Systemic risk and climate tipping points

A systemic risk is a risk that impairs the functioning of the system. The impairment could range from local inconvenience all the way through to a collapse of the entire system. Systemic risks are always with us and are not distinct events. They are ‘endogenous’ (originate from within the system) and the probability of a systemic risk rises and falls through time depending on how we influence the evolution of the system.

What is new here is the need to adopt systems thinking. So, we are no longer concerned with the risk to, or failure of, a component, but with the interconnectedness of that component with the rest of the system. Does the failure propagate through the system, or remain local? As an example, consider the global financial crisis. It could be argued that this is the closest humanity has got to a complete failure of the entire financial system. Individual mortgage defaults were amplified through securitisation vehicles, which blew up many off-balance sheet vehicles (‘SIVs’), which caused problems for the balance sheets of banks, which were either allowed to fail (Lehmans) and cause extreme market volatility, or were bailed out by taxpayers and the pumping of central bank liquidity on a scale never before imagined. Without the taxpayer bailouts, it is completely plausible that the banking system would have failed and, with it, the payments system – with consequences for the receipt of wages and pensions and the purchase of food and other essentials.

It is worth sitting with that thought for a while. It isn’t easy to imagine our entire system ceasing to function. And are we really claiming a bunch of bad weather could do that?

Well, yes… Here we need to introduce the notion of climate tipping points.

As an illustration of a tipping point, consider the sketch in the figure below. The ball is in equilibrium at point A, but if we disturb the system (apply energy) we will cause the ball to roll from side to side around point A. If we apply more and more energy eventually we will cause the ball to tip out of the basin around A and it will fall to point B. That becomes the ball’s new equilibrium state. Technically this tipping point is not irreversible. With enough energy we could move the ball up its new left-hand side and tip it back into state A. However, the energy required to do this is much greater than the energy required to move from A to B. We don’t know enough about climate tipping points to comment usefully on reversibility. We therefore invoke the prudence principle and state that it is much better to avoid tipping the system in the first place.

A climate tipping point is where a small amount of additional climate forcing (usually global warming) triggers a qualitative change in part of the climate system[1] (state B feels noticeably different to state A). Large parts of the climate system that can pass tipping points are called ‘tipping elements’[2], and it is also possible to tip the entire climate system[3]. The most policy-relevant tipping points are those that are likely to occur this century due to human activity, and that result in significant damages. They are (cryosphere) the melting of the Greenland, West and East Antarctic ice sheets, (circulation) the collapse of the Atlantic Meridional Overturning Circulation (AMOC), disruption of the West African and South Asian/Indian monsoons, and (biosphere) the large-scale dieback of the Amazon rainforest and of boreal forests. Forest dieback could be near term[4] and rapid, while the melting of the ice sheets will take centuries to complete. The collapse of a monsoon system could be very fast and therefore devastate food production for hundreds of millions of people.

The scientific assessment of climate tipping points has changed over time. When first considered by the IPCC 20 years ago they would have qualified as extreme risks – only likely if there was no mitigation and temperature was allowed to increase by 4C or more. Now they are considered to have a significant probability at current warming levels and a high probability at 2C or above. Further, the crossing of one tipping point can trigger a cascade of further tipping points. For example, the collapse of the AMOC[5] would fundamentally change the European climate, raise sea levels in the North Atlantic by 1m, and disrupt monsoons around the tropics. Can we predict how close we are to climate tipping points? Not really, because pre-tipping the observed changes tend to be smooth. When we observe abrupt changes, we have passed the tipping point.

The conversation about climate tipping points is a conversation about taking a climate system that has provided a pleasant niche in which humans have thrived, and moving it into a new state – hotter, more dangerous and less pleasant – with no path back. We have been living in an era of human-caused climate change (global warming). This is the good news, because if we have been causing it, we can stop causing it and there is a path back to the old, pleasant and less-dangerous niche. However, if we continue to force the climate system to warm, and we trigger a tipping point then we pitch ourselves into a new era. In that era, climate change will be partly human-caused and partly nature-caused. This is bad news, because we could drop our emissions to zero but we will not, then, be able to persuade nature to ‘un-tip’ herself. There will be no path back to our pleasant niche. This calls for grown up risk management, which in turn distinctly calls for sharp thresholds to be built into our models’ damage functions – even if we don’t have the first clue as to how punitive to make them.


[1] We suggest the best explanation (short and simple) from a climate scientist is Tipping points in the climate system, by Tim Lenton, 18 August 2021,

[2] Tipping elements in the earth’s climate system, Lenton et al 2008,

[3] Such as the on-set of ice-age cycles 2.5m years ago – see Tipping points in the climate system linked above

[4] In Amazon rainforest is losing its ability to recover from destruction, Financial Times (paywall), 7 March 2022, Alexandra Heal reports that the tipping point could be as little as 10 to 20 years away

[5] The currents are already at their slowest in the last 1,600 years, see Climate crisis: Scientists spot warning signs of Gulf Stream collapse, The Guardian, 5 August 2021