July 8, 2025
KARACHI – The city’s water crisis is not just about scarcity; for too long, decisions have been based on unsubstantiated assumptions and misplaced priorities.
Every summer, Karachi faces a predictable water crisis. As temperatures climb, neighbourhoods across the city brace for dwindling water pressure, erratic supply schedules, and increased dependence on costly private tankers. The city’s water utility, Karachi Water and Sewerage Corporation (KWSC), insists the primary problem is a stark mismatch between the city’s water supply — approximately 550-650 million gallons a day — and its ballooning demand, assumed to be double that, around 1,200 MGD.
New projects to increase the city’s water supply are thus championed as the silver bullet to the city’s perennial water crisis. Chief among these is the K-IV bulk water supply scheme, a mega-project to double the city’s water supply launched nearly two decades ago that has yet to be completed. Despite repeated delays and cost overruns, the project remains the centrepiece of the city’s water planning strategy.
But the water demand estimates used to justify the massive infrastructure deserve closer scrutiny. Our research group conducted over 600 household surveys and installed several smart water meters (the first scientific study conducted at this scale in Pakistan) to find that actual household water use is far below official estimates. In some cases, less than half of what is assumed.
If Karachi builds its infrastructure on exaggerated and unsubstantiated estimates, it risks repeating Pakistan’s costly mistakes from the power sector, where inflated demand estimates led to massive overspending and a glut of unused electricity that ordinary citizens are still paying for through soaring electricity bills.
The problem with the numbers
This is the rather simplistic way the KWSC estimates how much water is needed for Karachi: the population of Karachi multiplied by per-person daily water consumption.
The body assumes that each resident of Karachi’s estimated 22 million population uses 54 gallons or 204 litres per day, resulting in a demand estimate of around 1,200 million gallons per day (54 gallons per day * 22 million people).
However, the estimate for the per-person water consumption (54 gallons per day) is not based on any recent, empirical study for Karachi. In fact, residential connections across the city are unmetered, and no peer-reviewed analysis of actual household water use had ever been conducted until recently. So, because they don’t have any data, the KWSC relies on guesswork.
This isn’t a new problem. Since the 1980s, various studies and government reports have projected dramatically inflated water needs for Karachi. In 1985, a planning study forecasted that water demand would rise to 81 gallons per capita per day (gpcd) by 2025. Another estimate in 1999 projected the city’s total demand to reach nearly 1,340 MGD by 2010. These numbers were never grounded in direct measurement. Instead, they reflected an overly optimistic vision of urban development, with assumptions based on Western-style household consumption and full appliance access, which simply does not reflect how most people in Karachi use water.
These overestimations shape billion-rupee investment decisions. They justify large-scale infrastructure projects and continued borrowing, even when actual consumption may not warrant it. Misjudging demand risks funnelling scarce public resources into costly supply augmentation projects, when more targeted, cost-effective interventions could achieve greater impact.
The second major flaw in Karachi’s water demand projections is the assumption that all residents use water in the same way, regardless of season, income, or neighbourhood. The KWSC’s estimate — 54 gallons per capita per day — is applied uniformly across its entire population, as if every single person in Karachi needs the same amount of water every day. In reality, household water use varies widely across the city and changes substantially over the course of the year.
Our studies, one employing water use questionnaires and the other using high-resolution smart meters, both show that actual usage varies dramatically across different households and seasons. Water use tends to rise during hotter months, particularly when daily temperatures exceed 31 degrees Celsius, and declines in the winter. Wealthier households consistently use more water than poorer ones. Even within a single household, usage fluctuates across the week.
Therefore, treating water demand as a single number flattens all this complexity.
What the evidence suggests
The limits of Karachi’s demand projections become even clearer when we look closely at how water is used in low-income neighbourhoods. In Lyari, a township marked by dense housing, poor infrastructure, and limited piped supply, we surveyed over 600 households across the summer and winter seasons to understand how much water people use. Most households in the neighbourhood rely on a mix of unreliable piped supply, private vendors, and storage tanks.
Across our sample, the average daily use was just 16 gpcd in the summer and 13 gpcd in the winter — less than a third of what is assumed by the KWSC. Clearly, looking at just the current water usage underestimates the true water demand. That is, if more water were available, households would use more. We also test this hypothesis in our work and find that for the subsample of households who reported satisfaction with their water access, usage increases to 23 gpcd and 14 gpcd in the summer and winter months, respectively. These numbers suggest that even when water is fully available, actual demand remains far below official estimates.
