How can we save lives effectively? Climate and donations: orders of magnitude

Realities that coexist

Three findings are true at the same time, and none cancels out the others: living standards protect life; natural disasters kill far fewer people than a century ago; global warming makes some extremes more frequent, heat first of all. Looking at them together helps us understand why climate and development are not opposed.

1. Living standards, the first line of defence

Where GDP per capita is higher, fewer children and mothers die, fewer children go hungry, people live longer, and drinking water and electricity are almost everywhere. The gap is immense: according to our calculations on the same data (medians), around 1 child in 13 dies before the age of 15 in countries with less than $3,000 per capita, compared with 1 in 220 in those with more than $50,000, a risk almost 17 times higher. Max Roser, who compares the poorest and richest countries, writes ‘more than 20 times’ [10]. These are comparisons between countries, not proof that wealth explains everything; but it is wealth that pays for healthcare, vaccines, drinking water, sea walls and warnings.

Enlarge

2. Fewer deaths, despite more disasters in the records

Earthquakes, floods, storms, droughts: in the 1920s, natural disasters killed on average more than 500,000 people a year, around 26 per 100,000 inhabitants. In the 2010s, around 45,000 a year, the Haiti earthquake included, or 0.6 per 100,000: a rate divided by more than 40, while the world’s population quadrupled [11]. Not that hazards have weakened: weather forecasts, warnings, sturdier buildings, more productive agriculture and rescue services protect far better. As older records are incomplete, the decline is probably even underestimated.

The number of recorded disasters, by contrast, has risen four- to fivefold since the 1970s: this is the figure often cited, including by the UN.

Note: this chart counts recorded disasters, not actual disasters. The EM-DAT database has existed only since 1988 (the centre that maintains it, CRED, since 1973). With satellites, the internet and the growth of humanitarian aid, more and more small events, once ignored, have entered the records: the rise from the 1970s to 2000 comes mainly from this. EM-DAT itself advises against reading a trend into it before 2000 and classes these data as ‘historical’. Since 2000, there has been no clear increase: from 325 to 450 disasters or so a year [12]. This does not mean that extremes are not increasing: they are better measured by meteorological records, below.

3. What is really increasing: extreme heat, rainfall and droughts

The IPCC establishes this on the basis of records and models: each half-degree of warming makes heatwaves, heavy rain and, in some regions, droughts markedly more frequent and more intense [13]. How many times today, and tomorrow, does an event occur that would have happened once every ten years without human influence (number of times in ten years):

IPCC, 6^th Assessment Report, Working Group I, figure SPM.6: central estimates, warming measured relative to 1850-1900 (‘today’: level adopted in 2021). Extreme one-day rainfall gains around 7% in intensity per degree.

Heatwaves already kill in every region of the world (IPCC, very high confidence) [14]. These deaths, mainly cardiovascular, show up in mortality statistics far more than in disaster registers, which count them poorly and which the first chart only partly includes: this is one of the reasons why the two trends coexist. The Lancet Countdown estimates heat-related deaths at around 546,000 a year (2012–2021), with a rate up 23% since the 1990s [15]. Cold still kills far more, in the order of 4.6 million deaths a year against 0.5 million for heat (2000–2019), but the gap is narrowing: from the early 2000s to the late 2010s, the share of deaths attributable to cold fell by about 0.5 points and that of heat rose by about 0.2 points [16]. Warming prevents some of these deaths, mainly in temperate countries, and adds others elsewhere, mainly in hot, poor countries.

4. And tomorrow? What we can reasonably expect

The IPCC projects, with high confidence, that climate change and extreme events will significantly increase illness and premature deaths, with large differences depending on adaptation [14]. The most comprehensive quantified estimate, by Tamma Carleton and her colleagues, gives for 2100, under high emissions, 73 additional deaths per year per 100,000 inhabitants (from 6 to 101 for the central half of simulations), of the same order as all cancers today. Without rising incomes or adaptation, it would be 221; with moderate emissions, 11 [17]. The effects would be highly unequal: mortality would rise by 17% in Accra, Ghana, and fall by 15% in Berlin, thanks to milder winters.

What can reasonably be drawn from this: deaths from floods, storms and cyclones should continue to fall where incomes, warnings and emergency services improve. In Bangladesh, a cyclone could once kill hundreds of thousands of people; death tolls there are now hundreds of times lower [11]. Between the 1980s and 2007-2016, the mortality rate from climate hazards fell by a factor of 6.5 worldwide, but it remains 4.4 times higher in poor countries [18]; between 2010 and 2020, floods, droughts and storms killed 15 times more people in the most vulnerable regions [14]. Heat-related mortality, meanwhile, is expected to rise, all the more so if emissions are high and hot countries remain poor. The outcome depends on three levers at once: emissions, development and adaptation (heatwave plans and warning systems, which the IPCC judges effective).

Extremes are increasing, societies protect themselves better than before, and poverty remains the foremost vulnerability. Cutting emissions and lifting people out of poverty both protect lives: the question for this page remains where each euro saves the most.