Five groups, sixteen sub-models
Grouping follows the structure of the arrangement, not the name a programme gives itself. Two countries can call the same mechanism by different names, and one name can cover several mechanisms. The test applied here is what the state puts in and what it keeps.
Thirteen criteria, no weighting
The same criteria are applied to every group in the same order. Each is answered in one or two sentences and carries an evidence status. None is scored, and none is combined with another.
Economic impact means food security, not treasury return
Economic impact here is import substitution and national self-sufficiency. It excludes corporate tax receipts, lease and licence revenue to the state, fiscal return on public investment, and land sale proceeds. Where a model draws in private capital, that is recorded as development achieved — not as revenue received.
Employment is counted under social impact, as citizen livelihood and rural continuity, not as a contribution to the tax base.
Produced domestically 1.1 million t
National wheat area 183,000 ha at 6 t/ha 10,000 ha at the national yield → 60,000 t → displaces 1.9% of wheat imports
The criterion asks what an additional area of production actually removes from the import bill. Stated in tonnes displaced, not in hectares developed.
Domestic share today 1.1 ÷ 4.2 = 26%
Production needed for 50% coverage 2.1 million t Shortfall 1.0 million t → ~167,000 ha at 6 t/ha → close to doubling the existing wheat area
The criterion converts a coverage target into the area a model would have to reach. It is what turns a policy aspiration into a land requirement.
Barley imported, MY 2026/27 projected 4.5 million t
Corn imported, MY 2026/27 projected 4.66 million t Coverage on feed grains starts from zero — the largest uncovered volume in the tonnage
Coverage is not uniform across commodities, and the gaps are not where intuition puts them. Feed grain is the largest uncovered volume, yet barley is water-intensive — which is why the criterion is reported alongside water rather than on its own.
Wheat import origins 4 principal suppliers Resilience reads as origin concentration, not as import volume alone
Two commodities can be imported at similar tonnage and carry entirely different exposure. A model producing a commodity drawn from one origin improves resilience more than the same tonnage of a commodity drawn from four.
Production programme ended MY 2015/16, by decree
Quota reduction applied 12.5% a year from 2007/08 Volume was never the constraint. The water source was
Volume and reliability are separate readings. A model can deliver the first in full and still fail the second, because reliability is a property of the resource rather than of the mechanism.
Wheat delivered Jun – Oct — one season
Morocco lease terms by what is planted 17 / 25 / 40 years Annual crops yield inside one term; perennials require a term built around the delay
Time to production is what forces tenure length. It is also the criterion that separates models the state can test quickly from models it is committed to for a generation.
Area placed under lease ~112,000 ha
Investment committed 22.3 billion MAD ≈ 199,000 MAD per hectare of committed private investment
Capital mobilised is recorded as development achieved, never as revenue received. Nothing here is money reaching the treasury.
Saudi wheat quota, participating farms up to 5,000, plots to 50 ha Participation is set by the allocation rule, not by the mechanism
Competitive tender produces a low figure unless a tier is reserved. The same instrument reaches citizens or excludes them depending on one drafting decision.
Across ~112,000 ha ≈ 0.56 jobs per hectare, contracted rather than projected
Employment counts here as citizen livelihood, not as a contribution to the tax base. Density varies sharply by production system — protected agriculture and centre-pivot cereals are not comparable on this criterion.
Companies and organisations up to 400 ha
Special projects up to 4,000 ha Ceiling to floor = 800 : 1 — the scale advantage sits inside the rule itself
Equity is measurable before a single hectare is allocated, because it is written into the size bands. This is the criterion most often decided by drafting rather than by outcome.
Settler households ~112,600
Roads, water, schools, clinics public grant Continuity is built where holdings are standardised and cannot consolidate
The contrast is with freehold transfer, where nothing prevents consolidation afterwards and concentration is the default direction of travel.
Delivered as inputs in kind + technical support + offtake Transfer occurs where technical capacity is bundled, not where it is merely required
A training obligation written into a lease and never audited transfers nothing. The criterion asks who is actually standing in the field.
