Uncommon truths: Hot, Hot, Hot!
We are seeing extreme climatic conditions but 2027 could be even worse. The trend looks set for a 3.5-4.2C change by 2100. This suggests we need massive mitigation and adaptation spending.
Hot, Hot, Hot (borrowed from The Merrymen) could describe the situation in the Middle East. The Brent first futures price has risen back to near $100, bond yields are up, the dollar is strengthening and stock markets are down. However, implied volatility indices such as VIX and MOVE suggest there is little panic. We are monitoring the situation but still expect global economic resilience and stick with our preference for cyclical assets. However, that could change if the Strait of Hormuz and Red Sea remain closed into Q4.
But the title of this document refers to the extreme temperatures seen in Europe and North America (accompanied by wildfires in Canada, France and Spain, extremely low rainfall in the UK, low rivers in many parts of Europe and spikes in excess deaths in countries such as Belgium, France, Germany and Spain). Based on data from the UK Meteorological Office’s Hadley Centre, I reckon the average global temperature (sea and air surface) in January-May was 1.43°C above the 1850-1900 average for the time of year. That is bad but was exceeded in 2024 (1.59), 2025 (1.52), 2016 (1.48) and 2020 (1.45).
Then came June, the hottest ever in Western Europe and the second hottest globally, with the surface air temperature 1.39°C above the 1850-1900 average, according to the EU’s Copernicus Climate Change Service (June 2024 was the hottest at 1.50°C above the 1850-1900 average).
Unfortunately, it could get even hotter. Though 2026 may not turn out to be the hottest year on record, I think 2027 could be, due to building El Nino conditions. The global sea-surface temperature averaged 20.86°C in June, the hottest June on record for extra-polar (60°S-60°N) oceans, with high temperatures particularly notable in the Pacific, including some of the Nino 3.4 region used to monitor El Nino events. From July onwards, the historical peaks came in 2023, a period that preceded the 2023-24 El Nino event associated with the record temperatures of 2024.
To that building El Nino event must be added the overall warming trend (see Figure 2b). Are rising temperatures (and broader climate change) caused by human activity? Perhaps it is just coincidence, but the atmospheric concentration of CO2 reached a new high of 427.35 parts per million (ppm) in 2025, according to the US National Oceanic and Atmospheric Administration. Figure 1 shows this to be well above the norms of the last 800,000 years. As CO2 concentration is correlated to CO2 emissions in the previous one hundred years, I believe that rising CO2 emissions have contributed to rising temperatures (molecules of greenhouse gases such as CO2 absorb energy, thus holding heat in the atmosphere).
Note: “Actual” data is from the year -803,719 (i.e. 803,719 B.C.) to 2026. Data is not available for all years, so the date axis is not to scale. Data is shown for each year from 1750, using simple interpolation to fill any gaps. Data from 1958 to 2025 is based on observations at the US National Oceanic and Atmospheric Administration’s (NOAA) Mauna Lao Observatory on Hawaii. The 2026 datapoint (429.7 ppm) is based on the forecast change for 2026 produced by the UK’s Meteorological Office. Data prior to 1958 is derived from ice core records, as provided by NOAA Earth System Research Laboratories. Projections assume that CO2 concentration is determined by emissions in the previous 100 years (using an econometric relationship derived from data since 1750). Projections rely on forecasts of future CO2 emissions by low, middle and high-income countries (the global total being an aggregation of the three): “recent trends” assumes a continuation of recent trends in declines in the CO2 intensity of GDP and growth in GDP per capita, whereas “optimistic” assumes a more aggressive reduction in CO2 intensity (see the detailed explanation in the appendix). In both cases, population forecasts are taken from the UN’s World Population Prospects 2024. “Industrial revolution” is the period 1760-1840. Source: NOAA, Our World in Data, UK Meteorological Office, United Nations, World Bank, LSEG Datastream and Invesco Strategy & Insights
The good news is that we can do something about it. However, CO2 emissions hit a new high in 2025 (41.0bn tonnes, according to the Energy Institute Statistical Review of World Energy). I reckon they will continue climbing, using a model that calculates CO2 emissions as the product of population, GDP per capita and the CO2 intensity of GDP.
