to chart how public policy processes anticipate, process and respond to threats which are inherently unpredictable and (conveniently) elusive to define, both at the point a disaster is unfolding, and after once the committees start to circle.

Amartya Sen and Scarcity
Amartya Sen believed that famines were influenced not just by the reduced supply of food, but also by the market responding to segments of the population with more buying power ('command'), causing prices to rise beyond the reach of less affluent consumers (1995:215-20). If one applies Sen’s theory, using nursing services as an analogy to food, we see that the disadvantaged were placed in the greatest danger by relying on the overstretched public hospitals and aged care facilities. Medical professionals, as economic agents themselves, were naturally inclined to work for private instead of public providers, who in turn relied on contracted foreign professionals from developing countries to meet labour shortfalls (passing on skill shortages internationally). Like marginalised peasants without tradeable skills, the elderly are dependent on familial and state support, but to make matters worse their needs (cleaning, supervision etc.) are labour intensive. A labour shortage cannot be solved as quickly as a capital infusion or air-dropping sacks of grain, especially if families free ride state support, labour unions demand reduced hours and labour productivity cannot be boosted (one seriously doubts nursing homes can be automated).

Sen’s thesis becomes multi-dimensional when workers (like public hospital doctors) are hired from developing countries - the problems of the marginalised in a dominant country are solved by extracting the productive component from poorer countries. Flanagan (2004:109-28) illustrates this paradox with child care and subservient gender roles in her provocative tract ‘How Serfdom Saved the Woman’s Movement’. One inequality subsumes another.

Randy Shilts and Marginalisation
Randy Shilts’s account of the early days of AIDS showed that the mainstream media greatly underreported the silent epidemic until heterosexuals and celebrities began dying (1987). Governments at all levels were ignoring the problem or failed to coordinate their resources correctly. Just as risk assessments can be corrupted by subjectivity, prejudice and politics, the dangers afflicting marginalised communities may be ignored by the media, a vital institution that should identify governmental oversights and generate community censure when necessary.

Could the elderly be ‘marginalised’ in a country known for reserving park benches for its veterans ? The lack of social networks and financial resources led to many victims to depend on the publicly-funded, under-resourced health care system, and the intersectionality was also evident in Chicago where many elderly were too poor to afford air conditioning. The other feature of age discrimination that led to deaths was ‘invisibility’, in that rising mortality amongst the age was largely unreported by the media, and thus became a non-issue to government. Like America’s gay community in the early 1980s, the elderly in France had a low profile, and their deaths were downplayed not just as a natural consequence of hot weather, but also through subtle discrimination. The underreporting can be attributed to a systemised and subconscious form of discrimination through Harris’s ‘fair innings argument’ : the death of somebody younger is more tragic and less inevitable than for somebody older who has already lived through a greater part of their expected lifespan (1997:669-72).

No major French magazine featured the heatwave as a cover story. Boyer scanned France’s print media, and noted that media reports on the heatwave largely focussed on the effect heat had on agriculture, footballers and animals during in the first half of the heatwave (2003:876) . While King has identified an unprofessionally close relationship between the French government and sections of the French press, who jointly connived to suppress reports about a extortionist threatening the country’s rail networks (2004:12), in the second half of the heatwave the media was robustly critical of the government. The situation was framed by the media as a human interest story, then as a political moot point (with some writers pontificating about public attitudes to the elderly), rather than as an ongoing disaster requiring a collectively supported solution.

Samantha Power and Denial
Samantha Power considered how governments not just selectively interpret events, but how pressure is put to bear on public servants not to inform government about facts that would oblige them to take action that interferes with other political imperatives. During the massacres in Rwanda, for example, State Department officers were routinely told to downplay atrocities so that the President would not be compelled to send peacekeepers, an option highly unpopular in the United States since Somalia. She developed a framework that explains how deniability is artificially established by governments (2003a).

Denial of Fact. As primary collators of information, governments can be more holistically aware of situations than other institutions, and may choose to ignore or refute facts. In France the government knew that the number of heat-related illnesses were rising on the 8th August (when it had conducted its own survey over the previous two days), but did not admit its hospitals were in crisis until much later.

continued as French heatwave disaster of August 2003 - II