CHART PROVIDED BY DAWN
The findings also reveal stark differences. Not every household is the same. Richer households in one area can invest in infrastructure, secure informal connections, or organise collectively to bring in tankers. Poorer families, in contrast, pay more per unit of water, endure longer wait times, and face greater exposure to unsafe sources.
By portraying Karachi’s water crisis as a system-wide shortfall, the KWSC shifts attention away from its failings in distribution, leakage control, and service reliability. The story of an overwhelming supply-demand gap becomes a convenient excuse, obscuring the fact that many residents are left underserved, not because there isn’t enough water, but because water is mismanaged and unfairly allocated. Solving Karachi’s water crisis will require more than new pipelines — it will require a fundamental shift in how we understand who needs water, how much, and at what cost.
While household surveys help establish broad patterns of water use, they are limited by recall bias and self-reporting. They also represent primarily low-income households. To develop a more robust and holistic picture of water usage in the city, we developed and deployed a low-cost, resilient smart metering system in middle to high-income households across Karachi. These devices recorded actual water flow every 30 seconds, providing millions of data points across 17 months in 23 households. This is the first study of its kind to generate daily-scale water use data in a South Asian city facing intermittent supply.
The findings further undermine the case for uniform, inflated demand projections. We observed significant variation in daily water use both across and within households. Only two out of the 23 households had average water use higher than 54 gpcd. Overall, the average across the entire sample was just 22 gpcd, less than half of the figure used by the KWSC.
The usage also fluctuated with temperature. Once daily highs crossed 31 degrees Celsius, household water use rose sharply, indicating heat-related behavioural changes. Fridays also stood out, with a clear bump in water use likely tied to religious practices. But even during the hottest days or highest-use periods, few households approached the levels that the utility uses in its citywide projections.
CHART PROVIDED BY DAWN
The sample size of the smart water meters study is not large enough for us to definitively establish Karachi’s true water demand. However, even with the current sample, we have enough statistical power to show that household usage is far lower than the KWSC’s assumptions and varies considerably across seasons and households.
A more holistic accounting of the city’s water needs must survey more households and incorporate commercial and industrial consumption. It is the KWSC’s responsibility to gather this evidence before making multi-billion-rupee infrastructure commitments.
The corporation, like other utilities in the region, estimates total citywide demand by taking residential usage and adding a fixed percentage for system losses and industrial use. But this approach assumes all components grow at the same pace and ignores underlying structural changes. For example, Karachi’s industrial landscape is evolving. Global economic shifts and changing water-use technologies may reduce industrial consumption even as population growth increases residential demand. Yet these trends are not accounted for in the fixed-percentage planning model. Without better data, the city is flying blind.
Moreover, the assumed per capita use in Karachi, 54 gpcd, is higher than what is reported for many cities. In Shiraz, Iran, it is 35 gpcd; in Mumbai, the utility plans for 39 gpcd. Karachi’s assumptions are not just wrong, they are also outliers compared to other cities. If actual usage is lower and more varied than assumed, the projected supply-demand gap narrows significantly. The case for large-scale supply augmentation, such as the K-IV, begins to look far less certain. Its projected cost has ballooned from Rs25 billion to more than Rs190 billion, yet the project remains incomplete.
More concerning than the delays and overruns, however, is the way the K-IV has come to dominate the city’s water planning agenda. It has drawn political attention and institutional focus away from more immediate and manageable reforms like improved billing or equitable distribution. This is the risk with mega-projects: they become both a symbol and a distraction, offering the illusion of progress while sidelining practical solutions.
Refocusing the water agenda
Karachi does face real water shortages, and new supply investments may well be necessary given the impact of climate change. But the case for costly supply augmentation must be grounded in evidence. Additionally, for any water supply project, two important questions must be addressed. First, who will benefit from this new water? Will it go to neighbourhoods that are currently underserved, or will it reinforce existing inequalities? Second, how much will the people of Karachi pay for these projects, and for how long?
The city urgently needs investment, but not only in pipes and pumps. Allocating water more equally across the city and pricing reforms would likely deliver a greater impact on people’s well-being, and crucially, do not need billions of rupees to carry out. These are management challenges, not capacity ones. Prioritising them could improve outcomes for the majority of Karachi’s residents without locking the city into unsustainable capital projects. There are indications that the recently revamped KWSC has started addressing some of these issues.
Clearly, Karachi’s water crisis is not just about scarcity. For too long, decisions have been based on unsubstantiated assumptions and misplaced priorities. Karachi can either recalibrate its water management strategy based on accurate, data-driven realities, or it can continue down a financially unsustainable path of infrastructure expansion built on misleading assumptions.
The choice the city makes today will determine whether billions of rupees are wisely invested or recklessly wasted.