Cape Town came to the brink of “Day Zero” in 2018, resulting in the city becoming a global case study for emergency water saving as residents had to cut back on their consumption drastically.

A pivotal study found that average households in freestanding homes reduced water use from about 183 litres per person per day before the drought to 84 litres 1 in 2018, a significant decline of roughly 54% that highlighted both the severity of the crisis and the scale of behavioural change achieved (Brühl and Visser, 2021). This kind of shift matters for anyone interested in behavioural economics, because it raises a fascinating question: when a crisis passes, do people keep the new habits, or do they slide back to old behaviour?

The answer, at least in Cape Town, shows that some rebound has happened, but not a full return to old consumption patterns. The City of Cape Town’s 2025 Water Outlook notes that post-drought ‘bounce back’ in demand appears to have tapered down, and that per capita demand has settled at about 160 litres per person per day (l/c.d), still well below the 200–225 litres per person per day (l/c.d) seen before the drought (Figure 1) (City of Cape Town, 2025). The report also suggests there has been a lasting drop in discretionary use, especially outdoor use like irrigation, which implies that some water-wise habits have endured (City of Cape Town, 2025).

Figure 1: City of Cape Town per capita water demand (l/c.d)

Source: City of Cape Town. Cape Town Water Outlook. June 2025.

Habits do not just change because people suddenly become more virtuous. They change because the environment around decision-making changes. During the drought, Cape Town residents were hit by a powerful mix of incentives and cues: rising tariffs, restrictions, fear of shortage, public messaging, social pressure and constant feedback about dam levels. Research on the drought argues that the city’s response combined with price increases, lower pressure, restrictions and mass communication helped produce the steep reduction in water use (Brühl and Visser, 2021).

It is interesting to note that even after the immediate crisis faded, household water use did not revert fully to pre-drought patterns. This suggests the drought did more than temporarily scare people, it may have reset expectations about what counts as reasonable water use. In behavioural terms, the crisis has likely changed consumers’ water usage reference point. Long showers, green lawns and frequent hose use were no longer perceived as neutral; they came to be seen as wasteful.

A natural question one then asks is, what kinds of behavioural nudges work when it comes to water? The evidence suggests that social comparison is one of the strongest tools. People often do not know whether their own use is high or low, but they care when they learn they are using more than similar households. Randomised trials found that home water reports comparing households with peers recorded a reduction of water use by around 2.7% in one study, and by roughly 4.8% in another drought-period study (Bernedo et al., 2014; Ferraro and Price, 2011).

This matters because water use is partly invisible as a significant share of household water consumption goes ‘down the drain’. Social-comparison messages make the invisible visible. They also activate a powerful behavioural mechanism- most people do not want to be using far more than their neighbours.

A second promising nudge is social recognition. A Cape Town-based experimental paper found that the most effective nudges were ones that officially recognised a household’s conservation efforts, followed by messages appealing to the public good. In that study, social recognition performed better than other treatments, particularly among wealthier households (Brick et al., 2023). This is classic behavioural economics. Recognition changes behaviour not by punishing people, but by tapping into identity and status. People generally like to be seen as responsible citizens. In a city with high inequality like Cape Town, such is important. The same study found that responses differed by income group, with the poorest households being largely unresponsive to these nudges, probably because they were already using very little water and had limited room to cut back further (Brick et al., 2023).

Such insights point to an important policy lesson: nudges are not magic, and they are not equally effective for everyone. The nudges tend to work best where people still have discretionary consumption to trim. If a household is already close to subsistence use, information alone will not achieve much. In those cases, infrastructure, service quality and affordability matter more than clever messaging.

A third lesson from the research is that feedback works better when it is timely and concrete. A long-term study from Valencia (Spain) found that smart meter feedback and digital engagement helped many households sustain lower water use over time. Nearly half of participating households maintained meaningful reductions for more than two years after the programme began (Cominola et al., 2021).

This is encouraging as one of the biggest worries in behavioural policy is rebound: people responding first, then drifting back to their previous state. The Valencia evidence suggests that when feedback is regular, personalised and easy to interpret, conservation can become more durable (Cominola et al., 2021).

So, did Cape Town residents go back to old water consumption levels after Day Zero? Not fully. The city’s own recent reporting suggests a partial rebound, but also a meaningful long-term shift downward in per capita use and discretionary demand (City of Cape Town, 2024; City of Cape Town, 2025). Behavioural economists can at least be optimistic due to such findings. It shows that extreme events can create a window in which new habits form and that good policy can help lock some of them in.

The deeper lesson is that water demand is not just a technical issue. It is a behavioural one. People respond to prices, yes, but also to norms, recognition, salience and feedback. In other words, how we frame water use can shape how much water we use.

References

Bernedo, M., Ferraro, P.J. and Price, M.K. (2014). The persistent impacts of norm-based messaging and their implications for water conservation. Link: https://static1.squarespace.com/static/5926f8f75016e1695b3108ff/t/5fbe8066f8cdb769c6b8517d/1606320230225/Bernedo+et+al+JoCP+paper.pdf

Brick, K., De Martino, S. and Visser, M. (2023). Behavioural nudges for water conservation in unequal settings: Experimental evidence from Cape Town. Link: https://www.sciencedirect.com/science/article/pii/S0095069623000700

Brühl, J and Visser, M. (2021). The Cape Town drought: A study of the combined effectiveness of measures implemented to prevent “Day Zero” Link: https://www.sciencedirect.com/science/article/pii/S2212428421000025

City of Cape Town. (2024). Water Outlook 2024. Link: https://resource.capetown.gov.za/documentcentre/Documents/City%20research%20reports%20and%20review/Water_Outlook_March_2024.pdf

City of Cape Town. (2025). Water Outlook June 2025. Link: https://resource.capetown.gov.za/documentcentre/Documents/City%20research%20reports%20and%20review/Water-Outlook-June-2025.pdf

Cominola, A., Giuliani, M., Castelletti, A., Fraternali, P., Herrera Gonzalez, S. L., Guardiola Herrero, J. C., Novak, J. and Rizzoli, A. E. (2021). Long-term water conservation is fostered by smart meter-based feedback and digital user engagement. Link: https://www.nature.com/articles/s41545-021-00119-0

Ferraro, P.J. and Price, M.K. (2011). Using non-pecuniary strategies to influence behavior: Evidence from a large-scale field experiment. Link: https://www.nber.org/papers/w17189