Morocco’s drip conversion: efficient farms, emptier aquifers

Drip irrigation delivers water to the root rather than the field, and applies less of it per hectare. That makes subsidising the conversion an attractive policy: farmers get higher yields, the state gets a saving, and nobody has to be told to stop.⁶

Morocco ran that policy at national scale under the Plan Maroc Vert, through the Programme National d’Economie d’Eau en Irrigation, with donor support.⁴ It was planned to convert about 555,000 hectares between 2008 and 2020.² Coverage went from 9 per cent of the irrigated area in 2008 to about 43 per cent in 2020 and roughly 794,000 hectares, about half, by the end of 2023.¹ As conversion programmes go, it worked.

The state paid most of the bill. Subsidies ran at 80 per cent above five hectares and 100 per cent below.⁶ They are now tiered at 100, 75 and 60 per cent by farm size, with a ceiling of MAD 38,000 a hectare for the smallest farms and MAD 28,500 for farms of five to twenty hectares.³ Two tracks ran in parallel: individual conversion, which farmers applied for directly, and collective schemes that modernised whole irrigation sectors.³⁵

One thing the design did not include was a limit on how much water a farmer could then take. A plan to phase in irrigation tariff increases, attached to World Bank policy lending for the Plan Maroc Vert, was withdrawn in 2011 after vocal farmer opposition, as unrest swept the region.⁴ The subsidy schedule also pays for densifying date-palm groves, and drip lets farmers densify orchards generally, which raises water use per hectare.³⁶

Where drip came with a reliable surface supply, it did what it promised. In the PROMER project, groundwater abstraction in Tadla fell 43 per cent, which the completion report attributes to on-demand surface water service. Groundwater consumption fell 61 per cent on an experimental remote-sensing estimate. Small-farm production rose 166 per cent in Tadla and 142 per cent in Doukkala. Internal rates of return re-estimated at closing came in between 16.5 and 25.5 per cent, above appraisal.⁴

The clearest evidence is a natural experiment inside the programme. Remote sensing and field surveys reported in the 2023 completion report of the successor Large Scale Irrigation Modernization Project found total consumption in the PROMER area of Tadla down about 15 per cent between 2015 and 2021, with productivity up 47 per cent, while Haouz showed no such fall. The difference the report identifies is that Tadla adopted water quotas and Haouz did not.⁷ The technology was the same in both.

Nationally, demand did not fall. Morocco’s Cour des comptes reported in its 2023–2024 annual report that the modernisation and drip promotion effort had not stabilised irrigation demand, and that the problem of groundwater overexploitation had worsened.¹ The World Bank says the same thing in a background note to its country climate and development report, naming the mechanism as Jevons’ paradox, with groundwater withdrawals running 28 per cent above renewable resources.⁵

The aquifer record supports them. At one upstream piezometer in the Souss the rate of decline accelerated from 1.9 metres a year between 1998 and 2008 to 3.5 metres a year between 2012 and 2020. Across the Souss-Massa plain the correlation between the rainfall index and the groundwater index is 0.07, which the authors read as factors other than meteorological drought driving the fall.⁸

The saving was spent on more farming. Efficient irrigation makes a hectare cheaper to water, so farmers plant more of it and choose thirstier, more valuable crops.⁵⁶ Nothing in the programme prevented that, and the subsidy for densification encouraged it.

The money also did not reach the farmers it was aimed at. The World Bank’s background note records 395,000 hectares of individual uptake, 117 per cent of target, while adoption was slower for the 70,000 small farmers targeted in collective schemes.⁵ A joint FAO and World Bank note describes the drip subsidies as mainly for groundwater-based systems, and larger operations as the main beneficiaries of farm equipment support.⁹ Under the later Large Scale Irrigation Modernization Project, approved requests for improved irrigation technology reached 42 per cent of target, and 8 per cent in Doukkala.¹⁰

Measurement was never built. Groundwater monitoring under PROMER amounted to meters on 20 farms in Haouz and eight in Tadla.⁴ As of June 2026, none of the 4,000 wells targeted for metering in Chtouka had been fitted.¹¹ And in November 2015 the World Bank removed the promotion of more sustainable irrigation water use from PROMER’s development objective. Its completion report presents the change as realism about timeframes rather than a change in the theory of change.⁴

Efficiency is not a water policy on its own. It is a way of making each hectare cheaper to irrigate, and the response to a cheaper input is to use more of it. The transferable finding here is unusually clean, because Morocco ran both arms of the experiment: drip with quotas cut consumption in Tadla, and drip without them did not in Haouz.⁷

Pair the subsidy with a cap, and make the cap measurable before the equipment goes in. A programme that funds conversion but not metering has no way to know what it bought, and no way to enforce a limit it might later want. Twenty-eight meters across two irrigated regions is not monitoring.⁴

Watch who collects. A capital subsidy claimed by application favours farmers with the capital to co-invest and the borehole to irrigate from, which helps explain why individual uptake outran its target while the smallholder collective track fell short.⁵⁹ If the equity objective matters, the collective route needs its own delivery machinery rather than the same window.