16 economies withdraw more freshwater than their available renewable supply provides each year — and in 12 of them, agriculture alone already exceeds it

Say "water-stressed Gulf" and the picture is a desert city: air-conditioned towers, hotel fountains, swimming pools set in the sand. We pulled the complete water-withdrawal record for every economy that reports one, and the water at the top of the list is not going to the city. It is going to the field. The measure is SDG indicator 6.4.2 — "level of water stress" — compiled by the UN's FAO through its AQUASTAT system and republished in the World Bank's World Development Indicators. It is a ratio: all the freshwater a country withdraws in a year, divided by the renewable freshwater available to it after the flow the environment needs is set aside. Below 100%, a country withdraws less than that available renewable supply; above 100%, more. The excess withdrawal comes from beyond that environmental-flow-adjusted budget in one of three ways the ratio cannot tell apart: simply withdrawing into the share set aside for the environment (a country can cross 100% while still below its total renewable resources), drawing renewable rivers and aquifers down faster than they recharge, or tapping fossil groundwater that doesn't refill on a human timescale ^mmodeled. (Desalinated seawater is a separate, non-conventional supply that sits outside this ratio's numerator — it lets an arid state meet demand without a freshwater withdrawal, so it cannot by itself push the ratio above 100%.) The ratio alone does not identify which source fills the gap, and cannot on its own prove it ^sspeculative. We read the whole series, not a leaderboard of the worst. Of the 217 economies the World Bank tracks, 176 carry a 2022 figure; 41 report none, and every claim here is scoped to the 176 that do. Across those 176 the median economy withdraws just 10.7% of its renewable supply, and 117 of them — two in three — sit below 25%. The UN puts the global figure near 18% ^mmodeled. Against that backdrop, 16 economies stand above 100%. Ranked by ratio: Kuwait 3,850%, the United Arab Emirates 1,510%, Saudi Arabia 974%, Libya 817%, Qatar 431%, Yemen 170%, Algeria 145%, Egypt 141%, Turkmenistan 135%, Bahrain 134%, Syria 124%, Uzbekistan 123%, Sudan 119%, Oman 117%, Pakistan 110% and Jordan 105%. Geographically it is the arid belt — the Middle East and North Africa, the two irrigated Central Asian republics, and Pakistan — not a scatter of rich outliers. Now the part a single ranking hides. In 14 of the 16, agriculture is the single largest use of water — more than half of every liter withdrawn. The only two exceptions are the smallest, Qatar and Bahrain, Gulf city-states where households and municipal supply dominate and farming is about a third of the draw. Industry barely registers anywhere on the list: it peaks at 12.4% in Oman and stays under 8% in 15 of the 16. The heavy draw is not thirsty cities and it is not factories. It is agriculture. Push the arithmetic one step further and the point sharpens. Multiply each country's farm share of withdrawals by its overall stress ratio, and in 12 of the 16, agriculture *by itself* draws more than the country's entire available renewable supply. Pakistan withdraws 94% of its water for farming at a 110% overall ratio — so agriculture alone runs at about 103% of its renewable supply. Egypt's farms alone reach 112%, Uzbekistan's 113%, Sudan's 114%, Turkmenistan's 125%, Saudi Arabia's 795%, Kuwait's roughly 2,400%. Whatever their households and industries add sits on top of a farm draw that has already broken the budget. (Algeria at 98% and Oman at 94% fall just short of that farm-alone threshold and are not counted among the 12.) It is not a rich-country story either. The 16 split 6 high-income, 4 upper-middle, 3 lower-middle and 3 low-income — Kuwait and Sudan indicted by the same measure. And the largest total withdrawal belongs to none of the petrostates: Pakistan withdraws 179.3 billion cubic meters a year, 94% of it for agriculture — more than Egypt (77.5), Uzbekistan (42.9) and Sudan (26.9) combined, the next three by volume. (This is total withdrawal, not the excess-over-renewable, which the ratio does not resolve into a volume.) One blind spot belongs at the top, not buried. This ratio cannot tell a country drawing on fossil groundwater that will never refill apart from one over-pumping renewable aquifers and rivers faster than they recharge — or encroaching on the flow reserved for the environment. All read as "over 100%," and the ratio does not distinguish them, or say how much of the excess is permanent loss of a finite stock versus a renewable source drawn too hard. That distinction, not the ranking, is what a reader should weigh — and among these economies this piece isolates it at source only for Pakistan, whose Indus-plain groundwater the Notable documents; for the others it is widely reported but not established by this dataset.