Of course, the outcome depends on the assumptions. UN Medium Variant projections suggest the world’s population will rise from 8.2bn in 2025 to a peak of 10.3bn in 2084. Luckily, that peak is around 1bn lower than predicted in 2015 (total population was then expected to reach 11.21bn in 2100 and to go higher). Nevertheless, if real incomes (and spending) continue to rise, that 26% gain in population will require massive technological shifts to stabilise CO2 emissions. Figure 2a shows that we are on the right path, with a gradual decline in the CO2 intensity of economic activity. Technological change will hopefully drive it even lower.
Unfortunately, it isn’t happening fast enough. If CO2 intensity follows recent trends, I estimate that annual CO2 emissions will double by the time they peak in 2072 (assumptions are in the appendix). Based on my models, this, and the emissions of recent decades, will result in a further rise in the atmospheric concentration of CO2 (see “Projected (recent trends)” in Figure 1). This leads me to conclude that by 2100 the global temperature will be 4.2C higher than the 1850-1900 average, based on the model shown in Figure 2b (which shows the variance versus 1961-1990).
A highly optimistic scenario, that sees high-income country gross emissions fall to zero by 2060 and a doubling of the rate of decline in CO2 intensity in low- and middle-income countries, gives the result that global CO2 emissions fall from here and would halve by 2091. But CO2 concentration would continue to climb because of past emissions (see “Projected (optimistic)” in Figure 1) with a peak in 2088. Nevertheless, I still predict a temperature gain of 3.5C versus 1850-1900 by the end of the century, which shows the scale of the challenge facing us.
Even on my most optimistic scenario (accepting the simplicity of my models), the temperature change outcomes would be dramatic, as would the potential implications for weather pattern volatility, rising sea levels, health, agricultural production and migration flows. The 2100 temperature projections are roughly 0.5C higher than what was suggested just five years ago, which shows how little progress we are making.
The world may need big investment in carbon reducing and carbon removing technology, from reforestation, through electrification of transport systems to renewable energy sources (as covered in The 21st Century Portfolio, November 2019, and in various editions of Economic Transition Monitor). That could continue to present investment opportunities in related technologies and industries. Meanwhile, large scale adaptation spending (infrastructure, food sciences etc.) may be needed as we learn to live in a changing world.
Unless stated otherwise, all data as of 24 July 2026.
Notes: Figure 2a shows the CO2 intensity of GDP annually from 1990 to 2025 for low, middle and high-income countries (as currently defined by the World Bank). Figure 2b shows annual data from 1850 to 2100. It shows the historical global temperature variance ("Temp variance"), which is the global average land-sea temperature anomaly relative to the 1961-1990 average temperature in degrees Celsius, median estimate, as provided by UK Met Office Hadley Centre. "Fitted temp variance" is the result of a regression analysis that fits historical temperature variance to atmospheric CO2 concentration (using the natural logarithm of the 100-year moving average of concentration, on the assumption that temperature at any moment is determined by CO2 concentration during the previous 100 years). "Predicted (recent trends)" applies that fitted relationship to our forecast of CO2 concentrations, assuming that recent trends in CO2 intensity and GDP per capita continue, though with some convergence between World Bank income groups after 2050 (see appendix for details). "Predicted (optimistic)" assumes a doubling of the rate of decline in CO2 intensity (with the added assumption that high-income CO2 emissions trend to zero in 2060). Source: NOAA, Our World in Data, UK Meteorological Office, United Nations, World Bank, LSEG Datastream and Invesco Strategy & Insights
Investment risks
The value of investments and any income will fluctuate (this may partly be the result of exchange rate fluctuations) and investors may not get back the full amount invested.