Minimum viable area, protected agriculture 2 ha Spread of 500 : 1 — an area allocated for cropping removes a grazing block, not a field
Displacement is read against the production system already present, not against the area's classification. The Kingdom's 1968 ordinance is recorded as having reshaped tenure relations and contributed to the decline of pastoral nomadism — the clearest instance of a transfer mechanism acting on customary use.
Implementation reference
Where a model has actually been implemented, the country and programme are named at the point of use and the source is linked. Where no implementation has been located, the entry says so plainly and the model is carried as a structure rather than as a record.
| What the entry shows | What it means |
|---|---|
| A country named | The model has been implemented there. The outcome described is what was recorded, with the source linked. |
| Variant of the group | A real mechanism in use, but not operated as a separate named programme. Its distinction from the parent model is structural. |
| No known implementation | No programme operating this model has been located. It is set out so the field of options is complete, and marked so it is not mistaken for a record. |
Where resource stewardship sits
Water and land condition are not scored as a third pillar. They are conditions on a model working, and they are answered inside each group's key success factors. Three questions run for every group:
| Incentive alignment | Does the model reward water efficiency, or reward volume? |
| Term against horizon | Does the tenure term granted exceed the water source's remaining horizon? |
| Enforcement | What can the state actually do if the resource is degraded? |
Weighting social against economic impact is a political judgement
A composite score would require deciding how much citizen participation is worth against how much import substitution. That decision belongs to the Ministry, not to an analysis. Producing a single number would embed a policy choice inside what presents as a technical output, and would conceal the trade-off the comparison exists to show.
This follows the same rule applied in the evaluation itself, where capability, water, land rights and crop suitability are reported as four independent answers and never averaged.
Area type, not area size
Each model states which area types it suits, read from the classification the evaluation already produces. No areas are quantified here — the ledger belongs to the evaluation. What matters for model choice is the kind of constraint an area carries, because that is what opens or closes a mechanism.
| Area type | Read from | What it means for model choice |
|---|---|---|
| Ready now | C1–C2 · W1 · E1 | Capable, watered and legally clear. Every model in the catalogue is available. |
| Blocked on paperwork only | E2, all else clear | Cannot be sold or leased until tenure is registered. Blocks every transfer mechanism until resolved — and is the most actionable finding in the study, because it is administrative rather than physical. |
| Finite water horizon | W2y | Usable now on a declining non-renewable source. Argues against perpetual transfer, since the state would part with land whose productive life it cannot guarantee. Favours terms matched to the supply horizon. |
| Conditional water | W3 · W4 | Restricts the field to models that carry their own water solution. |
| Marginal, viable with investment | C3 · S3 | Requires investment to become productive. Favours models where the state shares development risk. |
| Rangeland | Grazing production system | A different operator class entirely, and the only type for which a grazing instrument rather than a cropping instrument applies. |
| Not developable | CN · W4 · E3 | No model applies. |
This module does not re-open whether land is suitable
The evaluation answers whether agriculture is possible, watered, permitted and suited to a crop. It never answers whether it is profitable. This catalogue answers only by what mechanism land reaches an operator, and takes the evaluation's findings as given.
Keeping the two apart is what stops a permission question being answered with an economic argument, or the reverse.
Every documented claim, traceable to its source
Each source carries a number used throughout this application. Where a model has been implemented, the numbered chip beside it opens the source describing that implementation. Statements that follow from how a mechanism is built carry no source, because none exists to carry — and where no implementation has been located at all, the entry says so.
Eleven of sixteen sub-models have been implemented somewhere
Much of what can be said about an allocation mechanism follows from how it is built rather than from what it produced. Only where a programme actually ran does a recorded outcome exist — and for five of the sixteen sub-models, no implementation has been located at all.
That balance is itself worth stating. Presenting structural reasoning and recorded outcome at identical weight would conceal it.