The following is analysis, not fact. Read any single seat on this list and you get the cliché — dry Gulf, extravagant cities. Read all 176 at once and the intuition inverts twice. First, the water at the extreme is agricultural, not municipal: in 14 of the 16 economies over 100%, farming is the majority use, and industry is a rounding error. Second, the overdraft is not a marker of wealth. It tracks aridity and agriculture, and it catches low-income Yemen, Syria and Sudan in the same net as high-income Saudi Arabia and Kuwait. A reader who starts from "rich desert states waste water" ends the full read somewhere else: the common thread is a farm sector drawing on water that isn't there. The farm-alone figure is the piece only the complete arithmetic surfaces, and it has a policy edge. Where agriculture by itself exceeds the available renewable supply — 12 of the 16 — cutting household and municipal use cannot on its own bring total withdrawal back under the renewable budget, because farming has already overshot it before a household draws a drop. (The exception inside the 12 is Qatar, where municipal use dominates and farming clears 100% only off an extreme overall ratio — there, household measures would move the total more.) That is a different problem from the one the "water-stressed city" frame implies, and for most of the 12 it points at a different lever: what is grown, and how it is watered, not how long the tap runs at home. What the data establish, and what they don't. They establish, by definition and by re-derivable arithmetic, that these 16 economies withdrew more than their available renewable supply (renewable resources after the environmental-flow requirement SDG 6.4.2 sets aside) in 2022, that farming was the majority draw in 14 of them, and that farming alone exceeded supply in 12. They do not establish that any given country's water table is falling this year — a country drawing hard on renewable rivers and aquifers, or reading high in a single dry year, can sit over 100% without permanently mining a finite stock — nor do they explain why the geography is arid or assign blame for the policy that irrigates a desert. The ratio locates the overdraft and names its dominant user. It does not diagnose the cure. We hold no brief for any government on this list. The same measure that flags petro-rich Kuwait flags war-strained Syria and low-income Sudan, and we report all of them the same way — as the arithmetic of their own reported withdrawals against their own renewable supply.

The strongest objection is that a single ratio flattens two physically different situations. A country over 100% because it over-pumps renewable aquifers and rivers — stocks that recharge, if drawn less hard — is not in the same danger as one over 100% because it mines fossil groundwater that took millennia to accumulate and will not refill. Both draw beyond the renewable, environmental-flow-adjusted budget, but only the fossil case is, in the stock sense, self-terminating. Pakistan's Indus plain is the depletion case this piece can source (see Notable), while the split for the North African and Central Asian examples is widely reported but not established by this dataset. (Desalination doesn't enter this ratio: built outside the freshwater system, it lowers the freshwater a country must withdraw rather than raising the figure.) Ranked purely by ratio, Kuwait's 3,850% looks like the crisis — but the more alarming cases for permanent depletion may be the ones just over the line, where the water comes out of a finite underground bank. The number we lead with does not carry that distinction, and a reader should not infer that a higher ratio means a nearer cliff. Second, this is one year against a long-run average. Renewable freshwater resources are estimated as a multi-year normal; a single year over 100% can reflect a dry spell as much as structural overuse. A ratio that sits above 100% year after year is unsustainable by definition, but our pull is the 2022 reading, and we scope the claim to it. Third, the "agriculture alone exceeds supply" finding is arithmetic on two published shares — farm share of withdrawals times the overall ratio — both rounded in the source. The clear cases (Kuwait, Saudi Arabia, Libya, Turkmenistan, Uzbekistan, Sudan, Egypt, Pakistan) survive any rounding; the borderline ones do not, which is why Algeria (98%) and Oman (94%) are excluded from the 12 rather than rounded up. So the 12 are the cases robust to the source's rounding — Algeria and Oman are the only borderline exclusions, both within a few points of the line — not a precise census. Fourth, the measure itself embeds assumptions. SDG 6.4.2 subtracts an estimated environmental-flow requirement and leans on AQUASTAT's country reporting, some of which is modeled or lagged. The threshold of "100%" is exact; the inputs feeding it are estimates, and a different environmental-flow assumption would move the borderline entries more than the extremes.